Comparative Governmentand Politics

An Introduction6th Edition

Rod Hague and Martin Harrop

Comparative Government and Politics

Visit the companion website for thistext at:

http://www.palgrave.com/politics/hague

COMPARATIVE GOVERNMENT AND POLITICSFounding Series Editor: The late Vincent Wright

Published

Rudy Andeweg and Galen A. IrwinGovernment and Politics of the Netherlands

Nigel BowlesGovernment and Politics of the United States (2nd edition)

Paul BrookerNon-Democratic Regimes: Theory, Government and Politics

Robert ElgiePolitical Leadership in Liberal Democracies

Rod Hague and Martin HarropComparative Government and Politics (6th edition)

Paul HeywoodThe Government and Politics of Spain

B. Guy PetersComparative Politics: Theories and Methods[Rights: World excluding North America]

Tony SaichGovernance and Politics of China (2nd edition)

Anne StevensThe Government and Politics of France (3rd edition)

Ramesh ThakurThe Government and Politics of India

Forthcoming

Judy BattGovernment and Politics in Eastern Europe

Robert LeonardiGovernment and Politics in Italy

Comparative Government and PoliticsSeries Standing OrderISBN 0–333–71693–0 hardcoverISBN 0–333–69335–3 paperback(outside North America only)

You can receive future titles in this series as they are published by placing a standing order. Please contact your bookseller or, in the case of difficulty, write to us at the address below with your name and address, the title of the series and an ISBN quoted above.

Customer Services Department, Macmillan Distribution LtdHoundmills, Basingstoke, Hampshire RG21 6XS, England

Comparative Government and Politics

An Introduction

6th Edition

Rod HagueandMartin Harrop

© Rod Hague and Martin Harrop 1982, 1987, 2001, 2004© Rod Hague, Martin Harrop and Shaun Breslin 1992, 1998

All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission.

No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limitedcopying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP.

Any person who does any unauthorized act in relation to this publication may be liable to criminal prosecution and civil claims for damages.

The authors have asserted their rights to be identified as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988.

First edition 1982Second edition 1987Third edition 1992Fourth edition 1998Fifth edition 2001Sixth edition 2004

Published byPALGRAVE MACMILLANHoundmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010Companies and representatives throughout the world

PALGRAVE MACMILLAN is the new global academic imprint ofSt. Martin’s Press LLC Scholarly and Reference Division andPalgrave Publishers Ltd (formerly Macmillan Press Ltd).

ISBN 1–4039–1314–5 hardcoverISBN 1–4039–1315–3 paperback

This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources

A catalogue record for this book is available from the British Library.

10 9 8 7 6 5 4 3 2 113 12 11 10 09 08 07 06 05 04

Printed in China

PART I Foundations1 Politics and government 12 The state in a global context 173 Democracy 354 Authoritarian rule 515 The comparative approach 69

PART II Politics and society 6 Political culture 897 Political communication 1058 Political participation 122

PART III Linking society and government9 Elections and voters 145

10 Interest groups 16611 Political parties 185

PART IV Government and policy12 Constitutions and the legal framework 20913 Federal, unitary and local government 22814 Legislatures 24715 The political executive 26816 The bureaucracy 29017 The policy process 309

Summary of Contents

v

vi

List of illustrative material viiiPreface x

PART I Foundations

1 Politics and government Politics 3Government 4Governance 6The state and sovereignty 7Nations and nationalism 8Nations and states 10Power 12Authority 13Legitimacy 15Key reading 16

2 The state in a global contextHow the state emerged 17The Western state 19The post-colonial state 21The state and international organizations 25The state in the global economy 30Key reading 34

3 DemocracyDirect democracy 36Representative and liberal democracy 38Waves of democratization 39New democracies 43Semi-democracy 46Key reading 49

4 Authoritarian ruleTraditional authoritarian rule 51Communist, fascist and military rule 53The Arab and Muslim worlds 62China in transition 66Key reading 68

5 The comparative approachWhy compare? 69The risks of comparison 71

Comparing institutions, societies and states 75Techniques of comparison 79Key reading 85

PART II Politics and society

6 Political culture The civic culture 89Political trust and social capital 90Postmaterialism 93Political culture in new democracies 95Political culture in authoritarian states 96Elite political culture 98Political socialization 100Islam and the West 102Key reading 104

7 Political communication Development of the mass media 106The media: contemporary trends 108The media in established democracies 112The media in new democracies 115Public opinion 116The media in authoritarian states 119Key reading 121

8 Political participationParticipation in established democracies 122Participation in new democracies 124Social movements 127Participation in authoritarian states 131Political violence and terror 134Revolution 136Key reading 141

PART III Linking society and government

9 Elections and votersScope and franchiseElectoral systems: legislatures 146Electoral systems: presidents 154

Contents

Voting behaviour 156Turnout 159Referendums 160Elections in new democracies 163Elections in authoritarian states 164Key reading 165

10 Interest groupsClassifying interest groups 167Channels of access 169Iron triangles and issue networks 172Conditions of influence 175Pluralism and corporatism 177Interest groups in new democracies 181Interest groups in authoritarian states 182Key reading 184

11 Political parties Party organization 186Selecting candidates and leaders 189Membership and finance 191The social base 193Party systems 194Parties in new democracies 200Parties in authoritarian states 201Key reading 205

PART IV Government and policy

12 Constitutions and the legal frameworkConstitutions 210Judicial review and constitutional courts 212Judicial activism, independence and

recruitment 216Administrative law 221Law in new democracies 223Law in authoritarian states 224International law 225Key reading 227

13 Federal, unitary and local government Federalism 228Federal-state relations 231Assessing federalism 235

Unitary government 236Local government 239Subnational government in new

democracies 244Subnational government in authoritarian

states 245Key reading 246

14 Legislatures Structure 248Functions 252Recruitment 259Legislatures in new democracies 263Legislatures in authoritarian states 264Key reading 267

15 The political executive Presidential government 268Parliamentary government 274Semi-presidential government 281The executive in new democracies 284The executive in authoritarian states 285Key reading 289

16 The bureaucracy Evolution 291Recruitment 292Organization 293Accountability 299New public management 301Bureaucracy in new democracies 303Bureaucracy in authoritarian states 304Key reading 307

17 The policy process Stages of the policy process 309Public policy in established democracies 315Public policy in new democracies 319Public policy in authoritarian states 322Key reading 324

Appendix: list of boxes 325References 326Index 348

CONTENTS vii

viii

MapsMap of the world xii

4.1 Post-communist Eastern Europe and Central Asia 55

4.2 Asia and the Middle East 6510.1 The European Union 173

Tables3.1 The British electorate as a percentage

of the adult population, 1831–1931 424.1 Post-communist states in Eastern

Europe and the former Soviet Union 544.2 Islam and democracy, 2001 626.1 Confidence in political institutions,

2001 927.1 The development of communication

media 1077.2 Countries in which a majority of

people use the internet, 2002/03 1097.3 Time taken for new media to reach

50 million households in the USA 1117.4 Countries ranked by press freedom,

2001/02 1168.1 Female representation in national

legislatures, 1950s–2000s 1258.2 The ten larger demonstrations against

the Iraq War, by city, 15–16 February 2003 130

8.3 World Trade Center fatalities by nationality, September 11, 2001 135

8.4 The ten most lethal governments, 1900–87 136

9.1 The Canadian elections of 1993 and 2000 150

9.2 Explicit thresholds of representation in some PR systems 152

9.3 How the additional member systemworks: Germany, 2002 153

9.4 Declining turnout at national elections, 1950s–1990s 159

9.5 Turnout in some countries with compulsory voting 161

11.1 Falling party membership in some European democracies, 1960–99 192

13.1 Some federations in established democracies 230

13.2 Subnational government in unitary states: some European examples 238

13.3 Average population of elected local authorities in some European democracies 241

14.1 Selection to the upper chamber in some established democracies 250

14.2 Average length of service in national parliaments 263

16.1 The structure of Germany’s Ministry of Economics and Labour 296

17.1 Introduction of social insurance to some democracies 316

17.2 Case-by-case privatization in the United Kingdom, 1980s 319

Figures1.1 Number of states belonging to the

United Nations, 1949–2002 82.1 Number of intergovernmental

organizations, 1909–97 263.1 The direct democracy of ancient Athens 373.2 Liberal democracy: the separation of

powers in the United States 414.1 The ending of military rule in Latin

America 605.1 Easton’s model of the political system 775.2 Population and assembly size 836.1 Almond and Verba’s theory of the

civic culture 906.2 Americans’ trust in the ‘government in

Washington’, 1964–2002 916.3 How elite political culture affects

political stability 996.4 Political learning and participation

across the life cycle 1017.1 The transmission model of political

communication 105

List of illustrative material

8.1 Patterns of participation in democracies 1238.2 The J-curve theory of revolutions 1389.1 Methods for selecting presidents 155

10.1 A peak association: Britain’s CBI 16910.2 Channels of interest group influence 16910.3 Iron triangles: how subgovernments

operated in the USA 17411.1 A bell-shaped distribution: parties

converge at the centre 19611.2 Organization of the Chinese

Communist Party 20412.1 Establishing constitutional courts in

Western Europe 21512.2 Levels of judicial activism in selected

democracies 21812.3 Administrative courts in France 22214.1 Population and assembly size 24814.2 Typical steps in making a law 25615.1 Presidential government 26915.2 Parliamentary government 27415.3 Formateurs and coalition governments 27715.4 Semi-presidential government 28215.5 Typical executive structure in

communist states 28716.1 Delimiting the bureaucracy 29016.2 Founding of cabinet-level departments

in the USA 29417.1 Stages of the policy process 309

DebatesState sovereignty or humanitarian intervention? 31‘Asian democracy’ 48Authoritarian rule as a recipe for economic development 57

The political significance of the internet 111Bowling alone 126Compulsory voting 161Is the USA a pluralist political system? 179Parties as vote-maximizers 197A political role for the judiciary 219A federal Europe? 237The decline of legislatures? 265Presidential government for a new democracy 273A representative bureaucracy? 295

ProfilesThe world 24Mexico 45Nigeria 61Germany 94Russia 128The United States 147The European Union 176Italy 198South Africa 213Canada 232United Kingdom 261France 283Japan 305

50 years of The European Union 29The European Court of Justice 217The European Parliament 255

BoxesA full list of boxes is provided in the Appendix.

LIST OF ILLUSTRATIVE MATERIAL ix

This edition retains the purpose of its predeces-sors: to provide a wide-ranging, contemporary andclearly written introductory text for courses incomparative politics, and for other introductorycourses in politics and political science. We haveretained the framework and chapter division usedin the previous edition but we have added 14 newsections and subsections addressing current devel-opments in the subject and the literature. We havealso rewritten many existing sections and revisedthe entire book to provide more recent examplesand references; to enhance clarity; and to takeaccount of suggestions from readers and our ownevolving appreciation of the subject matter.

It might be helpful to outline the thinkingbehind the new sections. In Part I, we have added‘Nations and states’ to the opening chapter,seeking to distinguish more sharply between twoconcepts that can no longer be presented as acompound ‘nation-state’. We have restructuredChapter 2 to present the state in a more historicaland global context, with new sections on ‘Howthe state emerged’, ‘The Western state’ and ‘Thepost-colonial state’. We hope the changes hereblend this chapter’s global themes more success-fully with the book’s comparative approach, asrecommended by Haynes (2003). Chapter 3 onauthoritarian rule includes extended coverage ofcontemporary authoritarian regimes with a par-ticular focus (in the light of increased interestsince 9/11) on ‘The Arab and Muslim worlds’.We have also added ‘China in transition’ not leastto show that authoritarianism is not just a featureof Islamic societies. The material on China, inthis section and elsewhere, also reflects thecountry’s growing importance in the worldeconomy.

In Part II, the chapter on political culture nowincludes a section on ‘Political trust and socialcapital’ as well as an introduction to culturalaspects of the relationship between ‘Islam and theWest’. Within the political communicationchapter, we have extended the material on mediaimpact through ‘Reinforcement, agenda-setting

and framing’. And the participation chapter nowcovers ‘Social movements’.

The decline in formal political participation incontemporary democracies is an issue that inter-ests students – and rightly so. This theme featuresmore significantly in Part III, with ‘Turnout’ and adebate topic on compulsory voting added to thechapter on elections and voters. Falling partymembership and its implications are carriedthough to the updated chapter on parties whichnow covers both ‘Party membership and finance’and ‘Selecting candidates and leaders’.

We have also strengthened Part IV on the struc-

Preface

Chapter 1 Politics and governmentNations and states

Chapter 2 The state in a global context How the state emergedThe Western state The post-colonial state

Chapter 4 Authoritarian rule The Arab and Muslim worlds China in transition

Chapter 6 Political culturePolitical trust and social capital Islam and the West

Chapter 7 Political communication Reinforcement, agenda-setting and framing

Chapter 8 Political participation Social movements

Chapter 9 Elections and votersTurnout

Chapter 11 Political partiesSelecting candidates and leaders Party membership and finance

Chapter 15 The political executive Presidential government: Brazil

B OX P. 1

New sections and subsections inthis edition

x

tures of government, aware that what should bethe core of the book was in danger of slidingtowards its periphery. In particular, we have addeda subsection on presidentialism in Brazil to thechapter on the executive, enabling us to broadenthe discussion of presidential government beyondthe United States. Our classroom experience isthat student interest in President Lula da Silva canbe carried though to a broader appreciation of thediversity in presidential systems. To reflect recentresearch on the parliamentary executive, we havealso rewritten ‘Minority and coalition govern-ment’.

We continue in this edition to use as the orga-nizing framework within each chapter a three-partdivision of states into established democracies,new democracies and authoritarian states (seeBoxes 3.1 and 4.1). Given that most new democ-racies have now survived for a number of years, wehave in this edition placed our discussion of themimmediately after the section on establisheddemocracies. This strikes us as a more intuitivesequence.

We invite you to visit our website. It supportsthis book specifically and your access to it is freeand unrestricted. A major new feature for thisedition is an entirely new chapter – only availableon the website – on comparative politicaleconomy.

We would like to thank our publisher StevenKennedy and his reviewers for their constructiveadvice; Keith Povey and Glynis Harris for theircopy-editing skills; and Tim Flower of FloridaInternational University for advising us on recentchanges to the decree powers of the Brazilian pres-ident. We would also like to acknowledge a moregeneral debt to the thousands of political scientistswho provided the findings and insights on whichthis edition, in particular, is based. Without theireffort, this book could not exist.

In all its manifestations, this book has now soldwell over 100,000 copies. We are grateful for thesupport reflected in this figure and also for thecorrections and suggestions provided by the manyteachers and students around the world who usethe book. We continue to welcome all feedback,not least because some errors of fact or interpreta-tion are bound to have crept into this new edition.Please contact Martin Harrop at

School of Geography, Politics and SociologyUniversity of Newcastle Newcastle upon Tyne EnglandNE1 7RU

e-mail Martin.Harrop@newcastle.ac.uk

Rod Hague Martin Harrop

PREFACE xi

� An extra comparative political economy chapter� Hague and Harrop’s Guide to Comparative Politics

on the internet � A searchable bibliography for each chapter� Chapter summaries � Essay questions and reading.

B OX P. 2

Key features of the websitehttp://www.palgrave.com/politics/hague

xii

Areas featured in Profiles are shaded on map

1 Canada p. 2322 European Union p. 1763 France p. 283

4 Germany p. 945 Italy p. 1986 Japan p. 305

7 Mexico p. 458 Nigeria p. 619 Russia p. 128

10 South Africa p. 21311 United Kingdom p. 26112 United States p. 147

World p. 24

Map of the World

7

9

12 5

10

1

3

8

2

11

6

4

In this part we set out the foundations of compar-ative government and politics. Chapter 1 outlinesthe key concepts of the subject while Chapter 2focuses in more detail on the state: its emergence,character and alleged crisis. The subsequent

chapters discuss the two main ways of organizingpower in the state: democracy (Chapter 3) andauthoritarian rule (Chapter 4). In Chapter 5, weconclude the part by discussing the comparativeapproach used in the book.

Part I

FOUNDATIONS

This page intentionally left blank

Politics 3Government 4Governance 6The state and sovereignty 7Nations and nationalism 8Nations and states 10Power 12Authority 13Legitimacy 15Key reading 16

In this book we examine the organization of poli-tics in countries around the world. We focus on

how nations solve the core political problem ofmaking collective decisions. But we cannot jumpstraight into these issues. For just as whatastronomers ‘see’ in the sky depends on the type oftelescope through which they peer, so too does anyinterpretation of politics depend on the conceptsthrough which we approach the topic. Indeed, inpolitics it often seems as though everyone has theirown telescope – and claims that their own instru-ment is best!

In politics, major concepts remain at the fore-front of discussion in a way that does not normallyapply to more scientific disciplines. Politicalanalysis is far more than mere opinion; yet even so,conclusions vary with the analyst rather more thanis comfortable for those who advocate a strictlyscientific approach to the subject. Comparativepolitics, based on a range of countries, is especiallysuited to the task of revealing contrasting perspec-tives on our subject matter. So in this chapter wediscuss some central concepts of the discipline, notso much to establish ‘correct’ definitions as tointroduce our own interpretations.

Politics

To start at the beginning: what is politics? We caneasily list, and agree on, some examples of political

activity. When the American president andCongress start their annual tussle over the federalbudget, they are clearly engaged in politics. Whenterrorists crashed hijacked planes into the WorldTrade Center and the Pentagon in 2001, their actswere patently political. The heartland of politics,as represented by such examples, is clear. However,the boundaries of the political are less precise.When one country invades another, is it engagedin politics or merely in war? Would politics occurif resources were unlimited? Is politics restricted togovernments or can it also be found in families,universities and even seminar groups?

A crisp definition of politics – one which fitsjust those things we instinctively call ‘political’ – isimpossible. Politics is a term with varied uses andnuances. Perhaps the nearest we can come to acapsule statement is this: politics is the activity bywhich groups reach binding collective decisionsthrough attempting to reconcile differences amongtheir members. Four significant points inhere inthis definition (Miller, 1991):

� Politics is a collective activity, involving peoplewho accept a common membership or at leastacknowledge a shared fate. Robinson Crusoecould not practise politics.

� Politics presumes an initial diversity of views,about goals or means or both. Were we in agree-ment all the time, politics would be redundant.

� Politics involves reconciling such differencesthrough discussion and persuasion. Communi-cation is therefore central to politics.

� Political decisions become authoritative policyfor a group, binding members to agreementsthat are implemented by force if necessary.Politics scarcely exists if decisions are reachedsolely by violence, but force, or its threat,underpins the execution of policy.

The necessity of politics arises from the collec-tive character of human life. We live in groups that

Chapter 1

Politics and government

3

must reach collective decisions about sharingresources, about relating to other groups and aboutplanning for the future. A family discussing whereto take its vacation, a country deciding whether togo to war, the world seeking to limit the damagecaused by pollution – all are examples of groupsseeking to reach decisions which affect all theirmembers. As social creatures, politics is part of ourfate: we have no choice but to practise it.

Indeed, the Greek philosopher Aristotle (384–322 BC) argued that ‘man is by nature a politicalanimal’ (1962 edn, p. 28). By this he meant notjust that politics is unavoidable but also that it isthe essential human activity, the feature whichmost sharply separates us from other species. ForAristotle, people can only express their nature asreasoning, virtuous beings through participatingin a political community. Politics is what we arefor.

Of course, members of a group rarely agree, atleast initially, on what plan of action to follow.Even if there is agreement over goals, there maystill be a skirmish over means. Yet a decision mustbe reached, one way or the other, and once madeit will commit all the members of the group. Thuspolitics consists in procedures for allowing a rangeof views to be expressed and then combined intoan overall decision. As Shively (2002, p. 11)points out,

political action may be interpreted as a way towork out rationally the best common solution toa common problem – or at least a way to workout a reasonable common solution. That is, poli-tics consists of public choice.

By debating the options, the quality of the finalchoice should improve as the participants to thediscussion become both better informed and morecommitted to the agreed course of action. In otherwords, good politics yields policy which is bothwell-designed and well-executed.

But the members of a group share some interestsbut not others. A collective decision will typicallyengage both common and conflicting interests.Deciding to expand higher education is one thing;working out who should pay for it is quiteanother. A decision will affect all, and may evenbenefit all, but not everyone will gain equally.

Most often a course of action will produce bothwinners and losers.

Here we arrive at what is the essence of thesubject: politics is about reaching decisions whichimpinge on both the shared and the competinginterests of the group’s members. Indeed someauthors define political situations as those in whichthe participants mix common and competinginterests. ‘Pure conflict is war’, wrote Laver (1983,p. 1). ‘Pure cooperation is true love. Politics is amixture of both.’

So one aim of politics is compromise: to reachan agreement acceptable to all even if the firstchoice of none. Thus, Crick (2000, p. 21) definespolitics as the ‘activity by which differing interestswithin a given unit of rule are conciliated bygiving them a share in power in proportion totheir importance to the welfare and the survival oftheir community’. Crick’s definition is somewhatidealistic; it seems to dismiss the possibility of pol-itics occurring at all in dictatorships. But he issurely right to stress that politics involves (if notexclusively) negotiation, bargaining and compro-mise. And the venue for such discussions is gov-ernment.

Government

Groups must not only reach decisions on theircommon affairs, they must also work out howtheir decisions are to be agreed and implemented.Small groups can often reach agreement byinformal discussion, without needing to developspecial procedures for decision-making. And theiragreements can be self-executing: those who makethe decision put it into practice themselves.However, these simple mechanisms are impracticalfor large groups, which must develop special insti-tutions for making and enforcing collective deci-sions. By definition, these bodies form thegovernment – the arena within which politicalissues are resolved.

Once government reaches a decision, it must beput into effect. In Easton’s famous definition(1965a and b), ‘politics is the authoritative alloca-tion of values’. Values are allocated through imple-menting decisions, not just by making them. Taxesmust be raised as well as set; wars must be fought

4 FOUNDATIONS

and not merely declared. Here we encounter theharder edge of politics. Public authority – ulti-mately, force – is used to implement collectivedecisions. If you break the rules, the governmentmay put you in prison; at any rate, it is the onlybody with the authority to do so. The words ‘poli-tics’ and ‘police’, it is well worth noting, comefrom the same root.

Further, your government has not explicitlyasked you whether you would like to abide by itslaws, pay its taxes or die in its wars. You may haveplayed no part in shaping the laws of your countrybut you are still expected to abide by them. Andeven if you leave one country, you will be subjectto the government of another. From governmentthere is no escape. You cannot – in the contempo-rary world – choose a life without government.

DefinitionA government consists of institutions respon-sible for making collective decisions for society.More narrowly, government refers to the toppolitical level within such institutions.

In popular use, ‘the government’ refers just tothe highest level of political appointments: to pres-idents, prime ministers and cabinet members. Butin a broader sense government consists of all orga-nizations charged with reaching and imple-menting decisions for the community. Thus byour definition public servants, judges and thepolice all form part of the government, eventhough such people are not usually appointed bypolitical methods such as election. In this broadsense, government provides the landscape of insti-tutions within which we experience publicauthority.

Given the special authority of government, whyshould individuals ever agree to cede theirautonomy to such a body? One argument for gov-ernment, much favoured by contemporary econo-mists, is the efficiency gained by establishing astandard way of reaching and enforcing decisions.If every decision had to be preceded by a separateagreement on how to reach and apply it, politicswould be tiresome indeed. Efficiency gains meanthat people who disagree on what should be donecan nonetheless agree on a mechanism forresolving their disagreement.

In addition, government offers the benefits ofsecurity and predictability (Peters, 1999). In awell-governed society, citizens expect laws to bedurable, or at least not to be changed arbitrarily;they know that rules apply to other people as wellas to themselves; and they have grounds forexpecting that decisions will be enforced fairly. Inthese ways, government serves as an escape hatchfrom the gloomy state of nature envisaged by theEnglish philosopher, Thomas Hobbes (1588–1679):

Hereby it is manifest, that during the time menlive without a common power to keep them allin awe, they are in that condition which is calledwar; and such a war, as is of every man, againstevery man.

(Hobbes, 1651, p. 100)

Without government, Hobbes continued, thelife of man is ‘solitary, poor, nasty, brutish andshort’. Only when government overcomes the warof all against all, can society – including industry,science and culture – flourish. Thus governmentcreates a framework of settled order within whichendeavours such as a free market and a welfarestate can emerge.

Once government is established, it may ofcourse have unforeseen consequences. The dangerof Hobbes’s common power is that it will abuse itsown authority, creating more problems than itsolves. As one of Hobbes’s critics pointed out,there is no profit in avoiding the dangers of foxesif the outcome is simply to be devoured by lions(Locke, 1690). This point is one on which the 130million people murdered by their own govern-ment during the lethal twentieth century woulddoubtless agree, were they in a position tocomment.

Further, even when a government does secureinternal peace, it may simply turn its attention toexternal war. The twentieth century was an era ofwarfare states as well as welfare states. Govern-ment, then, is a two-faced, high-risk enterprise,offering the rewards of peace but also the dangerof intensified conflict. Our aim in studying gov-ernment should be to work out how to controlHobbes’ common power while also securing itsbenefits.

POLITICS AND GOVERNMENT 5

Governance

An old word enjoying renewed popularity, gover-nance refers to the activity, process or quality ofgoverning. The term directs our attention awayfrom the institutions and powers of governmenttowards the task of public regulation, a functionwhich government may share with other actors.Because governance is more abstract than govern-ment, denoting an activity rather than an institu-tion, we need here to clarify its meaning and toexplore why the term has become more commonin political analysis.

As Pierre and Peters (2000, p. 7) say, governanceis ‘a notoriously slippery concept’. In essence, theword encourages us to focus on the wide range ofactors involved in regulating modern societies.Depending on the particular sector, these actorsmight include employers, trade unions, the judi-ciary, professional employees, journalists and evenacademics. In areas such as health care or educa-tion, these expert participants form a specialistnetwork which delivers substantial self-regulation.Because professions such as medicine and law donot take kindly to instructions from government,political institutions such as the executive and thelegislature are just particular actors in these net-works and by no means always the commandingplayers; hence the need for the broader term.

Understood as the task of managing complexsocieties, governance involves the coordination ofboth public and private sector bodies; it is theability to get things done without the capacity tocommand that they are done (Rhodes, 1996).Governance implies persuasion exerted through anetwork, rather than direct control over a hierarchy.

The term grew in popularity in the final twodecades of the twentieth century as Westerndemocracies lost some confidence in the ability oftheir governments to directly manage economicproduction and welfare provision. As a result,more emphasis was placed on government as aregulator (e.g. of telecommunications networks)rather than as a provider (e.g. through a state-owned telephone company).

Governance is also the preferred term whenexamining the activity and effectiveness of govern-ment, rather than just the institutions themselves.In this context, governance refers to what govern-

ments do and to how well they do it. For example,many international agencies suggest that ‘effectivegovernance’ is crucial to economic development innew democracies. Thus, the World Bank (1997, p. 1) argued in an influential report that ‘the stateis central to economic and social development, notas a direct provider of growth but as a partner, cat-alyst and facilitator’. The focus here is on govern-ment policies, activities and achievements, not itsinternal organization or its direct provision ofgoods and services.

DefinitionGovernance denotes the activity of making col-lective decisions, a task in which governmentinstitutions may not play a leading, or even any,role. In international relations, for example, noworld government exists to resolve problemsbut many issues are resolved by negotiation – acase of governance without government.

And it is the field of international relationswhich offers the best examples of governance. Thereason is clear: there is no world government, noinstitution making enforceable decisions for theworld as a whole. Even so, many aspects of globalrelations are regulated by agreement. One exampleis the internet, a massive network of linked com-puters beyond the control of any one government.Yet standards for connecting computers and datato the internet are agreed, mainly by privateactors. Thus we can speak of the governance, butnot the government, of cyber-space (Hall andBiersteker, 2002).

Similarly, international institutions haveemerged to formulate rules in many other areas:for instance, the World Trade Organization(WTO) works to reduce trade barriers between itsmember states. However, such bodies are certainlynot governments; they have limited powers, espe-cially in enforcement, and they lack a police forceto enforce their will. Indeed, the WTO reachesdecisions – when it does so at all – by the cumber-some method of consensus.

So the emerging pattern, in international andperhaps also in national politics, is rules withoutrulers, order without orders, governing withoutgovernment. In a word: governance (Rosenau,1992).

6 FOUNDATIONS

The state and sovereignty

The state is now the dominant principle of polit-ical organization on the world’s landmass. Themain exceptions are a few remaining colonies(such as Britain’s Gibraltar) and territories cur-rently administered by the United Nations (suchas Bosnia). In addition, some territories are volun-tarily subject to partial external control – forexample, Puerto Rico is affiliated with the UnitedStates – while others claim substantial autonomywithin a larger state – for instance, Hong Kongand Tibet within China. But leaving such anom-alies to one side, the world is parcelled up intoseparate states which, through mutual recognition,form the international system (Krasner, 2001).

The state is a unique institution, standing aboveall other organizations in society. The state canlegitimately use force to enforce its will and citi-zens must accept its authority as long as they con-tinue to live within its borders. As Edelman(1964, p. 1) writes,

the state benefits and it threatens. Now it is ‘us’and often it is ‘them’. It is an abstraction, but inits name men are jailed, or made rich on defensecontracts, or killed in wars.

The state is more than its government. The termdenotes the ensemble formed by combining gov-ernment, population and territory. It is the statethat claims a monopoly of authorized force, pro-viding a mandate which government then putsinto effect. To bring out the distinction betweenstate and government, note that all countries havesomeone who serves as head of state but that thisperson is not usually head of the government.European monarchs are examples: they symbolizethe state but leave prime ministers to control thelevers of power. In short, the state defines thepolitical community of which government is theexecutive branch.

A central feature of the state is its capacity toregulate the legitimate use of force within itsboundaries. In describing states as ‘bodies ofarmed men’, the Russian revolutionary VladimirLenin (1870–1924) articulated this link betweenthe state and violence. But although the state canand does employ coercion, it is far more than a

mere band of hoodlums. Because the state is basedin a fixed territory, it has a long-term incentive toincrease the wealth of its people and therefore ofitself. To use Olson’s phrases (2000), states are ‘sta-tionary bandits’ rather than ‘roving bandits’ andthey behave better as a result.

DefinitionThe state is a political community formed by aterritorial population which is subject to onegovernment. A country usually refers to a state’sterritory and population, rather than its govern-ment. In international law, a state’s territoryextends to its airspace, continental shelf and ter-ritorial waters.

Most importantly, the state claims not just thecapacity but also the right to employ force. As theGerman sociologist Max Weber (1864-1920)noted, the exclusive feature of the state is its inte-gration of force with authority. As Weber wrote, ‘astate is a human community that (successfully)claims the monopoly of the legitimate use of phys-ical force within a given territory’ (Gerth andMills, 1948, p. 78). Any state must successfullyuphold its claim to regulate the authorized use ofcoercion within its domain. When the state’smonopoly of legitimate force is threatened, as in acivil war, its existence is at stake. While the con-flict continues, there is no legitimate authority.Contrary to Lenin, a body of armed men is not astate because the law of the gun is no law at all.

Much of the theoretical justification for the stateis provided by the idea of sovereignty. Indescribing the state, we must therefore unpack thisrelated notion. As defined by the French philoso-pher Jean Bodin (1529–96), sovereignty refers tothe untrammelled and undivided power to makelaws. Echoing Hobbes, the English jurist WilliamBlackstone (1723–80) argued that ‘there is andmust be in every state a supreme, irresistible,absolute and uncontrolled authority, in which theright of sovereignty resides’.

The word ‘sovereign’ originally meant the oneseated above. So the sovereign body is the oneinstitution within a country which is not subjectto higher authority – and that body is, by defini-tion, the state. As Bodin wrote, the sovereign can‘give laws unto all and every one of the subjects

POLITICS AND GOVERNMENT 7

and receive none from them’. By consecratingcentral authority in this way, the legal concept ofsovereignty contributed powerfully to the develop-ment of the European state.

Sovereignty originally developed in Europe tojustify the attempt by monarchs to consolidatecontrol over unruly kingdoms. Indeed, the Britishmonarch is still known as the ‘sovereign’. But asdemocracy gained ground, so too did the beliefthat elected parliaments acting on behalf of thepeople are the true source of sovereignty. Themeans of acquiring sovereignty evolved althoughthe need for Blackstone’s ‘supreme authority’,wherever located, remained unquestioned inBritain and France.

Beyond Europe, however, the notion of sover-eignty remained weaker. In the federal ‘UnitedStates’, for instance, political authority is sharedbetween the central and state governments, alloperating under a constitution made by ‘we, thepeople’ and enforced by the Supreme Court. Inthese circumstances, the idea of sovereignty isdiluted and so too is the concept of the state itself.

DefinitionSovereignty refers to the ultimate source ofauthority in society.The sovereign is the highestand final decision-maker within a community.Sovereignty is a legal title which is possessed inits entirety or not at all; a state cannot be partlysovereign. Internal sovereignty refers to law-making power within a territory. External sovereignty describes international recognitionof the sovereign’s jurisdiction over its territory.The phrase ‘the sovereign state’ reflects bothdimensions.

Contemporary discussions of sovereignty distin-guish between internal and external aspects. Thelaw-making body within the state possessesinternal sovereignty: the right to make and enforcelaws applying within its territory. By contrast,external sovereignty is the recognition in interna-tional law that a state possesses authority over aterritory. By implication, the state is answerablefor that jurisdiction in international law. Externalsovereignty is important because it allows a stateto claim the right both to regulate affairs within itsboundaries and to participate as an accepted

member of the international system. In this way,the development of the international system hasstrengthened the authority of states in thedomestic sphere; indeed, internal and external sov-ereignty represent two sides of a single coin.

Nations and nationalism

A nation is more easily recognized than defined.We all know that France is a nation but we find itharder to specify what characteristics justify thisdesignation. A single language, for example, isoften taken as evidence of a common nationalityand many countries share a name with their lan-guage, France among them. But in France, as insome other states, nation-builders created a sharedlanguage, rather than the other way round. Afterthe revolution of 1789, the country’s centralizingelite insisted that only French should be taught inschools, thus overwhelming traditional dialects. Acommon language is as much an achievement as acondition of nationhood.

In any case, a single language is not necessary fora shared nationality. Switzerland, for instance, isindisputably a single nation even though French,German and Italian are spoken within its borders.Similarly, of the many countries in which Englishis the leading language, only one comprises theEnglish nation. In truth, seeking to identify anation using any single marker, whether language,history or ethnicity, is fruitless. Nations are imag-

8 FOUNDATIONS

200

150

100

50

0

Figure 1.1 Number of states belonging to the UnitedNations, 1949–2002

Note: Switzerland joined in 2002.Source: United Nations at http://www.un.org/overview/growth.htm.

1949 1959 1969 1979 1989 1997 2000 2002

Number of member states

ined communities and a nation is often viewed asany group which upholds a claim to be regarded assuch (Anderson, 1983).

In two ways, though, we can be a little moreprecise. First, nations are peoples with homelands.As Eley and Suny (1996, p. 10) put it, a nation –like a state – implies ‘a claim on a particular pieceof real estate’. Here the origin of the word ‘nation’is relevant: it derives from a Latin term meaningplace of birth. This link between nation and placeis one factor distinguishing a nation from anethnic group or tribe. Where a tribe can up sticksand move home, a nation remains tethered to itsmotherland, changing shape mainly throughexpansion and contraction.

Second, when a group claims to be a nation, itasserts a right to self-determination within itshomeland. The group uses or invents a sharedculture to justify its claim to political autonomy. Itis this assertion of self-rule that gives the idea of anation its political character. A group united by acommon language, history or ethnicity becomes anation by achieving or seeking control over its owndestiny, whether through independence or devolu-tion. In short, nations ‘are spatially concentratedand culturally distinct groups attempting to attainpolitical self-determination’ (Hechter, 2000, p. 14).

Some examples will clarify these points. Todescribe French-speaking Canadians as a separatenation, as opposed to a linguistic community,indicates a demand for autonomy if not indepen-dence for this group. In a similar way, to refer tothe indigenous people of the Americas as Indiannations, as opposed to tribes, also implies a claimto self-determination. Note that French Canadiansand American Indians are also linked to particularareas, thereby satisfying the territorial condition ofnational status.

DefinitionA nation is defined by Guibernau (1999, p. 14) as‘a human group conscious of forming a commu-nity, sharing a common culture, attached to aclearly demarcated territory, having a commonpast and a common project for the future andclaiming the right to rule itself’.

So much for what a nation is. The next question,far more difficult, is: why do nations exist? When

and why did they come into being? Here debatecentres on whether nations should be seen asancient or modern (Motyl, 2002). On the onehand, nations are sometimes viewed as creatures ofantiquity, emerging from the primeval soup of pasttimes. Smith (1998), for instance, points out thatseveral large ethnic groups (‘nations’?) are indeed ofdistant origin. Examples include the Jews and theChinese. This position is the ‘primordial’ view ofnations, meaning ‘existing from the beginning’.

On the other hand, a modernist approach linksnations to more recent ideas of self-determination.From this perspective, nations are made ratherthan found. They may indeed draw on ancientcultures but their core lies in justifying self-gov-ernment amid a world of states. Nations, after all,assert statehood and states themselves are productsof modernity (Mann, 1995).

In addition, a national identity serves broadermodern functions. It unites people who do notknow each other but who nonetheless find them-selves yoked together under common rulers andmarkets (Gellner, 1983). It was, for example, onlyafter the creation of the USA following thePhiladelphia convention of 1787 that anAmerican identity emerged to supplement loyaltyto the 13 founding states. The American nationfollowed the founding of the central government.

As in white, male America, national loyaltiesglued large populations together under a veneer ofequality. According to Anderson (1983, p. 49),this integration was itself made possible by themodern development of literacy and mass com-munication using a single national language:

The convergence of capitalism and print tech-nology on the diversity of human languagecreated the possibility of a new form of imag-ined community which set the stage for themodern nation.

Certainly, many nations have been constructedin the course of recent struggles. In the nineteenthand especially the twentieth centuries, many colo-nial peoples marched to independence under thebanner of national liberation. Often, thesefreedom-claiming ‘nations’ were as artificial as theboundaries originally imposed by colonial rulers.In a process of nationalism from above, the leaders

POLITICS AND GOVERNMENT 9

of independence movements sought to constructnational identities on extremely insecure founda-tions. Even so, the constructions served theirimmediate purpose: nationalism did indeed proveto be the grave-digger of empires (Tivey, 1981).Dating the precise origin of nations is impossiblebut their starring role surely lies in modern times.

Definition Nationalism, the key ideology of the twentiethcentury, is the doctrine that nations are entitledto self-determination. Gellner (1983, p. 1) writesthat nationalism ‘is primarily a political principle,which holds that the political and national unitsshould be congruent’.The significance of nation-alism is that it offers one answer to a questionbeyond the reach of democracy: who are ‘thepeople’ who are to govern themselves?

Even more than nations themselves, nationalismis a doctrine of modernity. Like some other ‘isms’,nationalism emerged in the nineteenth century toflourish in the twentieth. But unlike some otherideologies, the principle of nationalism is reassur-ingly straightforward. It is simply the doctrine thatnations do have a right to determine their owndestiny – to govern themselves. The British philos-opher John Stuart Mill (1806–73) was an earlyadvocate of this position. He argued that ‘wherethe sentiment of nationality exists in any forcethere is a prima facie case for uniting all themembers of the nationality under the same gov-ernment, and a government to themselves apart’(Mill, 1861, p. 392).

This principle of national self-determinationproved to be highly influential in the twentiethcentury. It provided a justification for redrawingthe map of Europe with the final collapse of theAustro-Hungarian Empire in 1918. The UnitedNations Covenant on Civil and Political Rights(UNHCHR, 1966) offered further support to theprinciple of national self-government:

all peoples have the right to self-determination.By virtue of that right they freely determinetheir political status and pursue their economic,social and cultural rights.

Nations and states

National identities are a potent political force yetthe aspirations of nationalism have not been fullyachieved. Mill’s hope that each nation should formits own state remains unfulfilled. Indeed, tounderstand contemporary politics, we mustexamine the various ways in which nations andstates can combine. Box 1.1 sets out four situa-tions. The first two rows in the box describe typesof state (nation-state, multinational state) whilethe latter two rows denote additional possibilitiesfor national groups (stateless nation, diaspora).

The first and most straightforward category isthe traditional nation-state. Here each countrycontains only the people belonging to its nation.The French Revolution of 1789 established theidea that the state should articulate the interestsand rights of citizenry bound together by a shared

10 FOUNDATIONS

Description Type of Definition Example

Nation-state State A state with its own nation Iceland

Multinational state State A state with more than one nation Scotland and Wales in the United Kingdom

Stateless nation Nation A nation which lacks its own state and Palestinians, the Kurdswhose people are spread across several countries

Diaspora Nation A nation dispersed beyond a home state Jews

B OX 1 . 1

Nations and states

national identity. In today’s world, an example of apure nation-state is Iceland – a small countrywhose population shares such a well-documenteddescent that its birth records provide a perfect lab-oratory for genetic research. Despite being Mill’sfavoured option, the traditional nation-state is lessdominant today. Nearly 3 per cent of the world’spopulation now live outside the country of theirbirth and, in consequence, many states have sig-nificant national minorities (IOM, 2003).

Even though such figures encouraged Anderson(1996) to pronounce the ‘crisis of the hyphen’, thenation-state is probably still an appropriate termwhen discussing the many countries in which onenationality remains dominant, politically andnumerically. In France, Germany and Israel, forexample, the state remains rooted in the soil of astrong national identity despite the presence ofsignificant minorities – Algerians and Moroccansin France, Turks in Germany and Palestinians inIsrael (Smooha, 2002). In essence, these countriesremain nation-states, even if they lack the ethnichomogeneity of Iceland.

The second category in Box 1.1 is the multina-tional state. In this form, more than one nation isfundamental to a country’s politics and assimila-tion to a dominant nationality is not a realisticoption. Britain, for instance, is divided betweenEnglish, Welsh, Scottish and Irish nationals;Canada between English- and French-speakers;and Belgium between Dutch- and French-speakers. As these examples demonstrate, multina-tional states can achieve internal peace andpolitical stability.

But there are, of course, other cases wherenational divisions within a country have led toconflict. For instance, Balkan states such as Bosniaand Croatia experienced vicious conflicts betweenCroat, Muslim and Serbian national groups inthe1990s. A key issue in comparative politics is toidentify the governing arrangements which con-tribute most to political stability in multinationalstates.

One device available here is to separate citizen-ship from nationality. This distinction is subtlebut increasingly important. Citizenship refers tothe rights and obligations flowing from member-ship of a state; it represents a political and legalstatus which can, in principle, be shared by people

with different national identities. In this way, citi-zenship can offer a protective roof under whichdifferent nationalities can live together.

Definition A citizen is a person accorded the legal rightsand duties flowing from membership of a state.In contrast to subjects, citizens are related to thestate as equals and their rights are set out in law.Reflecting the traditions of ancient Greece andRome, citizens possess the right and perhaps theobligation to participate in the affairs of theirrepublic, including waging its wars (Heater,1999).

For example, many immigrants from the Indiansubcontinent who have become citizens of theUnited Kingdom think of themselves as Britishcitizens but as Indian or Pakistani, rather thanEnglish, nationals. In a similar way, citizens of theEuropean Union can now vote in local andEuropean elections when resident in a memberstate other than that of their home nationality(Day and Shaw, 2002).

The ‘nation of nations’ in the United Statesachieves a similar blend through a different route.‘Hyphenated Americans’ – immigrants who retaintheir traditional national loyalty while alsoembracing the citizenship and credo of their newhomeland – are a familiar category in the USA.Such illustrations show how the capacity toacquire citizenship, without imposing a specificnational identity, can be a flexible technique ofintegration within multinational states.

This book centres on states, whether nation-states or multinational in character. But there is adanger in this state-centred approach: namely, thatwe ignore how nations can cut across state bound-aries. Our third category, the stateless nation, refersto just such a situation. Here a nation lacks ahome state of its own. The Kurds, for example,have inhabited a mountainous region of Asia forover 4,000 years and their national identity iswell-established. But the emergence of modernstate frontiers has not proved kind to the Kurds.The national homeland of Kurdistan is nowdivided between Iran, Iraq, Syria and Turkey, withlittle immediate prospect of a Kurdish nation-state.

POLITICS AND GOVERNMENT 11

The Palestinians are an equally tragic instance ofa stateless nation. After the postwar creation ofIsrael, the Palestinians became an uprooted Arabpeople, living mainly in refugee camps not just intraditional areas of Palestine but also in other Arabcountries such as Jordan. However, a sense ofnational identity has developed among thePalestinians, paralleling a growing demand forstatehood. The prospects of their claim succeedingseem greater than for the Kurds (Telhami, 2001).

Our final category is the diaspora. This termrefers to a group which retains its national identitydespite being widely distributed beyond its homecountry. The Jews are the leading case. Since itscreation in 1949, the modern state of Israel hasserved as a Jewish homeland. However, the Jewishdiaspora is widely spread, with only a minority ofthe world’s Jews living in Israel itself. The Chinesediaspora, encompassing economically significantminorities in Malaysia, Thailand and other Asiancountries, is another important example (Luce andSevastopulo, 2003).

In Islamic theory, the Muslim world – known asDar al-Islam – also forms a single entity whichextends far beyond the religion’s Arabian homeland.But Islam is not so much a diaspora centred onSaudi Arabia as a religion which transcends states.This transnational ethos may have been a back-ground factor permitting the mobilization of radicalMuslims from a range of countries into interna-tional terrorist organizations (Mandaville, 2003).

Using modern communications and transport,the members of a diaspora can keep in touch withtheir homeland, illustrating the feasibility of whatAnderson (1998) calls long-distance nationalism.With increased mobility, these cross-border com-munities are growing in importance, diluting thetraditional concept of the nation-state.

Power

Power is the currency of politics. Just as moneypermits the efficient flow of goods and servicesthrough an economy, so power enables collectivedecisions to be made and enforced. Withoutpower, a government would be as useless as a carwithout an engine. Power is the tool that enablesrulers both to serve and to exploit their subjects.

Some authors define politics in terms of power.For Hay (2002, p. 3), politics is ‘concerned withthe distribution, exercise and consequences ofpower’. On such accounts, politics is found notjust in government but also in the workplace, thefamily, the university and indeed in any otherarena in which power is exerted. Such a view isprobably too catholic; those who study politics areprimarily interested in the flow of power in andaround the state. Morgenthau’s definition of poli-tics (1966, p. 63) as concerned with the nature ofpower ‘with special emphasis on the power of thestate’ perhaps comes closer to the mark. But all areagreed that power is central to politics. How thenshould power be defined and measured?

One indicator of power is simply the ability of acommunity to achieve its aspirations. In this sense,describing the United States as a powerful countrysimply means that it has the capacity to achieve itsobjectives. Notice that the emphasis here is onpower to rather than power over – on the capacityto achieve goals, rather than to exercise controlover other countries or people (Box 1.2).

This ‘power to’ approach is associated with theAmerican sociologist Talcott Parsons (1967). Heinterpreted power as the capacity of a governmentto draw on the obligations of its citizens so as toachieve collective purposes such as law and orderand protection of the environment. From this per-spective, political systems resemble energy sources:the more power they deliver, the better. After all,the more powerful the government, the more effec-

12 FOUNDATIONS

Power The production of intended effects

Power to The ability to achieve one’s goals

Power over The ability to overcome opposition

Incentive- Exerting power through the use of shaping inducements and threats

Preference- Exerting influence through shaping persuasion and control of the

climate of opinion

B OX 1 . 2

The language of power

tive it should be at achieving the goals of the com-munity. For Parsons, political power is a collectiveresource rather than an attribute of individuals.

The German-born political theorist HannahArendt (1906–75) made a similar point in definingpower as ‘not just the ability to act but the abilityto act in concert’ (1966, p. 44). A group whosemembers are willing to act together possesses morehorsepower – an enhanced capacity to achieve itsgoals – than does a group dominated by suspicionand conflict. Thus Arendt viewed power and vio-lence as enemies rather than siblings: ‘power andviolence are opposites; where the one rules, theother is absent. Violence can destroy power; it isutterly incapable of creating it’ (1966, p. 56).

This view of power has exerted some influencebut remains incomplete. Power, like politics, has aharder edge. Politics is more than a technical taskof implementing a vision shared by a wholesociety. It is also an arena of conflict over whichgoals to pursue. Politics is substantially a matter ofwhose vision triumphs, a point that must bereflected in any definition of power. From this per-spective, power consists of the ability to get one’sway, to impose one’s opinions, to overcome oppo-sition. The underlying view of power here assumesconflict rather than consensus. In Dahl’s famousdefinition (1957), power is a matter of gettingpeople to do what they would not otherwise havedone. Dahl’s definition is neutral as to means;power is equated with influence, however exerted.

Of course, power and influence can be exertedby a variety of methods. One form is incentive-shaping: that is, altering the incentives con-fronting those subject to power. This can includethe threat of punishment for disobedience or thepromise of reward for acceding to a request. Ineither case, A seeks to alter the context withinwhich B acts rather than B’s overall political views.American presidents, for example, spend a signifi-cant amount of time on such activities. Theythreaten, bribe and cajole overseas leaders, wealthycorporations, leading members of Congress andanyone else who can help them achieve their goals.Indeed, their success in office depends in part onthe skill with which they shape the incentivesfacing other political actors.

But power can also be exerted by shaping prefer-ences rather than incentives. Here A’s effort goes

into shaping what B wants to achieve rather thanthe context within which B behaves. For example,an American president may seek to persuade theSecretary-General of the United Nations that theUSA’s intervention in another country’s affairs ismorally justified. Here power shades into influ-ence, exerted through persuasion by discussionand debate.

On a wider scale, preference-shaping can takingthe form of controlling the overall climate ofopinion within which preferences are formed.Here influence arises from agenda-setting: that is,controlling what issues are addressed and how theyare interpreted. For example, George W. Bushsought to convince his public that the invasion ofIraq in 2003 was a response to a threat toAmerican security; this battle for public opinionproved to be more prolonged and more difficultthan the initial invasion itself. On a smaller scale,parents exert influence over their children byshaping the way they see the world, not just bydirectly giving them incentives to behave in a par-ticular way.

Awareness of this point led Lukes (1974) to con-clude that A exerts power over B when A affects Bin a manner contrary to B’s interests, even if B isunaware of the damage caused. So, the manager ofa nuclear power plant that leaks radioactivity intothe surrounding community has exercised powerover the residents, even if the population isunaware of the contamination. The difficulty hereresides in specifying what a person’s ‘true’ interestsare, a challenging task once we move beyond phys-ical well-being.

Even so, we must accept that controllingpeople’s knowledge and attitudes is the most effi-cient way to control them. As the French philoso-pher Michel Foucault (1977) reminded us,supplying the framework within which an issue isapproached is a potent form of control.

Authority

Authority is a broader notion than power. Wherepower is the capacity to act, authority is the rightto do so. Authority gives the holder the right toexercise power, just as owning property offers theright to decide how that property is used.

POLITICS AND GOVERNMENT 13

Authority exists when subordinates acknowledgethe right of superiors to give orders. So a generalmay exercise power over enemy soldiers but hedoes not have authority over them; this isrestricted to his own forces.

When writers such as Parsons and Arendt arguethat power is a collective resource, they mean thatpower is most effective when converted intoauthority. Yet authority is more than voluntarycompliance. To acknowledge the authority ofrulers does not mean you agree with their deci-sions; it means only that you accept their right tomake decisions and your own duty to obey.Relationships of authority are still hierarchical.

DefinitionAuthority is the right to rule. Strictly, authority isthe right to act, rather than the power to do so.However, authority creates its own power so longas people accept that the authority figure hasthe right to make decisions.

The German sociologist Max Weber (1922) pro-vided a path-breaking analysis of the bases ofauthority. He distinguished three ways of vali-dating political power (see Box 1.3).

The first type is by reference to the sanctity oftradition. This authority is based on ‘piety forwhat actually, allegedly or presumably has alwaysexisted’ (Weber, 1923, p. 296). Traditional rulersdo not need to justify their authority; rather, obe-dience is demanded as part of the natural order.For example, monarchs rule because they alwayshave done so; to demand any further justificationwould itself challenge traditional legitimacy – andwould meet a firm response. Traditional authorityis usually an extension of patriarchy: that is, theauthority of the father or the eldest male. Weberoffers several examples of paternal relationships:

patriarchy means the authority of the father, thehusband, the senior of the house, the eldersibling over the members of the household; therule of the master and patron over thebondsmen, serfs, and freed men; of the lord overthe domestic servants and household officials, ofthe prince over house- and court-officials.

(Weber, 1923, p. 296)

While such illustrations may seem old-fash-ioned, in reality traditional authority remains themodel for many political relationships, especiallyin non-democratic countries. In the Middle East,for example, ‘government has been personal, andboth civil and military bureaucracies have beenlittle more than extensions of the leader’ (Bill andSpringborg, 2000, p. 152). The leader takes careof his followers and so on down the chain. Theserelationships are presented as familial but in prac-tice they are based on inequality: the strong lookafter the weak in exchange for their loyalty. Whenentire political systems operate on the principle oftraditional, patriarchal authority, they are termed‘patrimonial’.

Charismatic authority is Weber’s second form ofauthority. Here leaders are obeyed because theyinspire their followers, who credit their heroeswith exceptional and even supernatural qualities.Where traditional authority is based on the past,charismatic authority spurns history. The charis-matic leader looks forward, convincing followersthat the promised land is within reach. A keypoint here is that, contrary to popular use,charisma is not for Weber an intrinsic quality of aleader. Rather, charisma refers to how followersperceive such figures: as inspirational, heroic andunique. So there is little point in searching for per-sonal qualities that distinguish charismatic fromordinary leaders; rather, the issue is the politicalconditions which bring forth a demand for charis-matic leadership.

14 FOUNDATIONS

Type Basis Illustration

Traditional Custom and the Monarchy established way of doing things

Charismatic Intense commitment Many to the leader and revolutionary his message leaders

Legal– Rules and procedures; Bureaucracyrational the office, not the

person

B OX 1 . 3

Weber’s classification of authority

Generally, charismatic leaders emerge in times ofcrisis and upheaval. Jesus Christ, MahatmaGandhi, Martin Luther King, Adolf Hitler andAyatollah Khomeini are illustrations. Khomeini,for instance, was a Muslim cleric and exiled herowho returned in triumph to take over the govern-ment of Iran following the revolutionary over-throw of the Shah in 1979.

Charismatic authority is short-lived unless it canbe transferred to a permanent office or institution.‘It is the fate of charisma’, wrote Weber (1922, p. 129), ‘to recede with the development of per-manent institutional structures.’ This process iscalled the routinization of charisma. For example,Ayatollah Khomeini succeeded in establishing atheocratic regime (system of government) in Iran,dominated by the Islamic clergy, which outlastedthe Ayatollah’s death in 1989. But as memories ofthe regime’s founder recede, so younger genera-tions increasingly question the political authorityof religious leaders. Authority in the Iranian theoc-racy is not fully routinized.

The third and final base for authority in Weber’sscheme is called legal–rational. Here obedience isowed to principles rather than to people. Theresult is government based on rules, not tradi-tional or charismatic leaders. Legal–rationalauthority inheres in a role or a position, not a spe-cific person.

Indeed, a major virtue of legal–rationalauthority is that it limits the abuse of power.Because it derives from the office rather than theperson, we can speak of officials ‘going beyondtheir authority’. Setting out the extent of an office-holder’s authority reveals its limits and so providesthe opportunity for redress.

In this way, legal–rational authority is a founda-tion of individual rights. Weber believed legal-rational authority was becoming predominant inthe modern world and certainly it has become thedominant form in established democracies. IndeedWeber’s homeland of Germany is the best exampleof a Rechtsstaat, an entire state based on law.

Legitimacy

We must introduce one final concept in thischapter: legitimacy. This notion is a close cousin

of authority but there is a significant difference inthe context in which the terms are used.Legitimacy is normally used in discussing an entiresystem of government, whereas authority oftenrefers to a specific position. When the authority ofa government is widely accepted by those subjectto it, we describe it as legitimate. Thus we speak ofthe authority of an official but the legitimacy of aregime.

Although the word legitimacy comes from theLatin legitimare, meaning to declare lawful, legiti-macy is much more than mere legality. Legality isa technical matter. It denotes whether a rule wasmade correctly – that is, following regular proce-dures. By contrast, legitimacy is a political ques-tion. It refers to whether people accept the validityeither of a specific law or, more generally, of theentire political system.

DefinitionA legitimate system of government is one basedon authority: that is, those subject to its rule rec-ognize its right to make decisions.

Regulations can be legal without being legiti-mate. For example, the majority black populationin white-run South Africa considered the country’sapartheid laws to be illegitimate, even thoughthese regulations were made according to thecountry’s then racist constitution. The same couldbe said of many laws passed by communist states:properly passed and even obeyed but not acceptedas legitimate by the people.

While legality is a topic for lawyers, political sci-entists are more interested in legitimacy: in how aregime gains and sometime loses public faith in itsright to rule. Legitimacy is judged in the court ofpublic opinion, not in a court of law.

Legitimacy is a crucial concept in understandingboth the stability and the effectiveness of govern-ments. In a famous analysis, Lipset (1960, p. 77)argued that ‘legitimacy involves the capacity of thepolitical system to engender and maintain thebelief that the existing political systems are themost appropriate ones for the society’. Howexactly political systems succeed, and sometimesfail, in creating a perception of their own appro-priateness is the subject of Chapter 6.

POLITICS AND GOVERNMENT 15

Key reading

Next step: Shively (2002) is a clear and wide-ranging introduction to politics.

Crick (2000) is a lively and argumentative exami-nation of the nature of politics. On specific topics,governance is covered by R. Rhodes (1996) andPierre and Peters (2000). For the state, Hall andIkenberry (1989) and Poggi (1990) are goodstarting points while van Creveld (1999) traces thestate’s rise and alleged fall. Hinsley (1986),

Hoffman (1998) and Krasner (1999) survey sover-eignty from historical, sociological and interna-tional perspectives respectively. For nations, seeSmith (1998) and Guibernau (1999). Nationalismis examined by Breuilly (1993) and Hechter(2000) while Heater (1999) introduces citizen-ship. See Kennedy and Roudometof (2002) fordiasporas. On power, see Poggi (2001); Haugaard(2002b) is an annotated collection. Watt (1982)introduces authority while Lipset (1960) providesthe classic account of legitimacy.

16 FOUNDATIONS

How the state emerged 17The Western state 19The post-colonial state 21The state and international organizations 25The state in the global economy 30Key reading 34

Although we now take for granted the divisionof the world into states, we should not assume

that the state always was the dominant principle ofpolitical organization, nor that it always will be.There was a world before states and, as advocatesof globalization tirelessly point out, there may be aworld after them too (Guéhenno, 1995).

Before the state, formal government consisted,in the main, of kingdoms, empires and cities.Some of these units were substantial by any stan-dards. For example, the ancient Chinese empire‘proved capable of ruling a population that eventu-ally grew into the hundreds of millions over aperiod of millennia – albeit control was not alwayscomplete and tended to be punctuated by recur-ring periods of rebellion’ (van Creveld, 1999, p.36). Ancient history quickly disposes of the ideathat all modern states are larger and more stablethan every traditional political system.

Nor does the contemporary democratic statehold a patent on self-government by free citizens.Democratic ideas emerged in the classical cities ofAthens and Rome, several centuries before thebirth of Christ, and have merely been recoveredand adapted to modern conditions.

Yet the modern state remains a unique politicalform, distinct from all preceding political systems.Melleuish (2002, p. 335) argues that ‘the develop-ment of the modern state can be compared to theinvention of the alphabet. It only happened oncebut once it had occurred it changed the nature ofhuman existence for ever.’

Today’s states possess the sovereign authority torule a specific territory though their own special-ized institutions. States form an abstract entity

with a special legal status, a notion which contrastswith more personal rule by traditional kings andemperors. The modern idea of the state developedin Europe between the sixteenth and eighteenthcenturies. Indeed, the use of the word ‘state’ as apolitical term only came into common use inEurope towards the end of this period (Dyson,1980, p. 26).

In this chapter, we portray some of the historicaland contemporary forces shaping the state. Ouraim is to present the state not just as an abstractidea but also as a force that has moulded, and is inturn shaped by, the modern world. We begin byreviewing the emergence, growth and partialretreat of the state in its Western heartland. Wethen examine the export of the state from itsEuropean home to the rest of the world throughcolonialism. With this historical context set, weturn to two contemporary challenges facing thestate: international organizations and the worldeconomy.

How the state emerged

The state emerged from the embers of medievalEurope (c.1000–c.1500). In the Middle Ages,European governance had been dominated by twoinstitutions, the Roman Church and feudalism,which together left no room for monarchies todevelop into sovereign states.

The Church formed a powerful transnationalauthority placed above mere monarchs. Kingswithin the Christian commonwealth were consid-ered to be secular agents of the Church’s higherauthority (Figgis, 1960). So strong were theseexternal limits on monarchs that some authorswho believe that global forces are constrainingtoday’s rulers describe this process as ‘the newmedievalism’ (Slaughter, 1997).

Further, within their nominal territories, kingswere further constrained by feudal noblemen who

Chapter 2

The state in a global context

17

exerted extensive authority over men of lowerrank. In this decentralized national setting, theking frequently learned that he needed hisnoblemen rather more than they needed him(Strayer, 1965).

The problem before us, then, is to explain howmodern states emerged from this dual configura-tion of church and feudalism. How did Europeanstates shake off these medieval restraints tobecome the defining political units of the modernworld?

War and reformation

If any single force was responsible for the transi-tion to the modern state, that factor was war. AsTilly (1975, p. 42) writes, ‘war made the state, andthe state made war’. The introduction of gun-powder in the fourteenth century transformedmilitary scale and tactics, as organized infantryand artillery replaced the knight on horseback.The result was an aggressive, competitive andexpensive arms race. Between the fifteenth andeighteenth centuries, military manpower in Franceand England grew almost tenfold (Opello andRosow, 1999, p. 50).

New technology required fresh thinking fromrulers. Kings needed administrators to recruit,train, equip and pay for large standing armies.Reflecting these new benefits of scale, units of ruleincreased in size. The number of independentpolitical units in Europe fell from around 500 in1500 to just 25 by 1800, as a medieval architec-ture of principalities, duchies and bishoprics gaveway to a more recognizable framework of largercountries (Tilly, 1975, p. 24). (Note, however, thattwo major European states, Germany and Italy,did not unify until the second half of the nine-teenth century.)

With the growth of bureaucracy, local patternsof administration and justice became moreuniform. Rulers began to establish formal diplo-matic relations with their counterparts abroad, acore feature of the contemporary state system. Theoutcome was the more centralized monarchieswhich developed in England, France and Spain inthe sixteenth century and which flourished in theseventeenth century. In France, for instance, LouisXIV of France (r.1643–1715) became known as

the Sun King: the monarch around whom therealm revolved.

Just as war-making weakened the feudal pillar ofthe medieval framework, so the Reformationdestroyed its transnational religious foundations.From around 1520, Protestant reformers led byMartin Luther condemned what they saw as thecorruption and privileges of the organized Church.Their reform movement exerted profound politicalconsequences, shattering the Christian common-wealth as war developed between Protestant andCatholic rulers, notably in the Thirty Years’ War(1618–48) in German-speaking Europe.

This conflict was finally ended by the Peace ofWestphalia (1648), an important if occasionallyoverstated chapter in the book of the state(Osiander, 2001). Westphalia is considered pivotalbecause it permitted rulers themselves to regulatethe public exercise of religion within their king-doms, thus rendering secular authority superior toreligious edict. The medieval idea of the Church asa transnational religious authority, superior tosecular rule, fell apart. The threat posed byWestphalia to papal supremacy doubtless explainsthe vigour of Pope Innocent X’s criticisms: ‘null,void, iniquitous, unjust, damnable, reprobate,inane, empty of meaning and effect for all time’(van Creveld, 1999, p. 82).

Sovereignty, contract and consent

As central authority developed in Europe, so didthe need for its theoretical justification. Thecrucial innovation here was sovereignty, as latertamed by the notions of contract and consent. TheFrench philosopher Jean Bodin (1529–96) made acrucial contribution to this centralizing ideology.He argued that within society a single authorityshould possess the untrammelled and undividedpower to make laws. In Bodin’s view, the sovereign– literally, the one seated above – should beresponsible for legislation, war and peace, appoint-ments, judicial appeals and the currency. Suchconcentrated authority is clearly far removed fromthe decentralized medieval framework ofChristendom and feudalism.

The English philosopher Thomas Hobbes(1588–1679) drove the argument forward. Heshared Bodin’s belief in the need for a powerful

18 FOUNDATIONS

sovereign. Without such a body to enforce thepeace, suggested Hobbes, life would be impover-ished indeed. But where Bodin’s sovereign stillderived his authority from God, Hobbes’s analysiswas firmly secular. Hobbes located the sovereign’sauthority in a contract between rational individ-uals seeking protection from each other’s mischief.If the sovereign failed to deliver social order,people would no longer be under an obligation toobey. In this way, the sovereign serves the people,no longer the other way round, and religionbecomes entirely a matter of inner conviction.

So, if Bodin was post-medieval in his thinking,Hobbes represents the first of the moderns. AsSkinner (1978, p. 349) puts it, ‘by the seventeenthcentury, we may be said to enter the modernworld: the modern theory of the state remains tobe constructed, but its foundations are complete’.

The vision of a government made by and for thegoverned was developed by John Locke (1632–1704), an English philosopher whose thinkingshaped one influential interpretation of theWestern state: namely, the liberal vision underpin-ning the American revolution. Locke argued thatcitizens possess natural rights to life, liberty andproperty. These rights must be protected by rulersgoverning through law. Citizens consent to obeythe laws of the land, if only by tacit means such asaccepting the protection which law provides. Butcitizens can withdraw their consent, and institutenew magistrates, should rulers violate the naturalrights of the citizens.

As if to mark the transition from traditionalkingdoms to modern states, Locke further insistedthat political rulers are not just benevolent father-figures caring for a notional household. Rather,political relationships are based on contract andconsent. In Locke’s writing, we see a modernaccount of the liberal state, with society placedbefore government and sovereignty limited bypopular consent.

These ideas of sovereignty, contract and consentwere reflected, in contrasting ways, in the twomost momentous and tangible affirmations ofmodernity: the American and French revolutions.The American Revolution, in which the colonistsestablished their independence from Britain andwent on to create the United States, gave sub-stance to Locke’s liberal interpretation of the state.

Thus, the Declaration of Independence (1776)boldly declared that governments derive ‘their justauthority from the consent of the governed’ whilethe American constitution (drafted 1787)famously begins, ‘We, the people of the UnitedStates’.

However, it was the French Revolution (1789)that made the most ambitious and radical attemptto reinterpret sovereignty in democratic terms.Finer (1997, p. 1516) has no doubt as to the revo-lution’s significance, describing it as ‘the mostimportant single event in the entire history of gov-ernment’.

In essence, the French mapped out the contoursof modern democracy. Where the Americanfederal government remained strictly limited in itsauthority, the French revolutionaries regarded acentralized, unitary state as the sovereign expres-sion of a nation consisting of citizens with equalrights. As national identity joined forces with thestate in this way, so sovereignty – once the deviceused by monarchs to establish their supremacyover popes and princes – was reinterpreted for anew democratic age. These principles wereexpressed in the Declaration of the Rights of Manand the Citizen, a document which served as pre-amble to the constitution of 1791 and whichFiner describes as ‘the blueprint of virtually allmodern states’ (Box 2.1).

The Western state

With the French Revolution, the theoreticalunderpinnings of the Western democratic statewere, in essence, complete. The detailed construc-tion work was completed in the nineteenth andthe first three quarters of the twentieth centuries,supported by growing nationalist sentiment. Onlylate in the twentieth century, and in the twenty-first, did the state begin to contract.

Expansion

Externally, the cage of the state became moreprecise during the nineteenth century, especially inEurope. Borders slowly turned into barriers asprecise maps marked out defined frontiers.Lawyers established that a state’s territory should

THE STATE IN A GLOBAL CONTEXT 19

extend into the sea by the reach of a cannonballand, later, above its land to the flying height of ahot-air balloon (Palan, 2002). Reflecting this newconcern with state boundaries, passports wereintroduced in Europe during the First World War.To travel across frontiers became a rite of passage,involving the formal permission of state officials asexpressed by a stamp in a passport. Such docu-ments remained necessary for overseas travel atleast until some member states of the EuropeanUnion abolished mutual border controls in 1990(Anderson, 1997).

Internally, government functions began toexpand. Many tasks we now take for granted aspublic responsibilities only emerged in the nine-teenth century. These included policing, elemen-tary education, gathering statistics and factoryregulation.

For most of the twentieth century, Westernstates bore ever deeper into their societies (Box2.2). Once again, this expansion was fuelled bythe demands of war. The First (1914–18) andSecond (1939–45) World Wars were total con-flicts, demanding unparalleled mobilization of cit-izens, economies and societies. These campaignswere fought by massive forces equipped with theindustrial weapons of tanks, planes and bombs.Such conflicts were extraordinarily expensive. As aresult, tax revenues as a proportion of nationalproduct almost doubled in Western states between1930 and 1945 (Steinmo, 2003, p. 213). Thetwentieth century was an era of war and thereforea century of the state.

Yet the onset of peace in 1945 did not initiallylead to a corresponding reduction in the role ofthe state. Rather, Western governments sought toapply their enhanced administrative skills todomestic requirements, taking responsibility formanaging the overall economy with the aim ofsecuring full employment. In Europe, the warfarestate gave way to the welfare state, with govern-ments accepting direct responsibility for pro-tecting their citizens from the scourges of illness,unemployment and old age (Flora and Heiden-heimer, 1981). In this way, the state led a postwarsettlement which integrated full employment andpublic welfare with an economy in which theprivate sector continued to play a substantial part.

Contraction

The expansion of the state proved to be expensiveand ultimately unaffordable. By 1975, the propor-tion of national income which West Europeancountries devoted to public social expenditurereached 30 per cent, a tenfold increase comparedto 1900 (Pierson, 1998). With social democraticgovernments redistributing income in the name ofgreater equality, the top rate of income tax inWestern countries reached an inhibiting 63 percent by the mid-1970s (Steinmo, 2003, p. 221).

20 FOUNDATIONS

1. Men are born and remain free and equal in rights.Social distinctions may be based only on consid-erations of the common good.

2. The aim of every political institution is the preser-vation of the natural and imprescriptible rights ofman.These rights are liberty, property, securityand resistance to oppression.

3. The source of all sovereignty lies essentially in theNation. No corporation or individual may exerciseany authority that does not expressly emanatefrom it.

4. Liberty is the capacity to do anything that doesnot harm others. Hence the only limitations onthe individual’s exercise of his natural rights arethose which ensure the enjoyment of these samerights to other members of society.These limitscan be established only by legislation.

5. The law is entitled to forbid only those actionswhich are harmful to society. Nothing not for-bidden by legislation may be prohibited and noone may be compelled to do what the law doesnot ordain.

6. Law is the expression of the general will. All citi-zens have a right to participate in shaping iteither in person, or through their representatives.It must be the same for all, whether it punishes orprotects.

Source: Article 6 above is an extract. For the concept of the general will mentioned in that clause, see Rousseau (1762). For the Declaration’s full text, see Finer (1997, p. 1538) orhttp://www.elysee.fr/ang/instit.

B OX 2 . 1

Declaration of the Rights andDuties of Man and the Citizen(1789), Articles 1–6

As public employment continued to expand, sofinancial pressures mounted. Following the oilcrises of the 1970s, speculation even began toemerge about whether governments might gobankrupt. Rather like the empires of old, nosooner had the Western state reached its full extentthan it began to look overstretched.

In consequence, the final decades of the twen-tieth century witnessed some retreat of the state,particularly in English-speaking countries wherethe conservative agenda linked with RonaldReagan (American president, 1980–88) andMargaret Thatcher (British prime minister,1979–90) gained ground. State-owned industrieswere sold, welfare provision was trimmed and thestate increasingly sought to supply public servicesindirectly, using private contractors. Significantly,military demands were for once consistent with adiminished state: spending on the armed forcesdeclined after the end of the Cold War.

To be sure, the retreat of the state was less pro-nounced in continental Europe than in the Anglo-American world. Even in the English-speakingdemocracies, only time will tell whether the state’srole has evolved – from producer to regulator –rather than declined. But as the violent twentiethcentury approached its unusually peaceful end,

some commentators discerned a fundamental shiftin the approach of the state. Rather than wagingwar and providing welfare, the state began to focuson meeting the challenges of an increasingly openworld economy. Just as medieval kings had pliedtheir restricted trade in the transnational frame-work of Christendom, so modern governments, itwas alleged, had fallen victim to a new master:globalization (Cerny, 1990).

The post-colonial state

The state was born in Europe and then exportedto the rest of the world by colonial powers,notably Britain, France and Spain. As Opello andRosow (1999, p. 161) write, ‘it is impossible tounderstand the development of modern stateswithout taking into account the way Europeanstates constructed an interconnected global orderby means of conquest, trade, religious conversionand diplomacy’. Of all Europe’s exports, the stateis perhaps the most important.

Most states in the world (including, of course,the USA) are former colonies. Countries without ahistory as a colony, leaving aside the ex-colonialpowers themselves, are few and far between: they

THE STATE IN A GLOBAL CONTEXT 21

Aspect Definition Examples

Centralization The centralization of power over a specified territory Law enforcement, border controls

Standardization Greater uniformity within society Common language, standard weights and measures, consistent time zones

Force Strengthened monopoly of legitimate force National police force

Mobilization Increased capacity to extract resources from society Taxation, conscription

Differentiation State institutions and employees are increasingly The idea of ‘public service’distinct from society

Functions Growth in the state’s tasks and its intervention in War-making, welfare provisionsociety

Size Expansion of the state’s budget and personnel Growth of public sector

Source: Adapted from Clarke (1995),Table 1, p. 12.

B OX 2 . 2

Aspects of the growth of the Western state, 1789–1975

include Japan and Thailand in Asia, Ethiopia inAfrica and Iran in the Middle East. In addition, afew former empires – notably Russia and China –have redefined themselves as states to fit thedemands of the current state-based internationalsystem (Oksenberg, 2001).

While the state form may have been successfullyexported from Europe, its substance has rarely fol-lowed. In many post-colonial countries, the statehas been superimposed on traditional ethnic,regional and religious divisions. Often, the statebecomes a prize for which the traditional leaders ofsuch groups compete, resulting in a lack ofautonomy for the state from social interests. Inthese circumstances, the state is more a resource tobe fought over than an actor in the fight.Government institutions are fragmented and thestate as a whole lacks the coherence and drive of itsEuropean forebears. This contrasting role of thestate is the key political contrast between Europeanand post-colonial countries.

How then colonies emerge into statehood?Described by Crawford (2002) as the largest singlechange in world politics over the last five hundredyears, this process took place in four waves spreadover two centuries (Box 2.3). The retreat from

empire by European powers after 1945 was cer-tainly the largest of these waves but by no meansthe only one.

The first wave of decolonization occurred earlyin the nineteenth century, in the Spanish andPortuguese territories of Latin America. Promptedby the American and French revolutions, theseearly wars of independence resulted in 15 new sov-ereign states. But these early battles for self-deter-mination had been initiated by a white economicelite in the colony’s capital. The wealth of this classderived from stringent control of large indigenouspopulations extracting commodities for export.

In Latin America, the outcome of independencewas not a Western-style constitutional state legit-imized by a contract between citizens and subjectto their consent. Rather, the result was continuedeconomic exploitation of native workforces withindeeply unequal societies governed by autocratslacking direct control over the interior (McCreery,2002). These authoritarian traditions remain significant in Latin America today, even thoughmost such countries have now embraced formaldemocracy.

The second and perhaps neglected wave of post-colonial statehood emerged in Europe and the

22 FOUNDATIONS

Wave of Main imperial Main locations of Approximate number Examples of newly decolonization powers colonies and of new states created independent states

imperial territories by decolonization

1810-38 Spain, Portugal Latin America 15 Argentina, Brazil

After the Ottoman, Austro- Europe (beyond 12 Austria, Finland,1914–18 Hungarian and its Western core), Poland,Turkey war Russian empires Middle East

1944–84 UK, France, Mainly Africa, Asia 94 Algeria, Angola,Belgium, Portugal and the Caribbean Congo, India,

Jamaica

1991 Russia Soviet Union 15 Kazakhstan, Latvia,republics (East Ukraine*Europe, Central Asia)

* See Table 4.1 (p. 54) for a full listSources: Adapted from Derbyshire and Derbyshire (1999) and Opello and Rosow (1999).

B OX 2 . 3

States from empires: waves of decolonization

Middle East with the final collapse of the multina-tional and religiously diverse Austro-Hungarian,Ottoman and Russian empires around the end ofthe First World War. The principle of national self-determination, espoused by American PresidentWoodrow Wilson and reflected by nationalist sen-timent within the imperial territories themselves,played a key role in redrawing the map of Europe.

Thus, the Austro-Hungarian Empire dissolvedinto five separate states: Austria, Hungary, Poland,Czechoslovakia and Yugoslavia. Turkey, the his-toric core of the Ottoman Empire, became a sov-ereign state in 1923. Ottoman territories in theMiddle East, including Iraq and Palestine, wereplaced under British or French control undermandates from the new League of Nations. TheRussian Revolution of 1917 provided an opportu-nity for Finland, Estonia, Georgia, Latvia,Lithuania and Ukraine to achieve at least tempo-rary independence.

With the exception of Turkey, however, theoutcome was rarely strong and stable statehood.Most of these new states were multinational incomposition; Czechoslovakia, for example, com-bined Czechs and Slovaks. National identitiesoverwhelmed the limited history of independentstatehood. Eventually, most of these countrieswere incorporated into Hitler’s Germany or theSoviet Union (Batt, 2003). It was only with thecollapse of communism in the 1990s and theentry of many of these post-colonial countries intothe European Union in 2004 that independentstatehood seemed finally to be assured.

The third and largest wave of state creationoccurred after 1945, with the retreat from empireby European states diminished by war. The exem-plar was Indian independence, achieved in 1947;many other colonies, in Africa, Asia, the MiddleEast (including Iraq again) and the Caribbean, fol-lowed suit. This wave of decolonization grew intoa veritable tsunami. Over 90 new independentstates were created between 1944 and 1984,around half the world’s current total (Derbyshireand Derbyshire, 1999). As a result, about one intwo of the world’s countries have existed as inde-pendent states for less than a century.

It is in this group of states that the colonial legacyis most pronounced, with ethnic groups strength-ened by imperial classification battling to control

the resources of the government. Once the contestis won, the dominant group or individual sees thestate as a mine to be exploited. The victor distrib-utes resources to its supporters, often copying thecoercive ruling style of the departing power. Theresult is governance far removed from that found inWestern states (Werbner and Ranger, 1996).

The fourth and final wave of state formationoccurred in the final decade of the twentiethcentury, with the collapse of communism. TheBaltic states, and a dozen soviet satellites in EastEurope such as Hungary, Poland and Romania,achieved effective independence with the dissolu-tion of the communist bloc previously dominatedby the Soviet Union. In addition, the Soviet Unionitself – in effect, a Russian empire – dissolved into15 successor states, including the Ukraine,Uzbekistan and of course Russia itself (Table 4.1,p. 54).

The experience of these new post-communiststates has been mixed. In the Baltic region, statessuch as Lithuania have gained economic and polit-ical stability from their proximity to the EuropeanUnion. However, central Asian republics such asUzbekistan show a more typical post-colonialpattern: small size, ethnic division, pre-industrialeconomies and autocratic rule. In the successorstates to the Soviet Union, these problems are gen-erally reinforced by the absence of any pre-colonialexperience as an independent state (White et al.,2003).

Overall, then, the contrasts between WestEuropean parent states and their post-colonialprogeny are deep-rooted. Post-colonial states rarelypossess the hard edge which their European fore-bears acquired during their own long and violentdevelopment. This contrast can be seen in thetreatment of borders. While European states werekeen to mark off their own frontiers, they inventedborders for their colonies which bore little relationto natural or social features. For instance, almosthalf the boundaries of African states today containat least one straight section and many nationalborders are treated with indifference by govern-ment and people alike. Some are completelyunguarded, hardly the sign of a state concerned todemonstrate its sovereignty over a defined territory.Sovereignty remains important as a title of state-hood, securing international recognition and access

THE STATE IN A GLOBAL CONTEXT 23

24 FOUNDATIONS

Following the Second World War, the planet under-went a sequence of political changes lasting into thefinal decade of the twentieth century. First, with decol-onization, a world of empires gave way to a world ofstates. Second, many of these newly independentcountries experienced military coups as the armyseized power, often with the implicit or explicitsupport of a superpower.Third, the collapse of commu-nism, and thus the end of the Cold War, created moreauthentic independence for former client states of theSoviet Union, leaving the USA as the world’s onlysuperpower. Finally, the retreat of the generals and ofruling communist parties led to a remarkable flow-ering of freedom and democracy in many countries.

But even in the early years of the twenty-first century itwould be simplistic to imagine that the globe now con-sists of free, democratic, sovereign and peaceful states.The Arab and Muslim worlds remain enclaves of eco-nomic stagnation under authoritarian rulers, fuellinghistorically grounded resentment against the West andunderpining terrorism within and beyond the region.

China, the world’s most populous country, remainsunder the direction of a political if no longer commu-nist elite even as it emerges as a global economic force.Many smaller post-colonial countries, particularly inAfrica, combine impoverishment with weak govern-ment control over society, creating conditions forethnic conflict both within and across state boundaries.

Even the countries of the developed world arebecoming caught in a web of connections betweenand beyond states that challenge traditional ideas ofsovereignty. A huge expansion of international com-merce in the second half of the twentieth centuryforced even the wealthiest countries to pay moreattention to international competition and multina-tional corporations. As the Cold War ended, mostcountries formed regional trade associations withtheir neighbours while the oldest states of all devel-oped the European Union, a unique amalgam of inter-governmental and federal principles.

There can be little doubt that such connectionsbetween states will need to deepen if the world is toovercome American indifference to the global envi-ronmental problems, including global warming, whichwill surely move up the political agenda as thetwenty-first century unfolds.

Sources: CIA World Factbook,http://www.cia.gov/cia/publications/factbook; Freedom House,http://www.freedomhouse.org.; Financial Times, 28 November 2003.

Note: Most figures are estimates for 2002 or 2003; some are for earlieryears.

Population: 63 bn (2003 estimate),increasing by about 1.2 per cent peryear.

Population composition: Asian 58 percent; African 12 per cent; European 10per cent; North American 5 per cent.

Life expectancy: 64 years (male 62;female 70).

Literacy: 77 per cent (male 83 per cent;female, 71 per cent).

HIV/Aids infection: 40 million,increasing by about 5 million per year.

Gross world product (GWP): about $49trillion, increasing by 2.7 per cent in2002.

Composition of GWP: agriculture 4 percent, industry 32 per cent, services 64per cent.

Language: Mandarin 17 per cent;English 6 per cent; Hindu 6 per cent;Spanish 6 per cent.

Religion: Christian 33 per cent(including Catholic 17 per cent);Muslim 20 per cent; No religion 15 percent; Hindu 13 per cent.

Unemployment: in the developingworld about 30 per cent unemployedor underemployed; in the developedworld typically 4 per cent–12 per centunemployment.

Inequality: the richest 20 per centreceive about 80 per cent of theworld’s income; the poorest 20 percent receive about 2 per cent of theworld’s income.

Environment: Depletion of minerals,forests, soil, wetlands, amimal andplant species and ozone layer. Globalwarming producing adverse changesto the climate.

Number of states: 1972 – 150; 2002 –192.

Number of democracies: 1988 – 66;1993 – 99; 1998 – 117; 2002 – 121.

Number of free countries: 1988 – 58;1998 – 81; 2002 – 89 (covering 46 percent of the population).

Profile T H E W O R L D

Further reading: Kennedy (1994), Lechner and Boli(2003), Meadows (1994).

to aid, while deterring some invaders (Sørensen,2004, p. 117). But the title’s significance restslargely in its symbolic value.

Internally, too, the rulers of many post-colonialstates – particularly in Africa – find that their pen-etration through their territory is limited. Controlmay not extend far beyond the capital, with gov-ernment outposts falling under the influence oflocal strongmen. The state is not so much acoherent actor as an arena in which groups andindividuals jostle for control over particular min-istries and local offices (Migdal, 2001). Theauthority of political rulers is sometimes subject tofurther competition from criminal gangs such asdrug cartels in Colombia or vigilante groups inAfrica. In view of such difficulties, Englebert(2002) suggests that in Africa territorial restruc-turing may be needed to bring about long-termand sustainable development.

Certainly, post-colonial states possess economicas well as political problems. Most are not justyoung but also small and poor, with unbalancedeconomies overdependent on the export price of asingle commodity. While a few such states possessa rich endowment of natural resources, even thisapparent advantage can turn into a resource curseas politicians compete to control commoditiessuch as oil and minerals so as to obtain resourcesto distribute as patronage, thus unbalancing devel-opment (Luong and Weinthal, 2001).

So the capacity, autonomy and stability of mostpost-colonial governments are more restrictedthan in Western Europe, where states drew on pre-vious traditions of central authority to develop assophisticated taxing, war-making and welfaremachines. The contrasts with Europe are too greatto justify the easy prediction that post-colonialstates will in due course develop along Europeanlines. As far as the state is concerned, Latouche’s(1996) ‘Westernization of the world’ may belimited to form rather than substance.

The state and international organizations

No state is an island entire unto itself. The firstwave of states originated in the requirement toraise men and taxes for foreign wars; later waves

emerged from the end of empires; and contempo-rary states continue to be buffeted by internationaland global winds. So, as Krasner (1999, p. 13)reminds us, ‘globalization is not new, challenges tothe authority of the state are not new, transna-tional flows are not new’.

But the nature and complexity of the externalchallenges confronting states have altered. Wheremedieval rulers in Europe had to contend with theChristian commonwealth, their equivalents todaymust respond to a range of newer forces. Theseinclude:

� regional bodies such as the European Union � intergovernmental organizations such as the

International Monetary Fund � multinational corporations such as Microsoft � international terrorist groups such as al-Qaeda � a massive growth of international trade in

goods and currencies in the second half of thetwentieth century.

In the remainder of this chapter, we examinehow the international and global context impingeson the operation of contemporary states, focusingon two key areas: international organizations andthe global economy. In both areas, we will see howthese transnational forces bear far more heavily onpost-colonial developing countries than on thedeveloped states of the West.

Definition Waters (2000) defines globalization as ‘a processin which the constraints of geography on socialand cultural arrangements recede and in whichpeople become increasingly aware that they arereceding’.

Intergovernmental organizations

The majority of established states belong to mostof the 250 or so intergovernmental organizations(IGOs) which now populate the internationalenvironment (Figure 2.1). These include the mul-tiple agencies of the United Nations system, suchas the International Labour Organization and theWorld Health Organization (WHO). At the veryleast, belonging to so many IGOs complicates the

THE STATE IN A GLOBAL CONTEXT 25

task of governance for states. They must arrange topay their subscriptions, attend meetings, identifytheir national interests, consult with domesticinterest groups, initiate some proposals, respondto others and implement agreements. Such activi-ties dilute the distinction between domestic andforeign policy and give an international dimensionto many, perhaps most, government activities.

What, though, is the broader impact of IGOs?Have they acquired some of the policy-makingrole of national governments or do they justexpress the interests and values of the most pow-erful states?

One perspective on IGOs is that they havebecome important political actors, exerting signifi-cant influence over their member governments,including those in the developed world. After all,setting up an IGO creates a body with its ownemployees and agenda. IGOs may be created bystates – indeed, they may be the state ‘gone global’– but, like children, they grow up to develop theirown interests and perspectives.

For developing countries in particular, the poli-cies of such bodies as the International MonetaryFund impinge directly on domestic policies. Forinstance, when the SARS virus began to spreadfrom China in 2003, intervention by the WorldHealth Organization soon led China to revise itsinitial evasiveness about the illness. An authori-tarian government found it could not adopt the

same indifference to the international communityas it practised with its own people.

DefinitionIntergovernmental organizations (IGOs) arebodies whose members include states. IGOs areestablished by treaty and usually operate byconsent, with a permanent secretariat.Theyinclude single purpose entities such as theInternational Telecommunications Union (estab-lished 1875), regional organizations such as theEuropean Union and universal bodies such asthe United Nations.International non-governmental organiza-tions (usually called NGOs) are private institu-tions with members or groups drawn from morethan one country. Examples are the InternationalRed Cross, Greenpeace and the Catholic Church.

Furthermore, even though most IGOs lack anenforcement mechanism, most states do complywith IGO decisions. The mechanisms of IGOgovernance – conferences, discussions, treaties andpronouncements – are characteristic of modernpolitics; they are an appropriate response to globalproblems by an international community lacking aworld government. IGOs may be less visible thanstates, and they certainly lack the legitimacy ofdirect election, but these points hardly justify theconclusion that IGOs are politically tame. Intimes of peace, the IGO network spins a web ofconstraints around even the strongest states.

There is, though, a more critical reading ofIGOs: namely, that they are mere decorationdesigned to conceal the continued pursuit ofnational self-interest, particularly by the oneremaining superpower. For example, whenPresident George W. Bush failed to achieve asecond UN resolution authorizing military inter-vention in Iraq in 2003, his decision was simple:invade anyway. Operating under the UnitedNations brand, as President Bush did during theGulf War in 1991, would have been useful butwas not deemed essential. Thus, sceptics of IGOinfluence allege that IGOs do not govern states; atmost, dominant states govern through IGOs. As arule, strong states comply with IGO recommenda-tions because they only commit themselves towhat they are already doing (Downs et al., 1996).

26 FOUNDATIONS

Figure 2.1 Number of intergovermental organizations, 1909–97

Source: Data from the Union of International Organizations athttp://www.uia.org/statistics/ using the series for conventional organiza-tions.

Nu

mb

er o

f in

terg

over

nm

enta

l o

rgan

izat

ion

s

400

300

200

100

01900 1925 1950 1975 2000

So a state-centred position is that leading coun-tries remain dominant, only agreeing in IGOs towhat is in their national interest. At most, IGOsare arenas through which states influence eachother, and through which stronger states exertauthority over their weaker brethren.

While sceptics are surely correct to suggest thatstrong states retain most of their traditionalautonomy, it is also the case that IGOs return spe-cific benefits to their members, enhancing theirability to achieve shared goals:

� Membership of universal organizations, espe-cially the UN, confirms to all and sundry thatnational rulers have acquired statehood andsovereignty.

� IGOs help to endorse unpopular policies, pro-viding national governments with both a con-science and a scapegoat: ‘of course, we don’twant to close your steel mill but the EuropeanUnion insists on it’.

� IGOs perform useful functions. Everyone gainsfrom a world telephone network and from safernuclear power.

Whether or not IGOs exert power over states,they have certainly affected the balance of forceswithin them. Specifically, IGOs tend to fragmentnational policy-making. In part this is because a

club-like spirit often develops among ministers in‘their’ IGO. For instance, finance ministers – neverpopular at home – are among friends at meetingsof bodies such as the International MonetaryFund. In this way IGOs lessen the fragile cohesionof national executives and perhaps even contributeto the emergence of a transnational political elite(Sklair, 2003). As a Dutch minister of agriculturesaid about the European Union,

In the Dutch Council of Ministers I met theministers from the departments, and I had todefend the farmers’ interest against other inter-ests, but in the European Council of Ministers[an EU body] I met only other ministers of agri-culture, and we all agreed on the importance ofagriculture.

(Andeweg and Galen, 2002, p. 169)

Given that IGOs tends to fragment nationalpolitical systems, we must ask which governinginstitutions gain, and which lose, from interde-pendence. Among the winners are the executiveand the bureaucracy (Box 2.5). These bodiesprovide the representatives who attend IGO meet-ings and conduct negotiations; they thereforeoccupy pole position. Protective interest groupsalso benefit, since they provide their governmentwith the information and expertise it needs to for-mulate a sensible negotiating position.

The judiciary is also growing in significance as aresult of IGO activity, partly because some influ-ential IGOs such as the World Trade Organizationadopt a highly judicial style, issuing judgements

THE STATE IN A GLOBAL CONTEXT 27

EMU European Monetary Union EU European UnionFTA Free trade associationIGO Intergovernmental organizationIMF International Monetary FundMNC Multinational corporationNAFTA North American Free Trade Association NATO North Atlantic Treaty OrganizationNGO Non-governmental organizationSARS Severe acute respiratory syndromeUN United NationsWHO World Health Organization WTO World Trade Organization

B OX 2 . 4

Some key acronyms in the globalorder

Winners Losers

Executive Legislature

Bureaucracy Parties

Judiciary

Protective interest groups

B OX 2 . 5

The impact of IGOs on nationalpolitics: winners and losers

on the basis of reviewing cases. In addition,national judges are increasingly willing to useinternational agreements to strike down the poli-cies of their home government.

As for losers, the most significant is surely thelegislature, which may only learn of an interna-tional agreement after the government has signedup to it. In some countries, Australia for one,international treaties are an executive preserve,enabling government to bypass the assembly bysigning treaties on proposals opposed by parlia-ment.

Political parties, too, seem to have lost groundunder pressure from IGOs. Like assemblies, theirnatural habitat is the state, not the internationalconference. While party groupings have developedin the EU, these are loose groupings lacking thecohesion and drive of national parties.

Regional organizations

Regional bodies are a specific form of IGO inwhich neighbouring governments join together forcommon purposes, most often trade. These agree-ments have multiplied in number and significancesince the end of the Cold War as countries whichonce sheltered under the skirts of a superpowerhave turned to their neighbours to find a responseto international economic pressures.

So what can a focus on regional organizationsadd to our understanding of the relationshipbetween IGOs and states? Most regional associa-tions are simple free trade areas, with littleprospect of institutional development. Theirpurpose is to secure gains from trade for theirmembers without compromising political sover-eignty. For that reason, their political impact islikely to be unplanned and indirect.

The North American Free Trade Association(NAFTA) is an example. Over a 15-year periodbeginning in 1994, NAFTA seeks to eliminatetrade tariffs between two developed states, theUSA and Canada, and a developing economy,Mexico. Even though the agreement does notcompromise traditional ideas of sovereignty, itsestablishment was controversial. American unionsfeared that free trade with Mexico would cause amigration of jobs to low-cost assembly sites on theMexican side of the border. In a famous phrase,

the independent politician Ross Perot referred tothe ‘great sucking sound’ of jobs being pulleddown to Mexico. When even simple free tradeareas create such debate, attempts at formal polit-ical integration between neighbouring countriesare likely to be impossible in most circumstances.

Even the European Union, undoubtedly themost developed example of regional integration inthe modern world, illustrates the problems ofdeepening integration. The effort to establish acommon currency among some member states,though ultimately successful, is one illustration.German citizens in particular judged that theintroduction of the euro in 2002 provided anopportunity for businesses to put up prices, a per-ception that no doubt reflected traditionalnational pride in the Deutschmark. As always, thespecific, short-term costs of change proved morevisible than the general, long-term benefit.

Just as IGOs are often conceived as an out-growth of state power, so regional integration inparticular flourishes on the foundation providedby strong and stable states. The cornerstone ofregional integration is established democraciescapable of making, implementing and sustainingagreements with other countries. States makeregions at least as much as regions limit states. AsHurrell (1995, p. 354) puts it,

It is no coincidence that the most elaborateexamples of regionalism have occurred in regionswhere state structures remain relatively strongand where the legitimacy of both frontiers andregimes is not widely called into question.

The European Union is a good example: stabledemocracies have come together, pooling somesovereignty, to create a regional body with uniquepowers. The regional organization with the mostelaborate institutions was constructed on the con-tinent which invented statehood.

By contrast, when a country faces severe internaltensions, it is unlikely to be able to build regionalalignments. For example, the post-soviet republicswhich formed the Commonwealth of Indepen-dent States in 1991 were so preoccupied withdomestic difficulties that they had little sparecapacity to develop their association in any mean-ingful way.

28 FOUNDATIONS

Like other IGOs, most regional organizationslack the legitimacy that only direct election canprovide. In contrast to established democracies,they suffer from a ‘democratic deficit’ which limitstheir legitimacy and even their visibility tonational populations (Dahl, 1999). Thus theEuropean Union may prove to be a false model forthe rest of the world. Certainly, it seems unlikely

that the turn to economic regions in today’s worldwill result in political federations of the type estab-lished in the USA in the eighteenth century. AsNAFTA shows, moving towards a free trade zonedoes not require political integration. Regionalfree trade agreements need only limited sup-porting organization.

THE STATE IN A GLOBAL CONTEXT 29

The European Union is the world’s most devel-oped example of regional integration. It repre-sents a deliberate attempt by European politiciansto bring peace to a continent with a long historyof war. From modest beginnings in the 1950s, theUnion has developed its institutions, acquired con-siderable policy-making authority, reducednational barriers to trade, established a new cur-rency and broadened its membership.

1951 Treaty of Paris signed by France, WestGermany, Italy, Belgium, the Netherlandsand Luxembourg.This set up the EuropeanCoal and Steel Commission (ECSC) whichincluded a supranational High Authority.

1957 The ECSC members sign the Treaty ofRome, establishing the EuropeanEconomic Community (EEC) and Euratom.

1965 The Merger Treaty combines the ECSC, EECand Euratom.

1973 Britain, Denmark and Ireland join the EEC.

1979 The European Monetary System (EMS) isagreed, linking currencies to the EuropeanCurrency Unit (ECU). First direct Europe-wide elections to the European Parliamentheld.

1981 Greece joins the EEC.

1986 Spain and Portugal join the EEC. Signing ofthe Single European Act, to streamlinedecision-making and set up a singlemarket by 1992.

1992 Treaty of Maastricht launches provisionsfor Economic and Monetary Union (EMU)and replaces the EEC with the EuropeanUnion (EU) from 1993.

1994 Austria, Finland and Sweden join the EU.

1997 Treaty of Amsterdam agrees to extend theUnion’s role in justice and home affairs andto enhance the authority of the EuropeanParliament.

1999 Launch of European Monetary Union(EMU), irrevocably linking 11 (now 12)national currencies to the euro.The euro islaunched as a virtual currency.

2000 Treaty of Nice agrees on institutionalreforms to prepare for enlargement,including a reallocation of member states’voting power and a reduction in the issuesrequiring unanimity.

2002 National currencies are withdrawn withinthe eurozone.

2003 Cyprus, the Czech Republic, Estonia,Hungary, Latvia, Lithuania, Malta, Poland,Slovakia and Slovenia sign an accessiontreaty to join the EU in 2004.

2003 Publication of a draft constitution forEurope.

Note: Nomenclature of the ‘European Union’ has varied over time,reflecting institutional and constitutional developments. For a map,see p. 173.

50 YEARS OF

T H E E U R O P E A N U N I O N

Non-governmental organizations

As private and unofficial bodies lacking the statusoffered by a membership of governments, wemight expect international non-governmentalorganizations (NGOs) to exert less influence ondomestic politics than IGOs. This conclusion isprobably justified, at least for the more powerfulstates. However NGO impact on post-colonialcountries can be considerable. Leading NGOssuch as the Red Cross and Médecins SansFrontières can apply substantial pressure in coun-tries where the international community is alreadyengaged in supplying services as a result of civilwar or humanitarian crisis (Adams, 2003).

In such conditions, NGOs have become impor-tant actors as executors of IGO, especially UnitedNations, policy. By the mid-1990s, over 10 percent of all public development aid was distributedthrough NGOs, compared to less than 1 per centin 1970 (Weiss and Gordenker, 1996). About 10super-NGOs, including CARE, Save the Childrenand Oxfam, dominate the distribution of aid incomplex emergencies.

In acting as UN subcontractors, NGOs can insome ways substitute for government. Forinstance, NGOs coordinated primary education innorthern Sri Lanka after civil war started there in1987. As a channel for distributing aid and imple-menting associated policy, NGOs possess clearattractions to outside donors: they are more effi-cient and less corrupt than many domestic govern-ments and they are also more sensitive to localpolitical conditions than are military forces.

All this gives NGOs considerable clout in theleast developed countries on which aid is concen-trated. Fernando and Heston (1997, p. 8) go sofar as to suggest that ‘NGO activity presents themost serious challenge to the imperatives of state-hood in the realms of territorial integrity, security,autonomy and revenue’. Furthermore, the numberof NGOs has continued to expand over the last 20years whereas the number of IGOs has fallen(Figure 2.1, p. 26).

The state in the global economy

While international organizations can provide aconcrete challenge to state authority, the global

economy is a more abstract, but no less important,influence on national governments. The ‘globaleconomy’ is a convenient phrase to describe thegrowing trend for economic activity to operatebetween countries (e.g. international trade) andeven beyond them (e.g. currency trading).Particularly in the second half of the twentiethcentury, international trade grew apace while pro-duction and especially finance broke free ofnational restraints. Has the global economy nowescaped state control, reducing political sover-eignty to a redundant fiction?

As with other aspects of the global environment,the impact of an interdependent economy variesbetween the economically developed Westernstates and the less developed, mainly post-colonialcountries. The global economy presents states inthe developed world with a reasonably favourablebalance of opportunities and threats. But the leastdeveloped countries remain in a dependent posi-tion, surviving by exporting basic foodstuffs orminerals in competition with other equally poorstates. We will therefore assess the globaleconomy’s impact on the developed and devel-oping worlds separately, beginning with theaffluent West.

Developed countries

Perhaps the main influence of today’s open tradingworld on developed states in the West has been tomodify their policy agendas. International compe-tition remorselessly exposed economic weaknessesthat could be disguised in an era of nationalmarkets and state-owned monopolies. From the1980s, rulers in many developed states placedgreater emphasis on economic competitiveness(Cerny, 1990). In a process of competitive deregu-lation, public ownership has now been reduced byprivatization; budget deficits have been cut back;corporate and individual tax rates have beenreduced; and opaque share-trading and corporatereporting requirements have been modified in linewith international (in practice, mainly American)standards.

Two key channels through which external eco-nomic forces exert influence in developed statesare multinational corporations (MNCs) andglobal financial flows. Consider MNCs. The

30 FOUNDATIONS

THE STATE IN A GLOBAL CONTEXT 31

‘The time of absolute exclusive sovereignty has passed;its theory was never matched by reality.’ (BoutrosBoutros-Ghali, speaking in 1992 when Secretary-General of the United Nations)

The history of the international system has beenbased on non-intervention in sovereign states.International law, for example, makes little referenceto how states should organize themselves internally(Rich, 2001).This tradition was reinforced during theCold War by the division of much of the world intotwo opposed armed camps. But the collapse of com-munism opened up new possibilities for intervention.So is the international community now justified inintervening in another state’s affairs – by force if nec-essary – to reduce acute human suffering? Is savingstrangers now a practical possibility?

The case forAll governments should reduce human sufferingwhen they are able to do so.Whether that sufferingoccurs inside or beyond a state’s boundaries is irrele-vant; the value of a human life is the same every-where.This view is far from new: in the seventeenthcentury, the Dutch jurist Hugo Grotius (1583–1645)had argued that sovereignty was not the highest law.

Hoffman (1995, p. 35) is among those who has rein-forced this opinion:‘the state that claims sovereigntydeserves respect only as long as it protects the basicrights of its citizens.When it violates them, the state’sclaim to sovereignty falls.’ Oxfam adopts a similarposition:‘we do not accept that the principle of sover-eignty should block the protection of basic rights,including the right to emergency relief and safety’(Harriss, 1995, p. 3). Sovereignty, it is argued, does notentitle a state to decline humanitarian aid, since theresult is added suffering for the people who are thestate’s tangible expression. Also, when the state hasfailed, as several have done in Africa, arguably there isno sovereignty left to violate.

The case againstBy itself, ethics makes poor politics. At issue is not theprinciple of intervention but its likelihood of success.There is simply no world force capable of providingeffective humanitarian intervention. Instead, an inter-

national consensus has to be built behind interven-tion on each occasion and such agreements soon dis-integrate when problems arise in the field.

Somalia is an example (Mayall, 1996, p. 9).When televi-sion images of starvation generated a public demandfor the international community to do something, theUN authorized military intervention to support thedistribution of aid.The project was a disaster. An inad-equate number of troops was despatched in 1992after a long delay to carry out an unfamiliar missionwith insufficient local knowledge in a hostile situa-tion. After a local warlord killed 20 Pakistani soldiersserving with the UN, and a captured American pilotwas paraded in public, the United States decided tobring the boys back home.The entire UN missionfolded in 1995.

Although the American-led invasion of Iraq in 2003lacked clear UN backing, this intervention also con-firmed the dangers inherent in such projects. SaddamHussein was efficiently removed from power but theinvading troops then found themselves immersed inchallenging peacekeeping duties for which they hadnot been trained.

AssessmentIf the principle of intervention is accepted, the nextsteps will be practical.Will powerful states ever permitthe emergence of a global military force? Shouldintervention be subcontracted to regional bodies, aswith NATO’s air-strikes into Bosnia to enforce the 1995Dayton Peace Agreement? In addition, the interna-tional community will need to work out under whatconditions intervention is justified. For example, vanEijk (1997) suggests that any humanitarian interven-tion must pass seven specific tests: (1) human rightsmust be threatened, (2) no alternative solution is avail-able, (3) intervention must receive wide internationalsupport, (4) intervention must be justified to the UN inadvance, (5) the extent of intervention must be mini-mized, (6) intervention must be able to make a con-structive contribution and (7) intervention must lastno longer than is required.

Further reading: Holzgrefe and Keohane (2003), Owen(2002), Wheeler (2002).

DEBATE

STATE SOVEREIGNTY OR HUMANITARIAN INTERVENTION?

balance between MNCs and governments raises,in modern form, the age-old question of the rela-tionship between economic and political power. Inits current form, the logic governing this relation-ship is simple: capital is mobile, labour less so andstates not at all. Companies can move – or justthreaten to move – their factories and technologybetween countries. But states, by definition, areanchored to their territory.

However, compared to the vulnerable states ofthe developing world, the relationship betweenWestern governments and MNCs is balanced.MNCs must sell their products in the affluent andsophisticated markets of the developed world.Aware of their attractions, host governments canstrike a deal. For instance, they may require localfactories of the MNC to buy a certain share oftheir resources from domestic suppliers. In addi-tion, states can join together in an alliance such asthe European Union to create a larger marketwhich gives the member states, acting together,more bargaining power with MNCs. As withmany relationships, each side needs the other.

International financial flows are the secondmajor way in which markets constrain policy-making in developed countries. Governmentsfund their debt by borrowing, often from overseas.The less confidence financial markets possess in acountry’s government or economy, the higher therate of interest they will demand for lending tothat country. This relationship is sustainedthrough rating agencies, such as Standard andPoor’s, which evaluate the creditworthiness ofcountries as well as companies.

In similar fashion, a country’s currency willdecline in value if dealers in foreign exchange loseconfidence in its government. Because the weightof money in the financial market is far greaterthan that in the vaults of central banks, centralbanks are unable to stem such losses. The outcomemay be an unwanted devaluation for those govern-ments that seek to maintain a fixed exchange rate.

A final effect of economic globalization on thedeveloped world has been to increase economicinequality within borders, both at individual andregional level. Unskilled workers in developedcountries face downward pressure on incomesfrom the low-wage economies of the developingworld, notably China. However, firms are willing

to pay more to recruit and retain managers –whatever their nationality – who can increase cor-porate profitability. Partly as a result of theseforces, economic inequality within Britain at theend of the twentieth century was greater than atthe start.

Globalization also increases inequality acrossregions within states. Core regions possess thelocation and skills needed to prosper fromincreases in trade. In particular, financial centressuch as the City of London and Wall Street marchto the drum of the global economy, partlydelinking from their national homeland (Sassen,2002). But peripheral regions, rather like individ-uals lacking relevant skills, find themselves trappedin a cycle of decline. Eventually, we may find thatthe categories ‘rich’ and ‘poor’ apply less to coun-tries than to sectors and regions within them, pro-ducing massive new problems of politicalmanagement for national governments.

Developing countries

Most developing countries find themselves in amore exposed position in relation to external eco-nomic forces. For better or worse, the impact ofthe global economy is far greater than on devel-oped states. Partly, this weakness arises from thedependence of many developing countries on asingle product: for example, Nigeria’s exportsconsist overwhelmingly of oil. Alternatively, adeveloping country may depend on a singlemarket for its exports: for instance, most ofMexico’s exports go to the USA.

Many commentators suggest that through suchconcentrated trade patterns the economies of theleast developed countries are locked into depen-dence on rich countries. The result is unbalanceddevelopment. Dependency theorists argue there-fore that what has emerged is not a neutral globaleconomy but rather a new form of economic colo-nialism in which the developed world shapes thestructure of client economies in the developingworld. This is not development but rather whatFrank (1969, p. 16) called ‘the development ofunderdevelopment’.

The IGOs which have emerged to regulate theglobal economy according to the liberal principleof free trade are particularly influential in vulner-

32 FOUNDATIONS

able developing economies (Box 2.6). The abilityof IGOs to grant or withhold loans gives them anoversight role over the least solvent states whichamounts to veto power. A crucial test of the skillof the leaders of such countries is their ability tonegotiate not just with these economic IGOs butincreasingly with sources of private capital fromoverseas.

Consider, by way of example, the impact of theWorld Trade Organization (WTO). Its function isto provide a way of resolving trade conflictsbetween member states and to seek further reduc-tions in tariffs and other barriers to trade. In prac-tice, this means that member states gainpreferential access to each other’s markets. Withmost countries requiring access to lucrative NorthAmerican and European markets, membership ofWTO is highly desirable. But in order to join, asRussia wants to do, states must meet specific con-ditions. In essence, the price to governments ofentering the global economy is to open theirdomestic market. Thus the WTO has leverage notjust over its 146 members but also over the 50 orso states which have yet to join.

The weaker the finances of a developingcountry, the more leverage international agenciescan exert. IGO impact has been particularlystrong in those countries most affected by the1980s debt crisis, a problem which arose whenWestern banks increased lending to developingcountries which then failed to meet repayments.Many of the least developed states remain inmassive debt. As late as 1995, the developingworld as a whole owed $1.5 trillion; interest pay-ments stood at $232 billion per year. Like anyother debtor, the least developed states wereunable to bargain about the terms of any furtherfinancial aid.

Bodies such as the World Bank and the IMF aregenerally willing to provide additional support butthis debt restructuring is conditional on domesticeconomic reform, often including privatization,lower trade tariffs and more transparency in allo-cating government contracts. For example, theinternational community (led by the IMF andJapan) offered a $16 billion loan to the Thai gov-ernment in 1997. The conditions attached to theloan were extensive and specific, including:

� halving the country’s trade deficit � balancing the budget� speeding up privatization� allowing foreign ownership of banks � floating the currency.

DefinitionConditionality is the practice of attachingstrings to aid. Political conditionality mightinclude a requirement to legalize opposition andhold competitive elections. Economic condition-ality often takes the form of a specified struc-tural adjustment programme, typicallyinvolving privatizing state companies, intro-ducing more competition and opening domesticmarkets to overseas companies.

Governments which build domestic politicalsupport by using the economy to reward theirsupporters find the injunction to reduce theircontrol striking at the heart of their power. Ineffect, the international agencies require suchcountries to restructure their politics as well as

THE STATE IN A GLOBAL CONTEXT 33

IGO Full title Function

IMF International To promoteMonetary Fund international monetary

stability and cooperation

IBRD International To promote economic Bank for recovery and develop-Reconstruction ment and Development (‘World Bank’)

WTO World Trade To supervise and Organization promote international

trade

Note: The WTO was founded in 1995 to replace GATT (GeneralAgreement on Tariffs and Trade) which, like the IMF and the WorldBank, had been established at Bretton Woods (Jackson, 1998).

B OX 2 . 6

Intergovernmental organizationsand the world economy

their economy. Mahathir bin Mohamad,Malaysia’s Prime Minister from 1981 to 2003, wasskilled at exploiting this dependence for domesticpolitical purposes. He likened the internationalpolitical economy to ‘a jungle of ferocious beasts’,claiming that the IMF was willing ‘to subvert oureconomy’ in order to prove the validity of its owntheories.

Attacks by developing countries on such bodiesas the WTO have not gone unsupported bygroups in the developed world. To date, the mostprominent example of a backlash was the ‘battle inSeattle’ in 1999, when a WTO conference wasmarked by violent protests outside the conferencehall. Over 5,000 anti-globalization protestorsmarched through the streets and over 500 werearrested as police in riot gear fired tear-gas andplastic bullets. The activists included unionmembers, environmentalists, students and indige-nous people. The conference itself ended in acri-mony, with many developing countries objectingto what was seen as an American attempt toimpose its own agenda.

Seattle demonstrates that, even more than theEU, IGOs such as the WTO suffer from a democ-ratic deficit which limits their scope. Four yearsafter Seattle, a two-pronged attack by developingcountries and NGOs was able to exploit thisweakness, leading to the failure of WTO tradetalks. In Cancún, Mexico, in 2003, the USA andespecially the EU came under attack for contin-uing to subsidize their own farmers, thus discrimi-nating against agricultural exports from thedeveloping world. Since WTO agreements requireconsensus, delegates from poorer states were ableto prevent an agreement. Cancún provides tenta-tive evidence of a modification in the balance ofpower between IGOs and developing countries.

But of course it is not just IGOs which imposeconditionalities on vulnerable countries. Since theend of the Cold War, individual states – includingestablished democracies such as the UK – havesubjected bilateral aid to specific conditions.Donor governments which turned a blind eye tothe internal politics of developing countries arenow willing to attach conditions to aid, strings

which may take the form of pressure for politicalchange.

For example, sanctions may be imposed oncountries with poor human rights records (such asChina) or corrupt governments (Kenya). Stateswhich have invaded other countries (Iran), whichsponsor international terrorism (Syria), which areruled by undemocratic means (Libya) or which arestill under communist control (North Korea,Cuba) find themselves isolated from, or entirelywithout influence within, the international system.

Thus dominant countries can directly affect thepolitical situation confronting national leaders inpoorly regarded developing states. The worst casesbecome pariah states, kept out of the internationalcommunity altogether.

Key reading

Next step: Opello and Rusow (1999) is a wide-ranging but accessible historical introduction tothe state.

Van Creveld (1999) and Pierson (2004) also exam-ines the rise of the state but the most monumentalhistory of government, sparkling with insight, isFiner (1997). Turning to the contemporary positionof the state, both Sørensen (2004) and Paul (2004)examine its alleged transformation in a global era.On globalization itself, Waters (2000) offers a clearintroduction while Lechner and Boli (2003) is awide-ranging reader. On international organiza-tions, Diehl (2001) is a useful introduction, whileBoli and Thomas (1999) examine non-govern-mental organizations specifically. Holzgrefe andKeohane (2003) cover the many dilemmas ofhumanitarian intervention while Chatterjee andScheid (2003) focus on the ethical dimension. Onregionalism, Breslin et al. (2002) blends theory withcases; see also Hettne (1999). On internationalpolitical economy, Stubbs and Underhill (1999) is ahelpful edited collection. Strange (1994) and themore sceptical Hirst and Thompson (1999), haveeach generated considerable debate on theautonomy of states in a global economy.

34 FOUNDATIONS

Direct democracy 36 Representative and liberal democracy 38Waves of democratization 39New democracies 43Semi-democracy 46Key reading 49

We live in an era of democracy; for the firsttime in history, most people in the world

live under tolerably democratic rule. This upsurgein democracy reflects the transformation of theworld’s political landscape in the final quarter ofthe twentieth century. Over this short period, thenumber of democracies more than doubled.Democracy expanded beyond its core of WesternEurope and former settler colonies to embraceSouthern Europe (for example Spain), EasternEurope (for example Hungary), Latin America (forexample Brazil), more of Asia (for exampleTaiwan) and parts of Africa (for example SouthAfrica).

This shift to democracy, while important initself, will also have international ramifications. Itis likely to contribute to peace and prosperity sincedemocracies rarely go to war with each other andare more likely to form trade agreements than arenon-democracies (Huth and Allee, 2003). The ter-rorists who attacked the United States onSeptember 11, 2001, we should note, originatedfrom authoritarian rather than democratic coun-tries.

As democracy continues to spread, so it becomesmore varied (Box 3.1). Understanding the formstaken by democracy in today’s world is therefore acentral task for comparative politics. In thischapter, we examine the established democraciesof Europe and its settler colonies, with theiremphasis on representative and limited govern-ment. We then discuss the newer democraciesemerging from the ashes of communist and mili-tary rule. Finally, we assess those awkward semi-democratic regimes – Russia is an example – that

straddle the border between democratic andauthoritarian rule. But to begin we must explorethe origins of democracy itself. And that task musttake us back to the fifth century BC to the world’smost influential example of self-government:ancient Athens.

Chapter 3

Democracy

Form Definition

Direct The citizens themselves assemble democracy to debate and decide on collective

issues

Representative Citizens elect politicians to reach democracy collective decisions on their behalf,

with the governing parties held toaccount at the next election.

Liberal The scope of democracy is limited democracy by constitutional protection of indi-

vidual rights, including freedom ofassembly, property, religion andspeech

New A democracy in which an authori-democracy tarian legacy continues to influence

political action and debate.Democracy is not the only game intown

Established A consolidated democracy which democracy provides an accepted framework for

political competition.The outcomeof free elections is accepted by thelosers as well as the winners

Semi- An illiberal democracy in which democracy elected presidents do not respect

individual rights, or in which electedgovernments form a façade behindwhich traditional rulers continue toexercise effective power

B OX 3 . 1

Forms of democracy

35

Direct democracy

Democracy is a form of government offering aworkable solution to the fundamental politicalproblem of reaching collective decisions bypeaceful means. But it is also an ideal and an aspi-ration. So we cannot understand democracysimply by looking at contemporary examples.Judged against the democratic ideal, even the mostsecure ‘democracies’ are found wanting. Indeed,the tension between high ideals and prosaic realityhas itself become part of the democratic condition(Dahl, 2000).

So what, then, is the core principle of democ-racy? The essential idea is self-rule: the word itselfcomes from the Greek demokratia, meaning rule(kratos) by the people (demos). Thus democracy inits literal and richest sense refers not to the elec-tion of the rulers by the ruled but to the denial ofany separation between the two. The modeldemocracy is a form of self-government in whichall adult citizens participate in shaping collectivedecisions in an environment of equality and opendeliberation. In a direct democracy, state andsociety become one.

The birthplace of democracy is ancient Athens.Between 461 and 322 BC, Athens was the leadingpolis (city-community, often translated as city-state) in ancient Greece. Poleis were small indepen-dent political systems, typically containing anurban core and a rural hinterland. Athens, one ofthe larger examples, held only about 40,000 citi-zens. Especially in the earlier and more radicaldecades of the period, the Athenian polis operatedon the democratic principle summarized byAristotle as ‘each to rule and be ruled in turn’ (seeBox 3.2). This principle applied across all theinstitutions of government within the city-com-munity. All citizens could attend meetings of theassembly, serve on the governing council and siton citizens’ juries. Because ancient Athens con-tinues to provide the archetypal example of directdemocracy, we will look at its operation in moredetail (Figure 3.1).

History has judged there to be no more potentsymbol of direct democracy than the Ekklesia(People’s Assembly) at Athens. Any citizen aged atleast 20 could attend assembly sessions and thereaddress his peers; meetings were of citizens, not

their representatives. The assembly met around 40times a year to settle issues put before it, includingthe recurring issues of war and peace which werecentral to the prospects and prosperity of the polis.In Aristotle’s phrase, the assembly was ‘supremeover all causes’ (1962 edn, p. 237); it was the sov-ereign body, unconstrained by a formal constitu-tion or even, in the early decades, by written laws.

But the assembly did not exhaust the avenues ofparticipation in the Athenian democracy.Administrative functions were the responsibility ofan executive council consisting of 500 citizensaged over 30, chosen by lot to serve for a one-yearperiod. Through this device of rotation, thecouncil exemplified the principle of direct democ-racy: government by, and not just for, the citizens.Hansen (1991, p. 249) suggests that about one inthree citizens could expect to serve on the councilat some stage in their life, an astonishing feat ofself-government entirely without counterpart inmodern democracies.

A highly political legal system provided the finalleg of Athens’s complex democracy. Juries ofseveral hundred people, again selected randomlyfrom a panel of volunteers, considered lawsuitswhich citizens could – and frequently did – bring

36 FOUNDATIONS

� All to rule over each and each in his turn over all.� Appointment to all offices, except those requiring

experience and skill, by lot.� No property qualification for office-holding, or

only a very low one.� Tenure of office should be brief and no man

should hold the same office twice (except militarypositions).

� Juries selected from all citizens should judge allmajor causes.

� The assembly should be supreme over all causes.� Those attending the assembly and serving as

jurors and magistrates should be paid for theirservices.

Source: Aristotle, The Politics, Book VI.

B OX 3 . 2

Aristotle’s characterization ofdemocracy

against those considered to have acted against thetrue interests of the polis. The courts functioned asan arena of accountability through which topfigures (including generals) could be brought tobook.

Thus the scope of the Athenian democracy wasextraordinarily wide, providing an all-encom-passing political framework within which citizenswere expected to develop their true qualities. Forthe Athenians, politics was intrinsically anamateur activity, to be undertaken by all citizensto develop both themselves and the broader com-munity.

Of course, we should not blind ourselves toserious flaws in Athens’s little democracy:

� Citizenship was restricted to a small elite: it wasa birthright of males whose parents were them-selves citizens. Most adults – including women,slaves and foreign residents – were excluded.Women played no significant public role andcritics allege that slavery was the platformwhich allowed citizens time to devote to publicaffairs (Finley, 1985).

� Participation was not in practice as extensive asthe Athenians liked to claim. Most citizens wereabsent from most assembly meetings even afterthe introduction of a payment for attendance.

� Athenian democracy was hardly an exercise in

lean government. A modern management con-sultant would conclude that the system was atime-consuming, expensive and over-complexmethod of governing a small society. Its applic-ability to a modern world in which people arecommitted to paid work, and the affluenceresulting therefrom, is questionable.

� The principle of self-government did notalways lead to decisive and coherent policy.Indeed, the lack of a permanent bureaucracyeventually contributed to a period of ineffectivegovernment, leading to the fall of the Athenianrepublic after defeat in war.

Perhaps Athenian democracy was a dead-end inthat it could only function on an intimate scalewhich precluded expansion and proved inherentlyvulnerable to predators. As Finer (1997, p. 368)observes, ‘the polis was doomed politically if itexpanded and doomed to conquest if it did not. Ithad to succumb and it did.’ Yet for over 100 years,the Athenian democratic experiment survived andprospered. It provided a settled formula for ruleand enabled Athens to build a leading position inthe complex politics of the Greek world. Athensproves that direct democracy is, in some condi-tions, an achievable goal.

Certainly, Finer (1997, p. 371) was correct inacknowledging the Athenian contribution to

DEMOCRACY 37

Figure 3.1 The direct democracy of ancient Athens

Assembly People’s CourtCouncil

Adult male citizens

Women, slaves, non-residentsand children

selection by lotfor a one-yearperiod

right toattendand speak

Note: Citizenship was a birthright which could not normally be acquired by other means.

selection by lot(for a specificcase) from apanel of volunteers

set agenda for

Western politics: ‘the Greeks invented two of themost potent political features of our present age:they invented the very idea of citizen – as opposedto subject – and they invented democracy’.

Representative and liberal democracy

The contrasts between the classical democracy ofancient Athens and the modern democracies oftoday’s world are clear. Most obviously, citizenshipis no longer an elite status but has been extended tothe vast majority of the adult population. But twoother contrasts are equally important.

First, today’s democracies are representativerather than direct. The democratic principle hastransmuted from self-government to elected gov-ernment. This transformation can be seen in thecontrasting treatment of elections offered byancients and moderns. The Greeks viewed elec-tions as an instrument of aristocracy: a device forselecting qualified people for technical tasks whichrequired an unfortunate departure from self-gov-ernment. However, as the phrase ‘representativedemocracy’ shows, the modern world regards elec-tions as an expression rather than a denial ofdemocracy.

Second, modern democracy is based on a liberalphilosophy in which the state’s scope is restrictedby the constitution. This limit is based on a dis-tinction between public and private that wouldhave been unacceptable in classical Athens wherecitizens who lived an entirely private life were dis-missed as idiotes. Today’s democracies are liberaldemocracies and it is the constitution as much asthe legislature that is ‘supreme over all causes’.

In this section, we examine how these modernconcepts of representation and liberalism weregrafted on to the original democratic idea so as todeliver a new hybrid. The requirement for thisnew form was clear. In contrast to the littledemocracy of Athens, any modern version ofdemocracy must be compatible with the muchlarger states found in today’s world.

In reinventing democracy for the modern era,the key breakthrough was to modify traditionalideas of representation. In itself, the idea of leadersrepresenting their community in a wider assemblywas nothing new. In Europe, for example,

medieval monarchs had summoned leaders of thevarious estates of the realm – lords, commonersand representatives of the cities – to help themwith their tasks of raising revenues and fightingwars. But unlike the Athenian assembly, themembers of these royal advisory assemblies weresummoned or self-appointed, not elected. Theywere not elected by those they represented, norwould they have deigned to have been so.

Indeed, representation was still viewed as a desir-able brake on democracy. Thus James Madison, anarchitect of the American constitution, judged thatrepresentation served to ‘refine and enlarge thepublic views, by passing them through the mediumof a chosen body of citizens, whose wisdom may bestdiscern the true interest of their country’ (Madison,1787, p. 45). At this stage, then, representation wasstill a device for limiting ‘pure democracy’.

But in the nineteenth century, stimulated by theFrench Revolution of 1789 and by the diffusion ofpower brought about by mass literacy and indus-trialization, the notion of turning assemblies intorepresentative bodies elected from a wide franchiserapidly gained ground. One of the first authors tograft representation on to democracy was theBritish-born pamphleteer and international revo-lutionary Tom Paine. In his Rights of Man (firstpublished in 1791 or 1792), Paine wrote:

The original simple democracy . . . is incapableof extension, not from its principle, but fromthe inconvenience of its form. Simple democ-racy was society governing itself without the aidof secondary means. By ingrafting representa-tion upon democracy, we arrive at a system ofgovernment capable of embracing and confeder-ating all the various interests and every extent ofterritory and population.

(Paine, 1984 edn, p. 180)

Scalability has certainly proved to be a key strengthof representative democracy. The conventionalwisdom in ancient Athens was that the upper limitfor a republic was the number of people who couldgather together to hear a speaker. However,modern representative government allows massivepopulations (such as 1.05 billion Indians and 290million Americans) to exercise some popularcontrol over their rulers. And there is no upper

38 FOUNDATIONS

limit. In theory, the entire world could become onegiant representative system. Adapting Tom Paine’sphrase, representative government has proved to bea highly convenient form.

As ever, intellectuals were on hand to secure thetransition of representative democracy from aninherent contradiction to a workable system of rule.Prominent among them was Joseph Schumpeter(1883–1965), an Austrian-born economist whobecame an academic in the United States.

Schumpeter (1943, p. 269) conceived of democ-racy as nothing more than party competition:‘democracy means only that the people have theopportunity of refusing or accepting the men whoare to rule them’. He wanted to limit the contribu-tion of ordinary voters because of his jaded view oftheir political capacity: ‘the typical citizen dropsdown to a lower level of mental performance assoon as he enters the political field. He argues andanalyzes in a way that he would recognize as infan-tile within the sphere of his real interests. Hebecomes a primitive again.’

Reflecting this jaundiced view of the public,Schumpeter argued that elections should not beconstrued as a device through which voters electrepresentatives to carry out their will; rather, therole of elections is simply to produce a govern-ment. From this perspective, the elector becomes apolitical accessory, restricted to selecting frombroad packages of policies and leaders prepared byrival parties. Representative democracy is merely away of deciding who shall decide:

The deciding of issues by the electorate [is made]secondary to the election of the men who are todo the deciding. To put it differently, we nowtake the view that the role of the people is toproduce a government . . . And we define thedemocratic method as that institutional arrange-ment for arriving at political decisions in whichindividuals acquire the power to decide by meansof a competitive struggle for the people’s vote.

(Schumpeter, 1943, p. 270)

These ideas represent a considerable thinning ofthe democratic ideal envisaged in classical Athens.But that is not all. The second distinctive featureof modern democracy, its liberal character, con-tributes a further qualification to strict rule by the

people. Like representative democracy, liberaldemocracy is a compromise: it seeks to integratethe authority of democratic governments withlimits on the scope of their action.

The central feature of liberal democracy islimited government. The goal is to secure indi-vidual freedom, including freedom from unwar-ranted demands by government itself. The objectis defensive: to protect the population from itsrulers and minorities from the danger of majoritytyranny (Held, 1996). Liberal democracy is a set-tlement between individual liberty and collectiveorganization which reflects the key issues involvedin its emergence. These issues include the desire toentrench religious freedom, to protect against therecurrence of tyranny and to secure the rights ofproperty against the mob. All these elements werecentral to the design of the American system ofgovernment, the most liberal (and perhaps theleast democratic) of all the democracies.

In place of the all-encompassing scope of theAthenian polis, liberal democracies are govern-ments of laws rather than men. Even elected rulersare subject to constitutions that almost alwaysinclude a statement of individual rights. In theory,citizens can use domestic and international courtsto uphold their rights when the governmentbecomes overbearing. In this way, a liberal demo-cracy is democracy disarmed.

Both the representative and liberal elements ofmodern democracy dilute the original principle ofself-rule. We find in contemporary democracies aform of rule in which decision-making is theresponsibility of governments rather than the gov-erned and in which the public sphere is limited byprotecting the rights of citizens in general and ofproperty-owners in particular. The watering downis considerable but the outcome is a flexible andscalable political system which is coming to domi-nate the world.

Waves of democratization

How then were these principles of representativeand liberal democracy implemented in the transi-tion to democracy? When and how did modernestablished democracies emerge? As with thephases of decolonization discussed in the last

DEMOCRACY 39

chapter, so too did democracies emerge in a seriesof distinct waves (Box 3.3). As defined byHuntington (1991, p. 15),

A wave of democratization is a group of transi-tions from nondemocratic to democraticregimes that occur within a specified period oftime and that significantly outnumber transi-tions in the opposite direction during thatperiod . . . Three waves of democratization haveoccurred in the modern world.

The first modern democracies emerged in the‘first long wave of democratization’ between 1828and 1926. During this first wave nearly 30 coun-tries established at least minimally democraticnational institutions, including Argentina,Australia, Britain, Canada, France, Germany, theNetherlands, New Zealand, the Scandinaviancountries and the United States. Some of thesefledgling democracies were later overthrown byfascist, communist or military dictatorships duringHuntington’s ‘first reverse wave’ from 1922 to1942.

However, democracy did consolidate in the earliest nineteenth-century democratizations, in-cluding the United States and the UnitedKingdom. We will examine these two transitionsof the first wave in more detail, not least becausethe USA remains the leading example of liberaldemocracy while Britain usefully illustrates repre-sentative democracy.

The emergence of democracy in the UnitedStates was rapid but it was a transition nonethe-less. The founders had thought of political leader-ship in non-democratic terms, as the duty of adisinterested, leisured gentry. However, the ideathat citizens could only be represented fairly bythose of their own sort quickly gained ground,supported by the egalitarian spirit of a frontiersociety. The suffrage quickly extended to nearly allwhite males. But some groups had to wait untilthe twentieth century for the full franchise.Women were not offered the vote on the sameterms as men until 1919 and the black franchisewas not fully realised until the Voting Rights Actof 1965 (Dahl, 2001).

Today, the USA gives us the clearest picture of aliberal democracy in which limited government isentrenched by design. The Founding Fatherswanted, above all, to prevent dictatorship, in-cluding tyranny by the majority. To prevent anygovernment – and especially elected ones – fromacquiring too much power, the constitution set upan elaborate system of checks and balancesbetween the institutions of government (Figure3.2). Because power is so fragmented, the dangerof any particular faction manipulating publicauthority for private ends is much reduced. Powerchecks power to the point where it is often diffi-cult for the government to achieve even neededreforms. The constitution placed governmentunder law before government by all the people. Inthis way, the liberal dimension of America’s

40 FOUNDATIONS

Wave Period Examples

First 1828–1926 Britain, France, USA

Second 1943–62 India, Israel, Japan, West Germany

Third 1974–91 Southern and Eastern Europe, Latin America, parts of Africa

Note:The first wave was partly reversed between 1922 and 1942 (for example, in Austria, Germany and Italy) and the second wave similarlybetween 1958 and 1975 (for example, in much of Latin America and post-colonial Africa). A return to authoritarian rule after a democratic inter-lude is termed backsliding.

Source: Huntington (1991). For some criticisms of the wave approach, see Grugel (2002), pp. 32–7.

B OX 3 . 3

Huntington’s three waves of democratization

democracy emerged victorious over its representa-tive aspect. Only during periods of external threat,including post-9/11, do individual liberties comeunder threat (Lyon, 2003).

In Britain, by contrast, the outcome of thedemocratic transition was a less liberal but morerepresentative form of government. By the eigh-teenth century, the power of the monarch hadbeen checked by the authority of parliament.However, the rights of the individual citizen werenever stated as clearly as in the USA. Thewidening of the suffrage also occurred more gradually in the United Kingdom, with each step easing the fears of the propertied classes about the dangers of further reform (Table 3.1).As the House of Commons acquired democraticlegitimacy, so both the monarchy and the non-elected House of Lords retreated into the back-ground. As in the United States, theimplementation of democratic procedures inBritain continued well into the twentieth century

but the battle of principle was fought and won inthe nineteenth.

Yet if the USA emphasizes liberal democracy,Britain gives priority to its representative element.Where American democracy diffuses power acrossinstitutions, British democracy emphasizes thesovereignty of parliament. Representation operatesthrough parties that retain tight control over theirown members of the House of Commons. Asingle governing party wields extensive powersuntil the voters offer their verdict at the next elec-tion. Except for the government’s sense of self-restraint, the institutions that limit executivepower in the United States – including a codifiedconstitution, the separation of powers and feder-alism – are absent. Instead the electoral rules nor-mally ensure a secure majority of seats to thewinning party. In reality, the hallowed sovereigntyof parliament is leased to the party in office.

Far more than the United States, Britain exem-plifies Schumpeter’s model of representative

DEMOCRACY 41

Figure 3.2 Liberal democracy: the separation of powers in the United States

Constitution of the United States

The Federal Government Fifty States of the Union

Judicial Branch

Supreme Court ofthe United States

Congress approves appoint-ments, controls budget, canpass laws over president’sveto, impeach and remove

president from office

President canveto legislation

Executive Branch

The President

Executive Office ofthe President,

Cabinet,Departments,

Independent agen-cies

Judicial Branch

State SupremeCourt

Legislative Branch

State Assembly

Executive Branch

Governor

Court candeclare laws

unconstitutional

Presidentappointsjudges

Legislative Branch

The Congress– House– Senate

Senate con-firms judicial

appointments;can impeachand removejudges from

office

Court candeclare

presidentialactions

unconstitutional

divides power between

Each state has itsown constitution

democracy as an electoral competition betweenorganised parties. ‘We are the masters now’, trumpeted a Labour MP after his party’s triumphin the British election of 1945; similar thoughtsmust have occurred to many Labour MPs aftertheir party’s equally emphatic victory in 1997.From a comparative perspective, a governing party in Britain is still given an exceptionally freehand.

Despite these contrasts between Britain and theUSA, both countries are of course examples ofconsolidated democracies emerging during thefirst nineteenth-century wave of democratization.Huntington’s second wave of democratizationbegan in the Second World War and continueduntil the 1960s. Like the first wave, some of thenew democracies created at this time did not con-solidate. For example, elected rulers in severalLatin American states were quickly overthrown bymilitary coups. But established democracies didemerge after 1945 from the ashes of defeated dic-tatorships, not just in West Germany but also inAustria, Japan and Italy. These postwar democra-cies were introduced by the victorious allies, led bythe USA and usually acting with the support ofdomestic partners. Yet despite their partly imposedcharacter, these second-wave democracies didestablish firm roots, helped by an economicrecovery itself nourished by American aid. Duringthis postwar wave, democracy also consolidated inthe new state of Israel and the former BritishDominion of India.

Second-wave democracies differed in characterfrom their predecessors. Their liberal traditionswere somewhat weaker as representation throughparties proved to be the stronger suit. Parties hadgone unmentioned in the American constitutionbut by the time of the second wave they hademerged as the leading democratic instrument.Indeed, Germany’s Basic Law (1949) codifies theirrole: ‘the political parties shall take part in formingthe democratic will of the people’. In severalsecond-wave democracies, the importance of party was confirmed by the emergence of a singleparty which dominated national politics for a gen-eration: Congress in India, the ChristianDemocrats in Italy, the LDP in Japan and Labourin Israel.

The third wave of democratization, finally,

began in 1974 and continued until 1991. Its mainand highly diverse elements were:

� the end of right-wing dictatorships in SouthernEurope (Greece, Portugal and Spain) in the1970s

� the retreat of the generals in much of LatinAmerica in the 1980s

� the collapse of communism in the Soviet Unionand Eastern Europe at the end of the 1980s.

This third wave has transformed the global polit-ical landscape: the predominance of democraticforms in today’s world itself places added pressureon those non-democratic regimes that survive.

Within the third wave, it is only the SouthernEuropean group that provides consistently securecases of democratic consolidation, aided by mem-bership of the European Union and economicdevelopment. Elsewhere, in Eastern Europe andLatin America, many third-wave democracies havenot yet fully consolidated, if indeed they are everto do so at all. The category of new democracy –as also of semi-democracy – remains central to

42 FOUNDATIONS

Table 3.1 The British electorate as a percentageof the adult population, 1831–1931

Year Electorate(per cent of population aged 20+)

1831 4.4

1832 First Reform Act

1832 7.1

1864 9.0

1867 Second Reform Act

1868 16.4

1883 18.0

1884 Third Reform Act

1886 28.5

1914 30.0

1918 Vote extended to women over 30

1921 74.0

1928 Equal Franchise Act

1931 97.0

Note: In 1969, the voting age was reduced from 21 to 18.

Source: Adapted from Dahl (1998), fig. 2.

understanding these uncertain regimes. We willexplore this theme further by examining in moredetail the political and economic challenges facingthese new democracies.

New democracies

Just as the first wave of modern democracies repre-sented a severe dilution of the Athenian vision ofself-rule, so too many new democracies of the thirdwave are developing a further compromise withtheir own authoritarian histories. Certainly, thedistinctions between most new democracies andthe early modern democracies such as the UnitedStates and the United Kingdom remain important.In this section, we review the challenges facingthese new additions to the democratic family.

To be sure, many new democracies do seem tohave consolidated by one crucial test: a peacefultransfer of power through elections. For example,the South Korean presidential election of 1997witnessed the first peaceful transfer of power tothe centre-left in that country’s history. Similarly,Herbst (2001) notes that by 1999 a dozen Africanstates had experienced at least one change of gov-ernment through the ballot box. Peaceful transfershave also become almost routine in EasternEurope and parts of Latin America.

Yet even when elections have succeeded in thedelicate task of replacing a governing elite, mostnew democracies remain distinctive; the questionis not whether they will consolidate but whatexactly they are consolidating into. The difficultiesfacing new democracies can be grouped into twoclusters: the political problems associated with anilliberal inheritance and the economic problemscaused by the combination of limited develop-ment and extreme inequality.

The political challenge

Consider first the political challenges facing newdemocracies of the third wave. Reflecting anauthoritarian legacy, liberal ideas often remainweak. As Luckham and White (1996b, p. 7) pointout, the development of democracy requires morethan just competitive elections. It also requires theenforcement of legal restraint on state power, pro-

tection of civil rights, the establishment of rela-tively uncorrupt and effective bureaucracies, andthe imposition of democratic control over poten-tially authoritarian forces such as the military andthe security services.

DefinitionA democracy has consolidated when it providesan accepted framework for political competition.As President Havel noted in Czechoslovakia aftercommunism’s collapse, democratic consolidationrequires more than creating appropriate institu-tions:‘we have done away with totalitarianismbut we have yet to win democracy’.The standarddefinition comes from Przeworski (1991, p. 26):

Democracy is consolidated when under givenpolitical and economic conditions a particularsystem of institutions becomes the only gamein town and when no-one can imagine actingoutside the democratic institutions.

To the extent that democratic consolidationrequires substituting a government of laws for oneof men, the task is still incomplete in new democ-racies. The inheritance from the old regime con-tinues to limit progress. After all, rulingcommunist parties and military councils hadbrooked no interference from the judiciary andpaid no heed to constitutions, including state-ments of human rights. The agencies of repression– the military, the intelligence services and thepolice – were as strong as the mechanisms of rep-resentation were weak. However well-intentionedthe new rulers may be, constructing a liberaldemocracy from an authoritarian history is agreater challenge than the blank canvas facing theframers of the United States constitution,designing a new state where none of any signifi-cance had previously existed.

Take, for example, the post-military democraciesof Latin America. Here the generals still possessconsiderable prestige deriving from their historicalrole as providers of order to unstable societies.This status is sometimes reflected in a guaranteedbudget, seats in the legislature and virtual exemp-tion from civil law. Even in the civil arena, justicein much of Latin America remains underdevel-oped. Lower courts are often inefficient and

DEMOCRACY 43

corrupt and many cases do not arrive therebecause the police are themselves corrupt orbecause ordinary people regard the courtroom asthe prerogative of the wealthy. In urban slums, theconcept of individual rights does not apply, nomatter what fine words appear in the constitution.

In post-communist states, too, any national tra-ditions of rule by law were dulled by the totali-tarian experience. Ruling communist parties wereabove the law and public officials continue toregard their position as an opportunity to obtainprivate advantage. The combination of an inade-quate legal framework with systematic evasion andinadequate enforcement of those laws that do existis a difficult base from which to consolidate liberaldemocracy.

In Africa, the problems differ. Far fromachieving communist levels of penetration thoughsociety, the limited incision of the African stateinto its territory limits the impact of a transitionfrom authoritarian to democratic rule. For Africanpeasants, as for Latin American slum-dwellers,regime transitions in the capital city must seemremote indeed.

So in many new democracies, the tradition ofpower revolving around individuals – whethercommunist party bosses, the arrogant generals andlandowners of Latin America or the ‘big men’ ofAfrican politics – tends to subvert attempts toconsolidate the democratic framework.

The economic challenge

The economic difficulties confronting new democ-racies of the third wave are even more obviousthan the political challenges. These problemsconsist of a combination of poverty and inequality,exacerbated by severe economic decline in the earlyyears of the new democracy. Even in many of thelarger and more developed new democracies, livingstandards remain well below Western levels. In theUSA, gross domestic product (GDP) per head hadreached $36,300 by 2002; in the new democracyof neighbouring Mexico, the equivalent figure wasjust $9,000 and 40 per cent of the populationlived in poverty. European contrasts are equallystark: German GDP per head is almost double thefigure for post-communist countries such as theCzech Republic and Hungary.

Relative poverty goes hand in hand with greaterinequality. Many new democracies retain a largeagricultural sector, where sharp contrasts continuebetween a rich, powerful elite and a poorly edu-cated, and often powerless, population. Conflictsbetween landowners and dependent peasants areendemic in much of Latin America, for instance.As Vanhanen (1997) notes, such conditions favourneither the diffusion of power resources nor thedevelopment of mutual toleration and compro-mise which foster democratic consolidation.

Further, the ex-communist states in EasternEurope suffered enormous economic dislocationin the transition from the old order. As plannedeconomies began to be dismantled in tandem withdemocratization, unemployment soared. Through-out the post-communist world, the 1990s was adecade of deep economic decline in which the realsuffering of the many was exacerbated by theostentatious affluence of a few. Only in theopening years of the twenty-first century did eco-nomic growth return to most post-communistdemocracies, with Central and Eastern Europe as awhole now growing at a faster rate, albeit from alower base, than in the Western part of the conti-nent.

Lower levels of affluence in new democracies areimportant partly because a long research traditionclaims that economic well-being is the key todemocratic consolidation. In Political Man (1960,pp. 48–9), Lipset famously concluded that ‘themore well-to-do a nation, the greater the chancesthat it will sustain democracy’. Lipset demon-strated that stable democracies scored highest onsuch measures as income per person, literacy andthe proportion of the population living in cities.Following Aristotle, Lipset believed that a largemiddle class opposed to extremism was conduciveto democracy.

More recent research confirms the correlationbetween affluence and stable democracy, eventhough there are exceptions such as poor butdemocratic India. Marks and Diamond (1992, p. 110) seem to be fully justified in describing theconnection between affluence and democracy as‘one of the most powerful and stable relationshipsin the study of comparative national development’.

Crucially, the economic and political weaknessesof new democracies are linked. The absence of a

44 FOUNDATIONS

DEMOCRACY 45

‘Yes, it can be done,’ shouted the crowds in MexicoCity as they celebrated the downfall of the PRI afterthe presidential election of 2000. After 70 years inpower, the PRI not only lost a presidential election forthe first time but also no longer controlled eitherhouse of the national legislature.The world’s oldestruling party had suffered an historic reverse, defeatedby a centre-right coalition led by Vicente Fox.Thispeaceful transfer of power decisively confirmedMexico’s status as a new democracy. For students ofcomparative politics, Mexico offers a remarkably suc-cessful example of democratization.

The PRI had been founded in 1929 in the decade fol-lowing the radical Mexican revolution. Gradually,however, socialist principles were diluted as the PRIestablished a classic semi-democracy based onpatronage networks.The PRI distributed favours whilerepressing opposition and manipulating electionresults. In the 1950s and 1960s, the PRI seemed tohave discovered the perfect recipe for a stable dicta-torship. However, three problems recurred:

� continuing poverty for those excluded from the PRInetwork, reflected in periodic revolts

� increasing opposition from the expanding urbanmiddle class created by economic growth

� occasional economic crises when the PRI placed itspolitical objectives before sound economic policy.

With the political effectiveness of the PRI machinedecaying, Carlos Salinas (President, 1988–94) initiatedeconomic reforms, including privatizing major firmsand opening the economy to international competi-tion, not least through NAFTA. In contrast to the SovietUnion, where Gorbachev had initiated political reform

before restructuring the economy, liberalization inMexico preceded political change. As the PRI lostdirect control of economic resources, so its powers ofpatronage declined and voters became free tosupport opposition parties, especially in the cities.Independent trade unions began to form outside theenveloping embrace of the PRI.

But the PRI also introduced political changes thatserved to enliven a moribund opposition. By the1990s, the PRI no longer felt able to manipulate elec-tion results. In 1997, it lost its majority in the Chamberof Deputies after relatively fair elections.The decisiveelection in 2000 was overseen by an independentelection commission. So the PRI’s fall was partly self-induced: its leaders recognized that the tools neededto guarantee their party’s continued grip on powerwere hindering the country’s further development.

Mexico’s gradual moves to democracy seem to haveavoided what Baer (1993, p. 64) described as ‘thedilemma of all reforms from above, particularly inageing regimes: how to avoid unleashing a revolutionfrom below’. But it remains to be seen how far, and atwhat speed, democracy will consolidate in Mexico.ThePRI remains a significant force, controlling half thecountry’s 32 states. Mexico’s continuing problems –peasant revolts and urban squalor, drugs and crime,corrupt judges and incompetent police – mean that itremains premature to place the country in the samepolitical category as the USA and Canada, its NAFTApartners.

Population: 105m.Gross domestic product per head:

$9,000. Mexico is the world’s ninthlargest economy.

Form of government: a federal andpresidential republic.

Legislature: the 500 members of theChamber of Deputies are elected for asingle three-year period and the 128members of the Senate for a six-yeartenure.

Executive: the president, directly elected

for a non-renewable six-year term,heads both the state and the govern-ment, choosing the members of theCabinet.

Judiciary: headed by the Supreme Courtof Justice, the judicial system mixesAmerican constitutional principleswith the civil law tradition. In practice,both judicial independence and policeenforcement of law have been weak.

Electoral system: 300 members of theChamber represent single-member

districts, the other 200 are elected bythe list system of proportional repre-sentation.The Senate also operates amixed electoral system.

Party system: dominated by theInstitutional Revolutionary Party (PRI)until the 1990s. Other major partiesare the conservative National ActionParty (PAN), which formed part of Fox’sAlliance for Change in the 2000 elec-tions, and the left-wing RevolutionaryDemocratic Party (PRD).

Profile M E X I CO

Further reading: Camp (2002), Cornelius and Weldon(2004), Craske (2001), Levy and Bruhn (2001).

liberal political framework itself inhibits economicadvance. Weak legal systems restrict economicdevelopment because corporations lack confidencethat commercial disputes can be resolved fairlyand promptly through the courts. Close personalconnections develop between politicians in needof money and well-placed business executives whovalue political influence. These semi-corrupt net-works preclude the development of a clear frame-work for market regulation. Dominant firms withpolitical protection can see off competitive threats,preventing the development of a level playing fieldin which the most efficient companies canprosper. Scared off by corrupt and slow-movingbureaucrats, foreign investors are inclined to goelsewhere, especially as population and market sizeare relatively small in new democracies. As a result,both economic and democratic developmentwither, held back by the incomplete penetration ofliberal ideas and institutions.

The challenge of timing

We should mention one final factor affecting theconsolidation of third-wave democracies: thetiming of their transition. To be born into a worldwhich is already democratic is a mixed blessing.On the one hand, it increases the pressures on new democracies to consolidate too quickly.Populations value not just democracy but theaffluence they sense goes with it; and they demandboth now. On the other hand, an internationalenvironment supportive of democracy – and evenmore so since 9/11 – is beneficial to democraticconsolidation. To understand new democracies, itis helpful to explore the challenge of timing inmore detail.

The first wave democracies were not so muchadopting a new political order as inventing it. Aswe saw in discussing the United Kingdom, inno-vation was a leisurely, even evolutionary, process ofadapting old ideas to large states. By contrast,third-wave democracies were delivered into aworld where democratic ideas were alreadybecoming predominant. As a result, they areexpected to mimic established examples withoutthe economic resources and gradual maturationwhich helped the countries of the first wave. Bothdomestic and international audiences expect the

process of developing democracy to be collapsedinto a decade or two. The result is rushed ratherthan leisurely democratization. In the first wave,democracy could be an outcome but in the thirdwave it has to be an intention. As Hollifield andJillson (2000, p. 11) suggest,

The latest transitions to democracy haveoccurred with dizzying speed, giving the soci-eties involved little time to prepare for the leapto representative government. Whereas democ-racies in Western Europe, the United States andthe former British dominions had a gestationperiod of one or two centuries, in the third wavedemocratization has come virtually overnight.This has led to a great deal of improvisation andmany setbacks.

At the same time, democracies of the third wavehave one clear advantage over their predecessors: afavourable global and regional context. Leadingactors such as the United States and the EuropeanUnion, and sympathetic institutions such as theWorld Bank, began to promote democracy oncethe Cold War ended. This support began evenbefore September 11 gave the USA a furtherreason to promote democracy in the authoritarianMiddle East.

Often, a favourable regional context also easedtransition. Greece, Portugal and Spain – and morerecently post-communist Poland, Hungary andthe Czech Republic – undoubtedly benefited fromtheir position close to the heartland of Europeandemocracy. In a similar way, Mexico’s transitionfrom semi-democracy surely owes something to itstrading links with the USA, consolidated throughNAFTA. Indeed Diamond (1997, p. 39) suggeststhat ‘the greatest regional force for democraticconsolidation in the Americas may well be themove towards regional free trade’.

Semi-democracy

The final concept to explore in this chapter is thatof semi-democracy. This term lacks the theoreticalpurity of either democratic or authoritarian rule;its task is more descriptive. Semi-democracydenotes forms of government which, in practice,

46 FOUNDATIONS

blend both democratic and authoritarian ele-ments. In a semi-democracy, democratic legiti-macy is not wholly lacking; rather, it is acquiredand exploited in dubious ways and often remainscontested. As in Asian states such as Singapore andMalaysia, the dominant leaders use ethnic diver-sity, fear of political instability or the demand foreconomic development as reasons for departingfrom the liberal aspects of established democracies.

In a world dominated by democratic ideology,semi-democracy is a more likely outcome than areturn to authoritarianism for those new democra-cies that do not consolidate. This hybrid is notnew but it is becoming more prevalent. Carothers(2002, p. 18) observes that ‘what is often thoughtof as an uneasy, precarious middle ground betweendemocracy and dictatorship is actually the mostcommon political condition in countries of thedeveloping and postcommunist worlds’.

DefinitionA semi-democracy blends democratic andauthoritarian elements in stable combination.Although rulers are elected, they govern withlittle respect for individual rights and oftenharass opposition or even non-official groups.By contrast, a new democracy is one that hasnot yet had time to consolidate; that is, democ-racy has not become the ‘only game in town’. Inpractice, new democracies and semi-democra-cies show similar characteristics but a newdemocracy is transitional while a semi-democ-racy is not. Assuming a new democracy does notslide back into authoritarian rule, it will developinto either an established democracy or a semi-democracy.

The crucial point is that we should not think ofdemocracy and authoritarianism as sole and incom-patible ways of organizing government. Rather,each principle can provide pockets of power thatcan coexist, sometimes indefinitely, within the onepolitical system; once set, semi-democracy is astrong amalgam. Crouch (1996, p. vii), forinstance, shows how Malaysia’s ‘repressive–respon-sive’ regime combines democratic and authoritarianfeatures in a manner that ‘provides the foundationfor a remarkably stable political order’.

Similarly, Borón (1998, p. 43) refers to the

‘faulty democracies’ of Latin America in whichrulers, once elected, govern in an authoritarianstyle, showing little concern for mass poverty orlegal niceties. In these conditions, suggests Borón,democracy ‘endures but does not consolidate’.

The notion of semi-democracy also captures thepolitical reality of many states in sub-SaharanAfrica. As Herbst (2001, p. 359) writes, ‘it iswrong to conclude that African states are travel-ling between democracy and authoritarianismsimply because a majority of them belong toneither category. Rather, the current condition ofAfrican states could well prevail for decades’.

In understanding the operating methods ofsemi-democracies, it is useful to distinguish twovariants. In the first type, an elected party or leadersets the framework for political competition, gov-erning in an illiberal fashion. O’Donnell (1994, p.59) describes this format: ‘whoever wins electionto the presidency is thereby entitled to govern ashe or she sees fit, constrained only by the hardfacts of existing power relations and by a constitu-tionally limited term of office.’

Russia’s super-presidential system is an example.The president not only takes the lead in seeking toimpose solutions to national problems but moresignificantly is expected to do so. Boris Yeltsin(President 1991–2000) ruled in a highly personalway which inhibited the development of govern-ment institutions. Yeltsin’s successor, VladimirPutin, is an equally tough political operator. Inmany African countries, too, ‘presidents have aninordinate amount of power invested in them andlittle in the way of institutional provisions tocheck that power’ (May, 2000, p. 176).

One way in which these democratic despotsacquire at least some legitimacy is through semi-competitive elections. They use control overmoney, jobs, contracts, pensions, public housing,the media, the police, the electoral system and thecourts to deliver success, usually without any needto manipulate the election count. Egypt andTunisia are examples of countries where electionshave long been semi-competitive.

Note, however, that such methods are oftencombined with effective governance and afavourable disposition towards a dominant ruler orparty. For example, Singapore’s People’s ActionParty may manipulate elections in its favour but it

DEMOCRACY 47

48 FOUNDATIONS

Many Asian leaders reject aspects of the Westerndemocratic tradition.They claim to be building a dis-tinctive form of ‘Asian democracy’. For example, therulers of Malaysia and Singapore explicitly reject theWestern interpretation of liberal democracy based onindividual rights.They favour an approach that givesmore weight to Asian values, including respect forauthority, avoiding public conflict and accepting theprimacy of the group. Democracy is defined in almostfamilial terms, with the elected leader adopting apaternal style.The state leads society, and democracytherefore depends less on the independent groupsand associations which provide the foundation forWestern democracy.

The institutional consequences of ‘Asian democracy’include a subservient media and judiciary. In addition,the police and security forces become more aggres-sive in their approach to criminals and dissenters. Butis there really a distinctive form of democracy in Asiaor is democracy a universal principle?

The case forThe attempt to develop non-Western models ofdemocracy derives in part from the natural cynicism offormer subjects to their colonial masters. Asian leadersreject what they see as imperialist attempts to univer-salize Western democracy. Dr Mahathir, when PrimeMinister of Malaysia, condemned Western democracies‘where political leaders are afraid to do what is right,where the people and their leaders live in fear of thefree media which they so loudly proclaim as inviolable’.A former foreign minister of Vietnam exposed Westernhypocrisy more bluntly:‘Human rights? I learnt abouthuman rights when the French tortured me as ateenager’ Thompson (2001, p. 160). Further, the Asianmodel has delivered economic growth by allowingleaders to focus on long-term modernization free fromelectoral pressures.Thus Prime Minister Goh ofSingapore suggests that

our government acts more like a trustee. As a custo-dian of the people’s welfare, it exercises indepen-dent judgement on what is in the long-termeconomic interests of the people and acts on thatbasis. Government policy is not dictated by opinionpolls or referenda (Wang, 2002, p. v).

The case againstCritics allege that ‘Asian democracy’ is simply anexcuse for failing to move beyond semi-democracy.Putzel (1997, p. 253) roundly declares that ‘claims for“indigenous forms of democracy” appear to be nomore than justifications for authoritarian rule’. AndBrzezinski (1997, p. 5) suggests that ‘the “Asian values”doctrine is nothing but a rationalization for a certainphase of historical development’. By this he meansthat through accidents of history Western societieshave more experience with protecting individualfreedom. Asia, Brzezinski suggests, is still playingcatch-up, both economically and politically. Andresolving the financial crisis that engulfed East Asia in1997 required most countries in the region to some-what reduce state intervention in the economy,yielding a rather more liberal form of democracy.Finally, a Western human rights activist argues thatdemocracy and human rights are inherently universal:

There is nothing special about torturing the Asianway. Rape is not something that is done an Asianway. Rape is rape, torture is torture and humanrights are human rights.

(Vatikiotis, 1995, p. 98)

AssessmentThe debate on Asian democracy can not be resolvedeasily. It mixes ideology and colonial memories in anexplosive combination. But three points are clear. First,Asia has never been a single category. China is stillauthoritarian, Indonesia largely so, Singapore is asemi-democracy and Japan is an established democ-racy. Second, rather than referring to ‘Asian’ democ-racy, it might be more useful to consider the kind ofdemocracy best suited to economic development.The ‘Asian’ approach may be more effective in coun-tries which are still growing their industrial capacityeven if the Western model is more appropriate fordeveloped economies.Third, the liberal form ofdemocracy found in the West reflects a long-termproject to tame the power of secular and religiousrulers. If Asia is to ‘catch up’, it will take generations todo so.

Further reading: Bell et al. (1995), Diamond and Plattner(2001),Thompson (2001).

DEBATE

‘ASIAN DEMOCRACY’

has also ruled with competence. Similarly,President Putin’s willingness to pull all the leversof power available to him has not dented hissupport among the Russian public; it may indeedhave endeared him to them.

Once elected, semi-democratic presidents rulethe roost and the assembly and the judiciary arecowed into insignificance. Ordinary ‘citizens’ mayhave a vote and their rights are tolerably securewhen the interests of the regime are not at stake.But citizens are sensitive to the sound of gunfire.They know when to lie low. With good reason, thisform of semi-democracy is sometimes called illib-eral or electoral democracy (O’Donnell, 1996).

In a semi-democracy based on a dominant partyor individual, power is concentrated in a fewhands. But there is a second form of semi-democ-racy in which elected rulers have too little ratherthan too much power. Here elected rulers arepuppets rather than despots. In this version,‘power is shifted to the military, bureaucracy ortop business groups’ (Case, 1996, p. 439). Likeweak monarchs surrounded by powerfulnoblemen in medieval Europe, these elected politi-cians must continue to govern alongside military,ethnic, religious and regional leaders determinedto maintain their established privileges.

When the president is merely a frontman, theoutcome is an unconsolidated democracy in whichelections are established but do not function asdefinitive statements of who should exercise finaldecision-making power. In Thailand, Turkey andPakistan, for example, the military stands as aguardian of the nation, exerting a ‘silent veto’ overcivilian decisions (Gills et al., 1993, pp. 21–8).Such semi-democracies are sometimes calledsupervised or even façade democracies.

Some post-soviet republics, including centralAsian republics such as Kyrgyzstan andUzbekistan, exemplify this pattern of limitedauthority for elected politicians (Fairbanks, 2001).Real power comes from patronage and from dealswith regional and ethnic power brokers or evenwith criminal gangs. In this type of semi-democ-racy (‘disguised authoritarianism’ might be moreaccurate), power gives the capacity to take electedoffice but elected office does not add much power.In these circumstances, the president is merely themouthpiece of a dominant and corrupt elite, and

elections are just plebiscites to confirm the elite’schoice of top leader. Democracy fails because pres-idents are impotent rather than despots.

The ways of semi-democracy are a soberingreminder to those who take a naive view of thetriumph of democracy. Simplistic counts of thenumber of democracies tell only part of the story.As quantity has increased, quality has fallen. Whyis it, then, that so many new democracies turnout, on closer inspection, to be only semi-democ-ratic? There are two answers, the optimist’s andthe pessimist’s.

The optimist’s view is that semi-democracy ismerely transitional, a temporary staging post inthe world’s pilgrimage from authoritarian rule toestablished democracy. This scenario possesses acertain plausibility. After all, nearly all Westerndemocracies passed through a stage in which thecontest for power became open and legitimate, aphase which preceded the introduction of uni-versal suffrage. Even in the United States, democ-racy took decades to establish.

But it is prudent to consider a more pessimisticaccount of semi-democracy: that it is a stablemethod of governing poor and unequal societies,particularly now that blatant dictatorship hasbecome less acceptable. When poverty coincideswith extreme inequality, and when ethnic divisionsare strong, the prospects of creating a democraticcommunity of equals are slender indeed. Further,semi-democracy is usually sufficient for the rulingelite to meet the conditions of aid set by theWorld Bank, the IMF and donor governments.While these international bodies may welcomedemocracy, in practice they give higher priority toeconomic reform.

The heart of the matter is perhaps that semi-democracy is a tacit, but stable, compromisebetween domestic elites and international organi-zations. For such reasons, Case (1996, p. 464)concludes that semi-democracy is not ‘a mere waystation on the road to further democracy’.

Key reading

Next step: Dahl et al. (2003) is a wide-rangingcollection on the nature, conditions, proceduresand impact of democracy.

DEMOCRACY 49

Dahl (1989, 1998) offers lucid accounts ofdemocracy. The 1989 book is more advanced, the1998 volume more introductory. Other overallassessments include Arblaster (2002) and Held(1996). Democratization has spawned an out-standing literature: Huntington (1991) andO’Donnell et al. (1986) are influential, whilePridham (1995) is a collection of classic articles.Also on democratization, Grugel (2002) providesan introductory overview while Diamond (1999)

focuses on consolidation. Carothers (2002) dis-cusses the end of the transition paradigm as coun-tries enter a seemingly permanent grey zone ofsemi-democracy. For democracy in the developingworld, see Haynes (2001); for Islam and democ-racy, Diamond et al. (2003) and for democracyafter communism, Diamond and Plattner (2002).Gill and Marwick (2000) review Russia’s stillborndemocracy. Agüero and Stark (1998) remains anexcellent survey of post-transition Latin America.

50 FOUNDATIONS

Traditional authoritarian rule 51Communist, fascist and military rule 53The Arab and Muslim worlds 62China in transition 66Key reading 68

Brooker (2000, p. 1) writes that ‘non-democ-ratic government, whether by elders, chiefs,

monarchs, aristocrats, empires, military regimes orone-party states, has been the norm for most ofhuman history’. As late as 1981, Perlmutter (p. xi)could still claim that ‘the twentieth century is theage of political authoritarianism’. Certainly thatbrutal century will be remembered more for thedictatorships it spawned – including Hitler’sGermany, Stalin’s Russia, Mao’s China and PolPot’s Cambodia – than for the democratic transi-tions at its close.

But studying non-democracies remain far morethan an historical exercise. Authoritarian rulersmay be under more pressure as democratizationspreads but the species is far from extinct. Indeed,September 11, 2001 brought non-democraticregimes into sharper focus. Most of the terroristsinvolved in the attacks on the United States,including Osama bin Laden, were nationals ofSaudi Arabia, a leading example of an authori-tarian Islamic state. More generally, the Arabworld contains a high proportion of the world’sstock of both authoritarian governments and oil.

Other non-democratic regimes are also of inter-national significance. In China, for instance, anominally communist ruling elite continues togovern a quarter of the world’s population and arapidly expanding economy. China’s distinctivecombination of authoritarian politics with a partlyfree economy has acquired global significance par-ticularly since China joined the World TradeOrganization in 2001.

We begin by examining authoritarian rule in thetraditional style, a form common to non-democ-ratic rulers past and present, before turning to the

new forms of authoritarianism which emerged inthe twentieth century, namely communist, fascistand military rule (Box 4.1). We then considercontemporary authoritarianism in the Arab andMuslim worlds and we conclude with an examina-tion of China.

Traditional authoritarian rule

Traditional authoritarianism is a distinct form ofnon-democratic rule in which authority is owed tothe ruler himself rather than to a more abstractentity such as a communist or fascist party. Thepeople are subjects, not citizens, and the ruler isconstrained neither by law nor by competitiveelection. Nonetheless, the ruler is expected to takeresponsibility for his people, just as a father shouldlook after his children, in a format known as patri-monial rule. The abstract idea of a state linkingrulers and citizens is missing, as are such modernnotions as constitutions, rights, the separation ofpowers and the rule of law. Rather, the law (if itexists at all) expresses the wishes of the ruler.

Definition Weber’s notion of patrimonial rule is based onthe personal authority of a leading male whorules as if he were the head of a large family. As afather-figure, the ruler claims to care for hisdependants but at the same time his dominantposition affirms a relationship of inequality.Patrimonial rule is traditional – patrimony liter-ally means an inheritance from one’s father – butremains common in authoritarian regimes today(Gerth and Mills, 1948).

The major forms of traditional rule were chief-doms and monarchies, though these sometimesextended to larger empires such as Imperial China,Mesopotamia and the Aztecs (van Creveld, 1999).With the rise of the modern state, traditional rule

Chapter 4

Authoritarian rule

51

has become less common. However, presidentsgoverning in the traditional style are still commonin Africa and the Arabian Gulf.

Whether chief, monarch, emperor or president,the authority of the traditional ruler is technicallyunlimited. The leader’s authority is typically based

on religion, as with the divine right of kings inpre-modern Europe and the mandate of heavenclaimed by Chinese emperors. Succession is oftenbased on heredity, maintaining the fictionaldescent from God, but in practice is oftenacquired through usurpation. Succession can beeither to the eldest son or (as with many rulingfamilies in the Middle East) to the eldest capablerelative. The latter method is less clear-cut,inviting short-term conflict.

However, even when an outsider does becomeruler, the method of governing often continuesunchanged. Thus, traditional authoritarian rulecan provide a settled political framework, espe-cially for static agricultural societies with littleneed of government.

For all their theoretical authority, the power ofmost traditional non-democratic leaders is neitherunlimited nor arbitrary. For one thing, rulers whoclaim divine authority must ration those actionscontradicting the religious code. More important,in the absence of an extended bureaucracy, rapidtransport and modern communications, rulerslack direct means of controlling their subjects.They are forced to administer their kingdoms andempires indirectly, calling on the services of localleaders. Kings, and emperors even more so, havelittle choice but to govern by making deals withprovincial notables.

Finer (1997, p. 38) suggests that palace politicsis the characteristic mode of traditional authoritar-ianism. Befitting the personal character of authori-tarian governance, in palace politics the officers ofstate are nothing more than servants of the ruler.Thus, the Keeper of the King’s Purse and Ministerof Finance are one and the same. Because alle-giance is owed to the ruler rather than to rules,palace politics is based on personal relationships.Finer’s examples of pure ‘palace-type politicalsystems’ include ancient Egypt, the Roman empireand some eighteenth-century absolute rulers inEurope, such as the court of Louis XIV in France.

While palace politics can provide stable gover-nance, the danger of traditional authoritarian ruleis that it becomes insular and introverted; toomuch politics and not enough government.Further, the court constitutes a tax on society:money comes in from the authority of officials togrant licences and take bribes. This intensely polit-

52 FOUNDATIONS

Form Definition

Authoritarian (1) Any form of non-democratic rule rule. (2) Those non-democratic

regimes which, unlike totalitarianstates, do not seek to transformsociety and the people in it

Traditional Allegiance is owed to an individual authoritarian ruler who often claims a religious rule mandate passed down through

family succession.The chief,monarch or president rules his sub-jects as if he were the head of anextended family

Communist Political systems in which the states communist party monopolized

power, leading to an all-encom-passing bureaucratic state. Intheory, the object was to implementMarx’s vision of a classless society;in practice, the party sought toprotect its position through socialcontrol

Fascism An anti-liberal doctrine that glori-fied the nation and advocated awarrior state, led by an all-powerfulleader, to whom the masses wouldshow passionate commitment andsubmission. Advocated by Mussoliniin Italy and, supplemented by Aryanracism, the basis for NationalSocialism in Nazi Germany

Military rule Government by the military, oftenruling through a junta comprisingthe leader from each branch of theforces. Half the countries in Africawere under military control as lateas 1987

B OX 4 . 1

Forms of authoritarian rule

ical economy discourages economic developmentand is increasingly unpopular with internationalagencies. With good reason, traditional authori-tarian leaders rightly regard not just democracy,but also development and the agencies whichpromote it, with suspicion.

Communist, fascist and military rule

The twentieth century raised the political stakes.Unlike their traditional counterparts, modern dic-tators could exploit the political power of anextended state, the economic resources of theIndustrial Revolution and the communicationsfacilities of national media. These developmentspermitted unprecedented mobilization andcontrol of mass populations. Authoritarian leaderswere no longer just masters of their palace; theirdecisions now impinged directly on ordinarypeople. In extreme cases such as Stalin’s Russia andHitler’s Germany, the result was the systematicmurder of millions. So although much traditionalrule continued in the twentieth century, our focusin this section will be on that century’s new cast of dictators: communist, fascist and militaryleaders.

Both communist and fascist rulers claimed to beseeking a reconstruction of human nature andsociety. Communist states notionally aimed for aclassless utopia while fascist rulers sought to renewthe nation’s strength through submission to adominant leader. These bombastic declarationswere not always matched by political reality buteven so, such bold aspirations were far removedfrom the traditional authoritarian regime, with itsoverriding commitment to maintaining the ruler’sposition. Certainly, a distinctive feature of com-munist and fascist regimes was what Perlmutter(1981, p. xi) terms the ‘conspicuous political inno-vations’ of twentieth-century authoritarianism,namely:

the unopposed single party, the party-state,political police, the politburo [top party com-mittee], revolutionary command councils, stormtroops, political youth movements, cadres andgulags, propaganda machinery and concentra-tion camps.

Definition A totalitarian regime aims for total penetrationof society in an attempt, at least in theory, totransform it. As defined by Linz (2000, p. 4), atotalitarian system is ‘a regime form for com-pletely organizing political life and society’.During the Cold War, communist and fasciststates were bracketed as totalitarian, a connec-tion which served to link communist regimeswith disgraced fascism (Gleason, 1995).

Below, we will outline communist and fascistrule before turning to the large number of militarygovernments – usually authoritarian rather thantotalitarian – which ruled many developing coun-tries for part of the second half of the twentiethcentury.

Communist states

The 1917 October Revolution in Russia was adecisive event of the twentieth century. It signalledthe international advent of a regime, an ideologyand a revolutionary movement which sought tooverthrow the capitalist democracies of the West.Although communism failed to become a gov-erning force in the affluent West, communistpower did expand dramatically in Eastern Europeand Asia. The Union of Soviet Socialist Republics(USSR) – effectively a new Russian empire – wasformed in 1924, extending from the Ukraine inthe west to the central Asian republic ofKazakhstan in the east (Table 4.1 and Map 4.1).By area, the USSR became the largest country inthe world.

After 1945, Eastern European countries such asPoland and Romania became satellite territories ofthis new empire. In Asia, the Chinese revolutionof 1949 established an additional if distinctivecommunist state. During the Cold War, severaldeveloping countries such as Benin and the Congoalso declared a nominal Marxist allegiance. Untilthe decisive collapse of the communist order inthe late 1980s and early 1990s, regimes claimingMarxist inspiration ruled more than 1.5 billionpeople: about one in three of the world’s popula-tion.

In seeking to understand communist rule, weshould note the sharp contrasts between ideology

AUTHORITARIAN RULE 53

54 FOUNDATIONS

Table 4.1 Post-communist states in Eastern Europe and the former Soviet Union

State Population Gross domestic Ethnic groups comprising estimate, 2003 product per head, over 10% of the population,(million) 2002 2002 (listed by size of group)

Eastern European states formerly under the control of the Soviet Union

Albania 3.6 $4,500 Albanian

Bosnia and Herzegovina 4.0 $1,900 Serb, Bosnian, Croat

Bulgaria 7.5+ $6,600 Bulgarian,Turk

Croatia 4.4 $8,800 Croat

Czech Republic* 10.2+ $15,300 Czech, Moravian

Hungary* 10.0+ $13,300 Hungarian

Macedonia, Former Yugoslav

Republic of 2.1 $5,000 Macedonian, Albanian

Poland* 38.6 $9,500 Polish

Romania 22.4 $7,400 Romanian

Serbia and Montenegro 10.7 $2,370 Serb, Albanian

Slovakia* 5.4 $12,000 Slovak, Hungarian

Slovenia* 1.9 $18,000 Slovene

States formed from the Soviet Union

Armenia 3.3+ $3,800 Armenian

Azerbaijan 7.8 $3,500 Azeri

Belarus 10.3+ $8,200 Belarusian, Russian

Estonia* 1.4+ $10,900 Estonian, Russian

Georgia 4.9+ $3,100 Georgian

Kazakhstan 16.8 $6,300 Kazakh, Russian

Kyrgyzstan 4.9 $2,800 Kyrgyz, Russian, Uzbek

Latvia* 2.3+ $8,300 Latvian, Russian

Lithuania* 3.6+ $8,400 Lithuanian

Moldova 4.4 $2,500 Ukrainian, Russian,Moldovan/Romanian

Russia 144.5+ $9,300 Russian

Tajikistan 6.9 $1,250 Tajik, Uzbek

Turkmenistan 4.8 $5,500 Turkmen

Ukraine 48.0+ $4,500 Ukrainian, Russian

Uzbekistan 26.0 $2,500 Uzbek

* Joined the European Union in 2004.+ Falling population.

Source: CIA World Factbook at http://www.cia.gov/cia/publications/factbook.

and practice. In his theoretical writings, Karl Marx(1818–83) had envisaged an equal, classless andstateless utopia in which goods would be distrib-uted from each according to their ability to eachaccording to their need. In The CommunistManifesto (1848, p. 244), Marx and Engelsclaimed that ‘in place of the old bourgeois society,with its classes and class antagonisms, we shallhave an association, in which the free developmentof each is the free development of all’. In the tran-sition to this utopia, Marx suggested that the statewould be converted ‘from an organ superior tosociety to one completely subordinate to it’.

Practical revolutionaries, though, faced a moreimmediate problem: how to overthrow theexisting capitalist order. Here Vladimir Lenin(1870–1924), the Russian revolutionary, made apivotal contribution. He argued that the commu-nist party should serve as a vanguard organization,leading the workers into political activity thatwould further enhance their revolutionary con-sciousness. By assuming the party possessed adeeper understanding of the true interests of theworking class than did the workers themselves,

Lenin provided the crucial rationale for themonopoly position which communist partiescreated once in power. In this way, the dictator-ship of the party supplanted Marx’s utopiandreams.

In power, ruling communist parties dominatedsociety. Lenin’s view that the workers must beforced to be free simply resulted in no freedomwhatever. Communist regimes were stronglyauthoritarian, brooking no opposition, stage-man-aging elections, acting above the law, rewritingconstitutions, determining all major appointmentsto the government, controlling the media andspying on their populations. Far from disap-pearing as anticipated by Marx, the state under theparty’s tutelage became an enveloping presence.Economies were brought under public control aspart of the push to industrialize; the elaborate five-year plans produced in the Soviet Union wereundoubtedly the most ambitious, detailed andcomprehensive attempts at economic planning theworld has ever seen. The party controlled and thestate implemented.

This new form of party-state snuffed out inde-

AUTHORITARIAN RULE 55

Map 4.1 Post-communist Eastern Europe and Central Asia

7

Belarus

Ukraine

Poland

RomaniaHungary

Latvia

9

6

9

Bulgaria134

5

12

1

Black SeaCaspian

SeaGeorgia3

2

3

Turkmenistan

Uzbekistan

Tajikistan

Kyrgystan

Kazakhstan

R u s s i a

KEY:

1 Albania 2 Armenia 3 Azerbaijan 4 Bosnia-Herzegovina 5 Croatia 6 Czech Republic 7 Estonia 8 Former Yugoslav Republic

of Macedonia 9 Lithuania10 Moldova11 Slovakia12 Slovenia13 Serbia and Montenegro

8

11

pendent organizations, creating a social wastelandof distrust in which true political beliefs couldonly be expressed safely within the family (andsometimes not even there). Not only was activeopposition suppressed but explicit support – inthe form of attendance at demonstrations, party-led meetings and elections – was required. Thisinsistence on active if ritual support was one factordistinguishing totalitarian from merely authori-tarian regimes.

As communist states retreat into history, weshould be careful to avoid stereotyping their char-acteristics. Communist regimes varied amongthemselves and also over time. In parts of EasternEurope, for example Poland, local communistleaders governed with a lighter touch than in thecommunist heartland of the Soviet Union. Chinafollowed an even more independent path, with thetriumph of the communist revolution in 1949owing as much to nationalism as to Marxism, andresulting from the efforts of the army as well as theparty (Selden, 1995).

Similarly, communist states grew less totalitarianas they matured. Once the initial thrust to indus-trialize had been achieved, many such regimessettled into the routines of middle age. The SovietUnion is a striking example. During the 1930s,under Stalin’s brutal dictatorship, Russia hadachieved forced industrialization and the collectiveownership of agriculture. But after the tyrant’sdeath in 1953, a programme of ‘de-Stalinization’quickly emerged. Nikita Khrushchev, the newparty secretary, famously denounced Stalin in hissecret speech to the party elite in 1956. Terrorceased to be a routine political tactic and theSoviet Union came to offer a more predictableenvironment to its citizens.

Yet far from stabilizing the party’s control, theattempt to transform communist rule into morerational and orderly governance eventually provedits undoing. State-led planning achieved speedyindustrialization but proved incapable of deliv-ering the advanced products and services found inthe West. ‘Advanced socialism’ proved to be a con-tradiction in terms. Communism reached its deadend. The party lost its mission and continued torule only because it had done so in the past. Insuch circumstances, reform was always likely toescalate into revolution.

When Mikhail Gorbachev became GeneralSecretary of the Soviet Communist Party in 1985,his intention was modernization but the outcomewas dissolution. In Eastern Europe, communistrule fell apart in 1989 once the new Russian leadermade it clear that the USSR would no longerintervene militarily to protect the puppet rulers ofits satellite states. The following year, the SovietUnion itself dissolved into 15 constituentrepublics (Table 4.1). In Russia, by far the mostimportant of these republics, the CommunistParty was outlawed, a humiliating fate for whathad been the most powerful party on earth.

Even where nominally communist rule still sur-vives, as in China, Vietnam and Laos, most eco-nomic development now occurs outside the statesector. In the twenty-first century, communism’smajor significance lies in its legacy for successorregimes. As a system of rule and a method of eco-nomic organization, communism is finished. Itwas the future that didn’t work.

Fascist states

Fascism was the twentieth century’s secondremarkable contribution to authoritarian rule.Located at the extreme right rather than the farleft of the ideological spectrum, fascist regimesnonetheless sought – like communist states – todominate the societies they ruled. But fascistregimes were rarer and less stable than their com-munist equivalents. Although we can still observethe consequences of communism in the twenty-first century, fascism’s challenge was confined tothe period bordered by the two world wars.

Fascism began with the emergence of revolu-tionary groups (fascia) in Italy during the FirstWorld War (1914–18). As a serious force, it endedwith the defeat of Germany in 1945. Fascist ele-ments continued to be found in Spain underGeneral Franco and even Portugal under Salazar,two dictators whose right-wing rule continuedwell into the 1960s. But these were conservativeauthoritarian regimes rooted in the army and thechurch; they sought merely to recover traditionalnational glories rather than to build a new andself-consciously modern order (Linz, 2000).Similarly, the right-wing anti-immigrant partiesfound in contemporary Europe do not embrace all

56 FOUNDATIONS

AUTHORITARIAN RULE 57

Can authoritarian rule be defended as an effectivemethod for economic development? If so, we willhave a powerful critique of democracy’s claims to beuniversally the best form of government since wewould nearly all prefer to eat under a dictator than tostarve in a democracy. In addition, we could reason-ably anticipate that democracy is unlikely to consoli-date in poor countries over the long term if the cost isslower economic development.

The case forSørensen (1997a, p. 65) points out that ‘in the twen-tieth century there was no case of successful eco-nomic development without comprehensive politicalaction involving enormous state intervention in theeconomy’.The reason is clear: industrializationrequires massive investment in infrastructure such astransport, communications and education; initially,these can only be funded by the state. And authori-tarian rulers can generate the surplus needed forinvestment precisely because they can resist short-term pressures for immediate consumption. Simplyput, they can kick-start development because theycan ignore the squeals of those whose consumptionis initially held down.

Consider some examples.The communist revolutionin Russia initiated a remarkably rapid transformationfrom a rural to an industrial society. Similarly, theChinese economy has grown under communism attwice the rate achieved by democratic India. Between1960 and 1985, authoritarian Indonesia, Singaporeand South Korea were among the fastest-growingeconomies in the world. In parts of Latin America, too,technocrats operating under more or less authori-tarian governments succeeded in the final decades ofthe century in imposing coherent economic policy onunruly societies. Even some military rulers have initi-ated worthwhile modernization: for example, the landreforms introduced by General Abdel Nasser(President of Egypt, 1956–70) mean that nearly allEgyptians now have access to safe water, an accom-plishment as yet unmatched by India.

The case againstA few non-democratic regimes may initiate economicdevelopment but the vast majority do not. Many tradi-tional rulers, such as the ruling families in the MiddleEast, continue to resist modernization. Other dictators,for example Nigeria’s military ‘lootocrats’, set back eco-nomic development by decades. An overall assess-ment by Przeworski et al. (2000) concludes that there isno ‘cruel choice’ to be made between democracy anddevelopment. If industrialization really does requireforgoing immediate consumption, rulers shouldattempt to persuade the people of the need for sacri-fice, not impose dictatorial solutions. Besides, even ifnon-democratic rule can lead to industrialization, thatpoint does not excuse the abuses of power andhuman rights which are an inherent danger of authori-tarian regimes. For example, China’s path of commu-nist modernization involved the brutality of the GreatLeap Forward, in which around 40 million people diedbetween 1958 and 1963 as a result of a bungled exper-iment in forced collectivization.Who is prepared to say – indeed, who is entitled to say – that economicgrowth is justified at such a massive human price?

AssessmentPerhaps economic development in the twentiethcentury could only be achieved by a stable authori-tarian elite capable both of extracting resources forinvestment and of providing state leadership foremerging private industries. But in the twenty-firstcentury, globalization has given developing countriesaccess to new sources of capital through multinationalcorporations, overseas banks and the World Bank.Toaccess these resources, developing countries benefitfrom convincing lenders that their economy is market-based and that their politics takes the form of a toler-ably liberal democracy.The twentieth century mayprove to have been the pinnacle of ‘the developmentalstate’ led by authoritarian rulers; in the new century,markets and democracy may belong together indeveloping as well as developed countries.

Further reading: Przeworski et al. (2000), Robinson and White(1998), Sørensen (1997a).

DEBATE

AUTHORITARIAN RULE AS A RECIPE FOR ECONOMICDEVELOPMENT

aspects of the interwar ideology. Even thoughthese protest parties are frequently condemned asfascist, their essential character is post-fascist(Ignati, 1992).

Even in its interwar heyday, fascist regimes wererare on the ground. Mussolini’s leadership of Italy,lasting from 1922 until il duce was deposed in1943, is the main example. However even this dic-tatorship was fascist more by bombastic declarationthan by institutional reality. In Hitler’s Germany,the Nazi party espoused an ideology that certainlyincluded fascist principles. However, these ele-ments were blended with crude Aryan racism toform the compound known as national socialism.

Yet the fascist worldview cannot be ignored; itrepresents an important nationalist response to therise of communism. Its significance for the twen-tieth century – and for six million European Jewsin particular – was profound.

What, then, was the doctrine expressed by theclassic fascist regimes? Fascism was an extreme glo-rification of the nation, often defined in racialterms. The notional purpose was to create an all-embracing nation to which the masses wouldshow passionate commitment and submission. Anautocratic ruler and a single party would personifythe state. State and nation would become one.

Fascism lacked the theoretical sophistication ofcommunism; it offered an ideological impulsemore than a coherent plan. It sought to use thepower of the state, as revealed by the First WorldWar, to revive the countries defeated in that con-flict. Religion, liberalism, parliamentary democ-racy and even capitalism were condemned as weakdistractions from the key task of national revival.Fascists claimed that a strong, self-sufficient,warlike nation could mobilize the populationmore effectively – and in a more modern way –than any other type of regime. ‘Everything in thestate: nothing against the state: nothing outsidethe state’, said Mussolini. Fascism, not liberalism,was the defence which proud nations should adoptagainst the communist threat. In short, fascismwas the twentieth-century doctrine of nationalismtaken to extremes (Griffin, 2004).

In power, fascist regimes governed very differ-ently from ruling communist parties, even thoughboth forms are often grouped under the totali-tarian label. Certainly, fascist rulers were com-

mitted to mobilizing the population in an orga-nized effort at national rebirth, just as communistregimes claimed to be constructing a classlesssociety. And both ideologies gave primacy to poli-tics: nothing could compete with the authoritybestowed on the supreme fascist leader, just asruling communist parties dominated their ownsocieties.

But fascism lacked the organized character ofcommunist rule. It favoured the risky ‘leader prin-ciple’ in which governance depended on a singleindividual rather than a well-developed party.Hitler, for one, never showed much interest inadministration, preferring to leave his underlingsto fight their own bureaucratic battles (Kershaw,2000). Mature communist states often ran onauto-pilot, with an anonymous party functionaryin charge, but fascism – a doctrine of constantmovement and change – never developed compa-rable routines of rule.

Fascist parties were essentially personal vehiclesthrough which the leader managed his rise topower; unlike communist parties, they lost signifi-cance once the state was won. In power, neitherMussolini nor even Hitler achieved the domina-tion of society found under communism.Mussolini proved incapable of abolishing even theItalian monarchy while Hitler preferred to exploitrather than nationalize German industry. For allits impact on the twentieth century, fascist prac-tice often seemed to present politics as theatre:marches, demonstrations, symbols and speeches. Itwas no surprise that fascism’s collapse in 1945 pre-ceded that of its better-organized communistbogeyman.

Military rule

Military rule is our final form of twentieth-century authoritarian government. Most militaryregimes lacked the ideological underpinnings ofcommunism or even fascism; indeed, they typi-cally lacked any theoretical justification at all.Nonetheless, military government was an impor-tant aspect of twentieth-century government. AsPinkney (1990, p. 7) writes,

the involvement of soldiers in politics is notnew, and can be traced back at least as far as

58 FOUNDATIONS

Roman times. The phenomenon of militarygovernment, in the sense of a government drawnmainly from the army and using the army as itsmain power base, is much newer and belongsessentially to the last 50 years.

The contrasts between military regimes, on theone hand, and communism and fascism, on theother, are acute. Most military coups came later inthe century, between the 1960s and 1980s, and,more significantly, they occurred in post-colonialcountries in Latin America, Africa and parts ofAsia where the state had not achieved the penetra-tion found in Europe. While fascism and commu-nism sought to exploit the power of the modernstate, many military coups (especially in smallerAfrican countries) were made possible preciselybecause the state remained simple and underdevel-oped. An ambitious general just needed a fewtanks, driven by a handful of discontented officers,to seize the presidential palace and the single radiostation.

Yet because the post-colonial state’s penetrationthrough society remained limited, life outsidethe capital would continue unchanged after thecoup. Lacking the economic resources and gover-nance tools of modern states, most militaryrulers were modest in their policy aspirations.The state in uniform lacked the grand objectivesof both communist and fascist regimes; in somecases, the aim of the generals was little morethan to steal public money. Military governmentwas always authoritarian and sometimes brutal,not least during Latin America’s phase of repres-sive army rule from the mid-1960s to the mid-1970s. But, reflecting the societies in which themilitary came to power, army rule was rarelytotalitarian.

DefinitionA military coup is a seizure of political power bythe armed forces or sections thereof.The termconjures up images of a violent, secretive andunwelcome capture of power against the oppo-sition of civilian rulers. In fact, many coupsreplaced one military regime with another,involved little if any loss of life and were more orless invited by the previous rulers.

In many of the post-colonial countries created inthe 1950s and 1960s, generals soon seized powerfrom civilian rulers – and then from other gen-erals. Sub-Saharan Africa is the major arena. Here,68 coups occurred between 1963 and 1987(Magyar, 1992). But military takeover was notrestricted to new states. In Latin America, wherecolonies had gained independence in the nine-teenth century, only Mexico and Costa Rica wereimmune from military government in the postwarperiod. Military governments became far morenumerous than communist and fascist regimescombined.

DefinitionThe Cold War refers to the competition betweenthe United States and the Soviet Union whichlasted from the late 1940s to the Soviet Union’scollapse in 1991.The Cold War reached a highintensity of confrontation, particularly beforedétente began in the late 1960s. Its end was anevent of the first magnitude, releasing the wavesof globalization, regionalization, nationalism anddemocratization which characterize the twenty-first century.

Why did military coups cluster in the decadesfollowing the Second World War? As with otheraspects of politics during this era, the Cold Warwas a crucial factor. In this period, the UnitedStates and the Soviet Union were more concernedwith the global chessboard than with how post-colonial countries were governed internally. Eachsuperpower sought allies and did not enquireclosely into the background, civilian or military,of a country’s rulers. Thus, governing generalscould survive through the political, economic andmilitary backing of a superpower even thoughthey might lack support in their own country.Simple contagion, in which a coup in one countrywas emulated by its neighbours, was anotherinfluence.

Inclusionary and exclusionary regimes repre-sented the two extremes of military rule (Remmer,1989). In the former, the military leaders soughtto build a base of support among the political class– and even, on occasion, in the wider population– often by exploiting the population’s respect for astrong leader. Civilian politicians were represented

AUTHORITARIAN RULE 59

in a cabinet and the bureaucracy continued tomake important decisions.

The modernizing regime of Colonel Juan Perónin Argentina was an example. Perón came topower in a coup in 1943, serving as presidentbetween 1946 and 1955 and again in 1973–74.He undertook a populist programme of state-ledindustrialization based on a strong trade unionmovement and a commitment to social welfare forthe urban working class. This Peronist amalgam,based on a distinctively Catholic commitment toboth order and reform, continues to influenceArgentinian politics (Norden, 1996).

But most military governments were exclu-sionary rather than inclusionary. In classic authori-tarian fashion, the generals sought to preventpopular participation so as to entrench their ownposition. Opposition was always monitored andsuppressed as necessary. Consider GeneralPinochet’s bloody rule of Chile between 1973 and1989. Pinochet eliminated all potential sources ofpopular opposition. He exterminated, exiled orimprisoned thousands of labour leaders and left-wing politicians, concentrating power in the handsof his ruling military clique (Drake and Jaksic,1989).

The standard institutional form of an exclu-sionary military regime was the junta (council), asmall group made up of the leader of each branchof the armed forces. In Chile, Pinochet himselfacted as chief executive while a classic four-manjunta representing the army, navy, air force andnational police took over legislative tasks.

Just as military governments prospered duringthe Cold War, so they shrivelled after its close. AsWiseman (1996, p. 4) writes, ‘authoritarianAfrican political leaders [such as the generals] weremore strongly placed to resist the pressures ofAfrican democrats when they could turn tooutside pressures to help them stay in power’. By

the 1990s, these rulers could no longer rely ontheir sponsoring superpower; instead, condition-ality ruled the roost. Aid and technical assistanceflowed to civilian regimes that adopted democraticforms and offered at least some protection to civilrights. International bodies such as the WorldBank stipulated market-based economic policiesthat did not sit comfortably with military rule.

Just as contagion had accelerated the diffusionof military coups in the 1960s and 1970s, so alsodid it encourage generals to return to their bases inthe 1980s and 1990s. In Latin America, evenbefore the Cold War ended, and later in most ofAfrica, the military withdrew from formal rule,transforming the pattern of government aroundthe world. The last Latin American generals wereback in their barracks by 1993 and any coupssince then have been sporadic affairs confined tosmaller countries in the region (Figure 4.1).

For now at least, military governments – likecommunist states – are known mainly for theirimpact on successor regimes. So we conclude thissection by examining the difficult legacy of mili-tary rule for contemporary civilian leaders. Themain problem is that long periods of army rule ledto an interweaving of civilian and military power.In many Latin American countries, senior officershad become accustomed to such privileges as

� guaranteed seats in the cabinet � a high level of military expenditure� sole control of the security agencies � personal profit from defence contracts � exemption from civilian justice.

The ending of military government did not meanan end to these resources. Indeed, some of theseprivileges were entrenched before military rulerscould be persuaded to relinquish their occupancyof the state.

60 FOUNDATIONS

Figure 4.1 The ending of military rule in Latin America

Ecuador Peru Bolivia Argentina Brazil Surinam Chile ParaguayUruguay

1979 1980 1982 1983 1985 1987 1990 1993

Note: Ecuador experienced a military coup in 2000.

AUTHORITARIAN RULE 61

In April 2003, Olusegun Obasanjo was reelectedcivilian president of Nigeria in elections considered tohave been only partly rigged. Hardly an inspiring sum-mation, we might think, except that this was the firsttime since independence that two successive elections had occurred without an intervening periodof military rule (even in 2003, the two leading candi-dates were former military heads of state). Nigeriaillustrates many of the difficulties faced by countrieswith an authoritarian history in consolidating democracy.

Under General Sani Abacha, the country’s militaryruler from 1993, governance had been corrupt, sordidand self-serving. Nigerians called the regime a lootoc-racy because it consisted of stealing public assets forpersonal benefit. After the general’s death in 1998 –popularly known as God’s coup – his wife was caughtfleeing to Saudi Arabia with 38 suitcases full of foreigncurrency. A son was intercepted carrying the consider-able sum of $100m in cash. Later governments havedevoted considerable effort to securing the return ofthese monies from deposit accounts in the City ofLondon.

The transition to civilian rule under Obasanjo wasremarkably smooth, raising hopes that Nigeria wouldbegin the road to recovery after decades of authori-tarian misrule.Yet so far the results have been limited,demonstrating not just the long-term damageinflicted by the military but also the deeper problemsfacing many African states. Apart from an expandingmobile telephone system, Nigeria’s economy remainsin poor condition. In the oil-producing Niger Delta,wealthy executives employed by multinational com-panies extract the vital commodity while local peoplesubsist in squalor amid a degraded environment.Thegovernment’s foreign debt remains an imposing$30bn, denting its international credibility.

The infrastructure which might permit rapid eco-nomic recovery has also decayed. Electricity is irreg-ular while endemic corruption scares off many foreigninvestors; Nigeria is probably the most corruptcountry in the world.The civil service is massivelyoverstaffed, with many illiterates appointed to postsrequiring documents to be processed. Inefficiency isrife: over $5bn of public money has been invested in asteel mill that has not yet produced any steel. Militaryequipment is in a chronic condition: the navy hasmore admirals than seaworthy ships. In a major oil-producing country, petrol is sometimes rationed.

Ethnic and religious conflict, superimposed onprovinces operating in a federal framework, holdsback post-military recovery.The central governmentbecame an arena for conflict between regions andbetween ethnic groups, leading to civil war in 1967.Even today, divisions between Hausa-Fulani,Yorubaand Ibos are entrenched. Because a gain for one groupis defined as a loss by the others, the national interestis subordinated to conflicts between North and Southand between Muslims and Christians.These divisions,leading to around 10,000 deaths in Obasanjo’s firstadministration, have intensified with the introductionof traditional Islamic law to some northern states.

The transition from authoritarian rule has thrownNigeria’s continuing difficulties into sharper relief. Anaimless continuation of the status quo is perhaps themost likely prognosis but neither national disintegra-tion, nor even another phase of military rule, can beruled out. Partly free elections notwithstanding,Nigeria seems to be incapable of developing into aconsolidated, united democracy.

Population: 134m, the tenth largest inthe world (HIV/Aids, 6 per cent).

Gross domestic product per head:about $345 but with markedinequality.

Main groups: Hausa-Fulani (29 per cent)in the north,Yoruba (21 per cent) in

the southwest and Ibo (18 per cent) inthe southeast.

Religions: Muslim 50 per cent, Christian40 per cent, traditional religions about10 per cent. Nigeria has the fifth largestMuslim population in the world.

Form of government: a presidential

republic. Civilian rule was reintro-duced in 1999 following 15 years ofmilitary rule.

Territorial basis of power: federal, withthe number of states increasing from12 in 1967 to 37 in 2003.

Profile N I G E R I A

Further reading: Holman and Wallis (2000), Maier (2002),Momoh and Adejumobi (2002).

Chile illustrates the difficulties of full disengage-ment. Before returning power to civilians in 1980,General Pinochet ensured that the new constitu-tion secured military autonomy. The armed forceswere granted exemption from prosecution incivilian courts and retained their position as guar-antors of the ‘institutional order’ and ‘nationalsecurity’. Similarly, Ecuador’s armed forces wereguaranteed 15 per cent of the country’s oil rev-enues until 2010. Such conditional transitions,characteristic of Latin America, helped the shift to,but weakened the depth of, the post-militaryregime. They signal the continued perception ofthe military as a source of order for the nation andthey leave a difficult bequest for new democracies(Pion-Berlin, 2001).

The Arab and Muslim worlds

In the twenty-first century, authoritarian regimesform a more diverse group than ever before; nolonger are their ranks dominated by military gov-ernments and communist party states. Instead, weare presented with a varied collection includingChinese Communist Party leaders, Pakistani gen-erals, Iranian clerics, Saudi princes and assortedauthoritarian presidents in some of the smallerstates of Central Asia, Africa and Latin America.These rulers have little in common beyond theirrejection of Western democracy. It is tempting todismiss this ragged band as twentieth-century left-overs, soon to fall victim to an American-inspiredembrace of democracy and capitalism. But such ajudgement is certainly premature, involving a riskybet on yet another wave of democratization.

In this section, we will focus on the authori-tarian regimes of the Arab and Muslim worlds, themain enclaves of non-democracy today. These twocategories overlap but not completely. The Arabworld is centred on the Arabian peninsula andNorth Africa. It includes Algeria, Egypt, Libya,Iraq, Morocco, Saudi Arabia and Syria. There isno established democracy in Arab countries.

Muslim countries – those with an Islamicmajority – include the Arab heartland but extendbeyond it. For example, Indonesia, Pakistan andTurkey are populous Islamic but non-Arab coun-tries. Again, most but not all Muslim countries are

authoritarian, with democracy confined to part ofthe Islamic periphery, notably Turkey (Map 4.2).

Table 4.2 shows the strong statistical relation-ship between Islam and non-democratic rule. Itdemonstrates not only that Islamic democraciesare rare but also that Muslim countries compriseone in two of the world’s authoritarian regimes.

With their massive oil reserves, the Arab andMuslim worlds have always attracted interest fromWestern commentators. But this attention was ofcourse magnified by 9/11 and the resultingAmerican-led invasions of Afghanistan and Iraq.Suddenly, the authoritarian governments of theArab and Muslim worlds became a focus ofWestern interest, with some American policy-makers beginning to seek ways of promoting bothdemocracy and market economies in a regioncharacterized by poor governance and low growth.

Why have Arab and Muslim countries resistedthe waves of democratization which have lappedagainst the shores of other authoritarian states?There can be little doubt that Islam and democ-racy are difficult bedfellows. In Islam – unlikeChristianity – religious and secular authority arecombined rather than separated. Religious valuessuffuse politics, limiting the space for an indepen-dent political will expressed through democraticmeans. Just as Islam dominates culture, so reli-gious figures take the lead in guiding politics.Yayla (2002, p. 3), for example, concludes that ‘inall Islamic countries decision-making is over-cen-tralized, power-sharing mechanisms are very few,civil society is extremely weak and the sponta-neous forces of society are strictly limited’.

Certainly, the form taken by Islam varies acrosstime and space. Contemporary Turkey demon-strates that tolerably democratic politics can be

62 FOUNDATIONS

Table 4.2 Islam and democracy, 2001

Is the government Countries with an Non-Islamic elected by Islamic majority countriesdemocratic means?

Yes 11 110

No 36 35

(Total number of countries) (47) (145)

Source: Adapted from Karatnycky (2002).

achieved in an Islamic society. But even in Turkeythe combination has been difficult to sustain. Attimes, Turkey’s army has intervened to maintainthe secular vision of the country’s modernfounder, Kemal Atatürk (Altunisik and Kavli,2003). Elsewhere in the Muslim world, Islamictraditions have hindered the spread of democracy.

But we should also recognize the West’s histor-ical role in shaping the political environmentwithin the contemporary Arab and Muslimworlds. At least until 9/11, most Western powershad shown little interest in democracy promotion.Instead, Western influence has worked in threemain ways against the establishment of democracyin the East.

First, the Middle East consists of what are, ineffect, post-colonial states. As elsewhere, Europeimposed the state form on areas that had previ-ously been organized differently – mainly asprovinces of the Turkish Ottoman Empire. Withthe collapse of this Empire after the First WorldWar, the British and the French became effectivemasters of the region. Operating under a mandatefrom the League of Nations, Western powers ruledin colonial style, creating arbitrary state bound-aries, new administrative centres and a forcefulbureaucracy more concerned with internal securitythan social development. The absence of anindigenous state tradition and the post-coloniallegacy combined to inhibit the development ofstable democracies of the region.

Second, the Middle East has proved to be afulcrum of world politics where forms of govern-ment have historically taken second place tosuperpower strategy. ‘For the last two centuries,’ asBrown (1984, p. 3) observes, ‘the Middle East hasbeen more consistently and more thoroughlyensnared in great power politics than any otherpart of the non-western world.’ The region’s oilreserves, and the continuing conflict over Israel,have certainly engaged Western attention. Butwhen its strategic interests were at stake, the West– and especially the USA – showed little concernabout the internal organization of Arab states.

For example, the United States was content tobuild a relationship with the authoritarian rulers ofoil-rich Saudi Arabia, even though the presence ofAmerica troops between 1991 and 2003 in acountry containing the holy Islamic cities of Mecca

and Medina fuelled resentment throughout theMuslim world. Strategic calculation has dominatedthe West’s approach to the Middle East, asPresident George W. Bush acknowledged in aspeech to the National Endowment for Democracyin November 2003:

sixty years of Western nations excusing andaccommodating the lack of freedom in theMiddle East did nothing to make us safe –because stability cannot be purchased at theexpense of liberty.

Third, the contemporary vigour of Islam, partic-ularly in its fundamentalist forms, is in part a reac-tion against Western preeminence. As theeconomic, scientific and military superiority of theWest has become apparent, so radical and explic-itly anti-Western variants of Islam have gainedground. Some Muslims take refuge in the era longgone when Islamic civilization was indeed moreadvanced than the West, concluding that theirtask is to restore this pure culture against itsexternal desecration. Once Islamic voices articulatethis anti-Western turn, the task of importingdemocracy from its European and Americanheartlands becomes even more challenging.

With these points in mind, let us review threevaried examples of authoritarian rule in Islamiccountries: Saudi Arabia, Iran and Pakistan (Map4.2). Saudi Arabia is at the centre of the Arabworld while Iran and Pakistan are major examplesof non-Arab Muslim countries.

Saudi Arabia

Saudi Arabia, possessed of the world’s largest oilreserves, is of particular interest as a major source ofboth personnel and funding for anti-Western ter-rorism. The country also exemplifies authoritarianrule in the region, with the government led by thecautious and conservative Saud family. Advocates ofWahhabism, the dominant and puritanical strain ofIslam found in the kingdom, have been particularlyactive in promoting Islam internationally.

Saudi Arabia’s political style reflects the influ-ence of King Abdul Aziz Ibn Saud. He led theSaudi state from its inception in 1902 until hisdeath in 1953. In true patrimonial style, Ibn Saud

AUTHORITARIAN RULE 63

ran his kingdom as a gigantic personal household,using marriage as a vital political tactic. In total,he took 300 wives drawn from all the powerfulfamilies in the state. The king was in a position tocontrol the political and personal fortunes of allthe leading figures, and did so. The carrot wasused more than the stick but the ruler’s monopolyof both devices enabled him to combine rewardsfor supporters with ruthlessness towards oppo-nents (Kostiner and Teitelbaum, 2000).

Like most authoritarian rulers, Ibn Saud wasmore concerned with protecting his position thandeveloping his kingdom. Politics came beforepolicy. Such developments as education (or morerecently television and the internet) are always per-ceived as a threat by traditional rulers. Their mainconcern, after all, is to maintain the population’sdependence on their own control of wealth andpatronage. While American colonists raised the cry,‘no taxation without representation’, the oil-inspireddeal the Saudi king offered to his people was ratherdifferent: ‘no taxation, therefore no representation’.

Slowly, however, social pressures in Saudi Arabiahave begun to build. A predominantly young pop-ulation, limited by inadequate education, hasbegun to experience unemployment even in acountry which is still the world’s largest oilexporter. Internal surveillance is extensive, themedia practise self-censorship, male domination isvirtually complete and political parties are banned.In such a controlled environment, society beyondthe ageing ruling family consists largely of themosque, meaning that radical Islamic movementsprovide one of the few outlets for expressionwhich is formally separate from the state:

The religious opposition groups are the onlyones that have regular meeting places where theycan assemble and have at their disposal anetwork not fully subject to the state. The moreoppressive the regime, the more it helps the fun-damentalists by giving them a virtual monopolyof opposition.

(Lewis, 2003, p. 102)

Given the long-term alliance between Saudi’sruling family and the United States, verbal andphysical assaults on the West by youthful membersof such movements also represent implicit cri-

tiques of Saudi Arabia’s own autocratic rulers(Niblock, 2003). But Islamic terrorists have alsosought to strike at Saudi Arabia itself. One of al-Qaeda’s earliest attacks, in 1995, was against theSaudi National Guard; more recently, compoundsfor foreigners have been targeted.

So the country is in an awkward position asboth a source for and a victim of terrorist activity.With even the USA now beginning to look morecritically at the country’s funding of radical Islamicgroups beyond its borders, the foundations of theHouse of Saud are beginning to shake, if notcrumble.

Iran

Iran (formerly Persia) is our second example ofauthoritarian rule in an Islamic society. Althoughthe Iranian population is predominantly Persianrather than Arab, the main contrast with SaudiArabia – and with other Muslim countries – lies inthe direct political role played by Iran’s religiousleaders. Whereas in Saudi Arabia, the royal familyand the Wahhabis coexist uneasily, Iran exemplifiesthat rarest form of authoritarian rule: theocracy.The country illustrates with exceptional clarity theclose relationship in Islam between church andstate. However, even in Iran the political authorityof the clerics (ayatollahs and mullahs) remainscontested; indeed, increasingly so.

Definition A theocracy is government by religious leaders.Although Christianity separates political and reli-gious roles, clerics play a direct political role insome other religions. In ancient Israel, forexample, God’s laws were expounded andapplied by holy men.The regime established inIran after the overthrow of the Shah in 1979 is amore recent example of theocratic rule.

Iran’s theocracy was a child of the 1979 revolu-tion, the last great insurrection of the twentiethcentury. In this revolution, Ayatollah Khomeini, a76-year-old cleric committed to Islamic funda-mentalism, overthrew the Shah of Iran. The Shah,an absolute monarch whose family had ruled thecountry since 1926, had supported Western-styleeconomic development. In reaction the revolu-

64 FOUNDATIONS

tionaries advocated a traditional Islamic republicfree from foreign domination. ‘Neither East norWest’ became the slogan. In power, the ayatollahscreated a unique Islamic state in which they gov-erned directly rather than by overseeing secularrulers.

Iran’s post-revolutionary constitution did incor-porate a directly elected presidency and assembly.Yet the real power lay with the clerics, expressed inpart through a 12-member Council of Guardianswhich certifies that all bills and candidatesconform with Islamic law. In strictly enforcing tra-ditional, male-dominated Islamic codes, the aya-

tollahs permeated society in a manner reminiscentof totalitarian regimes. The Interior Ministry stillmakes extensive use of informants while the stateemploys arbitrary arrests and even assassination asa form of control through terror. These are classicsigns of totalitarianism.

But as with many radical Islamic movements,Iran’s revolution was backward-looking, seeking torecreate the religion’s former glories. Rule by aya-tollahs has not delivered economic growth, even ina country with considerable oil reserves, and Iran’spolitics has turned into a lengthy battle betweenthe now traditional clerics and liberal reformers.

AUTHORITARIAN RULE 65

Iran

Black SeaCaspianSea

Iraq

SaudiArabia Pakistan

C h i n a

Kazakhstan Mongolia

Afghanistan

India

Thailand

Myanmar(Burma)

Vietnam

Turkey

Bangladesh

Syria

I n d o n e s i a

Map 4.2 Asia and the Middle East, highlighting states discussed in this chapter

The liberals are led by Mohammad Khatami, amoderate advocate of Islamic democracy who wasfirst elected president in 1997. In a country wheretwo-thirds of the population are under 25, itseems unlikely that religious leaders will be able toresist further reform indefinitely (Schirazi, 1998).

This division between an authoritarian establish-ment and a young population is also found inSaudi Arabia and other Islamic countries. But thebattle of generations and outlook is waged moreopenly in Iran than elsewhere, with the youngrelying on the internet, satellite television andmobile phones to circumvent official censorship.Although no counter-revolution to 1979 is guar-anteed, how the conflict of generations is resolvedin Iran will, in time, resonate through the Muslimworld, possibly providing the opening for a moredemocratic politics throughout the region.

Pakistan

Pakistan provides our final case of non-democraticrule in an Islamic country. Pakistan is located inAsia rather than the Middle East but its popula-tion is overwhelmingly Muslim.

Unlike Iran, which has a long history as a sover-eign state, Pakistan is more typically post-colonial.The country was created by the British in 1947from the Muslim provinces of colonial India. Thename ‘Pakistan’ is taken from the northwestprovinces of British India: Punjab, Afhghan,Kashmir, Sindh and Baluchistan. In 1971, thecountry’s separate eastern wing broke away toform independent Bangladesh.

In some ways, contemporary Pakistan is a curi-ously old-fashioned example of authoritarian gov-ernment. For one thing, it provides a rarecontemporary instance of military rule.Throughout its post-colonial history, military andcivilian rule have oscillated in what is a large, poor,unequal and virtually feudal state lacking the oilwealth of Saudi Arabia and Iran. The current gov-ernment, led by General Pervez Musharraf, is thefourth military regime since Pakistan was createdfrom the partition of India in 1947. It dates froma coup in 1999 which followed several years ofineffective civilian rule, including a setback in thelong-running conflict with India over Kashmir.

Together with the bureaucracy, Pakistan’s army

has long seen itself as the guardian of the nationalinterest – and the common weal is indeed aconcept remote from the workings of civilian poli-tics in the country. Money has become the corepolitical currency, with allegiances simply boughtand sold. In these circumstances, it is neither diffi-cult nor even implausible for the army to presentitself as national guardian. Pakistan provides acontinuing example of a political system in whichthe military supervises domestic politics, exerting asilent veto even when civilian rulers are nominallyin charge (Constable, 2001).

In addition, President Musharraf has courtedAmerican support to bolster his own politicalposition, a tactic associated with authoritarianrulers during the Cold War. Musharraf did notoppose the American invasion of Afghanistan. Inconsequence, America has so far had little choicebut to acknowledge Pakistan’s military regime (andits nuclear weapons), again in a manner reminis-cent of Cold War realities. As the United Statescontinues its attack on terrorism and its causes, itremains to be seen whether Pakistan’s rulinggeneral can continue to balance internal Islamicpressures against his dependence on the USA.

China in transition

Just as the Arab and Muslim worlds attractedmore Western attention after 11 September, so thePeople’s Republic of China (PRC) has drawn moreWestern interest with its emergence at the start ofthe twenty-first century as a powerful force in theglobal economy. China’s population, estimated at1.28 bn and growing at 0.6 per cent per year, isalready the world’s largest. Its increasingly openeconomy has grown fourfold since 1978 and islikely to become the world’s largest in the first halfof the century. The country is the world’s fourthlargest exporter, with particular strengths inassembly and manufacturing, resulting in massivereserves of American dollars (‘China lends whileAmerica spends’). China’s massive trade surpluswith the USA attracts increasing criticism fromAmerican labour unions. Already a regional power,the country is destined to become a world force –perhaps the world force – over the course of thepresent century (Kennedy, 1993). The world is

66 FOUNDATIONS

learning the truth of Napoleon’s observation:‘when China wakes, she will wake the world’.

Contemporary China is now rediscovering historicstrengths. As Manion (2004, p. 422) points out,

imperial China was the longest-lasting majorsystem of government in world history,enduring as a centralized state for more thantwo millennia until the fall of the Qing, the lastdynasty, in 1911.

This period established an authoritarian tradi-tion in which the emperor governed through anelaborate social hierarchy, supported by a smallbureaucratic elite, all rationalized by theConfucian philosophy of harmony and piety.Democracy has played little role in Chinese gover-nance, past or present, a fact which may haveimportant if unpredictable consequences for theworld as the country engages with a mainly demo-cratic external environment.

The seizure of power by the Chinese Com-munist Party (CCP) in 1949, following nearlyfour decades of internal upheaval, reinforced theauthoritarian traditions established under theemperors. Initially, the CCP followed the Sovietmodel of industrialization and collectivization ofagriculture. However, from the late 1950s MaoZedong followed an increasingly independentstrategy in which politics took priority, culmi-nating in the Cultural Revolution (1966–77)which reduced the country to near anarchy. It wasonly in 1978, two years after Mao’s death, that thecurrent era of economic modernization began. Allthese developments, including the contemporarycommitment to economic growth, have been initi-ated from the top and have reflected the internalpolitics of the party.

Contemporary China is of particular interest tostudents of authoritarian regimes. Together withVietnam, it has helped to define a distinctive tran-sition in which ruling communist parties loosentheir control of the economy while retaining adominant political position. The key questionfacing China’s ruling party is whether its politicalmonopoly can be sustained as the economy con-tinues to grow. The answer may be that sinceChina is still a developing economy, there remainsroom for a dominant party to oversee economicgrowth. As with communist Europe, the decisive

moment may not arrive until the party comes tobe seen as a brake on further development.

So far, at least, Mao’s successors have shown con-siderable skill in giving priority to economicgrowth while maintaining the party’s leadership ofsociety. Reformers have reduced the state’s role indirect economic production while creating a some-what more predictable legal environment for trans-actions that are not politically sensitive. Reformistpropaganda slogans have included ‘to get rich isglorious’ and even ‘some get rich first’, astonishingcontrasts to the theme of ‘politics in command’adopted during Mao’s Great Leap Forward of 1958.

China’s political system is no longer communistin any traditional sense. With the acceptance ofprivate wealth, the country has become one of themost unequal in Asia (Saich, 2004). Successfulbusinessmen are officially labelled model workersand awarded May Day medals. At the party’s con-gress in 2002, an entrepreneur even won a seat onthe party’s central committee. However, governanceremains deeply authoritarian. The CommunistParty is above the law because the party still makesthe law. Its members occupy many leading posts inthe public sector, reflecting the traditional commu-nist theory that the party should guide the state.

The key reform has been the reduction ofcentral control from Beijing, rather than the intro-duction of markets. As a result, local state officialshave gained a strong role in economic develop-ment through licensing and regulation. On theground, informal networks of power-holdersdetermine ‘who gets rich first’. These alliances arecomposed of well-placed and increasingly well-heeled men in the party, the bureaucracy and thearmy. So far, China’s transformation has involvedthe decentralization of economic, and to someextent political, power more than a shift towards aWestern market economy operating within therule of law. China is situated somewhere betweenMarx and the market, an unusual politicaleconomy with distinctly Chinese characteristics.

True, a growing number of thriving companiesoperate outside the inefficient and overmannedstate sector. However, even supposedly privatecompanies operate in a context where local polit-ical influence is crucial (Dittmer and Gore, 2001).Overseas companies arrive in China expecting tofind a clear distinction between public and private

AUTHORITARIAN RULE 67

spheres; they quickly discover that the two sectorsremain interwoven in the country’s socialistmarket economy. As ever in China, informal polit-ical connections (guanxi) remain important foreconomic success.

China’s rulers are aware of the tensions inducedby decentralization. One problem unleashed by theloosening of central political control has been anexplosion of corruption at lower levels. In the newenvironment, public employees are quick to recog-nize that their position can be turned to their ownadvantage. In one port city, a smuggling racketresponsible for about 10 per cent of China’s totalimports of gasoline was found to include thedeputy mayor, the head of customs and over 100other officials, six of whom were sentenced to deathafter the scheme was uncovered (Gong, 2002).

While the central party elite is concerned aboutcorruption, its policy often seems to consist of littlemore than exemplary punishment for the few casesthat happen to come to light. Some officials evenblame the problem on foreigners: ‘you can’t openthe door without letting in a few flies’. But suchglib assertions merely increase suspicion. Wang(2002, p. 3) notes that ‘the continuing invocationof socialist values in an increasingly capitalistsociety has only deepened cynicism, allowingneither socialist nor capitalist values to gain a firmfoothold’. Here, perhaps, is one of the party’smajor dilemmas: it can only attract members byoffering opportunities to acquire resources but thedubious manner in which these are obtainedincreases the distance between party and society.

An additional difficulty is the growing contrastbetween the richer coastal regions and the poorerinternal provinces, a division which potentiallythreatens the survival of the state itself. In the1990s, mechanization and the introduction of theprofit motive resulted in the loss of about sixmillion jobs per year from the countryside. Theresulting movement of the rural unemployed to thecities created an impoverished floating populationof about 200m migrants seeking work, a resentfulgroup which will be further enraged if the manyremaining state-owned enterprises (SOEs) are shutdown. The danger of unrest arising from reducingthe public sector too rapidly is one reason whyChina’s rulers must steer a delicate middle course,reforming at a socially acceptable rate.

In theory, China’s long-delayed entry into theWorld Trade Organization in 2001 shouldencourage further moves towards a genuinemarket economy. The entry agreement, after all,required the government to protect private prop-erty rights, separate government from enterprise,limit bureaucratic corruption and reduce the roleof the military in business (Fensmith, 2001).

Yet the withering away of the party, and thereduction of the state to the role of umpire, are stillfar distant. China’s entry to the WTO took 16 yearsto negotiate and will doubtless take even longer toimplement. In any case, the idea of ‘politics incommand’ is entrenched in Chinese history; thecountry’s rulers have traditionally acted as guardiansof stability in what is a fragmented but dynamiccountry. China’s continuing political experiment isof worldwide significance but the final outcome – ifthere is to be one – is most unlikely to be aWestern-style liberal democracy in which a marketeconomy operates under the rule of law.

Key reading

Next step: Linz (2000) is an influential andinsightful guide to authoritarian rule.

Brooker (2000) is a wide-ranging source on non-democracy. Classic works on totalitarianisminclude Arendt (1966) and Friedrich andBrzezinski (1965); Gleason (1995) is a morerecent review. For fascism, see Griffin (2004) and,for communist states, Harding (1984). Kershawand Lewin (1997) compare these two forms ofdictatorship. For twentieth-century authoritari-anism, see Perlmutter (1981 and 1997). Jacksonand Rosberg (1982) remains the key account ofpersonal rule in Africa, while Chebabi and Linz(1998) examine sultanistic regimes. On militarygovernments, see Finer (1962) and for Africaspecifically, Keih and Ogaba (2003). For militarydisengagement, consider Howe (2001) for Africa,Cottey et al. (2001) for Eastern Europe and Silva(2001) for Latin America. Owen (2000) offers auseful background on Middle Eastern politicswhile Lewis (2002 and 2003) places Islamic poli-tics in the context of September 11. On China,Saich (2004) is a reliable guide.

68 FOUNDATIONS

Why compare? 69 The risks of comparison 71Comparing institutions,societies and states 75Techniques of comparison 79Key reading 85

The goal of comparative politics is to encompassthe major political similarities and differences

between countries. The task is to understand themixture of constants and variability which charac-terizes the world’s governments, bearing in mindthe global, regional and national contexts withinwhich they operate. Given this definition of com-parative politics, the comparative approach issimply the family of strategies and techniqueswhich advance understanding within this field.

Grander definitions are of course available. Thecomparative approach can with justification beregarded as the master strategy for drawing infer-ences about causation in any area of study. Afterall, experiments and statistical analysis designed touncover relationships of cause and effect mustinvolve a comparison between observations,whether of individuals, social groups or countries.All investigations of cause and effect are by naturecomparative.

But most political research has more modestambitions than the systematic analysis of causesand effects. In the main, the challenge is todescribe and to interpret rather than to explain ina traditional scientific sense. For example, identi-fying cross-national trends in electoral reform orpolitical participation is itself a worthwhile exer-cise, involving the ability both to immerse our-selves in the politics of particular countries and toidentify broader trends. Nor is there anythinglimited about such generalized description: thereis much sense in King et al.’s observation (1994,p. 44) that ‘good description is better than badexplanation’.

We proceed in this chapter from the general tothe specific. We begin by outlining the strengths of

comparison by asking the simple question, ‘whycompare?’ We then turn to some pitfalls of com-parative research. We follow by discussing differentlevels or units of comparison – institutions, soci-eties and states – and conclude by outlining themajor techniques used in the field: case studies,focused comparisons and statistical analysis.

Why compare?

To begin, we will raise the obvious question: whatis to be gained by comparing politics in differentcountries? Why comparative politics? The answer isthat a comparative approach broadens our under-standing of the political world, leading toimproved classifications and giving potential forexplanation and even prediction (Box 5.1). Wediscuss each of these virtues in turn.

The first strength of a comparative approach isstraightforward: it enables us to find out moreabout the places we know least about. This pointwas well-stated by our predecessor Munro. In1925 he described the purpose of his book onforeign governments as aiding ‘the comprehensionof daily news from abroad’ (p. 4). This ability tointerpret overseas events grows in importance as

Chapter 5

The comparative approach

� Learning about other governments broadens ourunderstanding, casting fresh light on our homenation.

� Comparison improves our classifications of polit-ical processes.

� Comparison enables us to test hypotheses aboutpolitics.

� Comparison gives us some potential for predic-tion and control.

B OX 5 . 1

The advantages of comparison

69

the world becomes more interdependent. No onetoday can afford the insular attitude of MrPodsnap in Dickens’s Our Mutual Friend:‘Foreigners do as they do, sir, and that is the endof it.’ In any case, Munro was perhaps a shademodest in stating his purpose. Even when thefocus is on just one overseas country, an implicitcomparison with the home country providesmuch of the background.

Understanding foreign governments not onlyhelps to interpret new developments, it also helpswith practical political relationships. For instance,British ministers have a poor track record innegotiations with their European partners partlybecause they assume that the aggressive tone theyadopt in the Commons chamber will also work inEU meeting rooms. Their assumption is incor-rect, showing ignorance of the consensual polit-ical style found in many continental democracies.What works at home often fails in an awayfixture.

Similarly, American students frequently puzzleat how the British parliamentary system candeliver stable government when the prime min-ister, unlike their own president, is constantly atthe mercy of a vote of confidence in theCommons. Because American parties are sodecentralized, the tendency is to underestimate thepowers of a governing party in Britain.

Conversely, British students are so accustomedto the importance of party that they experiencedifficulty in understanding why Congress and theWhite House continue to quarrel even when thesame party controls both institutions. The generalpoint here is well-made by Dogan and Pelassy(1990): through comparison, we discover our ownethnocentrism and the means of overcoming it.

A second advantage of comparison is that itimproves our classifications of politics. Forinstance, we can group constitutions into writtenand unwritten types, and electoral systems intoproportional and non-proportional formulae. Wecan then search for the factors which incline coun-tries to one form rather than the other. Similarly,once we classify executives into presidential andparliamentary systems, we can look at the conse-quences of each. Classification is inherently com-parative and it turns what is a constant within asingle country into a variable between them,

thereby providing the raw material from whichexplanatory ventures can be launched.

The potential for explanation is the third advan-tage of a comparative approach. Comparativeresearchers seek to understand a variety of politicalsystems not just for its own sake but also to for-mulate and test hypotheses about politics.Comparative analysis enables us to develop andscrutinize such questions as: do first-past-the-postelectoral systems always produce a two-partysystem? Are two-chambered assemblies only foundunder federalism? Are revolutions most likely tooccur after defeat in war?

As these questions illustrate, an hypothesis is arelationship posited between two or more factorsor variables: for example, between electoral andparty systems, or between war and revolution.Confirmed hypotheses are valuable nor just fortheir own sake but because they are essential forexplaining the particular.

Consider, for example, one specific question:why did a major socialist party never emerge inthe United States? An obvious answer is becausethe USA was built on, and retains, a strongly indi-vidualistic culture. This explanation may seem tobe particular but in fact it is quite general. Itimplies that other countries with similar valueswould also lack a strong socialist party. It also sug-gests that countries with a more collective outlookwill be more likely to support a party of the left.These comparative hypotheses would need to beconfirmed by looking at a range of countriesbefore we could claim a full understanding of ouroriginal question about the USA. So explainingthe particular calls forth the general; only theoriesexplain cases.

Generalizations, once validated, have potentialfor prediction. Here we come to our fourth andfinal reason for studying politics comparatively.The ability to predict is not only a sign of system-atic knowledge but it also gives us some base fordrawing lessons across countries. So, if we findthat proportional representation (PR) does indeedlead to coalition government, we can reasonablypredict at least one effect of introducing PR tocountries such as Canada which still use the plu-rality method. Equally, if we know that subcon-tracting the provision of public services to privateagencies raises the quality of delivery in one

70 FOUNDATIONS

country, we can advise governments elsewhere thathere is an idea at least worth considering. Orauthoritarian rulers might look to China for clueson how non-democratic rulers can reduce theircontrol over the economy while retaining theirown hold on political power.

In this way, lesson-drawing provides somecapacity to anticipate and even to shape the future(Rose, 1994). This function of comparative poli-tics was nicely stated by Bryce (1921, p. iv):

Many years ago, when schemes of politicalreform were being copiously discussed inEngland, it occurred to me that somethingmight be done to provide a solid basis for judg-ment by examining a certain number of populargovernments in their actual working, comparingthem with one another, and setting forth thevarious merits and demerits of each.

The risks of comparison

Any approach brings its own dangers and thebreadth inherent in comparative politics brings itsown risks (Box 5.2). Again, we can proceed byexamining these pitfalls one by one.

Knowledge requirements

By definition, any comparative study involvesmore than one country. A statistical analysis, suchas an examination of the relationship between eco-nomic development and democracy, may draw oninformation from every country in the world.Clearly, the knowledge requirements for researchin comparative politics can be substantial. Onecommon solution is to form a team of researchers,each expert in at least one of the countriesincluded in the study. In this way, knowledgedeficits can be reduced.

Fortunately, the idea that knowledge require-ments increase directly with the number of cases isa misunderstanding, albeit an understandable one.In comparative research, the focus should be onthe comparison as much as the countries. Breadthis at least as valuable as depth; indeed detail canoften distract from the larger picture.

For example, discussing the general trend

towards proportional representation does not callfor in-depth knowledge of the electoral system inevery country in the study. Similarly, we can drawconclusions about democratization in LatinAmerica without understanding the intricacies ofthe transition in every case. And those who debatethe relative merits of presidential and parliamen-tary government cannot possibly read all that hasbeen written about the operation of the executivein every country where these forms of governmenthave been tried.

The task in comparative politics is not to knowall there is to know. Rather, the skill is to use spe-cialist knowledge so as to produce new and moregeneral observations. What matters is knowingwhat needs to be known – and being able to findit out.

Understanding meaning

In comparing politics across countries, we shouldremember that the meaning of an action dependson the conventions of the country concerned.When British MPs vote against their party in theHouse of Commons, their acts are far more signif-icant than when American legislators do the ‘same’thing in the less partisan Senate. To take anotherillustration, many observers from democratic

THE COMPARATIVE APPROACH 71

� By definition, comparative research demandsknowledge of more than one political system.

� The ‘same’ phenomenon can mean differentthings in different countries, creating difficulties incomparing like with like.

� Globalization means that countries cannot beregarded as independent of each other, thusreducing the effective number of cases availablefor testing theories.

� The countries selected for study may be an unrep-resentative sample, limiting the significance of thefindings.

� Any pair of countries will differ in many ways,meaning we can never achieve the experimenter’sdream of holding all factors constant apart fromthe one whose effects we wish to test.

B OX 5 . 2

The difficulties of comparison

countries were shocked by the apparent indiffer-ence with which military coups used to be greetedin developing countries. They failed to recognizethat coups had become a regular and fairlypeaceful mechanism for the circulation of elites. Ina sense, coups had become the functional equiva-lent of elections in the West and should have beencompared accordingly.

Students of comparative politics must thereforebe capable of understanding how politics is viewedin different countries. Rather than assuming thatall politicians are motivated by, say, rational self-interest, researchers must recognize that politicalacts can mean different things in different placesat different times. For example, is politics in a par-ticular country based on class, race or religion?The answer can only be discovered by under-standing how people in that country constructand interpret their political world. As Green(2002a, p. 6) puts it, ‘actors have identities, world-views and cognitive frames, informed by culture,that shape perceptions and interests’.

For some authors, interpreting how politics isconstructed or interpreted in a particular countryis no mere preliminary to a comparative project;rather, it is the stuff of comparative research itself.For instance, do politicians in a particular countryview globalization as a danger to be resisted (a viewmany French politicians claim to support) or as aspur to essential domestic reform (as withMargaret Thatcher’s administration in the UnitedKingdom). From this perspective, there is no phe-nomenon of globalization independent of the ideaspolitical actors hold about it (Hay, 2002, p. 204).

DefinitionDifferent institutions are functionally equiva-lent when they fulfil the same role within thepolitical system. For instance, legislative commit-tees, an independent media and oppositionparties can each serve as scrutineers of govern-ment, holding the actions of the executive toaccount.Thus a comparative study of this topicmight need to compare different mechanisms ofscrutiny across the selected countries (Myers andO’Connor, 1998).

The implication of this interpretive approach isthat we should be wary of starting comparative

projects with excessively scientific pretensions.Our first task – some would say our main task – isto understand politics from the viewpoint of par-ticipants in the countries concerned.

DefinitionAn interpretive approach to politics emphasizesthe importance of grasping the ideas whichpolitical actors themselves hold about theiractivity.The assumption is that political realitydoes not exist independently of people’s ideas;rather, political discourse in a particular countrylargely defines that ‘reality’ (Bevir and Rhodes,2002).

This problem of the meaning and significance ofaction is particularly important in politics becausethe activity is partly conducted through coded lan-guage (Ch.7). Were the people who attacked theWorld Trade Centre and the Pentagon murderers,martyrs or both? Were American civil rightsactivists also black militants? Were active membersof the Irish Republican Army terrorists or freedom-fighters? How such actors are described reflectsexisting political opinions, raising doubts aboutwhether we can find, or should seek, a neutral lan-guage for interpreting our subject matter.

Globalization

Globalization poses a considerable challenge tocomparative politics, understood as the comparisonof separate states. Although 191 ‘independent’countries belonged to the United Nations by 2002,in reality far fewer cases are available to the studentof comparative politics. Countries learn from,copy, compete with, influence and even invadeeach other in a constant process of interaction.Even the states which provide the units of oursubject did not develop separately; rather, the ideaof statehood diffused outwards from its provingground in Europe. As Dogan and Pelassy (1990, p.1) say, ‘there is no nation without other nations’.The major transitions of world history – industrial-ization, colonialism, decolonization, democratiza-tion – unfolded on a world stage. In that sense wehave one global system rather than a world of inde-pendent states. Green (2002a, p. 5) puts the pointwell when he says it is ‘as if national polities are in

72 FOUNDATIONS

fact cells of a larger entity with a life all its own’.The implication is that we should study this largerorganism rather than its individual cells.

Specific institutional forms also reflect diffusion.The communist model was often imposed byforce of Soviet arms, the presidential system inLatin America was imported from the UnitedStates and the ombudsman was a device copiedfrom Sweden. The development of internationalorganizations, from the United Nations to theEuropean Union, also creates a newer layer of gov-ernance to which all states must react.

Why do connections between states constitute apitfall for students of comparative politics? Theanswer is provided by Tilly (1997): comparativepolitics traditionally presumes distinct and sepa-rate units of comparison, most often states, whichcan be treated as if they are independent of eachother. That assumption was always a simplificationbut in an interdependent and even global worldsuch a presumption has become positively mis-leading. More technically, treating countries asindependent artificially inflates the sample size instatistical analysis, resulting in exaggerated confi-dence in the significance of the results obtained.

How students of comparative politics will reactto the issue of globalization is a story still tounfold. It is one thing to agree with Jackson andNexon (2002, p. 89) that it is ‘difficult to clearlydifferentiate between the subject areas of compara-tive politics and international relations’. However,it is quite another to propose a merger betweenthese two distinct fields, each with its own historyand body of knowledge. How global processesinteract with state forms is clearly an importanttheme for both disciplines. But this relationshipflows in both directions: states shape the world atleast as much as the world reshapes states.

Nor should we forget the continuing fact thatstates remains the fundamental unit of rule in alldeveloped nations and most post-colonial coun-tries. The state’s unique capacities – to shape iden-tities, regulate societies, raise taxes and wage wars– remain unmatched and largely unthreatened.

Selection bias

We turn now to a more technical difficulty incomparative politics which nonetheless has impli-

cations for all those who practise the art. Ingeneral form, selection bias arises when the choiceof what to study, or even of how to study it, pro-duces unrepresentative results. This risk inheres inany study covering only a few countries.

The danger often emerges as an unintendedresult of haphazard selection. For example, wechoose to study those countries which speak thesame language, or which have good exchangeschemes, or in which we feel safe. As a result,large, powerful countries are studied more inten-sively than small, powerless ones.

One result is that the findings of comparativepolitics are weighted toward established democra-cies, a rare form of government in the expanse ofhuman history. A virtue of designs covering a largenumber of countries is that they reduce the risk ofselection bias. Indeed, if the study covers allcurrent countries, selection bias disappears – atleast so long as generalization is restricted to thecontemporary world.

But, alas, the problem may just resurface inanother form, through an unrepresentative selec-tion of variables rather than countries. The diffi-culty here is best approached through an example.Much statistical research in comparative politicsrelies on existing data collected by governmentsand international bodies with different interestsfrom our own. But the priorities of these organiza-tions tend to be financial, economic, social andpolitical – in that order. So financial and eco-nomic variables may receive more attention thanthey justify, and politics runs the risk of beingtreated as a branch of economics.

For instance, a large body of research examinesthe relationship between economic conditions andgovernment popularity as reported in opinionpolls (Norpoth, 1996). This work has producedworthwhile findings but its sheer quantity reflectsthe availability of regular statistics about economictrends more than the intrinsic significance of thetopic for comparative politics.

A final important form of selection bias comesfrom examining only positive cases, thus elimi-nating all variation in the phenomenon we seek toexplain. Because this is a common, noteworthyand avoidable mistake, it deserves careful consider-ation. King et al. (1994, p. 129) explain theproblem:

THE COMPARATIVE APPROACH 73

The literature is full of work that makes themistake of failing to let the dependent variablevary; for example, research that tries to explainthe outbreak of wars with studies only of wars,the onset of revolutions with studies only of rev-olutions, or patterns of voter turnout with inter-views only of nonvoters.

When only positive cases of a phenomenon arestudied, comparison is avoided and the conclu-sions become limited. Comparison is needed togive variation, so that we can then consider whatdistinguishes times of war from times of peace,periods of revolution from periods of stability andabstainers from voters.

Skocpol’s (1979) study of revolutions in France,Russia and China illustrates the difficulty. Herresearch design allowed her to identify the featurescommon to her revolutions, such as the declininginternational and domestic effectiveness of the oldregime. However, she was unable to say how oftena failing regime was followed by a revolution. Byselecting on the dependent variable, she wasrestricted to positive cases. As a result, she wasunable to assess how far government ineffective-ness covaried with revolutions.

DefinitionSelection on the dependent variable occurswhen only similar (usually positive) cases of aphenomenon are selected. By eliminating varia-tion, there is no contrast left to explain; we areleft with variables that do not vary. For example,a study of countries which have democratizedsuccessfully tells us nothing about the condi-tions of successful democratization.Those condi-tions can only be identified through acomparison with failed transitions.

Note: the dependent variable is simply thephenomenon we seek to explain – for example,whether democratization succeeds or fails.

History plays a trick on us by selecting on thedependent variable through evolution. Forinstance, in our discussion of the origins of themodern European state, we noted that the emer-gence of states flowed from the requirement formonarchs to mobilize men and materials for war.

However this point tells us nothing about howmany monarchs failed in the task, with the resultthat their proto-states disappeared into the waste-bin of history (Przeworski, 1995, p. 19). Similarly,we are often told that democracy is triumphant inthe world today but few pause to ask how manydemocracies were lost en route (Linz and Stepan,1978). Survivorship bias creates a danger ofdrawing false conclusions by treating those thatcomplete the journey as representative of those thatstarted out, ignoring the casualties on the way.

Too many variables, too few countries

This is a major problem for those who conceive ofcomparative politics as analogous to the experi-menter’s laboratory, in which researchers patientlyseek to isolate the impact of a single variable. Evenwith 191 sovereign states, it is impossible to find acountry which is identical to another in allrespects except for that factor (say, the electoralsystem) whose effects we wish to detect. For thisreason, political comparison can never be asprecise as the experiments conducted in thenatural scientist’s laboratory. We just do not haveenough countries to go round, a difficulty termedthe ‘small-N ’ problem (‘N ’ is the statistician’sterm for ‘number of cases’).

To make the same point from another angle, wewill never be able to test all the possible explana-tions of a political difference between countries.For example, several plausible reasons can beinvoked to ‘explain’ why Britain and the USA weretwo of the countries most sympathetic to intro-ducing the ethos of the private sector into therunning of public services during the 1990s.Perhaps the strength of these reforms reflected theright-wing ideology of Prime Minister Thatcherand President Reagan. Or perhaps the publicsector in these English-speaking countries was vul-nerable to reform because – unlike several democ-racies in continental Europe – its structure was notprotected by the constitution.

Here we have two potential explanations forAnglo-American distinctiveness, one based on ide-ology and the other on the constitution. Bothinterpretations are broadly consistent with thefacts. But we have no way of isolating which oneof the two is decisive. Ideally, we would want to

74 FOUNDATIONS

discover whether the public sector had beenreformed in those crucial countries where just oneof the two factors applied. But at this point wetypically discover that there are no such countries;we just run out of cases.

In such circumstances, we can of course resort toasking hypothetical ‘what if ’ questions. Forexample, would public sector reform have pro-gressed in Britain and the USA even without theright-wing leadership of Thatcher and Reagan?Such counterfactuals are a useful sharpening toolwhen cases run short and Tetlock and Belkin(1996) have developed useful guidelines forjudging the plausibility of any particular counter-factual. But by definition such thought experi-ments can never be tested against the acid test ofreality.

Comparing institutions, societies andstates

When we engage in comparative politics, what isit that we compare? In one sense, the answer isobvious: comparative politics adopts the country,rather than regions, intergovernmental organiza-tions or indeed the entire world, as its core unit.But even a country focus leaves us with a choicebetween three different levels of analysis: the insti-tutions of government, the social context of poli-tics and the state as a whole (Box 5.3). Nor aredecisions about the appropriate emphasis merelytechnical; in fact, the recent history of compara-tive politics is a story of shifts in the favoured

level. So as we examine each of these three tradi-tions of enquiry, we will also be covering the intel-lectual development of the subject.

Comparing institutions

The study of governing institutions will always bea central activity of political science. As Rhodes(1995, p. 43) writes,

If there is any subject matter at all that politicalscientists can claim exclusively for their own, asubject matter that does not require acquisitionof the analytical tools of sister fields and thatsustains their claim to autonomous existence, itis of course, formal political structures.

Because virtually all countries possess an executive,legislature and judiciary, such institutions arenatural units for comparative politics.

But how exactly should an ‘institution’ beunderstood? Usually, the term refers to the majororganizations of national government, particularlythose defined in the constitution. But the conceptalso radiates outwards in two directions. The firstextension is to other governing organizationswhich may have a less secure constitutional basis,such as the bureaucracy and local government.And the second extension is to other importantpolitical organizations which are not formally partof the government, notably political parties.

However, as we move away from the heartlandof constitutionally mandated structures, so theterm ‘organization’ tends to supplant the word

THE COMPARATIVE APPROACH 75

Level Definition Examples

Institution-centred The organizations of government The balance between president and and the relationships between them Congress in the USA

Society-centred How social factors influence The impact of education on voting individual behaviour in politics behaviour

State-centred How the priorities of the state Building the welfare state; regulating forimpinge on society economic competitiveness

B OX 5 . 3

Levels of analysis in comparative politics

‘institution’. Implicitly, therefore, an emphasis oninstitutions in political analysis affirms the originsof political studies in constitutional studies andthe unique character of those organizationscharged with the task of governance.

DefinitionAn institution is a well-established body, oftenwith public status, whose members interact onthe basis of the specific roles they performwithin the organization. In the study of politics,the core meaning of an institution is an organ ofgovernment mandated by the constitution.Institutionalization is the process by whichsuch organizations become entrenched byacquiring the capacity to shape the behaviour oftheir members.

The starting point of institutional analysis isthat roles matter more than the people whooccupy them. It is this assumption that enables usto discuss presidencies rather than presidents, leg-islatures rather than legislators and the judiciaryrather than judges (Ridley, 1975). Put differently,the capacity of institutions to affect the behaviourof their members means that politics, like othersocial sciences, is more than a branch of psy-chology. In one sense, institutions are little morethan accepted rules for interaction between theirmembers. But institutions can also be conceived aspossessing a history, culture and memory of theirown, frequently embodying founding values andtraditions and often simply growing ‘like coralreefs through slow accretion’ (Sait, 1938, p. 18).They often possess legal personality, acquiringtheir own rights and duties under law.

The value of institutions in political affairs liesin their capacity to make long-term commitmentswhich are more credible than those of any singleemployee, thus building up trust. For example,because governments are less likely than theiremployees to go bankrupt, they can normallyborrow money at lower rates than are available toindividual civil servants. Similarly, a governmentcan make credible promises to repay its debt over aperiod of generations, a commitment that wouldbe beyond the reach of an individual debtor. So inand beyond politics, institutions help to gluesociety together, extending the bounds of what

would be possible for individuals acting alone(Johnson, 2001).

An institutional approach implies that organiza-tions do indeed shape behaviour. Employeesacquire interests such as defending their organiza-tion against outsiders and ensuring their own per-sonal progress within the structure. Crucially,institutions define interests. As March and Olsen(1984, p. 738) conclude in their influentialrestatement of the institutional approach:

The bureaucratic agency, the legislative com-mittee and the appellate court are arenas forcontending social forces but they are also collec-tions of standard operating procedures andstructures that define and defend interests. Theyare political actors in their own right.

Further, institutions bring forth activity whichtakes place simply because it is expected, notbecause it has any deeper political motive. When alegislative committee holds hearings on a topic, itmay be more concerned to be seen to be doing itsjob than to probe the topic itself. Much politicalaction is best understood by reference to its appro-priateness within the organization, an objectivethat is separate from the institution’s ostensiblepurpose.

Similarly, when a president visits an area devas-tated by floods, he is not necessarily seeking todirect relief operations or to achieve any purposeother than to be seen to be performing his duty ofshowing concern. In itself, the tour achieves thegoal of meeting expectations arising from theactor’s institutional position.

DefinitionThe logic of appropriateness refers to actionswhich members of an institution take to conformto its norms. For example, a head of state willperform ceremonial duties because it is an offi-cial obligation. By contrast, the logic of conse-quences denotes behaviour directed atachieving an individual goal such as promotionor reelection. March and Olsen (1996) point outthat much behaviour within institutions is gov-erned by appropriateness rather than conse-quences.

76 FOUNDATIONS

This emphasis within the institutional frame-work on the symbolic or ritual aspect of politicalbehaviour contrasts with the view of politiciansand bureaucrats as rational, instrumental actorswho define their own goals independently of theorganization they represent. Institutions are farmore than the theatre within which a politicalactor performs; they also shape the script (Peters,1999).

Comparing societies

In the 1960s and 1970s, however, the focus ofcomparative politics moved away from the insti-tutional level. Decolonization spawned manynew nations in which the formal structures ofgovernment proved fragile. The failure of newlydesigned democratic institutions to take root informer colonies led naturally to a concern withthe social foundations of democracy. Institutionsseemed irrelevant to the question, ‘why diddemocracy consolidate in the West but not inpost-colonial Africa?’ Nor were the forms of gov-ernment of particular importance in communiststates where the ruling party was the real wielderof power; in practice, ‘governing’ institutionsserved the party.

In addition, the Second World War had stimu-lated new developments in social science tech-niques (e.g. opinion surveys) which youngerscholars were keen to apply to politics. So thequalitative study of institutions lost prominence asresearchers used the comparative method in asearch for quantitative generalizations about polit-ical attitudes and behaviour.

In this shift to a more society-centred approach,Easton’s systems model of the political system ledthe way (1965a and b). Although few political sci-entists explicitly use Easton’s framework today, hiswork still forms part of the vocabulary of politicalanalysis. In particular, ‘the political system’ hasbecome a widely used phrase.

Easton conceived of politics in terms of its rela-tionship with the wider society (Figure 5.1). Hispolitical system consists of all those institutionsand processes involved in the ‘authoritative alloca-tion of values’ for society. Specifically, the politicalsystem takes inputs from society, consisting ofdemands for particular policies and expressions ofsupport. The political system then converts theseinputs into outputs – that is, into enforceablepolicies and decisions. These outputs then feedback to society so as to affect the next cycle ofinputs.

THE COMPARATIVE APPROACH 77

Figure 5.1 Easton’s model of the political system

Demands,Supports

Thepoliticalsystem

Authoritativedecisions

GATEKEEPERS

OUTPUTSINPUTS

FEEDBACK

For instance, trade unions may demand and winlegislation achieving improved health insurancefor their members. In operation, however, thispolicy may add to employers’ costs, leading theirrepresentatives to request tax relief from the gov-ernment.

Inputs from society do not automatically reachthe decision-makers. Rather, inputs are regulatedby gatekeepers, such as parties and interest groups,which can bias the system in favour of certaindemands and against others. In office, a left-wingparty may listen more to the trade unions while aright-wing administration might only hear theemployer’s voice. In essence, however, Easton’smodel views the political system as a mechanismfor converting demands from society into concretepolicies. The real driver is the inputs rather thanthe institutions. Indeed, the institutions of govern-ment are reduced to little more than a ‘black box’in an abstract diagram. In practice, Easton’s modelalso proved to be too static, premised on theachievement of equilibrium between inputs andoutputs. It offered little insight into politicalchange.

DefinitionBehaviouralism was a school of thought inpolitical science which emphasized the study ofindividuals rather than institutions.The focus wason voters rather than elections, legislators ratherthan legislatures and judges rather than the judi-ciary.The aim was to use statistical techniques todiscover scientific generalizations about indi-vidual political attitudes and behaviour.

Society-centred analyses formed part of thebehavioural revolution in politics, an approachwhich offers a useful contrast with institutionalanalysis. The central tenet of behaviouralists wasthat ‘the root is man’ rather than institutions(Eulau, 1963). People are more than badges of theinstitution they work for; they possess and indeedwill always create some freedom to define theirown role. The higher the position in an organiza-tion, the more flexibility the occupant possesses.Those at the very top can even reshape the institu-tion itself.

Organizations were not ignored altogether butthe study of assemblies, for instance, moved away

from formal aspects (e.g. the procedures by whicha bill becomes a law) towards legislative behaviour.Thus, researchers investigated the social back-grounds of representatives, their individual votingrecords and how they defined their own roleswithin the institution (Wahlke et al., 1962). In thestudy of the judiciary, too, scholars began to takejudges rather than courts as their level of analysis,using statistical techniques to assess how the socialbackground and political ideology of justicesshaped their decisions (Schubert, 1972).

The disregard of institutions by much society-centred analysis of the 1960s now seems extremebut its effect in broadening horizons represents apermanent and positive legacy for comparativepolitics. Note, for example, that the parts of thisbook examining the relationship between societyand politics take up more space than the sectionsdealing with government institutions. One legacyof society-centred analysis, and of the behaviouralmovement that went with it, is that comparativegovernment is now irreversibly embedded in com-parative politics.

Comparing states

In the 1980s, attention once more returned to thestate. ‘Bringing the state back in’ became a stirringrallying-cry in comparative politics (Evans et al.,1985). Partly, this shift reflected a belated recogni-tion that the state is the single central concern ofpolitical study. In addition, statistical and behav-ioural studies had run out of steam, becomingincreasingly technical and failing to engage withpolitical change.

Yet the new focus on the state represented morethan a return to descriptive studies of governmentinstitutions. Rather, the state as a whole, ratherthan its specific manifestations, became the levelof analysis. The focus lay not so much on institu-tional detail but on the state as an active agent,shaping and reshaping society. Thus the state pre-sents an additional level of analysis in comparativepolitics, distinct from both institutions and indi-viduals.

Where society-centred analysis saw the state asembedded in society, the state-centred approachsaw society as part of a configuration definedlargely by the state itself. The state acts

78 FOUNDATIONS

autonomously and is not just imprisoned by socialforces. In particular, the state is seen as using itsadministrative capacity and monopoly of legiti-mate force to bring about fundamental socialchanges. For example, Skocpol (1979) showedhow successful revolutionaries such as the RussianBolsheviks and Iranian mullahs used their controlof the state to produce total transformations ofsociety.

But examples are not confined just to post-revo-lutionary situations. Throughout the world, stateshave played a large role in the shift to an industrialeconomy, most recently in Asian countries such asJapan and Singapore. Such large-scale transforma-tions require a public effort going beyond thework of a particular institution or ministry. Stateshave also led the introduction of mass educationand welfare states, achievements that are againmisread if they are attributed simply to specificdepartments of state.

State-centred analysts suggest that the uses towhich public power has been put by those chargedwith its exercise cannot be understood by theroutine analysis of specific institutions. Rather, thestate itself must provide the unit of analysis. Theinterests of the state can be identified and analysedwithout declining into institutional detail, just as acountry’s ‘national interest’ is a useful guide to itsforeign policy even though that policy is in prac-tice the work of many hands.

Thus Marxists, for example, suggest that thestate performs underlying functions in supportingcapitalism which run deeper than the manifestpurpose of any particular element within it.Burnham (1982, p. 75) illustrates this perspective:

Any state fulfils three basic functions. First, itdefends the basic needs and interests of thosewho control the means of production within thesociety in question. Closely associated with thisis the second function of the state: achievinglegitimacy for itself and ensuring socialharmony. Finally, no state can survive if itcannot adequately defend itself, and the domi-nant powers in the economy and society, fromexternal attack.

When we do use states as our unit of analysis, itsoon becomes clear that their powers (if not their

functions) are a variable rather than a constant.For example, in communist countries, the statewas an overarching influence, pervading virtuallyall aspects of life. In established democracies, thestate is less dominant though even here a contrastis sometimes drawn between strong states such asFrance and Japan whose rulers are expected to leadsociety and weaker states such as the USA wherethe government is more a servant of society(Dyson, 1980). In much of the developing world,the state is less important still; its writ may notrun far beyond the capital city and a few othertowns. These contrasts in state power would againbe obscured if we concentrated solely on com-paring government institutions.

Techniques of comparison

This section turns from strategy to techniques.Comparative politics offers a wide repertoire oftechniques and here we examine three majormethods: case studies, focused comparisons andstatistical analysis (Box 5.4). These devices rangefrom intensive scrutiny of one country (casestudies) or a small number (focused comparisons)to systematic analysis of variables drawn from alarger number of countries (statistical analysis).

Case studies

A case is an instance of a more general category. Toconduct a case study is therefore to investigatesomething with significance beyond its ownboundaries. For instance, lawyers study caseswhich are taken to illustrate a wider legal prin-ciple. Physicians study a case of a particularailment because they want to learn how to treatsimilar instances in the future. So a project turnsinto a case study only when it becomes clear whatthe study is a case of.

As Scarrow (1969, p. 7) points out, case studiesmake a contribution to general knowledge of poli-tics if ‘the analysis is made within a comparativeperspective which mandates that the description ofthe particular be cast in terms of broadly analyticconstructs’. In other words, a single case shouldoffer a detailed illustration of a theme of widerinterest. For instance, we could take the United

THE COMPARATIVE APPROACH 79

States as an example of presidential government,Canada as an illustration of federalism or Sweden’sSocial Democrats as an example of a dominantparty. Thus cases can be deliberately chosen asexamples of broader phenomena.

Because case studies locate their findings in awider context, they are a tool of comparative poli-tics, even though only one example is examined.In that sense, comparative politics resemblesanthropology, a discipline where case studies ofparticular communities have nonetheless cumu-lated into a body of general knowledge (Peters,1998, p. 61).

In the absence of overarching theory, casestudies are the building blocks from which weconstruct our understanding of the political world(Yin, 2003). We usually proceed by comparingcases rather than by making deductions from firstprinciples. In consequence, much comparativepolitical analysis takes the form not of relatingcases to abstract theory, but simply of drawinganalogies between the cases themselves. Forinstance: how did the process of state-buildingdiffer between post-colonial states of the twentiethcentury and the states of early modern Europe?What are the similarities and differences betweenthe Russian and Chinese revolutions?

Reflecting this pragmatic approach, Khong(1992) suggests that much political reasoning,especially in foreign policy, is by analogy.Decision-makers and analysts look for earlier criseswhich resemble the current one, so that lessonscan be learned and errors avoided. This strategyreflects common sense but runs the risk of over-

weighting history. Just because democracy failed inLatin America in the 1960s, we should not conclude that it will do so again in the changedcircumstances of the twenty-first century. Justbecause the United States lost the war in Vietnam,we should not infer that America cannot winother wars on foreign soil. The art of comparingcases is to be as sensitive to differences as to simi-larities; and the contrasting location of the cases intime is often a major distinguishing feature.

In practice, case studies are multi-method, usingthe range of techniques in the political scientist’stoolkit. This kit includes: reading the academic lit-erature, examining secondary documents (forexample newspapers), searching for primary mate-rial (for example unpublished reports) and ideallyconducting interviews with participants and otherobservers in the country under scrutiny. In otherwords, scholars of cases engage in ‘soaking andpoking, marinating themselves in minutiae’ (Kinget al., 1994, p. 38). Case studies aim to provide adescription which is both rounded and detailed, agoal which the anthropologist Clifford Geertz(1973) famously defined as ‘thick description’.They blend well with the current emphasis oninterpretive explanation in politics.

By definition, all case studies possess broadersignificance but this added value can be acquiredin various ways. Box 5.5 outlines four types of casestudy. A case can be useful either because it is rep-resentative – a typical, standard example of a widercategory – or because it is prototypical, deviant orarchetypal.

Of these designs, the representative case is the

80 FOUNDATIONS

Number of Case- or Strategycases variable-

centred?

Case studies One Case Intensive study of a single instance with wider significance

Focused comparisons A few Case Qualitative comparison of a few instances

Statistical analysis Many Variable Quantitative assessment of the impact of variables

B OX 5 . 4

Major techniques in comparative politics

most common. It is the workhorse of case studydesigns, as useful as it is undramatic. Oftenresearchers will use their own country as a repre-sentative case. For example, researchers may beinterested in female political participationthroughout the democratic world but choose tostudy the phenomenon in their own country indetail. The home country is the research site butthe hope is that the results will contribute to abroader, cross-national debate. Of course, nocountry is exactly like any other but a collection ofrepresentative case studies will provide the rawmaterial for later distillation.

By contrast, a prototypical case is chosen notbecause it is representative but because it isexpected to become so. As Rose (1991, p. 459)puts it, ‘their present is our future’. Studying anearly example can help us to understand a phe-nomenon which is growing in significance else-where.

In the nineteenth century, for instance, theFrench scholar Alexis de Tocqueville (1835, Ch. 1)studied America because of his interest in the newpolitics of democracy. He wrote, ‘my wish hasbeen to find there [in the USA] instruction bywhich we [in Europe] may ourselves profit’. DeTocqueville regarded the United States as a har-binger of democracy and a potential guide toEurope’s own future.

More recently, Carey (1996) examined the termlimits on the tenure of Costa Rican legislators inorder to identify the possible effects of introducingsimilar limits to state assemblies in the USA. Likede Tocqueville before him, Carey realized that pro-

totypical case studies offer opportunities forlesson-drawing. Rather than speculate about thepossible effects of term limits in the United States,relevant evidence could be gathered from an earlyadopter. In this way, later adopters can learn fromthe experience of the innovator (Rogers, 1971).

The purpose of a deviant case study is very dif-ferent. It deliberately seeks out the exceptional andthe untypical, rather than the norm: for instance,the countries which remain communist, or whichare still governed by the military, or which seem tobe immune from democratizing trends. Deviantcases are often used to tidy up our understandingof exceptions and anomalies. We might ask whyIndia is an exception to the thesis that democracypresupposes prosperity. Or we could seek toexplain one of Switzerland’s several anomalies:why did such a small country adopt a federalarchitecture?

Normal science, suggests Kuhn (1970), proceedsin exactly this way, with researchers seeking toshow how apparent paradoxes can be resolvedwithin a dominant intellectual tradition. Deviantcases always attract interest and, by providing acontrast with the norm, can help our under-standing of typical examples. But since the excep-tional is always more exotic than the typical, thedanger of over-studying deviant cases is that com-parative politics turns into a collection of curios.

Reflecting the tendency for comparative politicsto move from cases to theory, rather than viceversa, some of the most important examples incomparative politics are best conceived as arche-typal cases, our final form of case study. The idea

THE COMPARATIVE APPROACH 81

Type Definition Example

Representative case Typical of the category Finland: coalition government

Prototypical case Expected to become typical USA: a pioneering democracy

Deviant case The exception to the rule India: democracy in a poor country

Archetypal case Creates the category USA: presidential government

Further reading:Yin (2003), Gomm and Hammersley (2000).

B OX 5 . 5

Some types of case study

here is that an archetypal case generates the cate-gory of which it is then taken, in a somewhat cir-cular way, as representative.

Take the French Revolution. This episodealtered the whole concept of revolution, recon-structing the idea as a progressive, modernizingforce. In this way, the French Revolution madepossible all the modern revolutions which fol-lowed. In similar fashion, the American presidencydoes far more than illustrate the presidentialsystem of government; it is the model which influ-enced all later attempts to create similar systems,notably in Latin America.

Focused comparisons

Focused comparisons fall between case studies andstatistical analysis. They are ‘small N’ studies con-centrating on the intensive comparison of anaspect of politics in a small number of countries.Most often, the number of countries is either two,a paired or binary comparison, or three, a trian-gular comparison. The emphasis is on the compar-ison at least as much as on the cases; otherwise,the design would be a multiple case study.Countries are normally selected to introduce varia-tion into the dependent variable, thus overcomingan inherent limit of the single case study.

To illustrate the technique, consider two exam-ples of the strategy, each using paired comparison.First, Kudrle and Marmor (1981) compared thegrowth of social security programmes in theUnited States and Canada. They sought to under-stand Canada’s higher levels of spending and pro-gramme development, concluding that theelements of left-wing ideology and conservativepaternalism found there were the key contrastswith the USA.

Second, a classic study by Heclo (1974) com-pared the origins of unemployment insurance, oldage pensions and earnings-related supplementarypensions in Britain and Sweden. In both coun-tries, Heclo concluded, the bureaucracy was themain agency of policy formulation. In contrast toKudrle and Marmor, and indeed to many focusedcomparisons, Heclo’s project was unusual inseeking to explain a similarity rather than a differ-ence between the countries examined.

Focused comparisons such as these have proved

to be the success story of comparative politics inrecent decades (Collier, 1991). Like case studies,they remain sensitive to the details of particularcountries and policies but in addition theydemand the intellectual discipline inherent in thecomparative enterprise. That is, the dimensions ofcomparison must be addressed, similarities anddifferences identified and some effort made toaccount for the contrasts observed.

Also, focused comparisons remain sensitive tohistory. Indeed, the format works particularly wellwhen – as with Heclo – a few countries are com-pared over time, examining how they vary in theirresponse to common problems such as the transi-tion to democracy. If it is difficult to produce apoor case study, it is virtually impossible to deliveran uninteresting report using focused comparison.

How should countries be selected for a focusedcomparison? A common strategy is to select coun-tries which, although differing on the factor understudy, are otherwise similar. This is a ‘most similar’design. With this approach, we seek to comparecountries which are as similar as possible in, say,their history, culture and political institutions, sothat we can clearly rule out such common factorsas explanations for the particular difference ofinterest to us. For instance, Kudrle and Marmor’sstudy (1981) sought to explain the contrasts insocial security programmes between Canada andthe United States, two countries which are similarin many other ways. Or we might seek to explainwhy Britain managed a more peaceful transition todemocracy than Germany, comparing two largecountries which share a European heritage.

However, even with a most similar design manyfactors will remain as possible explanations for anobserved difference and usually there will be nodecisive way of testing between them. Theproblem of too many variables and too few coun-tries cannot be sidestepped; in practice, the valueof a focused comparison lies in the journey ratherthan the destination.

A ‘most different’ design takes a contrastingapproach. Here, the object is to test a relationshipby discovering whether it can be observed in arange of different countries. If so, our confidencethat the relationship is real, and not due to bothfactors depending on an unmeasured third vari-able, will increase (Peters, 1998).

82 FOUNDATIONS

For example, Rothstein (2002) examines theevolution of social and political trust in two con-trasting democracies, Sweden and the UnitedStates, assuming that any trends shared betweenthese two very different countries should also beobservable in other democracies. In a similar way,if we were to find that the plurality method ofelection were associated with a two-party systemin the diverse group of countries employing thatmethod – including Canada, India and the UnitedKingdom – our confidence in the robustness ofthis relationship would increase. A most differentdesign is the basis of much statistical research, towhich we now turn.

DefinitionA most similar design takes similar countries forcomparison on the assumption, as Lipset (1990,p. xiii) puts it, that ‘the more similar the unitsbeing compared, the more possible it should beto isolate the factors responsible for differencesbetween them’. By contrast, the most differentdesign seeks to show the robustness of a rela-tionship by demonstrating its validity acrossdiverse settings (Przeworski and Teune, 1970).

Statistical analysis

Far from being unrelated, statistics and politics arelinked by origin. The word ‘statistics’ originallymeant the science dealing with facts about thestate. The growth of statistics formed part of therise of the state itself, particularly through thedevelopment of the census as a basis for taxingproperty and people (Fransmyr et al., 1990). Incomparative politics, statistical research is lesscommon than in the behavioural era but itremains a significant and worthwhile strand.

In contrast to the techniques reviewed so far, thestatistical approach is based on variables ratherthan cases. Specifically, the object is to explore thecovariation between variables. Some of thesefactors are usually measured on a quantitative scale(such as percentages) but it is also possible to drawup tables based on qualitative variables (forexample, whether a country employs parliamen-tary or presidential government).

In such analyses, one variable is dependent –

that which we seek to explain. The others areindependent or explanatory – those factors whichwe believe may influence the dependent variable(Box 5.6). Examples of such work in comparativepolitics include tests of the following hypotheses:

� the more educated a population, the higher itsproportion of post-materialists

� the higher a person’s social status, the greaterhis or her participation in politics

� the more affluent a country, the more likely it isto be an established democracy

� presidential government is more like to be over-thrown than is parliamentary government.

To illustrate the statistical approach, consider anexample. Figure 5.2 is a scatterplot showing therelationship between the number of members in anational assembly (the dependent variable) and acountry’s population (the independent variable).Such a plot displays all the information about

THE COMPARATIVE APPROACH 83

700

600

500

400

300

200

100

0

Figure 5.2 Population and assembly size from fig.14.1 (p. 248), showing the line of best fit and high-lighting two outliers

10 20 30 40 50 60Population (million)

Line of best fit

Size of assembly(number of members)

North Korea

Zaire

•••••••••••

••

••

••••

•

••••

••

•••••• • ••

•••

• • ••

•

•••

•

•

•

••

•

•

•

•••••

••

•

•••

these two factors for the countries surveyed.Clearly, the larger the population, the larger theassembly.

However, the content of the graph can be sum-marized more precisely. This is achieved by calcu-lating a regression line: that is, the line giving thebest fit to the data. This line, also shown in thefigure, is defined by a formula linking the vari-ables. Given such an equation, we can use thepopulation of any particular country to predict itsassembly size; indeed, these predictions can easilybe made from the regression line using a ruler.

A regression equation allows us to identifydeviant cases, known as outliers in statistics. Thegreater the difference between the predicted andthe actual assembly size for a given legislature, thegreater the need for additional explanation, thusproviding a link to case analysis. In our example,North Korea’s Supreme People’s Assembly is farlarger than would be expected for a country with apopulation of just 23 million. How can weaccount for this outlier? Many communist statesadopted large assemblies as a way of reducing anythreat they might pose to the party’s power. Suchan interpretation would provide a plausiblestarting point for further investigation.

Statistical methodology can provide precise sum-maries of large amounts of data using standardtechniques whose use can be checked by otherresearchers. But the approach carries two mainrisks. The first is that a strong correlation betweentwo variables may arise simply because bothdepend on a third, unmeasured factor. Forexample, the relationship between proportionalrepresentation (PR) and multi-party systems

might arise because both factors emerge in dividedsocieties, not because PR itself increases thenumber of parties. Or minority ethnic statusmight be correlated with low turnout simplybecause ethnic minorities are concentrated amongthe low-voting poor.

In principle, the solution to spurious correlationis simple: include all relevant variables in theanalysis, for statistical techniques can effectivelycontrol for such problems. In practice, not all therelevant variables will be known and spurious cor-relation is a continuing danger. However, even aspurious correlation may be a useful if risky basisfor prediction.

Second, even if a relationship is genuine, thedirection of causation remains to be established.Suppose we find that democracies have higherrates of economic growth than authoritarianregimes. We still face a problem of interpretation.Does the correlation arise because democraciesfacilitate economic growth or because a high rateof growth fosters a stable democracy? A case canbe made either way and by itself a correlation willnot provide the answer.

But not all statistical work is concerned with esti-mating the impact of one factor on another. Innew areas, simple counts provide a useful begin-ning. As Miller (1995) points out, asking the plainquestion ‘how many of them are there?’ is wellworthwhile. For instance: how many federationsare there? How many Arab states are democratic?What was the probability that an authoritarianregime in 1980 had become democratic by 2000?Such questions must be answered if we are toachieve our objective of comprehending the vari-

84 FOUNDATIONS

Type of variable Definition Example

Dependent variable The factor we seek to explain Party voted for

Independent variable The factor believed to influence the Social classdependent variable

Intervening variable A factor through which the independent Attitudes to party policyvariable influences the dependent variable

B OX 5 . 6

Dependent, intervening and independent variables

ability of the political world. Just as straightforwardcase studies often contribute more to comparativepolitics than elaborate attempts at theory-testing,so simple counting often offers more than sophisti-cated statistical analysis. As Peters (1998, p. 191)wisely says, ‘statistical analysis may not be every-thing but it is certainly something’.

Key reading

Next step: Peters (1998) is a thorough and judi-cious discussion of the comparative method.

Dogan and Pelassy (1990) is a challengingaccount of the comparative approach, morephilosophical than Peters but insightful even so.

For general overviews, Lijphart (1971), Keman(1993) and Mair (1996) are all worth reading.Landman (2003) usefully combines issues ofmethod with detailed examples. King et al.,(1994) contribute much to comparative researchstrategy. Green (2002b) examines the interpretiveapproach to comparative politics. For the institu-tional approach, see Peters (1999) and Lowndes(2002). For the state-centred approach, see Evanset al., (2003). Yin (2003) is the standard sourceon case studies. For focused comparisons, seeRagin (1994a) and Ragin et al. (1996). Penningset al. (1999) look at statistical methods in thecontext of comparative politics. We should alsomention Wiarda (2004), an extensive and expen-sive edited collection covering approaches to com-parative politics.

THE COMPARATIVE APPROACH 85

The relationship between politics and society hasalways preoccupied political thinkers. Governmentdoes not operate in isolation, unaffected by thesociety of which it forms part. This part locatespolitics in a broader social framework. Chapter 6

looks at the attitudes of people toward govern-ment, Chapter 7 discusses communication flowsbetween politics and society while Chapter 8examines how and whether people take part inpolitics.

Part II

POLITICS AND SOCIETY

This page intentionally left blank

The civic culture 89Political trust and social capital 90Postmaterialism 93Political culture in new democracies 96Political culture in authoritarian states 97Elite political culture 98Political socialization 100Islam and the West 102Key reading 104

The assault on the United States on September11, 2001 killed more people than any other

single terrorist attack in history. So it is natural toask what caused these events. And what motivatedthe terrorists to sacrifice their own lives in theprocess? Such questions are simple but theanswers, of course, will be complex, involvingmany factors and disciplines.

Still, we would probably want to begin with thepolitical culture of the terrorists and of others intheir network. Political culture is, after all, ‘thesum of the fundamental values, sentiments andknowledge that give form and substance to polit-ical processes’ (Pye, 1995, p. 965). So the startingpoint would be to describe the ‘values, sentimentsand knowledge’ of the terrorists, beginningperhaps with their perceptions of the UnitedStates, which gave form and substance to theirattack. In this way, political culture would help us to identify the first link in a long chain of causation.

The events of 9/11 also show the importance ofstudying political culture comparatively. A fullunderstanding of those events must surely involvecomparing the cultures of Islam and Christianity,East and West and – perhaps above all – the dis-possessed and the powerful. But before we canproceed to a discussion of what Huntington(1996) claims is a ‘clash of civilizations’, we mustexplore how political culture has been treated intraditional political science. Here, the focus hasbeen on how values, sentiments and knowledge

influence politics within rather than betweenstates. In examining political culture within estab-lished democracies, we begin with Almond andVerba’s classic account before turning to ideas thathave attracted more recent attention: trust, socialcapital and postmaterialism.

The civic culture

The classic study of political culture and democ-racy is Almond and Verba’s The Civic Culture(1963). Based on surveys conducted during1959–60 in the USA, Britain, West Germany,Italy and also Mexico, this landmark investigationsought to identify the political culture withinwhich a liberal democracy is most likely todevelop and consolidate. Their study provides ahelpful introduction to the topic.

Almond and Verba’s argument is based on a dis-tinction between three pure types of politicalculture: the parochial, subject and participant. In aparochial political culture, first of all, citizens areonly indistinctly aware of the existence of centralgovernment, as with remote tribes whose existenceis seemingly unaffected by national decisions madeby the central government. Parochial cultures havebeen rare in established democracies but elementscan be found in isolated rural communities or inthe growing number of inner city areas where gov-ernment is remote from people’s lives.

In a subject political culture, second, citizens seethemselves not as participants in the politicalprocess but as subjects of the government, as withpeople living under a dictatorship. Although wemay not associate subject cultures with democra-cies, subject attitudes may be growing amongyoung people, many of whom remain distant frompolitics even though they recognize government’simpact on their lives.

The third and most familiar type is the partici-pant political culture. Here, citizens believe both

Chapter 6

Political culture

89

that they can contribute to the system and thatthey are affected by it.

It would be natural to assume that people withparticipant attitudes are the model citizens of astable democracy. But the interest of Almond andVerba’s study rests precisely in their rejection ofsuch a proposition. Rather, the authors proposethat democracy will prove most stable in societiesblending different cultures in a particular mix theytermed the ‘civic culture’ (Figure 6.1). The idealconditions for democracy, they suggest, emergewhen subject and parochial attitudes provideballast to an essentially participant culture.

In this civic culture, many citizens are active inpolitics but the passive minority, whetherparochials, subjects or both, provides stability tothe system. Further, the participants are not soinvolved as to refuse to accept decisions withwhich they disagree. Thus the civic cultureresolves the tension within democracy betweenpopular control and effective governance: it allowsfor citizen influence while retaining flexibility forthe governing elite.

Armed with this theory, Almond and Verba setout to discover through opinion surveys whichcountries in their study came closest to theirmodel of the civil culture. Britain, and to a lesserextent the United States, scored highest. In bothcountries citizens felt they could influence thegovernment but often chose not to do so, thusgiving the government its required agility. By con-trast, the political cultures of West Germany, Italyand Mexico all deviated in various ways from theauthors’ prescription.

Like most original works, Almond and Verba’s

study attracted considerable scrutiny (Barry,1988). Critics alleged that attempting to portray anational political culture was as vague an under-taking as investigating national character. Perhapsthe research should have focused more on subcul-tures of race and class within the societies exam-ined. Nor did the authors offer a detailed accountof the origins of political culture. It is possible,after all, that citizens believe they can influencegovernment just because they can actually do so, apoint that would suggest political culture reflectsas much as it influences government. Nor did theauthors focus on the evolution of political culture,a theme which – as we will see – has characterizedmuch later discussion in this area.

Political trust and social capital

Times move on. In the decades following Almondand Verba’s study, many established democracieshit turbulent waters: Vietnam and studentactivism in the 1960s, the oil crisis of the 1970s,the anti-nuclear and ecology movements of the1980s, privatization and cutbacks to the welfarestate in the 1990s, terrorism in the 2000s.

As Almond and Verba (1980) noted in an initialupdate, such events left their mark on Westernpolitical cultures. More recent research in the areahas therefore focused on whether establisheddemocracies have suffered a decline in political andsocial trust. And the answer, in general, is that theyhave, although the fall focuses on the public’s confi-dence in the performance of democratic institutionsrather than on the principle of democracy itself.

For example, Norris (1999, p. 20) concludesthat overall public confidence in such institutionsas parliament, the civil service and the armedforces declined between 1981 and 1991 in each ofthe 17 countries she examined. Today’s disillu-sioned democrats, as Norris calls them, may becynical but they remain committed to democracy’sideals:

democratic values now command widespreadacceptance as an ideal, but at the same time citi-zens have often become more critical of theworkings of the core institutions of representa-tive democracy.

90 POLITICS AND SOCIETY

The civicculture

Stabledemocracy

Figure 6.1 Almond and Verba’s theory of the civicculture

Parochialculture

Participantculture

Subjectculture

The United States clearly illustrates the declineof trust in government. In 1964, three-quarters ofAmericans said that they trusted the federal gov-ernment ‘to do the right thing’; by 1994, at thebottom of the cycle, only a quarter did so (Figure6.2). As Wuthnow (2002, p. 59) points out, muchof this decline was brought about by specificevents such as the Vietnam War and Watergate,with partial recoveries during periods of peace andprosperity. Thus, trust recovered somewhat as theeconomy and stock market boomed in the late1990s. Despite the intelligence failings exposed bySeptember 11, faith in government received amassive short-term boost following the attacks asAmericans rallied round the flag (Brewer et al.,1993). But contemporary faith in national govern-ment remains well below the levels recorded in thelate 1950s, when Almond and Verba issued theirpositive appraisal of America’s civic culture.

The long-term trend in political trust slopesdown in other democracies too. In the UK, forexample, trust in government fell from 47 per centin 1987 to 28 per cent in 2001 (Bromley andCurtice, 2002). So both the American and theBritish ‘civic cultures’ have witnessed a shifttowards more sceptical and instrumental attitudesto politics since Almond and Verba’s study.

When we probe in more detail into these broadtrends, the picture becomes more negative.Surveys from a number of European democracies

suggest that the public places more trust in theinstitutions of law and order, such as the militaryand the police, than in the agencies of representa-tion such as parties. As Table 6.1 shows, politicalparties come bottom of the confidence league inall the countries examined. So public support forcore functions of the state remains generally highbut those mechanisms which provide democraticcontrol over rulers are rated poorly. This patternconfirms the public’s jaundiced view of democraticperformance.

What are the consequences of falling confidencein political institutions? This question has preoc-cupied Putnam (2002) who suggests that a cultureof trust oils the wheels of collective action,enabling projects to be initiated which would notbe feasible in a society where mutual suspicionprevailed. Declining faith in government is there-fore a form of political deflation, reducing thecapacity of the political system to achieve sharedgoals.

In an influential study using Italy as his labora-tory, Putnam (1993) attempted to test these ideasto show how a supportive social environmentdirectly enhances the performance of a politicalsystem. In their original work, Almond and Verbahad portrayed Italy as a country whose people feltuninvolved in, and alienated from, politics. At thetime, Italy showed strong elements of the subjectand parochial cultures. Putnam revisits Italy’s

POLITICAL CULTURE 91

80

70

60

50

40

30

20

10

01964 1966 1968 1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002

Vietnam war

Watergate

Percent

11 September 2001

Figure 6.2 Americans’ trust in the ‘government in Washington’, 1964–2002

Note: Per cent trusting ‘the government in Washington to do what is right’ just about always or most of the time.Sources: National Election Studies, University of Michigan; Brookings Institute. See also Brewer et al. (2003).

political culture, paying more attention to diver-sity within the country. He demonstrates how cul-tural variations within Italy influenced theeffectiveness of the 20 new regional governmentscreated in the 1970s. Similar in structure andformal powers, these governments nonethelessvaried greatly in performance. Some (such asEmilia-Romagna in the North) proved stable andeffective, capable of making and implementinginnovative policies. Others (such as Calabria in theSouth) achieved little. What, asks Putnam,explains these contrasts?

DefinitionSocial capital refers to a culture of trust andcooperation which makes collective action pos-sible and effective. As Putnam (2002) says, it isthe ability of a community to ‘develop the “I” intothe “we”’. A political culture with a fund of socialcapital enables a community to build politicalinstitutions with a capacity to solve collectiveproblems. Bonding social capital is sustained bynetworks of people drawn from similar back-grounds while bridging social capital bringstogether dissimilar types.

Putnam finds his answer in political culture. Heargues that the most successful regions have a posi-tive political culture: a tradition of trust and coop-eration which results in high levels of socialcapital. By contrast, the least effective govern-ments are found in regions lacking any traditionof collaboration and equality. In such circum-stances, supplies of social capital run low and gov-ernments can achieve little.

But where does social capital itself come from?How does a community establish a foundation ofmutual trust? Like Almond and Verba before him,Putnam’s answer is historical. Somewhat contro-versially, he attributes the uneven distribution ofsocial capital in modern Italy to events deepwithin each area’s history (Morlino, 1995). Themore effective governments in the north draw on atradition of communal self-government datingfrom the twelfth century. The least successfuladministrations in the south are burdened with along history of feudal, foreign, bureaucratic andauthoritarian rule. Thus, Putnam’s analysis illus-trates how political culture can be used as thedevice through which the past is taken to influ-ence the present.

The idea of social capital extends the idea of

92 POLITICS AND SOCIETY

Table 6.1 Confidence in political institutions, 2001

France Germany Sweden United Kingdom

High confidence Civil Police Police, Military,

(greater than 50%) service, Courts Police

Police,

Military,

Media

Low confidence Courts, Courts, Military, Courts,

(30–49%) Parliament Military, Parliament, Civil service,

Parliament Civil service Parliament

Very low confidence Parties Civil service, Media, Media,

(below 30%) Media, Parties Parties

Parties

Note: Ranking is based on the proportion of the sample expressing confidence in the institution. Institutions within each cell are ranked by confi-dence level.

Source: Data from Inoguchi (2002,Table 9.4b).

trust beyond its political domain into the widerfield of social relationships. Putnam suggests thatthe declining faith of Americans in their govern-ment is matched by a corresponding social change:namely, a fall in trust in other people. In theUnited States, those who give a positive responseto the survey question ‘Generally speaking, wouldyou say that most people can be trusted?’ declinedfrom 46 per cent in 1972 to 34 per cent in 1994(Wuthnow, 2002, p. 71). Such findings have ledto extensive debate about the danger of socialcapital disintegrating in the USA.

But there is little evidence suggesting that thedecline in social trust in the United States ismatched in other democracies. In a study ofpublic opinion in the member states of theEuropean Union, Newton (1999, p. 175) found‘no evidence of a general decline in social trust.On the contrary, nine of the EU twelve showhigher levels of trust in 1993 than 1976.’ InSweden, two in three people say they trust others,notwithstanding the assassination of the country’sprime minister in 1986 and its foreign minister in2003. The equivalent figure for trust in the USA,overall a more violent country than Sweden, is farlower (Rothstein, 2002, p. 320).

Within Europe, only Britain has shared theAmerican pattern of a marked decline in interper-sonal trust. The populations of both countries, itmay be significant to note, are heavy viewers oftelevision, a behaviour which reduces socialcontact and communication.

Nor do the connections between social andpolitical trust seem to be strong. Newton foundthat ‘the correlations between trust in people andtrust in government are so small that they can beignored’. Rothstein’s study of Sweden (2002, p.321) also found only ‘a weak correlation’ betweenthese concepts. Certainly, declining faith in gov-ernment is an important trend in the democraticworld. But social trust, and with it social capital,appears to be a separate issue.

Postmaterialism

One factor which helps to account for develop-ments in political culture, at both mass and elite

level, is postmaterialism. Along with the themes ofpolitical trust and social capital, this notion illus-trates how political scientists have sought to incor-porate change into their understanding of politicalculture.

From the late 1940s to the early 1970s, theWestern world witnessed a period of unprece-dented economic growth. ‘You’ve never had it sogood’ became a cliché that summarized the experi-ence of the postwar generation. This era – longbefore 9/11 and the wars resulting from it – wasalso a period of relative international peace. Acohort grew up with no experience of world war. Inaddition, the newly instituted welfare state offeredincreased security to many Western populationsagainst the scourges of illness and unemployment.

According to Inglehart (1971, 1997), thisunique combination of affluence, peace and secu-rity led to a silent revolution in Western politicalcultures. He suggests that the emphasis on eco-nomic achievement as the main priority is makingway for an emphasis on the quality of life:

in a major part of the world, the disciplined,self-denying and achievement-oriented norms ofindustrial society are giving way to the choicesover lifestyle which characterize post-industrialeconomies (Inglehart, 1997, p. 28).

From the 1960s, a new generation of postmate-rialists emerged: young, well-educated people withconcerns centred on lifestyle issues such as ecology,nuclear disarmament and feminism. Whereprewar generations had valued order, security andfixed rules in such areas as religion and sexualmorality, postmaterialists take political and reli-gious authority for granted. They give priority toself-expression and flexible rules. Postmaterialistsare elite-challenging advocates of the new politicsrather than elite-sustaining foot soldiers in the oldparty battles. They are more attracted to single-issue groups than to the broader packages offeredby political parties. A loaf of bread does not satisfypostmaterialists; it must also be wholemeal,organic and preferably hand-baked!

Based on extensive survey evidence, Inglehartshows that the more affluent a democracy, thehigher the proportion of postmaterialists within its

POLITICAL CULTURE 93

94 POLITICS AND SOCIETY

The Federal Republic of Germany boasts the thirdlargest economy in the world. Its skilled employees,based in capital-intensive factories, produce manufac-tured goods for sale at premium prices throughoutthe world. Germany’s exports have exceeded itsimports each year since 1955 and the country’s ‘eco-nomic miracle’ supports extensive public services;both contributed to the growth in the legitimacy ofthe Federal Republic founded in 1949 on the ashes ofHitler’s Third Reich. Only in the later 1990s, as unem-ployment passed the four million mark, and the costsof reunification began to mount, did the German eco-nomic model and its generous welfare system beginto experience stress.

Germany has been a leading player in the develop-ment of the European Union. Its political commitmentto a united Europe is entrenched in its constitutionand Germany has been willing to support its objec-tives with hard cash. Because Germany naturally viewsEuropean developments through the lens of its ownsystem of government, the country’s political institu-tions are of continental significance.

Seeking to avoid the political instability of the WeimarRepublic (1919–33), which had contributed to the Naziseizure of power, the framers of the postwar constitu-tion made the chancellor the key figure in the newrepublic.The chancellor determines governmentpolicy, appoints cabinet ministers, heads a staff of 500and can only be removed from office when parlia-ment also demonstrates a majority for a named suc-cessor.Within a parliamentary framework, Germanyoffers a distinctive form of ‘chancellor democracy’.

Most of the republic’s six chancellors have beenstrong figures, further enhancing the status of theoffice. For example, Helmut Kohl (Chancellor, 1982–98)was the dominant force behind the rapid but peacefulunification of Germany following the dramaticopening of the Berlin Wall in 1990.Though unificationhas proved costly for the Western provinces, andrequired massive restructuring in the former GermanDemocratic Republic, the result is the largest countryin Western Europe – and a state which is strategicallypositioned at the heart of the European continent.

For students of political culture, the country’s postwarhistory shows how the legitimacy of a political systemcan flow from successful economic performance.Between 1959 and 1988 the proportion of Germansexpressing pride in their political institutionsincreased from 7 to 51 per cent. Over a similar period,support for a multiparty system grew from 53 to 92per cent.This success story has not been repeated soeasily in the former East Germany, and some recentreductions in welfare have dented the previously cosycompact between the government and its citizens inthe Western provinces. Even so, the emergence of asupportive public in the Federal Republic since 1949offers hope to other transitional countries seeking tobuild a democratic culture on an authoritarian history.

Population: 82m.Gross domestic product per head:

$26,600 (much lower in former EastGermany).

Form of government: a constitutionaland parliamentary federal republic.

Legislature: the 603-memberBundestag is the lower house.Thesmaller and weaker upper house, theBundesrat, represents the 16 federalLänder (states).

Executive: the Chancellor leads acabinet of between 16 and 22 minis-

ters. A President serves as ceremonialhead of state.

Judiciary: Germany is a state based onlaw.The Federal Constitutional Courthas proved to be highly influential asan arbiter of the constitution and,externally, as the model of a constitu-tional (rather than an American-stylesupreme) court.

Electoral system: the Bundestag iselected through an influential addi-tional member system which has nowbeen adopted in over 20 countries

(p. 148). Members of the Bundesrat arenominated by the Länder; hence, theBundesrat is never dissolved.

Party system: the leading parties arethe SPD (Social Democrats) and theCDU (Christian Democrats).Traditionally, these have governed incoalition with the smaller FDP (FreeDemocrats). However, the Greensjoined a coalition with the SPD in1998, a coalition which was renewedafter the 2002 election.

Profile G E R M A N Y

Further reading: Conradt (2001), Helms (2000), Schmidt(2003).

borders. The United States was in the vanguard. Inthe early 1970s, American postmaterialists wereconcentrated among yuppies – young, upwardlymobile urban professionals, especially those in thewealthiest state of all, California. Three decadeslater, this baby-boom generation retains a rela-tively progressive outlook despite its unparalleledaffluence.

In Europe postmaterialism came first to, andmade deepest inroads in, the wealthiest democra-cies such as Denmark, the Netherlands and WestGermany. The affluent Scandinavian countries(except Norway) also proved receptive to thesevalues (Knutsen, 1990). Postmaterialism remainsless common in poorer democracies with lowerlevels of education: for example, Greece.

DefinitionPostmaterialism is a commitment to radicalquality of life issues (such as the environment)which can emerge, especially among the edu-cated young, from a foundation of personal secu-rity and material affluence. Postmaterialistsparticipate extensively in politics but they areinclined to join elite-challenging promotionalgroups rather than traditional political parties(Inglehart, 1997).

If other things remain equal, postmaterial valueswill become more prominent. When Inglehartbegan his studies in 1970–71, materialists out-numbered postmaterialists by about four to one inmany Western countries. But by 2000 the twogroups were much more even in size, a majortransformation in political culture. Populationreplacement will continue to work its effect. AsInglehart (1999, p. 247) notes, ‘as the youngerbirth cohorts replace the older more materialistcohorts, we should observe a shift towards thepostmaterial orientation’.

The unerring expansion of education gives post-materialism a further boost. Experience of highereducation is the best single predictor of a postma-terial outlook. Indeed ‘postmaterialism’ can belargely understood as the liberal outlook inducedby degree-level education, especially in the artsand social sciences. Such values are then sustainedthrough careers in professions where knowledge,rather than capital or management authority, is

key (Farnen and Meloen, 2000). The march ofhigher education, and the changing structure ofthe workforce, may be more important mecha-nisms of cultural change than the factors Inglehartemphasizes, namely peace and affluence.

Although postmaterialism is normally inter-preted as a value shift among the population, itsmost important effects may be on political elites.Inglehart’s infantry are an active, opinion-leadinggroup and already his shock troops have movedinto positions of power, securing a platform fromwhich their values can directly affect governmentdecisions.

For instance, the 1960s generation retainedtouches of radicalism even as it secured the seduc-tive trappings of office. Thus, Bill Clinton (born1946, the first president to be born after the war)offered a more liberal agenda to the Americanpeople than did his predecessor in the WhiteHouse, George Bush (born 1924). These two menbelonged to different parties, to be sure, but theyalso represented contrasting generations. A similarclaim can be made about Britain by comparingTony Blair (born 1953) with his predecessor JohnMajor (born 1943).

However, the political success of George W.Bush (born, like Clinton, in 1946) reminds usthat postmaterialism may not carry all before it. In the short and medium term, direct politicaldevelopments exert more influence over the cultural mood than long-term forces such as post-materialism.

Political culture in new democracies

In new democracies, political culture offers lesssupport to the system of government than is thecase in established democracies. In part, this weak-ness derives from mere unfamiliarity with a neworder. New rulers lack the authority which accruesnaturally to a regime with a record of success. Atthe same time, they must confront the excessiveexpectations initiated by the overthrow of the oldrulers; public opinion may expect too much, tooquickly and above all too easily. In addition, theinheritance from the old regime is likely to be acynical and suspicious attitude to politics, and aculture which is more parochial than participant.

POLITICAL CULTURE 95

The communist legacy, in particular, is farremoved from Almond and Verba’s civic culture.

Since new democracies lack a reservoir of good-will built up over generations, attitudes to thepolitical system depend more on current perfor-mance. A new democracy which literally deliversthe goods will engender supportive attitudescapable of sustaining it in the future. As Diamondand Lipset (1995, p. 751) write, ‘for the long-runsuccess of democracy, there is no alternative toeconomic stability and progress’.

Two examples of successful democratic consoli-dation, Germany and Spain, confirm this point.West Germany’s success in translating its eco-nomic miracle of the 1950s and 1960s intofavourable attitudes to its new democracy was tes-tament to the power of economic performance toreshape political culture (see the profile on p. 94).

In a similar if less dramatic way, the consolida-tion of Spain’s new democracy in the final quarterof the twentieth century owed much to economicdevelopment which had been inhibited – thoughnot prevented – under the authoritarian rule ofGeneral Franco. Economic and political liberaliza-tion marched hand in hand. So in both Germanyand Spain, strong economic performance con-tributed to the consolidation of a democraticculture.

Compared to these success stories, post-commu-nist and post-colonial regimes have experiencedmore difficulty in delivering the performanceneeded to strengthen democratic commitment.Consider the post-communist countries. In thegiddy moment of revolution in 1989, expectationsran away with themselves. With a measure offreedom achieved, the people expected affluenceto follow, seemingly unaware of the magnitude ofchange needed to transform an inefficient state-run economy into a vibrant free market.Anticipating effortless wealth, many in EastEurope simply encountered long-term unemploy-ment. As de Tocqueville noted long ago (1856),dashed expectations are politically more damagingthan outright fatalism.

Initial reserves of goodwill were soon runningshort as factories closed, infrastructure decayedand the welfare safety net provided by commu-nism disappeared. Far from delivering affluence,the new democracies seemed to deliver insecurity

and poverty to many, and illicit wealth only to afew. As one frustrated politician said, ‘Whenpeople had security, they wanted freedom; nowthey have freedom, they want security.’

By the first decade of the twenty-first century,however, post-communist states had clearlydiverged in the development of their political cul-tures. Most East European countries, for examplePoland, had witnessed considerable economicrecovery after the initial meltdown of the early andmid-1990s. Total production at last exceededlevels achieved under communism. The popula-tion began to learn that over the longer term,democracy could deliver an improved standard ofliving for most, if not all, the people. Prospectivemembership of the European Union offered afurther stimulus to embracing a democraticculture.

In post-communist Asia, however, politicalleaders saw no reason to imitate Western models.Primitive pre-industrial economies declined ratherthan developed and authoritarian rule seemed thesurest guarantee of political stability. In any case,even the pre-communist heritage simply offeredweak foundations on which to build a democracy;Tismaneanu (1995) describes the political culturesof the Asian republics, such as Kazakhstan andUzbekistan, as ‘antidemocratic, anti-liberal andethnocentric’. With Russia and China as powerfulneighbours, regional politics in Asia were also lessfavourable to constructing a democratic culturethan was the case in East Europe. In such circum-stances, it would have required an astonishing eco-nomic transformation to induce a democraticorientation in either the political elite or thegeneral population. Such agricultural societies mayneed a generation of industrialization, quite pos-sibly state-led, before a democratic culture canemerge.

Political culture in authoritarian states

In the mature democracies of the West, politicalculture contributes to the stability of government,offering broad support to those charged with thetask of ruling. Authoritarian rulers, by contrast,face characteristic problems arising from theirunwillingness to confront the challenge of the

96 POLITICS AND SOCIETY

ballot box. Lacking the legitimacy which flowsfrom free election, such rulers must find otherways of responding to the political culture of thesocieties they govern. Broadly speaking, theiroptions are threefold: to ignore, to manipulate orto seek to transform the existing political culture.Each approach merits separate discussion eventhough in practice non-democratic rulers oftencombine these strategies.

Ignoring political culture

Disregarding the political culture of the widersociety is the tactic favoured by most authoritariangovernments. Military rulers, for example, ride topower on a tank and show little concern for theniceties of political culture. Their task is to protecttheir own back against challengers seeking to sup-plant them. Far from seeking to draw supportfrom the wider culture, military rulers typicallyseek to isolate the mass population from engage-ment with government, thus shrinking the polit-ical arena.

In extreme cases, tyrants (civil or military)demand the submission of the populace, not itssupport. Yet it is a tribute to the power of politicalculture that such repressive survival strategiesrarely succeed over the long term. In practice,naked power only prospers when wrapped in legit-imacy’s clothes.

Manipulating political culture

The second approach is to exploit the politicalculture by selectively emphasizing its authoritarianelements. This strategy can be more effective overthe long term. As Eckstein (1998a) remarks, anauthoritarian government which is congruent withcultural values may prove to be more stable than ademocratic regime which remains unnourished bythe wider culture.

For instance, traditions of deference, and of per-sonal allegiance to powerful individuals, are a cul-tural resource which many leaders in Asia andLatin America have exploited to the full to sustaintheir power. Loyalty to the national leader is pre-sented as a natural outgrowth either of the submis-sion of the landless peasant to the powerfullandowner or of the unforced obedience of the

child to its parent. The ruler is father and/or chiefpatron to the nation, providing security and sta-bility but not democratic accountability.

In pre-democratic Mexico, for example,scholars suggested that ‘underlying values werefundamentally authoritarian in the sense thatMexican children learned in the family to acceptthe authority of their fathers and they later trans-ferred this acceptance of authority to politicalleaders, including the president’ (Turner, 1995,p. 209).

Seeking to transform political culture

The most interesting approach to political culturein non-democratic regimes is to seek to reshapethe country’s values. By definition, totalitarianregimes sought to transform the political values oftheir subjects. In Hitler’s Germany, for instance,all textbooks had to conform to Nazi ideology andpupils were trained in arithmetic using examplesbased on ‘the Jewish question’.

But it was communist regimes which made themost systematic and long-lasting effort at trans-forming political culture. Their starting point wasthat the state must restructure the way peoplethink and behave. As Meyer (1983, p. 6) com-ments, communist revolutions were originallyintended as cultural revolutions. Through educa-tion and persuasion, the aim was to create a newcommunist personality which would flourish in aclassless, atheist society, free of the poisons inhaledunder capitalism.

Take the Soviet Union and China as examples.In both countries the post-revolutionary commu-nist rulers sought to increase mass participation inpolitics. Mass campaigns ensured that everyonebecame involved in politics. Yet the anticipatedtransformation of political culture never cameabout. Eventually, mass participation took on apurely ritual form, based on passive obedience topower rather than active commitment to commu-nism. Fear created citizens who outwardly con-formed but in reality adopted strategies designed toensure their own survival: two persons in one body.

So communist reconstructions of politicalculture rarely succeeded in transforming mightinto right. Instead, they often strengthened criticalattitudes to politics and depleted social capital by

POLITICAL CULTURE 97

creating a social environment in which no onecould be trusted. This negative climate continuesto hold back the development of participatory cul-tures in post-communist countries.

Elite political culture

Although the impact of mass political culture onpolitical stability has been debated widely, the sig-nificance of elite political culture has beenaddressed less often. Yet in countries with aparochial or subject political culture, elite politicalculture is primary. Even where mass attitudes topolitics are well-developed, as in consolidateddemocracies, it is still the views of the elite whichexert the most direct effect on political decisions.As Verba (1987, p. 7) writes, the values of politicalleaders can be expected to have both ‘coherenceand consequences’. Political leaders have, forexample, proved central to recent democratic tran-sitions. In this section, we examine elite politicalculture, again focusing on its consequences forpolitical stability.

Elite culture is far more than a representativefragment of the values of the wider society.Throughout the world the ideas of elites are dis-tinct from, though they overlap with, the nationalpolitical culture. For example, leaders in democra-cies generally take a more liberal line on social andmoral issues. Stouffer’s (1966) famous survey ofAmerican attitudes to freedom of speech, con-ducted in 1954, confirmed this point. Stouffershowed that most community leaders maintainedtheir belief in free speech for atheists, socialists andcommunists at a time when the public’s attitudeswere much less tolerant. By the 1980s, surveysrevealed a striking increase in the Americanpublic’s support for free speech (Weissberg, 1998).Nonetheless, it was crucial to the cause of freespeech in the United States during the 1950s thata majority of the political elite resisted the strongpressure from Senator Joe McCarthy’s populistanti-communist witch-hunt (Fried, 1990).

In a similar way, many leaders of post-commu-nist countries accept the need for a thorough tran-sition to a market economy even while massculture continues to find reassurance in theequality of poverty practised under communism.

DefinitionElite political culture consists of the beliefs, atti-tudes and ideas about politics held by those whoare closest to the centres of political power.Thevalues of elites are more explicit, systematic andconsequential than are those of the populationat large.

One reason for the liberal and sophisticatedoutlook of political leaders is their education; inmost democracies, politics has become virtually agraduate profession. The experience of higher edu-cation nurtures an optimistic view of humannature, strengthens humanitarian values andencourages a belief in the ability of politicians tosolve social problems (Farnen and Meloen, 2000).Indeed the contrast between the values of the edu-cated elite and the least educated section of thepopulation is itself a source of tension in manypolitical cultures.

In assessing the impact of elite political cultureon stability, three dimensions are crucial:

� Does the elite have faith in its own right torule?

� Does the elite accept the notion of a nationalinterest, separate from individual and groupambitions?

� Do all members of the elite accept the rules ofthe game, especially those governing thetransfer of power? (see Figure 6.3.)

The first and perhaps most vital componenthere is the rulers’ belief in their own right to rule.The revolutions of 1989 in Eastern Europe dra-matically illustrate how a collapse of confidenceamong the rulers themselves can enforce a changeof regime. As Schöpflin (1990) points out,

an authoritarian elite sustains itself in power notjust through force and the threat of force but,more importantly, because it has some vision ofthe future by which it can justify itself to itself.No regime can survive long without someconcept of purpose.

In the initial phase of industrialization, commu-nist rulers in the Soviet Union and Eastern Europe

98 POLITICS AND SOCIETY

had good reason to believe their new plannedeconomies were producing results. But by the late1980s progress had given way to decline: industrialplanning had reached a dead end. As any remainingsupport from intellectuals faded, so party officialsbegan to doubt their own legitimacy. In the end,communist rule was toppled so easily because ithad already become enfeebled. Communist rulerswere aware that they had become a barrier to,rather than a source of, progress. Elite values hadceased to underpin the system of government. Bycontrast, economic growth continues apace in con-temporary China, sustaining the elite’s confidencein its own legitimacy.

A more recent example of elite political culturesustaining non- or semi-democratic government isrule by experts. Latin America provides the bestrecent instances of this technocratic culture. The‘techno-boys’ were a cohort of European- orAmerican-trained graduates (mainly in economicsor engineering) who influenced economic policy-making in much of the continent, notably Chile,in the final decades of the twentieth century.While communist rulers were losing faith in theirright to rule, the techno-boys possessed every con-fidence in the validity of their own prescriptions.

When supported by a strong political leader,these specialists were able to impose harsh mone-tary remedies on countries where financial disci-pline had often taken second place to politicalrequirements. Because the authority of these spe-cialists derived in part from their self-belief,Centeno and Silva (1998, p. 3) were surely justified

in concluding that ‘an elite culture links all of thedifferent historical apparitions of expert rule’ onthe continent. Rule by experts provided an instancewhere the technical, depoliticized views of an edu-cated elite came to dominate the political agenda.

DefinitionTechnocracy is rule by experts, a temporary formof rule that sometimes emerges after a period ofpoor governance.The term implies rule by spe-cialists with expertise in non-political subjects,often economics and engineering.The worditself was coined by William Smyth, an engineerbased in California, who founded Technocracy,Inc. in 1919.Today, a technocrat denotes a specialist with a job or outlook that is technicalrather than political.

The second aspect of elite political culture lies inits approach to the national interest. At issue here isthe attitude of rulers to the government posts theyhold. Is public service seen as just that – a way ofserving the national interest? Some national bureau-cracies, from France to Pakistan, have seen them-selves as guardians of the nation even to the pointof protecting their country from ‘mere politicians’.Latin America’s technocrats are again an example:they assumed that their assessment of the country’slong-term economic interests should take priorityover the preferences of specific social groups.

More often in the developing world, the state isseen by its ruling elite as a seam of scarce resourcesto be mined for the benefit of the rulers, theirfamilies, their constituents and members of theirethnic group. This approach predominates whereeconomic resources are scarce and – just as impor-tant – where state institutions are weak. Bothfactors apply to many post-colonial countries. Inpost-communist countries, too, officials who sur-vived the collapse of the old order often gainedpersonally from acquiring public assets throughcorrupt privatizations.

It would of course be naive to suppose thatpoliticians anywhere are guided solely by thenational interest. However, at a minimum, elitevalues should not condone self-interested behav-iour which threatens the collective interest. Whenexposed, corruption should generate criticism, nota mere shrug of the shoulders.

POLITICAL CULTURE 99

Elite acts on national

interest

Political stability

Figure 6.3 How elite political culture affects political stability

Elite believesin its right

to rule

Elite politicalculture

Elite acceptsthe rules ofthe game

The third dimension of elite political culture,and one with obvious implications for politicalstability, is the attitudes which politicians hold tothe rules of the game. A range of possibilities existshere. Is elite competition absolute, as in dividedcountries such as Northern Ireland where gains toone side (Protestant or Catholic) were traditionallyviewed as losses by the other? Alternatively, isstrong party conflict moderated by agreement onthe rules of the competition, as in mainlandBritain, thus rendering political stability consistentwith vigorous political debate?

The consequences of these attitudes to the polit-ical game are highly significant. As an example of unmoderated conflict, consider America’sWatergate scandal (1972–73), during whichPresident Nixon’s Republican supporters engagedin illegal acts such as break-ins and phone-tapsagainst their Democratic opponents. This unhappyepisode reflected the President’s own stark view ofpolitics: ‘us’ against ‘them’. Nixon was willing todispense with the normal rules to ensure that hisenemies ‘got what they deserved’. In the USA, ofcourse, most politicians do support the rules of thegame as set out in the country’s constitution. IfNixon’s attitudes predominated among America’selite, its democracy would be far less stable.

When party or group leaders are willing to com-promise to allow the expression of other interestsand values, the prospects for political stabilityimprove. In a classic analysis, Lijphart (1977) sug-gested that just such an accommodating attitudeprevailed among group leaders in divided societiessuch as Austria and the Netherlands in the 1950sand 1960s. Then, religion still strongly dividedthese countries, with considerable physical separa-tion between the various communities. In theNetherlands, for example, the Catholics,Protestant and secular communities formed sepa-rate subcultures or ‘pillars’. The issue was howthese pillars could be integrated to form a stabledemocratic culture.

Crucially, the leaders of each pillar accepted theright of each group to a fair share of stateresources. The leaders privately agreed amongthemselves, and in private, on how to slice thenational pie. However, each group then controlledthe distribution of its own resources; so thatCatholics, for example, might give higher priority

to welfare while socialists allocated more money toeducation. In this way, a culture of accommoda-tion among the elite allowed separate and evenhostile communities to live together within theone state. Today, religious divisions have softenedbut compromise remains a key theme in Dutchpolitical culture.

Elite accommodation as a formula for managingdivided but separate communities is still relevanttoday, particularly in post-conflict situations wheredistrust between subcultures remains high. Forexample, the device might have value in Sri Lankafollowing its long period of fighting between theSinhalese majority and the minority Tamils. Elitecompromise offers a form of informal federalism,casting light on the key issue of how hostile com-munities can learn to live together even when thefoundations are too insecure to support the elabo-rate architecture of formal federation.

Political socialization

Political socialization is the means by which polit-ical culture is transmitted across the generations. Itis a universal process. All societies must find a wayof passing on the skills needed for people toperform political roles, varying from voting at anelection to governing the country. The key pointabout socialization is that it is largely an uncon-trolled and uncontrollable process. No matter howmuch rulers try, they find themselves unable todominate either its process or content. By itsnature, therefore, socialization serves to replicatethe status quo. As a result, political culturebecomes a stabilizing force, providing a majorbarrier against planned change.

DefinitionPolitical socialization is the process throughwhich we learn about politics. It concerns theacquisition of emotions, identities and skills aswell as information. Its main dimensions arewhat people learn (content), when they learn it(timing and sequence) and from whom (agents).

Learning a political culture is very different fromacquiring formal knowledge of politics asobtained, say, from this book. Political socializa-

100 POLITICS AND SOCIETY

tion is a more diffuse, indirect and unplannedprocess. It involves the development of politicalemotions and identities (what is my country? myreligion? my party?) as well as the acquisition ofinformation. Political socialization takes placethrough a variety of institutions – the family, peergroup and workplace – and is as much influencedby the context of communication as its content.For example, children’s attitudes toward politicsare influenced by their experience of authority athome and at school at least as much as by thepreachings of parents and teachers.

Most studies of political socialization are based onthe primacy model. This holds that basic politicalloyalties are formed when young. Childhoodlearning is ‘deep learning’ because it provides aframework for interpreting further informationacquired in adulthood. Core political identities aredeveloped in early childhood, when the family isthe critical influence on the child. In late childhood,these attachments are supplemented by a markedincrease in information. The main effect of adoles-cence is to refine the child’s conceptual under-standing, building on information already obtained.

These three stages of socialization – early child-hood, late childhood and adolescence – preparethe child for political participation in adult polit-ical life (Figure 6.4). Adult experiences modify butrarely transform the outlook secured when young.

Some authors use Asian cultures to illustrate thecentrality of childhood to political socialization.

The argument advanced is that strong family tra-ditions encourage a group-centred style of adultpolitics in which deference to authority places aleading role. For instance, Pye (1985) suggests that

the cornerstone of powerbuilding in the Asiancultures is loyalty to a collectivity. Out of theneed to belong, to submerge one’s self in a groupidentity, is power formed in Asian cultures.

But what is the origin of this need to belong andconform? Pye suggests that the answer lies in theexperience of childhood. The Asian child, perhapsmore than elsewhere, finds unconditional love andattention from the family. As a result, the childrespects and does not question parental authority,leading to similar deference to political rulers laterin life. This acceptance of benevolent leadership(or perhaps more accurately a reluctance to expressopen dissent) is supposedly characteristic of ‘Asiandemocracy’.

Although most research on political socializationhas focused on children, we must remember thatthe process is lifelong; basic political outlooksmature in response to events and experience, andpolitical learning does not stop at childhood’s end(Conover and Searing, 1994). Indeed we can con-trast the primacy model with an alternative recencymodel. This is the idea that current informationcarries more weight just because it is contemporary.What we see on television today matters more than

POLITICAL CULTURE 101

Political learningPolitical participation

Early Late Adolescence Young Middle Old agechildhood childhood adulthood age

high

low

Figure 6.4 Political learning and participation across the life cycle

submerged childhood memories; in other words,adult reality packs more punch than childhoodmyths. In the eighteenth century, Montesquieu(1748) observed that

we receive three educations: one from ourparents, another from our teachers, and a thirdfrom the world. The third contradicts all thefirst two teach us.

The recency approach undoubtedly carries someplausibility. Adult experiences of such majorevents as war or depression surely leave their mark,helping to shape the political outlook of a genera-tion. There can be no doubt, for example, thatSeptember 11, 2001 coloured the approach ofAmerican politicians to subsequent foreign policydecisions, just as the scars of Vietnam discouragedthe previous generation of leaders from militaryadventures. But crises such as 9/11 are exceptional.Routine events come and go but most often theyare viewed through – rather than overriding –established perspectives, notably those acquiredduring childhood.

Islam and the West

To conclude our discussion, we examine the valueof political culture in understanding the relation-ship between Islam and the West. Such a compar-ison has acquired added significance since 9/11but also possesses two further advantages. First, ittakes us beyond the state towards a more globalperspective. Second, the comparison raises thetopic of religion, a dimension of culture that wehave not so far addressed. To what extent, then,should the current division between the Muslimand the Western worlds be viewed as a conflict ofpolitical cultures?

Huntington (1996) is the main proponent of theproposition that the division between Islam andthe West is cultural or, to use his term, civiliza-tional. Huntington’s view derives from a broaderanalysis, published before September 11, 2001, inwhich he suggests that cultures based on civiliza-tions rather than countries will become the leadingsource of conflict in the twenty-first century.

According to Huntington, the old base of inter-

national conflict between the communist SovietUnion and the free market United States isexhausted. But the end of the cold war, he sug-gests, does not mean the end of cultural divisions.Rather the focus will shift from a battle of ideolo-gies to a clash of civilizations, an irreducible divi-sion between the world’s major cultures, includingIslam and the West (Box 6.1). Since such group-ings are supranational, Huntington (1996, p. 20)implies that political culture has escaped itsnational moorings to embrace wider but still com-peting identities:

A civilization-based world order is emerging:societies sharing cultural affinities cooperatewith each other; efforts to shift societies fromone civilization to another are unsuccessful; andcountries group themselves around the lead orcore states of their civilization.

Between the contradictory worldviews of thesecivilizations, suggests Huntington, there is littlecommon ground or room for compromise. Asglobalization proceeds, interaction and frictionwill intensify, producing a high potential for con-flict. He notes, for example, how cultural kinshipinfluences the choice of sides in recent wars: ‘inthe Yugoslav conflicts, Russia provided diplomaticsupport to the Serbs . . . not for reasons of ideology or power politics or economic interestbut because of cultural kinship’ (1996, p. 28).Huntington is also sceptical of pragmatic efforts toswitch civilizations, suggesting that the reasonAustralia failed to reinvent itself as an Asiancountry was simply that it’s not.

In discussing the specific relationship betweenIslam and Christianity, Huntington draws onthese civilizational themes. The transnational char-acter of civilizations is indeed exemplified by thesereligions, each of which pre-dates the emergenceof states. Thus, in medieval Europe, Christendomstood above kingdoms in the political hierarchy.Similarly, contemporary Muslim countries forman Islamic domain in which a shared religiouscommitment transcends national divisions.

Although the origins of the conflict betweenIslam and the West may lie in religion,Huntington argues that the contemporary divisionis cultural or civilizational rather than religious.

102 POLITICS AND SOCIETY

With the Islamic world falling ever further behindthe West in science, technology and wealth, it isno longer the West’s Christian foundations, butrather its secular character as exemplified byAmerican materialism, which has become thetarget of Islamic criticism.

As would be expected for a civilizational divide,differences in socialization underpin and perpet-uate these cultural differences. Western educationis avowedly secular, allowing schooling to concen-trate on scientific knowledge and technicaltraining. But in many Muslim countries, literalinstruction in the Koran (Islam’s holy text)remains a major theme of education, ill-preparingyoung people – male as well as female – for themodern world.

The upshot is that Huntington (1996, p. 217)portrays Islam and the West as civilizations lockedin permanent and seemingly unavoidable conflict:

The underlying problem of the West is Islam, adifferent civilization whose people are convincedof the superiority of their culture and are obsessedwith the inferiority of their power. The problemfor Islam is the West, a different civilizationwhose people are convinced of the universality oftheir culture and who believe that their superior,

if declining, power imposes on them the obliga-tion to extend that culture throughout the world.

Although Huntington’s analysis is clear andstraightforward, its sweeping character also pro-vides its weakness. Indeed, the widespread criti-cism of his discussion helps to identify the limitsof political culture as a tool for political analysis.Huntington’s work is pitched at an extremelygeneral level, showing insufficient sensitivity tovariations within civilizations. Just as Almond andVerba underplayed subcultures within the coun-tries they surveyed, so Huntington discounts vari-ation within civilizations across space and time.

Stepan (2001, p. 234) is surely closer to themark when he interprets Islam as ‘multivocal’,capable of varying its voice across place and time.In similar fashion, Feldman (2000) describes Islamas a ‘mobile idea’. Thus, both Turkey and SaudiArabia are Muslim countries but Turkey’s state issecular and largely democratic whereas SaudiArabia is an authoritarian regime leading a societydominated by a severe form of Islam. The reactionto September 11 confirms Islam’s multivocal char-acter: the hijackers undoubtedly drew on one anti-Western dialect within Islam but most Muslims,like most Christians, regarded the attacks asmorally unjustified (Saikal, 2003, p. 17).

So, like any other dimension of political culture,such as America’s commitment to freedom or theFrench idea of equality, Islam is a resource whichcan be used and developed in innumerable waysaccording to political circumstances to whichHuntington devotes insufficient attention. Inpolitical analysis, we should avoid an ‘essentialist’reading of Islam focused on its inherent character-istics which assumes all believers speak with onevoice (Lane and Ersson, 2002, p. 158).

Over time, too, the nature of Islam, and its rela-tionship with the West, has varied. The potentialfor conflict with the West may be inherent but thispotential often remains latent. Saikal (2003, p. 24)writes that ‘since the advent of Islam in the earlyseventh century, relations between its domain andthe largely Christian West have been marked bylong periods of peaceful coexistence but also bymany instances of tension, hostility and mutualrecrimination’. As long as civilizations are con-ceived as static, it is difficult to account for vari-

POLITICAL CULTURE 103

Huntington defines civilizations as the broadest cul-tural entities in the world; they are ‘cultures writlarge’. He divides the world into six or seven majorcivilizations:

1. Western2. Japanese 3. Islamic 4. Hindu5. Slavic–Orthodox 6. Latin American 7. (possibly) African

Such divisions pose special problems for torn coun-tries located on the fault-lines between civilizations.Mexico (situated between the West and LatinAmerica) and Turkey (on the border between theWest and Islam) are examples of such split countries.

B OX 6 . 1

The clash of civilizations?

ability in the relationship between them.Huntington’s expansive claim that the West’sproblem is ‘not Islamic fundamentalism but Islam’does indeed involve a breathtaking dismissal ofentire centuries:

The causes of this ongoing pattern of conflict[between Island and the West] lie not in transi-tory phenomena such as twelfth centuryChristian passion or twentieth century Muslimfundamentalism. They flow from the nature ofthe two religions and the civilizations based onthem.

(Huntington, 1996, p. 210)

Rather than regarding Muslim fundamentalismas ‘a transitory phenomenon’, we should seek tolocate its emergence in the events of the twentiethcentury, an approach which takes us away frompolitical culture towards more specific themes inpolitical history. Brzezinski’s list (2002, p. 18), forinstance, would probably be acceptable to many:

Arab political emotions have been shaped by theregion’s encounter with French and British colo-nialism, by the defeat of the Arab effort toprevent the existence of Israel and by the subse-quent American support for Israel and its treat-ment of the Palestinians, as well as by the directinjection of American power into the region.

So political culture (or equivalent terms such ascivilization) can only take us so far. Culture identi-fies the general climate but fails to offer specificforecasts. As Roy (1994, p. viii) observes, ‘culture

is never directly explanatory and in fact concealsall that is rupture and history: the importation ofnew types of states, the birth of new social classesand the advent of contemporary ideologies’. Byitself, terms such as ‘political culture’ and ‘civiliza-tion’ are blanket explanations, offering wide cov-erage but also obscuring the intricate detailunderneath.

Key reading

Next step: Pharr and Putnam (2000) is animportant comparative study of political culture inestablished democracies, focusing on increasing dis-satisfaction with the performance of government.

Norris (1999a) is a useful supplement to Pharrand Putnam while Crothers and Lockhart (2002)is a wide-ranging reader on the concept of politicalculture. Almond and Verba (1963, 1980) remain auseful starting point; see also Lane and Ersson(2002). For social capital, see Putnam (1993);Putnam (2002) is an interesting comparativeattempt to apply the concept beyond the UnitedStates. The key source on postmaterialism isInglehart (1971, 1990, 1997). On political culturein post-communist Europe, Pollack et al. (2003) isan extensive, survey-based collection. Haynes(1998) is an accessible introduction to religionand politics; Madeley (2003) is a comprehensivereader on the topic. Huntington (1993, 1996) isthe main source on the clash of civilizations butsee also more grounded works such as Saikal(2003) or even, perhaps, Lewis (2003).

104 POLITICS AND SOCIETY

Development of the mass media 106The media: contemporary trends 108The media in established democracies 112The media in new democracies 115Public opinion 116The media in authoritarian states 119Key reading 121

Society, and with it politics, is created, sustainedand modified through communication.

Without a continuous exchange of information,attitudes and values, neither society nor politicswould be possible. As Williams (1962, p. 11)writes, ‘What we call society is not only a networkof political and economic arrangements, but also aprocess of learning and communication.’ Insimilar vein, Habermas (1978) suggests thatdemocracy can best be understood as a form ofcommunication in which citizens inform, educateand become reconciled to each other in theprocess of reaching collective agreements. ForHabermas, as for many others, democracy is aform of political communication.

Because communication is central to politics,the ability to control its content, style and flow is acrucial if indirect source of power. The willingnessof people to listen to what their leader has to say isperhaps the ruler’s most important resource.Getting issues on to the agenda, and keeping themthere, is a necessary objective for any politicianseeking radical change. The additional ability toframe issues – to determine how they are discussed– adds further influence since, as Foucault (1977)remarks, those who define situations therebycontrol them.

The main business of politicians is communica-tion; their task is to signal their agendas, policiesand strategies to other players of the politicalgame. Indeed, in public affairs words often speaklouder than actions. For example, statements byAmerica’s Federal Reserve Board and its chairmanAlan Greenspan seem to be at least as influential

in determining expectations of interest rates in themoney market as the Board’s decisions on whatthe current rate should be (Beattie, 2003). Thesignals are as important as the actions they fore-shadow.

Even when politicians do act, their actionsconvey meanings that transform their behaviourinto communication. When President Trumanauthorized atom bombs to be dropped onHiroshima and Nagasaki in August 1945 he wasalso indicating to Japanese opponents his willing-ness to continue using these new weapons of massdestruction on civilian populations. The Japanesegot the message and quickly surrendered. In short,political activity is invariably a form of communi-cation and the analyst’s first task is always to inter-pret (‘decode’) the message so that its underlyingthemes (‘subtexts’) become apparent.

Once the message itself is understood, the tradi-tional ‘transmission model’ offers a useful guide toresearch (Figure 7.1). This account interprets com-munication as consisting of who says what towhom, through which media and with whateffects. The model distinguishes five aspects to anycommunication: a sender (who?), a message(what?), a channel (how?), a receiver (to whom?),and a presumed impact (with what effects?). Forexample, a local party (the sender) might dis-tribute a leaflet (the channel) advocating voting ata forthcoming election (the message) to its ownsupporters (the receivers), with the result thatturnout increases (impact).

This simple account is often criticized as one-directional – that is, it fails to recognize that mostcommunication involves continuing interaction

Chapter 7

Political communication

Sender ➨ Message ➨ Channel ➨ Receiver ➨ Impact

Figure 7.1 The transmission model of political communication

105

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 105

between the participants. The model also payslimited attention to the multiple meanings embed-ded in most political communication. Yet despitethese weaknesses the transmission model doesbreak down the process of communication into itscomponent parts. In this way, it assists research.

Development of the mass media

To understand contemporary trends in the com-munications media, we must begin by placing themass media in a historical context. This task is apolitical assignment since the expansion of masscommunication was intimately linked to thegrowth of the state, especially in the nineteenthand twentieth centuries. A history of politics mustalso be, in part, a history of communication.

Two preliminary points should be made. First,the impact of each new medium, from writingthrough printing to television, has generated morecontroversy than consensus. The impossibility ofisolating pure media effects means there is moresense in asking in what ways, rather than to whatextent, specific media have altered politics.

Second, we must beware of technological deter-minism. Technical advances are rarely the soleexplanation of innovations in communication.Typically, new media have emerged in response toa recognized social need. For instance, the tele-graph, the telephone and the internet were eachconsciously developed to improve communicationacross distance. If communication technologieshave transformed the social and political world, itis because society has been in search of change(Williams, 1989).

Table 7.1 outlines the major developments inthe history of the media. The first innovationshown there, namely writing, was the most impor-tant of all. Ong (1982) argues that the inventionof writing in the early civilizations of the NearEast in the fifth millennium before Christ changedeverything. He suggests that ‘by separating theknower from the known’, writing permitted thedevelopment of abstract concepts, objectivity andthe accumulation of detailed knowledge.

Certainly, the long-term political impact ofwriting has been profound. It permitted therecord-keeping which is a foundation of the

modern state; without writing, states could nothave penetrated society to the extent that they did.Writing also made possible the transmission ofboth information and values over large distances,as with the spread of Christianity through monas-teries and church schools in early Europe (Mann,1986, p. 335).

However, the bridge to the era of modern com-munication was provided not by writing itself butrather by Gutenberg’s fifteenth-century inventionof printing to paper using movable type. Thisinnovation enabled written knowledge to reach awider market.

Yet it was not so much the invention of writingand printing, but rather the later extension of lit-eracy to the wider population, which marked offthe current era of both communication and states.Mass literacy in a common language was central tothe successful development of contemporarystates, facilitating administration of large areas andencouraging the development of a shared nationalidentity. By the end of the eighteenth century,‘signing literacy’ had reached about 80 per cent ininnovative Sweden. Other European countries,and New England, achieved mass literacy in thenineteenth century, following the introduction ofcompulsory primary education. Mass literacy wasa function, an achievement and an affirmation ofthe modern state, just as it remains an aspirationfor many countries in the developing world today.

In particular, widespread literacy in a shared language made possible the popularization ofnewspapers, the key development in political com-munication during the nineteenth and early twen-tieth centuries. Originally, newspapers had beensmall-circulation party journals containing lengthyeditorials and dreary reports of politicians’speeches (Dooley and Baron, 2001). But advancesin printing (e.g. the steam press) and distribution(e.g. the railway) opened up the prospect of pop-ulist and profitable papers funded by advertising.By growing away from their party roots, papersbecame not just more popular but also, paradoxi-cally, more central to politics.

In compact countries with national distribution,such as Britain and Japan, newspapers built enor-mous circulations which largely continue to thisday. Newspaper owners became powerful politicalfigures. In interwar Britain, for example, four

106 POLITICS AND SOCIETY

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 106

newspaper barons – Lords Beaverbrook, Rother-mere, Camrose and Kemsley – owned newspaperswith a total circulation of over 13 million,amounting to one in every two daily papers sold.The press barons were committed to the circula-tion of their papers more than to the parties theysupported; under competitive pressure, theyreduced the political coverage in their publica-tions. But they were also willing to use theirorgans to advocate favoured and often idiosyn-cratic causes, such as Rothermere’s Anti-Waste

Campaign. Stanley Baldwin, a prime minister ofthe time, described these papers as

engines of propaganda for the constantlychanging policies, desires, personal wishes andpersonal desires of [their owners] . . . What theproprietorships of these papers is aiming at ispower and power without responsibility – theprerogative of the harlot throughout the ages.

[Curran and Seaton, 2003, p.64]

POLITICAL COMMUNICATION 107

Table 7.1 The development of communication media

Date Development

Fifth millennium BC Writing systems develop in the earliest urban civilizations of the Near East,notably Phoenicia.

About 750 BC The first modern alphabet, based on sounds rather than symbols, emerges in Greece.

1450 AD In Mainz, Johannes Gutenberg invents printing with movable metal type. Book production is aided by new techniques for the mass manufacture of paper.

Nineteenth century Compulsory primary education introduced, and mass literacy achieved, in Western states.

The telegraph – a way of sending information across wires using electric signals – permits international dissemination of news, decisions and instruc-tions (e.g. by the British empire to its colonies).

Later nineteenth century and Popular newspapers emerge, often with mass circulation. New railway networks early twentieth century allow national distribution.

1930s Radio’s golden age. For the first time, politicians broadcast directly into electors’homes.

1950s–1960s Television becomes the most popular, and usually the most trusted medium in Western countries. Election campaigns are soon fought on, and not just reported by, television. Entertainment programmes from the USA are widely exported, contributing to the diffusion of American values.

1970s–1980s The television audience begins to fragment, with an increase in the number and commercialization of channels; distribution by cable and satellite; and wide-spread use of video.

1990s Internet access reaches more affluent and educated groups in Western societies, representing a further opening of international communication.Emergence of mobile telephony.

2000s The internet reaches the mass population in Western societies. Broadband facilitates distribution of streaming media such as television, films, radio and telephone. Mobile telephony becomes standard in the West but China becomes the world’s largest single market for mobile phones.

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 107

For better or worse, control over the means ofmass communication had become, as it remains, amajor political resource.

DefinitionMass media refers to methods of communica-tion that can reach a large and potentially unlim-ited number of people simultaneously.Televisionand newspapers are the most important exam-ples; others are posters, radio, books, magazinesand cinema. An email to a friend is personal, notmass, communication, but sending a message toall members of an email list is mass communica-tion.

Although newspapers remain important chan-nels of political communication, in the twentiethcentury their role was of course supplanted by anew dominant medium: broadcasting. Cinemanewsreels, radio and then television enabled com-munication with the mass public to take place in anew form: spoken rather than written, personalrather than abstract and, on occasion, live ratherthan reported. Oral communication reasserteditself, though now in a form which could escapethe confines of the small group.

Significantly, the two major ideologies foundedin the twentieth century – communism andfascism – consciously sought to mobilize popularsupport through exploiting broadcasting. In thesebrave new worlds of total and totally controlledcommunication, mass media would be used tocreate people with the knowledge and attitudesspecified by the political elite. Such visions provedto be fanciful but the broadcast media, like news-papers earlier, did serve as agents of national inte-gration. In most countries, a small number ofnational channels initially dominated the airwaves,providing a shared experience for a dispersed pop-ulation of everything from national events topopular entertainment.

The political impact of broadcasting was imme-diate. Politicians had to acquire new communica-tions techniques. A public speech to a live audienceencouraged expansive words and dramatic gesturesbut a quieter tone was needed for transmissiondirect to the living room. The task was to conversewith the unseen listener and viewer rather than todeliver a speech to a visible audience gathered

together in one place. The art was to talk to themillions as though they were individuals.

President Franklin Roosevelt’s ‘fireside chats’,broadcast live by radio to the American populationin the 1930s, exemplified this new, informalapproach. The impact of Roosevelt’s somewhatfolksy idiom was undeniable. He talked not somuch to the citizens but as a citizen and wasrewarded with his country’s trust. In this way,broadcasting transformed not just the reach but alsothe tone of political communication (Barber, 1992).

The media: contemporary trends

In the first decade of the twenty-first century, thethree major trends in communications are com-mercialization, fragmentation and globalization(Box 7.1). The combined political effect of thesedevelopments is to reduce national politicalcontrol over broadcasting, permitting consumerseither to choose their own political programmingor increasingly to escape from politics altogether.If the mass media performed a nation-buildingfunction in the twentieth century, their emergingimpact in the new century is to splinter the tradi-tional national audience, fragmenting the sharedexperience induced by twentieth-century broad-casting and perhaps contributing thereby todeclining political participation.

108 POLITICS AND SOCIETY

Trend Definition

Commercialization The decline of public broadcastingand the rise of for-profit media

Fragmentation The increased range of channelsand an enhanced ability toconsume programmes on demand(e.g. videos)

Globalization Improved access to overseasevents and media in the globalvillage

B OX 7 . 1

Contemporary trends in masscommunication

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 108

Commercialization

Communication is increasingly treated as theimportant business it has become. The AmericanFederal Communications Commission estimatesthat worldwide revenues from the mass mediaexceeded $1 trillion per year by the start of the1990s; the figure rises each year (Tracey, 1998, p.8). Commercialization also allows new mediamoguls, such as Rupert Murdoch, to buildtransnational broadcasting networks, achieving ona global scale the prominence which the news-paper barons of the nineteenth century hadacquired at national level.

In the Americas, Canada excepted, broadcastinghad always been commercially led but in WesternEurope commercialization has been a disputedpolitical development. The first television stationsin Europe had been national and publicly owned.The British Broadcasting Corporation (BBC) wasa notable example. Such stations were controlledby public supervisory boards; were funded by aspecial licence fee; and adopted a public serviceethos which aimed to educate and inform, ratherthan merely to entertain. However in the 1970sand 1980s new commercial channels were intro-duced and advertising was added to many publicstations (though not to the BBC’s main channels).

These developments threatened previously cosylinks between parties and broadcasters and repre-sented a shift to conceiving the viewer in lesspolitical terms, as a consumer rather than acitizen. In this increasingly commercial environ-ment, Tracey (1998) claims that public servicebroadcasting has become nothing more than ‘acorpse on leave from its grave’. Even the BBC hasbecome more commercial, establishing numerousspecialist channels and creating one of the world’smost successful websites.

In Rich Media, Poor Democracies, McChesney(1999) argues that commercialization shrinks thepublic space in which political issues are discussed.Channels in search of profit will devote little timeto serious politics. What coverage there is will beinfotainment – information as entertainment.Profit-seeking media have no incentive to supplypublic goods such as an informed citizenry and ahigh electoral turnout, traditional concerns ofpublic media. For those who view democracy as a

form of collective debate, engaging most of thecitizenry in matters of common concern, mediacommercialization is a challenge indeed.

Against which, commercial broadcasters canreply that it is preferable to reach a mass audiencewith limited but stimulating political coveragethan it is to offer extensive but dull political pro-gramming which, in reality, only reaches the smallminority who are already strongly interested inpublic affairs (Norris, 2000).

Fragmentation

Consumers are increasingly able to watch, listen toand read what they want, rather than what they aregiven. Long gone are the days when Americanviewers were restricted to the three large networks(ABC, CBS and NBC) and when British viewerscould choose only between the public BBC and thecommercial ITV. Distribution of programmes bycable and satellite allows viewers to receive a greaterrange of content, local and overseas as well asnational. The advent of video recorders in the1980s introduced another distribution channel: thelocal video store. The internet allows interested citi-

POLITICAL COMMUNICATION 109

Table 7.2 Countries in which a majority of peopleuse the internet, 2002/03

Internet penetration Countries

70% or higher Sweden (76%)

60–69% Hong Kong (67%), Australia (64%), Netherlands (64%),USA (63%), Denmark (60%)

50–59% Switzerland (59%), UK (58%),South Korea (56%), Singapore (55%), New Zealand (55%),Germany (54%), Canada (53%), Finland (51%), Norway (51%)

Note: The ten countries with the most internet users are, in order,USA, China, Japan, Germany, UK, South Korea, France, Italy, Canadaand India.

Source: Data from Internetworldstats at http://www.internetworld-stats.com.

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 109

zens to access newspapers, radio and television fromanywhere in the world (Table 7.2). Unlicensedpirate stations further complicate the picture (Soley,1998), as does the emergence of mobile telephonesand email as channels of communication. Thecombined result of these developments is a moresplintered audience. During the 1990s, forinstance, the audience share of the big threeAmerican television networks dropped by a third.

The political implications of this shift frombroadcasting to narrowcasting are substantial. Themedia cease to be agents of national integration.Instead, they function to bring together groupswith narrow interests, both within and acrosscountries. Freedom of speech – including thefreedom to express racist and sexist views –emerges by default, due to the practical impossi-bility of regulating the babble of sounds emergingfrom diverse media and often delivered over theinternet. Governments and politicians encountermore difficulty in reaching a mass audience whenthe audience can defend itself through remotecontrols and sophisticated video recorders. Theelectorate becomes harder to reach, forcing polit-ical parties to adopt a greater range and sophistica-tion of marketing strategies, including the use ofpersonalized techniques such as direct mail.

In this more fragmented environment, politi-cians will have to communicate in a manner, andat a time, of the voters’ choosing. They will haveto continue their migration from television newsto higher-rated talk -shows. The sound-bite, neverunimportant, will become even more vital aspoliticians learn to articulate their agenda in a 30-second commercial. In short, just as the balancewithin the media industry has shifted from publicservice to private profit, so the emphasis withinpolitical communication will move from parties tovoters. Politicians rode the emergence of broad-casting with considerable success in the twentiethcentury but they are finding the going tougher inthe new millennium.

Globalization

In the global village, the world has been com-pressed into a television screen. In 1776 theEnglish reaction to the American Declaration ofIndependence took 50 days to filter back to the

United States. In 1950 the British response to theoutbreak of the Korean War was broadcast inAmerica in 24 hours. In 2003 British andAmerican viewers watched broadcasts of the IraqWar at the same time. We now take for grantedthe almost immediate transmission of newsworthyevents around the world.

Further, access to mass media is no longer a pre-serve of the developed world. Around half of thepopulation in developing countries now haveaccess to television; there are already more sets inChina than in the USA. Even in poor areas ofIndian cities, entrepreneurial ‘cable wallahs’ haveput a single satellite dish on to a roof and thenillicitly strung wires round the neighbourhood,charging affordable fees to local residents. WhereTV ownership remains low, as in Africa, many vil-lages possess a shared radio (Thomas, 2003). Theability to receive, if not transmit, information isbecoming global at a faster rate than is the case formany other goods and services.

Even wars have become, in part, battles of com-munication. During the conflict in Kosovo in1999, television stations worldwide were suppliedby NATO with video footage of high-techweapons hitting specified objectives. At the sametime, Serbia offered pictures of civilians ‘taken out’by missiles destroying wrong targets. In the Iraqwar of 2003, reporters were embedded in thecoalition’s military units, distributing live andoccasionally dramatic coverage to viewersthroughout the world (Iraq excepted).

The prime impact of the global village has beento encourage more open and informed societies. Itis now harder than ever for governments to isolatetheir populations from international develop-ments. Even communist states found it difficult tojam foreign radio broadcasts aimed at their people.Discussing the collapse of communist states,Eberle (1990, pp. 194–5) claimed that ‘thechanges in Eastern Europe and the Soviet Unionwere as much the triumph of communication asthe failure of communism’.

Recent technological developments also facilitateunderground opposition to authoritarian regimes.A small group with a fax machine and internetaccess now has the potential to draw the world’sattention to political abuses, providing sourcematerial for alert journalists. Burma’s military

110 POLITICS AND SOCIETY

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 110

POLITICAL COMMUNICATION 111

By the twenty-first century, the internet had becomeone of the most talked-about communications chan-nels, not just because of its explosive growth but alsobecause it permits the dissemination of any kind ofdata: images, music, speech, text and video. But whatwill be the long-term significance of the internet forpolitics?

The case forThe internet has spread more rapidly than previousinnovations such as radio and television (Table 7.3)Already, internet access has become routine in thedeveloped world, at least among the young andbetter-educated. By 2002/03, a majority of the population in 15 countries had used the internet(Table 7.2).

The internet goes further than any previous mediumin overcoming distance, paying scant regard tonational boundaries. During the war in Kosovo in1999, for instance, it was as easy for users to access theviews of the Yugoslav, American or British govern-ments – or of all three.Thus the web weakens thehold of states over their populations and fragmentspolitical communication, allowing specialized com-munities to exchange information and ideas in amanner which overwhelms national boundaries.

State censorship will become impractical; nationallibel laws carry little sway on the internet frontier. If agovernment wants its economy to benefit fromglobal e-commerce, it must pay the political price ofallowing its citizens access to the net. Governmentsmay be unable to tax e-commerce on electronic ser-vices offered across borders; the internet will bringabout the final triumph of business over politics.And the regulation of the internet itself will remain in the hands of technical committees rather thanstates, a leading case of governance without governments.

The case againstThe long-term political impact of the internet is exag-gerated. Users can surf the world, but how manyAmericans, in particular, will do so, given the predomi-nance of American sites? In any case the sitesreceiving most hits are those from governments, cor-

porations and institutions which are already best-known. In that sense, the internet is a secondary orreinforcing medium.

Nor can a computer network overcome differences oflanguage; secondary nets for languages other thanEnglish are recreating rather than bypassing linguisticdivisions. And for all the talk of electronic democracy,official referendums by internet are nowhere to befound. Note, too, that in the past, newspapers, radioand television all improved coverage of far-off affairsbut none succeeded in displacing the state; whyshould the technical act of linking computerstogether be any different?

Table 7.3 Time taken for new media to reach 50 million households in the USA

Medium Years

Radio 38 Television 13 Cable 10 Internet 5*

*estimate for 1995–2000.

Source: Data from Financial Times, 28 July 2000, citing Morgan Stanley Dean Witter.

AssessmentFirm judgements are premature.The lesson of pre-vious innovations is that projections of their impactproved to be wildly inaccurate. Some people thoughtradio would change the world; others said the tele-phone would never catch on. But two points are clear.First, technology does not determine content.Themessages which travel round the internet will reflectforces beyond the net itself. Second, access to theinternet requires a telecommunications infrastructurewhich barely exists in many developing countries. Atthe end of the twentieth century, most people in theworld had never made a telephone call, let alonesurfed the net, and wireless telephony offered themost immediate prospect for improving interactivecommunication in developing countries.

Further reading: Hall and Biersteker (2002), Norris (1999b),Sunstein (2001).

DEBATE

THE POLITICAL SIGNIFICANCE OF THE INTERNET

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 111

rulers, China’s communist government and SaudiArabia’s ruling families have all suffered from over-seas groups in this way.

Internet communication also allows extremistorganizations to communicate beyond nationalboundaries. Groups seeking suicide bombers canask for volunteers from across the world. Also, oddlinks emerge between remarkably disparategroups: for instance, Western anti-Semites andIslamic fundamentalists opposed to Israel. In theera of the internet, neither dictatorial nor democ-ratic governments can control global informationflows.

The media in established democracies

The media possess a natural vitality in consoli-dated democracies, where freedom of expression islegally protected. In this section, we examine thepolitical game played between journalists andpoliticians in established democracies. We thendiscuss the distinct character of the two mainmedia, television and newspapers, before turningto the vexed question of the media’s impact onvoting behaviour.

The media game

The relationship between politicians and journal-ists is increasingly central to the character of elitepolitics in modern, communication-rich democra-cies. Recognizing the influence of the media, gov-ernments and parties make enormous efforts toinfluence coverage. Governing parties, in partic-ular, devote considerable attention to informing,cultivating and seeking to manipulate the journal-ists whose reports achieve national coverage. Thehumble government press office, now populatedby highly paid spin doctors, has never been moreimportant.

Consider the media game played between gov-ernment (say, a White House spokesperson) andjournalists (say, a White House correspondent).This contest is classically political, mixing sharedand competing objectives. The government needscoverage and the media must fill its space, creatinga shared interest in news. But the governmentseeks favourable coverage while journalists are after

a big story, objectives that are rarely consistentsince, as every reporter knows, good news rarelymakes the headlines. The game is further compli-cated by competition among journalists them-selves, giving the administration leverage it canexploit by placing stories with compliant corre-spondents.

But as with so many political relationships, thegame of government–media relations is repeatedeach day, allowing long-term norms of acceptablebehaviour to develop. A spokesperson who givesout misleading information, or no information atall on a significant topic, will lose credibility. Sobriefers accept that over the long run it pays to beaccurate and even-handed in distributing informa-tion; the facts must be correct if rarely complete.Equally, competition among journalists reduces asthey learn to hunt as a pack. Correspondentsknow that their key task is to avoid missing thestory that everyone else has. Obtaining exclusivesis an optional extra.

Thus, in practice, stable routines develop whichreduce the game of government–media relations toa few well-understood ground rules. The outcomeis not so much collusion as contained competi-tion; sometimes, not even that. Feeding off eachother, journalists and politicians find themselveslocked in a permanent if occasionally uncomfort-able embrace.

DefinitionSpin doctor is a critical term applied to publicrelations experts working for politicians.The spindoctor’s job is to encourage favourable mediacoverage for a party or its leader.The termderives from the spins applied by baseballpitchers and was first applied to politics in 1977by the novelist Saul Bellow. Although the phraseitself is new, public relations has always beenimportant in politics. Nor should we overesti-mate the impact of spin. Indeed, public relationscan become counterproductive when itbecomes its own story, a problem which afflictedTony Blair’s Labour government in Britain (Bayley,1998).

Of course, the government invariably has a headstart over opposition parties in media presenta-tion. Statements by presidents and prime ministers

112 POLITICS AND SOCIETY

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 112

are always more newsworthy than those made bytheir political opponents. Journalists have nooption but to cover a prime ministerial pressrelease, even if they are fully aware of the party-political motives underlying its timing. Many aprime minister has arranged a prestigious overseastrip in the run-up to an election, keen to exploitthe domestic impact of a handshake with theAmerican president. Coverage of oppositionparties is inevitably a lower priority for the media(Rudd and Hayward, 2001).

Television and newspapers

In nearly all established democracies, television hasbecome the preeminent mass medium. Televisionis a visual, credible and easily digested formatwhich reaches almost every household, providingthe main source of political information. In elec-tion campaigns, for instance, the television studiohas become the main site of battle. The party glad-iators participate through appearing on interviews,debates and talk shows; merely appearing on tele-vision confirms some status and recognition oncandidates. Ordinary voters consume the election,if at all, through watching television. Local partyactivists, once the assault troops of the campaign,have become mere skirmishers. In short, televisionhas ceased to cover the campaign, it has becomethe campaign.

Of course, the political significance of televisiongoes far beyond elections. Much larger, if stilldeclining, numbers watch the evening newscasts

and it is here, in deciding what to cover and whatto leave out, that television exerts most influence(Box 7.2). Through their assumptions about news-worthiness, editors resolve their daily dilemma ofreducing a day’s worth of world events to less than30 minutes on the evening news.

Because news programmes focus on the excep-tional, their content is invariably an unrepresenta-tive sample of events. For example, policy fiascosreceive more attention than policy successes.Similarly, corruption is a story but integrity is not.Necessarily, television is a distorting mirror on theworld. The more compressed news coveragebecomes, the less accurate the TV lens must be.

Despite the primacy of television, it would bewrong to discount the political significance of thesecond mass medium, newspapers. Even today,quality newspapers possess an authority springingfrom their longevity. More important, perhaps,the press remain free of the tight regulation stillapplied to national broadcasters. In nearly alldemocracies, newspapers are freer with commentthan is television. In an age when broadcastersdominate the provision of instant news, the morerelaxed daily schedule of the press allows printcolumnists to offer interpretation and evaluation.

Television tells us what happened; at their best,newspapers can place events in context. Broadcastnews can only cover one story at a time whereasnewspapers can be scanned for items of interestand can be read at the user’s convenience.Newspapers offer a luxury which television canrarely afford: space for reflection. For such reasons,quality newspapers remain the trade press of poli-tics, read avidly by politicians themselves.

In many if not all democracies, newspaper circu-lations remain large. In Japan, Britain andScandinavia, most adults still read a daily news-paper. In Japan, unusually, the public still reliesmore on the national press than on television forits information and some studies there indicatethat Japanese newspapers exert more influenceover the electorate’s agenda (Feldman, 1993, p.24).

In Britain and Scandinavia, newspapers retain atleast some loyalties to the parties from which theyoriginally emerged. When a national British paperswitches its party support, as the best-selling Sundid to Labour in 1997, the shift commands atten-

POLITICAL COMMUNICATION 113

� Will the story have a strong impact on our audi-ence?

� Does the story involve violence? (‘if it bleeds, itleads’)

� Is the story current and novel?� Does the story involve well-known people? � Is the story relevant to our audience?

Source: Adapted from Graber (1997), pp. 106–8, based on studies inthe USA.

B OX 7 . 2

Some tests used by journalists todetermine newsworthiness

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 113

tion elsewhere in the media, illustrating hownewspapers remain significant political players intheir own right.

Furthermore, newspapers also influence televi-sion’s agenda: a story appearing on TV’s eveningnews often begins life in the morning paper. Andwhen an elector sees a story covered both on tele-vision and in the press, the combined impactexceeds that of either medium considered alone.In countries with a lively press tradition, newspa-pers retain a political significance greatly in excessof their circulation.

Reinforcement, agenda-setting andframing

What is the media’s impact on those exposed to it?By virtue of its general nature, this questionremains without an agreed answer. Even so, it isworthwhile examining how scholars have thoughtabout media effects and the mechanisms throughwhich such effects are achieved.

In the 1950s, before television became preemi-nent, the reinforcement thesis held sway (Klapper,1960). The argument here was that party loyaltieswere transmitted through the family and that,once developed, such identities acted as a political

sunscreen protecting people from the harmfuleffects of propaganda (Box 7.3). People saw whatthey wanted to see and remembered what theywanted to recall. In Britain, for example, manyLabour supporters read left-wing papers whilemost Conservatives bought a paper close to theirown outlook. Given such self-selection, the mostthe press could do was to reinforce their readers’existing dispositions.

The reinforcement theory proved its value half acentury ago and even today many studies of mediaimpact find only limited effects (Gavin andSanders, 2003). Even so, reinforcement is surely toolimited as an account of media effects today. Partyloyalties are now weaker, and television more perva-sive, than in the 1950s. For this reason, the agenda-setting view of media impact has gained ground.

The agenda-setting perspective contends thatthe media (and television in particular) influencewhat we think about, though not necessarily whatwe think. The media write certain items on to theagenda and by implication they keep other issuesout of the public’s gaze. As Lazarsfeld and Merton(1948) put the point, ‘to the extent that the massmedia have influenced their audiences, it hasstemmed not only from what is said but more sig-nificantly from what is not said’.

114 POLITICS AND SOCIETY

Concept Definition Comment

Reinforcement The media People read newspapers which support their existing outlook strengthen existing (selective exposure). In addition, people interpret information to opinions render it consistent with their opinions (selective interpretation)

and forget information that runs counter to their beliefs (selective recall)

Agenda-setting The media influence The compressed nature of television news means its coverage is what we think and highly selective. Reported events are widely discussed by the public talk about but non-reported events lose all visibility

Framing How an event is A frame focuses on particular aspects of a problem, its origins,narrated as a remedies and evaluation. It encourages viewers and readers to coherent story portray the topic in a similar wayhighlights particular features of it

B OX 7 . 3

Media effects: reinforcement, agenda-setting and framing

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 114

In an election campaign, for example, televisiondirects our attention to the candidates and to therace for victory; by contrast, the issues are often ofsecondary concern. Walter Lippman’s (1922) viewof the press is applicable to the media generally: ‘itis like a beam of a searchlight that moves restlesslyabout, bringing one episode and then another outof the darkness and into vision’.

The frame of a story – the way in which reportsconstruct a coherent narrative about the events –can also make a difference, especially whenrepeated. Are immigrants presented as a stimulusto the economy or as a threat to society? Is a warportrayed critically (Vietnam in its later stages) orpositively (Iraq in the invasion phase)? Is a con-victed criminal confronting the death penaltyfacing his just deserts or cruel and unusual punish-ment? As the concept of a ‘story’ suggests, thejournalist must translate the event covered into anorganized narrative which connects with theviewer or reader (Entman, 1993).

So the journalist’s words, as much as the cameraoperator’s images, help to frame the story, pro-viding a narrative which encourages a particularreaction from the viewer. In turn, the frame mayexert a priming effect, encouraging the viewer toarticulate opinions which might otherwise havegone unexpressed (Norris et al., 2003).

The media in new democracies

An assessment of the position of the media in newdemocracies depends heavily on the point of com-parison. When we draw a contrast with the pre-ceding regime, media independence seems to be ahallmark of the new order. As communist andmilitary rulers depart, so censorship gives way to amultitude of new publications, peddling every-thing from news to pornography. Old people arebemused, just as the young are amused, buteveryone can see that the climate has changed.Indeed, an increasing spirit of media adventure isone of the first cracks to appear in pre-democraticstructures; journalists are quick to sense theiropportunity when authoritarian rule begins tocrumble.

Yet even after several years in a new democracy,the position of the media usually remains weaker

than in established democracies. Judged by thestandards of most consolidated democracies,media autonomy remains limited. Governingparties continue to exert influence over broad-casting media while political entrepreneurs useownership of newspapers to maximize their influ-ence. Far from becoming the fourth estate, themedia in many new democracies are pawns in apolitical game played above their heads.

Russia is an example of a post-communist semi-democracy where pressures on the media, frompowerful business people as well as politicians,remain intense. This influence derives preciselyfrom the centrality of television to political com-munication in Russia. As in many poor countries,broadcasting is the main way of reaching a large,dispersed and impoverished population for whomfree television has more appeal than paid-forpapers. The television audience in Russia fornightly news programmes is substantial indeed,matching that in the United States for the SuperBowl (Zassoursky, 2002).

For good reason, then, Moscow’s power-holderscompete intensely for control of a pluralistic, butfar from independent, television system. In thisruthless game, politicians in government still holdthe upper hand. Over 100 laws refer to the massmedia, giving rulers legal scope to close down anyparticular channel they deem to be unfriendly.And the public television stations – ORT andRTR – still experience direct political interference.

But in contrast to the communist period, theRussian government no longer monopolizes themedia. Politicians on the make will seek to acquiretheir own newspapers and broadcast channels andthen to exploit them in building their politicalbase. The leading Russian banks also own manyprivate television stations, which dutifully propa-gate their masters’ interests.

In such a context, self-censorship – the voice inthe editor’s head which asks ‘am I taking a risk inpublishing this story?’ – remains rife. The Russianexperience demonstrates that lack of direct govern-ment control does not guarantee freedom for themedia; indeed, Russia comes close to the bottomin expert rankings of press freedom (Table 7.4).

Vladimir Putin’s success in the presidential elec-tion of 2000 owed much to his control of publictelevision, whose news broadcasts in the final days

POLITICAL COMMUNICATION 115

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 115

of the campaign portrayed his opponents as sym-pathetic to gays and Jews. However, even the pres-ident was unable to suppress the surge of mediacriticism that followed an explosion abroad aRussian submarine, with the loss of all 118 liveson board, a few months later. While Putin’s elec-tion demonstrates beyond doubt that the Russianmedia are subject to political influence that canalter election outcomes, the Kursk disaster showedthat such control is markedly less complete than inthe communist period.

In post-communist Europe, journalistic stan-dards remain low, reflecting decades in whichreporters appointed by the communist party justreproduced party platitudes. Easy editorializing stilltakes priority over the hard graft of news-gathering.There is little tradition of media scrutiny of rulersto carry over from the pre-communist era. In some

countries, media freedom amounts to little morethan an opportunity for journalists to bribe publicfigures. In impoverished Romania, for instance,newspapers threaten unfavourable coverage ofpoliticians unless they are paid hush money.

Entrenching the freedom of the media to investi-gate and criticize public conduct is inevitably agradual affair in new democracies, requiring gov-ernment restraint, media professionalism and inde-pendent courts. Despite a rhetorical commitmentto free speech, the impulse of many leaders is stillto manipulate and even control (Milton, 2000).

Public opinion

Especially in democracies, politics is a battle forinfluence over the important but imprecise terrainof public opinion. Governments, parties andinterest groups jostle to influence public opinion,seeking to persuade the public to adopt theiragenda, their frame of reference and their policypreferences. Yet the notion of public opinion itselfremains somewhat opaque. One approach ishighly pragmatic: public opinion is simply whatthe public thinks about a given issue, nothingmore and nothing less. This down-to-earth viewwas reflected in the definition offered by SirRobert Peel, twice prime minister of Britain in thenineteenth century. He referred to public opinionas ‘that great compound of folly, weakness, preju-dice, wrong feeling, right feeling, obstinacy andnewspaper paragraphs’ (Green, 1994, p. 213).

A second view of public opinion is more ambi-tious, linking the idea of a ‘public’ to the views ofan informed community which shares basic polit-ical principles. Lippman (1922) illustrates thismore refined approach. He writes that ‘a body ofmen are politically capable of public opinion onlyso far as they are agreed upon the ends and aims ofgovernment and upon the principles by whichthose ends shall be attained’. Here public opinionis interpreted as the considered will of a cohesivegroup. This perspective derives from the model ofa self-governing republic and treats public opinionas far more than an opinion poll finding. Rather,public opinion acquires moral weight and anyminorities within the wider community shouldnormally accept the public’s verdict.

116 POLITICS AND SOCIETY

Table 7.4 Countries ranked by press freedom,2001/02

Top 13:

1= Finland, Iceland, Netherlands, Norway5 Canada6 Ireland7= Denmark, Germany, Portugal, Sweden

11 France12= Australia, Belgium

Selected others:

17 USA21 UK26 Japan 39 South Korea54 Brazil76 Mexico99 Turkey

121 Russia125 Saudi Arabia138 China

Note: Based on a survey of 139 countries using national experts.The criteria of press freedom include a supportive legal environ-ment, no attacks on journalists and media organizations, and thepossibility of redress when such assaults occur.The USA is rankedonly 17th mainly because journalists who refuse to reveal theirsources can be prosecuted.

Source: Data from Reporters without Borders at http://www.rsf.fr.

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 116

DefinitionPublic opinion can simply refer to (1) ‘the rangeof views on some controversial issue held bysome significant portion of the population’(Qualter, 1991). More ambitiously, the term candenote (2) the informed judgement of a commu-nity on an issue of common concern, where thatjudgement is formed in the context of sharedpolitical goals.

In democracies, public opinion extends to virtu-ally the entire adult population. Nearly all adultscan vote, their views are represented in frequentopinion polls and politicians facing reelection havean incentive to study the findings. Even in democ-racies, however, a minority underclass exists whichrarely votes, lives in no-go areas for pollsters anddoes not follow politics at all. In authoritarianregimes, of course, national politics engages farfewer people and public opinion shrinks.

Given that public opinion is an importantelement of democratic politics, how exactly does itexert its influence? In a sense public opinion per-vades all policy-making. It forms the environmentwithin which politicians work. Thus publicopinion sits in on many government meetings eventhough it is never minuted as a member. In suchdiscussions public opinion usually performs one oftwo roles, acting either as a prompt or as a veto.‘Public opinion demands we do something abouttraffic congestion’ is an example of the former;‘public opinion would never accept restrictions oncar use’ illustrates the latter. So as Qualter (1991,p. 511) writes, ‘while public opinion does notgovern, it may set limits on what governments do’.

Yet public opinion is never all-powerful, even inliberal democracies. Four factors limit its influ-ence:

� The impact of public opinion declines as issuesbecome more detailed. Voters are concernedwith goals rather than means; with objectivesrather than policies. ‘What policies politiciansfollow is their business; what they accomplish isthe voters’ business’ (Fiorina, 1981). A fewimportant objectives preoccupy the public butmost policies are routine and uncontroversial.In detailed policy-making, expert and organized

opinion matters more than public opinion.� The public is often ill-informed, and this, too,

limits its impact. Most Americans, for example,are unable to name their members of Congress.Similar findings from other democraciesconfirm the ignorance of large sections of thepublic, especially on remote issues involvingforeign policy. Limited knowledge is anotherreason why public opinion functions moreoften, and more appropriately, as an agenda-setter than as a policy-maker.

� Public opinion can evade trade-offs but govern-ments cannot (though they sometimes try). Thepublic may want both lower taxes and increasedpublic spending but rulers must make a choice.Further, the risks associated with a policy arepoorly assessed by the public but require closeattention from decision-makers (Weissberg,2002).

� Politicians respond to their perceptions ofpublic opinion but these interpretations can beinaccurate, derived as they are from the dis-torting lens of interest groups and the selectivetelescope of the media. Politicians typicallyrespond to mobilized opinion rather than topublic opinion (Herbst, 1998).

So even in the most democratic of countries,government by public opinion remains a distantdream or nightmare. Even so, the idea of publicopinion has gained further currency as opinionpolls, citizen juries and focus groups have devel-oped to study it. Indeed there are few areas in poli-tics where a concept is so closely linked to how it ismeasured. In modern democracies, public opinionis both measured by, and partly composed of,reports of investigations into its content (Box 7.4).

Consider opinion polls, and sample surveys, themost accurate method of identifying what peopleprofess. Although the public itself remains res-olutely sceptical of samples, their accuracy is nowwell-attested, at least in predicting election out-comes. In the United States, the average error inpredicting the major parties’ share of the vote atpostwar national elections has been around 1.5 percent. This accuracy is impressive even if not alwayssufficient for the television networks to pick theright winner on election night. Precision is similarin other democracies. Counter-intuitive it may be,

POLITICAL COMMUNICATION 117

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 117

but 1,000 people carefully selected for an opinionpoll can accurately represent the whole population.

A well-chosen sample will certainly provide amore reliable guide to public opinion than theself-selected readers who answer write-in polls inmagazines; the listeners who call radio discussionshows; or the voters who contact their representa-tive about their pet topic. In fact, a major virtue ofopinion polls is that they capture the views of themany bored and uninterested electors who wouldotherwise be left out of the debate.

Opinion polls contribute to the democraticprocess in three ways. First, they bring into thepublic realm the voices of people who would oth-erwise go unheard. They are the only form of par-ticipation, apart from the ballot box itself, inwhich all count for one and none for more thanone. Second, polls are based on direct contactbetween interviewers and the public; they getbehind the group leaders who claim to speak ‘on

behalf of our members’. Third, polls enable politi-cians to keep in touch with the popular mood andthey give some insight into the reasons for electionresults. In these ways, opinion polls oil the wheelsof democracy.

Yet it would be wrong to overstate the value ofopinion polls in defining the ‘mood of the people’.Like students taking a test, interviewees in asurvey are answering questions set elsewhere. Pollsare commissioned not by ordinary people but byparty officials and journalists in the capital city. Asa result, people may never have thought about atopic before they are invited to answer questionson it. They may give an opinion when they havenone or they may agree to a statement because it isthe easiest thing to do (‘yea-saying’).

For these reasons, Ginsberg (1986) suggests thatelites use polls to ‘pacify or domesticate opinion’by reducing citizens to passive respondents whovoice opinions in response to a narrow range oftopics selected by the elite itself. Certainly, onedanger of opinion polls is that they construct, andeven shape, public opinion at the same time asthey measure it.

Because opinion polls do not give respondents achance to discuss the issue before expressing theirviews, the technique is criticized by those whofavour more ambitious interpretations of publicopinion. When public opinion is defined as com-munity judgement of an issue against the standardof the common interest, opinion polls are foundwanting. Where, ask the critics, is the equivalentof the vigorous debate which preceded themoment of decision in the Athenian assembly? Tobe useful, it is argued, public opinion must havebeen shaped in the press of collective discussion,not simply measured through individual question-naires.

Building on this richer view of public opinion,scholars have developed the idea of a deliberativeopinion poll or citizen jury (Fishkin, 1991). Thistechnique involves exposing a smaller sample ofelectors to a range of viewpoints on a selectedtopic, perhaps through presentations by expertsand politicians. With awareness of the problemestablished in this way, the group proceeds to adiscussion and a judgement. Opinion is only mea-sured when the issues have been thoroughly aired.As Fishkin (1991, p. 1) explains,

118 POLITICS AND SOCIETY

� An opinion poll is a series of questions asked in astandard way of a systematic sample of the popu-lation.The term usually refers to short surveys ontopical issues for the media. Polls are usually con-ducted face to face or by telephone (Mann andOrren, 1992).

� A sample survey is conducted using similarmethods as opinion polls but involves a moredetailed questionnaire, often conducted for gov-ernment or for academic researchers (Broughton,1995).

� In a deliberative opinion poll, or citizens’ jury,people are briefed by, and can question, expertsand politicians on a given topic before opinionsare measured.The technique seeks to measurewhat public opinion would be if the public werefully informed on the issue (Fishkin, 1991).

� A focus group is a moderated discussion amonga small group of respondents on a particulartopic. A focus group is a qualitative device used toexplore the thinking and emotions lying behindpeople’s attitudes. It is an open-ended techniquethat has found favour with party strategists(Greenbaum, 1998).

B OX 7 . 4

Measuring public opinion

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 118

an ordinary opinion poll models what the publicthinks, given how little it knows. A deliberativeopinion poll models what the public wouldthink, if it had a more adequate chance to thinkabout the questions at issue.

Deliberative polling can therefore be used toanticipate how opinion might develop on newissues. It is also particularly useful on issues with alarge technical content, for example nuclearenergy or genetic testing. Though not widely used,citizens’ juries are an ingenious attempt to over-come the problem of ill-informed replies whichbedevils conventional opinion polls.

A focus group, finally, is similar in some ways toa deliberative poll. In a focus group, a researchergathers together a small group of people – nor-mally eight to ten – for a general discussion of atopic in which they share an interest. Examplesmight include lapsed Democrats, young non-voters or college drop-outs. The idea is to explorethe perspectives through which the participantsview the issue, sometimes for its own sake andsometimes in preparation for drafting a question-naire on the same topic.

The advantage of a focus group is that it stimu-lates those involved to react to each other’sthoughts, thus providing richer information thancan be obtained through opinion polls and samplesurveys. But the findings are qualitative ratherthan quantitative and may be unrepresentative ofthe wider population (Greenbaum, 1998).

The media in authoritarian states

If democracy thrives on the flow of information,authoritarian rulers survive by limiting free expres-sion, leading to journalism which is subdued evenwhen it is not subservient. Far from acting as thefourth estate, casting its searchlight into the darkercorners of government activity, the media inauthoritarian states defer to political power. Lack ofresources within the media sector usually increasesvulnerability to pressure. Official television stationsand subsidized newspapers reproduce the regime’sline while critical journalists are harassed and theentire media sector develops an instinct for self-preservation through self-censorship.

In her study of sub-Saharan Africa before thewave of liberalization in the 1990s, Bourgault(1995, p. 180) identified several means used byauthoritarian leaders to limit independent jour-nalism. These included:

� broad libel laws � states of emergency � licensing of publications and journalists� heavy taxation of printing equipment � a requirement to post bonds before new publi-

cations can launch� restricted access to newsprint � the threat of losing government advertising

Sub-Saharan Africa also demonstrates the lack ofresources which holds back the development ofthe media sector – and authoritarian states aremost often found in relatively poor countries. Ashortage of resources limits journalistic initiativeand increases vulnerability to pressure; impover-ished journalists may even be reduced to pub-lishing favourable stories in exchange for money.

Pressures from authoritarian leaders on themedia are by no means restricted to Africa. Inpost-communist central Asia, journalists continueto practise the delicate art of coexisting with non-democratic rulers. Large parts of the media,including news agencies and printing presses,often remain in state hands, giving the authoritiesdirect leverage. The state also typically retainsownership of a leading television channel. Foley(1999) reports that

from Kazakhstan to Kyrgyzstan and Tajikistan toBelarus and Ukraine, the story is a dismal one:tax laws are used for financial harassment; abody of laws forbids insults of those in highplaces; compulsory registration of the media iscommon. In Azerbaijan, as in Belarus, one-manrule leaves little room for press freedom.

The justification for restricting the freedom ofthe media typically refers to an overriding nationalneed such as social stability, nation-building oreconomic development. A free press, like a com-petitive party system, is presented as a recipe forsquabbling and disharmony. Many of these ‘justi-fications’ for controlling the media may just be

POLITICAL COMMUNICATION 119

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 119

excuses for continuing non-democratic rule. Evenso, we should not assume that the Western idea ofa free press has universal appeal. Islamic states, inparticular, stress the media’s role in affirming reli-gious values and social norms. A free press is seenas an excuse for licence; why, the question is asked,should we import Western, and particularlyAmerican, ideas of freedom if the result is justscandal-mongering and pornography? In someMuslim societies, the result is what Mernissi(1993, p. 68) terms ‘TV Islam’, a strange brew ofdull religious programming and inoffensiveAmerican movies.

As with other aspects of politics, totalitarianstates developed a rather more sophisticatedapproach to their dominance of the media. Masscommunication was expected to play its part inthe transformation of political culture. The mediawere therefore brought under tighter control thanin authoritarian systems. Unrelenting penetrationof mass communication into everyday life was acore component of the totalitarian system. ‘It isthe absolute right of the state to supervise the for-mation of public opinion’, said Joseph Goebbels,Hitler’s Minister of Propaganda.

It is significant however that neither communistnor fascist regimes regarded the media, by them-selves, as a sufficient means of ideological control.Communist states supplemented mass propagandawith direct agitation in places where people gath-ered together; fascist regimes valued direct controlthrough public meetings rather than impersonalbroadcasts through the media.

In communist states, the dual dimensions ofpolitical communication were propaganda andagitation. Propaganda explained the party’smission and instructed both the elite and themasses in the teachings of Marx, Engels andLenin. To achieve their propaganda objectives,ruling communist parties developed an elaboratemedia network with radio, posters, cinema andtelevision all reinforcing each other. Even art hadto become ‘socialist art’, playing its part in the rev-olution of hearts and minds. Communist societieswere short of many things but propaganda was notone of them.

Where propaganda operated through the media,agitation operated at local level. Agitation soughtto mobilize the masses behind specific policies

such as increased production. The party sought toplace an agitator in each factory, farm and militaryunit. Together, propaganda and agitation domi-nated the flow of information; their combinedeffect would be all the greater because, in thenature of a total regime, no dissenting voices werepermitted.

DefinitionPropaganda is communication intended topromote a particular cause by changing atti-tudes and behaviour; today, the term also impliesthe absence of a balanced perspective.The wordis religious in origin: in the seventeenth century,the Catholic Church established a College ofPropaganda to propagate the faith. Propagandawas an important feature of communication in totalitarian regimes but it has also beenimportant to democracies in times of war.

The communist experience offers an importantreal-life experiment of the attempt to use themedia as a tool for influencing the public. What,then, was the impact of such propaganda? It seemsto have helped with agenda-setting by disguisinglocal problems such as accidents, poverty or pollu-tion which could be hidden from people in otherparts of the country. It may also have scored somesuccess in highlighting achievements such asindustrialization. And, just as advertising byparties in democracies is often aimed at their ownactivists, so communist propaganda may havehelped party members to keep the faith.

But the communist experience also revealed thelimits of media power. A cynical public was noteasily fooled. Grandiose statements about nationalprogress were too often contradicted by the grimrealities of daily life under communism. Whileofficial reports on Western cities showed images ofdrunks and beggars in shop doorways, the viewerswould gawp at the unimaginable luxuries ondisplay in the shop windows.

In any case, as communist states became inert,so propaganda became empty ritual. Eventually,the main function of propaganda became that ofshowing the population that the party still ruled;whether the propaganda was convincing almostceased to matter. Propaganda became less a mech-anism of conversion and more a symbol of

120 POLITICS AND SOCIETY

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 120

control. Like the drizzle, nothing could be doneabout it so it was ignored.

It is noteworthy, however, that the remainingstates with a nominal communist allegiance havekept close control over the means of mass com-munication. In China, access to information hastraditionally been provided on a need-to-knowbasis. The country’s rulers remain keen to limitdissenting voices even as the party’s tolerance ofnon-political debate increases (Box 7.5). Forinstance, critics refer to the Great Firewall ofChina, a phrase that describes the state’s attemptsto control internet use beyond the natural barrier

of language. Although the regime is keen topromote e-commerce, internet users who searchfor ‘inappropriate’ topics such as democracy orTibetan independence will find their access tosearch engines withdrawn. The excuse offered isthat the authorities are merely protectingChinese people from unwholesome Westerninfluence.

Key reading

Next step: Norris (2000) offers an interestingcomparative account of the media’s role in Westerndemocracies.

Communication may be central to politics but thedefinitive contemporary treatment of the topic isstill to be written. Williams’s short but influentialstudy (1962) is still relevant to students of politics.Mann’s history of power (1986) includes consider-able material on communication. However, mostmodern studies focus on the media rather thancommunication generally. Gunther and Mughan(2000) offer a comparative collection; see alsoBennett and Entman (2001) and Street (2001).On the USA, Graber (1997) is an authoritativetext while Barber (1992) provides a lucid history.Norris et al. (2003) compare the frames used inmedia coverage of 9/11 with earlier terrorist inci-dents. On public opinion, Glynn et al. (1998) is awide-ranging collection while Manza et al. (2002)examine the impact of polls on policy. Fishkin(1991) looks at citizen juries. Away from the West,McCargo (2002) examines the media in Asia andBourgault (1995) discusses sub-Saharan Africa.Milton (2000) is an overview of the media in post-communist states.

POLITICAL COMMUNICATION 121

As an example of how authoritarian governmentsseek to limit media freedom, consider the followingrules introduced by the Chinese government in 2000to regulate internet content providers.Websiteproviders in China are required:

� to obtain government approval before cooper-ating with businesses overseas

� to avoid content that subverts state power orharms the country’s reputation

� to avoid content that supports banned cults � to record the content of their site� to record all website visitors for 60 days� to hand over records to the police on demand.

Even so, internet access is expanding rapidly inChina, allowing a sophisticated minority to circum-vent the party’s attempt to restrict full informationto its own elite.

Source: Adapted from Financial Times, 3 October 2000.

B OX 7 . 5

Limiting the damage: China andthe internet

14039_13153_10_Ch7.qxd 5/4/04 2:00 pm Page 121

Participation in established democracies 122Participation in new democracies 124Social movements 128Participation in authoritarian states 131Political violence and terror 134Revolution 136Key reading 141

An environmentalist lying in front of workersconstructing a new road, a citizen contacting

a legislator about a missing social security cheque,a terrorist plotting a suicidal assault on a despisedregime – all are examples of individual participa-tion in politics. This chapter examines generalcitizen involvement in politics; we reserve partici-pation through elections for the next chapter.

Patterns of participation differ greatly betweendemocratic and authoritarian governments (Box8.1). In established democracies, voluntary partici-pation is the norm. People can choose whether toget involved (for example by voting or abstaining)and how to do so (for example by joining a partyor signing petitions). The main exception to thevoluntary nature of participation is compulsory

voting, found in a few democracies such asAustralia and Belgium.

In authoritarian regimes, by contrast, participa-tion is diminished in quality if not quantity. Mostnon-democratic rulers hang a ‘keep out’ sign overthe political sphere, often limiting formal partici-pation by ordinary people to rigged elections.Only in totalitarian regimes did non-democraticrulers seek to mobilize the people behind theregime’s effort to transform society, whethertowards a communist or fascist ideal.

In this chapter, we examine political participa-tion in democratic and non-democratic settingsbefore turning to the less orthodox forms of par-ticipation represented by violence, terror and revo-lution.

Participation in established democracies

The most striking fact about political participationin established democracies is how far it falls shortof a participatory ideal. Voting in national elec-

Chapter 8

Political participation

Regime type Amount of Character of Purpose of participationparticipation participation

Democratic Moderate but Mainly voluntary To influence who decides and what decisions are declining reached

Authoritarian Low Manipulated To protect rulers’ power and to offer a democratic façade

Totalitarian High Regimented In theory:to transform societyIn practice:to demonstrate rulers’ power

B OX 8 . 1

Patterns of political participation, by type of regime

122

tions is normally the only activity in which amajority of citizens engages. Throughout thedemocratic world anything beyond voting is thepreserve of a minority of activists. These partici-pants are outnumbered by the apathetics: peoplewho neither vote nor even follow politics throughthe media. With turnout and party membershipfalling in most democracies, even the traditionalactivists seem to be reducing their involvement informal politics (Mair and van Biezen, 2001).

In a renowned analysis, Milbrath and Goel(1977, p. 11) divided the American populationinto three groups, a classification which has sincebeen applied to participation in other democra-cies. These categories were:

� a few gladiators (around 5–7 per cent of thepopulation) who fight the political battle

� a large group of spectators (about 60 per cent)who watch the contest but rarely participatebeyond voting

� a substantial number of apathetics (about one-third) who are withdrawn from politics.

Milbrath’s labels were based on an analogy withRoman contests at which a few gladiators per-formed for the mass of spectators but some apa-thetics did not even watch the show (Figure 8.1).

DefinitionPolitical participation is activity by individualsformally intended to influence who governs orthe decisions taken by those who do so. Citizenscan be classified both by the extent of their polit-ical involvement (e.g. gladiators, spectators andapathetics) and by the form their engagementtakes (e.g. conventional, unconventional or both).

In every country, political gladiators are far froma cross-section of society. In most democracies,participation is greatest among well-educated,middle-class, middle-aged white men.Furthermore the highest layers of political involve-ment show the greatest skew. As Putnam (1976, p.33) put it,

The ‘law of increasing disproportion’ seems toapply to nearly every political system; no matterhow we measure political and social status, the

higher the level of political authority, the greaterthe representation for high-status social groups.

What explains this bias in participation towardsupper social groups? Two factors seem to be par-ticularly influential: political resources and polit-ical interest.

Consider, first, the question of resources. Peoplein high-status groups are equipped with suchpolitical assets as education, money, status andcommunication skills. Education gives access toinformation and, we trust, strengthens the abilityto interpret it. Money buys the luxury of time forpolitical activity. High status provides the oppor-tunity to obtain a respectful hearing. And commu-nication skills such as the ability to speak in publichelp in presenting one’s views persuasively. Addedtogether, these resources provide a tool-kit foreffective political intervention.

Second, high-status individuals are more likelyto be interested in politics. They possess themotive as well as the means to become involved.No longer preoccupied with the daily struggle,they can take satisfaction from engagement in col-lective activity (Inglehart, 1997). The wealthy arealso more likely to be brought up in a family, andto attend a school, where an interest in currentaffairs is encouraged. So upper social groups showan interest in politics and can afford to put theseconcerns into practice.

We can apply this dual framework – of politicalresources and political interest – to the problem ofwhy women are still under-represented at higherpolitical levels. In 2000, women made up just 14per cent of the world’s legislators, double the

POLITICAL PARTICIPATION 123

Gladiators

Spectators

Apathetics

Figure 8.1 Patterns of participation in democracies

figure 25 years earlier but a large under-representa-tion (Table 8.1). There is no doubt that, as agroup, women still possess fewer politicalresources than men. Across the world, they haveless formal education (even though this differencehas reversed among the young in many establisheddemocracies). Further, interest in formal politics issometimes limited by childbearing and home-making responsibilities.

In addition to these underlying factors, malepreeminence in politics tends to be self-perpetu-ating. Some women lack the confidence needed tothrow themselves into the hurly-burly of formal,male-led politics while many still face the highhurdle of discrimination from sexist male politi-cians. These gatekeepers claim that women areunsuited to politics – and then use the scarcity ofwomen in high office to prove their point (Burnset al., 2001).

Institutional factors, some of which are rathereasy to modify, also help to explain female under-representation. Three such features are:

� The electoral system. Women do best under theparty list version of proportional representation,a method that allows party officials to select agender-balanced set of candidates. Scandinaviais a prominent example (Table 8.1).

� Quotas. These ensure women make up a speci-fied proportion of a party’s candidates or, moreradically, of its elected representatives. InNorway, for example, the leading parties intro-duced quotas as early as 1973.

� Turnover of legislators. Low turnover, as in theUnited States Congress, creates a recruitmentbottleneck which enables men, once elected, toremain in post for decades.

DefinitionPolitical exclusion refers to those people whoare effectively excluded from participation in col-lective decision-making because they occupy amarginal position in society. Migrant workers,prisoners, drug addicts and those who do notspeak the native language are examples ofgroups confronting this difficulty.

The emphasis of research on political participa-tion is on explaining what distinguishes the gladia-

tors from the spectators. But what about the apa-thetics, the people who do not participate at all?This group raises the emerging problem of polit-ical exclusion. The apathetics in effect excludethemselves – or perhaps are excluded – from thenormal means by which citizens collectively shapetheir society (Verba et al., 1995).

A typical non-participant might be an unem-ployed young man with no qualifications, inhab-iting a high-crime inner-city neighbourhood,often from a minority culture and perhaps noteven speaking the dominant language. Usuallysuch people are not registered to vote, often theydo not watch the news on television and rarely dothey read a national newspaper. Their politicalactivity, if any, is often irregular and spasmodic, aswith riots.

Just as higher social groups possess both theresources and the interest to overparticipate, so ashortage of resources and lack of interest in theremote goings-on of national politics explain theunderparticipation of those near the bottom ofincreasingly unequal Western societies. The emer-gence of a two-tier political system, characterizedby a majority which votes at national elections anda non-participating underclass, is a growing chal-lenge to the assumption of political equality onwhich democracy is based.

Participation in new democracies

The participation challenge facing new democra-cies is considerable. The transition from authori-tarian rule requires the population to learn newstyles of political participation: votes must be cast,parties organized and leaders recruited to politicaloffice. Cynicism must give way to a measure ofengagement. Creating democratic institutions is ashort-term constitutional task but embedding thehabit of voluntary participation is a long-termnecessity.

Consider, for example, post-communist democ-racies. The old style of regimented participationquickly disintegrated as communism fell apart,partly in response to dramatic street protests in thecapital cities. Once new and nominally democraticinstitutions had been created, the task was to con-solidate the new order by developing structured

124 POLITICS AND SOCIETY

forms of voluntary participation through parties,elections and interest groups.

This task has proved to be demanding. The pop-ulations of many Eastern European states hadexperienced regimented participation under com-munist rule and seen mass participation on televi-sion during its collapse. But this outburst ofengagement did not last long. Once the commu-nists were ousted, participation soon began todecline, reflecting the communist legacy of polit-ical cynicism among large sections of the public.In Czechoslovakia, Illner (1998, p. 75) notes that‘more active participation was reduced again afterthe short period of post-1989 exhilaration and thetraditionally detached attitudes of the populationtowards public involvement were continued’. Inmany countries traditions of paternalism and elitearrogance reestablished themselves.

Reflecting both communist and pre-communistinfluences, many more people express an interestin politics than actually take part. This discrep-ancy is especially marked in Russia. Two-thirds of

Russians say they regularly or sometimes discussnational problems but fewer than one in fivebelieve they can influence public decisions.Disengagement seems to reflect not so much lackof interest as lack of belief in the ability to make adifference (Remington, 2004, p. 385). So a stablesystem of voluntary participation requires the rela-tionship between state and society to be com-pletely recast, a task requiring far greater skill thanmerely chopping up the rotting timbers of thecommunist state.

The problem is to build a ‘civil society’ regulatedby law but remaining separate from the state. Sucha society provides opportunities for people to par-ticipate in collective activities that are neither pro-state nor anti-state but simply non-state. Butdeveloping such social capital is a conspicuouschallenge in countries emerging from commu-nism. Under communist rule, civil society hadbeen demobilized. It had been stood down so thatthe rulers could directly control the individual:‘everyone was supposed to be the same – working

POLITICAL PARTICIPATION 125

Table 8.1 Female representation in national legislatures, 1950s–2000s

Year of female Current electoral Female representation Change (mostsuffrage system* in the legislature (%)# recent election

minus 1950s average)Average in Most recentthe 1950s election

Sweden 1919 List PR 12 45 +33Denmark 1915 List PR 8 38 +30Finland 1906 List PR 13 38 +25Netherlands 1919 List PR 8 37 +29Norway 1913 List PR 5 36 +31Germany 1918 Mixed 8 32 +24New Zealand 1893 Mixed 4 28 +24Australia 1902 AV 0.2 25 +25Canada 1917 Plurality 1 21 +20United Kingdom 1918 Plurality 3 18 +15USA 1920 Plurality 3 14 +11Ireland 1918 STV 3 13 +10France 1944 Majority 4 12 +8Italy 1945 Mixed 6 12 +6

* For definitions of electoral systems, see p. 148 (AV, alternative vote; PR, proportional representation; STV, single transferable vote).# In bicameral assemblies, figures are for the lower house.The most recent election is as at 2003 .

Sources: Inter-Parliamentary Union (2000), http://www.ipu.org; McAllister and Studlar (2002).

126 POLITICS AND SOCIETY

‘Americans of all ages, all stations in life, and all types ofdisposition are forever forming associations . . .Thus themost democratic country in the world now is that inwhich men have in our time carried to the highest per-fection the art of pursuing in common the objects ofcommon desires’ (de Tocqueville, 1835, p. 513)

In influential publications, Putnam (1993, 2000) ques-tions whether de Tocqueville’s thesis about the UnitedStates still applies. Putnam claims that ‘something hashappened in America in the past two or threedecades to diminish social engagement and civic con-nectedness’, with damaging consequences for thequality of politics. He suggests that Americans nowspend more time watching Friends than making them.Is Putnam right? If so, does his thesis also apply toother democracies?

The case forPutnam marshals considerable evidence from theUSA to illustrate his theme. For instance, between the1960s and the end of the 1980s:

� voter turnout declined by nearly a quarter (and byrather more among young electors)

� the proportion of Americans claiming to haveattended a public meeting ‘in the last year’ fell bymore than a third

� the proportion of people engaged in regular volun-teer work dropped by a sixth

� the proportion agreeing with the survey statement‘most people can be trusted’ fell from a half to athird.

Putnam believes that the significance of these find-ings lies in their implications for social capital (see p.92). Face-to-face communication nurtures commit-ment to the common good; allows networks todevelop from which new projects can emerge;permits individuals to develop their skills, knowledgeand understanding; and generally encourages thegive-and-take which is a hallmark of democracy.Whatkind of ‘democracy’ awaits, he asks, if most people arenow ‘bowling alone’?

The case againstAll arguments that ‘things are not what they were’ aresuspect. Golden ages are more often tricks of memorythan historical realities. Even in the 1950s, commenta-tors were bemoaning the rise of the ‘lonely crowd’,‘mass society’ and the ‘inner-directed personality’(Riesman, 1950). It is much more likely that socialactivity has changed its form rather than its extent. AsPutnam (2002) himself recognizes, some types ofgroup, such as crime-watch groups, health clubs,support groups (e.g. of crime victims) and publicinterest groups have expanded.

With the advent of mobile phones, email and theinternet, people can associate with like-minded indi-viduals elsewhere in the country rather than face toface. As a result, they depend less on old-fashionedneighbours.There is no reason to suppose that thischange reflects any loss of ‘social capital’. Indeed, theability to join, benefit from and then leave specific,instrumental and short-term networks in response tochanging individual needs may increase the ‘effi-ciency’ of social relationships compared to an erawhen people relied on a few neighbours.

AssessmentEven if social participation is declining in the USA, it isdifficult to see what can be done to reverse the trend.Any political implications need political solutions. IfAmerica is to solve its turnout problem, for instance,the solution surely lies in improved registration proce-dures, not in persuading people to invite their neigh-bours to a barbecue. Further, the long-run effects of9/11 are as yet unclear. On the one hand, the attacksmay encourage families to retreat further into thesafety of their own home, diminishing social participa-tion. On the other hand, the immediate effect seemedto be a rush of civic engagement, citizen surveillanceand support for the government. At least in the shortterm, these factors reduced the extent to whichAmericans could be said to be ‘bowling alone’.

Further reading: Pharr and Putnam (2001), Putnam (1993,2000, 2002).

DEBATE

BOWLING ALONE

for the state, on a salary, on a leash’ (Goban-Klasand Sasinka-Klas, 1992). Creating a participatorycivil society from the communist inheritance is adifficult and continuing task.

DefinitionCivil society consists of those groups which are‘above’ the personal realm of the family but‘beneath’ the state.The term covers public orga-nizations such as labour unions, interest groupsand, on some definitions, recreational bodies.However, firms are usually excluded becausethey are not voluntary bodies emerging fromsociety. Italians say that if the state is representedby the palazzo (the palace), civil society is foundin the streets of the piazza (the square). By con-trast, totalitarian regimes contain only rulers andruled, a civic deficit which leaves a difficult chal-lenge for post-totalitarian rulers.

Despite the absence of a communist legacy, thestory is similar in many of the African countriesseeking to nurture new avenues of participationfollowing the retreat of the generals in the 1980s.Again, the act of overthrowing the old order didstimulate significant mass participation, at least inmajor cities, providing echoes of the originalstruggle for independence in the 1950s and 1960s.As Wiseman (1995, p. 5) writes,

the prodemocracy movements of most Africanstates in the late 1980s represented a remarkablecoming-together of political participation by arange of social groups. Prominent among themwere church leaders and professional associationsof lawyers, journalists, students and medicalstaff.

Yet the difficulties of entrenching voluntary par-ticipation in Africa remain substantial. The coreproblem of poverty narrows horizons while illit-eracy provides a barrier to voluntary participation.Further, the national government in many Africanstates has limited functions and weak penetrationoutside the capital. The political culture remainsstrongly parochial. Such participation as emerges,at least beyond voting, is directed towardsinformal politics in ethnic groups and containedthrough patron–client networks. These factors all

suggest that voluntary participation in the newAfrican democracies is unlikely to match even theundemanding levels of consolidated democracies.

Social movements

In 1996, 325,000 Belgians, 3 per cent of thecountry’s population, marched through Brussels ina peaceful protest against an inadequate publicinvestigation into a sexually motivated childmurder. This White March (white was worn as asymbol of purity) was an example of a socialmovement, a form of participation that has cometo mount a significant challenge to the ‘official’political system in established democracies.

Social movements (also called popular move-ments) consist of people from outside the main-stream who come together to seek a commonobjective though an unorthodox challenge to theexisting political order. The White March was atypical movement in that many participants con-sidered themselves to be conventional citizens whowould not ordinarily take an active part in politics.Their protest was both political and anti-political.Social movements espouse a political style whichdistances them from established channels, therebyquestioning the legitimacy as well as the decisionsof the government. The members of social move-ments adopt nonconformist forms of participationsuch as demonstrations, sit-ins, boycotts and polit-ical strikes. Some such acts may be illegal but, incontrast to other criminal behaviour, the actors’motives are political.

DefinitionSocial movements are defined by Tarrow (1998,p. 4) as ‘collective challenges by people withcommon purposes and solidarity in sustainedinteraction with elites, opponents and authori-ties’.They are political entities with a coating ofanti-politics. Rather like protest parties, move-ments can emerge and disappear at speed, afeature which distances them from the morestatic concept of civil society.

We can contrast social movements with bothinterest groups and political parties (Box 8.2).Like parties whose origins lie outside the legisla-

POLITICAL PARTICIPATION 127

Russia is a vast country with an imperial and authori-tarian past. It is a place full of problems, potential andparadoxes. By area Russia is the largest country in theworld, almost twice the size of the United States.Thepopulation includes numerous minorities, with 36national groups containing more than 100, 000people. Russia’s rulers have in the past been autocraticempire-builders, basing their imperial expansion oncontrol of a serf society and (until the communist era)a rural economy.Thus Russia’s experience with com-munist dictatorship represented a culmination of afamiliar authoritarian pattern. Internationally, theUnion of Soviet Socialist Republics (USSR, 1922–91),encompassing 15 European and Asian republics,formed what amounted to a Russian empire.

Of all the post-communist countries, Russia raises themost interesting questions about the relationshipbetween participation and democracy. For manyobservers a country with Russia’s autocratic pastcannot expect to develop the voluntary participationunderpinning established democracies.The country’shistory has bequeathed a political culture based on afatalistic acceptance of strong personal leadership.Russian critics of a competitive party system cite thesaying,‘a child with six nannies is sure to lose an eye’.This desire for a single powerful leader was, it isclaimed, reinforced by the economic decline andethnic divisions of the early post-communist years.Eckstein (1998b, p. 377), for example, suggests that

It would be harder to think of a less likely case forsuccessful democratization than Russia. Support forthe idea of democracy is at best tenuous and instru-mental, it has no strong roots in an accustomedcentral political culture and the social context isinhospitable to it in almost all respects.

Yet this assessment may be too gloomy. In any largecountry, attitudes to politics are neither simple noruniform. Among younger generations, and in theMoscow region, support for democracy is firmer.Thedesire for effective government in post-communistRussia may be a short-term response to social break-down, not an indicator of deep-seated authoritari-anism. Russia’s past may be non-democratic but thisdoes not prevent its educated population learningfrom established democracies elsewhere. Further, bythe mid-1990s Russia had succeeded in holding aseries of reasonably free elections to both the presi-dency and the Duma. Since then, a return to economicgrowth has led to some consolidation of the post-communist order.

Whatever the future may hold for post-communistRussia, it is surely too fatalistic to claim that an author-itarian past rules out the possibility of buildingdemocracy in Russia.The real question, perhaps, is justhow democratic Russia’s ‘democracy’ will prove to be.

Vladimir Putin, president since Boris Yeltsin’s suddenresignation in 1999, has governed in ruthless style,exploiting his influence over the media and arrestingpowerful businessmen who might threaten his posi-tion. Putin is popular and effective but it is doubtfulwhether his rule is subject to the limits associatedwith a fully constitutional democracy.

Population: 145m (and falling).Gross domestic product per head:

$9,300.Form of government: federation of 21

autonomous republics.Executive: formally semi-presidential

but with a strong presidency.Theprime minister heads the Council ofMinisters and succeeds the presidentif needed (no vice-president).

Assembly: the Duma (lower house)contains 450 members elected by anadditional member system.TheFederal Council (upper house) con-tains two members from each of 89geographical units.

Judicial branch: based on civil law.Headed by a Constitutional Court and,for civil and administrative cases, aSupreme Court. Substantial lawlessness.

Natural resources: massive reserves ofoil, gas, coal, timber and minerals.Russia is the world’s second largestexporter of oil.

Environment: extensive pollution,deforestation and contamination(including local radioactivity).

Profile R U S S I A

Further reading: Eckstein (1998b), Petro (1995),Remington (2004), Rose and Munro (2002).

ture, movements emerge from society to challengethe political establishment where their voice isunheard. But movements are also more looselyorganized, typically lacking the precise member-ship, subscriptions and leadership of both partiesand interest groups.

Like interest groups, social movements typicallyfocus on a single issue: for example, nuclear disar-mament, feminism or the environment. Again likeinterest groups, social movements do not seekstate power; rather they seek to influence govern-ment, usually by claiming their voice has previ-ously been ignored. But where protective interestgroups seek precise regulatory objectives, move-ments are more diffuse, often seeking to articulatea common identity, as with gay pride. Socialmovements therefore share the explicitly politicalfocus of parties. However, unlike parties move-ments do not seek to craft distinct interests into anoverall package; rather, they claim the moral highground in one specific area.

Social movements are sometimes described as‘new social movements’ or as practising ‘new poli-tics’. Yet in truth there is little that is new here.Single-issue protest movements must surely be asold as politics itself. Many contemporary politi-cians (and your authors) were schooled in the suc-cessful anti-apartheid and anti-Vietnam protests ofthe 1960s and 1970s. In the 1950s, an earlier gen-eration (but not your authors) wrote about thethreat to political stability posed by mass move-ments such as the Nazis. Kornhauser (1959), for

instance, argued that mass movements were basedon people at the margins of society, includingunemployed intellectuals, isolated workers and theperipheral middle class.

Delving further into the past yields more exam-ples: for example, the anti-slavery, suffragette andtemperance movements. In Sweden, the late nine-teenth century has been labelled the age of associa-tions since this was the era when numerousfolkrörelser (popular mass movements) emerged,initially to challenge the state. Examples includethe farmers’ movement, the temperance move-ment and the free churches. Rothstein (2002, p.299) suggests that if any organizations could besaid to be the owner of Sweden’s extensive stock ofsocial capital, it is the folkrörelser.

However, some features of the contemporaryenvironment certainly encourage social move-ments to flower on a national or even global scale.One such factor is the ease of direct communica-tion among sympathizers. Take the movement inBritain in 2000 against the country’s high tax onpetrol (Joyce, 2002). A diverse network of Britishfarmers and road hauliers blocked petrol refineriesin a coordinated series of protests, quickly creatinga national fuel crisis. This movement made a sharpimpact on both public opinion and the govern-ment. Its capacity to expand so rapidly, albeitbuilding on foundations previously established,owed much to mobile (cell) phones. A contempo-rary newspaper report suggested that in thisprotest

POLITICAL PARTICIPATION 129

Political goals, structure and means Social movements Protective interest Political partiesgroups

Seek to influence the government? Yes Yes Yes

Seek to become the government? No No Yes

Focus on a single issue? Yes Yes No

Formally organized, led and funded? No Yes Yes

Tactics used Unconventional Conventional Conventional

Further reading: Zilliacus (2001). On protective interest groups, see p. 168.

B OX 8 . 2

Comparing social movements, political parties and interest groups

mobile phones played a key part. The organisersran an informal but effective ‘bush telegraph’.Anyone they contacted had the numbers of adozen supporters. And they would each have adozen more. A single call could mobilise hun-dreds of people within an hour

(Doherty et al., 2003, p. 8).

In addition, global communication rendersinternational coordination perfectly feasible.Compared to parties or interest groups, socialmovements cross borders with relative ease. Thebest example here is surely the protest in February2003 against the American-led war in Iraq. Over asingle weekend, an estimated six million peopletook part in demonstrations in 600 or so citiesthroughout the world (Table 8.2.). The ability ofsocial movements to mushroom in this way – evenwithout any central organization or leader – con-firms their capacity to articulate authentic publicconcern at a transnational level. Social movementsare a form of participation that is well-suited to amore global world.

Established democracies hold no monopoly oversocial movements. In the 1970s and 1980s, theyalso became a significant feature in many authori-tarian and developing states. In contrast toWestern democracies, however, these movements

in developing countries have been the territory ofthe poor, as people facing acute problems of dailylife collaborate to improve their living conditionsin a hostile political environment. The urban poororganizing soup kitchens, the inhabitants ofshanty towns lobbying for land reform, groups ofmothers pressing for information on their sonswho ‘disappeared’ under military rule – all wereexamples of this blossoming of popular politicalactivity.

It is natural to imagine that the existence ofpoverty and oppression will provide an encour-aging environment for the collective self-organiza-tion practised by social movements. But inpractice the daily struggle for survival can leavelittle time, energy and resources for politics of anykind. In reality, a policy entrepreneur, oftenexternal to the community concerned, frequentlyemerges to mobilize the downtrodden and tobring their concerns into public focus. The policyentrepreneur’s task is to create social capital bydeveloping resourcefulness among people lackingformal political experience.

DefinitionA policy entrepreneur is a person or group withthe ‘appropriate political experience, vision andresources to help a particular constituency’(Zirakzadeh, 1997, p. 12).When such skills arelacking, human suffering is unlikely by itself togenerate a sustained movement for its ameliora-tion.

The entrepreneur’s role can be played by organi-zations or individuals. In Latin America undermilitary rule, development agencies and radicalpriests in the Catholic Church encouraged thedevelopment of local movements as they sought tocarve political space for themselves by assertingtheir identity as indigenous peoples, gays, femi-nists and environmentalists. Here, external organi-zations played an influential part.

But an inspirational individual can also serve ascatalyst. For example, the contribution of CesarChavez proved decisive in the attempt to unionizeseasonal migrant farm-workers in Californiaduring the 1960s. Previously, these poorly paidfruit-pickers, including many illegal immigrantsfrom Mexico, had been cowed by vigilante groups

130 POLITICS AND SOCIETY

Table 8.2 The ten largest demonstrations againstthe Iraq War, by city, 15–16 February 2003

City Estimated number of protestors

Rome As many as 3,000,000Barcelona 1,300,000London 1,000,000Madrid 600,000Berlin 500,000Paris 200,000Sydney 200,000Damascus As many as 200,000Melbourne 160,000New York 100,000

Note: Figures are estimates reported in the Financial Times, 17February 2003. Protests in smaller countries included Amsterdam(70,000 participants), Oslo (60,000) and Stockholm (35,000).Tocompare with the first Gulf War of 1991, see Koopmans (1999).

acting with the implicit approval of local police.Chavez succeeded in enlisting financial and legalsupport from outside bodies such as churches,universities and the liberal wing of the DemocraticParty. Exploiting media interest, he led a successfulnational boycott of California grapes, eventuallyforcing the farm owners into concessions. In afashion characteristic of many movements, Chavezcreated a cross-class coalition in which those withresources were mobilized to help those without(Jenkins, 1983).

Two cautionary points are necessary in assessingthe overall significance of social movements.First, in the developing world, many community-based movements have lacked the desire or themeans to engage with national politics. A cultureof anti-politics – also found in many post-com-munist countries – limits the movements’ widerimpact. Indeed the democratic transition hastaken some wind from their sails. In Uruguay, forexample, ‘democracy brought, by a curious twist,the disappearance of many grass-roots move-ments that had been active during the years ofdictatorship’ (Canel, 1992, p. 290). Movementsrise and fall but parties and interest groupsremain.

Second, we must beware of exaggerating thecontrast between social movements and orthodoxpolitics. At one level, of course, the whole ratio-nale of the movements is to threaten the domi-nance of existing elites. They provide a ‘people’schallenge’ to the iron triangles of government,protective interest groups and mainstream parties.Yet beneath the surface the distinction is muddied.Kornhauser’s fears notwithstanding, social move-ments seem to have extended the repertoire of par-ticipation more than they have endangeredpolitical stability (Ibarra, 2003). For example, theSwedish folkrörelser have long collaborated withthe state, functioning as schools of democracy forcitizens (Rothstein, 2002, p. 296).

In general, the unconventional activists of theera of protest in the West during the 1960s resem-bled the orthodox activists of an earlier era. In themain, they were well-educated, articulate youngpeople from middle-class backgrounds. And morethan a few leaders of the new politics are switchingto orthodox politics as they age; many a protestactivist of the 1960s had turned into a party leader

by the century’s end. Prominent examples includeJoschka Fischer, Germany’s foreign minister since1998, and Peter Hain, a minister in Tony Blair’sgovernment in Britain. For all the stylistic contrastbetween politics old and new, one function of themovements has been to provide a training groundfor future national leaders. Conversely, protestactivism has entered the mainstream; it is nolonger confined to younger generations (Norris,2002).

Participation in authoritarian states

In contrast to democratic governments, authori-tarian rulers seek either to limit genuine participa-tion in politics or, in the case of totalitarian states,to direct it through tightly controlled channels. Ineither case, the object is to minimize any threatwhich unregimented participation might pose tothe regime. But even in non-democracies, thelimits and nature of participation are often subjectto an implicit dialogue as activists test the bound-aries of the acceptable. Further, as societies gov-erned by authoritarian rulers grow more complex,so rulers often come to realize that responding topopular pressure on non-sensitive issues canenhance political stability.

Patron–client networks

The main technique for controlling participationin authoritarian states is the patron–clientnetwork. Clientelism, as this practice is oftencalled, is a form of political involvement whichdiffers from both voluntary participation in liberaldemocracies and the regimented routines of totali-tarian states. Although patron–client relationshipsare found in all political systems, the developingworld (and especially authoritarian regimes withinit) offers the fullest expression of such relation-ships. In many developing countries, networks ofpatrons and clients are not just the main instru-ment for bringing ordinary people into contactwith formal politics but also the central organizingstructure of politics itself. These networks are thefunctional equivalent of institutions in establisheddemocracies, integrating people through well-understood practices.

POLITICAL PARTICIPATION 131

So what exactly are patron–client relationships?They are informal hierarchies fuelled by exchangesbetween a high-status ‘patron’ and ‘clients’ oflower status. The colloquial phrase ‘big man/smallboy’ conveys the nature of the interaction. Patronsare landlords, employers, party leaders, govern-ment ministers or most often ethnic leaders.Lacking resources of their own, clients gatherround their patron for protection and security.Political patrons control the votes of their clientsand persuade them to attend meetings, join orga-nizations or simply follow their patron around in adeferential manner. In Sri Lanka, for instance,patrons with access to the resources of the statelargely decide how ordinary people vote(Jayanntha, 1991).

DefinitionClientelism is a term used to describe politicssubstantially based on patron–client relation-ships.These relationships are often traditionaland personal, as in the protection provided totenants by landowners in developing countries.But they can also be more instrumental, as withthe resources which dominant parties inAmerican cities provided to new immigrants inexchange for their vote.Where clientelism iscommon, it pervades the political culture,affirming the inequality from which it springs.

In return for their clients’ unqualified support,patrons offer access to jobs, contracts, subsidies,physical protection and even a guarantee of foodin hard times. In other words, they provide secu-rity for poor people who, lacking organized pro-tection through collective insurance or the rule oflaw, are in a vulnerable position. Patrons can thenexploit their local power-base to strike deals withministers in the national government, offering thesupport of their clients in exchange for a share ofthe government’s resources. In this way, patronagenetworks grow and decline according to theirpatron’s skill, just as more orthodox businesses riseand fall with the quality of their management.

The patron’s power, and its inhibiting effect ondemocracy, is nicely illustrated in this quotationfrom Egypt’s President Abdul Nasser, interviewedin 1957 when he was still a reforming leader(Owen, 1993):

We were supposed to have a democratic systembetween 1923 and 1953. But what good wasthis democracy to our people? You have seen thelandowners driving the peasants to the pollingbooths. There they would vote according to theinstructions of their masters. I want the peasantsto be able to say ‘yes’ and ‘no’ without this inany way affecting their livelihood and dailybread. This in my view is the basis for freedomand democracy.

Participation through patronage is a devicewhich appeals particularly in authoritarian settingsbecause it links elite and mass, centre andperiphery, in a context of inequality. Althoughinequality provides the soil in which patronagenetworks flourish, they still act as a political glue,binding the ‘highest of the high’ with the ‘lowestof the low’ through membership of a patron’snetwork.

Military governments

The military governments which populated largesections of the developing world in the second halfof the twentieth century provide useful examplesof participation – or more commonly non-partici-pation – in authoritarian polities. As one mightexpect, most military governments adopted whatRemmer (1989) calls an exclusionary approach topopular participation. These regimes discouragedmass involvement in politics because they fearedits consequences for their own position.

Chile between 1973 and 1989, under GeneralPinochet, was an extreme but revealing case.Pinochet sought not just to govern withoutpopular involvement but to suppress all potentialsources of popular opposition. He exterminated,exiled or imprisoned thousands of labour leadersand left-wing politicians, concentrating power inthe hands of his own military clique. Pinochethimself acted as chief executive while a four-manjunta representing the army, navy, air force andnational police performed legislative functions.

As with juntas elsewhere, the purpose was tomonitor political participation so as to identifythreats to the military’s position, for example fromstudents or organized workers. By such means,ordinary citizens were persuaded to keep their

132 POLITICS AND SOCIETY

heads down and to steer clear of any engagementwith military might.

However, not all military regimes were as brutalas Chile’s. More interesting, perhaps, are the fewwhich adopted Remmer’s inclusionary approach toparticipation. Some military leaders did attempt tobuild a base of support among the political class,and even on occasion among the wider popula-tion. The usual approach here was to seek toexploit the population’s respect for a strong leader.Such inclusionary regimes were often organized onpresidential lines, based around a dominant rulercommitted to national development and a strongcountry.

The modernizing regime of Abdel Nasser inEgypt was such a case. Nasser was an army colonelwho led a coup in 1952, became president in1956 and remained in post until his death in1970. He sought to broaden his political appealbeyond the military, though as a result his regimebecame presidential and bureaucratic, rather thannarrowly military. However, even in Nasser’s Egypt(as in contemporary China) any political competi-tion was confined to elite groups. As Lesch (2003,p. 589) says,

The government demobilized and depoliticizedthe public. The regime offered access to jobs,socioeconomic equity and national indepen-dence: in return, the public was expected to bepolitically quiescent.

So even those military leaders who initially soughtto mobilize the population behind their reformagenda often found that participation declinedover time, resulting in the more limited and con-trolled engagement characteristic of authoritarianregimes.

Totalitarian governments

It is when we turn to totalitarian governments thatwe find the most ambitious attempts to developmass participation in non-democratic systems. Incommunist and fascist states, participation wasboth more extensive and more regimented than indemocracies. This profile of participation – highin quantity, low in quality – was a central featureof totalitarian rule.

For the classic case, we must look back to com-munist states, particularly in their earliest andmost vigorous decades. At first glance, participa-tion left established democracies in the shade.Under communism, citizen activity certainlyoutscored the level found in today’s democracies.Ordinary people sat on comradely courts, admin-istered elections, joined para-police organizationsand served on people’s committees covering localmatters. This apparatus of participation derivedfrom the Marxist idea that all power at every levelof government should be vested in soviets (coun-cils) of workers and peasants.

DefinitionRegimented participation is elite-controlledinvolvement in politics designed to expresspopular support for the notional attempt by therulers to build a new society. Its purpose is tomobilize the masses behind the regime, not toinfluence the personnel or policies of the gov-ernment. Regimented participation was charac-teristic of totalitarian regimes.

However, the calibre of participation in commu-nist states did not match its extent. To ensure thatmass engagement always strengthened the party’sgrip, party members guided all popular participa-tion. At regular meetings of women’s federations,trade unions and youth groups, party activistswould explain policy to the people. But communi-cation flowed only from top to bottom. So themembers of such groups eventually behaved theway they were treated: as passive recipients ratherthan active participants. Because the party con-trolled participation so tightly, cynicism soonreplaced idealism. Only careerist die-hards wereprepared to invest their full energy in the charade.

Eventually, some ruling parties did allow moreparticipation but only in areas that did notthreaten their monopoly of power. Especially inEastern Europe, industrial managers were givenmore say in policy-making as political participa-tion became more authentic on local, specific andeconomic matters. But these modest reforms werenot matched in national politics. Because no realchannels existed for airing grievances, people wereleft with two choices: either to shut up and get onwith life or to express their views outside the

POLITICAL PARTICIPATION 133

system. For all the notional participation, commu-nist governments chose to ignore popular indiffer-ence to their rule.

The trajectory of participation in communistChina illustrates some of these themes. In theearly decades of the People’s Republic, participa-tion through accepted channels was required as away of demonstrating active support for theCommunist Party’s goals. The question was notwhether citizens participated but to what extent.In particular, Mao Zedong developed the doctrineof the ‘mass line’ by which party officials wereexpected to learn from the people while retainingtight control over policy formation.

But Mao became dissatisfied with the caution ofthe political elite emerging from this party-ledframework. In 1966, he launched the CulturalRevolution, encouraging the masses to turn oncorrupt power-holders within the party, the work-place or the family. Even harmless intellectualsfound themselves condemned as ‘the stinkingninth category’. The outcome was an orgy ofuncontrolled participation, including violentscore-settling, which threatened the social fabricand caused permanent damage to the party’s repu-tation.

Following this exceptional episode, political par-ticipation developed along the more predictablelines associated with most authoritarian regimes.After Mao, the leadership gave priority to eco-nomic reform over political purity, discouragingany form of participation which might threatengrowth. Political passivity became acceptable.More recently, the party has opened some socialspace in which sponsored groups in areas such aseducation and the environment can operate withrelative freedom.

However, channels for explicit opposition toparty rule remain closed. Memories remain of theTiananmen Square massacre of 1989, when thearmy’s tanks turned on pro-democracy demonstra-tors in Beijing. Local protests – against corrup-tion, unemployment, illegal levies or non-paymentof wages or pensions – continue but outside theframework of accepted political activity. Thesedemonstrations have not been directed againstnational rulers but they provide one moreheadache for the country’s hard-pressed rulers inBeijing.

Political violence and terror

The forms of participation we have examined sofar operate within a peaceful framework. Evenwhen social movements engage in illegal acts, theirprotests remain largely civil. Yet when orthodoxpolitics leaves conflicts unresolved, and sometimeseven when it does not, the outcome can be vio-lence – by the state against its own people, by citi-zens seeking to change government policy or byone social group against another. To appreciate thefull repertoire of activities falling under theheading of political participation, we must alsoconsider the role of violence in politics.

DefinitionPolitical violence consists of ‘those physicallyinjurious acts directed at persons or propertywhich are intended to further or oppose govern-mental decisions and public policies’(LaPalombara, 1974, p. 379). Political terror, asubcategory of political violence, occurs whensuch acts are aimed at striking fear into a widerpopulation. Both violence and terror can be com-mitted by as well as against the state. Genocide,in which the state is again usually implicated, isthe deliberate extermination of a large propor-tion of a people, nation, race or ethnic group.

The events of September 11, 2001 brought vio-lence and terror into focus (Lutz and Lutz, 2004).The assaults on New York and Washington wereunprecedented in the number and range ofnational origins of the victims but they built on atradition of political violence (Table 8.3). TheWorld Trade Center had first come under attackfrom Islamic terrorists in 1994 when a bomb inthe Center’s car park killed six people. Further, thedangers posed by al-Qaeda were well-recognized:early in 2001, 40 CIA officers were working onthe bin Laden case (US Congress, 2002). And, ofcourse, such activity did not cease on 9/11. Thefollowing year, bombs in the Indonesian resort ofBali killed 202 people, including many Australiantourists. Attacks have also taken place in othercountries, notably Saudi Arabia, Spain and Turkey.So while September 11 was certainly a bolt, it canhardly be said to have come from the blue.Understanding September 11 requires, among

134 POLITICS AND SOCIETY

other things, placing that case of terrorism in acomparative context.

Political violence is as ancient as politics itself.Many terms used today in describing terrorismderive from these earliest cases. For example, theZealots were political activists who resistedRoman rule in Palestine. In the era of theCrusades, the Assassins (literally, hashish-eaters)were a Muslim sect who believed their religiousduty was to hunt down Christians. The Thugswere religiously motivated bandits in central andnorthern India.

The central point about political violence(including 9/11) is that it must be viewed throughthe conventional lenses of political analysis, notthrough a distorting filter that ‘explains’ violencesolely as the product of irrational fanaticism. AsClausewitz said of war, violence is ‘a continuationof politics by other means’. The threat and use offorce is a way of raising the stakes; it extends butrarely replaces conventional politics. Most politicalviolence is neither random nor uncontrolled buttactical. When farmers block a road, or when thesecret police beat up a student activist, or evenwhen terrorists blow up a plane, the act carries adeliberate political signal. Even when so-called

‘uncontrolled’ violence erupts between ethnicgroups, the disturbances are usually initiated –though not carried out – by political leaders.

Campaigns of terror exemplify the use of vio-lence for tactical political ends. By creating aclimate of fear, such terrorist acts are intended tocoerce a wider target into submitting to its aims.To take an influential example, the Reign of Terrorunleashed throughout France after 1789 was adeliberate policy of the Jacobin revolutionaries. AsDavies (1996, p. 706) notes,

The Terror was not confined to the destructionof the revolution’s active opponents. It wasdesigned to create such an atmosphere of fearand uncertainty that the very thought of opposi-tion would be paralysed. It produced a climateof spies, informers and unlimited suspicion.

It is precisely this desire to influence the widerpolitical atmosphere which converts casual bru-tality into political terror. Paradoxically, ‘random’violence is a systematic technique for inducinganxiety among a broader population, a goalachieved by exploiting the oxygen of publicity. Inthat sense, the assault on the United States on9/11 was supremely effective. Watching the col-lapse of the Trade Center towers, television viewersthroughout the Western world said, and wereintended to say, ‘there but for fortune go you or I’.From a political and economic perspective, theimpact of September 11 lay less in the eventsthemselves, severe though these were, than in theinevitable reaction to them. These included animmediate slump in air travel, the cost of addi-tional security measures, America’s invasion ofAfghanistan and secondary terrorism such as theanthrax spores sent through the post to addressesin the United States.

In the aftermath of 2001, it is natural to inter-pret terror as a tactic employed by non-state actorsagainst the state. But again such a conclusionwould be too narrow. Terrorism can also be spon-sored by states against other states, as with Iran,Libya and Syria or as seen in the indirect supportoffered by Afghanistan and Saudi Arabia to the9/11 terrorists.

Most important of all, states can terrorize theirown people. A vital lesson of the twentieth century

POLITICAL PARTICIPATION 135

Table 8.3 World Trade Center fatalities by nationality, September 11, 2001

Nationality Number killed

United States 2106United Kingdom 53India 34Dominican Republic 25Jamaica 21Japan 20China 18Colombia 18Canada 16Germany 16Other 290

Note: Based on over 90 per cent of the death toll of over 2,800 fromaround 80 countries, including those on the planes. Deaths at thePentagon and from the hijacked plane which crashed in Pennsylvaniaare not shown. For an analysis of global responses to terrorism, seeBuckley and Fawn (2003).

Source: Data from New York City Health Department, 18 April 2002.

is that the state is in a unique position to exploitpolitical violence for its own ends. As the monopo-list of authorized coercion, states are well-positionedto intimidate their citizens, and over the course ofthe twentieth century many did so. Preciselybecause political violence is an organized technique,its practice increased with the growth of the state inthe last century. Rummel (1997) estimates thatbetween 1900 and 1987, almost 130 million peoplewere killed by the very institution intended toprovide peace, law and order (Table 8.4).

To demonstrate the link between violence andgovernment, consider the extreme case of genocide.A well-organized state and a population accus-tomed to obedience can make the systematicdestruction of an ethnic group entirely feasible.The Nazi genocide of the Jews is of course the best-known example but it is neither the only case northe most recent. The central African state ofRwanda is a more contemporary example. There,in a few weeks in 1994, most of the minority Tutsi,and some Hutu considered to be sympathizers,were butchered. In total, about 800 000 people, orone in ten of the population, were massacred.

We can use three particular features of the

Rwandan tragedy to illustrate broader characteris-tics of genocide: its planned character, long devel-opment and dependence on the state. First, thebutchery was far from a spontaneous outburst ofethnic hatred. Although Hutu peasants dutifullyimplemented the killings, the orders came fromgovernment and military leaders who told thepeasants to ‘clear the bush’ and ‘pull up the roots’(i.e. kill women and children too).

Second, the Rwandan case illustrates the deep-seated origins of genocide. In Rwanda, the tradi-tional balance between Tutsi and Hutu had beendistorted by Belgian colonialists, contributing tomass slaughters of Hutu in 1972 and 1988, andinitiating the dynamic that culminated in 1994.

Third, an effective system of local rule, alsoinherited from the colonial era, provided theadministrative means to implement genocide. Inthe end, the Tutsi could not escape with their livesbecause they could not escape from the state(Mamdani, 2001).

The cultural residue of officially sanctioned massmurder can be profound, with the emergence oftraumatized, broken societies such as Cambodia (2million dead, 1975–79). How, then, can popula-tions be protected against this threat from the verybody that is supposed to protect them? Rummel(1997) suggests that democracy is the key. Henotes that democracies rarely fight each other andhardly ever kill their own people. In a democracy,the ballot replaces the bullet; it is absolute powerwhich creates the conditions for democide. Thus,the transition to democracy in the final decades ofthe twentieth century may reduce state-sponsoredpolitical violence and render the twenty-firstcentury a time of relative peace in comparisonwith the barbarities of preceding decades.

Rummel’s point may also apply to the form ofterror against the state practised by organizationssuch as al-Qaeda. Were the authoritarian regimesof the Middle East to be replaced by prosperousdemocracies, terrorist organizations would surelyfind the ground beneath them becoming infertile.

Revolution

Occasionally, political violence extends to the gov-erning framework itself as the entire political order

136 POLITICS AND SOCIETY

Table 8.4 The ten most lethal governments,1900–87

Country Years Rate* Total killed(%)

Cambodia 1975–79 8.2 2,000,000Turkey 1919–23 2.6 703,000Yugoslavia 1941–45 2.5 655,000Poland 1945–48 2.0 1,585,000Turkey 1909–18 1.0 1,752,000Czechoslovakia 1945–48 0.5 197,000Mexico 1900–20 0.4 1,417,000USSR 1917–87 0.4 54,769,000Cambodia 1979–87 0.4 230,000Uganda 1971–79 0.3 300,000World 1900–87 0.2 129,909,000

* Per cent of its population that a regime murders each year, onaverage.

Note: Murderous governments after 1987 include Armenia,Azerbaijan, Burundi, Bosnia, Croatia, Iran, Iraq, Rwanda and Somalia.

Source: Rummel (1997).

becomes a matter for dispute. When the existingstructure of power is overthrown, leading to along-term reconstruction of the political, socialand economic order, we can speak of a revolution.Such episodes are rare but pivotal, inducing broadand deep alterations in society. The majorinstances – France, America, Russia, China, Iran –have substantially defined the modern world. Wetherefore conclude this chapter with an assessmentof the nature and causes of revolutions, focusingin particular on France and Russia.

Although changes of the magnitude needed toqualify as a revolution usually require violence, itis debatable whether violence should be built intothe definition of the term. The question here iswhether a ‘peaceful revolution’ is a contradictionin terms. On the one hand, we can certainly makea case for the possibility of revolutions withoutviolence. After all, the ancient world used the term‘revolution’ just to refer to a circulation in theruling group, howsoever induced. To ‘revolve’ isliterally to move round, and in this traditionalsense there was no necessary link between revolu-tion and disorder.

Recent experience confirms that major politicalchanges can occur without large-scale violence.The collapse of communism in Eastern Europe in1989, leading to the fall of the Soviet Union in1991, was a major reform initiated by peacefulmeans, with Czechoslovakia’s Velvet Revolution aparticular case in point. Yet 1989 surely qualifiesas a year of revolutions when such efforts are mea-sured by their impact rather than by the violenceof their birth.

DefinitionSkocpol (1979, p. 4) defines revolutions as ‘rapid,basic transformations of a society’s state andclass structures; and they are accompanied andin part carried through by class-based revoltsfrom below’. Goldstone (1991, p. 37) suggests rev-olutions consist of three overlapping stages: first,state breakdown; second, the ensuing strugglefor power; and third, a radical reconstruction ofthe state.

On the other hand, the contemporary use of theterm ‘revolution’ still connotes transformationthrough violence, a change in meaning which

reflects the experience of the modern world. Afterthe seminal French Revolution of 1789, the worldneeded a special term to capture dramatic, seismicshifts in the social and political order; and ‘revolu-tion’ in the modern sense was born (Lachmann,1997).

Social psychological and structural theories

The twentieth century vastly increased the world’sstock of revolutions (Box 8.3), stimulating asearch for general theories of these events. Here wewill consider two such accounts: first the socialpsychological account associated with Gurr; and,second, Skocpol’s structural interpretation.

The social psychological theory focuses on indi-vidual motivations rather than social groups. Itseeks to answer such questions as: What inspirespeople to participate in revolutionary activity?Why do some people sometimes feel so stronglyabout politics that they are willing to give time,energy and ultimately their lives to achieve change?

In his study of the French Revolution, de

POLITICAL PARTICIPATION 137

Country Year Outcome

Mexico 1910 A populist revolution leading to rule by the Institutional Revolutionary Party (PRI) which continued until 2000

Russia 1917 The world’s first communist state, lasting until 1990

Turkey 1922 A secular nation-state built amid the ruins of the Ottoman Empire

China 1949 The People’s Republic of China, led by a party which is now communist only in name

Iran 1979 An Islamic theocracy led by Ayatollah Khomeini. Religious rulers are now under pressure from more liberal reformers

B OX 8 . 3

A century of revolutions

Tocqueville (1856) noted that grievances patientlyendured become intolerable once a brighter futurebecomes possible. As part of the movementtowards a more scientific approach to politicalanalysis in the 1960s, Gurr (1980) sought todevelop de Tocqueville’s insight.

Gurr argued that relative deprivation was thekey to revolutions. For Gurr, political instabilityonly results from deprivation when combinedwith a belief that conditions are worse than theycould and should be. What matters is not absolutedeprivation, a condition which often breedsresigned passivity, but relative deprivation: a sensethat rewards fall below expectations or entitle-ment. The point of comparison can vary, ofcourse. Deprivation may be relative to past times,to an abstract standard of justice, or to the rewardsaccruing to other groups serving as points of refer-ence. But comparison of some kind is integral tothe idea of relative deprivation.

When relative deprivation is widespread, suggestGurr, instability can result. But Gurr is a littlemore specific. The most explosive situation, heargues, arises when a period of improvement is fol-lowed by a decline in the ability of the regime tomeet rising demands. Such a situation creates adangerous gap between expectations of continuedimprovement and the reality of decline. Theseconditions produce a revolutionary gap betweenexpectations and achievement (Figure 8.2).

Davies (1962) sums up the implications of thispsychological approach: ‘revolutions are mostlikely to occur when a prolonged period of eco-nomic and social development is followed by ashort period of sharp reversal’. This hypothesis isknown as the J-curve theory, with the top of the ‘J’indicating an abrupt halt to a previous period ofrapid growth.

The contribution of this social psychologicalapproach lies in demonstrating that how peopleperceive their condition is more important thanthe actual condition itself. Relative deprivation iscertainly a background factor in many revolutions.Peasant frustrations, in particular, were involved inthe French, Russian and Chinese examples. Note,however, that by citing actual revolutions, we arerestricting our attention to positive cases (p. 73).Relative deprivation may often be extensivewithout signalling uprisings of any kind. Its status

is more likely to be that of a necessary rather thana sufficient condition of revolution.

DefinitionRelative deprivation arises when peoplebelieve they are receiving less (value capability)than they feel they are entitled to (value expecta-tions). Relative deprivation breeds a sense ofresentment which contributes to political dis-content. By contrast, absolute deprivation oftenleads to a struggle for survival and, as a result,political passivity. As the Russian revolutionaryTrotsky (1932/3, p. 103) wrote,‘the mere exis-tence of privations is not enough to cause aninsurrection; if it were, the masses would alwaysbe in revolt’.

Also, although the social psychological accountprovides some insight into the conditions of polit-ical instability and violence, it seems incapable ofexplaining revolutionary progress and outcomes.Whose discontent matters? How and why douprisings turn into revolutions? How is discontentchannelled into organized opposition movements?Why is such opposition usually suppressed butsometimes not? Because relative deprivation hasno answer to these important questions, it is betterregarded as a theory of political violence in generalrather than of revolutions in particular.

138 POLITICS AND SOCIETY

Time

What people expect

What people get

Revolutionarygap

⎫⎪⎪⎬⎪⎪⎭

High

Low

Figure 8.2 The J-curve theory of revolutions

Note: ‘What people expect’ and ‘What people get’ refers to expectationsof government.

Given the limitations of the psychologicalapproach, in the 1970s the study of revolutionsturned away from broad psychological theoriesand returned to a more fine-grained historicalexamination. Skocpol’s (1979) influential discus-sion of the French, Russian and Chinese revolu-tions represented a culmination of this morepolitical structural approach.

For Skocpol, the causes of revolutions cannot befound in the motives of the participants. Whatmatters are the structural conditions: that is, therelationships between groups within a state and,equally important, between states. The back-ground to revolution, suggests Skocpol, is pro-vided by a regime that is weak internationally andineffective domestically. The classic revolutionsoccurred when well-organized revolutionaries suc-ceeded in exploiting peasant frustration with anold order which had lost its capacity to competewith more developed international competitors.With the landed aristocracy resisting economicmodernization, and offering only limited supportto the imperial ambitions of the regime, the oldorder becomes vulnerable to insurrection:

Caught in cross-pressures between domesticclass structures and international exigencies, theautocracies and their centralized administrationsand armies broke apart, opening the way for rev-olutionary transformations spearheaded byrevolts from below.

(Skocpol, 1979, p. 47)

Seizing power from a failing regime can be quitestraightforward. The real revolution begins as thenew rulers develop and impose their vision onsociety and, in particular, on opposition groups.Revolutions do not stop with the taking of power,as the social psychological theory seemed to imply,but only start at this point. Skocpol tells us muchabout how discontent is mobilized into politicalactivity and how that activity is turned into a revo-lutionary transformation. In this way, she bringspolitical, and especially state-centred, analysis backto the study of revolutions.

The French Revolution

Let us conclude by considering the French and

Russian revolutions in the context of the social psy-chological and structural theories we have intro-duced. Our scrutiny must certainly begin withFrance in 1789, as this was the defining revolutionof modern times. Indeed, it is no exaggeration tosay that both the modern concept of revolutionand modernity itself were born in France in 1789.

What, then, were the contours of this landmarkepisode? Before the revolution, France still com-bined an absolute monarchy with feudalism.Governance was a confused patchwork of local,provincial and royal institutions. However, in the1780s the stormclouds gathered: the old regimecame under pressure as the monarchy became vir-tually bankrupt. Then, in 1788, a poor grainharvest triggered both peasant revolts and urbandiscontent. The revolution itself was initiated afterthe Third Estate of the Estates General, convenedin May 1789 for the first time in over 150 years,declared itself to be France’s National Assembly.

There followed a half-decade of radical reformin which, amid the enormous violence of theTerror, the old institutions (including themonarchy) were torn down and the foundationsof a modern state constructed. After a furtherperiod of instability, Napoleon instituted a periodof authoritarian rule lasting from 1799 to 1814.Universal male suffrage was not adopted until1848 and the conflict between radicals and conser-vatives embedded in the revolution remainedimportant to French political debate for the nexttwo hundred years.

Marx described the revolution as a ‘giganticbroom’ sweeping away ‘all manner of medievalrubbish’. The shockwaves of the events in Francecertainly reverberated throughout Europe as rulingclasses in other countries saw their very existenceimperilled. The revolution’s mixed outcomesnotwithstanding, its progressive character is indis-putable:

� Politically, the revolution destroyed absolutemonarchy based on divine right. It establishedthe future shape of liberal democracy: popularsovereignty, a professional bureaucracy and aliberal philosophy.

� Economically, the revolution weakened aristo-cratic control over the peasantry, helping tocreate the conditions under which market rela-

POLITICAL PARTICIPATION 139

tions could spread and capitalism would even-tually emerge.

� Ideologically, the revolution was stronglysecular, fathering individual rights enforceablethrough codified law. It was also powerfullynationalist: the nation became the transcendentbond, uniting all citizens in patriotic fervour.

What light, then, can the social psychologicaland structural theories offered by Davies andSkocpol cast on the French Revolution? Davies’sconcept of relative deprivation certainly helps inunderstanding the peasant revolts, driven as theywere by the conjunction of a failed harvest and aseries of failed government policies which hadraised but not satisfied popular expectations.

But there seems little doubt that Skocpol’s per-spective offers deeper insight. The internationaldimension behind 1789 can be seen in France’sgenerally unsuccessful competition with Englandin the eighteenth century. The limited domesticeffectiveness of the monarchy was then demon-strated by its inability to pay for foreign adven-tures, resulting in a fiscal crisis that came to a headin the late 1780s. Further, Skocpol’s emphasis onthe complex dynamics and long-term effects ofrevolution certainly fits the French experience.The French Revolution became far more than apeasant revolt.

The Russian Revolution

Just as the French Revolution mapped the con-tours of liberal democracy, so the RussianRevolution of October 1917 established theworld’s first communist state. It signalled theadvent of a regime, an ideology and a revolu-tionary movement which sought to overthrowWestern democracy. The Russian Revolution was apivotal event of the twentieth century.

The Russian Revolution swept away the ram-shackle and decaying empire of the Tsar, just as itsFrench counterpart had destroyed medieval rem-nants in that country over a century earlier.Although Russia underwent significant state-spon-sored industrialization towards the end of thenineteenth century, the political frameworkremained conservative and autocratic. Under theTsarist bureaucracy, Russian society had remained

inert. In particular, the mass of poorly educatedpeasants remained locked in serfdom, dependenton the landowning nobility. Russia was an impor-tant imperial power but, domestically, its political,economic and social structures were falling behindits Western competitors.

The country was ripe for revolution and was rec-ognized to be such at the time. But in Russia,unlike France, a dress rehearsal occurred before therevolution proper got under way. An unsuccessfulwar with Japan stimulated a naval mutiny in1905, leading to a failed insurrection. The govern-ment quickly concluded the war with Japan,freeing loyal troops to suppress what the regimeclaimed was a ‘passing squall’ of domestic rebel-lion.

A decade later, the military disasters of the FirstWorld War, and the resulting economic andadministrative chaos within Russia, could not bedismissed so easily. By 1917, the capacity of thecentral government to rule Russia had virtuallydisintegrated. The Tsarist regime collapsed inMarch, to be replaced by a weak provisional gov-ernment. Reflecting Lenin’s decisive leadership,the Bolsheviks succeeded in November inreplacing this administration, though to say thatthey seized ‘power’ would be to understatedomestic disorder (‘Bolshevik’, meaning majority,was the name of a radical faction in the broadMarxist movement).

In Russia, as in France, the revolution was madenot by an insurrection but by the transformationof the political order which followed. Even beforeLenin ended Russia’s involvement in world war in1918, the country was consumed by civil warlasting until 1921. The decisive outcome was thereestablishment of central authority in the form ofa communist dictatorship. The Bolsheviks, despitetheir democratic and anarchist origins, developedinto a communist party wielding a monopoly ofpower. In this fateful transition, Lenin’s notion ofthe vanguard party – an elite body of revolution-aries which claimed to understand the long-terminterests of the working class better than that classitself – provided the crucial rationale.

Why did the first major communist revolutionof the twentieth century occur in one of the less-developed links in the capitalist chain? After all,Marx himself had predicted that such events

140 POLITICS AND SOCIETY

would occur in advanced capitalist countries withan industrial working class, not in largely unmod-ernized agrarian societies such as Russia. Here,Davies’s theory of relative deprivation providesuseful service. Although nineteenth-century Russiaremained a poor country with a serf economy, theTsar did institute important reforms between1860 and 1904, including notional freedom forthe serfs, the introduction of a modern legalsystem, state-sponsored industrialization andgeneral liberalization.

But these reforms served only to induce relativedeprivation. Although the peasants were theoreti-cally free, most remained burdened by debt. Manyflocked to the expanding cities in fruitless searchof a better life. Rapid economic expansion in thefinal decade of the 1890s, stimulated by industrialdevelopment, turned into recession in the firstdecade of the twentieth century as Western moneymarkets contracted. Expectations raised by polit-ical reforms were dashed by the Tsar’s inability topush ahead with further, more radical changes. Inthese ways, reform oscillated with repression in aperfect formula for fostering relative deprivation.

Where Davies is less successful is in explainingwhy an attempted revolution failed in Russia in1905 but another succeeded just twelve years later.At this point, Skocpol’s emphasis on regime col-lapse amid international failure comes back intoplay. With Russia’s armies outclassed by the supe-rior equipment and support available toGermany’s forces, the Tsar’s regime disintegrated

from within, leading to a power vacuum whichLenin’s Bolsheviks were able to exploit during thesummer and autumn of 1917.

Key reading

Next step: Putnam (2000, 2001) is an inter-esting assessment of the decline of social participa-tion in the USA, with an eye to its politicalimpact.

Verba et al. (1995) also examine the United Statesin their major study of participation. For a compa-rable British investigation see Parry et al. (1992).For gender differences specifically, see Burns et al.(2001). On social movements, see Tarrow (1998)for a general account, Zirakzadeh (1997) for acomparative study, della Porta et al. (1999) for theglobal context; and Ibarra (2003) and Norris(2002) for the uncertain relationship between themovements and orthodox democracy. The litera-ture on participation in post-communist societiesis now expanding; on Russia, see Eckstein(1998b). For participation in African transitions,see Bratton and van de Walle (1997, ch. 4). Lutzand Lutz (2004) and Whittaker (2004) examineterrorism while Buckley and Fawn (2003) considerresponses to September 11. On revolutions, seeSkocpol (1979) for a classic analysis, Foran (1997)for theoretical accounts and Halliday (1999) forthe links between revolutions and world politics.

POLITICAL PARTICIPATION 141

This page intentionally left blank

In this part, we examine the formal mechanismsthrough which society influences government. Webegin with the main linking device in establisheddemocracies: elections. We then turn to interestgroups, discussing how such associations protect

their own turf and, in some cases, promote theirvision of the public good. And we conclude withan analysis of political parties, many of which orig-inate in society but which, in power, are chargedwith directing the state and thereby leading society.

Part III

LINKING SOCIETY AND GOVERNMENT

This page intentionally left blank

Scope and franchise 145Electoral systems: legislatures 146Electoral systems: presidents 154Voting behaviour 156Turnout 159Referendums 160Elections in new democracies 163Elections in authoritarian states 164Key reading 165

As Katz (1997, p. 3) writes, ‘elections are thedefining institution of modern democracy’.

During the brief period of an election campaign,voters are the masters and are seen to be so. Toexamine the electoral process is therefore toanalyse the central device which has made repre-sentative democracy a feasible proposition forlarge countries.

The most obvious function of elections is toprovide a competition for office and a means ofholding the winners to account. In addition, anelection campaign permits a dialogue betweenvoters and candidates, and so between society andstate. Like coronations of old, competitive elec-tions also endow the new office-holders withauthority, contributing thereby to the effectivenesswith which duties can be performed (Ginsberg,1982).

We should note, however, that not all electionsare competitive. Most authoritarian rulers main-tain a legislature, and typically employ controlledelections as the means of recruitment to theirassemblies. Even these non-competitive electionscan provide a measure of legitimacy with the inter-national community, as well as a panel of docilerepresentatives who can safely be permitted toraise harmless grievances emanating from theirlocal area.

But it is competitive elections in a democraticsetting that provide the main focus of this chapter.We begin with the neglected issues of the scopeand franchise of elections. We then turn to elec-

toral systems, voting behaviour, turnout and refer-endums. The final sections discuss the specificcharacteristics of elections in new democracies andauthoritarian states.

Scope and franchise

An important question to ask about elections inestablished democracies is how wide is their scope.Which offices are subject to election is almost asimportant as who has the right to vote. Comparethe United States and Britain. The USA is uniquein its massive range of elected offices, rangingfrom the country’s president to the local dog-catcher. In total, the USA possesses more than500,000 elected offices, a figure reflecting a strongtradition of local self-government. Britain wouldneed over 100,000 elected positions to match theAmerican ratio of elected posts to population. Yetin Britain, as in many non-federal democracies inthe EU, voting has traditionally been confined toelections for the European and national parlia-ments and for local councils (however, the Labourgovernment elected in 1997 did introduce electedparliaments to Scotland and Wales). Similarly,Australia has many more elected posts than doesNew Zealand, even when Australia’s larger popula-tion is taken in account.

Other things being equal, the greater thenumber of offices subject to competitive election,the more democratic a political system becomes.However, there are dangers in electionitis. One isvoter fatigue, leading to a fall in attention, turnoutand the quality of choice. In particular, the leastimportant elections tend to become second-ordercontests: that is, their outcomes reflect the popu-larity of national parties even though they do notinstall a national government. The difficulty withsuch second-order contests is that they weaken thelink between performance in office and the voters’response (Anderson and Ward, 1996).

Chapter 9

Elections and voters

145

Many American electors, for example, still vote astraight party ticket for all the offices included ona single ballot. This easy option involves noscrutiny at all of specific contests and can lead to aparty winning many extra posts on the coat-tailsof a popular candidate for the White House.Similar processes operate in Europe. Elections tothe European Parliament become referendums onnational governments, although their supposedpurpose is to elect a member for the EuropeanParliament.

The franchise (who can vote) is another impor-tant, and rather underemphasized, aspect of elec-tions. Following a reduction in the voting age inthe 1960s or 1970s, the franchise in most democ-racies extends to nearly all citizens aged at least 18.A few countries are now contemplating a furtherreduction in the voting age to 16.

This wide franchise is fairly recent, particularlyfor women. Few countries can match Australiaand New Zealand where women have been elec-tors since the start of the twentieth century. Insome countries, women did not gain the vote onthe same terms as men until after 1945, reflectingmale recognition of women’s contribution to thewar effort (Table 8.1).

The main remaining exclusions from suffrageare criminals, the insane and non-citizen residentssuch as guest workers. Yet in each of these areasthere may still be room for further progress.Should the electoral process adopt techniquesenabling people with even severe learning difficul-ties to express preferences? Is denial of the votereally an appropriate response to citizens convictedof a criminal offence?

And how should the franchise respond to inter-national mobility? Should non-citizen residents begranted the vote in the country where they live,work and pay taxes? And should citizens livingoverseas (not just government employees such assoldiers on foreign service) retain the vote in theircountry of citizenship? If we answer yes to bothquestions, migrants would be able to vote in twocountries, giving them twice the electoral weightof the stay-at-homes. Such questions illustrate inpractical form the difficulties in reconciling states,and elections organized in them, with thedemands of a more global world.

Some political systems are adopting a more flex-

ible approach to these contemporary franchiseissues. Within the European Union, citizens ofcountry A who reside in country B can vote inlocal and European, but not national, elections incountry B (Day and Shaw, 2002). And since 1975American citizens living abroad, now numberingover 7 million, retain there vote in federal elec-tions in the USA.

Electoral systems: legislatures

Most controversy about electoral systems centreson the rules for converting votes into seats. Suchrules are as important as they are technical. Theyform the inner workings of democracy, sometimesas little understood by ordinary voters as theengine of a car but just as essential to the opera-tion of the political machine. In this section, weexamine the rules for translating votes into seats inparliamentary elections (Box 9.1).

In elections to the legislature, the main questionis whether an electoral system ensures that theseats obtained by a party are directly proportionalto votes received. Proportional representation (PR)means that a mechanism to achieve this goal isbuilt into the allocation of seats. However, someelectoral systems – namely, the plurality andmajority methods – are non-proportional. Theyoffer no guarantee that parties will receive thesame share of seats as of votes. We examine theolder non-proportional formats before turning toPR.

Plurality and majority systems

In non-proportional systems, parties are notrewarded in proportion to the share of the votethey obtain; instead, ‘the winner takes all’ withineach district, whether a Canadian riding, anAmerican district or a British constituency. Thesesystems take one of two forms: plurality ormajority.

In the plurality (also called ‘first-past-the-post’)format, the winning candidate is simply the onewho receives most votes in each electoral district.A plurality of votes suffices; a majority is unneces-sary. Despite its antiquity and simplicity, the plu-rality system is rare and becoming rarer. It survives

146 LINKING SOCIETY AND GOVERNMENT

ELECTIONS AND VOTERS 147

With the end of the Cold War, the United States becamethe world’s one remaining superpower.This uniquestatus is based partly on the country’s hard power: alarge population, the ability to project military forceanywhere and a dynamic economy accounting for athird of the world’s total production.Yet America’s softpower is also significant. Its leading position in the tech-nology, media and telecommunications sector is under-pinned by a strong base in science and universityeducation. Its culture, brand names and language haveuniversal appeal.

Above all, America retains a faith that it is bound tolead, a confidence demonstrated by its robust reactionto 9/11. Military interventions in Afghanistan (2001) andIraq (2003) confirm that the USA remains willing to useits hard power, even at the risk in the latter case ofdamage to its reputation (Nye, 2002).

The internal politics of the United States is therefore ofvital interest. Ironically, the world’s No. 1 operates apolitical system intended to frustrate decisive policy-making. By constitutional design, power is dividedbetween federal and state governments.The centre isitself fragmented between the executive, legislatureand judiciary. American politics is extraordinarily plural-istic; reforms are more easily blocked by interest groupsthan carried through by the executive.

The president, the only official elected by a nationalconstituency, finds his plans obstructed by a legislaturewhich is among the most powerful and decentralized inthe world. Substantial reforms such as the New Dealrequire a major crisis, such as the Depression, to bringforth that rare consensus which generates rapid reform.Excepting such times of crisis, Washington politics is a

ceaseless quest for that small amount of commonground on which all interests can agree.The presidentmay lead the world but the separation of powers meanthat in domestic politics he is a supplicant like anyother.

Similar paradoxes abound in the American experiencewith elections:

� The United States has over 500,000 elected offices,more than anywhere else, yet turnout is low for mostof them, including the presidency.

� A premise of equality underlies elections yetSouthern blacks were effectively denied the voteuntil the Voting Rights Act of 1965.

� The ‘log cabin to White House’ ideal is widelyaccepted but money is increasingly necessary,though not sufficient, for electoral success.

� Many states use devices of direct democracy (e.g. therecall of the California governor in 2003) but the con-stitution does not allow national referendums.

� Elections are expected to involve debates betweencandidates and parties but in many contests adver-tising by interest groups overwhelms the candidates’voices.

� Elections should create legitimacy for the winner butthe many confusions of the 2000 presidential elec-tion – including the fact that more electors voted forAl Gore than for George W. Bush – hardly contributedto the authority with which the eventual winnerentered the White House.

Population: 290m.Gross domestic product per head:

$37,600.Form of government: a presidential,

federal republic.Legislature: the 435-member House of

Representatives is the lower house.The 100-member Senate, perhaps theworld’s most influential upperchamber, contains two directly electedsenators from each state.

Executive: the president is supported bya massive apparatus, including the400-strong White House Office, and

the Executive Office of the President,numbering around 2,000.The Cabinetis far less significant than in, say,Britain.

Judiciary: the Supreme Court heads adual system of federal and statecourts.This nine-member body cannullify laws and actions runningcounter to the constitution. Manypolitical issues are resolved throughthe court system.

Electoral system: the USA is one of thefew countries still employing the plu-rality method.The president is elected

indirectly through an electoral college.A candidate who wins the key statesmay be elected through this collegeeven if he comes second in thepopular vote, as in 2000.

Party system: the Democratic andRepublican parties show extraordi-nary resilience, despite periodicthreats from third parties.The survivalof the major parties reflects ideolog-ical flexibility, an entrenched positionin law and the bias of plurality elec-tions against minor parties.

Profile T H E U N I T E D S TAT E S

Further reading: McKay (2001b), Wilson and Dilulio(2004)

148 LINKING SOCIETY AND GOVERNMENT

PLURALITY AND MAJORITY SYSTEMS

1. Simple plurality:‘first past the post’Procedure Leading candidate elected on first and

only ballot.Where used Nine countries, including Bangladesh,

Canada, India, UK and USA.

2. Absolute majority: alternative vote (‘preferentialvote’)Procedure Voters rank candidates. If no candidate

wins a majority of first preferences, thebottom candidate is eliminated and his orher votes are redistributed according tosecond preferences. Repeat until a candi-date has a majority.

Where used Australia.

3. Two-ballot systemsProcedure If no candidate wins a majority on the first

ballot, the leading candidates (usually thetop two) face a second, run-off election.

Where used Mali, Ukraine (1994 only). For the FrenchNational Assembly, all candidates winningthe support of more than 12.5 per cent ofthe electorate on the first ballot gothrough to the second round.The candi-date securing most votes wins this addi-tional ballot.

PROPORTIONAL SYSTEMS

4. List systemProcedure Votes are cast for a party’s list of candi-

dates though in many countries theelector can also express support for indi-vidual candidates on the list.

Where used Twenty-nine countries, including Brazil,the Czech Republic, Israel, Netherlands,South Africa and Sweden.

5. Single transferable vote (STV)Procedure Voters rank candidates in order of prefer-

ence. Any candidate needs to achieve a

set number of votes (the quota) to beelected. All candidates are elected whoexceed this quota on first preferences.Their ‘surplus’ votes (that is, the numberby which they exceeded the quota) arethen distributed to the second prefer-ences shown on these ballot papers.When no candidate has reached thequota, the bottom candidate is elimi-nated and these votes are also trans-ferred. Continue until all seats are filled.

Where used Ireland, Estonia (1990 only).

MIXED SYSTEMS

6. Mixed member majoritarian (MMM)Procedure Some candidates are elected for electoral

districts and others through PR. Electorsnormally have two votes. One is for thedistrict election (which usually uses theplurality method) and the other for a PRcontest (usually party list). In MMM, thesetwo tiers are separate, with no mecha-nism to achieve a proportional resultoverall.

Where used Seven countries, including Japan, Russiaand Thailand.

7. Mixed member proportional (MMP) or the‘Additional member system’ (AMS) Procedure As for MMM, except that the two tiers are

linked so as to deliver a proportionaloutcome overall.The party vote deter-mines the number of seats to be won byeach party. Elected candidates are drawnfirst from the party’s winners in the dis-trict contests, topped up as required bycandidates from the party’s list.

Where used Seven countries, including Germany, Italy,Mexico and Venezuela.

Note: The figures for the number of states employing each methodare not based on all countries but just on the 58 democracies (lowerchamber) examined in LeDuc et al. (2002b).

B OX 9 . 1

Electoral systems: legislatures

mainly in Britain and British-influenced statessuch as Canada, various Caribbean islands, Indiaand the United States. However, because India isso populous, around half the world’s people livingunder democratic rule still use first past the post(Lijphart, 1999).

The crucial point about the plurality method isthe bonus in seats it offers to the party leading invotes. In the Canadian election of 2000, forexample, the Liberals gained a majority of seats ona minority of votes (Table 9.1). To see how thisbias operates, consider an example in which justtwo parties, the Reds and the Blues, compete inevery constituency. Suppose the Reds win by onevote in every district. There could not be a closercontest in votes yet the Reds sweep the board inseats. This demonstrates the inherent bias of amethod which, within each district, offers every-thing to the winner and nothing to the losers.

The political significance of this amplifyingeffect lies in its ability to deliver government by asingle majority party. In parliamentary systemswith national parties, the plurality method is agiant conjuring trick, pulling the rabbit ofmajority government out of the hat of a dividedsociety. In Britain as well as Canada, a secure par-liamentary majority for the winner is customaryeven though a majority in the popular vote for asingle party is exceptional. This characteristic isdecisive for those who consider that the functionof an electoral system is to deliver majority gov-ernment by a single party.

It is, however, important to note that this ampli-fier works best when major parties competethroughout the country. The British contestbetween Labour and the Conservatives still fitsthis bill, enabling the swing of the pendulum todeliver a parliamentary majority first for one party,then for the other. But where parties are morefragmented, as in India, majority government isless likely. In India’s increasingly regional partysystem, plurality elections have not deliveredmajority government since 1989. The ability ofthe plurality method to produce majority govern-ment is often exaggerated because of a failure toconsider how the system works beyond its Britishhomeland. Weakening party loyalties, further-more, mean that a national competition betweenjust two strong parties is becoming less prevalent.

The capacity of first past the post to producemajority government, at least given national com-petition between two parties, is often presented asthe method’s greatest strength. But the pluralitysystem also possesses a remarkable weakness. Itoffers no guarantee that the party which wins themost votes will secure the largest number of seatsin the legislature. It is perfectly possible for aparty’s votes to be distributed so efficiently that itwins a majority of seats even with fewer votes thanits main competitor.

Again, take a hypothetical example. Suppose theBlues pile up massive majorities in their own geo-graphical stronghold while the Reds scrape homewith narrow wins throughout the rest of thecountry. The Reds could well win more seatsdespite obtaining fewer votes, reflecting the greaterefficiency of their vote distribution. This possi-bility arises because the average number of votesrequired to win a seat is not constant but variesbetween parties according to how their votes aredistributed across constituencies.

This bizarre situation has arisen twice in postwarBritish general elections. In February 1974, forexample, Labour won four more seats than theConservatives, and formed a minority administra-tion, even though the Conservatives won over200,000 more votes. For Lijphart (1999, p. 134),this possibility of ‘seat victories for parties that aremere runners-up in vote totals is probably the plu-rality method’s gravest democratic deficit’. If wewere designing an electoral system from scratch,we would surely reject a method in which theparty with most votes could fail to win most seats.

In addition, the importance of constituencyboundaries to the result of plurality elections givesincentives for parties to engage in gerrymandering– deliberate manipulation of boundaries so as toimprove the efficiency of a party’s vote.

Definition Gerrymandering is the art of drawing seat boun-daries in plurality elections to maximize the effi-ciency (seats-to-votes ratio) of a party’s support.The term comes from a constituency designed byGovernor Gerry of Massachusetts in 1812. It wasso long, narrow and wiggly that it reminded oneobserver of a salamander – hence gerrymander.

ELECTIONS AND VOTERS 149

The plurality system, familiar to us from Britainand the United States, is not the only form ofnon-proportional representation. There is also aless common but perhaps more democraticversion: the majoritarian method. As its nameimplies, this formula requires a majority of votesfor the winning candidate, an outcome normallyachieved through a second ballot. If no candidatewins a majority on the first round, an additionalballot is held, usually a run-off between the toptwo candidates.

Many countries in Western Europe usedmajority voting before switching to PR early inthe twentieth century. The system remains signifi-cant in France and its ex-colonies. For democrats,the argument for a majoritarian system is intu-itively quite strong: namely, that no candidateshould be elected without being shown to beacceptable to a majority of voters or even, wherethe rules so require, to a majority of electors.

Within the majoritarian category, the alterna-tive vote (AV) is a rather efficient way ofachieving a majority outcome in a single ballotwithin single-member seats. In this system,devised by an American academic in the 1870s,voters rank candidates in order of preference.However, lower preferences only come into play ifno candidate gains a majority of first preferenceson the first count (see Box 9.1). AV is used forAustralia’s lower chamber, the House ofRepresentatives. Compared to simple pluralityvoting, AV takes into account more information

about voters’ preferences but it is not necessarilymore proportional.

Proportional representation

We move now from non-proportional systems toproportional representation. PR is more recentthan non-proportional systems; it emerged in con-tinental Europe towards the end of the nineteenthcentury, stimulated by the founding of associa-tions dedicated to electoral reform. Even so, PR isnow more common than plurality and majoritysystems; it has been the method of choice for mostdemocratic countries since the early 1920s (Blaisand Massicote, 2002, p. 41). PR is the norm inWestern and now Eastern Europe. It also predomi-nates in Latin America.

The guiding principle of PR is to representparties rather than territory. The idea is straight-forward and plausible: namely, that parties shouldbe awarded seats in direct proportion to theirshare of the vote. In a perfectly proportionalsystem, every party would receive the same shareof seats as of votes; 40 per cent of the votes wouldmean 40 per cent of the seats.

Although the mechanics of PR are designedwith the principle of proportionality in mind,most ‘PR’ systems are not perfectly proportional.They usually offer at least some bonus to thelargest party, though less than most non-propor-tional methods, and they also discriminate bydesign or practice against the smallest parties. For

150 LINKING SOCIETY AND GOVERNMENT

Table 9.1 The Canadian elections of 1993 and 2000

1993 2000

Party Votes Number Votes Number(%) of seats (%) of seats

Liberal 42 177 41 173

Conservative 16 2 12 12

New Democratic Party 7 9 9 13

Reform/Alliance 19 52 26 66

Bloc Québécois 14 54 11 37

Total (98) (294) (99) (301)

Note: The bold entries show the extreme bias against the Conservatives in 1993 and the bias in favour of the Liberals in both elections.. Minorparties not shown.

these reasons, it would be wrong to assume thatany system labelled ‘proportional’ must be per-fectly so.

A single party rarely wins a majority of seatsunder PR. Hence majority governments areunusual and coalitions become standard. BecausePR usually leads to post-election negotiations inparliament about which parties will form the nextgovernment, it is best interpreted as a method ofselecting parliaments rather than governments.

How does PR achieve the goal of proportion-ality? The most common method, by far, is the listsystem. An elector votes for a slate of the party’scandidates rather than for just a single person. Thenumber of votes won by a party determines howmany candidates are elected from that party’s list.The order in which candidates appear on the list(decided by the party itself ) governs which peopleare elected to represent that party. For example,suppose a party wins 10 per cent of the vote in anelection to a 150-seat assembly. That party will beentitled to 15 members, who will be the top 15candidates on its list.

List systems vary in how much choice they givevoters between candidates on a party’s list. At oneextreme stand the closed party lists used inPortugal, South Africa and Spain. In these coun-tries, voters have no choice over candidates; theysimply vote for a party. This gives party officialsenormous, and perhaps excessive, control overpolitical recruitment.

However, most list systems do give voters at leastsome choice between candidates. Switzerland andLuxembourg operate exceptionally open lists inwhich electors are given the opportunity to voteeither for a party’s list or for an individual candi-date from the list. But most voters spurn thechoice; they adopt the simple procedure of votingfor a party’s entire slate.

List systems require multimember constituen-cies. Normally, the country is divided into a set ofmultimember districts and seats are allocated sepa-rately within each district. Using constituencies inthis way preserves some territorial basis to repre-sentation but, in the absence of an additionalcompensating device, reduces the proportionalityof the outcome.

The number of members returned per district isknown as the district magnitude. This figure is a

critical influence on how proportional PR systemsare in practice (Lijphart, 1994). As Farrell (2001,p. 79) observes, ‘the basic relationship for all pro-portional systems is: the larger the constituencysize, and hence the larger the district magnitude,the more proportional the result’. For example,Spain is divided into 52 small districts, returningan average of just seven members each. This smallmagnitude means that a party winning a limitedshare of the vote may not be entitled to any seatsat all. Indeed, in 2000 the Popular Party won amajority of seats with just 44 per cent of the vote,confirming the boost which proportional systemswith a small district magnitude give to the leadingparty.

However, in the Netherlands, Israel andSlovakia, the whole country serves as a single largeconstituency, extending proportionality even tosmall parties. The Israeli election of 1999, forexample, saw a total of 15 parties winning seats inthe 120-member Knesset. By dispensing withsmaller electoral districts, such countries eliminatethe vaunted link between constituency and repre-sentative found in single-member pluralitysystems.

DefinitionDistrict magnitude increases with to thenumber of representatives chosen for each electoral district. Under proportional representation, the more representatives to be elected for a district, the more proportionalthe electoral system can be.When the entirecountry serves as a single district, very smallparties can win representation in parliament. Bycontrast, when only three or four members areelected per district, smaller parties often fail towin a seat even with a respectable vote.Thus the‘proportionality’ of a PR system is not fixed butvaries with its district magnitude.

Most list systems add an explicit threshold ofrepresentation below which small parties receiveno seats at all, whatever their entitlement underthe list formula (Table 9.2). Thresholds, operatingat district or national level, help to protect the leg-islature from extremists. As Kostadinova (2002)observes, ‘the threshold is a powerful mechanismfor reducing fragmentation in the assembly. It can

ELECTIONS AND VOTERS 151

be and is manipulated by elites to cut off access toparliament for smaller parties.’

Turkey is a case in point. Its demandingthreshold of 10 per cent of votes cast was designedby the country’s secular rulers to exclude smallIslamic parties. However, such a high hurdle canhave capricious results. In the 2002 election – wonironically by a party with Islamic origins – over 40per cent of voters supported a party which did notwin any seats in parliament. Here is a striking caseof ‘PR’ delivering a most unproportional, andseemingly unfair, outcome.

Mixed systems

Plurality and PR systems are usually consideredalternatives yet a hybrid form has emerged whichcombines the two. This mixed method seeks thebest of both worlds (Shugart and Wattenberg,2000). It combines the geographical representa-tion of the plurality method with the party repre-sentation of PR.

Germany is the source of this ingenious com-promise. There, electors have two votes: one for adistrict candidate and the other for a regionalparty list. Half the seats in the Bundestag are filled

by candidates elected by plurality voting withineach electoral district. However, the party list voteis more important because it determines the totalnumber of seats to be awarded to each partywithin each region. Candidates from the party’slist are used to top up its directly elected candi-dates until the correct – that is, proportional –number of seats is achieved within each region.Should a party win more district seats than itsentitlement under the party vote, it retains theextra seats and the Bundestag expands in size. In2002, the SPD won four excess mandates, as theyare called, and the CDU one (Scarrow, 2002a).

Table 9.3 shows how the German electoralsystem operated in the 2002 election. Note thatwithout the party list component, the FreeDemocrats would not be represented in theBundestag at all. Similarly, the Greens would haveonly one seat, precluding them from achievingtheir current elevated position as a coalitionpartner of the SPD in Europe’s largest and wealth-iest democracy.

Because of the additional members drawn fromthe party list, Germany’s format is often describedas an additional member system (AMS). The com-promise character of AMS has encouraged othercountries, including Italy and New Zealand, toexperiment with similar methods, though some-times varying the balance between district and listmembers. The system has also been adopted forthe Scottish and Welsh assemblies in the UK.

Other countries have introduced the idea of sep-arate district and party list votes but without anytop-up device to achieve a proportional outcomeoverall. Countries using this non-proportionalmixed member majoritarian (MMM) systeminclude Japan, Thailand, Russia and several otherpost-communist countries.

Assessing electoral systems

In assessing electoral systems, an issue arising isthe relationship between electoral systems andparty systems. This topic remains a matter of con-troversy. In a classic work, Duverger (1954, p.217) argued that ‘an almost complete correlationis observable’ between the plurality method and atwo-party system; he suggested that this relation-ship, based on mechanical and psychological

152 LINKING SOCIETY AND GOVERNMENT

Table 9.2 Explicit thresholds of representation insome PR systems

Country Threshold (minimum share of total vote required to be awarded any list seats)

Turkey 10%

Moldova 6%

Czech Republic,Germany, Poland 5%

Hungary, Norway,Sweden 4%

Denmark 2%

Note: Hungary and Germany use mixed electoral systems. Germanparties winning three district contests achieve representation in theBundestag even if they fall below the 5% threshold on the partyvote.

Sources: Jasiewicz (2003), LeDuc et al. (2002a).

effects, approached that of ‘a true sociological law’.At the time, Duverger’s law seemed to favour theplurality system since PR and its accompanyingmultiparty systems were still found guilty by theirassociation with unstable coalition governments.

DefinitionDuverger (1954) distinguished two effects ofelectoral systems.The mechanical effect arisesdirectly from the rules converting votes intoseats. An example is the threshold for represen-tation used in many proportional systems.Thepsychological effect is the impact of the ruleson how electors cast their votes. For example, theplurality system used in Britain and the USA dis-courages electors from supporting minor partieswith little realistic chance of victory in the voter’shome district.

But in the 1960s a reaction set in againstattributing weight to political institutions such aselectoral systems. Writers such as Rokkan (1970)

adopted a more sociological approach, pointingout that social cleavages had produced multipartysystems in Europe long before PR was adoptedearly in the twentieth century. Bringing the argu-ment up to date, Jasiewicz (2003, p. 182) makesexactly the same point about post-communistEurope: ‘political fragmentation usually precededthe adoption of a PR-based voting system, notvice versa’.

Furthermore, Duverger’s law has declined inauthority with the emergence of more regionalparty systems in two countries still using pluralityelections: India and Canada. In a recent compara-tive study, Norris (2004, p. 164) does find that theevidence ‘supports Duverger’s generalization thatplurality electoral systems tend towards partydualism while PR is associated with multipar-tyism’. However, the observed difference is small.Electoral systems are only one of several influenceson the number of parties achieving representation.

Much ink has also been used on a related issue:the question of which is the ‘best’ electoral system.

ELECTIONS AND VOTERS 153

Table 9.3 How the additional member system works: the German federal election of September 2002

A B C D EParty Party list vote Number (and %) of Number of list seats Final number

(%) constituency seats awarded (to bring (and %) of seats won column E more in line for each party

with column B) (C + D)

Social Democrats (SPD) 38.5 171 (57.2%) 80 251 (41.7%)

Christian Democrats/Christian Social Union (CDU/CSU) 38.5 125 (41.8%) 123 248 (41.1%)

Greens 8.6 1 (0.3%) 54 55 (9.1%)

Free Democrats (FDP) 7.4 0 (–) 47 47 (7.8%)

Party of Democratic Socialism (PDS) 4.0 2 (0.7%) 0 2 (0.3%)

Others 3.0 0 (–) 0 0

Total 100 299 (100%) 304 603 (100%)

Notes:1. The PDS did not pass the 5% threshold required for the award of seats from its party list.2. The table shows the national outcome from an allocation of list seats which takes place at regional level.3. Turnout: 79.1%. Spoilt ballots: 1.2% (party list vote), 1.5% (constituency vote).

Source: Federal Statistical Office at http://www.destatis.de.

In truth there is no such thing; different horsesrun best on different courses. For example, incountries with intense social divisions such asNorthern Ireland, PR will provide at least somerepresentation for parties based on minoritygroups. In this way, the risk of majority tyranny isreduced.

By contrast, in the context of national competi-tion between two main parties, plurality systemsusually result in government by a single victoriousparty. Each new government can implement itspolicies without needing to water them downthrough compromises with coalition partners.This philosophy is seen in the UK, where thelosing party is willing to trust its opponents to geton with the task of governing, secure in theknowledge that its own time will come again.

It must be acknowledged, though, that the plu-rality method can thrust parties in and out ofoffice in an erratic and exaggerated way. Canada isa good example. In 1984, the Conservatives wonthree-quarters of the seats on just half the vote.Nine years later they were unmade in an aston-ishing election which reduced their representationto just two seats despite retaining 16 per centsupport (Table 9.1).

Whatever their theoretical weaknesses, electoralsystems tend to persist once established by afounding election. After all, parties elected topower under one system have no incentive tochange to another. Yet although electoral reform isuncommon, it is far from unknown. Japan, Italyand New Zealand each changed their electoralsystem in the 1990s. All three countries adoptedthe fashionable additional member system. Thesecountries therefore provide useful case studies ofthe impact of electoral reform. Over the shortterm, the effects seemed to be limited:

� In Japan, the object was to reduce the signifi-cance of money in elections by ending anunusual system that forced candidates from thesame party into competition with each other.Yet even today factionalism and corruptioncontinue to inhibit real policy debate (Reed andThies, 2002).

� Reformers in Italy wanted to escape from theunstable coalitions produced by PR. In theirplace, reformers sought to encourage a small

number of large parties that would alternate inpower, British-style. Yet the first governmentproduced under Italy’s new electoral law lastedjust eight months and the parties still rely onregional rather than national support (Katz,2002).

� In New Zealand, the motive was to reduce theunrestrained power of the single-party govern-ments produced by the plurality method.Again, however, the first parliamentary termunder the new system was hardly a greatsuccess. It was marked by a fractious andunstable coalition involving New Zealand First,a new party dominated by a leader who hadbeen expelled from the larger National Party(Denemark, 2002).

Over the medium term, however, electoralreform in these countries may deliver at least someof the desired benefits. Japanese debate nowaccepts the weaknesses of factionalism, even if fac-tions themselves remain robust. New Zealand isgrowing accustomed to coalitions and in 1999 itselectorate voted New Zealand First not just out ofoffice but almost out of parliament.

But it is Italy that provides perhaps the bestrecent example of electoral reform achieving someof its objectives. Although the country continuesto be governed by coalitions of regional parties,these groupings now represent broad left- andright-wing traditions and have achieved an alter-nation in office. As Newell and Bull (2002, p. 26)note,

The Italian election of 2001 resulted in alterna-tion in office between a coalition of the centre-right and an incumbent centre-left governmentseeking re-election. This was the first time thishad happened since the birth of the ItalianRepublic . . . The election brought to office apre-constituted coalition with an overallmajority.

Electoral systems: presidents

Electoral systems for choosing presidents receiveless attention than those for electing legislatures.Yet most countries have an elected president and

154 LINKING SOCIETY AND GOVERNMENT

sometimes, as in the Americas, the occupant hasclear executive powers. So this rather neglectedsubject certainly deserves attention.

In one sense, the rules for electing presidents arestraightforward. Unlike seats in parliament, a one-person presidency cannot be shared betweenparties; the office is indivisible. So PR is impos-sible and the main choice is between the pluralityand the majority method. However, in anothersense, presidential electoral systems are more com-plicated: many, including the USA, are based onindirect election through a special college.

We consider directly elected presidents first. AsFigure 9.1 shows, 61 of the 91 directly-electedpresidents in the world are chosen by a majoritysystem. The reason, presumably, is that it is moreimportant to confirm majority backing for a singlepresident than for every single member of a legis-lature. Most majority elections for president use asecond ballot of the top two candidates if neitherwins a majority of votes in the first round. France,once more, is a leading case.

Plurality contests, in which the candidate withmost votes on the first and only round wins, areless common. This technique saves the expense oftwo ballots but the winner may receive only asmall share of the vote when several serious candi-dates emerge. To take an extreme illustration,General Banzer won the Bolivian election of 1997with just 20 per cent support, hardly a ringingendorsement for a national chief executive.

A further requirement used in a few presidentialelections is to impose a turnout threshold of 50per cent. If turnout falls below that level, thecontest begins anew. Such requirements werecommon in communist states, where the partycould dragoon electors to the polls, but are a dan-gerous requirement in free elections. In the winterof 2002/03, Serbia and Montenegro each experi-enced two failed presidential elections.

As Figure 9.1 also shows, almost a third of presi-dents manage to avoid the perils of direct electionaltogether. Many of these are chosen via indirectelection where a special body (which may itself beelected) supposedly acts as a buffer against thewhims of the people.

The United States was once an example. Most ofthe Founding Fathers opposed direct election ofthe president, fearing the dangers of democracy.

After inconclusive debate at the consitutional con-vention, they eventually settled on a form of indi-rect election: an electoral college with delegatesselected by each state legislature as it saw fit. Inchoosing the president, the delegates were expectedto show wisdom beyond that of ordinary voters.Today, the college survives only as a proceduralrelic, with nearly all delegates routinely voting forthe candidate to which they are pledged.

However, votes for American presidential candi-dates are still counted by state, with the leadingcandidate in a state winning all its votes in thecollege. This federal feature means that a candi-date can still gain the White House despitereceiving fewer votes than his main opponent.Four presidential elections have led to just such anoutcome. In 2000, not only did Al Gore win morevotes than Bush, he would also probably have wonhad the United States employed the majority run-off method. Such an outcome would have resultedfrom first-round voters for the Green candidateswitching to Gore in the run-off. So we stillcannot regard the American format for electing

ELECTIONS AND VOTERS 155

Monarchies(38)

Other(12)

Presidentindirectlyelected or

non-elected(41)

Mixed orother

method(10)

Figure 9.1 Methods for selecting presidents

Pluralitymethod

(20)

Republics(132)

Countriesincluded in

study(170)

Presidentdirectlyelected

(91)

Majoritymethod

(61)

Majorityrun-off

(49)

Source: Blais, Massicotte and Dobrzynska (1997).

presidents as a straight forward example of directelection using the plurality method (Dahl, 2001,p. 81).

DefinitionIndirect election occurs when office-holders areelected by a body which has itself been chosenby a wider constituency. Indirect election waswidely used within communist parties as adevice for limiting democratic expression; eachlevel only elected the next level up.The device isalso employed in many presidential electionsand for upper houses of parliament.

Three other features of presidential elections areworthy of note: the length of the term, the possi-bility of reelection and the link with other elec-tions. On the first of these points, the presidentialterm is normally no shorter, and sometimeslonger, than for parliament. The longer the term,the easier it is for presidents to adopt a broad per-spective free from the immediate burden of reelec-tion. At just four years, the term of office ofAmerican presidents is unusually short. Thedanger is that year one is spent acquiring experi-ence and year four seeking reelection, leaving onlythe middle period as years of accomplishment.

Second, term limits are often imposed,restricting the incumbent to just one or twoperiods in office (Box 9.2). The fear is thatwithout such constraints presidents will be able toexploit their unique position to remain in officetoo long. Thus, the USA introduced a two-termlimit after Franklin Roosevelt won four electionsin a row between 1932 and 1944. Mexican presi-dents (like the deputies in parliament) cannotstand for reelection.

As with many institutional fixes, term limitssolve one problem at the cost of creating another.Lame-duck presidents lose clout as their termnears its end. For instance, the Korean financialcrisis of 1997 coincided with the end of the presi-dent’s non-renewable six-year tenure, adding touncertainty. Popular presidents, replete with confi-dence and experience, may be debarred fromoffice at the peak of their power. Also, presidentssubject to term limits sometimes seek to alter theconstitution in order to permit their own reelec-tion, a regular source of friction in Latin America.

Third, the timing of presidential elections is alsoimportant. When they occur alongside elections tothe assembly, the successful candidate is morelikely to be drawn from the same party as domi-nates the legislature. Without threatening the sep-aration of powers, concurrent elections limit thefragmentation of the presidential executive (Jones,1995a). This thinking lay behind the reduction ofthe French president’s term to five years in 2000,the same tenure as the assembly.

Voting behaviour

Given that voters have a choice, how do theydecide who to vote for? Although this is the mostintensively studied question in all political science,there is no single answer. As a broad summary,however, since 1945 electors in the establisheddemocracies have moved away from group andparty voting towards voting on issues, theeconomy, leaders and party competence. Franklin(1992) describes this process as ‘the decline ofcleavage politics and the rise of issue voting’.

For two decades after the Second World War,most studies of electoral behaviour disputed theintuitive proposition that voters do ‘choose’ whichparty to support. An influential theory of electoralchoice, originally developed in the United Statesin the 1950s, argued that voting was an act ofaffirmation rather than choice (Campbell et al.,1960). Voting was seen as an expression of aloyalty to a party, a commitment which was bothdeep-seated and long-lasting. This ‘party identifi-cation’, as it was termed, was acquired initiallythrough one’s family and then reinforced throughmembership of politically uniform social groups(for example, colleagues at work).

In the USA, party attachment was confirmed bythe traditional requirement to register as a partysupporter to be eligible to vote in its primaries.American electors learned – or were taught – tothink of themselves as Democrats, Republicans or,in a minority of cases, as Independents. This viewof the American voter as an habitual supporter ofa particular party is variously called the socializa-tion, Michigan or party identification model.

In Europe voting expressed loyalty to a socialgroup rather than a party. The act of voting flowed

156 LINKING SOCIETY AND GOVERNMENT

from one’s identification with a particular religion,class or ethnic group. Thus electors thought ofthemselves as Catholic or Protestant, middle-classor working-class; and they voted for parties whichexplicitly stood for these interests. In short, socialidentity anchored party choice. But whether theemphasis was placed on identification with theparty (as in the USA) or with the social group (asin Europe), voting was viewed more as a reflexthan as a choice. The electoral ‘decision’ was aningrained habit.

DefinitionParty identification is a long-term attachmentto a particular party which anchors voters’ inter-pretations of the remote world of politics. Partyidentification is often inherited through thefamily and reinforced by the elector’s socialmilieu. It influences, but is separate from, elec-toral choice.The stability of party identificationwas used to explain the continuity of Westernparty systems in the 1950s and 1960s.

These models of party and group voting hadbecome less useful by the 1960s and 1970s. Thosedecades witnessed partisan dealignment: that is,the weakening of the ties which once boundvoters, social groups and political parties together.Particularly in the third quarter of the century, the

proportion of party identifiers declined in manyestablished democracies (Box 9.3).

What caused this decline in party loyalties? Onefactor was political: the decay was not uniform buttended to be focused on periods of disillusionmentwith governing parties. In the USA, for instance,the fall of party identification was sharpest duringthe period of the Vietnam War and the associatedstudent protests.

Another factor in party dealignment was thedeclining capacity of social cleavages to fashionelectoral choice. In Europe, class and religiousidentities became less relevant to young, well-edu-cated people living in urban, mobile and moresecular societies. Class voting, in fact, declinedthroughout the democratic world, allowingDalton (2002, p. 193) to conclude that ‘class-based voting . . . currently has limited influence instructuring voting choices’. New divisions, such asthat between employees in the public and privatesectors, cut across rather than reinforced the tradi-tional class cleavage (Knutsen, 2001). Television –more neutral and leader-centred than the press –also loosened old party loyalties.

The consequences of dealignment are stillemerging but will surely be substantial. Much ofthe democratic world has already witnessed theemergence of new parties with a more radicalcomplexion, such as the Greens, and a decline not

ELECTIONS AND VOTERS 157

Country Method of election Term (years) Reelection permitted?

Argentina Electoral college 6 After one term out

Brazil Run-off 5 After one term out

Finland Plurality 6 Yes

France Run-off 5* Yes

Mexico Plurality 6 No

Russia Run-off 5 One term out required after two terms served

United States Electoral college 4 Two-term limit

* Reduced from seven years by constitutional amendment in 2000.

Source: Adapted from Jones (1995b). See also Nurmi and Nurmi (2002).

B OX 9 . 2

Methods for electing presidents: some examples

just in turnout but also in active participation incampaigns. Electoral volatility has grown, split-ticket voting (where relevant) has increased andmore electors are deciding how to vote closer toelection day. Also, candidates and leaders seem tohave grown in importance relative to the partiesthey represent (Dalton et al., 2000).

However, in none of these areas have we witnessed a complete transformation. In mostdemocracies, most of the time, most electors stillgo to the polls to support the same party forwhich they have always voted. Dealignment, we should remind ourselves, refers to the weak-ening of existing patterns and not to either theirdisappearance (non-alignment) or the creation ofnew links between cleavages and parties (realign-ment).

DefinitionPartisan dealignment refers to the weakeningof bonds between (a) electors and parties, and(b) social groups and parties. In most establisheddemocracies, such links have declined instrength but they have not disappeared; elec-torates are dealigning rather than dealigned.

The decay of group and party voting has ledpolitical scientists to focus on the question of howvoters do now decide. The contemporaryemphasis is on four factors: political issues, theeconomy, party leaders and party image. Fiorina’stheory (1981) of retrospective voting capturesmany of these themes. As developed by Fiorina,retrospective voting means casting one’s ballot inresponse to government performance; his phrasetells us much about the character of contempo-rary electoral behaviour. Electors do form ageneral assessment of the government’s recordand, increasingly, they vote accordingly. A vote isno longer an expression of a lifelong commit-ment, rather it is becoming a piece of businesslike any other. The elector asks of the govern-ment, ‘what have you done for me (and thecountry) lately?’

Retrospective voting helps to explain why eco-nomic conditions, particularly disposable income,unemployment and inflation, seem to have such aconsistent impact on the popularity of govern-ments (Dorussen and Taylor, 2002). More voters

now proceed on the brutal assumption that gov-ernments should be punished for bad times andperhaps also rewarded for economic advance.Especially where a single party forms the govern-ment, more voters are happy to judge by results;they are now fairweather friends only (Nadeau etal., 2002). The feel-good factor, however, is notjust a matter of objective economic performance;voters’ perceptions of the economy are the keybattleground and here politicians have some roomfor manoeuvre.

In this more pragmatic era, electors assess thegeneral competence of parties. Increasingly, theyask not just what a party proposes to do but alsohow well it will do it. Given that parties are lessrooted in ideology and social groups than in thepast, their reputation for competence in meetingthe unpredictable demands of office becomes acrucial marketing asset. So party image becomescrucial. The skill is to generate trust in one’s ownside and especially to create doubts about one’sopponents. Given volatile and sceptical voters,gaining credibility – or at least more than one’s

158 LINKING SOCIETY AND GOVERNMENT

Decline over 10% Decline of 1–10% No decline

Austria Australia Belgium

Canada Finland Denmark

France Japan

Germany Netherlands

Ireland New Zealand

Italy Norway

Sweden United Kingdom

USA

Note: Figures are based on a standard survey question askingpeople whether they ‘think of themselves’ as, for example, aDemocrat or a Republican. Decline is measured between an initialsurvey (1967-78, depending on country) and a later survey (1991-98).

Sources: Bentley et al., (2000); Dalton and Wattenberg (2000).

B OX 9 . 3

Decline in party identification,1970s–1990s

opponents – is the cardinal objective (Bowler andFarrell, 1992).

Turnout

Despite rising levels of education, turnout isfalling throughout most of the democratic world.(Table 9.4). In 19 democracies, turnout fell onaverage by 10 per cent between the 1950s and the1990s (Wattenberg, 2000). For example:

� In the USA, turnout among the adult popula-tion declined from 63 per cent in the presiden-tial election of 1960 to 49 per cent in 1996,recovering but modestly to 51 per cent in theclose contest of 2000.

� Turnout among registered electors at Britishgeneral elections declined from 78 per cent in1992 to 59 per cent in 2001, a loss of 7 millionvoters in just nine years.

� In national elections in Switzerland, as in manycountries at regional and local contests,abstainers now comprise the majority.

Why has turnout fallen? Precisely because thephenomenon is so widespread, it is difficult topinpoint its causes. However, there seems littledoubt that declining turnout forms part of a widertrend in the democratic world: namely, a growing

distance between voters, on the one hand, andparties and government, on the other. It is surelyno coincidence that turnout has fallen as partydealignment has gathered pace, as party member-ship has fallen and as the class and religious cleav-ages which sustained party loyalties in the earlypostwar decades have decayed.

Franklin (2002, p. 174) links the decline ofturnout to the diminishing significance of elec-tions. He suggests that the success of manydemocracies in establishing welfare states andachieving full employment in the postwar eraresolved long-standing conflicts between capitaland labour. With class conflict disarmed, electorshad fewer incentives to vote on election day. AsFranklin writes, ‘elections in recent years mayshow lower turnout for the simple reason thatthese elections decide issues of lesser importancethan elections did in the late 1950s’.

Declining satisfaction with the performance ofdemocratic governments has also played a part(Norris, 1999a). Especially in the 1980s, populartrust in government and parties fell in manydemocracies, reflecting the growing complexity,internationalization and corruption of governance.Even though mass support for democratic princi-ples remains strong, growing cynicism about gov-ernment performance has probably led morepeople to stay away from the polls (Putnam et al.,2000).

Declining turnout does not seem to reflect anydecline in political interest. On the contrary,Dalton et al. (2000, p. 56) show that withincreasing education ‘political interest is generallyincreasing over time’. Rather, electoral participa-tion has been caught up in wider shifts in polit-ical behaviour. Younger generations increasinglyfollow and discuss events as forms of participa-tion in themselves. In addition, because youngercohorts tend to view elections as part of the official political system, participating in them has less appeal than joining informal social movements.

Turnout varies not just over time but alsobetween countries. How then can we explaincross-national difference in turnout? Here a cost–benefit analysis is useful (Downs, 1957). That is,turnout tends to be higher in those countrieswhere the costs or effort of voting are low and the

ELECTIONS AND VOTERS 159

Table 9.4 Declining turnout at national elections,1950s–1990s

Decline over 10% Decline of 1–10% No decline

Austria Australia DenmarkFrance Belgium SwedenJapan CanadaNew Zealand FinlandSwitzerland GermanyUnited Kingdom IrelandUSA Netherlands Norway

Note: Figures are based on a comparison between the first twoelections in the 1950s and the last two elections before 2000. Mostof the fall occurred in the late 1980s and the 1990s.

Source: Adapted from Bentley et al. (2000) and Wattenberg (2000).

perceived benefits high (Box 9.4). On the costside, turnout is greater in those countries wherevoting is permitted through the mail and over theweekend. It is reduced when the citizen is requiredto take the initiative in registering as an electorwith the local administration, as in the USA. Onthe benefit side, proportional representationenhances voting by ensuring each ballot affects theoutcome (Lijphart, 1999). Also, the more impor-tance people attach to an election, the higher theturnout; presidential contests, for example, typi-cally secure a higher ballot than parliamentarycontests (Norris, 2003).

At the level of the individual, variations inturnout reflect the pattern found with other formsof political participation. Specifically, high turnoutreflects political resources and political interest.Thus, educated, affluent, middle-aged citizenswith a strong party identification, and those whobelong to a church or a trade union, are particu-larly prone to vote.

By contrast, abstention is most frequent amongpoorly educated, unemployed young people whobelong to few organizations and who have noparty ties. Minorities are particularly likely toabstain. For instance, Hispanics form 10 per centof the American population but provided only 4per cent of the voters in the American presidentialelection of 2000 (Conway, 2001, p. 81).

Referendums

Elections are instruments of representative democ-racy; the role of the people is only to decide whowill decide. By contrast, referendums, and similardevices such as the initiative and the recall, aredevices of direct democracy, enabling voters todecide issues themselves. A referendum involves areference from another body, normally the legisla-ture or the government, to the people for resolu-tion. The device therefore provides a practicalcounter-example to the common argument thatdirect democracy is impossible in large states.

Further, technology has opened up the possi-bility of voting in referendums through conve-nient electronic devices such as digital television,personal computers and mobile phones (Budge,1996). Referendum democracy has become tech-

nically feasible; the question now is whether it ispolitically wise.

Referendums vary in their status. Their outcomemay be binding, as with constitutional amend-ments requiring popular approval, or merely con-sultative, as with Sweden’s vote in 1994 onmembership of the European Union. A bindingreferendum will normally be triggered automati-cally under the constitution whereas a consultativereferendum is typically an option for a hesitantgovernment.

In a few countries (now including NewZealand), referendums can also be initiated bycitizen petition, a device extending popular influ-ence to the political agenda itself. The initiative, assuch citizen-initiated ballots are known, is usedwidely in Switzerland. It has also been adopted bymany Western states in the USA, notablyCalifornia. For instance, Proposition 13 in 1978limited property taxes in California, launching asequence of taxpayers’ revolts in Western democra-cies.

We should note two unusual forms of the initia-tive. First, Italy uses what has become known asthe abrogative referendum. Five regional councils

160 LINKING SOCIETY AND GOVERNMENT

Features of the electoral system Features of individuals

Compulsory voting Middle age

Proportional representation Strong party loyalty

Postal voting permitted Extensive education

Weekend polling Attends church

Elections decide who governs* Belongs to a union

Automatic registration Higher income

* Examples of elections which do not decide who governs are thoseto the American Congress and the European parliament.

Source: Franklin (2002).

B OX 9 . 4

Features of the electoral systemand of individuals which increaseturnout

ELECTIONS AND VOTERS 161

As turnout declines across most of the democraticworld, so attention focuses on potential solutions.Technical fixes, such as putting polling stations insupermarkets and permitting voting by email, mayhave a role to play. Countries using plurality or majorityvoting could also boost turnout by introducing pro-portional representation, in which every vote counts.But has the time come to consider the most radicaland effective solution of all: compulsory voting?

The case for Most citizens acknowledge obligations to the statesuch as paying taxes, obeying the law and evenfighting in war.Why therefore should they not acceptwhat Hill (2002) calls the ‘light obligation and unde-manding duty’ of voting at national elections?Currently, abstainers take a free ride on the backs ofthe conscientious.

By definition, an election based on a full turnoutwould be representative of the population and ensurea measure of political equality. It would enhance theauthority of the government, a collective benefit. Ifrequired to vote, disengaged groups such as theyoung and ethnic minorities would be drawn into thepolitical process, reducing current divisions.

As with military service, people who object on prin-ciple to participating could be exempted. Also, therequirement could merely be to cast a ballot, not nec-essarily to fill it out (the Australian approach).Alternatively, the ballot could include an ‘against all’option, as in Russia. Further, the mere existence ofcompulsory voting would provide sufficient incentiveto participate without any great need to punish non-voters. If compulsory voting can work in Australia,where the fine is a mere $20 if no satisfactory expla-nation is provided for abstention, it can surely workelsewhere.

The case against Mandatory voting is a denial of the liberty whichforms an essential component of liberal democracy.Requiring people to participate is a sign of an authori-tarian regime, not a democracy. Paying taxes andfighting in battle are duties where every little helps,and where numbers matter. However, elections in all

Table 9.5 Turnout in some countries with compulsory voting

Country Turnout at most recent Year of legislative election (%) election

Australia 94.9 2001Belgium 96.3 2003Turkey 76.9 2002Brazil 78.5 1998Greece 75.0 2000Mexico 57.2 2000

Note: Turnout in the 2002 presidential election in Brazil was 79.5%.

Source: International Institute for Democracy and Electoral Assistanceat http://www.idea.int.

democracies still attract more than enough votes tofill the posts. It is unlikely that an increase in turnoutwould raise the quality of choice; indeed, requiringthose with least political interest and knowledge totake part could lead to worse decisions. Furthermore,the law would need to be enforced; otherwise theentire legal system would suffer.

Compulsory voting deals with symptoms rather thancauses. It distracts from the real task of attractingpeople back to the polls of their own volition. In anycase, why worry? Perhaps non-voting just reflects con-tentment with life. Krauthammer (1990) may havebeen right in claiming that ‘low voter turnout meansthat people see politics as quite marginal to theirlives, as neither salvation nor ruin.That is healthy.’

Assessment Compulsory voting does make a difference: when theNetherlands made voting optional in 1970, turnoutfell considerably (Andeweg and Irwin, 2002, p. 74).However, the opposite reform – introducing manda-tory voting where it did not previously exist – has notbeen tried recently.The danger is that it would back-fire, increasing political distrust between governmentand governed.There is surely a case for testing otherreforms designed to increase turnout before resortingto compulsion.

Further reading: Franklin (2002), Hill (2002), Katz (1997, pp.243–5), van Deth (2000).

DEBATE

COMPULSORY VOTING

or 500,000 electors can initiate a popular vote butonly on whether to repeal an existing law (Uleri,2002). So, unlike normal initiatives, the abroga-tive referendum does not permit the people toraise new issues.

Second, some 15 American states make provi-sion for recall elections. These are ballots onwhether an elected official should be removedfrom office. A ballot is held on a petition fromaround 25 per cent of the votes cast for the rele-vant office at the previous election. Originallydesigned to dismiss corrupt politicians, the recallhas rarely been used.

However, a ballot to recall Democratic GovernorGray Davis did take place in California in 2003,following a petition by more than one million reg-istered voters. This recall election allowednumerous Republican candidates (includingArnold Schwarzenegger, the eventual victor) tostand as a potential replacement for Davis withoutgoing through the extensive vetting needed to wina party nomination for an ordinary gubernatorialelection. This short cut enabled Schwarzenegger toexploit his celebrity status, propelling him into thegovernorship.

Referendums are growing in popularity. Mostreferendums held in the twentieth centuryoccurred after 1960 and most democracies held atleast one referendum in the final quarter of thecentury (LeDuc, 2002). Switzerland headed the

list, holding 72 referendums between 1975 and2000. In 2003, it decided nine referendums atonce, on issues ranging from abandoning nuclearpower to banning car use on four Sundays eachyear. However, few countries have made morethan occasional use of the device and the USA hasheld none at all at national level. Despite the tra-dition of direct democracy in some Americanstates, the constitution makes no provision fornational referendums.

What is the contribution of referendums todemocracy and governance? How desirable is it totransform citizens into legislators? On the plusside, referendums do seem to increase people’sunderstanding of the issue, their confidence intheir own political abilities and their faith in gov-ernment responsiveness (Bowler and Donovan,2002). Like elections themselves, referendumsprovide an education for those who take part inthem.

But there are also reasons for caution. By itsnature, the referendum treats issues as isolatedtopics, ignoring the implications for other areas.What would happen, for instance, if the votersdecide both to raise teachers’ salaries and to lowertaxes? Further, voters are often reluctant toembrace change, turning referendums into aninstrument of conservatism as much as democracy(Kobach, 1997).

Despite their democratic credentials, theoutcome of optional referendums can be influ-enced by government control of timing. In 1997the British government only held a referendum inWales on its devolution proposals after a similarvote in Scotland, where support for devolutionwas known to be firmer. In 2003, EasternEuropean countries began a sequence of referen-dums on joining the EU in Hungary, hoping theresult there would influence the outcome in othercandidate states where public opinion was moresceptical (in the event, all the accession countriesholding referendums voted in favour).

More crudely, rulers can simply ignore the resultof a referendum. In 1955, Swedes voted decisivelyto continue driving on the left; the country nowdrives on the right. A quarter of a century later,Swedes voted to decommission its nuclear powerstations; it took almost twenty years before thefirst reactor closed. Alternatively, a referendum can

162 LINKING SOCIETY AND GOVERNMENT

Referendum – a vote of the electorate on an issue ofpublic policy such as a constitutional amendment.The vote may be binding or consultative.

Initiative – a procedure which allows a certainnumber of citizens (typically around 10 per cent inAmerican states) to initiate a referendum on a giventopic.

Recall – allows a certain number of voters todemand a referendum on whether an elected officialshould be removed from office.

Sources: Bowler et al. (1998), Cronin (1989).

B OX 9 . 5

The referendum, initiative andrecall

be repeated until the desired outcome is obtained.Ireland, for instance, only ratified the Nice Treatyon the European Union in 2002, at the secondtime of asking.

In addition to these difficulties, referendums canbe easily hijacked:

� by wealthy companies waging expensive refer-endum campaigns on issues in which they havean economic interest

� by government control over wording as well astiming

� by intense minorities seeking reforms to whichthe majority is indifferent.

So referendums and other instruments of directdecision-making live uneasily in the house of rep-resentative democracy. Their main benefit is toprovide a double safety valve. First, a referendumallows governments to put an issue to the peoplewhen for some reason it is incapable of reaching adecision itself. Like a plumber’s drain-rods, refer-endums resolve blockages. Second, where the ini-tiative and the recall are permitted, aggrievedcitizens can use these devices to raise issues andcriticisms that might otherwise go unheard.

Elections in new democracies

Nothing seems to mark out a new democracy asclearly as the introduction of free, fair and com-petitive elections. And the first election followingthe withdrawal of the dictators is typically a highturnout affair marking the launch of a newregime. The significance of such founding elec-tions lies less in the result than in their capacity tolegitimize the new order. Founding elections areboth a referendum on, and a celebration of,democracy.

Examples of founding elections include SouthAfrica in 1994 and the first post-communist elec-tions in most of Eastern Europe in 1990.Throughout Africa, founding elections between1990 and 1994 were marked by exceptionally highturnouts, convincing victories for the winners and,most important, the peaceful ejection of sittingpresidents in 11 countries (Bratton, 1998). Truedemocracy seemed to have arrived at last.

DefinitionA founding election is the first election fol-lowing the transition from authoritarian todemocratic rule. Such watershed contests are apublic affirmation of the new regime; they arenormally high-stimulus, high-turnout events. Bycontrast, second elections are normally markedby lower turnout, some disillusionment and, insome cases, the return of electoral malpractice.

However, second and subsequent elections are amore convincing test of successful democratic con-solidation. The broad coalitions which broughtdown the old rulers soon fall apart as the heroes ofthe struggle – such as Nelson Mandela in SouthAfrica – gradually depart. Popular euphoria givesway to a more realistic assessment of the hard roadahead. In these more chastened circumstances, thequestion is whether elections in new democraciescontinue to provide a fair and accepted method ofreplacing unpopular rulers.

Certainly, elections in some new democracieshave acquired the routine character that reflectsconsolidation of the democratic order. When theelection itself ceases to be the issue, and the focusshifts instead to the competing parties, electionshave become an institutionalized part of an estab-lished democracy. In these circumstances, adecline in turnout may even indicate a maturingdemocracy. Here, for example, is Levitsky’s assess-ment (2000, p. 56) of the 1999 election inArgentina:

Perhaps the most striking change was theroutine, even boring, character of the election.In short, a central characteristic of the 1999election was the unprecedented degree to whichelectoral politics had become routinized.

Although Russia remains no more than a semi-democracy, Sakwa (2000, p. 85) writes in similarvein about its elections in 1999:

The elections were no longer so much about achange of regime as about a change of leaderswithin the system. This is no mean achievementand suggests that the Russian political system isbeginning to stabilize and mature.

ELECTIONS AND VOTERS 163

Yet in many other new democracies the qualityof elections seemed to decline in second and sub-sequent contests. Particularly in sub-SaharanAfrica, opposition boycotts, manipulation byexisting rulers and simple administrative incompe-tence came to the fore. International monitoringagencies have become increasingly critical of post-founding elections in many African countries.Rather than becoming the only game in town, thenature and significance of elections continue to becontested, suggesting that some new ‘democracies’are moving towards semi-democracy rather thanan established democracy.

Further, existing rulers seem to have recoveredtheir capacity to secure their own reelection. Theprevalence of this outcome suggests – even if itdoes not prove – electoral manipulation. AsBratton (1998, p. 65) notes, ‘in a “big man” polit-ical culture, it is unclear whether the re-election ofan incumbent constitutes the extension of aleader’s legitimacy or the resignation of the elec-torate to his inevitable dominance’. In an estab-lished democracy, leaders have learnt how to losebut, in many new democracies, cunning rulershave quickly mastered the art of winning.

It seems, then, that the consolidation of democ-racy must be judged against more demandingstandards than merely holding regular elections.Such contests are a necessary but not a sufficientcondition of democracy. As Rose and Shin (2001,p. 331) point out,

A spectre is haunting contemporary studies ofdemocratization: conventional influences, suchas the introduction of free elections, have not (orat least not yet) created political regimes thatmatch the standards of established democracies.

First-wave democracies, such as Britain and theUSA, established the rule of law and the principleof executive accountability to the legislature beforethey extended the vote to the general population.By contrast, many recent democracies introducedelections even though the idea of governmentunder law had still to be accepted. In these cir-cumstances, elections are in danger of beingdevalued. The risk is that they become an agent of,rather than a choice among, those who wieldpower.

Elections in authoritarian states

Although we have defined an election as a compe-tition for office, ‘elections’ in authoritarian regimesdo not provide genuine choice. However, only themost extreme dictator dispenses with electionsaltogether; the appearance of choice must be pre-served if only as a useful fiction (Liddle, 1996). Inauthoritarian systems, elections are often onlysemi-competitive, with the winner known inadvance and electoral malpractice playing its partin delivering the desired result. In totalitariansystems candidates are simply presented to thevoters for approval, without even the illusion ofchoice.

Elections in authoritarian regimes are moreoften ‘made’ than ‘stolen’ (Mackenzie, 1958). In amade or semi-competitive election, the dice ofresources, visibility and access to the media are soheavily loaded towards the current rulers that thedesired result is manufactured without resort toelectoral theft.

Semi-competitive elections mix choice andcontrol in the characteristic fashion of an authori-tarian or semi-democratic state. The ruling partyuses all the advantages of office, including effectivegovernance and a high-visibility leader, to ensureits reelection. Patronage is the party’s key resource;it is used either to reward loyal voters directly, aswith cash for votes, or to provide local notableswith jobs, contracts, access, influence, status andmoney in exchange for the votes of their clients. ‘Amobile phone for every shepherd’, promised thesuccessful candidate for the presidency of the ruralRussian republic of Kalmykia in 1993. ‘Aninternet connection for every school in Peru’, pro-claimed President Fujimori during his reelectioncampaign in 2000. In poor countries, suchpromises from incumbents are worth far more toelectors than similar statements from oppositioncandidates whose destiny is defeat.

Rulers also exploit their control over both themedia and the administration of the election.Opposition candidates find they are disqualifiedfrom standing; that electoral registration is ineffi-cient in their areas of strength; that they are rarelypermitted to appear on television; that they areharassed by the police; and that their leaflets andposters are mysteriously lost. The opposition loses

164 LINKING SOCIETY AND GOVERNMENT

heart because it knows its function is always tooppose but never to win.

By contrast, the incumbent president can exploitunique resources. These include unparalleled visi-bility built over time, easy access to television, theability to use the state’s coffers for the campaignand finally the capacity to call in political creditscarefully acquired while in office. Anticipating thepresident’s reelection, all the underlings will seekto help the campaign along, thus amplifying thefinal victory. Why back losers?

In its heyday, Mexico’s Partido RevolucionariaInstitucional (PRI) was one of the world’s mostsuccessful vote-winning machines, providing theclassic example of a party-based approach to semi-competitive elections. By winning 11 presidentialelections in a row before its historic defeat in2000, the PRI became a party of the state, givingit unique access to resources which it could passout through its intricate patronage network.However, in Mexico as in the larger LatinAmerican countries, semi-competitive electionsslowly became harder to manage as the electoratebecome more urban, affluent and educated. Withboth the domestic media and internationalobservers becoming more critical of blatant cor-ruption, and with privatization reducing theresources under the government’s control, the PRIbegan to fight cleaner elections, creating the con-ditions for its own defeat in the 2000 presidentialcontest.

While semi-competitive elections preserve anillusion of choice, elections in totalitarian regimeswere more brutal. In communist states, forinstance, there was no pretence that the rulingparty could be defeated or even opposed throughelections. In the Soviet Union, for instance, theofficial candidate was simply presented to the elec-torate for ritual endorsement. Soviet elections weregrim, ritualistic affairs, irrelevant to the real poli-tics taking place within the party. They were littlemore than an opportunity for the party’s agitatorsto lecture the population on the party’s achieve-ments (Zaslavsky and Brym, 1978).

Contemporary examples of choice-free electionsare confined to decaying communist dictatorships.In Cuba, for example, 609 candidates were put upfor election to the National Assembly in 2003;

there were exactly 609 seats to be filled. The onlychoice given to voters was whether to support all,some or none of the candidates.

Some communist states did introduce a measureof choice to their elections by allowing a choice ofcandidates from within the ruling party. Thesecontrolled candidate-choice contests were charac-teristic of communist Eastern Europe in the 1970sand 1980s. Central rulers found candidate-choiceelections useful in testing whether local party offi-cials retain the confidence of their communities.

This is one reason for the gradual introductionof such elections to some of China’s 930 000 vil-lages since 1987. In addition, elected village com-mittees help to build state capacity in a countrywhere power has traditionally operated on a per-sonal basis. However, even in contemporary Chinano explicit opposition to the party’s policy plat-form is permitted. As a result, there are few signsof elections in China threatening the party’scontrol.

Key reading

Next step: LeDuc, Niemi and Norris (2002) isan excellent comparative study of elections andvoting, reviewing a wide literature.

Katz (1997) links elections with broader democ-ratic themes while Ginsberg (1982) remains astimulating top-down view of competitive elec-tions. On electoral systems, Farrell (2001) is aclear introduction, Shugart and Wattenberg(2000) look at additional member systems specifi-cally and Norris (2003) examines the impact ofelectoral systems on party systems. Dalton andWattenberg (2000) is a comparative study ofvoting trends; see also Broughton (2002) forturnout and Evans (2003) for a UK-based text onvoting behaviour. On the United States see Asher(1988b) for presidential elections generally, andPomper et al. (2001) for the 2000 election. Forthe impact of campaigns, see Farrell and Schmitt-Beck (2002), and for the influence of leaders’ per-sonalities specifically, King (2002). Onreferendums, consider Gallagher and Uleri (1996)and Mendelsohn and Parkin (2001).

ELECTIONS AND VOTERS 165

Classifying interest groups 167Channels of access 169Issue networks 172Conditions of influence 175Pluralism and corporatism 177Interest groups in new democracies 181Interest groups in authoritarian states 182Key reading 184

Interest groups (also called pressure groups) are‘organizations which have some autonomy from

government or political parties and which try toinfluence public policy’ (Wilson, 1990b). Theypresuppose formal organization and thus can bedistinguished from social movements. Examplesinclude employers’ organizations, trade unions,consumer groups, bodies representing specificindustries or professions and campaigning organi-zations seeking to promote particular causes.

Like political parties, interest groups inhabit thespace between society and state, helping to linkthe two. But where political parties aspire tobecome the government, interest groups seek justto influence it. Reflecting this narrower focus,interest groups do not fight elections; instead, theytypically adopt a pragmatic and often low-keyapproach in dealing with whatever power structureconfronts them.

Although many interest groups go about theirwork quietly, their activity is nonetheless pervasivein established democracies. Their staff are to befound negotiating with bureaucrats over thedetails of proposed regulations, pressing their casein legislative committee hearings and seeking toinfluence media coverage of their position. Inauthoritarian regimes, however, interests are artic-ulated in a less public, more spasmodic and some-times more corrupt fashion. Interests are stillexpressed to government but often through indi-vidual firms or powerful individuals rather thanthrough organized interest groups.

There is a puzzle about how interest groups

succeed in developing in the first place. Olson(1968) argued that people have no reason to joininterest groups when the fruits of the group’sefforts are equally available to non-members. Whyshould a worker join a union when the wage rise itnegotiates goes to all employees? Why should acompany pay a fee to join an industry associationwhen the tax subsidies the group negotiates willhelp all the firms in that sector? Surely it would bemore rational to take a free ride on the efforts ofothers.

However, the fact is that new groups do emergeand grow; the green lobby, for instance, has grownenormously in recent decades. Perhaps this isbecause members receive the selective benefits ofdeveloping their skills and meeting new people aswell as contributing to a shared goal which is dearto their hearts (Moe, 1980).

DefinitionA free rider leaves others to supply collectivegoods – that is, benefits such as unpolluted airthat must be supplied to everyone if they aresupplied at all.The possibility of free ridingcreates incentives not to join interest groupssuch as labour unions that negotiate deals ofequal benefit to members and non-membersalike: for example, a safer working environment.

From an historical perspective, interest groupsdeveloped in a rather predictable way. In the West,they emerged in a series of waves formed by socialchange (e.g. industrialization) and by the expan-sion of state activity (e.g. public welfare). Periodsof social change raise new problems while anactive government gives people more hope of gainsfrom influencing public policy. Box 10.1 summa-rizes the development of interest groups in oneparticular example: the United States. MostWestern nations have followed a similar course,resulting in the mosaic of independent groupactivity making up contemporary society.

Chapter 10

Interest groups

166

The range and influence of modern interestgroups raises awkward questions about the distrib-ution of power in democracies. Favoured groupsacquire insider status and thus the potential toinfluence decisions at an early stage. Often interestgroups work in tandem with national govern-ments in developing positions for bodies such asthe European Union and the World TradeOrganization. Despite only representing smallminorities, interest groups are deeply entrenchedin policy-making, certainly more so than manysupposedly ‘sovereign’ parliaments. Interest groupactivity creates a system of functional representa-tion operating alongside electoral representation.

Here, then, is an important question to raiseabout democracies: do interest groups possesspower without accountability? Are interest groups,as Lowi (1969) contended, ‘a corruption of demo-cratic government’? Or does group access topolicy-making simply express the right to organizein defence of specific objectives? In short, dointerest groups help or hinder democracy?

Classifying interest groups

When we think of ‘interest groups’, the bodieswhich first come to mind are protective groupsarticulating the material interests of theirmembers: for instance, trade unions, employers’organizations, industry bodies and professional

associations of lawyers or physicians. Sometimescalled sectional or functional groups, these protec-tive bodies are founded to influence government.They have sanctions to help them achieve theirgoals. Workers can go on strike; medical practi-tioners can refuse to cooperate with a new pre-scription policy. Protective groups seek selectivebenefits for their members and insider status withrelevant government departments (Box 10.2).Because they represent clear occupational interests,protective associations are often the most influen-tial of all interest groups. They are well-estab-lished, well-connected and well-resourced.

But protective groups can also be based on local,rather than functional, interests. Thus, geographicgroups arise when the interests of people living inthe same location are threatened, for instance by anew highway or a hostel for ex-convicts. Becauseof their negative stance – ‘build it anywhere buthere’ – these geographical bodies are known asNIMBY groups (not in my back yard).

Of course, protective groups are not the onlytype of organized interest. Indeed, many groupsfounded since the 1960s are promotional ratherthan protective. Promotional groups advocateideas, identities, policies and values. Also calledattitude, cause and campaign groups, promotionalgroups include pro- and anti-abortion groups,organizations combating pornography and ecologygroups.

Promotional groups are most significant in

INTEREST GROUPS 167

Period Description Examples

1830–60 Founding of first national organizations YMCA and many anti-slavery groups

1880–1900 Creation of many business and labour National Association of Manufacturers, American associations, stimulated by Federation of Laborindustrialization

1900–20 Peak period of interest group formation Chamber of Commerce, American Medical Association

1960–80 Founding of many environmental and National Organization for Women, Common Cause public interest groups

Source: Adapted from Hrebenar and Scott (1997, pp. 13–15).

B OX 1 0 . 1

Waves of interest group formation in the United States

established democracies, where they are growingin number, significance and recognition by gov-ernment. Indeed, the increasing influence of pro-motional groups since the 1960s, especially in theUnited States, constitutes a major trend in interestgroup politics. The relative significance of protec-tive and promotional groups reveals much aboutthe nature of politics and policy-making in anyparticular country. In the United States, signifi-cantly, promoting the public interest is largely theresponsibility of a legion of private, promotionalgroups. For example, Common Cause describesitself as ‘a nonprofit, non-partisan citizen’s lob-bying organization promoting open, honest andaccountable government’ (Gardner, 2000).

The boundary separating protective and promo-tional groups is not well-defined. For example,because bodies such as the women’s movementand the gay lobby seek to influence publicopinion, they are often classified as promotional.However, their prime purpose is to promote theinterests of specific groups: they are, perhaps, bestviewed as protective interests employing promo-tional means.

Interest groups do not always lobby governmentdirectly. Often, they join together in federations orcoalitions with other groups to increase their effec-tiveness. Such bodies are known as peak associa-tions. Examples include the National Associationof Manufacturers in the USA and theConfederation of British Industry in the UK(Figure 10.1). The members of these peaks are not

individuals but firms or interest groups repre-senting specific industries. Trade unions respondsimilarly: Britain’s Trades Union Congress (TUC)consists of 69 affiliated unions with 6.7m indi-vidual members who do not themselves belongdirectly to the TUC. America’s AFL–CIO is simi-larly a coalition of individual unions.

DefinitionA peak association is an organization repre-senting the broad interests of capital or labour togovernment. Examples from the Netherlandsinclude the VNO–NCW (Association of DutchCompanies–Dutch Association of ChristianEmployers) and the FNV (Federation of DutchTrade Unions).The members of peak associationsare not individuals but other organizations suchas firms, trade associations or labour unions. Peakassociations are important components of cor-poratism.

Peak organizations are particularly important inpolicy-making. They are generally more attuned tonational government than are their members.They have a strong research capacity and talk thelanguage of policy as well as the language ofindustry. In some states, peaks play an integral rolelinking government with their own members and,through them, with the wider society. For thepeaks to acquire such a position, however, theymust achieve some autonomy from, as well ascontrol over, their members. As implied by the fre-

168 LINKING SOCIETY AND GOVERNMENT

Protective groups Promotional groups

Aims A group of – defends an interest A group for – promotes a cause

Membership Closed – membership is restricted Open – anybody can join

Status Insider – frequently consulted by Outsider – not consulted as often by government.government and actively seeks this role Emphasizes public opinion and the media

Benefits Selective – only group members benefit Collective – benefits go to both members and non-members

Focus Mainly national – group aims to influence More international – group may seek to influence national government bodies such as the EU and global public opinion

B OX 1 0 . 2

Protective and promotional groups

quency with which peak associations label them-selves ‘federations’ or ‘confederations’, in practicesuch centralization is often lacking.

The creation of distinct peak associations islargely a phenomenon of protective groups.Among promotional groups, collaboration has sofar taken the form of coalitions of existing groupsor ad hoc cooperation on specific campaigns. Forexample, a series of promotional groups mightjoin together to campaign against capital punish-ment or for writing off third world debt.

Channels of access

How are interests communicated to politicaldecision-makers? What are the channels through

which this process takes place? Figure 10.2 sets outthree mechanisms characteristic of establisheddemocracies: direct dealings with government,indirect influence through political parties andindirect influence through public opinion. In thissection, we will also look at specialist lobbyingcompanies which help to pilot their interest groupclients through these varied channels.

Direct discussion with policy-makers

The core business of most interest groups, espe-cially protective ones, is influencing public policy.Most of this activity focuses on the bureaucracy,the legislature and the courts. In establisheddemocracies, the bureaucracy is the main pressurepoint. Interest groups follow power and it is incivil servants’ offices that detailed decisions areformed. As Matthews (1989, p. 217) comments,

the bureaucracy’s significance is reinforced by itspolicy-making and policy-implementing roles.Many routine, technical and ‘less important’decisions, which are nonetheless of vital concernto interest groups, are actually made by publicservants.

Shrewd protective groups focus on the smallprint because it is difficult for them to control thebroad contours of policy, as set by elected politi-cians. But on matters of detail, most democraciesfollow a convention of discussion with organizedopinion through consultative councils or commit-tees; often the law requires such consultation.After all, the real expertise often lies in the interestgroup rather than the bureaucracy and, from theminister’s viewpoint, an agreement acceptable toall is politically safe.

Consultation is, for example, a key feature of theextraordinarily deliberative character of policy-making in Scandinavia (Blom-Hansen, 2000).Denmark has been termed ‘the consulting state’because of the extensive negotiations between thegovernment and the country’s 2,000 or so nationalgroups.

Even in France, where the higher bureaucracytraditionally prizes the autonomy of the state,extensive dialogue takes place between civil ser-vants and formally organized interests. This inte-

INTEREST GROUPS 169

Policy affectinginterestgroup

Figure 10.2 Channels of interest group influence

Public opinion

Interestgroup

Parties

Policy-makinginstitutionsassembly,executive,

bureaucracy,courts

180 tradeassociationmemberse.g. Food and Drink Federation

Figure 10.1 A peak association: Britain’s CBI

Confederation ofBritish Industry (CBI)

Covering companymembers with 4m

employees

Covering industries with 4m

employees

2,000companymemberse.g. Logica

gration is assisted by the role played by the state inorganizing the groups themselves. Any associationwanting legal rights must register with the state;by 2001, 750,000 had done so. Further, member-ship of chambers of commerce, agriculture andtrade is compulsory for relevant organizations suchas firms and farmers. These chambers haveacquired semi-public status, helping to implementpublic policy in consultation with the bureaucracy(Stevens, 2003).

Assemblies are an additional channel throughwhich interests and demands can be voiced. Butwhile the bureaucracy is invariably a crucial arenafor groups, the significance of the legislaturedepends on its political weight. A comparisonbetween the United States and Canada makes thepoint.

The American Congress (and especially its com-mittees) forms a vital part of the policy process.The separation of powers, the constitutional right‘to petition the government for a redress of griev-ances’, weak party discipline and strong commit-tees combine to create an ideal habitat for lobbyoperations. Many interest groups have access toindividual members of Congress and can arrangeto contribute to committee deliberations. Largefinancial contributions by political action commit-tees (PACs) mean that it can be politically difficultfor legislators to spurn group demands (Cigler andLoomis, 2002).

But Congress is a unique case. In most democra-cies, parliaments are more reactive than proactive;as a result, interest groups treat members of parlia-ment as opinion-leaders rather than decision-makers. For example, party voting is entrenched inthe Canadian parliament and lobbyists concen-trate their strongest fire on the bureaucracy. AsLandes (1995, p. 488) comments on Canada,

interest groups have an acute sense of smellwhen tracking the scent of power. Interactionwith the bureaucracy and not with MPs is thegoal of most groups and one reason why interestgroup activity is not highly visible to theuntrained eye.

If interest groups feel ignored in the policy-making process, they may still be able to seekredress through the courts. In the European

Union, an interest group that is unsuccessful athome can take its case to the European Court ofJustice. In the United States, business corporationsroutinely subject government statutes and regula-tions to legal challenge. Class actions are particu-larly common in America.

DefinitionA class action is a legal device initiated by com-plainants on their behalf and ‘for all others so sit-uated’.This mechanism enables legal costs andgains to be shared among a large group and pro-vides a lever by which interest groups canpursue their goals through the courts.The deviceis commonly used in the USA.

But just as the USA is exceptional in the powersof its legislature, so too does it possess a stronglegal culture. Elsewhere, the courts are growing inimportance but still as a remedy of last resort. InAustralia, for instance, the requirement for liti-gants to prove their personal interest in the casehinders class actions. When the outcome of a legalcase affects only the person initiating it, the policyand financial implications are far less than in aclass action covering all those in the same position.

Indirect influence through politicalparties

Interest groups and parties can overlap since bothare devices through which social forces seek toinfluence government. Britain’s labour movementhistorically regarded its industrial wing (the tradeunions) and political wing (the Labour Party) aspart of a single movement promoting working-class interests. In a similar way, the environmentalmovement has spawned both promotional interestgroups and green political parties.

But such intimate relationships between groupsand parties are the exception. Most interest groupsseek to hedge their bets rather than to developclose links with a political party. Loose, pragmaticlinks between parties and interests are the norm.In the United States, business and organizedlabour gravitate towards the Republican andDemocratic parties respectively but these are part-nerships of convenience, not indissoluble mar-riages. The traditional maxim of the American

170 LINKING SOCIETY AND GOVERNMENT

trade union movement has been to reward itsfriends and punish its enemies, wherever these areto be found.

American business is equally pragmatic. Despitean ideological affinity with the Republican Party,corporations still contribute heavily to the electioncoffers of many Democratic members of Congress.Whether representing capital, labour or neither,most interest groups give more to incumbents.They do not waste money on no-hopers, even ifthese doomed candidates are standing for the‘correct’ party.

The theme is similar, if less explicit, in othercountries. In Germany, for instance, the powerfulFederation of German Industry (BDI) certainlyenjoys close links with the conservative ChristianDemocratic Union. However, it wisely remains onspeaking terms with the more left-wing SocialDemocrats. The rule is that protective interestsfollow power, not parties.

Indirect influence through the massmedia

Press, radio and television provide an additionalresource for interest groups. By definition, mes-sages through the media address a popular audi-ence rather than specific decision-makers. Thusthe media are a central focus for promotionalgroups seeking to steer public opinion. Theirtarget is society as much as government.

In addition, promotional groups usually lackboth the resources and the access available to pro-tective groups, so free publicity becomes theirstock-in-trade. For instance, ecological groupsmount high-profile activities, such as seizing an oilrig to prevent it from being sunk, to generatefootage shown on television across the world. Incontrast to protective groups, promotional groupsview the media as sympathetic to their cause(Dalton, 1994).

Traditionally, the media are less important toprotective groups with their more specialized andsecretive demands. What food manufacturerwould go public with a campaign opposing nutri-tional labels on foods? The confidentiality of thegovernment meeting room is a quieter arena forfighting rearguard actions of this kind; goingpublic is a last resort.

But even protective groups are now seeking toinfluence the climate of public opinion, especiallyin political systems where legislatures are impor-tant. In the United States, most protective groupshave learned that to impress Congress they mustfirst influence the public. Therefore groups follow adual strategy, going public and going Washington.Interest groups in other democracies are beginningto follow suit. Slowly and uncertainly, protectivegroups are emerging from the bureaucratic under-growth into the glare of media publicity.

Lobbyists

The lobby is a term derived from the hall or lobbyof Britain’s House of Commons. Here peoplecould, and still do, approach members of parlia-ment to plead their cause. Although interestgroups are often their own best lobbyists, ourfocus here is on specialist companies whose job itis to open the doors of government to their clients.These lobbying firms are technicians of influence:hired guns in the business of political communica-tion. And they are growing in number though notnecessarily in influence.

The business of lobbying remains highly per-sonal. A legislator is most likely to return a callfrom a lobbyist if the caller is a former colleague.For this reason, lobbying firms are always on thelookout for former legislators or bureaucrats witha warm contact book. More than most profes-sions, lobbying is about who you know.

Why is lobbying an expanding profession? Threereasons suggest themselves. First, government reg-ulation continues to grow, often impactingdirectly on companies, interest groups and tradeunions. For instance, a decision to permit the saleof a new medicine can be a matter of life anddeath for the drug company as well as for poten-tial patients. One prosaic task of lobbying firms isto keep a close eye on proposed regulations ema-nating from legislative committees.

Second, public relations campaigns arebecoming increasingly sophisticated, often seekingto influence both the grass roots and the govern-ment in one integrated project. Professionaladvisers come into their own in planning anddelivering multifaceted campaigns, which can betoo complex for a client to manage directly.

INTEREST GROUPS 171

Third, many firms now approach governmentdirectly, rather than working through their tradeassociation. Companies find that using a lobbyingcompany to help them contact a governmentagency or a sympathetic legislator yields quickerresults than working through an industry bodywhich has to consider the concerns of all itsmembers. McKay (1997, p. 250) reports that inthe United States

since the 1960s there is overwhelming evidencethat individual firms have taken a more activepart in public policy-making. Most major cor-porations now have Washington offices andemploy professional lobbyists to advance andprotect their interests.

What then is the political impact of lobbyingcompanies? Is it now possible for wealthy interestgroups and corporations simply to pay a fee to alobbying firm to ensure a bill is defeated or a regu-lation deferred? On the whole, the answer is no.Lobbyists are inclined to exaggerate their ownimpact for commercial reasons but in reality mostcan achieve little more than access to relevantpoliticians and, perhaps, bureaucrats. Lobbyists,like interest groups, tend to cancel each other out.Often, the lobbyist’s role is merely to hold theclient’s hand, helping an inexperienced companyto find its way around the corridors of powerwhen the corporation comes to town.

In these ways, professional lobbying firms cancontribute to the effectiveness of political commu-nication, focusing the client’s message on relevantdecision-makers. They can help to ensure that theclient’s voice is heard but shaping the policy-maker’s response to the message is a far greaterchallenge. Allegations of sleaze notwithstanding,influence can rarely be purchased through a lob-byist but must come, if at all, from the petitioninggroup itself. And impact depends first and fore-most on the intrinsic strength of the case. To theexperienced politician, a convincing case directfrom the petitioner sings louder than yet anotherrehearsed presentation from a lobbying firm.

Of course, not everything in the lobby is rosy.Even if a company achieves no more for its client’sfee than access to a decision-maker, perhaps thatexchange in itself compromises the principle of

equality which supposedly underpins democracy.Groups unable to employ lobbyists are not neces-sarily denied political impact but they do have tooperate on a do-it-yourself basis. Further, thewidespread perception that influence can be pur-chased in itself damages the legitimacy of thedemocratic process.

Iron triangles and issue networks

In the democratic world, many interest groups arein virtually daily contact with government. So thequestion arises, what is the nature of such relation-ships? How should they be characterized? This is acore issue in the study of interest groups, bearingon the question of whether groups express orsubvert democracy.

Until recent decades, the relationship betweeninterest groups and the state was viewed critically.Within a particular sector, it was alleged, interestgroup leaders and senior civil servants formedtheir own small communities. All the membersknew each other well, used given names and triednot to upset each other. The participants devel-oped shared working habits and common assump-tions about what could be achieved. The actorslearned to trust each other and to respect eachother’s goals and confidences.

Shared interests predominated. For instance, theroad-builders and bureaucrats in the transportministry would seek ever larger highway budgets,fully aware that similar coalitions in other sectors– defence, say, or education – would be seeking tomaximize their own funding and autonomy.Business was done behind closed doors to preventpolitical posturing and to allow a quiet life for all.Insiders were sharply distinguished from outsiders.The golden rule was never to upset the apple cart.

American political scientists used the term ‘irontriangle’ to describe the particular form taken bythe relationship between groups and governmentin the USA. The three points on the triangle wereexecutive agencies or departments, interest groupsand congressional committees (Figure 10.3). Suchtriangles became an exercise in mutual back-scratching: the committee appropriated fundswhich were spent by the agency for the benefit ofinterest group members. Thus, the Department of

172 LINKING SOCIETY AND GOVERNMENT

INTEREST GROUPS 173

Founding members (1952 ECSC;1958 EEC and Euratom)

First enlargement (1973)

Second enlargement (1981)

Third enlargement (1986)

Incorporation of the territory of theGerman Democratic Republic into a unitedGermany (1990)

Fourth enlargement (1995)

Fifth enlargement (2004)

DenmarkIreland United

Kingdom

France

Germany

Sweden

Finland

Be.

L.

Ne.

Italy

Greece

SpainPort.

KEY:

Be. BelgiumL. LuxembourgNe. NetherlandsPort. Portugal

Austria

Malta(2004 entrant)

Cyprus

Estonia

Latvia

Lithuania

Poland

CzechRepublic

Slovakia

Hungary

Slovenia

Map 10.1 The European Union (see also 50 years of the European Union on p. 29)

Source: Adapted from Nugent (1999, 2004).

Agriculture, the relevant committees in Congressand farmers’ groups would collude on larger foodsubsidies. Each point on the triangle benefited.This was a game without losers – except for thetaxpayer who rarely knew what was going on.

Iron triangles were also called subgovernments,implying that each triangle formed a mini-govern-ment of its own, largely independent of policy-making in other sectors. The effect was tofragment policy-making, deflecting the politicalaims of the majority party in Congress or even ofthe president himself (Ripley and Franklin, 1991).

Fortunately, perhaps, these cosy iron trianglesand subgovernments have decayed in manydemocracies. Today, policies are subject to closerscrutiny by the media; new public interest groupsprotest loudly when they spot the public beingtaken for a ride; and some legislators are lesswilling to keep quiet when they see public moneybeing wasted. And as issues become morecomplex, so more groups are drawn into the policyprocess, making it harder to stitch together secretdeals among a few insiders. In the United States,where this trend has gone furthest, the committeebarons who used to dominate Congress have lostmuch of their power. The iron has gone out of thetriangle; now influence over decisions depends onwhat you know as much as who you know.

Reflecting these trends, the talk now is of ‘issuenetworks’. These refer to the familiar set of organi-zations involved in policy-making: governmentdepartments, interest groups and legislative com-mittees plus expert outsiders. However, an issue

network does not imply a close-knit subgovern-ment. In an issue network, the impact of an interestgroup varies from issue to issue, depending on itsexpertise. As Heclo (1978, p. 102) famously put it,

the notions of iron triangles and subgovern-ments presume small circles of participants whohave succeeded in becoming largely autonomous.Issue networks, on the other hand, comprise alarge number of participants with quite variabledegrees of mutual commitment . . . it is almostimpossible to say where a network leaves off andits environment begins.

The emergence of issue networks reflects a shiftto a more open policy-making style in whichback-scratching has become harder to sustain.Clearly, the idea of issue networks enables us toportray policy-making in democracies more posi-tively. A wider range of interests participate indecisions, the bias toward protective groups isreduced, new groups enter the debate and a soundargument carries greater weight. Networks operatein a non-hierarchical way, with resources such asexpertise and authority in effect exchangedbetween the participants.

DefinitionsAn issue network is an open and flexible combi-nation of players who collectively shape policy ina given sector, such as defence or agriculture(Heclo, 1978).Through their deliberations, themembers of a network in effect exchange theirvarying resources such as knowledge (e.g. acad-emic specialists), legitimacy (e.g. elected politi-cians), control over implementation (e.g. interestgroups) and the capacity to draft bills and regu-lations (e.g. bureaucrats).

Further, epistemic (knowledge-based) communi-ties can often place issues such as acid rain on thepolicy agenda at an early stage, perhaps counteringthe conservatism of protective interests. Withinthese broad communities of expertise, particularcoalitions emerge to advocate specific positions(Sabatier and Jenkins-Smith, 1993). The founda-tion of many issue networks in expert knowledgeat least opens up the possibility of more informedpolicy-making.

174 LINKING SOCIETY AND GOVERNMENT

Cabinet Departmente.g. Department of

Agriculture

Interest groupe.g. Farm Bureau,

Mid-America Dairymen

Congressional Committeee.g. House Agriculture

Committee, SenateAgriculture, Nutrition and

Forestry Committee

Figure 10.3 Iron triangles: how subgovernmentsoperated in the USA

Conditions of influence

There is no doubt that some interest groups exertmore influence over government than others. Inpart, no doubt, this difference reflects the varyingskill with which groups exploit the opportunitiesavailable to them but a deeper explanation mustlie in the underlying strength of interest groupsthemselves. So why are some interest groups morepowerful than others? The answer depends on fourmain features of the group: sanctions, legitimacy,membership and resources.

The ability of a group to invoke sanctions isclearly important. A labour union can go onstrike, a multinational corporation can take itsinvestments elsewhere, a peak association canwithdraw its cooperation with the government informing policy. As a rule, protective groups (suchas industry associations) can bring more sanctionsto bear than promotional groups (such as ecologymovements).

The degree of legitimacy achieved by a particulargroup is also important. The aphorism ‘what isgood for General Motors is good for America’expresses the point. Interests enjoying high pres-tige are most likely to prevail on particular issues.For example, professional groups whose membersstand for social respectability can be as militant onoccasion, and as restrictive in their practices, asblue-collar trade unions once were. But lawyersand doctors escape the public hostility that unionsattract.

DefinitionDensity of membership refers to the proportionof those eligible to join a group who actually doso. An encompassing membership gives moreauthority and, in turn, a stronger bargaining posi-tion with government.The declining density oftrade union membership in the final quarter ofthe twentieth century undoubtedly weakenedlabour’s bargaining power.

A group’s influence also depends on its member-ship. This is a matter of density as well as sheernumbers. The highest penetrations are usuallyachieved when, as with many professional bodiessuch as physicians and lawyers, membership is acondition of practice. By contrast, low density

reduces influence, especially when an occupationis fragmented among several interest groups.American farmers, for instance, divide betweenthree major organizations with lower total cov-erage than Britain’s National Farmers Union. Tobe sure, the larger American food producers arepolitically well-connected but the interests of agri-culture as a whole would be better served if allfarmers belonged to a single national association.

In the European Union, breadth of membershipis especially important for lobbying organizations.Groups which can demonstrate their support fromnational associations in most member states willreceive a more cordial reception from the policy-makers in Brussels.

The organizational resources available to aninterest group are the final factor affecting itsinfluence. Here as elsewhere, money talks but notalways loudly. With an annual budget of $40million, America’s National Rifle Association(NRA) can employ 275 full-time staff. Despitepublic sympathy, the coalition of gun controlgroups cannot match the NRA’s fire power. Yeteven in the USA, poor but skilful campaigners cangenerate free publicity and wide public sympathy.Ralph Nader’s Crusade for Car Safety in the 1960swas an early example. Nader exposed a dangerousdesign fault in some General Motors models, aflaw that the manufacturer had sought to keepaway from the public gaze. In this case, what wasgood for General Motors was not good forAmericans, as Nader proved.

Thus, just as the impact of lobbying firms isoften exaggerated, so too is the significance of themoney available to an interest group. As a rule,finance is rarely decisive. Consider the NRA oncemore. It would be naive to suppose that the causeof gun control in the USA is held back solely bythe NRA. Two other factors are just as important:(1) the difficulty of passing any legislationopposed by a significant minority within Congressand (2) an ambiguous constitutional reference tothe ‘right of the people to keep and bear arms’.The NRA has exploited both features but createdneither.

In short, all interest groups inherit an endow-ment of political resources which substantiallydetermines the influence they can exert. Externalcontext matters more than internal organisation.

INTEREST GROUPS 175

176 LINKING SOCIETY AND GOVERNMENT

The European Union (EU) is the most developedexample of regional integration in the world. Uniquelyamong regional bodies, the EU has developed institu-tions resembling the architecture of national govern-ments. It has a parliament and an influential Court ofJustice. Proposals for a codified constitution,expressing a commitment for members ‘to forge acommon destiny’ were put forward in 2003.The heartof the EU, however, is the European Commission (anEU body) and the Council of Ministers (an intergovern-mental body).The tension between these units –between the EU as a cohesive actor and the EU as anarena for negotiation between member states – iscentral to its functioning.

The EU’s emergence owes much to Europe’s history ofconflict. After the 1939–45 war, many Europeanleaders set out to create a unified continent withinwhich war would no longer be feasible: a UnitedStates of Europe. However, economic factors were alsofundamental. European economies needed to berebuilt after the war and then, to achieve benefits ofscale, integrated into a large, single market.

Later members, notably Britain, have emphasized theeconomic basis of the Union while rejecting thefederal vision. Margaret Thatcher (1988) expressed thisposition well: ‘willing and active cooperation betweenindependent sovereign states is the best way to builda successful European Community’. In developing thisline, Britain has exploited the continental notion ofsubsidiarity to argue that decisions should be taken atthe lower, national level wherever possible.

Reflecting the European tradition of social partner-ship, the EU has encouraged interest groups, espe-cially those with a Europe-wide perspective. Over 900interest groups are listed in the Commission’s direc-tory, mostly protective in character.The main peakassociations are the Union of Industrial andEmployers’ Confederation of Europe (UNICE), with 40employees, and the European Trade Union Federation(ETUC), with 45 staff.There are probably more lobby-ists working in Brussels than there are policy-makersemployed by the European Commission.

However, many European interest groups possess fewresources and they experience difficulty in takingpositions acceptable to members drawn from a rangeof countries. Increasingly, therefore, national interestgroups lobby directly in the EU, often working in part-nership with their home government. Other lobbyingcomes from regional and local governments and,increasingly, from individual companies. Externalbodies, such as Japanese trade associations andAmerican multinationals, are also active.

Most lobbyists focus on the Commission, the nearestequivalent to national executives though withouttheir strong party basis.The Commission is a smallbody which develops policy across most sectors. Itrelies even more heavily than national governmentson interest groups for information, support and legiti-macy.

Population of member states: 454mGross domestic product per head:

Following the 5th enlargement (2004),GDP per head fell further below thatof the USA and Japan.

Form of government: a uniqueregional body in which policy is madepartly by European Union institutionsand partly through negotiationsamong the 25 member states.

Executive: the Council of the EuropeanUnion, composed of heads of govern-ment, provides political drive.The

powerful European Commission,divided into 17 directorates, initiateslegislative proposals.

Assembly: the members of the unicam-eral European Parliament are directlyelected from each country for a five-year term.The number of MEPs fromeach country reflects its population.Though growing in significance, theParliament still lacks full control overthe Commission, the budget and evenlegislation.

Judicial branch: the influential EuropeanCourt of Justice, composed of onejudge from each member country, hassuccessfully developed the constitu-tional underpinnings of federalism.

Members: Founders: Belgium, France,Germany, Italy, Luxembourg,Netherlands. 1973: Denmark, Ireland,United Kingdom. 1981: Greece. 1986:Portugal, Spain. 1995: Austria, Finland,Sweden. 2004: Cyprus, Czech Republic,Estonia, Hungary, Latvia, Lithuania,Malta, Poland, Slovakia, Slovenia.

Profile T H E E U R O P E A N U N I O N

Further reading: Eising (2003), Greenwood (2003),Hosli et al. (2002), Mazey and Richardson (2001).

Pluralism and corporatism

In this section we consider the wider role ofinterest groups within the political system. Thismeans broadening the debate beyond specificgroups in order to characterize the overall patternformed by the links between interest groups andthe government.

One answer to this question, of course, is thatthe government stands above the groups. After all,the government, uniquely, is authorized by elec-tion and it alone is answerable to parliament andpeople. In this idealized account, the governmentforms its policy and interest groups enter thepolicy cycle only at the stage of implementation.The groups operate within the strategic frameworkprovided by the state. Eising and Cini (2002, p.171) cite France – the home of popular sover-eignty – as an example of a country which aspiresto this approach. French political tradition stressesthe importance of a collective will, articulated bythe state, which is superior to particular interests.

But such accounts, whether of democracies ingeneral or of France in particular, are unrealistic.They understate the role of the groups, and over-state the role of the government, in shaping policy.To accommodate this point, political scientistshave developed two broad models of the politicalrole of interest groups: pluralism and corporatism.These two models usefully direct our attention tothe contrasts between the pluralistic politics of theUnited States (see p. 179) and the more corporatepolitics found in much of continental Europe. Sothe issue here is to demonstrate two ways in whichinterest groups can be integrated into the heart ofpolitical decision-making.

DefinitionLiterally ‘rule by the many’, pluralism refers to apolitical system in which numerous competinginterest groups exert strong influence over aresponsive government.The state is moreumpire than player. However each of thesegroups concentrates on its own area (forexample education or health care) so that nosingle elite dominates all sectors.

The debate between pluralists and corporatistsgoes to the heart of a central question in politics:

the relationship between society and the state.Pluralists see society dominating the state; corpo-ratists view the state as leading society. At one levelthis difference is descriptive, reflecting contrastingassessments of the flow of influence between gov-ernment and interest groups. But at a deeper level,the debate reflects contrasting views of the properrole of government. Pluralists see the state’s task asresponding to interests expressed to it.Corporatists, by contrast, favour an organized,integrated society in which the state offers leader-ship in pursuit of a vision shared with society.

The pluralist view dominated early postwaraccounts of interest group activity. In this account,politics is seen as a competition between a multi-tude of freely organized interest groups. Theseorganizations compete for influence over a govern-ment that is willing to listen to all the voices it canhear in the political debate. In a pluralist system,the state is largely an arena for competitionbetween interest groups. The governing party is anarbiter, not an initiator. For Bentley (1908), anAmerican pioneer of this approach, ‘when thegroups are adequately stated, everything is stated.When I say everything, I mean everything.’ Allkinds of interest have their say before the court ofgovernment. Groups compete on a level playingfield, with the state showing little bias eithertowards interests of a particular type or towardsspecific groups within that type. As new interestsand indeed new identities emerge, groups form torepresent them. In pluralism, politics is a competi-tive market with few barriers to entry.

Pluralism brings healthy fragmentation acrossthe range of government activity since most inter-ests are restricted to a single policy sector. Indeedthe central tenet of pluralism is that no single elitedominates government. Rather, different interestgroups lead the way in each area of policy. Overall,pluralism depicts a wholesome process of dis-persed decision-making in which the variety ofgroups allows government policies to reflect notjust economic interests but also social diversity.

The significance of pluralism lies in its implica-tions for our understanding of contemporarydemocracy. Pluralists accept that, in practice,majority rule is an inaccurate account of howdemocracies work. Rather, pluralists judge thatdemocracy is, in reality, rule by a series of minori-

INTEREST GROUPS 177

ties, each operating in a particular area but subjectto the checks and balances of other groups oper-ating in the same sector, with the government asumpire. Furthermore, the strengths of this form ofgovernance are substantial. Dahl (1993, p. 706)summarizes what he sees as the strengths of plu-ralism, suggesting that groups have

served to educate citizens in political life,strengthened them in their relations with thestate, helped to ensure that no single interestwould regularly prevail on all important deci-sions, and, by providing information, discus-sion, negotiation and compromise, even helpedto make public decisions more rational andmore acceptable.

Corporatism provides a contrasting account ofthe role of interest groups in national politics.Where pluralism was inspired by the UnitedStates, corporatism was developed in Europe. Inthe Middle Ages, a corporation referred not to aprofit-making company but to a craft guild withcontrol over recruitment to, and regulation of, itsprofession. The idea that people should be repre-sented and controlled through such self-regulatingoccupational bodies remains important to moderncorporatism.

How does corporatism conceive the relationshipbetween interest groups and government? Themain contrast with pluralism is that where plu-ralism implies competition between groups, cor-poratism implies coordination and planning. Inconjunction with government, the peak associa-tions representing capital and labour make andimplement key policies, thus securing the socialorder in societies which often have a long historyof conflict. This is a top-down approach, requiringa high level of social and political organization,including compulsory membership of firms inemployers’ organizations and of employees intrade unions.

Though formally accountable to their members,a central role of the peaks in a corporate system isto carry their members with them after agreementshave been struck with the government. Issues suchas price levels, wage increases, tax rates andpension entitlements are all settled in tripartitediscussions among senior politicians, industrialists

and union leaders. Agreements are often then pre-sented as an agreed social pact or social contractbetween the actors in the social partnership.

Under corporatism, negotiations between thestate and recognized groups take an administra-tive, technical form. Policy-making is depoliticizedand electoral representation through parliamentbecomes less important. Policy is made in privatenegotiations in government ministries, not in thepublic debate of the assembly. Corporatism alsoimplies a hierarchy of groups, with the govern-ment dealing with the influential peak associa-tions, which then pass decisions on to theirmembers.

DefinitionIn a democratic context, corporatism (oftencalled social partnership in Europe) is a relation-ship between the state and interest groups inwhich major decisions on domestic mattersemerge from discussion between the govern-ment and peak associations representing themajor social partners: capital and labour. Inreturn for influence, the partners ensure thecompliance of their members.This system isoften called liberal or societal corporatism to dis-tinguish it from the state-dominated corporatismof fascism (Schmitter and Lehmbruch, 1979).

So in contrast to pluralism, which emphasizes anupward flow of preferences from group membersto their leaders and then on to government, corpo-ratism stresses the downward flow of influence.Politicians govern with interest groups and are cer-tainly not governed by them. The state retains atleast a coordinating role and may even governthrough interest groups, persuading the peaks toaccept necessary reforms.

An example will add some colour to this outline.Just as the United States is taken to exemplify plu-ralism, so Austria’s postwar republic reveals manycorporatist features. In common with most coun-tries adopting a corporatist approach, Austria’shistory is troubled indeed. Civil war in the 1930swas followed by annexation by Nazi Germanyand, after 1945, by a further ten years withRussian troops on its soil. Prompted by thishistory, the country’s two blocs – conservative/Catholic and socialist – eventually agreed to

178 LINKING SOCIETY AND GOVERNMENT

INTEREST GROUPS 179

Both as an ideal and as a description of how politicsworks, pluralism draws on American experience. Butcan American politics really be presented as an opencompetition between freely organized interests? Or isthe entire decision-making process in the hands of anelite which, in effect, excludes consideration of anypolicies which would threaten its own values? IsAmerican pluralism reality or myth?

The case for In the USA, interest group patterns come closer to thepluralist model than anywhere else. As de Tocqueville(1856) wrote,‘in no country in the world has the prin-ciple of association been more successfully used, orapplied to a greater multitude of objects, than inAmerica’. Petracca (1992, p. 3) makes the point moresuccinctly:‘American politics is the politics of inter-ests.’

Nowhere else are interest groups so numerous,visible, organized, competitive or successful.Tens ofthousands of groups, ranging from Happiness ofMotherhood Eternal to the United Autoworkers ofAmerica, seek to influence policy at federal, state andlocal levels. Nor are such groups confined to protec-tive economic interests. Promotional groups areuniquely prominent in the USA with over 500 groupsfocusing on environmental protection alone.

Washington politics reflects the competitive spiritthat is pluralism’s hallmark. If one interest groupseems to be gaining the upper hand, others will formto counter its influence.Vigorous, independent andcompetitive media are always willing to listen to newgroups with a story to tell.The separation of powersgives interest groups many points of leverage:Congressional committees, executive agencies andthe courts. American government is simply too frag-mented to be anything more than an umpire of groupdemands.

The case againstAll political systems generate myths and America’s ispluralism. In reality, pluralist ‘competition’ operateswithin an unquestioned acceptance of broad

American values favouring the free market and thepursuit of individual self-interest.The entire politicaldiscourse works within a narrow ideological range,shaping and limiting the demands expressed so as tobenefit, in particular, corporate USA.

Washington’s intricate political games are dominatedby middle-aged, middle-class English-speaking gradu-ates, a group which forms only a small minority of anethnically and linguistically diverse population. In anycase, some interest groups are wealthier and betterorganized than others so that the pluralist ideal ofequal representation for all groups is far from reality.And, as the frequency of inner-city disturbancesreveals, some interests are left out of the debate alto-gether. As Schattschneider (1942) claimed,‘the systemis skewed, loaded and unbalanced in favour of a frac-tion of a minority’. Schattschneider’s critical conclu-sion remains valid: some interests are organized intoAmerican politics but others are organized out.

AssessmentThe danger of the debate is that the word ‘pluralism’becomes evaluative rather than descriptive.Thosewho favour the American way see pluralism; thosewho are more critical discern a hidden elite.

But one conclusion, at least, seems clear.While over-seas observers may interpret American politics asmuch ado about nothing, that view is not shared bythe players themselves.To understand what happensin American politics (as opposed to what overseasobservers would like to happen), it is necessary toappreciate the vigorous competition betweeninterest groups, even if this debate does operatewithin in a limited framework. For the thousands ofgroups seeking government subsidies for the inter-ests they represent, American politics is a serious busi-ness.

Further reading: Cigler and Loomis (2002), Dahl (1961), Mills(1956), Schattschneider (1942).

DEBATE

IS THE USA A PLURALISTIC POLITICAL SYSTEM?

discard past differences in what proved to be a suc-cessful search for compromise.

Austria developed an elaborate corporate systemof elaborate and interlocking institutions lastinguntil the end of the century. The Catholic andsocialist parties, representing each bloc, dominatednational politics and usually governed in coalition,where careful compromises could be negotiated.Just as important, the peak associations – theÖGB on the employers’ side and the VÖI forlabour – were powerful and centralized bodieswith high membership density. The leader of thesetwo organizations would agree important deci-sions affecting the whole economy, with the chan-cellor (that is, prime minister) acting as chair ofthe meeting. Furthermore, all working peoplebelonged to one of three statutory chambers ofcommerce, covering business, labour and agricul-ture. These chambers provided an additionalchannel of communication between the state andsocial interests (Tálos and Kittel, 2002).

In general, corporatism is most likely to takeroot in countries with recent experience of dis-order and in which a strong state meets powerfulpeak associations with high membership density.The Scandinavian nations and the Netherlands, aswell as Austria, fit many of these conditions.Scandinavia exhibits a tradition of compromiseand consultation between government and interestgroups which some authors argue amounts toNordic corporatism. Kvavik (1976), for example,interpreted Norway in the 1970s as a ‘cooptivepolity’ in which interest groups were brought intopublic policy-making and implementation in adistinctly non-pluralist way. Sweden also involvesinterest groups in a remarkably deliberativeprocess of policy formation.

In the 1980s and 1990s, corporatism cameunder attack. Even in prosperous Austria, thosewho felt left out of the system began to voice theirprotest. The country’s extreme right FreedomParty became one of the most successful populistparties in Western Europe, eventually enteringgovernment in a remarkable coalition with theconservative People’s Party in 2000. This coalitionsymbolized a thawing of Austria’s previously frozencorporate system. The coalition was renewed after2002, but with the Freedom Party in a weakerposition, enabling the leader of the People’s Party

to begin more radical attacks on the country’sentrenched corporate traditions.

However, it was in the Anglo-American world,where corporatism had never been more than half-hearted, that the heaviest blows were struck. Asearly as 1982, Olson had claimed that corporatismwas a form of political sclerosis, reflecting thegradual accumulation of power by sectional inter-ests. Corporatism was viewed as inviting excessivestate intervention. The carefully crafted corporateconsensus between government, capital and labourwas held to inhibit the continuous economicchanges needed to remain competitive in anincreasingly global economy.

Reflecting this intellectual sea change, MargaretThatcher in Britain, and Ronald Reagan in theUnited States, launched a political assault in the1980s on the power of entrenched interests. Whilesuch attacks were focused on the privileges oforganized labour, some large companies were alsoaffected by more vigorous competition policy,

180 LINKING SOCIETY AND GOVERNMENT

Most corporatistAustriaDenmarkSwitzerlandGermanyFinalndBelgiumIreland

Many corporate featuresNorwaySwedenNetherlands

A few corporate featuresFranceItaly

Least corporatistCanadaUnited KingdomUnited States

Source: Adapted from expert assessments in Lijphart and Crepaz(1991).

B OX 1 0 . 3

Democracies grouped by theextent of corporatism

more open markets and a growing emphasis oncompetitive tendering in the public sector.Structural change, particularly the decline of heavyindustry with its strong unions, large companiesand powerful trade bodies, accelerated the decayof organized labour, weakening one of the corpo-rate pillars (Addison and Schnabel, 2003).

Writing on Scandinavia, Eriksen (1990) arguedthat ‘the post-corporate state may represent a neworder’. But it remains to be seen how far mostEuropean countries, with their long history of cor-porate thinking and the inherent value they placeon governing by consensus, have the desire, theability or indeed the need to change their ways.Blom-Hansen (2000), for one, argues that corpo-ratism is alive and well and living in Scandinavia,albeit in the form of continuing consultationrather than joint decision-making between groupsand government. And some countries, includingIreland and the Netherlands, have even developedor revived social partnerships as a route toimproved efficiency.

Interest groups in new democracies

The role to be played by interest groups in newdemocracies remains uncertain, intimately linkedto the extent to which democracy itself consoli-dates. Certainly, the emergence of groups actingindependently of the state was integral to theweakening of authoritarian rule. For instance,groups based around the green, peace andwomen’s movements developed in a number ofcommunist states in the 1980s, posing a challengeto party rule by their mere existence as non-partyorganizations. Eventually, some of these groupsbecame the catalyst of regime change. In Poland,the trade union organization Solidarity (sup-ported by the Roman Catholic Church) emergedin 1980 to assert the interests of Polish workers.Within the decade, it took over the reins of government.

However, groups such as Solidarity were notinterest groups in the Western sense; they began asde facto opposition groups and then became broadsocial movements. They sought to replace ratherthan to influence the communist government.Their historical task was pivotal but short-term

and went far beyond anything of which a narrowinterest would have been capable.

In the democratic era, the groups which werethe levers of regime change – including environ-mental and women’s groups – have generallydeclined. Similarly, the churches have paradoxi-cally declined because their position as a ‘semi-protected site of opposition’ to communist rule isno longer needed in the new, freer order(Fitzmaurice, 1998, p. 174).

As these social movements disappear and someinstitutional interests decay, so orthodox interestgroups, with their detailed and routine demands,should take root in democracy’s fresh soil. Ågh(1998, p. 22) suggests that the ‘the chief actors ofdemocratic transition are the parties, but those ofdemocratic consolidation are the interest organiza-tions and civil society associations, which providethe fine-tuning and effect the full accomplishmentof democratization’.

Yet this is precisely where the uncertaintyresides. As yet, interest groups are not consistentlydeveloping in new democracies along Westernlines, any more than political parties are growinginto the mass membership organizations oncefound in West Europe. Certainly, some of the‘older’ new democracies of Southern Europe –notably Spain and Portugal – have moved towardsthe interest group pattern found in consolidateddemocracies. Yet even in Spain, Heywood (1995,p. 243) notes that in a context of ‘the continuingprimacy of the state over civil society’, member-ship of voluntary associations is falling, not rising.Padgett (2000, p. 166) reports that trade unionmembership has also declined over the briefhistory of post-communist Europe.

Even in the eastern areas of Germany, now for-mally part of the highly organized Germaneconomy, large firms are preoccupied with directlobbying and small ones with survival, leavinglittle space for interest groups representing specificindustries or general business values. Padgett con-cludes that ‘nowhere in east/central Europe is eventhe semblance of a stable, fully functioninginterest group system’ to be found. Western peakassociations and chambers of commerce are weakor non-existent.

A similar story emerges from the post-militarydemocracies of Latin America. Hagopian (1998,

INTEREST GROUPS 181

p. 238) considers that the rise of markets meansthat ‘the corporate negotiating channels once openbetween unions and the state are being renderedobsolete’.

However, the transition to democracy did releasesome new forces. Many new democracies wit-nessed a release of traditional social groups ratherthan interest groups. In the post-communistworld, long suppressed groups based on ethnicityand nationality, have proved to be a potent forcefor (and response to) instability. For example,long-standing national differences brought aboutthe dissolution of Czechoslovakia into the CzechRepublic and Slovakia.

In much of Africa, too, group politics in an eraof post-military government continues to be basedon ethnicity rather than interests generatedthrough the workplace. Many poorer countrieslack the complex economy needed to developinterest group patterns found in affluent estab-lished democracies. In Botswana, for example,only two organizations – a business associationand a conservation group – employ staff to lobbythe government (Herbst, 2001). In addition, whensuch economic resources as are available are par-celled out on ethnic lines, or in any other personalor unregulated way, incentives for orthodoxinterest groups to develop are limited.

Russia’s semi-democracy is, as always, an inter-esting mixed case. Certainly, the separationbetween public and private sectors, so central tothe organization of interests in the West, has notfully emerged there. Particularly in the early post-communist years, ruthless business executives,corrupt public officials and jumped-up gangstersmade deals in a virtually unregulated free-for-all.Individual financiers pulled the strings of theirpuppets in government but the politics was per-sonal rather than institutional. In such an environ-ment interests were everywhere but interest groupswere nowhere. ‘Comrade Criminal’ was disin-clined to join trade associations.

However, the situation may be changing. Witheconomic recovery taking hold at the start of thetwenty-first century, some business associations ofa Western kind have emerged, even if they havenot yet secured political influence. Peregudov(2001, p. 268) even claims that ‘in Russia anetwork of business organizations has been created

and is up and running’. He suggests that thisnetwork is capable of adequately representingbusiness interests to the state but it has so farreceived little attention from President Putin, whocontinues to reward his business friends and, onoccasion, to arrest his enemies.

To assume that new democracies will in duecourse replicate the group patterns of establisheddemocracies places a large bet on both economicdevelopment and democratic deepening. Just asmany newer democracies led the way in devel-oping media-based rather than mass membershipparties, so too do direct links between businessowners and government leaders lead us to ques-tion whether the Western model of industry-widetrade associations will ever become a potent forcein new democracies.

Interest groups in authoritarian states

The role played by interest groups in non-democ-ratic states provides a sharp contrast to their posi-tion in established democracies. Authoritarianrulers see freely organized groups as a potentialthreat to their own power; hence, they seek eitherto repress such groups or to incorporate themwithin the power structure. In this section, we willexamine the workings of these strategies in author-itarian regimes before turning to the special case oftotalitarian states.

In the latter half of the twentieth century, manyauthoritarian rulers had to confront the challengeposed by the new groups unleashed by economicdevelopment. These included labour unions,peasant leagues and educated radicals. How didrulers respond to these new conditions? Onestrategy was to suppress such groups completely.Where civil liberties were weak and many groupswere new, this approach was feasible. For example,a strategy of repression was adopted by many mili-tary regimes. Military leaders often had their ownfingers in the economic pie, sometimes in tacitcollaboration with multinational corporations; therulers’ goal was to maintain a workforce that wasboth compliant and poorly paid. ‘Trouble-makers’seeking to establish labour unions were quicklyremoved.

On the other hand, authoritarian rulers could

182 LINKING SOCIETY AND GOVERNMENT

seek to manage the expression of these new inter-ests. That is, they could allow interests to organizebut seek to control them, a policy of incorporationrather than exclusion. By enlisting part of the pop-ulation, particularly its more modern sectors, intoofficially sponsored associations, rulers hoped toaccelerate the push toward modernization. Thisapproach was common in Latin America, wherethe state licensed, funded and granted a monopolyof representation to favoured groups.

Before the pro-market economic reforms of the1980s and 1990s, Mexico offered a particularworking of this tradition. Its governing system wasfounded on a strong ruling party (the PRI) whichwas itself a coalition made up of labour, agrarianand ‘popular’ sectors (the latter consisting mainlyof public employees). Favoured unions andpeasant associations gained access to the PRI.Party leaders provided resources such as subsidiesand control over jobs to these groups, in exchangefor their political support. In effect, Mexicobecame a giant patron–client network – a form ofcorporatism for a developing country. For themany people left out of the network, however, lifecould be hard indeed.

Just as corporatism is decaying in establisheddemocracies, so Mexico’s system – and many otherslike it – is also in decline. It was over-regulated,giving so much power to civil servants and PRI-affiliated unions as to deter business investment,especially from overseas. As the market sectorexpanded, so the patronage available to the PRIdiminished. In 1997, an independent NationalWorkers Union emerged to claim that the oldmechanisms of state control were exhausted, apoint which was confirmed by the PRI’s defeat in apresidential election three years later.

The position of interest groups in communiststates was even more marginal than in other non-democratic regimes. For most of the communistera, independent interest groups did not exist.Their absence was a deliberate result of commu-nist ideology. Communist states were led by theparty, not by society. Groups served the party, notthe other way round. Interest articulation by freelyorganized groups was inconceivable. Communistrulers sought to harness all organizations intotransmission belts for party policy. Trade unions,the media, youth groups, professional associations

were little more than branches of the party, servingthe great cause of communist constructionthrough social transformation.

However, the capacity to articulate interests didincrease as communist economies matured. Theuse of coercion and terror declined as conflict overpolicy became more visible. Institutional groupssuch as the military and heavy industry becamemore important as decisions became more tech-nical. In the 1970s and 1980s, sectional interestsbegan to be openly expressed, particularly inPoland, Hungary and Yugoslavia.

However, one sharp contrast with Western plu-ralism remained: ruling communist parties tried torestrict interest articulation to safe technicalmatters. The party continued to crack down vigor-ously on dissent going beyond these confines. Theobjectives of the communist state remainedbeyond criticism. Thus ‘socialist pluralism’, to theextent that it existed at all, remained far morelimited than its Western counterpart.

Even in reforming China, the Western notion ofan ‘interest group’ still carries little meaning.China’s Communist Party continues to providethe framework for most formal political activity.‘Mass organizations’ such as the All-ChinaFederation of Trade Unions and the Women’sFederation are led by party officials and transmitits policy. Saich (2004, p. 206) notes that ‘stateentities have given birth to many of the new socialorganizations in China’. These bodies aredescribed in China as GONGOs (government-organized non-governmental organizations).

Private business is intertwined with the statesector and is not represented through traditionalindustry associations. Too many deals are done‘through the back door’ to leave much space forpolicy-oriented interest groups. ‘Rightful resis-tance’ enables citizens to protest to higherauthority about lower officials exceeding theirlegal powers but such appeals operate on an indi-vidual rather than a group basis. As Manion(2004, p. 448) concludes, ‘for the most part, thefunction of interest aggregation is monopolized bythe communist party’.

In fascist theory, as under communism, the statedominated partial interests. Indeed, the centralpremise of fascist thought was that the state mustlead. But unlike communism, the fascist state

INTEREST GROUPS 183

sought to mobilize, rather than to destroy, groupactivity. In particular, it wanted to exploit, ratherthan take over, private industry. Fascism advocateda corporate relationship between state andindustry (Brooker, 2000, p. 156).

The theory was that the economy would bearranged by industrial sector. Within eachindustry, special corporations (committees orchambers) composed of employers, employees,party and government would plan production, setwages and prices and resolve disputes. In this ver-tical arrangement, horizontal conflicts betweenbusiness groups and labour unions would be over-come as both sides learned to serve national goalsas defined by the supreme leader. Thus corpo-ratism would secure the national interest by over-riding class conflict.

Such a system was formally implemented inItaly in the mid-1930s but in practice the 22 cor-porations established there were given only tokenpowers. Their impact was minimal. Corporatismwas even less significant in Nazi Germany.Indeed, the man charged by Hitler with intro-ducing the format there claimed that afterlearning of his assignment, ‘I did not sleep forseveral nights on account of the corporate systembecause I could not make head or tail of it’(O’Sullivan, 1986, p. 133). In practice, industrial

policy took the practical form of ensuring thatlarge privately owned manufacturers met thedemands of an expansionist regime. Corporateinstitutions took second place to the task ofserving the Nazi war machine.

Key reading

Next step: Many of the best studies of interestgroups continue to be about the United States;Cigler and Loomis (2002) is an excellent collec-tion.

Wilson (1990b) is a clear and straightforwardintroduction to interest groups. Hrebenar andScott (1997) is a good text on the USA, Eisingand Cini (2002) cover Western Europe while M.Smith (1995) is a brief but helpful introduction toBritain. On European corporatism, Berger andCompston (2002) is a useful collection whileArter (1999) and Blom-Hansen (2000) review therelevance of the concept to the Nordic states. Twouseful comparative studies are Wilson (2003) onthe relationship between business and politics andThomas (2001) on the links between parties andinterest groups. Brown (2001) covers groups, andmuch else, in post-communist Russia.

184 LINKING SOCIETY AND GOVERNMENT

Organization Type Comment

All-China Federation of Trade Unions Mass organization A traditional transmission belt for theparty

All-China Women’s Federation Mass organization Traditionally a party-led body, this federation has created some space for autonomous action

China Family Planning Association Non-governmental Sponsored by the State Family Planning organization Commission, this association operates at

international and local level

Friends of Nature Non-governmental Led by a well-connected and charismatic organization figure, Friends of Nature has mobilized

student support in defence of habitats threatened by illegal logging

Source: Adapted from Saich (2004), pp. 186–92.

B OX 1 0 . 4

Examples of social organizations in contemporary China

Party organization 186Selecting candidates and leaders 189Membership and finance 191The social base 193 Party systems 194Parties in new democracies 200Parties in authoritarian states 201Key reading 205

‘In this book I investigate the workings ofdemocratic government. But it is not institu-

tions which are the object of my research: it is noton political forms, it is on political forces I dwell.’So Ostrogorski (1902) began his pioneering com-parison of party organization in Britain and theUnited States. Ostrogorski was one of the first stu-dents of politics to recognize that parties werebecoming vital in the new era of democratic poli-tics: ‘wherever this life of parties is developed, itfocuses the political feelings and the active wills ofits citizens’.

Ostrogorski’s supposition that parties weregrowing in importance proved to be fully justifiedin the twentieth century. In Western Europe, massparties battled for the votes of enlarged electorates.In communist and fascist states, ruling partiesmonopolized power in an attempt to reconstructsociety. In the developing world, nationalist partiesbecame the vehicle for driving colonial rulers backto their imperial homeland. In all these areas,parties succeeded in drawing millions of peopleinto the national political process, often for the firsttime. The mass party proved to be the key mobi-lizing device in the politics of the twentieth century.

In standing between the people and the state,parties continue to perform four main functions:

� Ruling parties offer direction to government, per-forming the vital task of steering the ship ofstate.

� Parties function as agents of political recruit-ment. They serve as the major mechanism for

preparing and recruiting candidates for the leg-islature and the executive.

� Parties serve as agents of interest aggregation.They filter a multitude of specific demands intomore manageable packages of proposals. Partiesselect, reduce and combine policies.

� To a declining extent, political parties also serveas a brand for their supporters and voters,giving people a lens through which to interpreta complicated political world.

DefinitionSartori (1976, p. 63) defines a political party as‘any political group identified by an official labelthat presents at elections, and is capable ofplacing through elections candidates for publicoffice’. Unlike interest groups, which seek influ-ence only, serious parties aim to secure the leversof government. In Weber’s phrase, parties live ‘ina house of power’.

Political parties are complex organizations, oper-ating across domains but with the various levels ofthe party sharing a common identity and abroadly similar outlook. In the parliamentarydemocracies of Western Europe, parties are partic-ularly complex; Box 11.1 sets out the elements ofa typical major party in that region.

At national level, the main distinction isbetween the top party leaders (or ministers if theparty is in power), the members of the parliamen-tary party and the officials working at party head-quarters. But parties are also represented in otherelected domains, including the EuropeanParliament, and both regional and local govern-ment. This superstructure is supported at the baseby ordinary members, usually organized by area.Clearly, mapping a party’s structure, and under-standing the relationships between its parts, isitself a significant task.

In Western Europe, parties typically possess alarge if declining subscription-based membership,

Chapter 11

Political parties

185

a coherent ideology and strong discipline amongtheir members of parliament. By contrast, inNorth America, party organization is weaker andeven more decentralized. Canada as well as theUnited States lack the European tradition of alarge, dues-paying membership. Beyond the legis-lature, American and Canadian parties are largelydevices for organizing elections; they often seem tohibernate between campaigns. Volunteer helperswith campaigns often have no formal links withthe party and, in the USA at least, candidateschoose parties more than the other way round.

The question for the twenty-first century iswhether we are witnessing a crisis of parties andthe export to the rest of the democratic world ofthis North American format of weak, decentral-ized organizations. The evidence for such a crisis iscertainly mounting:

� The major parties no longer offer radically dif-ferent visions of the good society.

� Voters’ loyalties to party are weakening as tradi-tional social divisions decay.

� Party membership is falling and ageing.� Leaders increasingly communicate with electors

through television rather than the party.� Party income increasingly depends on state sub-

sidies rather than members’ subscriptions.

No longer do parties seem to be energetic agentsof society, seeking to bend the state towards theirmembers’ interests. Rather, they are in danger ofbecoming political pensioners, living off pastglories and facing an uncertain future. IfOstrogorski were writing today, would he stillinterpret parties as a ‘focus for the active wills’ ofthe citizens?

Party organization

As Panebianco (1988) reminded us, internal orga-nization is a key issue in the study of parties. Howis power distributed within the party? What is therelationship between leaders, members and parlia-mentarians? The answer to these questions,Panebianco claims, must be historical. He placesspecial emphasis on a ‘genetic’ account of partydevelopment, a term he uses to stress the impor-tance of the party’s founding moment in dealingout the power cards between the elements of partyorganization. These ‘continue in many ways tocondition the life of the organization even decadesafterwards’. In this section, we will examine a clas-sification of parties based on their origins beforeturning to the question of the distribution ofpower within them.

186 LINKING SOCIETY AND GOVERNMENT

Element Description Level

The party in government The prime minister and other ministers (or the leadership National(or opposition) of non-governing parties)

The party in parliament ‘Backbench’ members of the parliamentary party National

The party’s central Officials at party headquarters Nationalorganization

The party at regional and Party members elected to regional and local assemblies; Subnationallocal levels local officials and ordinary party members

The party in the European Members of the European Parliament, where SupranationalUnion representatives of a national party join broader,

transnational party groups

Source: Adapted from Cotta (2000).

B OX 1 1 . 1

Typical elements of a major political party in Western Europe

Types of party organization

Adopting Panebianco’s historical approach leads to athreefold distinction between elite, mass and catch-all parties (Box 11.2). Elite parties are ‘internallycreated’. They are formed by cliques within anassembly joining together to reflect common con-cerns and then to fight effective campaigns in anenlarged electorate. The earliest nineteenth-centuryparties were of this elite type: for example, theConservative parties of Britain, Canada and Scan-dinavia. The first American parties, the Federalistsand the Jeffersonians, were also loose elite factions,based in Congress and state legislatures. Reflectingthis character, elite parties are sometimes calledcaucus parties, the ‘caucus’ denoting a closedmeeting of the party’s members in the legislature.

Mass parties are a later innovation. They origi-nate outside the assembly, in groups seeking repre-sentation in the legislature for their interests andgoals. The working-class socialist parties thatspread across Europe around the turn of the twen-tieth century epitomized these externally createdparties. The German Social Democratic Party(SPD), founded in 1875, is a classic example.Such parties acquired an enormous membershipand, in contrast to elite parties, sought to keeptheir representatives in parliament on a tight rein.

These socialist parties exerted tremendous influ-ence on European party systems in the twentiethcentury. In particular, they stimulated many eliteparties to copy their extra-parliamentary organiza-tion. In Germany, again, the Christian DemocraticUnion (CDU) was created after the Second WorldWar to offer a broadly-based Christian alternativeto the SPD. Green parties, too, show features ofthe mass party, seeking representation for a newsocial interest. The United States, unusually, neverdeveloped substantial mass parties.

The catch-all party is a more recent form.Kircheimer (1966) used this phrase to describe theoutcome of an evolutionary path followed bymany parties, both elite and mass, in response topost-1945 conditions. The catch-all party is aresponse to a mobilized political system in whichgoverning has become more technical and inwhich electoral communication takes placethrough the mass media. Their leaders communi-cate with the voters through television, bypassingwhat is still a large membership. Such parties seekto govern in the national interest rather than as representatives of a social group. Catch-allparties seek electoral support wherever they canfind it; their purpose is not to represent but togovern.

POLITICAL PARTIES 187

Elite party Mass party Catch-all party

Emergence: 19th century 1880–1960 After 1945

Origins: Inside the assembly Outside the assembly Developed from existing elite or mass parties

Claim to support: Traditional status of leaders Represents a social group Competence at governing

Membership: Small, elitist Large card-carrying Declining, leaders become membership dominant

Source of income: Personal contacts Membership dues Many sources, including state subsidy

Examples: 19th-century conservative Socialist parties Many modern Christian and and liberal parties, many Social Democratic parties in post-communist parties Western Europe

Source: Adapted from Katz and Mair (1995).

B OX 1 1 . 2

Types of party organization

The broadening of Christian Democratic parties(such as the CDU in Germany) from religiousdefence organizations to broader parties of thecentre-right is the classic example of the transitionto catch-all status. Indeed the German notion of aVolkspartei (‘people’s party’) with its support froma wide range of people captures the catch-all idea.The subsequent transformation of several radicalsocialist parties into leader-dominated socialdemocratic parties, as in Spain and the UnitedKingdom, is another example of this shift.America’s parties, it might be argued, went straightfrom elite to catch-all status, omitting the massstage.

Power within the party

Given the complex nature of modern parties, it isnatural to ask where authority within them reallyresides. Of all the elements in Box 11.1, whichone really commands the party? The answer isnot always clear and the question itself is ratherblunt. American parties, in particular, are some-times seen as empty vessels waiting to be filled byfresh ideas and ambitious office-seekers. TheAmerican party is not controlled from any singlepoint; no one pulls the levers, no one rules theparty.

Yet much European research on parties doessuggest that, in general, authority within the partyflows from the top down, with the leaders whorepresent the party to the public playing a keyrole. In 1911, the German scholar RobertoMichels (1875–1936) published Political Parties,perhaps the most influential work on the distribu-tion of power within parties. Michels argued thateven organizations with democratic pretensionsbecome dominated by a ruling clique of leadersand officials. Using Germany’s Social DemocraticParty (SPD) as a critical case, Michels suggestedthat leaders develop organizational skills and aninterest in their own continuation in power. Theordinary members, aware of their inferior knowl-edge, accept their own subordination as natural,even in a party such as the SPD with democraticpretensions.

Michels’s iron law is that power within parties,as within other organizations, ends up in thehands of the leaders. The law is almost a century

old but still possesses considerable validity. Today,it helps to explain why Green parties, with a com-mitment to internal democracy that at leastmatches that of Michels’s SPD, have seen theirleaders acquire more authority as their organiza-tions have matured.

DefinitionMichels’s iron law of oligarchy states that ‘whosays organization, says oligarchy’.The leaders oforganizations, including political parties, developexpert knowledge, specialist skills and a commit-ment to their own power.Together, these factorsensure that even parties formally committed todemocracy become dominated by a ruling elite.(Note: oligarchy is rule by and for the few.)

More specifically, in the parliamentary systemsof Europe, the party leaders in the legislature arenormally the key actors. When their party is inpower, they usually provide the ministers of thegovernment, including the prime minister, with allthe publicity which flows from such positions.Even in opposition, the legislative leaders are oftenthe party’s public face. Perhaps surprisingly, thispoint applies to mass as much as to elite parties. Ina famous study of Britain’s Labour Party stronglyinfluenced by Michels, McKenzie (1955, p. 365)argued that authority within the party does andindeed should rest with its parliamentary leadership. McKenzie concluded that, ‘whateverthe role granted in theory to the extra-parliamen-tary wings of the [Conservative and Labour]parties, in practice final authority rests in bothparties with the parliamentary party in its leader-ship’.

We should, however, note that the party organi-zation outside the assembly retains some usefulpower cards. State financial aid normally goes tothe party bureaucracy, not to the party in parlia-ment. And only party officials can cope withincreasingly technical tasks such as recruitingmembers and raising funds through mail-shots, orarranging for advertising, briefings and press con-ferences during election campaigns. The contem-porary importance of these tasks is captured inPanebianco’s concept (1988) of an electoral–pro-fessional party centred around fighting electionsthrough the mass media.

188 LINKING SOCIETY AND GOVERNMENT

Selecting candidates and leaders

Elite recruitment is a vital and continuing func-tion of parties. Even as parties decline in otherways, they continue to dominate elections to thenational legislature from which, in most parlia-mentary systems, the nation’s leaders are drawn.Given that candidates who are nominated for safedistricts or who appear near the top of their party’slist are virtually guaranteed a place in parliament,it is the selectorate (selectors of candidates), notthe electorate, which is the gatekeeper to thehouse of power. As Schattschneider (1942, p. 46)famously wrote, ‘the nature of the nominatingconvention determines the nature of the party; hewho can make the nominations is the owner of theparty’. How, then, do the major parties select theircandidates and leaders?

Candidates

In choosing candidates for the legislature, partiesare conditioned by the electoral system. Under thelist form of proportional representation, partiesmust develop a ranked list of candidates, either atnational or more commonly at regional level, topresent to the electorate. This task usually involvescompromises brokered by party officials betweenthe various factions and interests within the party.

Take the Netherlands as an example. Hollanduses national party lists so each party needs to

prepare just one list of candidates covering thewhole country. In the major parties, a nominatingcommittee begins the selection process, examiningapplications received either from local branches ordirectly from individuals. A senior party board thenproduces the final ordering, with the party leaderserving as ‘list-puller’ – that is, occupying numberone position. The board will seek a balance:between incumbents and fresh candidates, and alsobetween genders. In 2002, for instance, the DutchLabour Party alternated male and female candi-dates in the first 40 positions on the list.

In a few countries, candidate lists are preparedthough a ballot of party members. This procedureis more democratic but, as Michels might havepredicted, it causes difficulties of its own. Ballotsadvantage celebrity over competence, and appli-cants’ wealth over their party experience. In prac-tice, membership ballots are also incompatiblewith the goal of a balanced list. Significantly, themajor Israeli parties introduced ballots ofmembers early in the 1990s but most withdrewthe procedure before the 1999 election. Thecurrent procedure of selection by committee in theNetherlands also reverses a previous policy ofmembership ballots.

In the few countries using plurality electionsbased on local electoral districts, the nominationprocedure is naturally more decentralized.Candidates must win selection by local partieskeen to guard their autonomy against encroach-

POLITICAL PARTIES 189

Electoral system used for How parties select candidates for these elections Examplenational elections

Proportional representation Party officials (or special party conventions) draw Netherlands(party list) up a ranked list of candidates

Plurality system* Local parties select the candidate, sometimes Canada, United drawing from a list prepared by head office Kingdom

Mixed system The party draws up a list for the PR element and Germany local parties select a candidate for the district contest

* In the USA, primary elections are held among a party’s registered supporters in the area. On electoral systems, see p. 148.

B OX 1 1 . 3

Selecting candidates for legislative elections

ment from headquarters. In Canada, for instance,constituency parties seek candidates with anattractive local profile, showing little concern fornational needs (Carty, 2002). In Britain, selectionis also though local associations, drawing on listsof potential candidates approved by party head-quarters. Some democratization has occurred evenwith local selection, however, with the nominationmeeting – previously the preserve of the localmanagement committee – sometimes now open toall local members.

The USA has developed the exceptional deviceof primary elections to broaden the selection ofcandidates in a plurality context. Primaries enablea party’s supporters in a state to decide its candi-dates for a subsequent general election. In theabsence of a tradition of direct party membership,a ‘supporter’, in most states, is defined simply andgenerously as anyone who declares, in advance, anaffiliation to that party. Thus primaries extend thepower of nomination outwards into the electorateitself. Originally introduced to formalize selectionprocedures and to weaken the control of corruptparty bosses, primaries are now well-entrenched(Ware, 2002). Yet as with membership ballots, pri-maries seem to be a mixed blessing. They takecontrol over selection away from the party itself,reducing its cohesion and giving an advantage tobetter-known and well-financed candidates.

An increasing number of countries operate amixed electoral system, in which electors vote forboth a party list and a district candidate. These cir-cumstances complicate the party’s task of selectingcandidates. It must both produce a national orregional list and also ensure local selection of con-stituency nominees. In this situation, individualpoliticians also face a choice: should they seek elec-tion via the party list or through a constituency?Many senior figures ensure they appear on bothparts of the ballot, using a high position on theparty’s list as insurance against restlessness in theirhome district. In the 1998 election in Germany,most constituency candidates also appeared ontheir party’s list (Roberts, 2002).

Leaders

The method of selecting the party leader and pres-idential candidate merits special attention. Just as

many parties now afford their ordinary members agreater voice in candidate selection, so too has theprocedure for selecting the party leader becomebroader. As Mair (1994) notes, ‘more and moreparties now seem willing to allow the ordinarymembers a voice in the selection of party leaders’,perhaps to compensate members for theirdeclining role in campaigns increasingly driventhrough the media. Yet whether this wider selec-tion process yields better results remains distinctlydebatable.

The most common way to choose the leader isto use a special party congress or convention (Box11.4). American parties have long selected theirpresidential candidates through party conventionsbut these meetings are no longer the effective siteof decision. The real choice is made by voters inthe primaries, with the convention itself trans-formed since the 1970s into a media event for theparty and its anointed candidate.

Canada traditionally provided a more typicalexample of the party congress. From the 1920s,leaders of the major parties were selected (andcould be deselected) though special conventions.

190 LINKING SOCIETY AND GOVERNMENT

Selection Countries in Total number body which most of parties using

major parties this method use this (based on 16 method democracies)

Party Finland, Norway, 37congress or Sweden, USAconvention

Rank-and-file Belgium 19members

Members of the The Netherlands,parliamentary New Zealand 17party

Party committee Italy 8

Source: Adapted from Hazan (2002, p. 124).

B OX 1 1 . 4

Selection of party leaders inestablished democracies

Delegates were either elected by constituencyassociations, appointed ex officio, or drawn fromparty groups such as those for young people. Butthese conventions became large and expensiveaffairs, with some candidates exerting unforeseeninfluence over delegate selection. As a result,Canada’s parties are experimenting with othermethods, including votes by local parties, perhapsrunning alongside a convention (Young andCross, 2002).

A ballot of party members is an increasinglypopular method of selecting leaders. In Belgium,for example, all the major parties have adoptedthis approach to choosing their party president.Britain’s Liberal Party introduced a membershipvote even earlier, in 1975. In some parties, a ballotis integrated with other methods. Thus, the BritishConservatives now allow party members to choosebetween the two leading candidates as selected byvote of members of parliament.

Both conventions and membership ballotsrequire potential leaders to reach large numbers ofactivists through the media. As Canada’sConservative leader Brian Mulroney said, ‘everynight I want to be on the 11 o’clock news. It’s fineto shake delegates’ hands but you can’t win themin five minutes. You have to reinforce it’ (Davis,1998, p. 192). These methods therefore simulatethe skills the eventual winner will need in ageneral election campaign.

Selection by the parliamentary party, theremaining widely used technique of selecting theleader, involves a much narrower constituency.This format is of course the traditional method,especially for elite parties with their assemblyorigins. The device is still used in several countries,including Australia, Denmark and New Zealand.Of course, the ability of potential leaders to instilconfidence in their parliamentary peers may saylittle about their capacity to win a general electionfought on television. Even so, colleagues in theassembly will have a close knowledge of the candi-dates’ abilities; they provide an expert constituencyfor judging the capacity to lead not only the partybut also, and importantly, the country.

What members of the parliamentary party givethey can also take away. Backbenchers, leadersshould always remember, are a potential executionsquad. Davis’s remark about Australia (1998, p.

195) applies more widely: ‘a party leader can beremoved from office during a single eveningsitting of the parliamentary caucus’.

Membership and finance

In the first decade of the twenty-first century,many parties in established democracies havefewer members but more money than at any timesince they became established. This revealingparadox tells us much not just about the changingcharacter of parties but also about their evolvingrelationship with society and state.

We look first at party membership. Table 11.1shows the marked and often dramatic declinefrom the start of the 1960s to the end of the1990s in the proportion of the populationbelonging to a political party. Even in Scandinavia,where party systems remain strong, ‘since the1970s and 1980s, membership decline has set inat an unprecedented rate’ (Sundberg, 2002, p.196). In Denmark, to take an extreme example,one in every five people belonged to a party in the1960s; by the 1990s, the ratio was one in twenty.Across the democratic world, millions of partyfoot soldiers have simply given up.

Further, the number of members playing anactive role in party affairs may be falling, andturnover increasing, with the emergence of creditcard members whose participation does not extendbeyond automatic renewal of an annual subscrip-tion.

Lacking a steady flow of young members, theaverage age of members has increased. By the late1990s, the mean age of members of Germany’sChristian Democratic Union (CDU), a majorEuropean party, had reached 54; fewer than one intwenty members was under 30.

We should locate this recent decline in a longerperspective. If statistics were available for theentire twentieth century, they would probablyshow a rise in membership in the first part of thecentury followed by a fall in the final third. Thedecline is from a peak reached, in many countries,in the 1970s. Perhaps it is the bulge in party mem-bership after the war, rather than the later decline,which requires explanation. Certainly, Putnam’scomment (2000, p. 24) about civic engagement in

POLITICAL PARTIES 191

the USA applies equally to party membershipthroughout the democratic world:

It is emphatically not my view that communitybonds in the United States have weakenedthroughout our history – or even throughoutthe last hundred years. On the contrary,American history is a story of ups and downs incivic engagement, not just downs.

The decline in membership has occurred intandem with dealignment among electors andsurely reflects similar causes. These include theweakening of social cleavages, the loosening of thebond linking trade unions and socialist parties, thedecay of local party organization in an era of televised election campaigns, and the appeal ofsocial movements rather than parties to youngergenerations.

But the consequences of falling membership aremore important than their causes. As parties’ linkswith society have weakened, so they have come todepend more on the state for their sustenance. It isthis transition from society to the state whichexplains the paradox of rising or stable revenues forparties in an era of falling membership. State

funding of national parties is now virtually uni-versal in established democracies, providing themain source of party revenue in Austria, Denmark,Finland, Norway and Sweden, and often reducingthe incentive to recruit new members. In Germany,‘parties are the self-appointed beneficiaries of extra-ordinarily generous public subsidies which inrecent years have provided on average 20–40 percent of total party revenues’ (Scarrow, 2002a, p.86). Only in the Netherlands, the UK and theUSA do membership contributions still clearlyexceed funding from the public purse.

DefinitionCartel parties are leading parties that exploittheir dominance of the political market to estab-lish rules of the game, such as public funding,which reinforce their own strong position (Katzand Mair, 1995). In politics, as in business, thedanger of such collusion is that it damages thecredibility of the participants over the longerterm.

Since most state support goes to the leadingparties, the effect is to reinforce the status quo.Thus one view of the transition to public funding

192 LINKING SOCIETY AND GOVERNMENT

Table 11.1 Falling party membership in selected European democracies, 1960–99

Total party membership as a percentage of the electorate

Beginning Beginning End of Decline of 1960s of 1980s 1990s 1960s–1990s

Austria 26 22 18 -8

Denmark 21 8 5 -16

Finland 19 13 10 -9

Belgium 8 9 7 -1

Norway 16 14 7 -9

Italy 13 10 4 -9

Netherlands 9 3 3 -6

Germany 3 4 3 0

UK 9 3 2 -7

Sources: Adapted from Mair (1994), table 1.1; Mair and van Biezen (2001), table 1; Sundberg (2002), table 7.10.

is that the state and the top levels of major partiesare tending to converge into a single, managerialsystem of rule: the party state. Governing partiesin effect authorize subsidies for themselves, aprocess captured by Katz and Mair’s idea (1995) ofa party cartel: ‘colluding parties become agents ofthe state and employ its resources to ensure theirown survival’. The largest parties become part ofthe political establishment. Seen as ‘them’ ratherthan ‘us’, parties find that their popular appealdiminishes still further.

The social base

Since most modern parties emerged from outsidethe assembly to express group interests, they natu-rally acquired a specific social base which con-tinues to influence their policies and outlook.Western European parties, in particular, retain afoundation in the social structure which givesthem not just their historic identity but also abedrock of electoral support.

These links between parties and social groupsusually develop at crucial points of conflict in acountry’s history. Such moments define new socialcleavages (that is, divisions) from which partiesemerge and which they then reinforce. Indeed,Lipset and Rokkan (1967) could claim in the1960s that party systems and social cleavages werelargely frozen in a framework of class and religionestablished forty years earlier. Box 11.5 shows thethree main cleavages from which many WesternEuropean parties have emerged. As always, theAmerican experience is distinctive.

DefinitionThe freezing hypothesis, advanced by Lipsetand Rokkan (1967), was that the party systemsand social cleavages of Western Europe in the1960s reflected those of the 1920s, so that theparty alternatives were becoming ‘older than themajorities of the national electorates’. Since the1960s, some thawing of these cleavages hasoccurred but most major parties have retained aleading position (Mair, 2001).

Since Lipset and Rokkan’s analysis, partysystems have continued to evolve. In particular, far

right parties have emerged in many WesternEuropean countries and prospered in some.France’s National Front was a significant and rela-tively long-lasting example from the final quarterof the twentieth century. In Austria, similarly, theFreedom Party won over a quarter of the vote in1999 and joined a coalition government.Switzerland’s People’s Party (SVP) achieved com-parable success in national elections in 1999 and2003. In several other countries, in Europe andbeyond, the extreme right succeeded in influ-encing the agenda of mainstream conservativeparties (Minkenberg, 2001).

Such parties blame immigrants, asylum seekersand other minorities for their supporters’ sense ofinsecurity in a changing world. Their social base isremarkably similar, drawing heavily on unedu-cated and unemployed young men. This con-stituency is particularly disillusioned withorthodox democracy and, as Lubbers et al. (2002,p. 370) report, ‘across Europe, the stronger thedissatisfaction with democracy, the larger thesupport for the extreme right’.

These protest parties can easily flourish in anunsettled society by proffering simple ‘solutions’ toirreversible changes. However, many have provedto be flash parties whose prospects are held backby inexperienced leaders with a violent or evencriminal background. Were the more extreme ofthese parties to expand to a point where theythreatened the existing order, many protest voterswould cease to vote for them.

DefinitionProtest parties exploit popular resentmentagainst the government or the political system,usually by highlighting specific ‘problems’ suchas high taxes or a permissive immigration orasylum policy.They are often short-lived flashparties which fall as quickly as they rise.Theirleaders are typically populist but inexperienced,with activists operating on the margins of thelaw.

Even though no one today would describe partysystems as frozen, the plain fact is that in the polit-ical market (as in many others) the major playershave retained their leading position. Decline doesnot imply disintegration. Sundberg’s assessment of

POLITICAL PARTIES 193

Scandinavian parties (2002, p. 210) rings true formost democracies:

Parties in Scandinavia remain the primary actorsin the political arena. To be old does not auto-matically imply that the party as a form of polit-ical organization is obsolete. The oldest carmakers in the world are more or less the sameage as the oldest parties in Scandinavia, yetnobody has questioned the capacity of thesecompanies to renew their models. The same istrue for political parties. They have developedtheir organizations and adapted their policies toa changing environment.

Party systems

To understand the political significance of parties,we must move beyond an examination of themindividually. Just as a football game consists of twoteams, so a party system consists of interaction

between several parties. Parties, like countries,copy, learn from and compete with each other,with innovations in organization, fund-raising andelection campaigning spreading across the partysystem. Similarly, legal regulation of parties – aprominent theme in the United States – applies toall parties. It is a property of the party system as awhole.

DefinitionA party system denotes the interactionbetween the significant political parties. In ademocracy, parties respond to each other’s initia-tives in competitive interplay.The party systemalso reflects legal regulation applying to allparties.

Like parties themselves, patterns of party compe-tition persist over time, forming part of the oper-ating procedures of democratic politics. We candistinguish the three overlapping formats found inBox 11.6. Note, however, that both dominant and

194 LINKING SOCIETY AND GOVERNMENT

Stage Definition Party cleavages created by this revolution

National revolution The original construction of the � Regional parties representing outlying state as a territory governed by � territories vs. parties representing the core a single central authority � region

• � Catholic parties representing traditional church � control vs. secular parties wanting more control

by the state

Industrial revolution The emergence of urban � Agrarian parties defending rural interests societies based on manufacturing � Socialist parties representing the new rather than extraction or � urban working class vs. conservative parties agriculture � representing employers’ interests

Post-industrial The transition to societies in � Green parties and a shift toward postmaterial ‘revolution’ which education and knowledge � values among many ‘socialist’ parties

replace capital and manufacturing as key resources

Sources: Adapted from Lipset and Rokkan (1967) for the national and industrial revolutions and Inglehart (1997) for the post-industrial ‘revolu-tion’. On Green parties in power, see Müller-Rommel and Poguntke (2002).

B OX 1 1 . 5

Major cleavages in the development of Western European societies andtheir impact on parties

two-party systems are now in decline and thatmultiparty systems, without a single dominantparty, have become the most common configura-tion in established democracies. The British andAmerican model of two parties engaged in a per-manent contest for power may still be influentialbut it has become rather unusual.

Dominant party systems

Here one party is a constant component of theexecutive, governing either alone or in coalition.This format was more common in the twentiethcentury than it is today. A rare contemporaryexample is the leading position acquired by theAfrican National Congress (ANC) in SouthAfrica. The ANC benefits from memories of itsopposition to apartheid, from its strong positionamong the black majority and from its use ofoffice to reward its own supporters. By contrast,the opposition parties are weak and divided, witha smaller social base.

Sweden provides an additional and unusualexample of a party that continues to dominatedespite operating in a competitive and well-regu-lated multiparty system. The Social DemocraticWorkers’ Party (Socialdemokraterna, SAP) hasformed all but a handful of governments since thewar. Between 1945 and 1998, the party averaged

44 per cent of the vote, the highest rating inWestern Europe (Bergman, 2000, p. 193).

SAP occupies a strong historical, sociologicaland ideological position in Sweden. Historically, itis the country’s oldest party, becoming the leadingparty as early as 1917. Sociologically, the party hasbenefited from support among both the workingclass and the large public sector. Ideologically, SAPfills a pivotal ideological position on the centre-left. Unlike South Africa’s ANC, the SocialDemocrats do not benefit from a link to thefounding of the nation but the party has suc-ceeded in capturing and reinforcing the popula-tion’s commitment to public welfare provision.More than most dominant parties, SAP has com-bined stable leadership with competent gover-nance.

The dominance of the Social Democrats is farfrom unlimited. SAP rarely secures a majority ofvotes and it usually lacks a parliamentary majority.Normally, it forms a single-party minority admin-istration, though often relying on parliamentarysupport from the Left (formerly Communist)Party, the other component of the socialist bloc.Just as Sweden itself is unusual in combining amultiparty system with a tradition of single-partygovernment, so its leading party has so far main-tained its dominant position in a genuinely com-petitive party system.

POLITICAL PARTIES 195

Type Definition Examples

Dominant party system One party is constantly in office, either Japan (Liberal Democrats), South governing alone or in coalition with Africa (African National Congress),other parties Sweden (Social Democrats)

Two-party system Two major parties compete to form Great Britain (Conservative and Labour),single-party governments United States* (Democratic and

Republican)

Multiparty system The assembly is composed of several Belgium, Netherlands, Scandinavia minority parties, leading to government by coalition or a minority party

* However, divided government means one party can control the presidency while the other has a majority in either or both houses of Congress.

B OX 1 1 . 6

Party systems in democracies

In the long run, dominant parties have tendedto fall victim to their own success. The verystrength of a dominant party’s position means thatfactions tend to develop within it, leading to aninward-looking perspective, a lack of concern withpolicy and even corruption. In Japan the conserva-tive LDP (Liberal Democrats), although still theleading political force, illustrates these themes.The LDP did not engage in ballot-rigging orintimidation, devices used by dominant parties insemi-democracies. But the constant tenure inoffice of the LDP between 1955 and 1993 didcreate the corrupt meshing of state and partywhich characterizes such systems. The LDP usedstate patronage to reinforce its strength, passingout resources such as campaign funds to its candi-dates through party factions.

Today, the LDP continues as a leading force butgoverning in coalition with other parties. Inpublic, some LDP leaders – including PrimeMinister Koizumi – express commitment toreform and after the 2003 election some analystsclaimed to have detected the beginnings of a tran-sition to a two-party system in which the LDPwould compete alongside the oppositionDemocratic Party of Japan.

India provides a clearer example of a diminished‘dominant’ party. From independence in 1947,Indian politics was led by the Congress Party, aparty which under Mahatma Gandhi had pro-vided the focus of resistance to British colonialrule. To maintain its leading position after inde-pendence, the party relied on a patronage pyramidof class and caste alliances to sustain a nationalorganization in a fragmented and religiouslydivided country. For two decades, Congressproved to be a successful and resilient catch-allparty, drawing support from all social groups. Butauthoritarian rule during Indira Gandhi’s State ofEmergency from 1975 to 1977 cost Congressdear. The party suffered its first defeat at anational election in 1977 and it now plays secondfiddle to the Hindu Bharatiya Janata Party (BJP).

Two-party systems

In a two-party system, two major parties of com-parable size compete for electoral support, pro-viding the framework for political competition.

The remaining parties exert little if any influenceon the formation and policies of governments(Sartori, 1976). Neither major party dominates byitself but in combination, they comprise a strongparty system.

Today, the United States is the surest example.Although American parties may seem weak byEuropean standards, a two-party system has been aconstant feature of American history. The currentparties – the Republicans and Democrats – havedominated electoral politics since 1860, assistedby the high hurdle that plurality elections set forminor parties. In addition, winning a presidentialelection is a political mountain which can only beclimbed by major parties capable of assembling abroad national coalition. The United States illus-trates how weak parties, lacking the ideologicalcohesion and mass membership found in WesternEurope, can nonetheless make a strong partysystem.

Britain is also considered to exemplify the two-party pattern. Certainly, the Conservative andLabour parties regularly alternate in office,offering clear accountability to the electorate.However third parties have gained ground; farmore so, indeed, than in the United States. In2001, the centre Liberal Democrats won 52 seatsin a parliament of 659 members, the highest pro-portion for a third party in over 50 years. The

196 LINKING SOCIETY AND GOVERNMENT

Figure 11.1 A bell-shaped distribution: parties converge at the centre

Number ofelectors

Ideological positionLeft Right

POLITICAL PARTIES 197

What are parties for? Their defining feature is to fightelections and many theorists assume, for simplicity, thatthe sole motive of parties is to maximize their vote. Buthow realistic is that assumption? Should parties be con-strued as seeking to influence government policy, ratherthan just maximizing their electoral support? This ques-tion of the fundamental goal of parties has caused con-troversy among political scientists seeking to constructtheories of party competition.

The case forThe classic study of how parties compete is AnthonyDowns’s An Economic Theory of Democracy (1957), one ofthe most influential works of political science publishedsince 1945. Downs assumes that in a political market,parties act in a rational, self-interested way. He defines aparty as ‘a team of people seeking to control the gov-erning apparatus by gaining office in a duly constitutedelection’.To maximize their control over the government,parties seek to maximize their vote, even in a multipartysystem:

The more votes a party wins, the more chance it has toenter a coalition, the more power it receives if it doesenter one and the more individuals in it hold office inthe government coalition. Hence vote-maximizing isstill the basic motive underlying the behaviour ofparties. (Downs, 1957, p. 159)

From this assumption, Downs deduces that as long aspublic opinion forms a symmetrical, bell-shaped distribu-tion around the midpoint of a left–right scale (Figure 11.1p. 196), parties in a two-party system will converge at themidpoint of the ideological spectrum. A party may startat one extreme but it will move toward the centrebecause more votes are to be won there than will be lostthrough its extreme supporters shifting to abstention.Evidence to support this prediction is legion: Bill Clintonand Tony Blair are just two examples of leaders who havewon elections by steering their parties to the middleground. Further, public funding for parties is now often inproportion to their electoral success, giving a furtherincentive for parties to maximize their vote.

The case againstCritics allege that Downs’s vote-maximizing assumptionis too extreme. It is often more accurate to see parties asoffice-seeking, a goal which requires not that votes bemaximized but just that a sufficient number be obtainedto win. In the United States, for instance, what matters is

winning the White House; the magnitude of the victory issecondary. In similar vein, the object of a party in a two-party parliamentary system is surely to win enough seatsto form the next government; again, the size of itsmajority matters far less. For such reasons, Riker (1962, p.33) distinguishes his position from Downs’s as follows:

Downs assumed that political parties seek to maximizevotes. As against this, I shall attempt to show that theyseek to maximize only up to the point of subjectivecertainty of winning.

More radically, parties may be modelled as seeking toinfluence public policy, rather than to obtain office or tomaximize votes. For example, in European countries gov-erned by coalitions (especially Scandinavia), the choice ofcoalition partners is influenced by agreements over thepolicies to be followed. More coalitions are formed fromparties with a similar ideology than would be the case ifparties were motivated solely by the short-run desire foroffice. Further, if parties followed the goal of power aboveall else, why would they ever resign from a coalition? Onereason why parties are concerned with policy is to reflectthe priorities of the members. Activists are often com-mitted to policy as well as power and, unlike companyemployees, volunteers can walk away if they dislike theirparty’s policy direction. De Swann (1973, p. 88), a leadingadvocate of this ‘policy-influencing’ position, suggeststhat

Considerations of policy are foremost in the minds ofthe actors . . . the parliamentary game is, in fact, aboutthe determination of major government policy.

AssessmentDowns’s vote-maximizing model may be simple but thatis one of the functions of a model: to simplify. In any case,his assumptions seem to yield reasonable predictions.Müller and Strøm (1999, p.305) find that ‘about half of theparties in the 10 Western European countries covered inour case studies used a strategy of vote maximization’. Bycontrast, only three parties acted as though policy influ-ence were their main objective. Even if specific counter-examples can be found, Downs’s model is still a usefulyardstick against which actual party behaviour can bemeasured.

Further reading: Downs (1957), Green and Shapiro (1994),Müller and Strøm (1999).

DEBATE

PARTIES AS VOTE-MAXIMIZERS

198 LINKING SOCIETY AND GOVERNMENT

Until 1992, Italy was a leading example of a dominant

party system.The role of the Christian Democrats (DC) in

several ways resembled the position of Japan’s LDP.The

DC was a leading player in all 47 Italian governments

between 1947 and 1992.The DC was a patronage-based,

Catholic catch-all party that derived its political strength

from serving as a bulwark against Italy’s strong commu-

nist party. Like the LDP in Japan, the DC slowly colonized

the state, with particular ministries becoming the prop-

erty of specific factions.The party used its control of the

state to reward its supporters with jobs and contracts,

with little regard for the public purse.This patronage

network spun across the country, providing a measure of

integration between the affluent North and the back-

ward South. However the DC’s rule was based on

unstable coalitions with a variety of partners; govern-

ments were short-lived and policy development con-

strained.

Between 1992 and 1994, this party system disintegrated.

Still the largest party in 1992, the Christian Democrats

had ceased to exist two years later. Its old sparring

partner, the communists, had already given up the ghost,

largely reforming as the Democratic Party of the Left

(PDS) in 1991.Why then did this party-based system col-

lapse? Donovan (1995) suggests that five catalysts initi-

ated the destruction of Europe’s oldest ruling elite:

� the collapse of communism

� the success of the new, regional Northern League in

elections in 1990

� referendums on electoral reform in the early 1990s

which revealed public hostility to the existing order

� vivid attacks by Francesco Cossiga (president 1985-92)

on the patronage power of the leading parties.

� the success of a newly assertive judiciary in exposing

political corruption.

Like many other dominant parties, the DC’s reliance on

patronage also came under pressure from the global

economy and, in Italy’s case, from EU pressure for a

genuine single market.

Although the collapse of the old system was decisive,

Italy’s new order is only now beginning to deliver more

stable government.The election of 1994, the first fought

under a new electoral system designed to reduce frag-

mentation, fell apart after seven months, the victim of tra-

ditional coalition infighting. It was replaced by an

astonishing crisis government of technocrats containing

no parliamentary representatives at all.The next election,

in 1996, did produce signs of consolidation.Two major

alliances emerged: the centre-left Olive Tree Alliance and

the more right-wing Liberty Pole. But it was not until

2001, when Berlusconi’s House of Freedom coalition won

a majority in both legislative chambers, that stable gov-

ernment seemed to become a serious possibility.

However, consolidation of the new order remains some-

what insecure as a result of Berlusconi’s own volatile tem-

perament, his controversial ownership of media

companies, a continuing debate about constitutional

reform, and the strength of the new regional parties.

Whatever the future may bring, Italy’s old mass parties

have disappeared for ever, replaced by the looser, leader-

dominated parties which are characteristic of the new

democracies founded in the 1990s.The rapid decline of

the DC showed just how vulnerable and outdated its

form of dominant party rule had become.

Population: 58m.Gross domestic product per head:

$25,000.Form of government: parliamentary,

with an indirectly elected presidentwho can play a role in governmentformation.

Legislature: the Chamber of Deputies(630 members) and the Senate (315)are elected simultaneously by popularvote for a maximum of five years. A billmust receive the positive assent of

both houses and (as in Australia) theCabinet is equally responsible to bothchambers, producing a stronglybicameral legislature.

Executive: the prime minister formallyappoints, but cannot dismiss, themembers of the large Cabinet.Coalition requirements limit the PM’schoice. An 800-strong Office of thePrime Minister provides support,which has given PMs more influenceover the machinery of government.

Judiciary: based on the civil law tradi-tion, Italy has both ordinary andadministrative judicial systems. A 15-member Constitutional Court haspowers of judicial review.

Electoral system (1994): an additionalmember system, with three-quartersof both houses elected by simplemajority and one-quarter by propor-tional representation.

Profile I TA LY

Further reading: Donovan (1995), Newell and Bull (2002),Partridge (1998).

Liberal Democrats have also progressed in localgovernment and, aided by proportional represen-tation, in the new assemblies in Scotland andWales.

Like dominant parties, the two-party formatnow appears to be in decline, kept alive only bythe ventilator of the plurality electoral method.Duverger’s law (1954) says that ‘the simplemajority single ballot [plurality] system favoursthe two-party system’. Thus the recent shift awayfrom plurality elections, as in New Zealand andSouth Africa, has damaged the prospects of two-party systems.

Even where a favourable electoral regime con-tinues, as in Canada, the two-party system hasbuckled. Traditionally dominated by the Con-servatives and Liberals, Canada’s Conservativeswere reduced to just two seats at the 1993 election.Two regional parties – the Bloc Québécois and theReform Party – have emerged as the main opposi-tion to the Liberals. Canada’s regional fragmenta-tion proved stronger than the integration offeredby the national two-party system.

Multiparty systems

In multiparty systems, the legislature comprisesseveral minority parties, resulting in coalitions or,less often, minority government by the leadingparty. Multiparty systems are a natural conse-quence of proportional representation, the domi-nant electoral system in continental Europe andScandinavia. To judge multiparty systems, a viewis therefore needed on the character and func-tioning of coalitions. How effective are they indelivering sound governance?

Answers to this question have evolved over time,largely in response to national economic perfor-mance. In the English-speaking world, coalitionswere once held to produce weak and unstable gov-ernment with confused accountability. If thingswent wrong, which party or parties in the coali-tion should be blamed? In some countries, notablyItaly, coalitions were slow to form but quick tofall, giving an appearance of continuing politicalinstability. All this is far removed from the concen-tration of power in a single-party cabinet inBritain or even the focus of responsibility in theUnited States on the White House. Herbert

Asquith, British prime minister 1908–16,expressed the English-speaking world’s suspicionof coalitions when he wrote, ‘Nothing is so belit-tling to the stature of public men, as the atmos-phere of a coalition.’

But opinions of coalitions became more positivein the 1960s as the postwar recovery of conti-nental economies took hold. In practice, coalitiongovernment did not lead to inconsistent, vacil-lating policies. In most of Scandinavia, forinstance, coalitions were composed of parties witha similar ideological persuasion. Policy was formedby consensus. Coalitions came and went butwithout threatening continuity of policy. Theclaim was that the careful, cautious governanceinduced by multiparty coalitions was well-suitedto complex societies with strong social divisions.Coalitions were seen as the anvil of democracy, aforge for manufacturing consensus.

However, the link between coalitions and weakgovernment resurfaced in the 1990s. This reinter-pretation reflected the tough 1990s agenda:cutting the government’s budget deficit, priva-tizing state-owned companies and reducingwelfare expenditure. The more parties there are ina government, it was now argued, the harder itbecame to reach agreement on such a reform pro-gramme. It was the traditional two-party systems,notably Britain and New Zealand, which pursuedthe new policies with most energy. ContinentalEurope, particularly Germany, lagged behind,leading to doubts about whether multipartysystems were sufficiently flexible to produce therapid policy changes needed to adapt to a globaleconomy. Blondel (1993), for one, argued that

the consensus mode of politics is not well-equipped to lead to long-term strategic action . . .its value appears to lie primarily in its ability tohandle deep social cleavages rather than policydevelopment.

Yet several continental countries are cautiouslybeginning to implement the reform agenda.Confronting rising unemployment, evenGermany’s Social Democratic/Green coalitionadvanced modest reform proposals, including lim-iting increases in health care spending, in 2003. Abalanced conclusion is perhaps that coalitions

POLITICAL PARTIES 199

produce continuity of policy which is helpfulwhen the economy is growing naturally, but areslower, though not necessarily less successful in thelong run, at reviving economies which have fallenon hard times.

Parties in new democracies

The distinction between established and newdemocracies is crucial to understanding thevarying significance of parties. Parties remain fun-damental to the politics and governance of consol-idated democracies but their significance issomewhat reduced in the newer post-communistand post-military democracies. In most newdemocracies, parties lack cohesion, a mass mem-bership and even an ideology. To return toOstrogorski’s distinction, they are often as much apolitical form as a political force – shells for ambi-tious politicians rather than disciplined actors inthemselves. Certainly, in some new democracies,particularly in Central Europe, parties and partysystems are becoming more settled. Even so, itwould be a Western conceit to imagine that thefuture course of party development will followthat of the established democracies.

The soft character of parties in new democraciesis illustrated in most post-communist states. Asthe national movements which initially seizedpower from the communists began to split, manynew parties appeared, representing specializedinterests such as peasants and ethnic minorities.Countries in transition from communism wereshort of many things but they did not lack forparties. However, most of these parties were of theelite type, lacking a mass membership and strongextra-parliamentary structures. Steen’s comment(1995, p. 13) about the Baltic countries – Estonia,Latvia and Lithuania – applies more generally tothe post-communist world: ‘the parties are morelike campaigning institutions before elections thanpermanent institutions propagating ideology’. Inthat respect, post-communist parties follow theAmerican rather than the Western Europeanmodel. The contrast with the tight control exertedby their communist predecessors could hardly besharper.

The failure of parties to penetrate post-commu-

nist societies reflected continuing suspicion of pol-itics and politicians among the population. Itremained difficult to enthuse electors who hadbeen denied a political voice during, and oftenbefore, the communist era. In addition, newlyfounded parties did not possess the same incen-tives to build a mass organization as had con-fronted European socialist parties a century earlier.Since voting had already extended to virtually thewhole population, there was no need for socialistparties to emerge to demand the suffrage for anexcluded working class. Parties in the CzechRepublic, Poland, Hungary and Slovakia soonobtained state subsidies, eliminating the financialneed to build a dues-paying membership. Theexistence of television provided a channel of com-munication from the leaders to the electorate,reducing the value of local activists. And the adop-tion of proportional representation enabled smallparties to survive and even prosper, giving littlereason for parties to combine into the largergroupings which could develop a coherent pro-gramme and present it to the electorate.

DefinitionSuccessor party is the term used to describe thenew parties which formed in post-communiststates from the old ruling communist parties.TheSocial Democrats in the Czech Republic andPoland are examples. Several successor partiesachieved a remarkable electoral recovery in themid-1990s. Ideologically, successor partiesaccepted the end of communism but many didquestion the speed of transition to a full marketeconomy.

Kitschelt et al. (1999, p. 396) even suggest thatthere may be advantages in the political flexibilityof soft party structures:

The absence of sunk costs in large membershiporganizations enables Eastern European democ-racies to enjoy the ‘advantages of backwardness’and frees its politicians from devoting their ener-gies to fighting armies of party functionaries.

Ironically, the organizational weakness of thenew parties allowed the communist successorparties to stage an unpredicted if temporary come-

200 LINKING SOCIETY AND GOVERNMENT

back in several countries. In the early to mid-1990s, successor parties won presidential or parlia-mentary elections in Poland, Hungary and Russia.The living corpse of communism rose from itsuneasy grave, brought back to life in a popularprotest against the mass unemployment of theearly post-communist period. This politicalrebirth, surely the most astonishing in the historyof parties, owed much to the communists’ inheri-tance of money, property, facilities, people andexpertise.

Yet the success of the successor parties itselfproved to be temporary, a product of economicdecline in the transition from communism. InEastern European countries where the living stan-dards of the majority finally began to improve, thedesire for a return to ‘the good old days’ weak-ened, especially among younger generations. In1997, the Polish successor party was itself votedout of office. In any event, traditional communistrule has gone, born and died in the twentiethcentury.

Particularly in those countries most influencedby the European Union, there are signs ofincreasing party stability. In the post-communistworld, most parties now fall into one of the sevenfamilies shown in Box 11.7. Bear in mind,however, that these groupings are broad and someparties, including many in the Balkans, are drivenmore by leadership and patronage than by theirideological colour. Post-communist party systemshave also become less fragmented as they havematured, often aided by an increase in the votingthreshold needed to win representation in parlia-ment. Typically, only one or two new partiesachieved seats in the assembly in the second andthird post-communist elections.

In any case, we must take care not to judge post-communist parties by an idealized image of partystability in established democracies. Lewis’s con-clusion (2003, p. 172) about parties in post-com-munist Europe serves us well: ‘many of theircharacteristics and defects are increasingly evidentin Western Europe too, and party development inCentral and Eastern Europe certainly appears tobe adequate to sustain continuing democraticdevelopment’.

Parties in authoritarian states

Parties are less significant in authoritarian regimesthan in democracies. In some cases, authoritarianrulers dispense with parties altogether, claimingthey have no value in a country which needs tounite behind its leader. Alternatively, the rulermaintains a system of personal rule behind thefaçade of a one-party state. As ever, totalitarianregimes are a special case. Especially under com-munism, parties were the central political instru-ment through which leaders sought, andsometimes achieved, total control over society. Inthis section, we will first examine the role ofparties in non-totalitarian authoritarian regimesbefore turning to the communist and fascist expe-rience.

Authoritarian parties

Some authoritarian regimes still get by with noparties at all. These are either pre-party or anti-

POLITICAL PARTIES 201

Party family Example Country

Communist Communist Party Czech successor party of Moravia and Republic

Bohemia

Social Social Democratic Czech Democratic Party Republic

Agrarian United Peasant PolandParty

Pro-market National Liberal Romanialiberal Party

Christian Christian AlbaniaDemocratic Democratic Party

Ethnic Party of the SlovakiaHungarian Coalition

Nationalist Party of Greater RomaniaRomania

Source: Adapted from Lewis (2003).

B OX 1 1 . 7

Party families in post-communistEurope

party states. Pre-party states are most commonlyfound in the Middle East: for example, SaudiArabia, Jordan and Kuwait. In these traditionalmonarchies, a ruling family dominates and partieshave yet to emerge or be permitted. Today, pre-party states are rare but internationally important,given their reserves of oil, the conflict with Israeland in some cases their terrorist links.

In the more common anti-party state, existingparties were banned when a new regime took over.For example, newly installed military rulersquickly moved to abolish parties, claiming that thenation could no longer afford the bickering andcorruption associated with them.

Many civilian authoritarian rulers have found asingle party useful as a disguise for personal ruleand as a technique for distributing patronage. Inpost-independence Africa, for example, the heroesof the nationalist struggle soon put a stop to partycompetition. With independence achieved, one-party systems were established, with the officialparty serving as the leader’s personal vehicle.

Such parties lacked presence in the countryside,were riven by ethnic and regional divisions,showed little concern with policy and became adevice by which the country’s hero-founder couldparcel out resources. True, the party was one of thefew national organizations, and proved useful inrecruiting supporters to public office, but theparty lacked cohesion, direction and organization.Indeed, when the founder-leader eventuallydeparted, his party would sometimes disappearwith him. When a coup overthrew KwameNkrumah in Ghana in 1966, his ConventionPeople’s Party also collapsed.

Leaving aside communist states, there are fewcases of authoritarian rule where the political party,rather than a dominant individual, is the truesource of power. One example, on the borderbetween authoritarianism and semi-democracy, isSingapore’s People’s Action Party (PAP). This partyhas ruled the island since it achieved independencefrom Britain in 1959, winning 80 out of 83 seatsin the 2001 election. PAP uses the characteristictechniques of a dominant party. Its attitude to theelectorate blends threats against hostile localitieswith benevolence to districts which dutifullyreturn PAP candidates. The party manipulatespublic opinion by limiting media freedom. It runs

smear campaigns against opposition figures,harassing them with lawsuits and tax fraud charges.

But unlike many other authoritarian parties,PAP also provides effective governance. Singaporeis a classic example of the Asian developmentalstate in which tight but paternalistic politicalcontrol has served as a foundation on which tobuild a modern trading and export economy. As incommunist China, a national commitment to eco-nomic development gives the dominant party con-tinued faith in its right to rule.

Ruling parties in authoritarian systems faceincreasing problems in maintaining their position.International pressures for genuine democratiza-tion are growing and election observers are castinga more critical eye over traditional techniques forfiddling the electoral books. Just as important,international pressures to reduce a bloated publicsector, stimulated by such bodies as the WorldBank, are reducing the supply of patronage onwhich dominant parties depend to buy politicalsupport.

Even ruling parties which have always been sym-pathetic to the private sector – such as Singapore’sPAP – may eventually fall victim to their own eco-nomic success. Control by a single party sitsuneasily alongside the requirements of anadvanced economy. And as Lipset (1960) arguedlong ago, a wealthy and educated citizenry willeventually become frustrated with the tightcontrol exercised by a single ruling party.

Communist parties

Communist states provide the limiting case of thecontrol that a single ruling party can achieve in anon-democratic environment. The mass party wasthe key innovation of the twentieth century andthe device reached its zenith in the dominance ofruling communist parties over state and society,notably in the Soviet Union (1917–91).

The monopoly position of ruling communistparties was justified by Lenin’s notion of the ‘van-guard party’: the idea that only the party couldfully understood the long-term interests of theworkingclass. Armed with this doctrine, theCommunist Party of the Soviet Union sought toimplement its vision of a total transformation ofsociety following the 1917 revolution.

202 LINKING SOCIETY AND GOVERNMENT

In this as in other communist party states, theparty exerted its control through a variety of means.The party acted as a watchdog over society, vettedappointments to all positions of responsibility, con-trolled the media and carried out agitation and pro-paganda activities. Initially, hardly any independentgroups were permitted and civil society faded,leading to an absence of virtually any associationsstanding between the family and the state.

As with all totalitarian regimes, coercion, andespecially the threat of it, helped communistparties to maintain their control. The secret policewere the main instrument of repression, as withthe feared NKVD (later KGB) in the SovietUnion. The KGB used a vast network of informersto identify, and then eliminate, ‘class enemies’ and‘poisonous weeds’. Where this system of socialcontrol operated fully, it provided the most sys-tematic penetration of society that any politicalparty anywhere has ever achieved.

DefinitionDemocratic centralism was a key feature ofcommunist party organization. It was based ontwo principles.The first was that lower levelsmust accept decisions made by higher levels (thecentralism dimension).The second was that eachlevel was elected by the one immediatelybeneath, thus forming a pyramid of indirect elec-tion (the democratic dimension). But only oneperson would be nominated for each election;and in reality this candidate was chosen fromabove. So ‘democratic centralism’ was centralismwithout democracy.

Reforms notwithstanding, the ChineseCommunist Party (CCP) still illustrates the elabo-rate internal hierarchy of ruling communistparties, with its base in democratic centralism(Figure 11.2). At the party’s base stand 3.5mprimary party organizations, found not just inlocal areas such as villages but also in factories andmilitary units. At the top, at least in theory, is thesovereign National Party Congress, a body ofaround 2,000 people which meets infrequentlyand for short periods. In practice, the Congressdelegates authority: to its Central Committee and,through that body, to the Politburo (literally,political bureau) and its Standing Committee.

This intricate pyramid allows the half-dozen or somen on the Standing Committee to exert enor-mous influence over the political direction of themost populous country on earth – in itself, aremarkable political achievement.

The party is massive. In 2000, its membershipstood at 63m, comparable to the total populationof the United Kingdom. And unlike most partiesin the democratic world, the CCP’s membership isstill growing. Indeed, it has the potential toincrease further since the current members are justan elite fraction of the total population andwomen remain heavily under-represented.

What then accounts for the party’s continuingappeal? The answer can hardly be ideology for theparty’s communist allegiance is now nominal.Indeed, Saich (2004, p. 95) reports that on histravels he has met ‘party members who areMaoists, Stalinists, Friedmanites, Shamans, under-ground Christians, anarchists, and social andliberal democrats’. At local level, many membersof the Communist Party are successful capitalistswith businesses to run.

Party members are united by ambition ratherthan ideology. The CCP remains the onlyacademy for the ambitious. Members are preparedto undergo a searching entry procedure, and thecontinuing obligation to participate in dull partytasks, in order to secure access to resources such aseducation, travel, information, expenses and auto-mobiles. Above all, members can exploitincreasing opportunities – whether legal, corruptor both – to acquire wealth. Those who are partic-ularly determined will seek out a patron to helpthem ascend the party hierarchy. Within theparty’s formal structure, it is personal patron–client networks, not political ideologies or policydifferences, that provide the guiding force.

In contrast to many other authoritarian parties,the CCP has sought since the 1990s to strengthenits control over its own members, the governmentand even public debate. Tighter political control isa response to economic liberalization. Party leadersfear that that the mass unemployment, ruraldepopulation and increased corruption associatedwith economic change will diminish the party’sstanding and eventually limit its politicalmonopoly. Comparative experience suggests theymay well be right.

POLITICAL PARTIES 203

Fascist parties

Showing the limits of the concept of totalitarianrule, fascist parties never achieved communistlevels of control of either state or society. In largemeasure, this weakness reflected the cult of the

leader (Führer-prinzip) within fascism. The fascistparty served its leader, not the other way round.For example, the party could not be a site ofpolicy debate because policy was the leader’s pre-serve. Where communist ideology venerated theparty as the foundation of progress, fascistthinking was oriented to the leader as the personi-fication of the Italian state or the German people.

The German experience confirms the tangentialrole of the party within fascist regimes. The rise topower of the Nationalsozialistiche DeutscheArbeiterpartei (NSDAP, Nazi Party) owed muchto its leader’s opportunistic political skill. Yet onceHitler achieved power in 1933, the party becameincreasingly marginal. Soldiers, as servants of thestate, were even prohibited from joining the party.

The regime focused more on personal loyalty toHitler, downgrading the party. State officials weremade to swear an oath of loyalty to Hitler himself,not to his party. The feared SS (Schutzstaffel or‘protection units’) may have begun as the NaziParty’s security force but they became Hitler’s ownhenchmen. The NSDAP became primarily a pro-paganda organization and even this function waseventually taken over by the government’s ownministry. As Mommsen (1997, p. 170) writes:

Many foreign observers, impressed by well-orga-nized mass rallies, took it for granted that theNSDAP exerted an authoritative influence ondecision-making; yet, in fact, it was increasinglycondemned to political sterility.

Thus the significance of political parties in total-itarian states is not an issue which can beaddressed without acknowledging the funda-mental contrasts between communist and fascistregimes. Under communism, the political partyreached its twentieth-century apogee, with Lenin’snotion of the vanguard party rationalizing com-munist command of both state and society. Bycontrast, the ruling fascist party occupied a mar-ginal position, becoming the vehicle of a supremeleader in a system that was both personal andstate-centred.

204 LINKING SOCIETY AND GOVERNMENT

Central PartyOffices (pro-

paganda,organization

etc.)

CountyDisciplineInspection

Commission

CentralDisciplineInspection

Commission

ProvincialDisciplineInspection

Commission

Figure 11.2 Organization of the Chinese CommunistParty

CentralAdvisory

Commission

ProvincialParty

Committee

County Party

Committee

MilitaryAffairs

Commission

GeneralSecretary

Secretariat Politburo

PolitburoStanding

Committee

NationalParty Congress

Provincial Party Congress

Counry Party Congress

Primary PartyOrganization

(township,army unit, etc.)

electsappoints

CentralCommittee

approves

Key reading

Next step: Webb et al. (2002) is an informa-tive collection examining the condition of partiesin a range of established democracies.

Gunther et al. (2002), a more thematic collection,helpfully supplements Webb et al. (2002). Classicworks on parties (and there are many) includeLipset and Rokkan (1967) on the social base,revisited by Karvonen and Kuhnle (2001); Sartori(1976) on party systems; Kircheimer (1966) on

the catch-all party; Bartolini and Mair (1990) onthe stabilization of European electorates; and Katzand Mair (1995) on cartel parties. Wolinetz(1997) is a reader gathering together work in theseareas. On more specific themes, Hazan (2002)examines candidate selection while Davis (1998)discusses leadership selection. Ignazi (2003) con-siders extreme right parties in Western Europe.Lewis (2003) and Kitschelt et al. (1999) coverpost-communist parties, van Biezen (2003) looksat new democracies and Brooker (2000) surveysparties in non-democratic regimes.

POLITICAL PARTIES 205

This page intentionally left blank

We examine here the key institutions of nationalgovernment: legislatures, the executive and thebureaucracy. In many political systems, these are thecore structures through which policy is shaped andpower is exercised. We also analyse territorial poli-tics, reviewing the relationships between central,provincial and local government. In the final

chapter we move from the structures of governmentto a direct focus on the policies that governmentpursue. But we begin with a chapter on constitu-tions and the legal framework. Essential foundationsof a liberal democracy, constitutions and the judi-ciary are growing in political significance as moregovernments come to operate under the rule of law.

Part IV

GOVERNMENT AND POLICY

This page intentionally left blank

Constitutions 210Judicial review and constitutional courts 212Judicial activism, independence and

recruitment 216Administrative law 221Law in new democracies 223Law in authoritarian states 224International law 225Key reading 227

Law and politics are closely linked. The develop-ment of liberal democracy has been an attempt

to ensnare absolute rulers in the threads of legalrestraint. In the words of A. V. Dicey (1885 p.27), the nineteenth-century British jurist, theobject was to substitute ‘a government of laws’ fora ‘government of men’. Constitutional rule,affording protection for individual rights and ameans of resolving disputes between citizens andstate, is the major accomplishment of liberal poli-tics. From this perspective, law is not so much sep-arate from politics as an achievement of it.

The academic study of politics began as abranch of law and belatedly these old friends arenow renewing their acquaintance. Four factorsseem to be involved in this rebirth of interest inthe legal dimension of politics:

� The late twentieth century witnessed an explo-sion of constitution-making among postauthor-itarian states, with 85 constitutions introducedbetween 1989 and 1999 (Derbyshire andDerbyshire, 1999).

� Several established democracies – includingBelgium, Canada, the Netherlands, and Sweden– adopted new constitutions in the final portionof the twentieth century. The European Unionfollowed with a draft constitution in 2003.

� Stimulated by the legal character of theEuropean Union and judicial activism in theUSA, judges have become more willing to stepinto the political arena.

� The expanding body of international lawincreasingly impinges on domestic politics, withjudges called on to arbitrate between the con-flicting claims of supranational and nationallaw.

For such reasons, Shapiro and Stone Sweet (2002,p. 1) suggest that ‘it will be increasingly difficultfor scholars who do research on government, orgovernance, to avoid encountering a great deal oflaw and courts’.

DefinitionConstitutional rule (or constitutionalism) isthe combination of habits, practices and valueswhich underpin government by law.Constitutions are not self-implementing butdepend on the support of the political elite toprovide an effective framework for the exerciseof power.When constitutionalism is absent, aconstitution becomes a mere parchment.

To understand the political role of constitutionswe must first introduce the idea of constitutionalrule or culture (also called constitutionalism).Constitutional rule is broader than the constitu-tion itself. It refers to a political environment inwhich the equal rights of individuals are not juststated but also respected. Specifically, rights can bedefended through the courts, thus converting adusty document into a political reality.

In theory, however, constitutional rule canoperate without a foundation in a specific consti-tution: indeed, the United Kingdom is such acase. There, the notion of parliamentary sover-eignty historically excluded the possibility ofacknowledging a higher, constitutional power.Even so, the freedoms of most citizens are reason-ably well-protected by a tradition of rule by law.

Conversely, the mere possession of a writtenconstitution does not guarantee constitutionalrule. Parchments depend on people for their

Chapter 12

Constitutions and the legal framework

209

implementation. Communist states frequentlyadopted new constitutions but these statementslargely reflected the current objectives of theruling party. To take a contrasting example, theAmerican constitution did not alter on September11, 2001, but the rights of many immigrants whofound themselves imprisoned for several monthswithout charge suddenly became less certain. Aswith most other countries facing external threats,constitutionalism temporarily took second placeto national security.

Constitutions

We can look at constitutions in two ways. Thefirst reflects their historic role as regulator of thestate’s power over its citizens. For the Austrianliberal Friedrich Hayek (1960), a constitution wasnothing but a device for limiting the power ofgovernment. In similar vein, Friedrich (1937)defined a constitution as ‘a system of effective, reg-ularized restraints upon government action’. Fromthis perspective, the key feature of a constitution isits statement of individual rights, particularlythose held against the state.

Certainly a bill of rights now forms part ofnearly all written constitutions (Box 12.1). Thefirst Bill of Rights comprised the ten amendmentsquickly appended to the American constitution in1791, covering such liberties as freedom of reli-gion, speech, the press, assembly and the ‘right ofthe people to keep and bear arms’.

Recent constitutions tend to be more ambitiousin their statements of rights, often imposing dutieson rulers such as fulfilling citizens’ social rights toemployment and medical care. Several post-com-munist constitutions have extended the listfurther, to include the right to a healthy environ-ment. Many scholars argue that the greater theemphasis on rights in a constitution, the moreactive the judiciary will need to become as it testsgovernment policy against such explicit standards(Shapiro and Stone Sweet, 2002, p. 177).

The second and somewhat neglected role of con-stitutions is as power maps, defining the structureof government. Constitutions articulate the path-ways of power, describing the procedures formaking laws and reaching decisions. As Sartori

(1994, p. 198) wisely observes, the defining featureof a constitution is this provision of a frame ofgovernment. A constitution without a declarationof rights is still a constitution, whereas a documentwithout a power map is not a constitution. A con-stitution is therefore a form of political engi-neering, to be judged like any other constructionby how well it survives the test of time. From thisperspective, the American version, still standingfirm after more than 200 years, is a triumph.

DefinitionA constitution sets out the formal structure ofthe state, specifying the powers and institutionsof central government, and its balance withother levels. In addition, constitutions expressthe rights of citizens and in so doing create limitson and duties for the government.

A traditional distinction contrasts written andunwritten constitutions. Yet no constitution iswholly unwritten; even the ‘unwritten’ British andNew Zealand constitutions contain much statuteand common law. A contrast between codified anduncodified systems is more useful. Most constitu-tions are codified; they are set out in detail withina single document. The flexible constitutions ofBritain and New Zealand are unusual in that theyare not formalized in this way. Sweden falls in

210 GOVERNMENT AND POLICY

� A preamble seeks popular support for the docu-ment with a stirring declaration of principle and,sometimes, a statement of the goals of the state.

� An organizational section sets out the powers ofthe institutions of government.

� A bill of rights covers individual and perhapsgroup rights, including access to legal redress, andthereby sets limits on government.

� Procedures for amendment define the rules forrevising the constitution.

Further reading: Duchacek (1991), Maddex (2000).

B OX 1 2 . 1

The standard format of constitutions

between the codified and uncodified categories: itsconstitution comprises four separate Acts passed atdifferent times.

Amendment

Procedures for amendment are an important com-ponent of the constitutional architecture. Mostconstitutions are rigid (or ‘entrenched’), thus ren-dering them more acceptable to the various inter-ests involved in their construction. An entrenchedconstitution offers the general benefit, muchprized by liberals, of predictability for thosesubject to it. A rigid framework also limits thedamage should political opponents obtain power,for unless they can clear the amendment hurdlethey too must abide by the values embedded inthe settlement.

DefinitionsFlexible constitutions can be amended in thesame way as ordinary legislation is passed.Britain is one of the few examples. Rigid consti-tutions are entrenched, containing a specialamendment procedure. Rigid constitutions areusually codified – set out in a single document –as in the United States.

Flexible constitutions, though rare, do offer theadvantage of ready adaptability. In New Zealand,this flexibility permitted a recasting of thecountry’s electoral system and government admin-istration in the 1980s and 1990s. In most othercountries such radical changes would haverequired constitutional amendment.

How are constitutions entrenched? This isachieved by setting a higher level and wider spreadof support than is required for the passage of ordi-nary bills. Typically, amendment requires both atwo-thirds majority in each house of parliamentand additional endorsement by the states in a fed-eration or, in unitary countries, by the peoplethrough a referendum (Box 12.2).

The amendment procedure offers clues as to thestatus of the constitution in relation to the legisla-ture. When modifications cannot be approved bythe legislature alone, the constitution standssupreme over parliament (Wheare, 1951). InAustralia, for example, amendments must be

endorsed not just by the national legislature butalso by a referendum achieving a double majority– in most states and also in the country as awhole. So far, only eight of 42 questions put to thevoters have passed this test.

In a few countries, however, special majoritieswithin the legislature alone are authorized toamend the constitution. In such a situation, thestatus of the constitution is somewhat reduced.Germany is a partial example: amendments simplyrequire a two-thirds majority in both houses. At thesame time, however, the nucleus of the Basic Law –

CONSTITUTIONS AND THE LEGAL FRAMEWORK 211

Country Amendments require the approval of

Australia both houses of parliament, then a referendum achieving majority support (a) overall and (b) in a majority of states

Canada both houses of parliament and two-thirds of the states containing at least half the population

Germany a two-thirds majority in both houses of parliament1

Spain a two-thirds majority in both houses of parliament and, if demanded by a tenth of either house, a referendum achieving majority support2

Sweden majority vote by two successive sessions of parliament with an intervening election3

USA a two-thirds majority in both houses of Congress and approval by three-quarters of the states4

Notes:1 The federal, social and democratic character of the German state,

and the rights of individuals within it, cannot be amended.2 ‘Fundamental’ amendments to the Spanish constitution must be

followed by an election, ratification by the new parliament and areferendum.

3 Instrument of Government and Riksdag Act (two of the four actscomprising the constitution).

4 An alternative method, based on a special convention called bythe states and by Congress, has not been used.

B OX 1 2 . 2

Entrenching the constitution:some examples

which sets out core rights – is accorded the ultimateentrenchment. It cannot be amended at all.

Although rigid constitutions may appear to beincapable of coping with change, in practice theyare more adaptable than they seem. Judicial inter-pretation is the key here. The American SupremeCourt, for example, has become skilled atadjusting an old document to fit new times, rein-terpreting a constitution designed in the eigh-teenth century for the fresh challenges of latereras. Thus one contrast between rigid and flexibleconstitutions is that in the former the judiciarymanages evolution while in the latter politicianstake the lead. In other words, rigid constitutionsexpress a more liberal interpretation of democracy.

Origins

Constitutions are a deliberate creation, designedand built by political architects. As the Englishpolitical theorist John Stuart Mill (1861) wrote,constitutions ‘are the work of men . . . Men didnot wake up on a summer morning and find themsprung up.’ How then do constitutions come intobeing? Under what circumstances do societies setabout redesigning their political order?

New constitutions typically form part of a freshstart after a period of disruption. Such circum-stances include regime change (for example thecollapse of communism), reconstruction afterdefeat in war (Japan after 1945) and the achieve-ment of independence (much of Africa in the1950s and 1960s). With the collapse of commu-nist and military regimes, the 1980s and 1990swere busy times for constitution-makers: 17 newones were introduced in Eastern Europe between1991 and 1995, and over 30 in Africa during the1990s (Vereshchetin, 1996).

Yet most constitutions experience a difficultbirth. Often, they are compromises between polit-ical actors who have been in conflict and who con-tinue to distrust each other. An example of suchnegotiated settlements is South Africa’s post-apartheid constitution of 1996. Against anunpromising backdrop of near slavery, this consti-tution achieved an accommodation betweenleaders of the white and black communities.

As vehicles of compromise, most constitutionsare vague, contradictory and ambiguous. They are

fudges and truces, wrapped in fine words (Weaverand Rockman, 1993). As a rule, drafters are moreconcerned with a short-term political fix than withestablishing a resilient structure of government. Inprinciple, everyone agrees with AlexanderHamilton (1788b, p. 439) that constitutionsshould ‘seek merely to regulate the general politicalinterests of the nation’; in practice, they are lengthydocuments reflecting an incomplete accommoda-tion between suspicious partners. Some topics areover-elaborated but other issues are left unresolved.The lauded American constitution of 1787,although shorter than most, is no exception. Finer’sdescription (1997, p. 1495) makes the point:

The constitution was a thing of wrangles andcompromises. In its completed state, it was a setof incongruous proposals cobbled together. Andfurthermore, that is what many of its framersthought.

The danger, then, is that a new constitution willnot grant the new rulers sufficient authority. TheAmerican constitution, for instance, divides powerto the point where its critics allege that the ‘govern-ment’, and specifically the president, is virtuallyincapable of governing (Cutler, 1980). Far from pro-viding a settled formula for rule, some constitutionsdo no more than pass the parcel of unresolved polit-ical problems to later generations. Too often, polit-ical distrust means the new government is hemmedin with restrictions, limiting its effectiveness.

The Italian constitution of 1948 illustrates thispoint. The hallmark of the Italian constitution isgarantismo, meaning that all political forces areguaranteed a stake in the political system. Thus thedocument establishes a strong bicameral assemblyand extensive regional autonomy. These checks onpower were intended to prevent a recurrence offascist dictatorship but in practice garantismo con-tributed to ineffective government and ultimatelyto the collapse of the First Republic in the 1990s.

Judicial review and constitutionalcourts

Constitutions are no more self-implementing thanthey are self-made. Some institution must be

212 GOVERNMENT AND POLICY

CONSTITUTIONS AND THE LEGAL FRAMEWORK 213

South Africa’s transformation from a militarized statebased on apartheid to a more constitutional orderbased on democracy was one of the most remarkablepolitical transitions of the late twentieth century.Thefact that the transition was largely peaceful, defyingnumerous predictions of an inevitable bloodbath, wasthe most astonishing fact of all.

Since white settlers came to South Africa in 1652, theyhad controlled the country by exploiting black labour.After 1945, the system of apartheid (apartness) institu-tionalized these racial divisions. Apartheid definedthree races – white, coloured and black – and out-lawed marriage between them. Apartheid’s survivalinto the 1990s showed that governments based onbrute power can last a long time.Yet change waseventually induced, by three main factors: the collapseof communism which destroyed the regime’sbogeyman; the imposition of sanctions by the EU andthe United States; and black opposition which beganto encompass armed resistance.

As so often, initial reforms merely stimulated demandsfor more and faster change. In 1990, ANC leaderNelson Mandela was released from prison after 26years, symbolizing recognition by the white rulers thatthe time had come to negotiate their own downfall.Four years later, Mandela became president of a gov-ernment of national unity, including the white-ledNational Party (NP), after the ANC won the first mul-tiracial elections.

In 1996, agreement was reached on a new constitu-tion that took full effect in 1999.The constitution tooktwo years to negotiate, reflecting hard bargainingbetween the ANC and the NP.The final 109-page doc-ument reflected the interests of the dominant ANC. Itincluded a bill of rights covering education, housing,water, food, security and human dignity.The NPexpressed general support despite reservations thatled to its withdrawal from government.

It remains to be seen how South Africa’s rainbownation will be able to reconcile constitutional liberaldemocracy with the political dominance of the ANC.The ANC’s strength in parliament is such that it couldby itself virtually achieve the two-thirds majorityrequired to amend the constitution, thus providing anexception to the generalization that fresh constitu-tions normally deny sufficient authority to a new gov-ernment. Further, South Africa remains a minefield ofunresolved social problems: crime, inequality, poverty,unemployment and Aids.Yet the country’s politics,more than most, should be judged by what precededit. By that test the achievements of the new SouthAfrica are remarkable indeed.

Population: 43m (of whom 20 percentare estimated to be HIV positive or tohave Aids).

Gross domestic product per head:$3,167.

Composition:

Form of government: a democracywith an executive president andentrenched provinces.

Legislature: the National Assembly, thelower house, consists of 400 memberselected for a five-year term.The presi-

dent cannot dissolve the assembly.The weaker upper house, the NationalCouncil of Provinces, contains ten del-egates from each of the nineprovinces.

Executive: a president heads both thestate and the government, ruling witha cabinet.The National Assemblyelects the president after each generalelection.The president cannot nor-mally be removed while in office.

Judiciary: the 11-memberConstitutional Court decides constitu-tional matters and can strike downlegislation. Power of appointmentrests jointly with the president and aspecial commission.The Supreme

Court of Appeal is the highest courton non-constitutional matters.

Electoral system: the NationalAssembly is elected by proportionalrepresentation. Provincial legislaturesappoint members of the NationalCouncil.

Party system: the African NationalCongress (ANC; 266 seats and 66 percent of the vote in 1999) has domi-nated the post-apartheid republic. Itgoverns in coalition with the mainlyZulu Inkatha Freedom Party (34 seats).The more liberal Democratic Party (38seats) forms the official opposition.

Profile S O U T H A F R I C A

Further reading: Butler (2004), Deegan (2001), Glaser(2001).

‘White’ 13%

‘Coloured’ 9%

‘Indian’ 3%

‘Black’ 75%

found to enforce the constitution, striking downlaws and practices that offend its principles. Thisreview power has fallen to the judiciary. With acapacity to override the decisions and laws pro-duced by democratic governments, unelectedjudges occupy a unique position both in andabove politics. Indeed, India’s Supreme Court iseven empowered to override amendments to theconstitution itself.

Judicial power, furthermore, is only partlylimited by the constitution. Inevitably, judicialinterpretation varies with the temper of the times.As the American Chief Justice Hughes onceremarked, ‘we live under a constitution. But theconstitution is what the judges say it is.’ Shapiro(1964) took this pragmatic view to its extreme,suggesting that the American Supreme Court is apolitical institution like any other. He regards theCourt’s function within government as ‘basicallysimilar to the city council of Omaha, the ForestryService and the Strategic Air Command’. Baskingin a privileged position, constitutional courtsexpress a liberal conception of politics, restrictingthe power of even elected rulers. In this way, judi-cial review both stabilizes and limits democracy.

In reality, judicial power is far from unqualified.For one thing, constitutions themselves do restrictwhat judges can plausibly say about them. Justicesare only unfree masters of the document whosevalues they defend (Rousseau, 1994). Moreimportant, the impact of a court’s judgmentsdepends on its status among those who carry outits decisions. After all, a court’s only power is itswords; the purse and sword belong elsewhere. Socourts seeking to protect their own standing mustfollow a delicate course, paying heed to theclimate of opinion without being seen to panderto it. As O’Brien (1993, p. 16) concludes of theAmerican Supreme Court,

the Court’s influence on American life is at onceboth anti-democratic and counter-majoritarian.Yet that power, which flows from givingmeaning to the constitution, truly rests, in ChiefJustice White’s words, ‘solely upon the approvalof a free people’.

The function of judicial review can be allocatedin two ways (Box 12.3). The first and more tradi-

tional method is for the highest or supreme courtin the ordinary judicial system to take on the taskof ensuring the constitution is protected. Asupreme court rules on constitutional matters justas it has the final say on other questions of law.Australia, Canada, India and the USA are exam-ples of this approach. Because a supreme courtheads the judicial system, its currency is legal caseswhich bubble up from lower courts.

DefinitionSmith (1989) defines judicial review as ‘thepower of ordinary or special courts to giveauthoritative interpretation of the constitutionwhich is binding on all the parties concerned’.This covers three main areas: first, ruling onwhether specific laws are constitutional; second,resolving conflict between the state and citizensover basic liberties; and third, resolving conflictsbetween different institutions or levels of gov-ernment.

A second and more recent method is to create aspecial constitutional court, standing apart fromthe ordinary judicial system. This approach origi-nated with the Austrian constitution of 1920 andis now much favoured in continental Europe.Spain, for example, has both a ConstitutionalTribunal to arbitrate on constitutional matters anda separate Supreme Court to oversee nationalcriminal law. South Africa adopts the same divi-sion of judicial labour.

In constitutional courts, proceedings are morepolitical and less legal: such courts can typicallyjudge the constitutional validity of a law, or issuean advisory judgment on a bill, without the stim-ulus of a specific legal case. Where a supremecourt is a judicial body ruling on political cases, aconstitutional court is more akin to an upperchamber of parliament. Indeed, the Austrianinventor of constitutional courts, Hans Kelsen,argued in 1942 that they should function as a neg-ative legislator, striking down unconstitutionalbills but leaving positive legislation to parliament.

We will first examine the supreme courtapproach, using the United States – the originaland most renowned example of judicial review – asan example. America’s constitution vests judicialpower ‘in one Supreme Court, and in such inferior

214 GOVERNMENT AND POLICY

Courts as the Congress may from time to timeordain’. Constitutional issues can be raised at anypoint in the ordinary judicial system, with theSupreme Court selecting for arbitration those casesthat it regards as possessing broad significance.

However, it is important to note that the consti-tution itself does not explicitly specify the Court’srole in adjudicating constitutional disputes.Rather, this function was gradually acquired bythe justices themselves, with Marbury v. Madison(1803) proving decisive. In this case, Chief JusticeMarshall struck down part of the Judiciary Act of

1789 as unconstitutional, thereby establishing theprinciple of judicial review. At the same time, heskilfully avoided offending the administration onthe specific issue, which involved the administra-tion’s disputed appointment of Marbury to alower-level judgeship. Honour was satisfied andthe Court avoided an immediate controversy.

The Supreme Court favours the doctrine of staredecisis (stand by decisions made – that is, stick toprecedent) but does sometimes strike out in newdirections. This ‘inconsistency’ has proved to be asource of strength, enabling the Court to adaptthe constitution to changes in national mood. Forexample, after its rearguard struggle against theNew Deal in the 1930s, the Court conceded theright of the national government to regulate theeconomy.

At other times the Court has sought to leadrather than follow. The most important of theseinitiatives, under the leadership of Chief JusticeWarren in the 1950s and 1960s, concerned blackcivil rights. In its path-breaking and unanimousdecision in Brown v. Topeka (1954), the Court out-lawed racial segregation in schools, dramaticallyreversing its previous policy that ‘separate but equal’facilities for blacks fell within the constitution.

Continental Europe, both West and East,favours constitutional rather than supreme courts.In Western Europe, such courts were adopted after1945 in, for instance, West Germany, Italy andFrance (Figure 12.1). They represented a con-scious attempt to prevent a revival of dictatorship,whether of the left or the right. Half a centurylater, the German example provided an influentialmodel for post-communist countries establishingnew constitutions.

Just as the USA illustrates the supreme courtapproach, so Germany exemplifies a successful

CONSTITUTIONS AND THE LEGAL FRAMEWORK 215

Supreme court Constitutional court

Relationship to other courtsHighest court of A separate body dealing withappeal constitutional issues only

StyleMore judicial More political (issues(decides cases) judgments)

RecruitmentLegal expertise plus Political criteria more political approval important

Normal tenureUntil retirement Typically one non-renewable

term (six to nine years)

ExamplesAustralia, United Austria, GermanyStates

B OX 1 2 . 3

Institutions for judicial review:supreme courts v. constitutionalcourts

Figure 12.1 Establishing constitutional courts in Western Europe

Austria* Germany* France Portugal BelgiumItaly Spain*

1945

1948

1949

1958

1976

1978

1985

* These countries also possessed similar, but somewhat ineffective, courts in the interwar period.

Source: Adapted from Stone Sweet (2000, p. 31).

constitutional court. With political power stillunder a cloud after the Second World War, thenew Federal Constitutional Court was charged notjust with constitutional review but also with main-taining the constitutional order against groupsseeking to overthrow it. It has done just that,banning both communist and neo-Nazi parties inthe 1950s. By nurturing Germany’s postwarrepublic, the Court has ensured its high standingwith the public. As Conradt (2001, p. 245) writes,

More than any other postwar institution, theConstitutional Court has enunciated the viewthat the Federal Republic is a militant democ-racy whose democratic political parties are thechief instrument for the translation of publicopinion into public policy. The court hasbecome a legitimate component of the politicalsystem, and its decisions have been accepted andcomplied with by both winners and losers.

The Court’s decisions have also impinged onother areas, including abortion, university reformand party funding. Between 1951 and 1990, theCourt judged that 198 federal laws (nearly 5 percent of the total) contravened the constitution. Inframing new bills, German policy-makers antici-pate the likely reaction of the Court; partly for thisreason, Kommers (1994) argues that its influenceis fully the equal of America’s Supreme Court.

It would be remiss to conclude this sectionwithout considering the European Court of Justice(ECJ). Just as the Federal Constitutional Court inGermany helped to shape the development of anew republic, so too has the ECJ contributed tothe strengthening of the European Union. Buildingon the German tradition of constitutional courts,the ECJ played a central role between the 1960sand 1980s in expanding the EU’s legal order. Thecumulative impact of its decisions amounted towhat Weiler (1994) terms a ‘quiet revolution’ inconverting the Treaty of Rome into what is, insome respects, a constitution for Europe.

The Court’s formal purpose is to settle conflictsbetween the member states of the union, betweenthe organs of the community and between thestates and the organs (Nugent, 2003). It consistsof one judge appointed from each member statefor a renewable six-year term, with broad experi-

ence more important than judicial expertise. Casescan be brought before it by member states or theother institutions of the Union. Like national con-stitutional courts, the ECJ can issue opinions aswell as decide cases.

Like many other courts, the ECJ’s decisions inits early decades consistently strengthened theauthority of central institutions. It insisted thatEuropean law applies directly to individuals, takesprecedence over national courts and must beenforced by these courts. Even though somemembers began in the 1990s to question both theCourt’s procedures and further expansion of itsauthority, the ECJ remains a major example ofjudicial influence.

Judicial activism, independence andrecruitment

Throughout the Western world, judicial interven-tion in public policy has grown, marking the shiftfrom government to governance. Judges havebecome more willing to enter political arenas thatwould have once been left to elected politiciansand national parliaments. For instance:

� The Australian High Court under Sir AnthonyMason (Chief Justice, 1987–95) boldly uncov-ered implied rights in the constitution whichhad remained undetected by its predecessors.

� Even though the Dutch constitution explicitlyexcludes judicial review, the Supreme Court hasproduced important case law on issues whereparliament was unable to legislate, notably inauthorizing euthanasia.

� The Israeli Supreme Court has addressed con-flicts been secular and orthodox Jews that main-stream politics has been unable to resolve(Hirschl, 2002).

Activism

What explanation can we offer for this significantjudicialization of politics? The key seems to lie inthe changing nature of politics, rather than devel-opments within the judiciary itself. First, theincreasing reliance on regulation as a mode of gov-erning has encouraged intervention by courts. A

216 GOVERNMENT AND POLICY

political decision to deny gay partners the samerights as married couples is open to judicial chal-lenge in a way that a decision to go to war or raisetaxes is not. Indeed, Majone (1996, p. 290) arguesthat ‘the progressive judicialization of regulatoryproceedings’ is how regulation gains legitimacy inthe absence of mechanisms of democratic control.

DefinitionJudicial activism refers to the willingness ofjudges to venture beyond narrow legal decisionsso as to influence public policy. It is the oppositeof judicial restraint, a more conservative philos-ophy which maintains that judges should simplyapply the law (including the constitution), irre-spective of policy implications and the judges’own values.The two phrases developed in con-nection with the American Supreme Court butpossess wider applicability in an era of judicialpolitics.

Second, the decay of left-wing ideology hasenlarged the judiciary’s scope. Socialists were sus-picious of judges, believing them to be unelecteddefenders of property specifically and the statusquo as a whole. So in Sweden, for instance, somedecline in the strength of social democratic princi-ples has given more room to a traditionallyrestrained judiciary (Holmstrom, 1995).

Third, international conventions have givenjudges an extra lever they can use to break free fromtheir traditional shackles. Documents such as theUnited Nations Universal Declaration of HumanRights (1948) and the European Convention onHuman Rights (1950) have given judges a quasi-judicial foundation on which to construct whatwould once have been viewed as excessively politicalstatements. Within the European Union, of course,national courts are obliged to give precedence toEuropean law, a factor which gives them a stickwith which to beat national governments. It is thispriority of European over national law that has

CONSTITUTIONS AND THE LEGAL FRAMEWORK 217

1952 Court of Justice established as part of theEuropean Coal and Steel Community.

1957 Treaty of Rome: the Court’s jurisdictionextends to the new European EconomicCommunity and Euratom treaties.

1963 Van Gend en Loos: European laws applydirectly to individuals, creating rights andobligations that national courts mustimplement.

1964 Costa v. Enel: European law takes priorityover national law.

1979 Cassis de Dijon: a product sold lawfully inone member state must be accepted forsale in other member states (‘mutualrecognition’).

1987 Foto-Frost: national courts cannot invali-date EU measures but must refer theirdoubts to the ECJ for resolution.

1988 Court of First Instance established toreduce the ECJ’s workload and to improveits scrutiny of factual matters.

1992 Francovich and Bonifaci v. Italy: whenmember states breach EU law, they mustcompensate those affected.

1992 Maastricht Treaty clarifies the ECJ’s juris-diction, e.g. to include treaty provisions forcloser cooperation between memberstates but to exclude the Common Foreignand Security Policy.

2003 Publication of draft constitution forEurope.

Further reading: Dehousse (1998), de Burea and Wieler(2002).

50 YEARS OF

T H E E U R O P E A N C O U R T O F J U S T I C E ( E C J )

enabled the Dutch court to overcome the ban onjudicial review in its national constitution.

Throughout the democratic world, the expan-sion of judicial authority has become self-rein-forcing. Stone Sweet (2000, p. 55) makes thepoint: ‘as constitutional law expands to more andmore policy areas, and as it becomes “thicker” ineach domain, so do the grounds for judicializeddebate. The process tends to reinforce itself.’

Sensing the growing confidence of judges inaddressing broader political issues, interest groups,rights-conscious citizens and even political partieshave become more willing to continue their strug-gles in the judicial arena. With the status of thejudiciary remaining high while trust in politiciansdeclines, the willingness of judges to make broadlypolitical decisions seems likely to increase further.

Of course, judicial activism has proceededfurther in some democracies than in others. Incomparative rankings of judicial activism, theUnited States always comes top (Figure 12.2).America is founded on a constitutional contractand an army of lawyers will forever quibble overthe terms. The USA exhibits all the features con-tributing to judicial activism. These include: awritten constitution, federalism, judicial indepen-dence, no separate administrative courts, easyaccess to the courts, a legal system based on judge-made case law and high esteem for judges.

Fewer of these conditions are met in Britain, acountry that is one of the few democracies tooperate without judicial review. However, even inBritain judicial activism has increased, reflectingEuropean influence. British judges were willingaccomplices of the European court as it sought toestablish a legal order applicable to all memberstates. The country’s adoption of the EuropeanConvention on Human Rights (ECHR) in 1998,and the decay of the royal prerogative whichallowed the state to claim to be above the law, alsoencouraged judicial assertiveness. Proposals for anew Supreme Court to act as a final court ofappeal, announced in 2003, may well add to thejudiciary’s lustre.

Formal statements of rights have also encour-aged judicial expansion in other English-speakingcountries. Consider Canada and New Zealand. InCanada a Charter of Rights and Freedoms wasappended to the constitution in 1982, givingjudges a more prominent role in defending indi-vidual rights. Canada’s court built on the charterto develop a particular interest in protecting therights of people accused of crimes (Morton, 1995,p. 58). New Zealand introduced a bill of rights in1990, protecting ‘the life and security of theperson’ and also establishing traditional and previ-ously uncodified democratic and civil rights.

These charters pose a difficulty in countries witha tradition of parliamentary sovereignty, such asNew Zealand and the UK. Ingenuity is needed tointegrate a bill of rights with the supposed sover-eignty of the legislature. How can the authority ofparliament be reconciled with the overarchingstatus of rights charters? Fortunately, New Zealandhas delivered an ingenious solution. There, theAttorney-General (a cabinet minister who bridgesthe political and judicial worlds) advises MPs onwhether legislative proposals are consistent withbasic rights. Technically, at least, parliamentretains sole responsibility for adjusting its billsaccordingly.

Britain has adopted a similar halfway house inincorporating the ECHR. In theory, the legislatureremains supreme but in practice MPs are unlikelyto override a judicial opinion that a bill contra-dicts protected rights. In this way, sovereignty canbe simultaneously defended and diluted.

218 GOVERNMENT AND POLICY

Figure 12.2 Levels of judicial activism in selecteddemocracies

MOST ACTIVE

United StatesCanada

AustraliaGermany

ItalyIsraelJapanFrance

United KingdomSweden

LEAST ACTIVE

Source: Adapted from Holland (1991, p. 2).

➠

➠

CONSTITUTIONS AND THE LEGAL FRAMEWORK 219

More than ever before, judges participate in politics,striking down policies and laws deemed to contra-vene the constitution. But why should the judiciary bepermitted to encroach on the authority traditionallyaccorded to the elected branches of government?Whatever became of democracy – of rule by thepeople?

The case forThe fundamental argument for judicial authority is aliberal one: that tyranny of the majority is tyrannynonetheless. Subjecting government to the rule oflaw is a core achievement of Western politics, anaccomplishment to be cherished rather than criti-cized. Dividing power between the legislature, theexecutive and the judiciary is in practice the only wayof containing it.

Of course, interpreting the constitution is bound tocause controversy but an independent judiciary iswell-suited to the job of arbiter. Judges are held inhigher esteem than politicians and judicial interpreta-tion is likely to be more stable and disinterested thanthat offered by elected politicians, thus providingmore continuity to citizens and businesses alike.

In any case, judicial authority is often exaggerated.Judges review after the event; political initiativeremains with the executive.The justices wisely keepaway from sensitive areas such as war and taxation.And even an independent judiciary does not escapepolitical influence. Politicians have a say in theappointment of senior justices, the government canalways seek to change the constitution if it disagreeswith judicial interpretation and in practice the execu-tive retains responsibility for crisis management.

The case againstIn a democracy, the people must be sovereign. If agovernment behaves poorly, the solution lies in thepolling booth, not the courtroom. As Bork (2003)argues, societies must be responsible for their ownmoral judgements and should avoid simply dele-gating difficult decisions to judges. Historically, judi-cial authority developed as a device enabling the

wealthy to protect their property in a democratic eraand even today the judiciary shows a bias towardsdefending established interests.This conservative dis-position is strengthened by the narrow social andeducational background of judges, most of whom arestill middle-aged to elderly middle-class white men.

Further, the modern judiciary places increasingreliance on international conventions, a form of con-temporary imperialism that enables judges to escapedemocratic control exercised through national elec-tions. From this perspective, judicial power is a mecha-nism through which global forces override electedgovernments.

Even if a mechanism of constitutional arbitration isneeded (and Britain has got by well enough by relyingon informal conventions), why should judges, withtheir narrow legal training, be given the job? Why notcall upon an upper chamber or that oldest of politicalinstitutions – a council of elders – to perform the task?

AssessmentThis debate would benefit from fuller appreciation ofthe constitutional courts that have emerged in manyEuropean countries. In the political basis of theirappointments, relatively rapid turnover of membersand reliance on a political style of operation, constitu-tional courts operate very differently from traditionalsupreme courts.

To a degree, constitutional courts have alreadybecome the third chamber of politics. Indeed, Shapiroand Stone (1994, p. 405) suggest that ‘in the context ofexecutive-dominated legislative processes, the impactof constitutional courts may at times overshadow thatof parliaments’.Those who want the judiciary to bekept out of politics are fighting not a losing battle butone that is already lost.

Further reading: Bork (2003), Griffith (1997), Stone Sweet(2002).

DEBATE

A POLITICAL ROLE FOR THE JUDICIARY?

Independence and recruitment

Given the growing political authority of the judi-ciary, the question of maintaining its indepen-dence gains in importance. Liberal democraciesaccept judicial autonomy as fundamental to therule of law. To achieve this end, security of tenureis protected. In Britain, as in the American federaljudiciary, judges hold office for life during ‘goodbehaviour’. Throughout continental Europe thestate exerts closer control, reflecting a philosophythat views the judiciary as an agent of the stateeven if operating autonomously within it. Even so,judges remain secure in their tenure.

How, though, is this security to be achieved? Theproblem is that if judges are chosen, promoted ordismissed by government, they may become exces-sively deferential to political authority. In Japan,for example, research suggests that lower-courtjudges who defer in their judgments to the rulingLiberal Democratic Party are more likely to begiven promotion (Ramseyer and Rasmusen, 2001).

Box 12.4 shows the four main solutions.Cooption by other judges offers the surest guar-antee of independence but democratic election, aspractised in some American states, is more(perhaps excessively?) responsive to popular con-cerns. The format preferred depends on the weightgiven to judicial independence, on the one hand,and responsiveness to public opinion or partybalance, on the other.

In practice, professional and political criteria canbe combined. Judges are often selected by the gov-ernment from a pool of candidates prepared by aprofessional body. In South Africa, for instance,the President of the Republic appoints seniorjudges after consulting a Judicial ServicesCommission which includes representatives fromthe legal profession as well as the legislature.

For courts charged with judicial review, selectionnormally involves a clear political dimension. Inthe USA, the stature of the Supreme Court is suchthat appointments to it (nominated by the presi-dent but subject to Senate approval) are key deci-sions. Senate ratification can involve a set-piecebattle between presidential friends and foes. Inthese contests, the judicial experience and legalability of the nominee may matter less than ide-ology, partisanship and a clean personal history.Even so, Walter Dellinger, former acting USSolicitor-General, argues that ‘the politicalappointment of judges is an appropriate “democ-ratic moment” before the independence of lifetenure sets in’ (Peretti, 2001). In European states,members of constitutional courts are usuallyselected by the assembly. So in both American andEurope, political factors influence court appoint-ments, precluding a sharp distinction betweenlegal and political realms.

But the tenor of judges’ decisions cannot alwaysbe predicted on appointment. In post, judges con-

220 GOVERNMENT AND POLICY

Method Example Comment

Popular election Some states in the USA Produces responsiveness to public opinion but at what price in impartiality and competence? May be accompanied by recall procedures

Election by Some states in the USA, This method was also formally used for senior judges in the assembly some Latin American communist states but in practice the party picked suitable

countries candidates

Appointment by Britain, Supreme Court ‘Danger’ of political appointments though most judges will be the executive judges in the USA (subject appointed by an earlier administration, providing continuity

to Senate approval)

Cooption by the Italy,Turkey Produces an independent but sometimes unresponsive judiciaryjudiciary

B OX 1 2 . 4

Methods of selecting judges

tinue to develop their outlook. The best exampleis Earl Warren, Chief Justice of the AmericanSupreme Court, 1953–69. Warren, a formerRepublican governor of California, was appointedby President Dwight Eisenhower but went on todismay the president with his liberal rulings,notably on school desegregation. Eisenhower waslater moved to describe Warren’s appointment as‘the biggest damn fool mistake I ever made’.

The crucial point here is tenure. American jus-tices hold office for life, subject only to ‘goodbehaviour’, giving time for their own jurispru-dence to evolve. By contrast, the constitutionalcourts of Europe usually limit their judges to oneterm of seven to nine years. This enforcedturnover in Europe reflects the more political envi-ronment of an exclusively constitutional court andlimits the long-term impact of any one judge.

Administrative law

Where constitutional law sets out the fundamentalprinciples governing the relationship betweencitizen and state, administrative law covers therules governing this interaction in detailed set-tings. If a citizen is in dispute with an agency ofgovernment over specific issues such as pensionsentitlements, student loans or redundancy pay-ments, resolution is much more likely to be in anadministrative court or tribunal than in theregular law courts. Just as the central issue of judi-cial review is enforcement of the constitution, sothe task confronting administrative law is regula-tion of the bureaucracy.

DefinitionAdministrative law sets out the principles gov-erning decision-making by public bodies, princi-pally the bureaucracy. For example, America’sAdministrative Procedure Act, 1946, requirescourts to hold unlawful any agency action that is‘arbitrary, capricious, an abuse of discretion, orotherwise not in accordance with law’.

The increase of government activity in the twen-tieth century led to an expansion of administrativelaw. Administrative regulation may lack the high-profile political activity of constitutional courts

but by subjecting the work of public officials toclear rules they perform a function that is essentialto a liberal society. They help to balance the rela-tionship between the state and its citizens.

Typical questions asked in administrative laware:

� competence: was an official empowered to makea particular decision?

� procedure: was the decision made in the correctway (e.g. with adequate consultation)?

� fairness: does the decision accord with naturaljustice?

� liability: what should be done if a decision wasincorrectly made or led to undesirable results?

Countries can handle the problem of legal regu-lation of the administration in three ways, withthe chosen method reflecting conceptions of thestate (Box 12.5). The first solution, common incontinental Europe, is to establish a separatesystem of administrative courts concerned exclu-sively with legal oversight of the bureaucracy. Aseparatist approach marks out a strong publicsphere governed by its own legal principles. Thisseparatist approach speaks directly to the problemsarising in public administration but runs the riskof boundary disputes over whether a case shouldbe processed through administrative or ordinarycourts.

CONSTITUTIONS AND THE LEGAL FRAMEWORK 221

Method Definition Example

Separatist Special codes and Francecourts

Integrationist Rely on ordinary Anglo-law and courts American

countries

Supervision A procurator assesses Russia the legality of administrative acts

Source: Bell (1991).

B OX 1 2 . 5

Methods for judicial regulation ofthe bureaucracy

France is the most influential example of thisseparatist model. It has developed an elaboratestructure of administrative courts, headed by theConseil d’Etat, founded in 1799 (Figure 12.3). Alladministrative decisions taken by ministers andtheir officials are subject to review by the Council,a wide remit which can lead to slow decisions.Nonetheless, by developing its own case law, theCouncil has established general principles regu-lating administrative power. The Council’s prestigeexpresses the autonomy of the public realm whilealso enabling it to check executive power.

The second solution, favoured in Anglo-American countries with a common law tradition,seeks to deny the distinction between public andprivate law. The idea here, rarely fulfilled in prac-tice, is that the same legal framework should coverpublic and private transactions alike. For instance,employment in the public sector should be regu-lated by the same laws as apply to the privatesector; no special codes should be necessary.Similarly, ordinary courts should be able to arbi-trate disputes between bureaucrats and citizens; noConseil d’Etat is required. And in the UnitedStates, the existing courts do spend a considerableamount of time reviewing rule-making and appli-cation by the numerous agencies of the Americangovernment.

One strength of this integrationist philosophy isthat it prevents boundary disputes and simplifiesthe judicial system. Above all, the integrationistphilosophy affirms a modest aspiration for thepublic sphere; the state must abide by the samelaws as its citizens. However, in reality, specialcourts are rarely avoided entirely. Thus even theUnited States has administrative courts dedicatedto taxation, military, bankruptcy and patent issues.In any case, special obligations on American gov-ernment, such as ensuring all citizens are accordeddue process of law, are explicitly imposed by theconstitution.

Even in the absence of special courts, moreinformal administrative committees invariablyemerge in their stead, dealing with appealsinvolving such areas as employment law and socialsecurity. Such committees (often called tribunals)have their own strengths: they are quicker, cheaperand more flexible – though often more secretive –than the courts of law. In Britain, most adminis-trative cases are resolved through these tribunals.

The third approach to formal regulation of thebureaucracy is through administrative supervision.Peter the Great introduced this device in Russia in1722; it is now used throughout Eastern Europe.An officer known as the procurator supervises thelegality of administrative actions and can suspendbureaucratic decisions pending judicial resolution.Russia revived the role of procurator in 1992. Theprocurator covers principles as well as cases and soperforms a broader function than an ombudsmanwho merely investigates citizens’ allegations of mal-administration arising from individual cases. Theprocurator can form general principles, acting as anagent of a central ruler who, as in Russia, is posi-tioned above the bureaucracy.

Whichever approach countries adopt to the reg-ulation of their public bureaucracy, there can beno doubting the importance of the task. Thetwentieth-century expansion of governmentcreated not only new rights for citizens but alsofresh opportunities for public agencies to evadethose rights. In a liberal state, administrativecourts and tribunals make by their mere existencea worthwhile statement that public servants areindeed expected to behave in a fair and reasonableway to the public they serve. It is for such reasonsthat in 1982 Lord Diplock, a senior British judge,

222 GOVERNMENT AND POLICY

Figure 12.3 Adminstrative courts in France

Council of State

Thirty-eight administrative

tribunals

Seven administrative courts of appeal

Note: Separate courts exist for constitutional cases and for civil andcriminal law.Source: Adapted from Elgie (2003).

said that he regarded the development of adminis-trative law ‘as the greatest achievement of theEnglish courts in my lifetime’ (McEldowney(2003, p. 3).

Law in new democracies

New democracies are political construction sites,in which constitutional engineers seek to build aframework of law, including enhanced respect forthe judiciary (Kolstø, 2000). Their task is arduous.Establishing a state based on law in a countrywhere brute power has previously ruled the roostis inevitably a long-term assignment. Agreeing thenew constitution is merely the beginning. Thesubsequent requirement is to ensure that an inde-pendent judiciary is available to enforce the neworder and, crucially, that all political actors abideby the court’s decisions.

The status of the judiciary needs to increasefrom its lowly position in the previous regime.The powerful must learn new limits to theirbehaviour and the less powerful must begin totreat the courts as a realistic form of redress. Theextent to which new democracies have entrenchedthe constitution and judicial autonomy is at bestvariable, as our analysis of the Spanish-speakingand post-communist worlds will show.

Spain’s transition to democracy, following thedeath of General Franco in 1975, is an example of adeliberate and ultimately successful attempt tobuild the rule of law into the country’s new politicalarchitecture. As Heywood (1995, p. 103) writes,

After nearly four decades of dictatorship underGeneral Franco, it was to be expected that the1978 constitution should place particularemphasis on legal accountability. A centralconcern of the new democracy was that itshould be established as a ‘state of law’ in thesense of its constitutional arrangements beingboth legally accountable and enforceable.

Spain’s constitution-makers established a newConstitutional Court with 12 members appointedfor nine years by parliament, the government andthe legal profession. To ensure its independence,tribunal members are debarred from holding any

other public office or from playing a leading rolein a political party or trade union. Even so, suspi-cions remained that early appointments reflectedparty patronage.

Although Spain’s new democratic order rapidlyconsolidated, the judiciary initially retained itspre-democratic image as remote, inefficient andconservative. It was not until judges began investi-gating allegations of corruption against leadingpoliticians in the early 1990s – thus demon-strating their practical as well as their constitu-tional autonomy – that their reputation began toimprove.

The rule of law sought and found in Spain hasproved more elusive in many post-military democ-racies in Latin America. Elected presidents inSouth America countries may be less cavalier withthe constitution than their strong-armed predeces-sors but even so several have treated the constitu-tion as a flexible document to be adapted to suittheir own political requirements. For example,some have sought to abolish term limits so thatthey can stand for re-election.

Other South American constitutions haveretained privileges for departing generals, thus per-petuating a sense of an institution remainingabove the law. For instance, Chile’s armed forceswere initially granted immunity from prosecutionin civilian courts, a tactic that effectively enabledformer generals to escape justice for politicalmurders committed during their tenure. InArgentina, similar legal exemptions dating fromthe 1980s still formed a running political soretwenty years later.

In addition to the difficulties of establishing theconstitution as an effective political framework,the rule of law is held back through most of LatinAmerica by the low standing and standards of thejudiciary. Prillaman (2000) chronicles the prob-lems: chronic inefficiency within the judicialsystem, the vulnerability of judges to politicalpressure, outdated laws and the public’s lack oftrust in legal remedies. These factors combine tolimit the judiciary’s significance in Latin Americangovernance, holding back not just democratic con-solidation but also economic advance.

In some South American countries, the rule oflaw remains but an aspiration. The police prob-ably commit almost as many crimes, including

CONSTITUTIONS AND THE LEGAL FRAMEWORK 223

murder, as they solve. Coordination between thepolice and judges is poor; sometimes both arebribed and threatened by drug barons. InColombia, one in three judges received a deaththreat in the 1980s. Informal arbitration andindigenous justice are often more accessible thanthe remote legal system (Domingo and Sieder,2001). In the absence of effective policing, the res-idents of shanty towns sometimes resort tolynching as a form of social control, bringingabout a further devaluation of official justice.

Although moves to reform the judiciary areafoot, and are actively promoted by internationalorganizations, the task is inevitably long-term.One legal expert estimated that reform of Chile’scriminal justice system would take at least 15 years(Frühling, 1998, p. 252). An authoritarian tradi-tion and a cultural emphasis on personal powerare infertile ground on which to nurture the ruleof law.

The post-communist experience with constitu-tions and law is again somewhat patchy. As withother aspects of democratic consolidation, CentralEurope has made more progress than the successorstates to the Soviet Union. In Hungary andPoland, new constitutional courts have met withconsiderable success. The Hungarian court hasmade significant judgments on such topics ascapital punishment, abortion, private propertyand the electoral system. In Poland, the court ‘hasbecome in a short time an important part of thegovernmental system’ (Fitzmaurice, 1998, p. 96).One factor contributing to popular acceptance ofthese new courts has been the adoption of aGerman innovation allowing citizens to petitionthe court directly if they believe their rights havebeen denied (von Beyme, 2003).

By contrast, many successor states to the SovietUnion have simply swapped one non-constitu-tional order for another. In many of these coun-tries, presidents adopt the imperious style ofFrance’s King Louis XIV: ‘it is legal because I wishit’. Even in Russia, ‘there has been and remains aconsiderable gap between individual rights onpaper and their realization in practice. The furtherone goes beyond Moscow and St Petersburg intothe provinces, the enforcement gap tends to growgreater’ (Sharlet, 1997, p. 134).

But we should beware of judging the constitu-

tional quality of post-communist regimes againstcontemporary Western standards. As Sharlet(1997, p. 134) reminds us,

while the Founding Fathers of the Americanrepublic quickly added the Bill of Rights to theirnewly ratified Constitution in the late eigh-teenth century, a number of these rightsremained essentially ‘parchment rights’ and didnot garner nationwide respect and judicialenforcement until well into the twentiethcentury. Is it surprising that Russia with itsthousand-year authoritarian past and long tradi-tion of legal nihilism should be proceedingslowly in Rule of Law development? Surely it ismore remarkable that Russia has made theprogress it has, including in the uncharted terri-tory of civil rights.

Law in authoritarian states

In authoritarian regimes, constitutions are feebledocuments. The nature of such states is that anyrestraints on rule go unacknowledged; power, notlaw, is the political currency. It follows that thestatus of the judiciary, as guardian of the law, issimilarly diminished.

Non-democratic rulers follow two broad strate-gies in limiting judicial authority. One tactic is toretain a framework of law but to influence thejudges indirectly. In Indonesia, the Ministry ofJustice still administers the courts and pays judges’salaries; the justices understand the implications.More crudely, unsatisfactory judges can simply bedismissed. Egypt’s President Nasser adopted thisstrategy with vigour in 1969. He got rid of 200 inone go: the ‘massacre of the judges’. In Uganda, anextreme case, the killing was real rather thanmetaphorical: President Amin had his ChiefJustice shot dead.

A second strategy is to bypass the judicialprocess altogether. For instance, many non-demo-cratic regimes use Declarations of Emergencywhich are exempt from judicial scrutiny. Onceintroduced, such ‘temporary’ emergencies drag onfor years, even decades. Alternatively, rulers canmake use of special courts that do the regime’sbidding without much pretence of judicial inde-

224 GOVERNMENT AND POLICY

pendence. Military rulers frequently extend thescope of military courts, often meeting in secret,to include civilian troublemakers. Ordinary courtscan then continue to deal with non-political cases.Even the United States proposed to use militarycourts to try more than 600 terrorist suspects heldin Guantanamo Bay, Cuba, following the war inAfghanistan.

Communist states offered a more sophisticateddowngrade of constitutions and the judiciary.Marxist theory explicitly rejected the Western ideaof constitutional rule with its emphasis on limitedgovernment, individual rights and private prop-erty. What does the Western tradition amount to,asked the communists, other than an affirmationof the status quo? In communist states, legal docu-ments were of little moment. Nothing couldhinder the party’s task of building socialism; theparty’s mission was too challenging and importantto be subject to formal limits. So communistregimes were ruled by the party rather than by law,a point confirmed by the reference in the constitu-tions themselves to the party’s leading role.Communist constitutions also stressed social andeconomic rights such as the right to work, in con-trast to the Western emphasis on more politicalliberties such as freedom of speech.

Just as communist states rejected the idea of aneutral constitution, so too did they dismiss theconcept of judicial independence. The judges, too,must contribute to building socialism – and pro-tecting the party’s position. Throughout the com-munist world, judges were selected for theirparty-mindedness and were expected to put thisvirtue to good effect in court.

Yet as with many other aspects of communistpolitics, this situation was beginning to alterbefore the 1989 revolutions. Even authoritarianregimes can discover the advantage of applying therules consistently. When one citizen sued another,or even when one enterprise sued another, theinterests of the communist party were best servedby resolving the issue through law. The courtstherefore observed a measure of ‘socialist legality’.

China provides a contemporary example of theevolution of communist thinking about constitu-tions and law. Constitutional revisions have laggedbehind the reformist thinking of the party’s elite.References to the dictatorship of the proletariat,

and to other Marxist sacraments, have finally beenremoved but the supremacy of the party is stillaffirmed in the constitution. In that sense, powerstill trumps the constitution and law remains abird in a cage (Lubman, 1999).

But the interpretation of law has become moresympathetic. In the early decades of the People’sRepublic, legal perspectives were dismissed as‘bourgeois rightist’ thinking. There were very fewlaws at all, reflecting a national tradition of unreg-ulated power; the judiciary was largely a branch ofthe police force. However, laws did become moreprecise and significant after the hiatus of theCultural Revolution in the 1960s. In 1979, thecountry passed its first criminal laws; later revi-sions abolished the vague crime of ‘counter-revolu-tion’ and established the right of defendants toseek counsel. Law could prevail to the benefit ofeconomic development.

Reform notwithstanding, Chinese politicsremains authoritarian. The courts are regarded asjust one bureaucratic agency among others; manyjudgments are simply ignored. There is no tradi-tion of individual rights, enforceable against thestate and party. Legal institutions remain less spe-cialized, and legal personnel less professional, thanin established democracies. The death penalty isstill used. Private property rights are still notsecure. Above all, party officials continue tooccupy a protected position above the law. Politicsstill comes first.

International law

We conclude this chapter with a section on inter-national law. For three reasons, international lawmust now receive attention from students of com-parative politics:

� International law helped to define the divisionof the world into the states which provide theunit of comparative politics. States were theonly, and remain the major, subjects of interna-tional law. Through international law, statesreinforced their dominant political position andit is through participation in an internationallegal community that statehood is formallyacquired.

CONSTITUTIONS AND THE LEGAL FRAMEWORK 225

� International law forms part of national law,often without any special mechanism of incor-poration. Many constitutions are explicit onthis point. The German constitution, forinstance, states that ‘the general rules of publicinternational law are an integral part of federallaw. They shall take precedence over the lawsand shall create rights and duties for the inhabi-tants of the federal territory.’

� National sovereignty notwithstanding, interna-tional laws can apply directly to individuals andtheir rulers. The famous example here is theNuremberg War Crimes Tribunal in 1946. Thistribunal declined to accept the defence thatindividuals who committed crimes againsthumanity were just obeying state orders. Withthe end of the Cold War, new international tri-bunals have been created.

International agreements constrain nationalpolicy-makers. Such accords set out objectives (forexample reducing carbon dioxide emissions)which national governments must – or at leastshould – put into effect. When states fail to abideby conventions they themselves have signed,affected individuals can in principle seek a remedythrough the courts, both national and interna-tional.

An example from Australia illustrates this pointand also draws out some broader domestic impli-cations of international law. After Australia’snational government ratified a protocol to theInternational Covenant on Civil and PoliticalRights in 1991, Nick Toonen, a gay rights activist,lodged a complaint against Tasmania’s prohibitionof sexual relations between men. Tasmaniaremained resolute but the federal government inCanberra passed a liberalizing law, overridingprovincial legislation.

DefinitionInternational law is the system of rules whichstates (and other actors) regard as binding intheir mutual relations. It derives from treaties,custom, accepted principles and the views oflegal authorities.The term ‘international law’ wascoined by the English philosopher JeremyBentham (1748–1832).

The Toonen case shows how international agree-ments can impinge on the balance between dif-ferent levels of government within a state. Indeed,critics allege that Australia’s federal governmentcould in theory use its expanding treaty commit-ments to interfere in virtually any area of activitywhich, under the national constitution, is suppos-edly reserved to the states (Scott, 1997). So inter-national law provides an indirect route throughwhich one level of government within a countrycan seek to influence another.

A further bypass around the blockage of statesovereignty has been achieved by setting up inter-national courts to try individuals directly for theiralleged crimes, even if these were committed inthe name of the state. Of course, the Nurembergtrials after the Second World War had establishedthat individuals are responsible for their own actsin war. That view itself departed from the notionof treating states as the only subjects of interna-tional law.

However, in the 1990s, the International Courtof Justice set up new tribunals, the first sinceNuremberg, to try war crimes suspects fromRwanda and the former Yugoslavia. In 1999, forinstance, a female ex-minister from Rwanda wasindicted for rape by the International CriminalTribunal on Rwanda for sexual assaults she con-doned during the 1994 genocide. However, thesetribunals are proving to be expensive and slow-moving. The Rwanda tribunal cost more than$550m over seven years but secured just nine con-victions, including three guilty pleas, during thisperiod. In the Yugoslav case, the trials of impris-oned Serbian leaders Slobodan Milosevic andVojislav Seselj aided public sympathy back home,with both leaders winning election to the Serbianparliament from their prison cells in 2003. Why,asked the Serbs, should these leaders be singledout, ignoring similar atrocities committed by theother sides?

One innovation of these new tribunals has beento indict suspects who were not already in custody.This raises the prospect of international snatchsquads removing people from their home countryto the site of the trial, further diluting nationalsovereignty. However, bringing these suspects tojustice has proved to be a political challenge. Suchintervention can stimulate a damaging reaction in

226 GOVERNMENT AND POLICY

the suspect’s own community. Yet the alternative isequally unpalatable. Failing to capture indictedsuspects, such as Bosnian leader Radovan Karadzicand his henchman Ratko Mladic, makes the inter-national community appear weak.

Further impetus to the development of an inter-national judicial regime came with the establish-ment in 2003 of the permanent InternationalCriminal Court (ICC). Based in the Dutch city ofThe Hague, the Court is charged with dealingwith cases of genocide, war crimes and crimesagainst humanity. The emergence of this interna-tional body is significant in itself but again itspractical impact remains to be seen. Already, theCourt has collided with the countervailing tradi-tion of national sovereignty. The USA has vigor-ously refused to accept the ICC’s jurisdiction.Fearing false accusations against and maliciousprosecution of its troops, the USA has soughtbilateral immunity deals with countries to which itdonates military aid. These agreements exemptAmerican citizens from any arrest warrants theICC may issue in the countries concerned, thuskeeping the United States beyond the Court’sambit (Schabas, 2001).

Key reading

Next step: Stone Sweet (2000) examines consti-tutional politics in five European political systems.

Shapiro and Stone Sweet (2002) is a collection onpolitical jurisprudence, including the UnitedStates. The special issue of Political Studies (1996)is an excellent survey of constitutionalism; for acollection focused on the USA, see Ferejohn et al.(2001). Maddex (2000) surveys 100 constitutions.Von Beyme (2003) covers constitution-making inEastern Europe. On the judiciary, Guarnieri andPerderzoli (2002), Jacob et al. (1996) and Tate andVallinder (1995) are cross-national studies. Forparticular countries and regions, see O’Brien(1993) and Shapiro (1990) on the AmericanSupreme Court; Solomon and Foglesong (2000)on judicial reform in post-communist Russia;Lubman (1999) on post-Mao China; andPrillaman (2000) on the judiciary in LatinAmerica. On international law, Bull (1977)remains a good starting point.

CONSTITUTIONS AND THE LEGAL FRAMEWORK 227

Federalism 228Federal–state relations 231Assessing federalism 235Unitary government 236Local government 239Subnational government in new

democracies 244Subnational government in authoritarian

states 245Key reading 246

Governing always has a territorial dimension.Rulers need to extract resources from their

territory while also retaining the willingness of thepopulation to remain within the state’s orbit. Toachieve these ends, the modern state consists of anintricate network of organizations, typically con-sisting of:

� the central government � the central government’s offices in the provinces � regional or provincial governments � local authorities� the European Union (for member states).

Governance in all established democracies com-prises several levels with extensive flows of com-munication, money and influence between them.Multilevel governance is the term used to describehow policy-makers and experts in particularsectors (e.g. transport, education) negotiate acrosslevels, seeking to deliver coherent policy in theirspecific area.

Clearly, the balance between these assorted tiersraises important questions of democratic gover-nance. Who initiates policy? Who funds it? Whoexecutes it? Who is accountable for it? And, inview of September 11, how well can such intricatenetworks respond to national emergencies?

In this chapter, we examine the two basic solu-tions to the territorial organization of power:federal and unitary government. We then turn to

the lowest level of authority within the state, localgovernment, before examining the patterns of sub-national government found in new democraciesand authoritarian regimes.

DefinitionMultilevel governance emerges when severaltiers of government share the task of regulatingmodern society. Expert networks form withinparticular policy sectors, cutting across tiers andoften extending to relevant interest groups.Multilevel governance inheres in a federation butthe term is also used in discussing unitary states,particularly in the context of the relationshipbetween the European Union and its mainlyunitary members (Hooghe and Marks, 2001).See also pluralism (p. 177).

Federalism

Federalism is a method for sharing sovereignty,and not just power, between governments withina single state. It is a constitutional device, presup-posing a formal political agreement establishingboth the levels of government and their spheresof authority. So, like the constitutions withinwhich they are embedded, federations are alwaysa deliberate creation. The United States gave usthe first modern constitution and, as part of the settlement reflected in that document, itestablished the world’s first and most influentialfederation.

In a federation, legal sovereignty is sharedbetween the federal (or national) government andthe constituent states (or provinces). Neither levelcan abolish the other. It is this protected positionof the states, not the extent of their powers, whichdistinguishes federations (such as the USA andCanada) from unitary governments (such as theUK and France). Multiple levels of governance are

Chapter 13

Federal, unitary and local government

228

integral to a federation whereas in a unitary systemsovereignty resides solely with the centre and lowerlevels exist at its pleasure.

A federal constitution allocates specific functionsto each tier. The centre takes charge of externalrelations – defence, foreign affairs and immigra-tion – and some common domestic functions suchas the currency. The functions of the states aremore variable but typically include education, lawenforcement and local government. Residualpowers may also lie with the states, not the centre(see Box 13.1).

In nearly all federations the states have a guaran-teed voice in national policy-making through anupper chamber of the assembly. In that chambereach state normally receives equal, or nearly equal,representation. The American Senate, with twosenators per state, is the prototype. In unitarystates, by contrast, the rationale for an upperchamber is less clear. Indeed, most legislatures inunitary states consist of only one chamber.

DefinitionFederalism is the principle of sharing sover-eignty between central and state (or provincial)governments; a federation is any politicalsystem that puts this idea into practice. A con-federation is a looser link between participatingcountries which retain their separate statehood.A confederation does not create a new centralgovernment with a direct relationship to its citi-zens (Lister, 1996).

The natural federal structure is for all the stateswithin the union to possess identical powers underthe constitution. However, reflecting national cir-cumstances, some federations are less balanced.Asymmetric federalism arises when some stateswithin a federation are given more freedom thanothers. In Canada, for example, Quebec national-ists have long argued for special recognition fortheir French-speaking province. Asymmetric feder-alism is a natural response to differences in power

FEDERAL, UNITARY AND LOCAL GOVERNMENT 229

Type of allocation Canada Germany

Exclusive jurisdiction (functions The national government The federal government’s allocated entirely to one level): exclusively controls 29 functions, responsibilities include defence,

including criminal law, the citizenship, immigration and the currency and defence. internal market The provinces control ‘all matters No specific powers are explicitly of a merely local or private nature granted to the Länder but the in the province’, including local provinces implement federal laws ‘in government their own right’

Concurrent jurisdiction (functions Both the national and provincial Concurrent powers include criminalshared between levels): governments can pass laws dealing law and employment. A

with agriculture and immigration constitutional amendment in 1969/70 created a new category of joint tasks, including regional economic policy and agriculture

Residual powers (the level The national parliament can make Any task not otherwise allocated responsible for functions not laws for the ‘peace, order and good remains with the Länderspecifically allocated by the government of Canada’constitution):

Source for Germany: Jeffery (2003)

B OX 1 3 . 1

The allocation of functions in the Canadian and German federations

and culture between regions within a federation.However, the solution carries its own dangers. Therisk is that a spiral of instability will develop as theless favoured states seek the privileges accorded tomore autonomous provinces.

Federations must be distinguished not just fromunitary states but also from confederations. In thelatter, the central authority remains the juniorpartner; it is merely an agent of the componentstates. The classic instance of a confederation isthe short-lived system adopted in 1781 in what isnow the United States. The weak centre,embodied in the Continental Congress, couldneither tax nor regulate commerce. It also lackeddirect authority over the people. It was the feeble-ness of the Articles of Confederation that led tothe drafting of a federal constitution, and to thecreation of the United States proper, in 1787.

Federalism is a recognized solution to theproblem of organizing the territorial distributionof power. Elazar (1996, p. 426) counts 22 federa-tions in the world, containing some two billionpeople or 40 per cent of the world’s population(for some examples see Table 13.1). Federalism isparticularly common in large countries, whethersize is measured by area or by population. Four ofthe world’s largest states by area are federal:Australia, Brazil, Canada and the United States. InIndia, ten out of 25 provinces each contain morethan 40 million people, providing the only real-istic framework for holding together such a largeand diverse democracy. Germany, the largest

European country by population, exemplifies fed-eralism on that continent.

In theory, there are two routes to a federation:either by creating a new central authority or bytransferring sovereignty from an existing nationalgovernment to lower levels. Stepan (2001, p. 32)calls these methods ‘coming together’ and ‘holdingtogether’. In practice, coming together predomi-nates; federalism is almost always a compactbetween previously separate units pursuing acommon interest. The United States, for instance,emerged from a meeting of representatives of 13American states in 1787. Similar conventions,strongly influenced by the American experience,took place in Switzerland in 1848, Canada in1867 and Australia in 1897/98.

So far, restructuring as a federation to hold adivided country together is a rare occurrence.Belgium is the main example. First established in1830, Belgium has been beset by divisions betweenits French- and Dutch-speaking regions.Constitutional revisions in 1970 and 1980devolved more power to these separate languagegroups. In 1983, Belgium finally proclaimed itself afederation of three parts: French-speaking Wallonia,Dutch-speaking Flanders and Brussels, the predom-inantly French-speaking capital located in theDutch-speaking part of the country (Arel, 2001).The Belgian experience does suggest that federationcan be an alternative to disintegration, a lesson ofvalue to other states confronting strong divisions ofethnicity, language, religion and culture.

230 GOVERNMENT AND POLICY

Table 13.1 Some federations in established democracies

Year established Area Population Number of states as a federation (sq. km, thousands) (million), 2002 within the federation

United States* 1776 9,373 290 50Canada* 1867 9,976 32 10Switzerland 1874 41 7 26Brazil 1891 8,512 182 27Australia 1901 7,687 20 6Germany* 1949 357 82 16India 1950 3,288 1,050 25Belgium 1983 30 10 3

* For profiles, see p. 147 (USA), p. 232 (Canada), p. 94 (Germany).

What, then, provokes distinct peoples to set outon the journey to a federation? Motives are moreoften negative than positive: fear of the conse-quences of remaining separate overcomes thenatural desire to preserve independence.Historically, the most common aspiration hasbeen to secure the military and economic bonusaccruing to large countries. More recently,however, interest in federalism has focused on itspotential in enabling people with conflicting iden-tities to continue to live together within a singlestate, as in Belgium.

Riker (1975, 1996) emphasizes the militaryfactor. He argues that federations emerge whenthere is an external threat. The American states,for instance, joined together in 1789 partlybecause they felt themselves to be vulnerable in apredatory world. When large beasts are lurking inthe jungle, smaller creatures must gather togetherfor safety. Or, as the American statesmanBenjamin Franklin (1706–90) put it, ‘we mustindeed all hang together or, most assuredly, weshall all hang separately’ (Jay, 1996, p. 142).

The military case for forming new federations iscurrently rather weak, given the paucity of majorwars in the twenty-first century. Even when acommon threat such as terrorism does emerge,coordination between existing states serves thepurpose more readily than sharing sovereigntythrough a new federation.

But it would be wrong to regard militarymotives as the sole reason for federation-making.The federal bargain has also been based on theeconomic advantages of scale. Even the Australianand American federalists felt that a commonmarket would promote economic activity. Butagain, federation now seems to be a rather convo-luted way of securing gains from trade.Straightforward free trade areas (FTAs) betweenneighbouring countries are proving to be a morepopular way forward. Unlike federations, FTAsimply no risk to sovereignty.

As military and economic arguments forforming new federations have weakened, sointerest in ethnic federalism has grown apace. TheBelgian experience shows that federations areuseful for bridging ethnic diversity within adivided society; they are a device for incorporatingsuch differences within a single political commu-

nity. People who differ by descent, language andculture can nevertheless seek the advantages ofmembership in a shared enterprise.

The Indian federation, for instance, has so faraccommodated growing religious tensions betweenHindus, Muslims, Sikhs and Christians. On asmaller scale, the Swiss federation integrates 23cantons, two and a half languages (German andFrench, plus Italian) and two religions (Catholicand Protestant). Building on such experience,advocates of ethnic federalism claim that it mayhelp to integrate ethnic groups in post-TalibanAfghanistan and post-Saddam Iraq. A Kurdishleader in Iraq expressed the spirit of ethnic feder-alism when, amid the jockeying for position afterthe American invasion in 2003, he said, ‘the newconstitution cannot be a constitution of themajority. It must be a constitution of the entirety’(Clover, 2003).

Federal–state relations

The relations between federal and state govern-ments are the crux of federalism. A federal frame-work creates an invitation to struggle between thetwo levels and indeed among the states themselves.Federalism creates both competition and themeans of its containment. In this section, we dis-tinguish between two traditional views of feder-alism before examining some practical similaritiesin the workings of contemporary federations.

Dual and cooperative federalism

It is helpful to distinguish between dual and coop-erative federalism. The former represents thefederal spirit and remains a significant theme inAmerican culture; the latter is an important idealwithin European thinking and takes us closer tothe realities of federal governance.

Reflecting the original federal principle as con-ceived in the United States, dual federalism meansthat the national and state governments operateindependently, each tier acting autonomously inits own sphere, and linked only through the con-stitutional compact. In the circumstances of eigh-teenth-century America, such separation was aplausible objective; extensive coordination

FEDERAL, UNITARY AND LOCAL GOVERNMENT 231

232 GOVERNMENT AND POLICY

Canada is a large country with a relatively small popu-lation. Its land mass is the second largest in the worldbut its population is little more than a tenth of itspowerful American neighbour. Most Canadians live inurban settlements in a 100-mile strip bordering theUnited States. Canada’s economy depends heavily onthe USA, a relationship reinforced by the formation ofthe North American Free Trade Association in 1994.

Reflecting British influence, Canada’s constitution,originally set out in the British North America Act of1867, established parliamentary government in a cen-tralized federation. Since then,‘Canada has movedfrom a highly centralized political situation to one ofthe most decentralized federal systems in the world’(Landes, 1995, p. 101).This reflects the central issue ofCanadian politics: the place within it of French-speaking Quebec.This division goes back to thecountry’s origins. From the sixteenth century, Franceand then Britain colonized the territories of Canada,inhabited at that time by around ten million indige-nous people. Britain finally defeated the French in1759.

For many Francophones, Canada consists of twofounding peoples – the British and French – whosestatus should be equal.This is taken to imply thatQuebec should be more than just one among tenprovinces. Since the 1960s a revived nationalist partyin Quebec has sought to implement this vision.However the federal response has been to decen-tralize power to all provinces, not Quebec only. In

Quebec itself, the PQ, elected to power in 1994, held aprovincial referendum in 1995 on ‘sovereignty associa-tion’ for Quebec.This would have combined politicalsovereignty for Quebec with continued economicassociation with Canada.The proposal lost by the nar-rowest of margins. Subsequently, the issue of constitu-tional reform declined in intensity, with the PQ votedout of office in 2003.

Even so, the regional foundations of Canada remaincentral to its politics. An earthquake election in 1993saw the national emergence not just of the PQ butalso of the Western-based Reform Party (now theCanadian Alliance).The Alliance is a populist partywhich seeks to make members of parliament moreresponsive to local electorates.Thus Canada’s contem-porary politics are an uneasy balance betweennational and regional parties, and of pro- and anti-system themes.The question ‘what is Canada?’remains permanently on the table, never to receive afinal answer.

Yet as Williams (1995, p. 69) writes,

despite all Canada’s domestic turmoil, it is endowedwith a responsive federal system. Over time this hascreated a good safe place to raise a family and earna crust, the ultimate test of a state’s obligations toits citizens.

Population: 32m.Gross domestic product per head:

$29,300.Self-reported ethnic origin (six largest

groups):

Form of government: a federal parlia-mentary democracy with tenprovinces. Most Canadians live inOntario or Quebec.

Legislature: the 301-seat House ofCommons is the lower chamber. Most

unusually for a federation, the 104members of the Senate, the upperchamber, are appointed by the primeminister, not selected by theprovinces.

Executive: the prime minister leads acabinet whose members he selectswith due regard for provincial repre-sentation. A governor-general servesas ceremonial figurehead.

Judiciary: Canada employs a dual(federal and provincial) court system,headed by a restrained SupremeCourt. In 1982 the country introducedthe Canadian Charter of Rights andFreedoms.

Electoral system: a plurality system withsingle-member districts.This producesmassive swings and distortions in rep-resentation in the Commons yet elec-toral reform is rarely considered (seeTable 9.1, p. 150).

Party system: the party system isstrongly regional, with the governingLiberals dominating the key provinceof Ontario.The main oppositionparties are currently the PartiQuébécois (PQ) and the Western-based Alliance.The once-powerfulConservatives and the left-wing New Democratic Party are smalleropposition parties.

Profile C A N A D A

Further reading: Bakvis and Skogstad (2002), Burgess(2001).

French, 4.7m

Canadian, 11.1m

Scottish, 4.2m

Irish, 3.8m

German, 2.7m

English, 6.0m

between federal and state administrations wasjudged to be neither necessary nor feasible.

In particular, the federal government wasrequired to confine its activities to functionsexplicitly allocated to it, such as the power ‘to layand collect taxes, to pay the debts and provide forthe common defence and welfare of the UnitedStates’. In the world after 9/11, ‘providing for thecommon defence’ has of course again become apivotal and complex task, calling for cross-levelcollaboration. However, James Madison, one ofthe Founding Fathers, believed that such timeswould be exceptional and that in the main thestates would be free to follow their own course(Hamilton, 1788a, p. 237):

The operations of the federal government willbe most extensive and important in times of warand danger; those of the State governments intimes of peace and security. As the formerperiods will probably bear a small proportion ofthe latter, the State governments will here enjoyanother advantage over the federal government.

Perhaps always a myth, dual federalism has longsince disappeared, overwhelmed by the demandsof an integrated economy and global war longbefore the arrival of terrorist threats. Even so, itexpresses an implicit if unrealizable strand infederal thinking.

DefinitionDual federalism, as originally envisaged in theUSA, meant that national and state governmentsretained separate spheres of action. Each levelindependently performed the tasks allocated toit by the constitution. Cooperative federalism,as practised in Germany, is based on collabora-tion between levels. National and state govern-ments are expected to act as partners infollowing the interests of the whole.

The second approach is cooperative federalism.This interpretation is favoured in the main byEuropean federations, especially Germany. Whilefederalism in the USA is based on a contract inwhich the states join together to form a centralgovernment with limited functions, the Europeanform (particularly in Germany) rests on the idea of

cooperation between levels. Such solidarityexpresses a shared commitment to a united society;federalism displays organic links that bind the par-ticipants together. The moral norm is solidarityand the operating principle is subsidiarity: the ideathat decisions should be taken at the lowest feasiblelevel. The central government offers overall leader-ship but leaves implementation to lower levels – adivision rather than a separation of tasks.

Cooperative federalism lacks the theoretical sim-plicity of the dual model but, as a more recentform, it provides a realistic account of how federalgovernance proceeds in practice. It also reflects adistinctly European approach born from the desireto overcome that continent’s long history of con-flict. Along with proportional representation andcoalition government, cooperative federalism aimsto ensure social peace by awarding major interestsa share in decision-making.

Consider German cooperative federalism inmore detail. Imposed by the Allies in 1949 as abarrier against dictatorship, federalism representeda return to the country’s strong regional roots.Federalism is reflected in the country’s officialtitle, the Federal Republic of Germany, and underthe Basic Law Germany’s federal character cannotbe amended. Federalism quickly became anaccepted part of the postwar republic (Umbach,2002).

From its inception, German federalism wasbased on interdependence, not independence. Allthe Länder (states) are expected to contribute tothe success of the whole; in exchange, they areentitled to be treated with respect by the centre.The federal government makes policy but theLänder implement it, a division of labourexpressed in the constitutional requirement that‘the Länder shall execute federal laws as matters oftheir own concern’. Because the provinces are theimplementing agent, they must first approve billsaffecting them through parliament’s upperchamber, the Bundesrat. Further, the constitutionnow explictly defines some ‘joint tasks’, such ashigher education, where responsibility is shared.

These mechanisms of interdependence havebeen reinforced by the powerful ConstitutionalCourt, which has encouraged the participants toshow due sensitivity to the interests of other actorsin the federal system.

FEDERAL, UNITARY AND LOCAL GOVERNMENT 233

Although German federalism remains far moreorganic than its American equivalent, its coopera-tive ethos has come under some pressure fromgrowing economic inequalities between Länder,particularly since unification. The political realityis that significant competition now exists along-side the tradition of solidarity (Jeffery, 2003).

Federalism in practice

Whatever their origins or philosophy, most federa-tions have witnessed growing interdependencebetween the two tiers. The German notion ofjoint tasks has become an administrative reality inall federations, a fact which led McDonald andMcDonald (1988) to write their obituary for tra-ditional dual federalism in the USA. The talk nowis of ‘intergovernmental relations’ in the UnitedStates, of ‘executive federalism’ in Canada and of‘multilevel governance’ in Europe. All such phrasespoint to an intermingling between tiers, implyingthat policy is made, funded and implemented bynegotiation between levels (Simeon and Cameron,2002).

In the game of central–state relations, thecentral government tended to gain influence formuch of the twentieth century. Partly, this trendreflected the centre’s financial muscle (see Box13.2). The flow of money became more favourableto the centre as income tax revenues grew with theexpansion of both the economy and the work-force. Income is invariably taxed mainly atnational level because otherwise people and corpo-rations could move to low-tax states. By contrast,

the states must depend for their own indepen-dently raised revenue on sales and property taxes, arevenue source that is smaller and less dynamicthan income tax. This shift in financial strength isclearly seen in Australia, where most of the states’revenue now comes from the federal government.

In the United States, an economic decline in theearly years of the twenty-first century led to afurther deterioration in the financial position ofthe states. Many found themselves hamstrung bythe requirement of their own state constitutions torun a balanced budget. This obligation does notapply to national government, a fact the free-spending George W. Bush sought to exploit in hiscampaign for reelection to the White House in2004. In the United States, as in most federations,states lack the financial flexibility available to thecentral government and in consequence they aremore sensitive to economic trends (Braun et al.,2003).

But the enhanced authority of the centre infederal systems has been more than a financialmatter. More than anything, it reflects the emer-gence of a national economy demanding overallplanning and regulation. To take just one example,when California experienced an electricityshortage in 2000, the federal governmentinevitably became involved as the state began todraw in power from surrounding areas. Similarly,power cuts on the Eastern seaboard three yearslater exposed the weaknesses of an electricity gridpartly controlled by fifty separate states. Equally,national planning is needed to forestall the absur-dity of highways ending at a state’s borders.

234 GOVERNMENT AND POLICY

Form of transfer Definition

Categorical grants For specific projects (e.g. a new hospital)

Block grants For particular programmes (e.g. for medical care)

Revenue-sharing General funding which places few limits on the recipient’s use of funds

Equalization grants Used in some federations (e.g. Canada and Germany) in an effort to harmonize financial conditions between the states. Can create resentment in the wealthier states.

B OX 1 3 . 2

Financial transfers from the federal government to states

The broader expansion of public functions hasalso worked to the centre’s advantage. The warsand depressions of the twentieth century invari-ably empowered the national government. Suchadditional powers, once acquired, tended to beretained. And the postwar drive to complete awelfare state offering equal rights to all citizens,wherever they live, enhanced the central powerstill further. Of the major Western federations,only Canada has seen a long-term drift away fromthe centre.

The changing balance between tiers of govern-ment is reflected in decisions by constitutionalcourts. Generally, the courts have acceded tocentral initiatives, particularly when justified bynational emergencies. In the rulings of the USSupreme Court, federal law prevailed over statelaw for most of the twentieth century. In Australia,decisions of the High Court have favoured thecentre to the point where some commentatorsregard federalism as sustained more by politicaltradition than by the constitution.

However, in the 1990s a number of courts didmake some efforts to restore at least someautonomy to the states. In the USA a more con-servative Supreme Court used states’ rights tostrike down a number of congressional laws. InUnited States v. Lopez (1995), for example, the jus-tices declared unconstitutional a federal lawbanning possession of a gun within 1,000 feet of aschool. The Court ruled such matters to be a pre-serve of the states. Temporarily, at least, ideologicalfashion had moved back in favour of the states(Goodman, 2001).

Assessing federalism

What conclusions can we reach about the federalexperiment? An overall assessment is particularlyrelevant in the twenty-first century as governmentsseek a way of enabling different ethnic groups tolive together in the same territory.

The case for federalism is that it offers a naturaland practical arrangement for organizing largestates. It provides checks and balances on a territo-rial basis, keeps some government functions closerto the people and allows for the representation ofethnic differences. Federalism reduces overload in

the national executive, preventing the over-central-ized character of some unitary states such asBritain and France.

In addition, the existence of multiple provincesproduces healthy competition and opportunitiesfor experiment. As the American Supreme CourtJustice Brandeis wrote in New State Ice Co. v.Liebmann (1932), ‘a single courageous state may .. . try novel social and economic experimentswithout risk to the rest of the country’. Citizensand firms also have the luxury of choice: if theydislike governance in one state, they can alwaysmove to another.

Above all, federalism reconciles two modernimperatives: it secures the economic and militaryadvantages of scale while retaining, indeed encour-aging, cultural diversity. For such reasons, forminga federation can certainly be considered a realisticand tested option for diverse and often dividedcountries. As van Deth (2000, p. xv) comments,‘the rewards of federalism are evident when thealternatives of secession, oppression and war areconsidered’.

But a case can also be mounted against feder-alism. Compared to unitary government, decision-making in a federation is complicated andslow-moving. When a gunman ran amok inTasmania in 1996, killing 35 people, federalAustralia experienced some political problemsbefore it tightened gun control uniformly across thecountry. By contrast, unitary Britain acted speedilywhen a comparable incident occurred in the sameyear at a primary school in Dunblane, Scotland.When the American president moved quickly to setup a new Department of Homeland Security afterSeptember 11, 2001, one of his motives was pre-cisely to improve coordination between tiers of gov-ernment in the American federation. It may well bethat some of the security failings revealed on 9/11will eventually be laid at federalism’s door.

Federal policy-making is focused on compro-mise rather than problem-solving. Federations areat risk from what Scharpf (1988) terms the jointdecision trap: the tendency for the quality – andeven the number – of decisions to decline as thenumber of interests increases. Why, for instance, isGermany’s new high-speed rail service betweenFrankfurt and Cologne slowed by a stop in thetiny town of Montabaur (population 13,000)?

FEDERAL, UNITARY AND LOCAL GOVERNMENT 235

Because otherwise there would be no stops inRhineland-Palatinate, a province through whichthe line passes.

Further, federalism complicates accountability.Negotiations between levels take place away fromthe public gaze, bypassing the legislature and pro-viding opportunities for politicians to pass thebuck to the other level. Also, federalism distributespower by territory when the key conflicts insociety are social (e.g. race and gender) rather thanspatial. Federations empower neither the nationalmajority nor the main minorities. Today, free tradeareas are a simpler method of securing the eco-nomic advantages of scale.

Fundamentally, though, any final judgement offederalism must depend on taking a view of theproper balance between the concentration and dif-fusion of political power. Should power rest withone body to allow decisive action? If so, federalismis likely to be seen as an obstacle and impediment,as an anti-democratic device. Alternatively, shouldpower be dispersed so as to reduce the danger ofmajority dictatorship? Through this lens, feder-alism will appear as an indispensable defence ofliberty.

Unitary government

Most states are unitary, meaning that sovereigntylies exclusively with the central government. In

this hierarchical arrangement, the national govern-ment possesses the theoretical authority to abolishlower levels. Subnational administrations, whetherregional or local, may make policy as well asimplement it but they do so by leave of the centre.

Unlike federations, a unitary framework is notalways a deliberate creation; rather, such systemsemerge naturally in societies with a history of ruleby sovereign monarchs and emperors, such asBritain, France and Japan. Unitary structures arealso the norm in smaller democracies, particularlythose without strong ethnic divisions, as inScandinavia. In Latin America, nearly all thesmaller countries (but none of the largest ones) areunitary. In most unitary states, the legislature hasonly one chamber since there is no need for astates’ house.

After the complexities of federalism, a unitarystructure may seem straightforward and efficient.However, the location of sovereignty is rarely anadequate guide to political realities; unitary gov-ernment is often decentralized in its operation.Indeed in the 1990s many unitary statesattempted to push responsibility for more func-tions on to lower levels. In practice, unitary states,just like federations, involve the detailed bar-gaining characteristic of all multilevel governance.

We can distinguish three broad ways in whichunitary states can disperse power from the centre:deconcentration, decentralization and devolution(see Box 13.3). The first of these, deconcentration.

236 GOVERNMENT AND POLICY

Method Definition Example

Deconcentration Central government functions are Almost 90 per cent of US federal civilian employees executed by staff ‘in the field’ work away from Washington, DC

Decentralization Central government functions are Local governments administer national welfare executed by subnational authorities programmes in Scandinavia

Devolution Central government grants some Regional governments in France, Italy and Spain decision-making autonomy to new lower levels

Note: Deconcentration and decentralization occur in federal as well as unitary states.

B OX 1 3 . 3

Methods for distributing power away from the centre

FEDERAL, UNITARY AND LOCAL GOVERNMENT 237

Our understanding of federalism can be usefullytested against the challenging case of the EuropeanUnion (EU). Leaving aside opinion on whether the EUought to develop as a federation, to what extent canthis singular entity already be regarded as a federa-tion? Have those sceptics, particularly in Britain, whowould prefer the EU to be merely an intergovern-mental organization already been left behind?

The case for Shared sovereignty is the core feature of federalismand by this test the Union already possesses a broadlyfederal character. A comparison with the USA demon-strates the point. Like the United States, the EuropeanUnion has a strong legal basis in the treaties on whichit is founded. In addition, plans for a formal constitu-tion are in hand.The influential European Court ofJustice, like America’s Supreme Court, adjudicates dis-putes between levels of government. In both cases,court decisions apply directly to citizens and must beimplemented by member governments. In neither theEuropean Union nor the USA is there any provision formember states to withdraw. Further, citizens of theEuropean Union, like those of the USA, are free toreside anywhere within the Union’s territory.

As with the American national government, the Unionhas specific policy responsibilities, notably for thesingle market. Externally, the EU is represented at inter-national bodies such as the United Nations and itmaintains over 130 diplomatic missions. In short, inEurope – as in America – sovereignty is already pooled.

The case against To describe the EU as a federation would be to exag-gerate its cohesion. Again, comparisons with theAmerican exemplar are helpful. In contrast to the USA,the EU does not possess a common currency. Swedenvoted against the euro in a referendum in 2003 andneither Denmark nor the UK has even put the matterto a vote.Where the USA was designed as a federationfrom the start, the EU has been built on agreeing con-crete policies, particularly for a single market, with afederation as a long-term aspiration.Where theAmerican federal government now levies nationaltaxes, the EU does not tax its citizens directly.The USA

fought a civil war to preserve its union; the EU has notfaced this vital test nor, without its own military force,is it in any position to do so. As Henry Kissinger, formerAmerican Secretary of State, is reported to have said,‘When I want to speak to Europe, who do I tele-phone?’

In world politics, America is a dominant force; withinthe EU, foreign and defence policy, including controlover military force, remains the preserve of memberstates.The United Kingdom took part in the Iraq Warof 2003 but most other member states kept theirtroops at home. Surely we would not describe theUSA as a federation if each state decided separatelywhether to go to war? In addition, states within theEuropean Union – unlike American states – stillpossess such important markers of external sover-eignty as membership of the United Nations.

Assessment There is, in principle, no reason why all federationsshould share the same division of functions betweenlevels. A unique federation is a federation still. In theeighteenth century, the USA was still described in theplural – these United States – yet no one would ques-tion that those same states had formed a federation(Forsyth, 1996). In a similar way, the EU represents apooling of some of its members’ sovereignty and con-sists of institutions – including a court, a parliamentand a form of executive – through which sovereigntyis exercised. In these respects, the Union is more thana confederation yet, as in the early decades of theUnited States, its members retain much of their tradi-tional autonomy.We can conclude with Burgess(2003, p. 73) that

there is no precedent for the creation of a multina-tional federation composed of 15 or more nationstates, with mature, social, economic political andlegal systems. In this regard the EU is a colossal andoriginal enterprise.

Further reading: Bednar, Eskridge and Ferejohn (2001),Burgess (2000), McKay (2001a), Nicolaidis and Howse (2001).

See also: EU profile (p. 176).

DEBATE

A FEDERAL EUROPE?

is purely an administrative matter, denoting thelocation of many central government employeesaway from the capital. The case for a deconcen-trated structure is that it spreads the work around,reduces costs and frees central departments tofocus on policy-making. There is, after all, noreason in principle why such functions as taxationor public pensions should be administered fromthe capital. Indeed, field offices of central depart-ments can benefit from local knowledge. As aFrench decree of 1852 pointed out, ‘one cangovern from afar but must be close to administer’.

The second, and politically more significant,way of dispersing power is through decentraliza-tion. Here, policy execution is delegated to subna-tional bodies such as local authorities or a range ofother agencies. In Scandinavia, for instance, localgovernments have put into effect many welfareprogrammes agreed at national level. In a similarway, local government in the UK serves as theworkhorse of central authority.

The third and most radical form of power dis-persal is devolution. This occurs when the centregrants decision-making autonomy (includingsome legislative powers) to lower levels. In theUnited Kingdom, devolved assemblies were intro-duced in Scotland and Wales in 1999. But thecontrast with federations remains. Britain remainsa unitary state because these new assembles couldbe abolished by Westminster through normal leg-islation. As the English politician Enoch Powellsaid, ‘power devolved is power retained’.

Spain is another example of a unitary state withextensive devolution. Spanish regions were createdin the transition to democracy following GeneralFranco’s death in 1975. Seeking to integrate a cen-tralist tradition with strong regional identities,

Spain’s constitution-makers created a complexsystem in which regions could aspire to varyingdegrees of autonomy. The historic communities ofthe Basque Country and Catalonia quickly pro-ceeded to the most autonomous level, with otherregions offered the prospect of a later upgrade.The result was asymmetric devolution within theframework of what is still, in theory, a unitarystate.

A distinctive trend within unitary states has beenthe development of at least one level of governmentstanding between central and local authorities.France, Italy and Spain have introduced electedregional governments. The smaller Scandinaviancountries took a different route, strengthening andrefurbishing their traditional counties. Even a smallcountry like New Zealand has developed 12 electedregional councils. In unitary states, the standardpattern now is to have three levels of subnationalgovernment – regional, provincial and local – as inFrance and Italy (Table 13.2). The result is a multi-level system that is even more intricate than thetwo levels of subnational authority – state and local– in a federation. Whether federal or unitary, mul-tilevel governance has become a refrain repeated inall established democracies.

The European Union has encouraged theregional level within its member states. TheEuropean Regional Development Fund, estab-lished in 1975, stimulated regions to lobby for aidby distributing aid directly to regions, rather thanthrough central governments. This squeeze on thecentre prompted some regional nationalists to pos-tulate a Europe of the Regions in which the EUand revived regions would exert a pincer move-ment on national power. The notion, somewhatexaggerated but significant in itself, was that the

238 GOVERNMENT AND POLICY

Table 13.2 Subnational government in unitary states: some European examples

France Italy The Netherlands Norway

Regional level 22 regions 20 regions – –Provincial level 96 départements 94 provinces 12 provinces 19 countiesLocal level 36,565 communes 8,074 communes 496 municipalities 448 municipalities

Note: Figures are from the late 1990s to the early 2000s.

European Union and the regions would graduallybecome the leading policy-makers, outflankingcentral governments which would be left with lessto do. In 1988, the EU further developed suchaspirations by introducing a Committee of theRegions and Local Communities (CoR), a bodycomposed of subnational authorities. However,the CoR has proved to be merely consultative.Federal Belgium apart, national executives haveshown no sign of withering away in the manneroriginally anticipated by regional enthusiasts(Bourne, 2003).

Although regional governments can pass laws intheir designated areas of competence, their maincontribution has been in economic planning andinfrastructure development. Such tasks are beyondthe scope of small local authorities but beneath thenational vision of the centre. The Spanish regionof Valencia, for example, has sought to improvetelecommunications, roads, railways, ports andairports. In Italy, too, regional authorities – at leastoutside the south – have made a notable contribu-tion, with some left-wing parties determined todisplay their competence through showpiece gov-ernance.

So the evolution of regional government sug-gests a move away from the original demand forgreater self-government toward a more administra-tive role, standing between other levels of govern-ment. In a system of multilevel governance,regions are indeed finding their level. As Balme(1998, p. 182) concludes of France, regional gov-ernments in Europe are becoming arenas in whichpolicies are formed, even if they seem unlikely tobecome decisive actors in their own right.

Local government

Local government is universal, found in federaland unitary states alike. It is the lowest level ofelected territorial organization within the state.Variously called communes, municipalities orparishes, local government is ‘where the day-to-day activity of politics and government gets done’(Teune, 1995b, p. 16). For example, September11, 2001 was certainly a global event but it wasNew York City that faced the most immediateconsequences.

At their best, local governments express thevirtues of limited scale. They can represent naturalcommunities, remain accessible to their citizens,reinforce local identities, offer a practical educa-tion in politics, provide a recruiting ground tohigher posts, serve as a first port of call for citizensand distribute resources in the light of specialistknowledge. Yet local governments also have char-acteristic weaknesses. They are often too small todeliver services efficiently, they lack financialautonomy and they are easily dominated by tradi-tional elites.

Thus the perpetual problem of this level of gov-ernment: how to marry local representation withefficient delivery of services. As Teune (1995a, p.8) notes,

there never has been a sound theoretical resolu-tion [at local level] to the question of democracyand its necessity for familiarity, on the one hand,with the size and diversity required for pros-perity, on the other.

The balance struck between intimacy and effi-ciency varies over time. In the 1980s, as nationalgovernments came under financial pressure, localauthorities were encouraged to become more effi-cient and customer-led, particularly in theEnglish-speaking world. But as the twenty-firstcentury got under way, so signs began to emergeof a rebirth of interest in citizen involvement,stimulated by declining turnout at local elections.In New Zealand, for instance, successful manage-rial reforms introduced in 1989 were followedthirteen years later by the Local Government Act2002. This law outlined a more expansive – andpossibly expensive – participatory vision for thecountry’s territorial local authorities.

The status of local government varies acrosscountries. As a rule, the position of local govern-ment is stronger in (1) European rather than non-European democracies, (2) Northern rather thanSouthern Europe and (3) countries where localgovernment can represent its community morebroadly than through merely providing designatedfunctions to local people (Box 13.4).

Consider, first, the contrast between Europeanand New World democracies. In most of Europe,local authorities represent historic communities

FEDERAL, UNITARY AND LOCAL GOVERNMENT 239

that pre-date the emergence of strong nationalgovernments. In addition, the unitary nature ofmost European states allows direct links to formbetween powerful local administrations (especiallylarge cities) and the centre. Reflecting this status,European constitutions normally mandate someform of local self-government. Sweden’sInstrument of Government, for instance, roundlydeclares that Swedish democracy ‘shall be realizedthrough a representative and parliamentary polityand through local self-government’.

In the New World, by contrast, local govern-ment reveals a more pragmatic, utilitarian char-acter. Local authorities were set up as needed todeal with ‘roads, rates and rubbish’. Special boards(appointed or elected) were added to deal withspecific problems such as mosquito control, har-bours and land drainage. The policy style was apo-litical: ‘there is no Democratic or Republican wayto collect garbage’. Indeed special boards wereoften set up precisely to be independent of partypolitics. Because Australia, Canada and the USAare federations, local government is the preserve ofthe states, often creating further diversity in orga-nization. For instance, the USA is governed atlocal level by a smorgasbord of over 80,000 cities,counties, school districts, townships and specialdistricts.

Second, the standing of local government varieswithin Europe, on a broad north–south axis. Inthe Northern countries, local authorities becameimportant delivery vehicles for the extensivewelfare states that matured after 1945. Servicessuch as social assistance, unemployment benefit,childcare and education are funded by the statebut provided locally, giving municipal authoritiesa significant role as the government’s front office.Also, a process of amalgamation lasting until thefinal quarter of the twentieth century led to a largeincrease in populations served by local councils,notably in the United Kingdom. This expansionallowed more functions to be provided by eachauthority.

In Southern Europe, by contrast, welfare statesare less extensive, with the Catholic Churchplaying a greater role in providing care, whilepublic services such as education remain under thedirect control of the central authority. In Italy, forinstance, teachers are civil servants rather thanemployees of local councils. Also, communes inSouthern Europe have remained small, especiallyin France (Table 13.3). Limited scale precludes anextensive administrative role except when neigh-bouring areas form collaborative syndicates tosupply utilities such as water and energy.

Third, constitutional provisions also affect the

240 GOVERNMENT AND POLICY

Status of local government Interpretation

Higher Lower

In European democracies In new world democracies In Europe, local governments often represent (e.g. Australia, USA) historic communities but in the new world, local

government is more utilitarian in character

In Northern Europe In Southern Europe Local governments administer the extensive welfare (e.g. Scandinavia) (e.g. France, Italy) states found in Northern Europe but perform fewer

functions in Southern Europe

When local governments When local governments General competence allows local authorities to take possess general cannot act ultra vires the initiative whereas ultra vires restricts them to competence to represent (‘beyond the powers’) designated functionstheir community

Sources: Adapted from Goldsmith (1996), John (2001) and Page and Goldsmith (1987).

B OX 1 3 . 4

Exploring the status of local government

position of local authorities. In continentalEurope, local governments often enjoy generalcompetence: that is, the authority to make regula-tions in any matter of concern to the area.Germany’s Basic Law, for instance, gives localcommunities ‘the right to regulate [all localmatters] under their own responsibility and withinthe limits of the law’.

But in other countries, including the UnitedKingdom, councils could traditionally onlyperform those tasks expressly designated by thecentre; any other act would be ultra vires (beyondthe powers). In the United States, Dillon’s rulesimilarly restricts local governments to tasks dele-gated by their particular state. Some countrieswhere ultra vires applies, including the UnitedKingdom and New Zealand, did establish a moreliberal legal framework at the start of the twenty-first century but without granting the full powerof general competence to local areas.

Three additional features of local governmentmerit consideration: internal structure, functionsand relationships with the centre.

Structure

The structure of local government has recentlyattracted attention, largely in an attempt to makedecision-making clearer to local electorates in anera of falling turnout. Just as political parties havesought to reverse a decline in membership bygiving their supporters more say in the selection ofcandidates and leaders, so local governments insuch countries as Italy, the Netherlands and theUnited Kingdom have experimented with directelection of the mayor.

There are, in fact, three broad ways of orga-nizing local government (Box 13.5). The first andmost traditional method, represented by Englandand Sweden, concentrates authority in a college ofelected councillors. This full council often oper-ates through powerful committees covering themain local services – housing and education, forexample – with professional appointees (such asarchitects and educational administrators) alsoplaying a significant role. The mayor plays amodest ceremonial role. Whatever virtues thisformat may have, its collegiate character presents arather opaque picture to the electorate.

Accordingly, a second method of organizationknown as the mayor–council system has attractedattention. This model is based on a separation ofpowers between a directly elected mayor, who isthe chief executive, and an elected council withlegislative and budget-approving powers. Thiselaborate structure is used in many large Americancities, notably New York, permitting a range ofurban interests to be represented within the frame-work. The mayor and council must negotiate, ifindeed they are not in conflict with each other.

The powers awarded to the mayor and councilvary across localities, defining ‘strong mayor’ and‘weak mayor’ variants of this format. Even beforeSeptember 11, the perceived success of New York’sRudy Giuliani encouraged some European cities,including London, to consider introducing electedmayors as a way of reviving a demoralized council-based system. However, London’s new mayor lacksthe strong powers of his equivalent in New York(Sweating, 2003).

The third structure for local government, againoriginating in the United States, is the council–manager system. Unlike the mayor–council

FEDERAL, UNITARY AND LOCAL GOVERNMENT 241

Table 13.3 Average population of elected localauthorities in some European democracies

Country Average population of local authorities (lowest level)

UK 137,000

Ireland 36,100

Sweden 33,000

Netherlands 32,466

Belgium 17,384

Finland 11,206

Norway 9,000

Germany 7,900

Italy 7,036

Spain 4,997

France 1,491

Source: Updated from John (2001), table 2.1. Figures are from thelate 1990s to the early 2000s, depending on country.

format, this arrangement seeks to depoliticize andsimplify local government by separating politicsfrom administration. This distinction is achievedby appointing a professional city manager, oper-ating under the elected council and mayor, toadminister the authority’s work. This council–manager format emerged early in the twentiethcentury in an attempt to curb corruption; it hasbeen widely adopted in western and south-westernstates in the USA. The model has corporate over-tones, with the voters (shareholders) electingcouncillors (board of directors) to oversee the citymanager (chief executive). However, the distinc-tion between politics and administration is diffi-cult to achieve in practice.

Functions

Leaving structure to one side, what is it that local governments do? Broadly, their tasks are twofold: to provide local public services (such asrefuse collection) and to implement nationalwelfare policies (Box 13.6). However, a staticdescription of functions fails to reveal how the roleof local government has evolved since the 1980s.

The major trend, especially prominent in theEnglish-speaking world and Scandinavia, has been

for municipal authorities to reduce their directprovision of services by delegating tasks to privateorganizations, both profit-making and voluntary.In Denmark, for example, many local govern-ments have set up ‘user boards’ in primary schools.These boards are given block budgets and theauthority to hire and fire staff (Bogason, 1996). In

242 GOVERNMENT AND POLICY

Structure Description Examples

Council Elected councillors operating through committees England (pre-Local Government Act supervise departments headed by professional of 2000), Swedenofficers.The mayor (if any) is largely a ceremonial post

Mayor-council* A mayor elected by the municipal area serves as About half of the 7,000 cities in the chief executive. Councillors elected from local USA, including Chicago and New York wards form a council with legislative and financial authority

Council–manager The mayor and council appoint a professional About 3,000 American cities, including manager to run executive departments Dallas and Phoenix

* Often subdivided into strong mayor and weak mayor systems according to the mayor’s executive autonomy.

Sources: Bowles (1998), Chandler (1993), Mouritzen and Svara (2002).

B OX 1 3 . 5

Structures for local government

Cemeteries Recreation

Economic development Refuse disposal

Environmental protection Roads

Fire service Social housing

Homes for the elderly Tourism

Libraries Water supply

Local planning Welfare provision

Primary education

Source: Adapted from Norton (1991), table 2.2.

B OX 1 3 . 6

Typical powers of localauthorities

a similar way, private firms located in the commer-cial district of some American cities have takenover much of the responsibility for funding andorganizing improvements to local services such asstreet-cleaning.

So the local authority’s role is changing from aprovider to an enabler. The council does not somuch provide services as ensure that they are sup-plied. In theory, the authority can become asmaller, coordinating body, more concerned withgovernance than government. More organizationsbecome involved in local policy-making, many ofthem functional (for example school boards)rather than territorial (for example county coun-cils). This shift represents an important transitionin conceptions of how local governments shouldgo about their task of serving their communities.It also represents some retreat of the service-pro-viding style of local government found inNorthern Europe to the more strategic gover-nance in Southern Europe and the more varie-gated patterns of local authority in the NewWorld.

DefinitionThe enabling authority is a term used to sum-marize one vision of local governance. Such anauthority does not provide many services itself.Rather, its concerns are to coordinate the provi-sion of services and to represent the communityboth within and beyond its territory. An enablingauthority is strategic, contracting out service pro-vision to private agencies, whether voluntary orprofit-making.

Relationships with the centre

How is local government integrated into thenational structure of power? This question is keyto appreciating how multilevel governance oper-ates in practice. The answer reveals contrastingnotions of state authority in the establisheddemocracies.

The relationship between centre and localityusually takes one of two forms: dual or fused. In adual approach, local government is seen as an orga-nization separate from the centre. In a fusedapproach, local and central government are regarded

as comprising a single network of state authority. Consider the dual system first. Here public

authority is seen as separated rather than inte-grated; it is as if there were two spheres ofauthority, connected for practical reasons only.Local governments retain free-standing status,setting their own internal organization andemploying staff on their own conditions of service.Employees tend to move horizontally – from onelocal authority to another – rather than vertically,between central and local government. Ultimateauthority rests with the centre but local govern-ment employees do not regard themselves asworking for the same employer as central civil ser-vants. National politicians rarely emerge fromlocal politics.

Traditionally, Britain was regarded as the bestexample of a dual system. Before the rise of themodern state, local magistrates administered theircommunities. This spirit of self-government sur-vives in a perception that central and local govern-ment are distinct, if intensely interdependent,spheres. Certainly, severe centralization underConservative administrations in the 1980sreduced local authorities to virtual servants ofcentral government. Even so, central and localtiers still do not form a single tier of publicauthority. Their continuing separation contributesto the rather weak notion of the state in Britain.

DefinitionA dual system of local government (as in Britain)maintains a formal separation of central andlocal government. Although the centre is sover-eign, local authorities are not seen as part of asingle state structure. In a fused system (as inFrance), municipalities form part of a uniformsystem of administration applying across thecountry, with a centrally appointed prefectsupervising local councils.

Under a fused system, by contrast, central andlocal government combine to form a single sphereof public authority. Both levels express the leadingauthority of the state. In some European coun-tries, such as Belgium and the Netherlands, themayor is appointed by central government and isresponsible to the centre for maintaining local lawand order. In addition, central and local authority

FEDERAL, UNITARY AND LOCAL GOVERNMENT 243

are fused in the office of the prefect, a centralappointee who oversees the administration of aparticular community and reports to the Ministryof the Interior. In theory, a prefectoral systemsignals central dominance by establishing a clearhierarchy running from national governmentthrough the prefect to local authorities.

France is the classic example of this fusedapproach. The system was established byNapoleon early in the nineteenth century andconsists of 96 départements in France, each with itsown prefect and elected assembly. The frameworkis uniform and rational but in practice the prefectmust cooperate with local councils rather thansimply oversee them. In fact, the prefect is now asmuch an agent of the département as of the centre,representing interests upwards as much as trans-mitting commands downwards. Although thepowers of the prefect have declined, the Frenchmodel remains influential. Many other countrieshave adopted it, including all France’s ex-coloniesand several post-communist states.

One effect of a fused system is the ease withwhich politicians can move between, or evenstraddle, the national and the local. France is againan interesting case. National politicians oftenbecome or remain mayor of their home town, afeature called the cumul des mandats (accumula-tion of offices). Typically, about one in twomembers of parliament will also be a local mayor.The cumul has come under attack because, afterall, multiple posts beget multiple salaries. Aninitial reform in 1985 restricted politicians toholding no more than two major elected posts atthe same time. However, even after the rules weretightened further in 2000, the most popular cumul– combining the office of local mayor with mem-bership of the National Assembly – is still per-mitted (Stevens, 2003, p. 170).

Subnational government in newdemocracies

In new democracies, subnational governmentbegins in an undernourished condition. Lackingresources, and inhibited by the legacy of centraldominance, local governance is initially a matterof doing whatever can be afforded. Only as the

new democracy consolidates, and economicreform brings results, do subnational governmentsbegin to develop more weight and coherence. Tothe extent that a new democracy fails to consoli-date, subnational government may continue in anuncertain position.

One index of democratic consolidation is pre-cisely the development of subnational institutionswith a stable relationship to the national govern-ment. In the actual transition to democracy,however, little attention is paid to regional andlocal levels. An outburst of local participation mayherald the new order but the new activists areoften inexperienced and short-lived amateurs. Thereal action occurs in the capital where the goal isto establish a new order of national power. Onlywhen this primary object has been achieved canthe focus turn to lower levels of governance.Establishing a uniform set of institutions beneaththe national level, with clear functions, soundfinancing and competitive elections, tends to be areflection as much as a dimension of democraticconsolidation.

Consider post-communist states. The strugglesof local government to become establisheddemonstrate the difficult inheritance left by col-lapsed dictatorships. Under communism, regionalparty bosses, not elected councillors, had wieldedpower. As a result, local government was massivelyunder-prepared for the post-communist era. ‘Weare starting from less than zero’, complained a cityofficial in Kiev, the capital of Ukraine, after thecollapse of communist rule there. Local authoritiespossessed little revenue while central governmenthad no money to spare. Often local governmentofficers worked from buildings that were still tech-nically owned by their former masters in theCommunist Party.

Despite chronic under-resourcing, local officialshad to confront the enormous welfare problemscreated by industrial collapse. They sought toresolve these difficulties in the context of societieswhich had gone without authentic local socialorganization for at least a generation. At locallevel, the organizational difficulties and humanconsequences of regime change became manifest.Millions of people confronted a diminishedquality of life as communist welfare provisioncame to an abrupt halt. The policy problems were

244 GOVERNMENT AND POLICY

national but the human consequences were local. In countries with a less severe history of com-

munism (such as Hungary and Poland), subna-tional government has now taken root, usuallydrawing on the pre-communist heritage. TakePoland as an illustration. In 1999 – a full ten yearsafter the collapse of communism – Poland reintro-duced its pre-communist subnational structure,with 16 voivodships (regions) given responsibilityfor economic development within their territory.The leader of each voivodship was no longerappointed by the centre but was drawn from anelected council. As with other Eastern Europeancountries, Poland’s focus on the regional level wasstimulated by the desire to join, and extractresources from, the European Union. Beneath theregions in Poland stand two reinvigorated levels:counties and communes. So Poland has finallymoved to the three-level subnational structurefamiliar in the West. However, many other post-communist countries – especially those remotefrom the EU – are still administered in a highlycentralized and authoritarian fashion, leaving littlescope for coherent governance at local level.

Some of the difficulties experienced in breathinglife into subnational government in post-commu-nist states are echoed in post-military LatinAmerica. There, a tradition of centralized personalpower persists into the democratic era, holdingback efforts to raise the status of elected local offi-cials. Mexico, for instance, is a federal and sub-stantially democratic state at national level but thepolitical legacy of centralized control by the PRIand the traditional sway of rural landowners stillprevent any sensible comparison with subnationalgovernance in North America.

Even where federalism is more authentic, as inBrazil, excessive spending at state level has causedconsiderable difficulties for national governmentsseeking to restore their credibility with interna-tional financial institutions. Where fundamentalproblems arising from economic mismanagementstill need to be resolved, there is a case forretaining the financial levers at national level.Argentine, for instance, suffered in negotiationswith the International Monetary Fund in 2002from excessive spending by the governor ofBuenos Aires, by far the largest province in thecountry. It is for these reasons that Domínguez

and Giraldo (1996, p. 27) question the desirabilityof decentralization in contemporary LatinAmerica:

Subnational governments can undermine theefforts of national governments to carry out eco-nomic reforms. Decentralization can also under-mine democratization by reinforcing the powerof local elites, their practices of clientelism andthe power of their military or paramilitary allies.Decentralization may some day empower ordi-nary citizens to take better charge of their gov-ernment, and permit a wider range of innovationat the local level, but there is still a long way togo before these promises are realized.

Subnational government in authoritarian states

In authoritarian and totalitarian regimes, the dis-tinction between local government and localpower is fundamental. The former is weak:authority flows down from the top and so bottom-up institutions of representation are subordinate.Where national power is exercised by the militaryor a ruling party, these institutions typically estab-lish a parallel presence in the provinces, wheretheir authority overrides that of formal state offi-cials. For a humble mayor in such a situation, themain skill required is to lie low and to avoidoffending the real power-holders.

But it would be wrong to conclude that authori-tarian regimes are highly centralized. Rather,central rulers – just like medieval monarchs – oftendepend on established provincial leaders to sustaintheir own, sometimes tenuous, grip on power.Central–local relations therefore tend to be morepersonal and less structured than in an establisheddemocracy. The hold of regional strongmen onpower is not embedded in local institutions; suchrulers command their fiefdoms in a personalfashion, replicating the authoritarian pattern foundat national level. Central and local rulers are inte-grated by patronage: the national ruler effectivelybuys the support of local bigwigs who in turnmaintain their position by selectively distributingresources to their own supporters. Patronage, notinstitutions, binds centre and periphery.

FEDERAL, UNITARY AND LOCAL GOVERNMENT 245

As always, however, totalitarian regimes offeredmore sophisticated excuses for a centralized order.Fascism, for example, sought unqualified obedi-ence to the state and its supreme leader. Unity inthe state implied a unitary state; dividing powerbetween multiple levels of government was dis-missed as liberal folly. In Italy, Mussolini relied oncentrally appointed prefects to run local areas,decreeing that the prefect represented the ‘entire,undiminished power of the state’. Spain underFranco proceeded similarly, with a civil governorin each province overseeing an elaborate structureof government outposts. These were fused systemspar excellence.

Given that the supreme leader could not in facttake all decisions, in practice the local representa-tives of the state exerted considerable influenceover their own areas. They sought to become theirown little dictators and they often succeeded. Soin fascist regimes, as in authoritarian ones, thedoctrine of central power resulted in a systemwhere its local exercise went unchecked.

It was perhaps only some communist regimes,most notably the Soviet Union, which achievedreal political centralization. In communist states,the leading role of the party in constructing asocialist utopia always took precedence over localconcerns. As long as the party itself remainedhighly centralized, national leaders couldcommand outlying areas through the party.

In the Soviet Union, the ability of communistleaders to command the vast territory of thecountry through the device of the party was anunparalleled organizational achievement.Technically, the ‘Union of Soviet SocialistRepublics’ was a federation but in reality anyattempt by a republic to exercise its constitutionalright to ‘freely secede from the USSR’ would havebeen crushed by force. Centralized control exer-

cised through the party overwhelmed federalistfiction. Although a measure of decentralizationwas introduced from the 1960s, Soviet federalismunder communist rule was dismissed by mostobservers as ‘false’, ‘façade’ or ‘incomplete’ – acover for what was in effect a Russian empire.

Even though many of the USSR’s constituentrepublics became independent states in the break-up of the Soviet Union in 1991, Russia itselfremains a country in which multiple nationalitiescontinue to jostle under authoritarian rule fromthe centre. A brutal example is Chechnya’s failedwar of independence, savagely repressed by aRussian elite (and public) determined to protectits own preeminence in a multinational state.Russian federalism is no longer a façade but it stillcoexists uneasily with ‘the dictatorship of power’(Stoliarov, 2002).

Key reading

Next step: Gagnon and Tully (2001) is a compar-ative study of multinational democracies, focusingon federalism and federations.

On federalism, useful comparative studies areStepan (2001), McKay (2001) and G. Smith(1995). Elazar (1996) provides an enthusiast’soverview while Wachendorfer-Schmidt (2000) andBraun (2000) compare the performance of federaland unitary states. Burgess (2000, 2003) surveysthe EU’s ‘federation’, Bakvis and Skotsgad (2002)examine the Canadian case and Kahn (2002)looks at Russia. Loughlin (2001) reviews subna-tional democracy in the European Union. John(2001) and Mouritzen and Svara (2002) surveylocal governance; see also the special issue of theInternational Political Science Review (1998).

246 GOVERNMENT AND POLICY

Structure 248Functions 252Recruitment 259Legislatures in new democracies 263Legislatures in authoritarian states 264Key reading 267

Legislatures are symbols of popular representa-tion in politics. They are not governing bodies,

they do not take major decisions and usually theydo not even initiate proposals for laws. Yet they arestill the foundation of both liberal and democraticpolitics. How then does their significance arise?The importance of legislatures rests on what theystand for rather than what they do. As Olson(1994, p. 1) writes,

legislatures join society to the legal structure ofauthority in the state. Legislatures are represen-tative bodies: they reflect the sentiments andopinions of the citizens.

In potential and often in reality, legislatures arethe authentic representative of the people’s will.For this reason they help to mobilize consent forthe system of rule. As representative democracyspreads throughout the world, so more legislaturesare gaining the political weight which comes fromperforming the role of standing for the people.

The origin of European parliaments lies inancient royal courts where monarchs would judgeimportant legal cases and meet with noblemen ofthe realm (Marongiu, 1968). Gradually theseassemblies became more settled, coming to repre-sent the various estates – the clergy, the nobilityand the towns – into which society was thendivided. In the thirteenth and fourteenth cen-turies, kings began to consult estate leaders moreconsistently on issues of war, administration,commerce and taxation. Myers (1975, p. 23)describes how these early ‘assemblies of estates’developed:

the leading members [of estates] might appear inthe assemblies either by virtue of their office orstatus, or because of election. At first the com-position and functions of such assemblies werevery ill defined and fluid, but gradually theysolidified into increasingly definite forms which,in a traditionally-minded society, came to beregarded as customary and therefore respected.

Although the initiative for calling these colloquialay with the king, a principle of Roman law wassometimes invoked in justification. This notionwas quod omnes similter tangit, ab omnibus compro-betur (what concerns all should be approved byall). So these early European assemblies wereviewed as possessing a right to be consulted longbefore they became modern legislatures with thesovereign authority to pass laws.

DefinitionA legislature is a multimember representativebody which considers public issues. Its mainfunction is to ‘give assent, on behalf of a politicalcommunity that extends beyond the executiveauthority, to binding measures of public policy’(Norton, 1990a, p. 1).The words used to denotethese bodies reflect contrasting origins:‘assem-blies’ assemble,‘congresses’ congregate,‘parlia-ments’ talk and ‘legislatures’ pass laws.We usethese terms interchangeably.

The history of legislatures in new republics such asthe United States is more straightforward. In theUSA, the Founding Fathers placed Congress at thecentre of their deliberations. A leading role for thelegislature was judged an essential defence againstexecutive tyranny; the list of powers which the con-stitution awards to Congress is longer and moredetailed than that given to the president. JamesMadison, an architect of the American constitution,declared that ‘in republican government, the legisla-tive power necessarily predominates’ (Hamilton,

Chapter 14

Legislatures

247

1788d, p. 265). And so, in the United States, it did.Congress remains the most influential legislature inthe world.

Structure

Only two things, writes Blondel (1973), can be saidwith certainty about every assembly in the world:how many members and chambers it has. Both areimportant aspects of parliamentary structure. But athird factor, the committee system, is an increas-ingly significant influence on the operation andeffectiveness of modern parliaments. In this section,we examine all three aspects of assembly structure.

Size

The size of an assembly, as indicated by member-ship of the more important lower chamber, reflectsa country’s population (Figure 14.1). In China,the world’s most populous country, the NationalPeople’s Congress has almost 3,000 members. Bycontrast, in the South Pacific island of Tuvalu(population 8,624), the assembly has a mere 12members (Taagepera and Recchia, 2002).

However, with legislatures – unlike countries –size rarely indicates strength. Rather, the oppositeapplies: very large chambers are rendered impotentby their inability to act as a cohesive body. Theyare in constant danger of being taken over bymore coherent actors such as political parties oreven by their own committees. Ruling communistparties, as in China, preferred a large legislatureprecisely because of its tendency to docility.

By contrast, a very small chamber – say, under100 – offers opportunities for all deputies to havetheir say in a collegial environment. A smallchamber may be entirely appropriate for the smallisland communities in the Pacific and theCaribbean where such legislatures abound. Inpractice, as Figure 14.1 shows, few lower chambersare more than 500–600 in size, and this is prob-ably a fair estimate of the maximum size for aneffective body.

Number of chambers

Should a legislature have one chamber or two? If

two, what role should the second chamber playand how should its members be selected? Theseold questions acquired practical significance in the1990s as a new wave of democratic constitutionsmoved parliaments closer to the centre of thepolitical stage.

Unicameralism is the norm today. By 2001,115 of the world’s 178 parliaments possessed onlyone chamber (Russell, 2001). This proportionrose in the decades following 1945 as severalsmaller democracies abolished their secondchamber, notably New Zealand in 1950,Denmark in 1954 and Sweden in 1970. Manysmaller post-colonial and post-communist statesalso embraced a single chamber. Bicameral legisla-tures are most often found in larger countries,particularly federations where the secondchamber expresses the voice of the componentstates (Massicotte, 2001).

248 GOVERNMENT AND POLICY

700

600

500

400

300

200

100

0

Figure 14.1 Population and assembly size

Note: Based on a sample of 70 states drawn from Derbyshire andDerbyshire (1999b), excluding countries with a population in excessof 60 million. Size of assembly is measured by the lower chamber forbicameral assemblies.

10 20 30 40 50 60

Population (million)

Size ofassembly

•

••••••••

•••

•

•••••

•••

•

•

••••••

••• •

•••

•••• ••

• ••

•

•••

•

• ••

•

•

•

•

•••

•

•••••

DefinitionAlthough some European assemblies originallycontained multiple chambers, one for each of thefeudal estates, most parliaments today are eitherunicameral (one chamber) or bicameral (twochambers). In bicameral legislatures, the first orlower chamber is typically called the chamber ofdeputies, national assembly or house of repre-sentatives.The second or upper chamber isusually known as the senate.The lower chamberis popularly elected; upper chambers vary in howtheir members are selected.

The choice between one and two chambers is notjust a technical matter of institutional design.Fundamentally, the decision reflects contrastingvisions of democracy. Unicameral parliaments arejustified by an appeal to a majoritarian reading ofpopular control. The proposition is that anassembly based on direct popular election reflectsthe popular will and should not be obstructed by asecond chamber. The radical French cleric AbbéSièyes (1748–1836) put the point well: ‘if a secondchamber dissents from the first, it is mischievous;and if it agrees, it is superfluous’ (Lively, 1991).

In addition to this traditional argument,modern analysts argue that a single chamber pre-cludes the petty politicking and point-scoringwhich become possible as soon as two houses arecreated (Tsebelis and Money, 1997). A singlechamber is also accountable, economical andcapable, in theory, of acting with despatch.

But the defenders of bicameral parliaments rejectboth the majoritarian logic of Abbé Sièyes and thepenny-pinching of accountants. Bicameralists stressthe liberal element of democracy, arguing that theupper chamber offers checks and balances. It candefend individual and group interests against apotentially oppressive majority in the lower house.The second chamber can also serve as a house ofreview, revising bills (proposed laws), scrutinisingconstitutional amendments and delaying intem-perate legislation – ‘a second chamber for secondthoughts’. In addition, a second house can sharethe workload of the lower chamber, conductdetailed committee work and assist with appoint-ments (e.g. to the judiciary). An upper chamberprovides a modern approximation to the tradi-tional idea of a council of elders. Reflecting these

points, Robert Cecil, thrice British prime ministerin the nineteenth century, declared that the Houseof Lords ‘represents the permanent, as opposed tothe passing, feelings of the nation’ (Russell, 2001).

Where legislatures do consist of two chambers,the question arises of the relationship betweenthem. Usually, the lower chamber dominates. Thisis especially so where the form of government isparliamentary. In such a system, the government’ssurvival depends on maintaining the assembly’ssupport and one chamber must become the focusof such accountability. If a government wereequally accountable to two houses, it might becaught in the grip of contradictory pressures,unable to command the confidence of one orother chamber. This task of sustaining or votingdown the government falls naturally to the lowerhouse, with its popular mandate. The dominanceof the first chamber can also be seen in other ways.It often has special responsibility for money bills;is the forum where major proposals are initiallyintroduced; and is entitled to override vetoes oramendments proffered by the second chamber.

In presidential systems, by contrast, the presidentis directly elected and his continuation in officedoes not depend on the legislature’s confidence.This forestalls any requirement for accountabilityto focus on a single chamber. In these conditions,the upper chamber may acquire more significance.The American Congress is the best illustration.With its constitutional position as representative ofthe states, the Senate plays a full part in thecountry’s legislative and budget-making process,with most bills ending up in a joint committee ofboth houses. Except for Italy, such strong secondchambers are confined to federations.

A bicameral structure raises the question of howthe members of the second chamber should bechosen. Some divergence with the lower house isneeded if the upper chamber is not simply tomirror the party balance in the first chamber. Thethree main methods used for the second chamberare (Russell, 2001):

� direct election (used by 27 of 66 upper houses) � indirect election through regional or local gov-

ernments (21/66) � appointment, usually by the government

(16/66).

LEGISLATURES 249

But even when election is the dominant prin-ciple, a contrast with the lower house is normallyachieved. Generally, the members of the upperhouse are given longer tenure: typically six yearscompared to three to five in the lower chamber.These differences in tenure ensure that elections tothe two chambers do not completely overlap, thusencouraging some variation in composition. Forexample, elections to the American Senate are heldevery two years, with only a third of the 100 seatsup for election at any one time. By contrast, theentire membership of the House ofRepresentatives must stand for reelection on anunusually short two-year cycle.

A federal structure also produces natural diver-gence between chambers. In federations, Canadaexcepted, elections to the upper chamber are orga-nized by state, with the number of senators perstate varying far less than population figureswould imply. The American Senate contains twomembers for each of 50 states; California with apopulation of 35 million has the same representa-tion as Nevada with only two million residents. In

more recent federations, membership is alsoweighted towards the smaller states but to a lesserextent. The German Bundesrat, for instance, offersthree to six seats to each Land, according to popu-lation. By contrast, the districts used for lowerhouse elections are in most countries more equalin population.

Committees

The final structural feature of legislatures to con-sider is its internal committee system. Given thecomplexity of modern politics, a powerfulassembly needs a well-developed committee struc-ture if it is to develop the detailed expertise neededfor real influence. Committees have become theworkhorses of effective legislatures. Yet committeeoperations lack the profile accorded to meetings ofthe whole chamber. So what exactly are legislativecommittees? And what purposes do they serve?

Committees are small workgroups of members,created to cope with the volume of parliamentarybusiness, particularly in the larger and busier lower

250 GOVERNMENT AND POLICY

Table 14.1 Selection to the upper chamber in some established democracies

Country Chamber Members Term (years) Method of selection

Australia Senate 76 6 Direct election by STV in each state†

Canada Senate 104 Retire by 75 Appointed by the prime minister

France Senate 321 9 Indirect election via départements‡

Germany Bundesrat 69 – Appointed by state governments

India Council of States 245 6 Indirect election via state assemblies+

Italy Senate 326 5 Direct election by a mixed member system*

Japan House of Councillors 252 6 Direct election by a mixed member system

Netherlands ‘First chamber’ 75 4 Elected by and from provincial councils(Senate)

USA Senate 100 6 Direct election by plurality voting in each state

Notes:† STV – single transferable vote (p. 148).‡ units of local government (100 in total, including 4 overseas territories).+ except for 12 presidential nominees.* elections are simultaneous with the lower house.The president can appoint 5 life members.

chamber. These workgroups traditionally took theform of standing committees (to consider bills andfinancial proposals) together with conference ormediation committees to reconcile differences inbills passed by the two houses of a bicameralassembly.

However, the more recent development has beenthe strengthening of select committees which scru-tinize government administration and investigatematters of concern, drawing on external witnessesas needed (Box 14.1). The growing status of selectcommittees reflects the increasing importance ofthe scrutiny function to contemporary legislatures.

A strong committee system largely defines aworking (committee-oriented) as opposed to atalking (chamber-oriented) assembly. Debate inthe chamber may fulfil parliament’s representativerole but it is in its committees that the legislaturegets down to the detailed business of scrutinizingbills, overseeing government and exploring policyoptions. Although members are usually allocatedto a committee in proportion to overall partystrength in the chamber, partisanship is usuallyheld in check.

DefinitionIn a talking assembly, such as the British Houseof Commons, floor debate is the central activity;it is here that major issues are addressed andreputations are won and lost. In a workingassembly, such as the American Congress, thecore activity takes place in committee rooms.There, legislators shape bills, authorize expendi-ture and scrutinize the executive.

The American Congress is unique in the impactof its committees. Although unmentioned in theconstitution, committees rapidly became vital toCongress’s work. ‘Congress in session is Congresson public exhibition, whilst Congress in its com-mittee rooms is Congress at work’, wroteWoodrow Wilson (1885, p. 79). Bryce (1921, p.68), a nineteenth-century British observer ofAmerican politics, described the House ofRepresentatives as ‘a huge panel from which com-mittees are selected’; his comment still applies andto both chambers. Committees are uniquely well-supported, employing over 3,500 policy special-ists. They decide the fate and shape of most

legislation. Further, each committee creates itsown subcommittees: by 1994 the Senate’s 20 per-manent standing committees had spawned a totalof 87 subcommittees.

So autonomous did these little legislaturesbecome that they reduced the overall coherence ofCongress. Their chairs became congressional lions,powerful and protective of their own territory. Bythe 1990s, however, party leaders were ready tofight back. They sought to rein in the committees.In the House, Republican Newt Gingrich, electedSpeaker in 1995, asserted party control overappointments to committee chairs and used partytask forces to drive legislation forward. But con-gressional committees are immensely resilient.They largely reasserted their control over bills onceGingrich’s star began to wane. However, onelegacy of the Republican revolution – a six-yearterm limit on committee chairs – may prove to belong-lasting, reducing the iron in the trianglelinking committees, departments and interestgroups (Evans, 2001).

Committees have less influence on legislation inparty-dominated legislatures. In the British Houseof Commons, for instance, government bills areexamined by standing committees which largely

LEGISLATURES 251

Type Function

Standing To consider bills in detail (‘permanent’) committee

Select committee � To scrutinize the executive,often one committee for each main government department

� Ad hoc committees to investigate particular matters of public interest

Conference or A joint committee to reconcile mediation differences in the version of a committee bill passed by each chamber

(bicameral legislatures only)

B OX 1 4 . 1

Types of parliamentary committee

replicate party combat on the floor of thechamber. These committees, unlike those ofCongress, do not challenge executive dominancein framing legislation. They are unpopular, unspe-cialized and under-resourced.

However, like many other legislatures theCommons has expanded its system of select com-mittees of scrutiny. Since 1979, select committeeshave shadowed all the main government depart-ments, probing government policy and moni-toring its implementation. The members of aselect committee can develop a shared outlookwhich moderates the war of the parties still wagedon the floor of the House. In the context of a tra-ditional talking assembly, Norton (1997, p. 166)suggests that Britain’s select committees ‘mark aremarkable advance in terms of parliamentaryscrutiny’.

Where the political style is less adversarial andpolicy emerges through agreement, committees area natural arena in which to seek compromise andconsensus. In these circumstances, influentialcommittees can coexist with strong parties. In theGerman Bundestag (lower house), party disciplineis firm but the committee members have moreregard for objectivity than point-scoring.Germany’s second chamber, the Bundesrat –whose members are appointed by state govern-ments – is even more committee-based. It onlyholds general plenary sessions once a month.

But it is Scandinavia that provides the bestexample of influential committees, again in com-bination with strong parties. Scandinavia’s maingoverning style, sometimes called ‘committee par-liamentarianism’, is one in which influentialstanding committees negotiate the policies andbills on which the whole parliament later votes. InSweden, for instance, committees modify aboutone in three government proposals and nearly halfthese changes are substantial (Sjolin, 1993, p.174). Many such modifications will result fromdiscussion with relevant interest groups. So legisla-tive committees are the vehicle for the extensiveconsultation which defines the character ofScandinavian policy-making.

Apart from the party system, the key to theinfluence of committees lies in three factors:expertise, intimacy and support.

� Expertise emerges over time from committeeswith specialized responsibilities and a clear fieldof operation. Expertise is most likely to developin permanent committees with continuity ofoperation and membership.

� Intimacy emerges from small size and is rein-forced by stable membership. Particularly whenmeetings take place in private, a small groupsetting can encourage cooperation and con-sensus, overcoming any initial hostility betweenmembers from competing parties.

� Support refers to the use of qualified staff toadvise committees. Expert researchers can helpbusy politicians to produce well-founded rec-ommendations.

Significantly, all three factors are present in theAmerican Congress.

Functions

The question of the functions of legislatures raisesthe important issue of their role in the politicalsystem. The key roles of modern parliament arerepresentation, deliberation and legislation. Otherfunctions, crucial to some but not all assemblies,are authorizing expenditure and scrutinizing theexecutive (14.2).

Through a discussion of these functions, we willsee how the significance of parliaments in estab-lished democracies extends well beyond thenarrow task of simply converting bills into laws.

Representation

We have suggested that the essence of assemblies isthat they ‘represent’ the wider society to the gov-ernment. But how can we judge whether, and howwell, that function is fulfilled? What featureswould a fully representative assembly exhibit?

One interpretation, plausible at first sight, isthat a representative assembly should be a micro-cosm of society. The idea here is that a legislatureshould be society in miniature, literally ‘re-pre-senting’ society in all its diversity. Such a parlia-ment would balance men and women, rich andpoor, black and white, even educated and unedu-

252 GOVERNMENT AND POLICY

cated, in the same mix as in society. How, after all,could a parliament composed entirely of middle-aged white men go about representing youngblack women – or vice versa? To retain the confi-dence of society, the argument continues, a repre-sentative assembly must reflect social diversity,standing in for society and not just acting on itsbehalf (Phillips, 1995).

But there is a considerable difficulty here.Achieving a legislature which mirrors societywould require interfering with the normal processof election. A microcosm could only be achievedby quota, not election. Assemblies in communiststates certainly achieved high levels of representa-tion for peasants, workers and women. But this

success was at the price of strict party control overnominations and elections. Indeed, a true micro-cosm is best obtained by random selection, dis-pensing with elections altogether, as with juries inthe English-speaking world. In any case, would wereally want a parliament containing its due pro-portion of the ignorant, the inarticulate and thecorrupt?

Representatives would also need to be replacedregularly, lest they become seduced by the trap-pings of office. Rotation is sometimes attemptedby new radical parties but not usually for long.Indeed, Kay (2003) suggests that the idea of amicrocosm is incoherent in itself: ‘those who takepart in politics are unrepresentative because, ifthey were representative, they would be at homewatching television’. In truth, the assembly asmicrocosm is an impractical ideal – if it is an‘ideal’ at all.

Definition A legislature would be a microcosm if it formeda miniature version of society, precisely reflectingsocial diversity. An exact microcosm is imprac-tical but there may still be value in ensuring thatall major groups achieve some parliamentarypresence (Phillips, 1995).

Representation in most assemblies operatesthrough party. Victorious candidates owe theirelection to their party and they vote in parliamentlargely according to its commands. In NewZealand, Labour members must agree to abide bythe decisions of the party caucus. In India, anextreme case, members lose their seat if they voteagainst their party, the theory being that they aredeceiving the voters if they switch parties aftertheir election. MPs are slowly becoming less reli-able lobby fodder but they are still defined, firstand foremost, by their party label.

Elsewhere, party discipline is combined with atleast some independence for members. In Franceand Germany, for instance, party obligations mustbe reconciled with the constitutional requirementthat members of the legislature owe allegiance tothe nation and not to any group within it. In prac-tice, party voting remains important, even if it isnot enforced with the eagerness found in NewZealand and India.

LEGISLATURES 253

Function Comment

Representation Most members articulate the goals of the party under whose label they were elected

Deliberation Debating matters of moment is the classic function of Britain’s House of Commons

Legislation Most bills come from the government but the legislature still approves them and may make amendments in committee

Authorizing Parliament’s role is normally expenditure reactive, approving or rejecting a

budget prepared by the government

Making In most parliamentary systems,governments the government emerges from (pp. 000–00) the assembly and must retain its

confidence

Scrutiny Oversight of government activity and policy is growing inimportance and is a task

well-suited to committees

Note: In parliamentary systems, the legislature is also responsiblefor making and sometimes breaking governments, a key functionexamined separately on pp. 000–00.

B OX 1 4 . 2

Functions of legislatures

Deliberation

An important function of many legislatures is toserve as a deliberative body, considering publicmatters of national importance. This functioncontrasts sharply with the microcosm and partyviews of representation and gives an additionalperspective on what parliaments are for.

In the eighteenth and nineteenth centuries,before the rise of disciplined parties, deliberationwas regarded as the central purpose of parliaments.The theory was that politicians should serve astrustees of the nation, applying exceptional knowl-edge and intelligence to the matters before them.Clearly, this philosophy of deliberation stands incomplete opposition to the idea of representationas a microcosm.

The British conservative Edmund Burke(1729–97) gave the classic exposition of the delib-erative account. Elected member of parliament forBristol in 1774, Burke admitted in his victoryspeech that he knew nothing about the con-stituency and had played little part in the cam-paign. But, he continued,

Parliament is not a congress of ambassadorsfrom different and hostile interests; which inter-ests each must maintain, as an agent and advo-cate against other agents and advocates; butParliament is a deliberative assembly of onenation, with one interest, that of the whole;where, not local purposes, not local prejudices,ought to guide, but the general good, resultingfrom the general reason of the whole. Youchoose a member indeed; but when you havechosen him, he is not a member for Bristol, buthe is a member of Parliament.

Deliberation of course continues today even ifits status is no longer as exalted as in Burke’s time.However, the deliberative style varies across coun-tries and here the contrast between talking andworking assemblies again proves useful.

In talking assemblies such as Britain’s and NewZealand’s, deliberation takes the form of debate inthe chamber. In Britain, key issues eventuallymake their way to the floor of the House ofCommons where they are debated with passionand often with flair. Floor debate sets the tone for

national political discussion, forming part of acontinuous election campaign. One of theachievements of the Commons is precisely itsability to combine effective deliberation, at leaston vital issues, with strong partisanship.

In ‘Considerations on Representative Govern-ment’ (1861), the English political philosopherJohn Stuart Mill (1806–73) makes the case forsuch a talking assembly:

I know not how a representative assembly canmore usefully employ itself than in talk, whenthe subject of talk is the great public interests ofthe country, and every sentence of it representsthe opinion either of some important body ofpersons in the nation, or of an individual inwhom such body have reposed their confidence.

By contrast, in working assemblies such as theAmerican Congress and the Scandinavian parlia-ments, deliberation is less theatrical. It takes theform of policy debate in the committee room. Thestyle here is more careful and detailed; deliberativein the literal sense of a careful consideration of thealternatives.

Legislation

Most constitutions explicitly assert the legislativefunction of parliaments. The end of absolute exec-utive power is affirmed by giving to parliament,and to it alone, the right to make laws. Arbitrarygovernment is replaced by a formal procedure forlaw-making. This painstaking process signals theimportance attached to government by rulesrather than individuals. Authoritarian rulersgovern by decree but in a democracy bills are scru-tinized and authorised by a congress of the nation.

Yet legislation is rarely the function where ‘legis-latures’ exert most influence. Indeed, after fiftyyears the European Parliament – admittedly aspecial case – does not even control the formallaw-making process in the European Union (see‘50 years of the European Parliament’). Atnational level, legislatures must approve bills buteffective control over legislation usually rests withthe government. Bills pass through the assemblywithout being initiated or even transformed there.

In party-dominated Australia, for instance, the

254 GOVERNMENT AND POLICY

government treats the legislative function of par-liament with virtual contempt. On a single nightin 1991 it sought to put 26 bills through theSenate between midnight and 3 a.m. In the erabefore New Zealand adopted proportional repre-sentation, a Prime Minister boasted that if an ideacame to him while shaving, he could have it onthe statute book by the evening, truly a case ofslot-machine law.

In Britain, similarly, the governing party domi-nates law-making. Ninety-seven per cent of billsproposed by government between 1945 and 1987became law. As Rose (1989, p. 173) said ofBritain, ‘laws are described as acts of Parliamentbut it would be more accurate if they were

stamped “Made in Whitehall” ’. In the party-dom-inated parliaments of Britain and some of its ex-colonies, the legislative function is reduced toquality control: patching up errors in bills pre-pared in haste by ministers and civil servants. Inlegislation, at least, these assemblies are reactiverather than active.

By contrast, in the parliamentary systems ofcontinental Europe, many legislatures do play amore positive role in legislation. Coalition govern-ments, influential committees and an elite com-mitment to compromise combine to deliver lawsacceptable to all sides. In a few countries, thismore flexible approach is reflected not just in thediscussion of bills but also in their initiation. In

LEGISLATURES 255

Although this time-line demonstrates thewidening competence of the European Parliament(EP), the assembly’s legislative authority remainslimited. However, as the EU itself has deepened, sothe parliament has become more willing to assertits expanding rights to at least be consulted onlegislation, the budget and appointments.

1952 Assembly of the European Coal and SteelCommunity established as an instrumentof scrutiny, with the right to dismiss theCommission in some circumstances.Assembly members are drawn fromnational legislatures.

1962 The Assembly is renamed the EuropeanParliament.

1970 The Treaty Amending Certain BudgetaryProvisions of the Treaties gives the EPmore influence over the budget.

1975 The Treaty Amending Certain FinancialProvisions of the Treaties gives the EP theright to propose modifications in areaswhere expenditure is not mandated byprevious agreements.

1979 First direct elections. Average turnout 67

per cent.

1980 Isoglucose judgment by the EuropeanCourt requires the EP to be consulted onproposed laws.

1986 The Single European Act initiates coopera-tion and assent procedures which give theEP more influence over legislation.

1992 The Maastricht Treaty requires members ofthe Commission to be approved by the EP.New co-decision procedure gives the EPsome veto authority over legislation.

1997 Treaty of Amsterdam extends the co-deci-sion procedure and formalizes the EP’sright to veto the nominee for Commissionpresident.

1999 All 20 commissioners (the SanterCommission) resign rather than face dis-missal by the EP for mismanagement.

1999 Fifth direct elections to the EP. Averageturnout 56 per cent.

Further reading: Judge and Earnshaw (2003),Scully (2003),Tsebelis and Garrett (1997).

50 YEARS OF

T H E E U R O P E A N P A R L I A M E N T

Switzerland, a bill may originate not just from theexecutive but also from members of either houseof the federal assembly or from any canton.

But it is in presidential systems, including theUnited States, that the assembly achieves mostautonomy in making law. The separation ofpowers and personnel inherent in a presidentialregime limits executive influence in the legislature.This institutional separation is often reinforced bydivided government. In the United States, oneparty frequently controls the White House whilethe other has a majority in either or both housesof Congress. In Latin America, where parties areweaker and proportional representation is thenorm for congressional elections, the president’sparty may be no more than a minor player in thelegislature.

Yet even in presidential systems the initiative inframing bills usually lies with the executive.Certainly, in the American Congress onlymembers of the House of Representatives can for-mally introduce bills. But the executive can easilyfind a friendly representative to initiate a bill on itsbehalf. The political reality is that bills are devel-oped by the administration and then transformedin Congress if indeed they do not expire in itsmaze of committees. ‘You supply the bills and wework them over’, a member of Congress said to anadministration official. The executive proposes butCongress disposes, usually by saying no.

Inevitably, this pluralistic process of law-makingreduces the coherence of the legislative pro-gramme. As President Kennedy said, ‘It is veryeasy to defeat a bill in Congress. It is much moredifficult to pass one’ (Eigen and Siegel, 1993, p.82). Or as Davy Crockett wrote in 1834, ‘Woebetide a bill that is opposed. It is laid aside forfurther time, and that never comes.’ The difficultywhich presidential systems experience in passinglaws stands in marked contrast to the tight controlover the legislative programme exerted by theruling party in the British parliamentary system.

Although the process by which a legislaturetransforms a bill into law naturally varies from onedemocracy to another, Figure 14.2 offers a generaloutline. The procedure is explicitly deliberative,involving several readings (debates) as the billmoves from the floor to committee and back again(Döring and Hallerberg, 2004).

Bicameral legislatures face an additionalproblem in realizing the legislative function. Whathappens should the second chamber amend thebill passed by the lower house? There must besome means of resolving such discrepancies. Inalmost all countries, the initial step is for theamended bill to return to the lower chamber for

256 GOVERNMENT AND POLICY

Figure 14.2 Typical steps in making a law

FIRST READINGDebate

VoteTo committee(s)

COMMITTEE(S)Detailed scrutiny

VoteTo floor

SECOND READINGDebate

Vote

THIRD READINGDebate

VoteTo second chamber

(bicameral assembly)

SECOND CHAMBERRepeats above process

To first chamber or joint committee of bothchambers (with modifications)

FIRST CHAMBERAccepts or rejects

modificationsVote

JOINT COMMITTEEReconciles the two

versionsVote

BILLSigned into law by head

of state

Source: Adapted from Mahler (2003) Table 4.9.

further discussion. But if the bill continues toshuttle between houses in this way until an agreedversion emerges, as in Italy’s strongly bicamerallegislature, the danger is that it will be delayed ornever become law at all. In Italy, a bill on rapeintroduced in 1977 did not become law until1995.

So most countries have developed a procedurefor short-circuiting the shuttle (Box 14.3). Themost common procedure is to employ a specialconference committee, containing an equalnumber of members from each chamber, toproduce an agreed bill. This compromise versionmust then be approved or rejected by the mainchambers without further modification.

The American, French and German legislaturesare among those employing conference commit-

tees. Conference committees are sometimesdescribed as third chambers for it is here that theultimate deals are struck and the decisive compro-mises made. They can be a vital arena in whichpolitical issues reach resolution.

In 2003, for example, the final shape of theGerman government’s Agenda 2010 economicreform package was formed in and around Room1.128 of the Bundesrat: the meeting place of theparliament’s 32-member mediation committee.The people meeting there represented not justtheir legislative chamber but also their party andtheir region. In modifying the executive’s pro-posals in a manner acceptable to the major inter-ests in Germany, they were shaping thegovernance of their country.

Authorizing expenditure

This is one of the oldest functions of parliamentand of the lower house in particular. The origin ofEuropean parliaments lay in the monarch’srequirement for money. Since, as Spencer Walpole(1881, p. 4) wrote, ‘the necessities of kings are theopportunities of peoples’, assemblies were able toestablish the right to raise grievances beforegranting supply (revenue). In Britain, this tradi-tion continued until 1982 with ‘supply days’during which the opposition could raise any issuesit wished.

The power to authorize spending may be one ofparliament’s oldest functions but in many democ-racies it has become nominal. Even more than thelaw-making function, it forms part of the mythrather than the reality of parliamentary power.Indeed lack of real financial control is a majorweakness of the modern assembly. Typically, theexecutive prepares the budget which is thenreported to parliament but rarely modified there.

Special rules usually set out how the legislaturemust handle the budget. In many countries(including Britain, New Zealand and Spain) par-liament cannot initiate its own expenditure pro-posals; it can only approve or reject spendingproposed by the government. In other cases, suchas Denmark and France, assembly proposals toincrease expenditure must be accompanied by off-setting reductions elsewhere (Döring, 2001).

So parliamentary approval is largely after the

LEGISLATURES 257

Procedure Comment Example

Indefinite Amended versions Italy is the main shuttle continue to shuttle example of this

between chambers rare procedureuntil agreement is reached (if ever)

Lower The lower house Czech Republic,house is decides whether Spain, United decisive to accept or reject Kingdom

amendments from the upper house

Conference A joint committee Many countries,or mediation of both chambers including committee negotiates a France,

compromise version Germany,which is then voted Switzerland on by each house and the USA

Vote of a The larger lower Australia, Brazil,joint session chamber exerts India, Norway of both most weight inchambers a joint vote

Source: Adapted from Tsebelis and Money (1997),Table 2.2a, pp.56–62.

B OX 1 4 . 3

Resolving differences in the versions of a bill passed by eachchamber

fact, serving to confirm budget compromisesworked out between government departments. Inmost democracies, once the budget reaches theassembly, it is a done deal. If the assembly beganto unpick any part of the budget, the wholepackage would fall apart.

Australia is an extreme case of governmentcontrol over the budget. Emy and Hughes (1991,p. 361) describe the political realities:

there is no suggestion of the House ofRepresentatives ever ‘refusing supply’ sincecontrol over the whole process of financialappropriations is firmly in the hands of the exec-utive. Only they may propose to spend money . . . Moreover, it seems members of the Housethemselves lack both the knowledge of, and aninterest in, the financial procedures which are,ultimately, crucial to the concept of parliamen-tary control.

The United States is once more the great excep-tion. Congress remains central to the confusedtangle of American budget-making. The constitu-tion places the power of taxation specifically withthe lower chamber: ‘all bills for raising revenueshall originate in the House of Representatives’.Similarly, all money spent by executive depart-ments must, under the constitution, be allocatedunder specific spending headings approved bymembers of Congress. No appropriations byCongress means no government programme. AsFlammang et al. (1990, p. 422) write, ‘without theagreement of members of Congress, no money canbe doled out for foreign aid, salaries for army gen-erals or paper clips for bureaucrats’. Given the sep-aration of powers, the executive cannot assumethat such agreement will be forthcoming.

The overall federal budget which the president isrequired by law to present to Congress each year isalso subject to congressional review. Furthermore,in the 1970s Congress created its own BudgetOffice in an effort to match the skills available onthe executive side. As a result, it can now deployconsiderable financial expertise in analysing thepresident’s proposals.

The result is that the annual American budgethas become an elaborate game of chicken. Theexecutive and the legislature each hopes the other

side will accede to its own proposals before themoney runs out and federal employees have to besent home without pay, as briefly occurred in1995. This rather alarming method of budget-making supports the proposition that financingthe modern state, like the law-making function, istoo important to be left to the assembly’s manyhands.

Scrutiny

The final function of legislatures is scrutiny (oroversight) of the executive. Although we are consid-ering this role last, such activity has been growingin significance and value. Its expansion gives the lieto unqualified assertions of legislative decline.

To emphasize the scrutiny function is to acceptthat the executive, not the legislature, mustgovern. However, the assembly can restate its keyrole as representative of the people by acting as awatchdog over the administration. Effective moni-toring can compensate for the downgrading of theassembly’s legislative and expenditure functions,providing a new direction to parliament’s work.

A modern assembly possesses three main instru-ments with which to monitor the executive:

� questions and interpellations (mainly used inparliamentary government)

� emergency debates and confidence votes(mainly used in parliamentary government)

� committee investigations (used in both presi-dential and parliamentary government).

Questions refer to direct queries of ministers. Inmany parliamentary systems, oral and writtenquestions are mainstays of oversight. In Britain,for example, members of the House of Commonsask a total of over 70,000 questions a year, keepingmany civil servants busy as they prepare answersfor their minister. In addition, all ministers mustface direct questioning in the Commons fromtime to time. Prime Minister’s Question Time,though now reduced to once a week, remains atheatrical joust between the PM and the leader ofthe opposition.

In other countries, questions have lower status.French ministries often fail to answer them at all,and in the Australian House of Representatives

258 GOVERNMENT AND POLICY

ministers give long prepared answers to questionsfrom their own side precisely so less time is left foropponents’ queries. But in some assemblies incontinental Europe, including Finland, France andGermany, the interpellation provides an alternativeform of interrogation. An interpellation is a sub-stantial form of question followed by a shortdebate and a vote on whether the government’sresponse is deemed acceptable.

DefinitionAn interpellation is an enquiry of the govern-ment, initiated by the opposition, which is fol-lowed by a debate and usually a vote on theassembly’s satisfaction with the answers given.This technique, often linked to a vote of no confi-dence, brought down several governments inthe French Third and Fourth Republics.

Emergency debates are a further, and higher-profile, way in which the legislature can call thegovernment to account. Normally a minimumnumber of members, and the Presiding Officer(Speaker), must approve a proposal for an emer-gency debate. Although the event normally endswith a vote – and a government win – the signifi-cance lies in the debate itself, and the fact of itscalling, rather than the outcome. An emergencydebate creates publicity and demands a carefulresponse from the government.

Votes of confidence or censure motions are theultimate test which a legislature can pose to theexecutive in a parliamentary system. Such motionsare not so much a form of detailed scrutiny as adecision on whether the government can continuein office. Again, special rules may apply: in Franceand Sweden, for instance, a majority of allmembers (not just those voting) is required todemonstrate the legislature’s loss of confidence. Inother countries, however, a confidence motion isnot specifically designated but is simply any voteon which the government would feel obliged toresign if defeated. In the Netherlands, this flexi-bility has led to rather odd motions about whetherother motions should be treated as censure votes(Andeweg and Irwin, 2002, p. 130). In somecountries, again including Sweden, votes of confi-dence can be directed against individual ministersand not just the government as a whole.

Above all, the rise of the scrutiny function hasstrengthened committees. Because the floor of thehouse is an inappropriate venue for detailedscrutiny, committee investigations are the key formof oversight. The American Congress is the classiccase here. Exceptionally, the constitution givesCongress – rather than the executive – responsi-bility for such important matters as commerce, thecurrency, defence and taxation. Of course,Congress does not carry out these functions itself;it delegates the tasks to the bureaucracy. Butbecause of its constitutional status, and the bud-getary authority flowing from it, Congress pos-sesses inherent powers of oversight. A British civilservant can afford to treat a parliamentary questionas a minor distraction but the head of a govern-ment agency in the United States knows that nextyear’s budget may depend on maintaining goodrelations with the relevant members of Congress.

The sheer extent of committee engagement inthe activity of American government is remark-able. On defence issues alone, 14 committees and43 subcommittees of Congress held hearings in1988, creating a large and possibly excessiveburden on the Pentagon. With more than 3,000committee staffers offering support, Congressachieves a unique level of scrutiny. However evenin the USA, committee oversight is limited. It canonly cast light on a few corners of a vast bureau-cracy; Congress often micro-manages departmentsrather than setting broad targets; and committeesmust concern themselves with finance and legisla-tion as well as scrutiny. In America, and even moreelsewhere, the government’s battalions outnumberthe forces available to the legislature.

Recruitment

In a democracy, by definition, virtually the entireadult population can stand for election. Legislativerecruitment is the process by which this huge poolof potential members of parliament is reduced tothe small number who achieve election. In parlia-mentary systems, government ministers are usuallyselected from the assembly so the legislaturebecomes the key channel of recruitment to polit-ical office. Its members constitute the shortlistfrom which national leaders are drawn.

LEGISLATURES 259

How then does this important process ofrecruitment operate? Box 14.4 outlines a four-stage model of political recruitment, based onlegal, social, party and electoral filters.

At the first stage, legal considerations reduce theentire population to those who are technically eli-gible to join the legislature. Many countries applymore stringent age limits for legislators than forvoters. In the United States, for instance, theminimum age remains 30 for the American Senateand 25 for the House, compared to 18 for voters.Some countries also impose citizenship and resi-dence requirements. In the USA, again, the consti-tution stipulates that members of Congress mustbe American citizens residing in the state they rep-resent.

At the second stage, a significant social filterreduces those who are eligible to stand to thosewho aspire to do so. Here, social influences onpolitical interest and ambition, such as beingbrought up in a political family, come into play.

At the third stage, a party filter reduces the aspi-rants to the small subset of candidates – those whosucceed in convincing the gatekeepers that theyshould be nominated for an electoral district orgranted a high place on a party list. Relevantfactors here include personal skills such as theability to present oneself well to the selectors aswell as party rules on such issues as genderquotas.

And the fourth and final electoral filter reducesthose who stand as candidates to those who winthe contest and become members of the nationalparliament.

This four-stage model, taking us from the eligi-bles to the elected, reveals that political recruit-ment is a far broader process than election alone.Indeed, the voters join the game only at the end,when most winnowing out has already takenplace. Over 130 million Americans were eligible tostand for president in 2000; in some states only sixcandidates appeared on the ballot; and only twonominees – those for the Democratic andRepublican parties – stood a realistic chance ofwinning.

In presidential systems, of course, the separationof executive and legislature means that peoplecannot belong to both at the same time. Althoughmembers of the legislature can resign to join the

government, and even stand for president, theassembly is generally less important as a recruitingagent to high office. In the USA, for instance,recent presidents have tended to be drawn fromstate governments rather than Congress: forexample, Jimmy Carter, Ronald Reagan, BillClinton and George W. Bush. The last presidentto move directly from Congress to the WhiteHouse was John Kennedy, in 1960. A career inCongress remains attractive in its own right; it ismore than a stepping stone to higher office (Davis,1998).

A similar point applies to the presidentialsystems of Latin America. There, too, joiningCongress is rarely seen as a direct route to a long-term position in the national government. But thestatus of assemblies in South America is lowerthan in the USA, reducing the calibre of themembers. Many legislatures are under-resourcedbodies operating in the shadow of assertive presi-dents. In addition, constitutions commonly limitmembers to one or two terms of office, ensuringturnover at the cost of expertise. In Mexico, in anygiven legislative term about eight out of tendeputies typically have no previous experience inthe Chamber of Deputies (Nacif, 2002).

260 GOVERNMENT AND POLICY

Segment of the Main influence Examplepopulation

Population Age and residence

↓ Legal rules requirementsEligibles

Growing up in a↓ Socialization

political familyAspirants

PartyGender quotas↓ gatekeepers

CandidatesElections Popularity of ↓ government

Members of the legislature

Source: Adapted from Norris (1996), fig. 7.1, p. 196.

B OX 1 4 . 4

Stages of recruitment to the legislature

LEGISLATURES 261

Britain is an established democracy whose politicalsystem has nonetheless been in transition.Traditionalmodels portrayed Britain as a centralized, unitarystate; as a two-party system; as an exemplar of parlia-mentary sovereignty in which ministers were held toaccount by the assembly; and as a political systemwhose uncodified constitution offered little formalprotection of individual rights.Yet the accuracy of allthese images has come under review, partly as a resultof the election in 1997 of a modernizing Labouradministration.

The centralized and even the unitary character of theUnited Kingdom has been put in question by creatingelected assemblies for Scotland and Wales.The two-party system has been challenged by the rise of thecentre-left Liberal Democrats. Parliamentary sover-eignty has been dented by British membership of theEuropean Union and a more assertive judiciary.Ministerial accountability has been complicated bythe delegation of government tasks to semi-indepen-dent agencies. And individual rights now receiveclearer protection from the incorporation into Britishlaw of the European Convention on Human Rights.

The new era of transition in British politics is certain toimpinge on its assembly.Traditionally, Britain’s parlia-ment (the oldest in the world) mixed omnipotence andimpotence in a seemingly impossible combination.Omnipotence, because parliamentary sovereignty,allied to an uncodified constitution, meant there couldbe no higher authority in the land. Impotence, becausethe governing party exercised tight control over itsown backbenchers, turning parliament into an instru-ment rather than a wielder of power.

In the twenty-first century, parliament’s position hasbecome less certain.The tired rituals of adversary poli-tics in the Commons have become less convincing,not least for the 260 new MPs elected in 1997 and the99 who followed in 2001.The notion that parliamentstill possesses some abstract quality called ‘sover-eignty’ still carries weight but, like many assemblies,Britain’s legislature runs the risk of being left behindby international integration.

But not all developments are negative. MPs them-selves have become more professional and com-mitted.The era of the amateur MP is over. And newselect committees have begun to enter the debateover policy.

To strengthen its position in an evolving politicalsystem, parliament will need to step further down theroad of reform. Besides its traditional function as atalking assembly, the legislature will need to becomea more effective working body.To influence a morecomplex decision-making process, committee reportsmust offer well-researched recommendations.

Yet even as the British parliament tries to broaden itsrepertoire, it will surely continue to do what it hasalways done best: acting as an arena for debatingissues of central significance to the nation, its govern-ment and its leaders.

Population: 60.1m.GDP per head: $25,500.Form of government: a parliamentary

democracy, with an hereditarymonarch playing a largely ceremonialrole.

Legislature: the House of Commons(659 members) is the dominantchamber.The House of Lords, the com-position and powers of which areunder review, acts in a revising andrestraining capacity.

Executive: the cabinet is the top deci-sion-ratifying body; the prime ministerselects and dismisses cabinetmembers. Most policy is formedoutside the full cabinet, either incabinet committees and surroundingdiscussions or in informal groups ledby the prime minister.

Judiciary: based on the common lawtradition. Britain’s membership of theEuropean Union and devolution toScotland and Wales enlarged the

scope of a traditionally restrained judi-ciary. In 2003, the government pub-lished proposals to introduce asupreme court.

Electoral systems: national and localelections still use the plurality system.The new Scottish and Welsh parlia-ments, first elected in 1999, use anadditional member system.The 1999election to the European parliamentused party list PR.

Profile U N I T E D K I N G D O M

Further reading: Butler and Kavanagh (2002), Dunleavyet al. (2003), Searing (1994).

Where American congressmen normally seekand achieve reelection, in Latin America’s legisla-tures many members are pursuing long-term polit-ical careers in their home province, perhaps asstate governor. In a grab-and-run exercise, they usetheir short period in the national legislature tolobby for their home district.

DefinitionTerm limits restrict elected politicians to a setnumber of terms of office, or ban reelectionwithout a break.They are commonly used inLatin America, for both legislators and presi-dents, in an attempt to prevent abuse of office.They are also employed in some states of theUSA.Term limits enforce turnover at the risk ofdamage to professionalism (Carey, 1998).

One finding of research into legislative recruit-ment stands out: in every country the profile ofparliamentarians is statistically unrepresentative ofthe wider society. Reflecting wider patterns of polit-ical participation, democratic assemblies remaindominated by well-educated, middle-aged whitemen. There are also party differences in representa-tion. Often, representatives of right-wing partiescome from business, especially finance, while edu-cational careers provide a fruitful recruiting groundfor parties of the left. In the United States, law isthe most common professional background(Norris, 1995). Throughout the world, a surprisingnumber of representatives are drawn from highlypolitical families, suggesting that politics risksbecoming, to an extent, an occupational caste.

An additional theme in the recruitment of legis-lators is the rise of the career politician. Where thestatus of the legislature is high and term limitsabsent, the professional politician is coming todominate the chamber. The local farmer, the loyaltrade unionist, the prominent business executivewho takes up politics for a few years – all have lostground to the professional politician for whompolitics is a sole and full-time career.

What are the consequences of the rise of thecareer legislator? On the one hand, such membersare assiduous and ambitious. They burrow away incommittees, serve their constituents and influencepublic policy. They are not content to be merevoting machines for their party.

DefinitionCareer politicians are described by King (1981,p. 255) as:

people committed to politics.They regard politics as their vocation, they seek fulfilmentin politics, they see their future in politics, theywould be deeply upset if circumstances forcedthem to retire from politics. In short, they arehooked.

On the other hand, as Berrington (1995, p.446) comments, ‘the career politician does notknow when to leave well alone’. An assemblypeopled by career politicians is a restless, assertiveinstitution. It lacks the judgement of those withwork experience beyond politics. A case can cer-tainly be made that politics is inherently a broadprofession and that assemblies benefit from a fewcolourful amateurs to offset the dark suits of thecareer politicians.

Of course, career politicians can only flourish inparliament when reelection prospects are good; ifmembers are automatically removed after eachelection, no parliamentary career is possible. Evenwithout term limits, electoral swings can some-times condemn many members to defeat. InCanada, for example, MPs serve an average of justsix years in post. Most members are politicaltourists, on vacation from other more satisfyingcareers. Inevitably, the esteem of the institutionsuffers.

DefinitionThe incumbency effect refers to the electoralbonus accruing to sitting members arising fromtheir visibility in their district, their ability tobring home the bacon and their access toresources for their re-election campaign.Theincumbency effect helps to explain the high re-election rate in most legislatures.

But such cases are exceptional. In parliamentswithout term limits, reelection is the norm. Thesuccess rate of incumbents in winning reelection isover 85 per cent in Germany, New Zealand andthe USA. Average length of service reaches 20years in the UK and 15 years in Japan (Table14.2), which is certainly sufficient time to develop

262 GOVERNMENT AND POLICY

a parliamentary career. Indeed, the question iswhether the incumbency effect damages legisla-tures by limiting fresh blood. One yardstick of thepolitical recruitment process is its ability to delivera balanced legislature blending experience withnewcomers.

Legislatures in new democracies

A transition to democracy provides new opportu-nities for the legislature. Since parliamentsembody the idea of representation, we shouldexpect their stature to grow in newly minteddemocracies. And the prestige of legislatures hasundoubtedly risen in almost all new democracies,particularly when the assembly played a major rolein the free for all of democratic transition.

However, parliaments remain weaker in newdemocracies than in established democracies. Inparticular, scrutiny remains a contested function,partly because rulers are reluctant to grant rightsof investigation and partly because many new par-liaments are only now developing well-supportedcommittees. As Diamond (1999, p. 98) com-ments,

In most new democracies, legislatures lack theorganisation, financial resources, informationservice, experienced members and staff to serveas a mature and autonomous point of delibera-tion in the policy process.

The political experience of new legislators drawninto the assembly by the drama of democratizationwas often limited and many have now left politics,to be replaced by a more professional cohort. Sothe revival of assemblies in post-authoritarian set-tings is only now beginning to gather speed, atheme we can develop by reviewing the assembly’srole in transitions in Southern Europe, LatinAmerica and Eastern Europe.

Southern Europe

In Greece, Italy, Portugal and Spain, parliamentshad to recover from highly subordinate positionsunder right-wing dictatorships. In the Spanishtransition, parliament did play a key role in devel-oping the new democratic system because it ‘wasthe only meeting place for the democratically-minded. At the outset of democracy, it was theplace where all the advocates of renewal met’(Giol, 1990, p. 96). Yet once the new democracyconsolidated, the esteem of parliament soondeclined as it became dominated by disciplinedparties and strong leaders.

Parliaments played a less crucial role in the otherSouthern European transitions and they remainweaker institutions than in Northern Europe.Pridham (1990, p. 246) notes that one feature ofthe new party-dominated democracies of SouthernEurope is the limited expectations held of theirparliaments.

Latin America

The position of legislatures in Latin America hasundoubtedly strengthened in the post-military era.As presidents have become more accountable, atleast in the larger and more democratic countries,so assemblies have asserted the authority whichflows from the separation of powers in a presiden-tial executive. In a previous edition, we wrote thatin a more democratic era, the realistic goal facingLatin American legislatures was not to gain sover-

LEGISLATURES 263

Table 14.2 Average length of service in nationalparliaments

Years

United Kingdom 20

Japan 15

New Zealand 12

United States (House) 12

United States (Senate) 11

Israel 11

Germany 8

Denmark 8

France 7

Canada 6

Source: Adapted from Somit (1994, p 13). Except where indicated, thefigure for bicameral assemblies refers to the lower house.

eignty but rather to find some ways of ‘restrainingthe prince and disciplining the powerful’(Chalmers, 1990). The legislatures in Brazil andMexico are achieving just that.

But it would be wrong to conclude that parlia-ments in these new democracies have achieved, orwill achieve, the exalted position of Congress inthe United States. In the main, Latin Americanconstitutions do not separate powers as completelyas in the USA. They give presidents greater controlover the budget and some latitude to governthrough decrees. Further, political traditioninclines the electorate to look to presidents ratherthan to multi-party legislatures for effective gover-nance. While resources and professionalism haveincreased in many Latin American legislatures,their expertise and influence still lag well behindthe American Congress.

Central Europe

In several Central European countries, such asPoland and Hungary, legislatures were beginningto achieve modest grievance-raising and policy-influencing powers even before the collapse ofcommunist rule. This experience undoubtedlyhelped to prepare these legislatures for the post-communist era. In several ways, parliaments werein a favourable position to influence legislation andpolicy in the first post-communist decade. UnlikeSouthern Europe, political parties and interestgroups remained weak, giving legislators excep-tional autonomy in their work. Yet this potentialwas only partly realized. High turnover, limitedcalibre, poor support and weak internal procedureslimited the assembly’s ability to confront assertivebut often divided executives (Ågh, 1996).

Only now, fifteen years or so after communism’sdownfall, are Central European assemblies begin-ning to settle into a more productive routine.More professional members are returning for addi-tional terms to work in a more stable politicalenvironment. Resources have increased; indeed,legislators are now relatively well-paid in compar-ison to many other professions.

Paradoxically, however, the influence exerted byindividual legislators is probably declining. Partiesare becoming better organized, placing more con-straints on their representatives. In addition, many

such countries have needed to make extensive legalchanges to meet the accession requirements set bythe European Union, a process in which the exec-utive has taken the lead, working round parlia-ment in many cases.

So the individual member is once morebecoming a foot soldier rather than a free-wheeling political entrepreneur – but that is thefate of many legislators in Western Europe too. AsKopecky (2003, p. 152) concludes,

parliaments throughout Central and EasternEurope have begun to resemble their sister insti-tutions in established democracies, both in theirinternal organization and procedures and in theexternal constraints normally placed upon suchinstitutions in modern democracy.

Legislatures in authoritarian states

Since assemblies are symbols of popular represen-tation in politics, their significance in authori-tarian regimes is inherently limited. Yet legislaturesare difficult to extinguish completely. In 1990only 14 out of 164 independent states had noassembly. Of those 14, only five (traditionaldynastic states in the Arabian Gulf ) had no experi-ence of assemblies at all.

Assemblies in non-democracies generally func-tion only as shadow institutions. Sessions are oftenshort and some members are simply appointed bythe government. In Mezey’s (1979) influential clas-sification, assemblies in non-democratic systemsplay only a marginal or a minimal role in policy-making (Box 14.5). Members concentrate onraising grievances, pressing constituency interestsand sometimes lining their own pockets. The rulersregard these activities as non-threatening; the realissues of national politics are left untouched.

Why, then, do authoritarian rulers bother withassemblies at all? Their value is fourfold:

� A parliament provides a fig leaf of legitimacy,both domestic and international, for theregime. The ruler can say to visiting dignitaries,‘Look! We too have an assembly, just like theBritish House of Commons and the AmericanCongress!’

264 GOVERNMENT AND POLICY

LEGISLATURES 265

The decline of parliament thesis maintains that legis-latures in established democracies are losing, or havelost, power to the executive. Petersson (1989, p. 96)notes that the argument is commonly encountered:‘every description of the form of government of themodern state seems to end up with a discouragingconclusion about the actual role of parliament’. Sincelegislatures express popular sovereignty, the implica-tion of the decline thesis is that democracy itself is inretreat. But is the assertion of a general loss of parlia-mentary authority correct?

The case forAs early as 1921, the British historian Lord Bryce (p.370) noted that ‘general causes ... have been tendingto reduce the prestige and authority of legislativebodies’. He referred specifically to the growth of disci-plined parties and to the increased complexity ofpolicy-making, both of which strengthen the execu-tive. In particular, assemblies have lost control of thelegislative process. Bills pass through the assembly ontheir way to the statute book but their origins lie else-where: in the executive, the bureaucracy and interestgroups.

Since then, growing interdependence between coun-tries has posed insuperable problems for assemblies.Parliaments are creatures of the nation; they arepoorly adapted to a global era. How can national par-liaments grapple with a world of international trade,intergovernmental deals, complex treaties and war?To be sure, a European Parliament now exists but itremains the runt in the litter of EU institutions, lackingstandard legislative powers. In today’s conditions, theexecutive is where the action is; parliaments havebecome political museum-pieces.

The case againstTo speak of the decline of legislatures in a global era istoo simple, for three reasons:

� Even if parliaments no longer initiate many bills,they continue to perform such functions as repre-sentation, scrutiny and, in parliamentary systems,recruitment to government office.

� The alleged golden age of parliaments was as longago as the mid-nineteenth century; there is littleevidence to show that parliaments have lost furtherground since 1945.

� In some ways assemblies are growing in impor-tance: as arenas of debate, as intermediaries in tran-sitions from authoritarian to democratic rule, asraisers of grievances and especially as agencies ofoversight.

Moreover, where the American Congress led the wayin equipping assembly members with the resourcesto do their jobs professionally, other legislatures arefollowing suit.Throughout the democratic world,backbench members have become more assertive:party leaders can no longer expect educated politi-cians to be totally deferential. Specialized committees,and members with a driving interest in policy, areincreasingly successful in contributing to politicaldebate.Televising proceedings has raised the profileof assemblies among voters. And the transition todemocracy has raised the standing of parliaments;when a new democracy consolidates, legislative pro-fessionalism eventually follows.

AssessmentThe role of parliaments is changing rather thandeclining.The reality is that legislatures do not initiatemany laws but that fact need not be mourned.Indeed, an assembly which really did settle the fate oflegislation would be a dubious asset.What could bemore odd, asked the English journalist Walter Bagehot(1826–77), than government by public meeting?

What modern assemblies can do, and what many arenow better equipped than ever to do, is to overseethe executive. Parliaments possess a unique authorityto force politicians and civil servants to account fortheir actions before a body which still represents thenation.They remain an essential element in the archi-tecture of democracy.

Further reading: Norton (1998), Sjolin (1993).

DEBATE

THE DECLINE OF LEGISLATURES?

� The legislature can be used to incorporate mod-erate opponents into the regime, providing aforum for negotiating matters that do notthreaten rulers’ key interests.

� Raising constituents’ grievances and lobbyingfor local interests provide a measure of integra-tion between centre and periphery and betweenstate and society. Such activity oils the politicalwheels without threatening those who controlthe machine.

� Even marginal and minimal assemblies providea useful pool of potential recruits to the polit-ical elite. Behaviour in parliament provides auseful test of reliability.

In totalitarian systems, assemblies were evenmore marginal than in authoritarian regimes.Fascist states, in particular, had no time for parlia-mentary debate – or indeed for parliaments.Fascism was state-centred and opposed to parlia-mentary government. Decisive action, not weakdebate, was the goal of a fascist regime. Power wasto be concentrated in a single leader, not dilutedthrough a separation of powers. The vital blood ofnational unity could not be thinned by the partydivisions observed within parliament. In short,fascism condemned sovereign legislatures as theinstitutional expression of liberalism. For suchreasons, Mussolini abolished the remnants of theItalian parliament in 1938, introducing a new ifequally minimal corporate institution in its place(p. 178).

In communist states, legislatures were nottreated with quite the same contempt. Indeed,communist regimes often produced statisticallyrepresentative assemblies. Women and favouredgroups such as industrial workers achieved greaterrepresentation than in Western legislatures.However, this feature reflected tight party controlof the nomination process and the limited polit-ical significance of the legislature. In practice, asocially representative assembly has indicated animpotent institution; this is one reason whyfemale representation initially fell after the end ofcommunism. The short sessions of communistassemblies (typically around ten days a year) weredominated by the party, which used them tooutline past successes and future targets. Standingovations were written into the script; free debatewas written out.

As communist regimes became somewhat morepluralistic in their later decades, so legislaturesacquired a measure of autonomy. China illustratesthis trend. In the twelve years before MaoZedong’s death in 1976, the National People’sCongress (NPC) did not even meet. However, inthe subsequent era of economic reform, the NPCbegan to flourish. A growing emphasis on the ruleof law raised the status of the legislature which hasalso begun to express popular hostility to corrup-tion. Many votes are no longer unanimous. Seniorfigures drafted into the assembly skilfully strength-ened the NPC’s position in Chinese politics, notby confrontation with the ruling party but by

266 GOVERNMENT AND POLICY

Type Definition Example

Active Assembly makes policy autonomously US Congress

Reactive Assembly largely responds to, but can also UK House of Commonsinfluence, government policy

Marginal Assembly is a minor partner in policy-making Many legislatures in communist states

Minimal Assembly is a rubber stamp under executive Many African states in the era of one-partydomination rule

Source: Adapted from Mezey (1979).

B OX 1 4 . 5

A policy classification of assemblies

assisting the transition to a market economy andby making efforts to encourage national integra-tion through links with subnational congresses.The number of support staff has increased, raisingthe Congress’s professionalism (O’Brien, 1990).

However the NPC, still the world’s largest legis-lature, remains strongly hierarchical in its internalfunctioning. Plenary sessions remain formal andinfrequent. More than in any democracy, legisla-tive influence operates through small subgroups ofsenior figures. These subgroups remain sensitive tothe party’s interests. As with Chinese politics gen-erally, authority within the NPC is focused on afew senior figures, not on the institution as awhole. In short, the NPC has become part of theChinese power network but even today its posi-tion cannot be understood through the Westernnotion of the separation of powers (Xia, 1999).

Key reading

Next step: Norton (1990a) is a helpful intro-duction, drawing together some classic writings onlegislatures.

Olson (1994) offers a useful comparative treat-ment of parliaments. Bicameralism and secondchambers have attracted a flurry of interest: seeTsebelis and Money (1997), Patterson andMughan (1999), Russell (2000a) and the specialissue of the Journal of Legislative Studies (Vol. 7,2001). Committees are examined comparativelyin Longley and Davidson (1998). Best and Cotta(2000) provide a comparative treatment of long-term changes in parliamentary careers. For studiesof assemblies in specific regions see Norton(1998) on Western Europe, Morgenstern andNacif (2002) on Latin America, Baaklini et al.(1999) on the Arab world and Norton andAhmed (1999) on Asia. Judge and Earnshaw(2003) survey the European Parliament. Themost intensively studied legislature remains theAmerican Congress: Dodd and Oppenheimer(2001) is a useful collection. For comparativestudies of assemblies in democratic transitions seeLiebert and Cotta (1990) on Southern Europeand Kopecky (2003) on Central and EasternEurope.

LEGISLATURES 267

Presidential government 268Parliamentary government 274Semi-presidential government 281The executive in new democracies 284The executive in authoritarian states 285Key reading 289

The political executive is the core of govern-ment, consisting of political leaders who form

the top slice of the administration: presidents andministers, prime ministers and cabinets. The exec-utive is the regime’s energizing force, setting prior-ities, making decisions and supervising theirimplementation. Governing without an assemblyor judiciary is perfectly feasible but ruling withoutan executive is impossible.

The political executive, which makes policy,must be distinguished from the bureaucracy,which puts policy into effect. Unlike appointedofficials, the members of the executive are chosenby political means, most often by election, andcan be removed by the same method. The execu-tive is accountable for all the activities of govern-ment; it is where the buck stops.

DefinitionThe political executive forms the top tier ofgovernment. It directs the nation’s affairs, super-vises the execution of policy, mobilizes supportfor its goals and offers crisis leadership.

The categories of democracy and authoritarianrule are defined by how the executive operates.Established democracies have succeeded in thedelicate task of subjecting executive power to con-stitutional limits. These restrain the exercise ofpower and set out rules of succession to executiveoffice. Both in theory and in practice, politicalleaders in liberal democracies are accountable fortheir conduct. In an authoritarian regime, by con-trast, constitutional and electoral controls areeither unacknowledged or ineffective. The scope

of the executive is limited by political realities butnot by the constitution.

In established democracies, executives fall intothree main groups: presidential, parliamentary andsemi-presidential. We examine each in turn beforeconsidering the executive in new democracies andin authoritarian states.

Presidential government

The world contains many presidents but fewexamples of presidential government. Any tin-potdictator can style himself ‘president’ and many doso. However, the existence of a self-styled presi-dent is an inadequate sign of a presidential system.

Presidentialism proper is a form of constitu-tional rule in which the chief executive governsusing the authority derived from direct election,with an independent legislature (Figure 15.1).Because both president and legislature are electedfor a fixed term, neither can bring down the other,giving each institution some autonomy. This sepa-ration of powers is the hallmark of the presidentialsystem and is typically reinforced by a separationof personnel. Members of the executive cannot sitin the assembly, creating further distance betweenthe two institutions. Similarly, legislators mustresign their seats if they wish to serve in the gov-ernment, limiting the president’s ability to buycongressional votes with the promise of jobs.

Contrasting methods of election yield a furtherdivergence in interests: legislators depend only onthe support of voters in their home district or statewhile the president – and the president only – iselected by national ballot.

In essence, then, presidential governmentdivides power against itself. It sets up a require-ment for the executive to negotiate with the legis-lature and, by this mechanism, seeks to ensure thatdeliberation triumphs over dictatorship. Defendersof the system argue that it captures the essence of

Chapter 15

The political executive

268

democracy, setting constitutional limits on execu-tive power. However, detractors suggest that presi-dentialism is a conspiracy against government.The dangers of that deliberation turns into stale-mate, that the democratic will is thwarted byexcessive fragmentation and that the system itselfbecomes unstable.

DefinitionPresidential government consists of three fea-tures:

� popular election of the president who directsthe government and makes appointments toit

� fixed terms of offices for the president and theassembly, neither of which can bring down theother

� no overlap in membership between the exec-utive and the legislature.

Presidential government predominates in theAmericas. To understand the system, we mustbegin with the United States, where the formatfirst emerged. We will then take Brazil as a con-trast to the American case.

The United States

When the framers of the American constitutionmet in Philadelphia in 1787, they found that theissue of the executive created a dilemma. On theone hand, the Founding Fathers wanted to avoidanything that might prove to be a ‘foetus ofmonarchy’. After all, the American revolution hadjust rid the new nation of England’s George III.On the other hand, the delegates also agreed withAlexander Hamilton that a single executive wasneeded for ‘decision, activity, secrecy anddespatch’. The founders decided on the presi-dency, an office that could provide prompt actionfor a republic in which Congress was nonethelessexpected to play the leading role.

But how should the president be selected? Herethe delegates were less clear. One possibility was toallow the national legislature to choose the chiefexecutive. A subcommittee of the constitutionalconvention did canvass this possibility, whichwould have led to some form of parliamentarygovernment. But the idea that the executiveshould be beholden to the assembly was not yetestablished, even in Britain. Had it been so, theconvention could well have plumped for thismethod of appointment.

Instead, rather late in the proceedings, the dele-gates settled on selecting the president through anelectoral college appointed by state legislatures.Without realizing the implications, the framershad stumbled across the essential feature of presi-dential government: separate election of executiveand legislature (Dahl, 2001).

The constitution tersely states that ‘theExecutive Power shall be vested in a President ofthe United States’. He (and so far all have beenmen) is chosen by popular election, operatingthrough what has become a nominal electoralcollege, for a four-year term. Under a constitu-tional amendment of 1951, presidents are limitedto two terms of office though this has not pre-vented several former presidents from playing anactive role in public affairs (Schaller and Williams,2003).

A vice-president serves as stand-by: a raven ofdeath just a heartbeat away from the WhiteHouse. Historically, he (and again all have beenmen thus far) has a one in three chance of moving

THE POLITICAL EXECUTIVE 269

President

chooses

Cabinet ministers

administer

Departments

Figure 15.1 Presidential government

Voters

elect

Legislature

on to the higher office. Presidential candidatesselect their own running-mate for the election andvoters now express support for a vice-presidentialcandidate in the same way as they choose betweenthe candidates for the senior post.

The president may only be dismissed byCongress through ‘Impeachment for, and onConviction of, Treason, Bribery, or other highCrimes and Misdemeanors’. So far, just two –including Bill Clinton in 1998 – have beenimpeached. Just as Congress cannot normallyremove the president, neither can the presidentdissolve Congress and call new elections. Theexecutive and the legislature are elected separatelyand survive independently until the next election.

Although the American presidency is often seenas a symbol of power, the institution was designedas part of a concerted attempt to control executivepretension. The small print of the constitutionhems in the office with restrictions:

� The president is commander-in-chief butCongress retains the power to declare war.

� The president can make government appoint-ments and sign treaties but only with Senateapproval.

� The president ‘recommends to Congress suchmeasures as he shall judge necessary and expe-dient’ but is offered no means to ensure his pro-posals are accepted.

� The president can veto legislation but Congresscan override his objections.

� Congress, not the president, controls the pursestrings.

President Kennedy summarized the peculiarambivalence of the office:

The president is rightly described as a man ofextraordinary powers. Yet it is also true that hemust wield those powers under extraordinarylimitations.

Two points flow from the president’s constitu-tional position. First, to describe the relationshipbetween the president and Congress as a ‘separa-tion of powers’ is misleading. In reality there is aseparation of institutions rather than of legislativeand executive powers. President and Congress

share power: each seeks to influence the other butneither is in any position to dictate. This separatedsystem, as C. Jones (1994) calls it, is subtle, intri-cate and balanced. It reflects a successful attemptby the founders to build checks and balances intoAmerican government.

This tension within the system, it is importantto note, continues even when one party controlsboth the White House and the Congress. AsMayhew observes in Divided We Govern (1991, p.135), ‘to suppose that an American party winningCongress and the presidency thereby wins theleeway of a British governing party is to bedeluded by the election returns’. Whatever theirparty, members of Congress have different elec-toral interests from the president. They are electedwithout term limits from local areas for distinctterms of two years in the House and six years inthe Senate (Box 15.1).

Second, in a system of shared control presiden-tial power becomes the power to persuade(Neustadt, 1991). As President Truman said, ‘theprincipal power that the president has is to bringpeople in and try to persuade them to do whatthey ought to do without persuasion’. In this taskof persuasion, the president can take two routes,best followed simultaneously: going Washingtonand going public (Rose, 2000).

Going Washington involves the president inwheeling and dealing with Congress and itsmembers, assembling majorities for his legislativeproposals. Hence the saying that the president isthe nation’s most important lobbyist. LyndonJohnson, a former leader of the Senate Democratswho succeeded to the presidency followingPresident Kennedy’s assassination in 1963, was amaster of this strategy. Johnson was, however, lesssuccessful with the wider public, eventuallydeciding not to seek reelection in 1968 when theVietnam War was at its height.

Going public means that the president exploitshis unrivalled access to the mass media to influ-ence public opinion and so persuade Washingtonindirectly. ‘The great communicator’ RonaldReagan (President, 1981–89) adopted thisapproach with great success. In addition, though,Reagan ensured his senior aides kept in closetouch with key members of Congress; in effect, hiswas a dual strategy.

270 GOVERNMENT AND POLICY

The paradox of the American presidency – aweak governing position amid the trappings ofomnipotence – is reflected in the president’ssupport network. To meet presidential needs forinformation and advice, a conglomeration of sup-porting bodies has evolved. Collectively known asthe Executive Office of the President, these bodiesprovide far more direct support than is available tothe prime minister in any parliamentary system.

Yet this apparatus of advice has often proved tobe a weakness. Many advisers are political out-siders, appointed by the president at the start ofhis tenure before his eye for Washington’s politicsis in. Far from helping the president, adviserssometimes end up undermining his position. TheWatergate scandal in the 1970s destroyed the pres-idency of Richard Nixon; the Iran-Contra scandalin the 1980s lay siege to the reputation of RonaldReagan.

One problem is that the presidential systemlacks a strong cabinet to offer a counterbalance topersonal advisers. In the USA, the cabinet goesunmentioned in the constitution and meetings arelittle more than a presidential photo-opportunity.Cabinet members often experience difficulty ingaining access to the president through his thicketof advisers. For more than any prime minister, thepresident stands alone.

Brazil

With democracy – of an illiberal kind – nowestablished in parts of Latin America, students ofpresidential government can broaden their hori-

zons beyond the USA. The Latin American experi-ence confirms that presidents often face consider-able difficulty in securing legislative support fortheir programme, even when the constitutionappears more favourable to the executive than inthe United States (Mainwaring and Shugart,1997b).

A comparison between Brazil and the USA isparticularly useful. These are two large and popu-lous former colonies though Brazil is a muchpoorer and far more unequal country. Both coun-tries are federations with strongly bicameral legis-latures. Brazil’s political difficulties are the morerecent; the military only withdrew from power in1985 and the current constitution dates from1988 (Martínez-Lara, 1996).

Where the American president is hemmed inwith restrictions, the Brazilian constitution seemsto offer the country’s president an arsenal ofweapons in dealing with Congress (Box 15.2). Inmany areas, the Brazilian president can issuedecrees: provisional regulations with the force oflaw. These stay in effect for 60 days without par-liamentary approval and can be renewed once. Hecan declare bills to be urgent, forcing Congress tomake a prompt decision on these proposals. He –and in some areas he alone – can initiate bills inCongress. And the president proposes a budgetwhich goes into effect, month by month, ifCongress does not pass a budget itself.

Yet despite their panoply of formal powers,Brazilian presidents face the same problem as theirNorth American colleagues: legislators who knowtheir own mind. Indeed, Brazilian leaders experi-

THE POLITICAL EXECUTIVE 271

Electoral unit Length of term Limit on reelection

Presidency A national vote, aggregated 4 years Maximum of two termsby state in an electoral college

Senate Direct popular vote in each of 6 years No limits50 states

House of Direct popular vote in each of Representatives 435 districts 2 years No limits

B OX 1 5 . 1

Separate elections in the United States

ence even greater difficulty in bending Congressto their will. The explanation for this contrast liesin Brazil’s fragmented multiparty system. In2002, 19 parties were represented in the Chamberof Deputies and ten in the Senate; as usual, thepresident’s party was in the minority in eachchamber. And the interpretation for this differ-ence rests in the electoral system. Where theAmerican Congress is elected using the pluralitymethod, elections to the Brazilian legislature useproportional representation with an open partylist.

Furthermore, party discipline within Brazil’sCongress is exceptionally weak. Deputies oftenswitch party in mid-term; unlike the country’ssoccer players, they do not even wait until the endof the season before transferring their loyalties.Deputies are more concerned to obtain resourcesfor their district than to show loyalty to theirparty. In Brazil, parties are not only morenumerous but also less cohesive than in the USA,a contrast which complicates the Brazilian presi-dent’s task.

Reflecting partisan fragmentation, Brazil’s presi-

dents must attempt to build informal coalitions.This requirement takes the form of appointingministers from a range of parties to the executivein an attempt to extract loyalty from its deputiesin Congress. It is as though a Republican presi-dent in the United States confronted a hostileCongress and was able to bolster his support byappointing some Democratic congressmen to hiscabinet.

In making coalitions, Brazilian presidents areassisted by a more flexible interpretation of theseparation of institutions than is found in theUSA:

In Brazil, ministers will occasionally resign theirgovernment positions just before an importantvote in the assembly, resume their legislativeseats, vote and then resign their legislative seatsand resume their ministerial posts again.

(Cox and Morgenstern, 2002, p. 459)

Thus Brazil shows that presidential governmentdoes not need to be single-party government.Rather, presidents rely on a multiparty governing

272 GOVERNMENT AND POLICY

Brazil USA

Can the president issue decrees? Yes, in many areas.Valid for 60 days, Norenewable once

Can the president initiate bills? Yes, exclusively so in some areas No

Can the president declare bills urgent? Yes No

Can the president veto legislation? Yes, in whole and part Yes

How can Congress override a veto? Absolute majority in joint meeting of Two-thirds majority both houses in each house

President’s control over the budget Stronger Weaker

Party system Multiparty Two-party

Does the president’s party usually No Sometimes possess a majority inCongress?

Party discipline in Congress Weak Stronger

Sources for Brazil: Ames (2002), Cox and Morgenstern (2001), Neto (2002).

B OX 1 5 . 2

Comparing presidential powers in Brazil and the USA

THE POLITICAL EXECUTIVE 273

In the final decades of the twentieth century, manycountries emerged from military or communist rule toembrace democracy. Constitution-writers in thesenew regimes faced the question of whether to adopta presidential or parliamentary form of governmentfor the new democratic order.The drafters frequentlysought advice from political scientists in establisheddemocracies. Had we been asked, should we haveargued for or against presidential government forthese new democracies?

The case forA presidential system offers the stability which is thefirst requirement for a new regime.The president’sfixed term provides continuity in the executive,avoiding the frequent collapse of governing coalitionsto which parliamentary governments are prone.Winning a presidential election requires candidates todevelop broad support across the country. Elected bythe country at large, the president can then take anational view, rising above the squabbles betweenminority parties in the assembly.

A president provides a natural symbol for a new order,offering a familiar face for domestic and internationalaudiences alike.The leader can pursue a steady coursein foreign policy, free from the volatility which wouldarise if the executive were directly accountable to afractious assembly. Since a presidential system neces-sarily involves a separation of powers, it should alsoencourage limited government and thereby protectliberty. Remember, finally, that the USA is the world’sdominant power and has sustained presidential gov-ernment for over 200 years; there is surely pragmaticgood sense in following No. 1.

The case againstPresidential government is inappropriate for a newdemocracy because only one party can win the presi-dency; everyone else loses. All-or-nothing politics isunsuited to a new regime where political trust has stillto develop. In addition, fixed terms of office are tooinelastic;‘everything is rigid, specified, dated’, wroteBagehot (1867).The deadlock arising when executive

and legislature disagree means that the new politicalsystem may be unable to address pressing problems.

There is a danger, too, that presidents will grow toobig for their boots: Latin American experience showsthat they sometimes seek to amend the constitutionso as to continue in office beyond their one- or two-term limit. In particular, a frustrated or ambitious pres-ident may turn into a dictator; presidentialdemocracies are more likely to collapse than are par-liamentary democracies (Cheibub, 2002).The USAremains the world’s only case of stable presidentialgovernment over the long term – an exception toadmire but not, it seems, to copy.

Presidential government involves betting the countryon one person, thus inhibiting the development ofthe rule of law in new democracies.The risks are alto-gether too great, leading Lijphart (2000, p. 267) toregard presidentialism ‘as a strongly negative featurefor the future of democracy’. Prudence mandates aparliamentary system governed by a broad coalitioncabinet.

AssessmentIn practice, the choices made by new democraciesseem to reflect history and geography more than theadvice of politics professors. Central European coun-tries such as the Czech Republic and Hungary haveadopted the parliamentary form which dominatesWestern Europe, mindful no doubt of their decision tojoin the European Union. By contrast, post-militaryregimes in Latin American countries have drawn ontheir own political histories, and the presence of theUSA to the north, to embrace presidential govern-ment. Only time will tell if Latin America has made theright choice.

Further reading: Andrews and Montinola (2004), Linz(1990), Lijphart (2000), Manuel and Cammisa (1999).

DEBATE

PRESIDENTIAL GOVERNMENT FOR A NEW DEMOCRACY

coalition as a technique for influencing the legisla-ture. These coalitions are, however, more informaland unstable than the carefully crafted interpartycoalitions which characterize parliamentary gov-ernment in Western Europe.

So Brazilian presidents face in stern form theinherent problem of presidential government: highexpectations but no guarantee that they will beable to deliver. The challenge confronting Lula daSilva, elected Brazil’s president in 2002 todomestic and international acclaim, was particu-larly acute. He faced a demanding cocktail ofpopular expectations resulting from the combina-tion of three factors: an election triumph, a cul-tural faith in strong leaders and a widespread beliefthat the state can and should resolve social prob-lems. As leader of the left-wing Workers’ Party(PT), Lula was elected on a programme of radicalchange. He was expected to improve the quality oflife of the poor, many of whom were unaware ofthe practical limits of his authority.

To achieve his political objectives, Lula had towork with an independently-minded legislature inwhich his own PT, although relatively disciplined,won only a fifth of the seats. He had also to con-front free-spending federal governors and interna-tional agencies concerned about the government’sdebt. In his first year in office, the president suc-ceeded in satisfying both the poor and theInternational Monetary Fund, a rare combinationindeed. But Brazil’s fragmented political systemsets many traps for incoming presidents. In Brazil,as elsewhere in Latin America, many strong menhave ended their careers as weak presidents(Hagopian, 2004).

Parliamentary government

Where presidents are separate from the assemblyand independently elected, the executive in parlia-mentary systems is organically linked to the legis-lature (Figure 15.2). The government emergesfrom the assembly and can be brought down by avote of no confidence. By the same token the gov-ernment can, in most cases, dissolve parliamentand call fresh elections. If the paradox of presiden-tialism is executive weakness amid the appearanceof strength, the puzzle of parliamentary govern-

ment is to explain why effective government canstill emerge from this mutual vulnerability of theassembly and the executive.

The crucial influence on the operation of parlia-mentary government is the party balance in parlia-ment. Where a single party holds a majority in theassembly (as normally in Britain), government canbe stable and decisive, perhaps even excessively so.But the British model of single-party majority gov-ernment, based as it is on a plurality electoralsystem, is exceptional. Most other European coun-tries employ proportional representation, a systemwhich rarely results in a majority of seats for oneparty. The usual outcome is a coalition government.

So parliamentary government is as variable inoperation as the presidential form. In effect, theparliamentary system has two variants, one basedon majority government and the other on coali-tions. We will examine these two forms separatelybefore raising two broader questions about theformat. How is power in parliamentary govern-ment distributed between the prime minister,cabinet and ministers? And where does the head ofstate fit into the scheme?

274 GOVERNMENT AND POLICY

Figure 15.2 Parliamentary government

Voters

Legislature

Governing party orparties

Prime minister

Head of State

Cabinet

elect

approves andcan remove

formallyapproves

DefinitionParliamentary government has three main fea-tures:

� The governing parties emerge from theassembly. Government ministers are usuallydrawn from, and remain members of, the legis-lature.

� The head of the government (called primeminister, premier or chancellor) and thecouncil of ministers (usually called thecabinet) can be dismissed from office througha vote of no confidence by parliament.Thepost of prime minister is normally separatefrom that of a ceremonial head of state.

� The executive is collegial, taking the form of acabinet in which the prime minister is tradi-tionally just first among equals.This pluralexecutive contrasts with the focus in presiden-tial government on a single chief executive.

Note: In Norway and the Netherlands, ministers are not drawn fromthe legislature but are accountable to it.Their system of government isstill considered to be parliamentary.

Source: Adapted from Lijphart (1992).

Majority government

Britain is the classic example of parliamentary gov-ernment based on a single party with a securemajority. The plurality method of election used ingeneral elections usually delivers a workingmajority in the House of Commons to a singleparty. The leader of this party becomes primeminister (PM), selecting twenty or so parliamen-tary colleagues from the same party to form thecabinet. The cabinet provides the formal lynchpinof the system; it is the focus of accountability toparliament and even the strongest PM cannotgovern without its support. The cabinet meetsweekly, chaired by the PM. The monarch now sitsabove the entire political process, meeting regu-larly with the PM but rarely if ever intervening inpolitical decisions.

Government accountability to the House ofCommons is tight. All ministers, including thePM, must regularly defend their policies ‘in thehouse’; the opposition will demand a vote of noconfidence whenever it senses an advantage from

launching an attack. However as long as the rulingparty retains its majority, this accountability willnot threaten the government’s survival. So inBritain, the governing party spans the cabinet andthe assembly. Through party discipline, the execu-tive dominates the Commons, controlling itsagenda and timetable. The cabinet is officially thetop committee of state but it is also an unofficialmeeting of the party’s leaders. As long as seniorparty figures remain sensitive to the views of theirbackbenchers (and often even if they do not), theycan control the Commons. The government doesindeed emerge from its womb in the Commonsbut it dominates its parent from the moment of itsbirth.

How is this control achieved? Each party has aWhip’s Office to ensure that backbenchers (ordi-nary MPs) vote as its leaders require. Even withoutthe attentions of the whips, MPs will generally toethe party line if they want to become ministersthemselves. In a strong party system such asBritain’s, a member who shows too much indepen-dence is unlikely to win promotion. In extremecases, MPs who are thrown out of their party fordissent are unlikely to be reelected by constituentsfor whom a party label is still key. In short, it is inmembers’ own interests to show public loyalty totheir party.

The importance of party to the effective opera-tion of parliamentary government is seen evenmore clearly in other British-style ‘Westminster’systems. In Canada, ‘party discipline is conformedto even more rigidly than in the UK’ (Brooks,1996, p. 100). In Australia, a cabinet is more ameeting of the ruling party than of the govern-ment; cabinet is captured by party. So where asingle party can count upon disciplined majoritysupport in the assembly, as in these Westminstersystems, parliamentary government can be deci-sive, perhaps excessively so.

Minority and coalition government

Most countries using parliamentary governmentelect their legislature by proportional representa-tion (PR), resulting in a situation where no oneparty gains a majority of seats. Here the tight linkbetween the election result and government for-mation weakens. In this more fragmented situa-

THE POLITICAL EXECUTIVE 275

tion, government takes one of three forms (Müllerand Strøm, 2000a):

� A majority coalition in which two or moreparties with a combined majority of seats jointogether in government. This is the mostcommon form of government across conti-nental Europe; it characterizes Belgium,Finland, Germany and the Netherlands.

� A minority coalition or alliance. These areformal coalitions or informal alliances betweenparties which, even together, still lack a parlia-mentary majority. Minority coalitions have pre-dominated in Denmark since the 1980s. Theywere also found in First Republic Italy(1947–92).

� A single-party minority government formed bythe largest party. Single-party cabinets com-prised about 30 per cent of continentalEuropean governments between 1945 and 1999and are customary in Sweden (where the SocialDemocratic Party has been preeminent) andNorway.

Custom and practice provide part of the expla-nation for these contrasts. Once one particulartype of government has emerged, this formatbecomes a point of reference for party leaders afterthe next election. But the constitution, throughboth its statements and its silences, is more funda-mental. The constitution lays out the hurdles anew government must clear before taking office.

As indicated in Box 15.3, some constitutionsdemand that the legislature demonstrates majority

support for the new government through a formalvote of investiture. Germany and, until 1995,Belgium are examples. This requirement for a positive vote by the assembly encourages broadcoalitions.

Since 1975, Sweden has adopted a less stringenttest: a vote is held but the test is negative. That is,the prime minister proposed can form a govern-ment as long as no more than half of the membersof the Riksdag object (Bergman, 2000). However,in other countries the constitution is silent on theprocedure for approving a new government. Inthese circumstances, the new government takesoffice, and continues in power, until it is voteddown by the assembly. Denmark and Norway areexamples here. These less demanding conventionsfacilitate the formation and survival of minoritygovernments, sometimes with the tacit support ofother parties.

How then do coalition governments emerge fol-lowing an election or indeed when a governmentfalls mid-term? What procedure do parties followin making a coalition? Again, the answer varies. Ina few countries, parties state their preferred part-ners during the election campaign, before theresults are known. In Germany and Norway, pre-election coalition statements are an establishedfeature, encouraged in Germany by a mixed elec-toral system which gives each elector two separatevotes (Saalfeld, 2000). When combined with therather small number of parties represented in theGerman Bundestag, these pacts simplify post-elec-tion coalition formation.

More often, the party composition of govern-

276 GOVERNMENT AND POLICY

Installation procedure Description Example

Positive investiture vote To take office, a new government must obtain majority Germanysupport in parliament

Negative investiture vote A new government takes office unless voted down by a Sweden majority in parliament

No investiture vote No formal parliamentary vote is required before a new Denmarkgovernment takes office

B OX 1 5 . 3

Procedure for installing a government in parliamentary systems when noparty possesses a majority of seats

ment is decided after the election, through intri-cate negotiations between the leaders of the rele-vant parties. While this activity is underway, theoutgoing government remains as a caretakeradministration, for an average period of 30 days.

Some constitutions specify a procedure whichthe parties follow in forming, and not merelyinstalling, a government. This usually involves thehead of state appointing the largest party, or itsleader, as formateur. The formateur’s task is to forman administration through negotiation (Figure15.3). The parties agreeing to go into coalitionwill detail a joint programme in a lengthy publicstatement. These package deals cover the policiesto be pursued and the coalition’s rules of conduct:for instance, how it proposes to resolve disputes.Questions of office – which party obtains whichministry and whether that party alone determinesthe appointment – are also negotiated if not published.

Definitions A minimum winning coalition (MWC) is a gov-ernment formed by the smallest number ofparties (typically two to four) which together cansecure a parliamentary majority.An oversize coalition contains more partiesthan are needed for a MWC.A rainbow coalition is a coalition combiningseveral colours from the ideological spectrum,often red (left-wing) and green (environmental).A formateur is a person or party charged by thehead of state with starting negotiations for acoalition.The formateur is usually the leader ofthe party with most seats in parliament.

Further reading: Laver (1996).

Why do some parties end up in a governingcoalition while others do not? The most commoncoalition contains the minimum number of partiesneeded to make a viable government. Partyinterest favours minimum winning coalitions sinceincluding additional parties in a coalition wouldsimply dilute the number of government postsobtained by each of the participating parties.

In addition, coalitions are usually based onparties with adjacent positions on the ideologicalspectrum. In most Scandinavian countries, coali-

tions draw only from parties within one of the twobloc: the bloc on the left, or the bloc on the right.Where such clearly defined blocs do not exist, thepreference for policy-connected coalitions benefitspivot or swing parties in the centre which are ableto jump either way. In Germany, for instance, thesmall liberal Free Democrat Party participated inmost coalition governments between 1949 and1998, sometimes with the more conservativeChristian Democrats and sometimes with themore left-wing Social Democrats.

Occasionally, coalitions emerge which span theideological spectrum. An influential example is thefive-party rainbow coalition, ranging from theconservative National Coalition to the LeftAlliance, which emerged in Finland in 1995 andwas renewed in 1999 (Jungar, 2002). One reasonwhy rainbow coalitions are unusual is that they areoversized, exceeding the minimum number of par-ticipants needed for a majority.

Coalition governments are frequently con-

THE POLITICAL EXECUTIVE 277

Figure 15.3 Formateurs and coalition governments

Election held

Head of state appoints formateur

Formateur proposescoalition

Vote of investiture (if required)

Head of state formally appoints

government

Government loses no confidence vote

and resigns

Source: Adapted from Gallagher et al. (2000) fig. 2-1.

demned as unstable, not least by English-speakingcritics of proportional representation (p. 152).How valid is this charge? Certainly, in a few coun-tries government duration has been measured inmonths rather than years: an average of fivemonths for the French Fourth Republic(1945–58) and of eight months for Italian govern-ments between 1948 and 1989. In these over-citedexamples, chronic instability certainly contributedto poor governance.

But in most of contemporary Europe, coalitiongovernments last a good deal longer, typically foraround two years (Müller and Strøm, 2000b).Some coalition agreements in the Netherlands(and most in Austria and France) include a clausestating that the partners will call an election if theydissolve the coalition. This election rule, as it iscalled, gives the partners an incentive to soldieron. Coalitions require compromise over policyand posts but compromise can be a route to sta-bility rather than instability.

Just as formal rules influence government for-mation, so too can they affect government sur-vival. Two procedures from Germany illustrate thispoint. First, the constructive vote of no confi-dence, under which a chancellor’s government canonly be brought down by electing his successorwith majority support, enhances governmentdurability. And second, the chancellor cannot dis-solve parliament and call new elections – or,rather, the chancellor can only request the presi-dent to dissolve the Bundestag if he has first lost avote of confidence. More often than not, Germancoalitions survive the four years until the nextscheduled election.

DefinitionThe constructive vote of no confidencerequires an assembly to select a new prime min-ister in order to dispose of the incumbent.Thepurpose is to prevent legislatures from actingdestructively by bringing down a governmentwithout any thought to its successor.The devicecomes from Germany but has also been adoptedin Hungary and Israel.

In any case, a slight modification of the partiesin government does not constitute a politicalearthquake in European countries accustomed to

coalitions. When a government loses a vote ofconfidence mid-session and resigns, normally as aresult of a defection of one of the coalition part-ners, the procedure for forming governmentsbegins anew, often without troubling the voters.Indeed, in Norway the parliament cannot be dis-solved during its four-year term. In some coun-tries, many of the same characters will return topower, often to the same ministries, so that con-tinuity is maintained. In Italy, for instance, onegovernment was defeated only for the exact sameparties to resume office when no other combina-tion proved to be feasible. Observers accustomedto single-party government often exaggerate theinstability associated with installing a new coalition.

Who governs?

Parliamentary government lacks the clear focus ofthe presidential system on a single chief executive.Rather, it involves a subtle, variable and evolvingrelationship between the prime minister, thecabinet and government ministers. For advocatesof parliamentary government, collective leadershipis its key strength, generating more deliberationand so resulting in fewer mistakes than occurunder a presidential format. By contrast, criticsallege that the emphasis on committee decision-making in parliamentary government yieldssecond-rate compromises and frequent opportuni-ties to pass the buck.

We can distinguish three broad patterns in theorganization of top decision-making in parliamen-tary systems:

� cabinet government: the cabinet determinesoverall policy through discussion

� prime ministerial government: the PM is thedominant figure, dealing directly with indi-vidual ministers

� ministerial government: ministers operate withlittle direction from the PM or the cabinet.

Finland provides an exceptionally clear case ofcabinet government. By law, the Finnish StateCouncil is granted extensive decision-makingauthority. In 1988, for example, it handled 4,472agenda items in 81 formal sessions (Nousiainen,

278 GOVERNMENT AND POLICY

1994, p. 91). Both the prime minister and indi-vidual ministers are subject to constraints arisingfrom Finland’s complex, and now rainbow, coali-tions. Prime ministers are little more than chairs ofcouncil meetings; individual ministers also findtheir hands tied by their party and its coalitionagreements. The pattern is similar in theNetherlands, where a small cabinet of about 15ministers meets about twice a week and altersabout one in five of the proposals put to it(Andeweg and Irwin, 2002, p. 116).

Germany, by contrast, is an example of primeministerial government, called chancellor democracyin Germany. The guiding principle here is hier-archy rather than collegiality. The Bundestag(lower house) appoints a chancellor, not a party,and accountability to the Bundestag is mainlythrough the chancellor. He answers to parliament;ministers answer to him. The strong position ofGermany’s chief executive derives from the consti-tution (Basic Law). This states that the ‘chancellorshall determine, and be responsible for, the generalpolicy guidelines’. So the buck stops with theChancellor, as it does with the American president.

Prime ministerial government is more likely toemerge when the premier has the capacity tochoose and dismiss ministers. In Britain, subjectto some constraints of party balance, PMs form agovernment of their choosing from the talentavailable in the parliamentary party. Cabinetreshuffles and government restructurings are bothwithin the prime minister’s gift. The Britishcabinet now acts mainly as a ratifying body – anumpire rather than a decision-taker. In Blair’sBritain, for example, cabinet meetings rarely lastmore than an hour; some are over in thirtyminutes (Holliday, 2000, p. 89).

Ministerial government arises when ministersoperate without extensive direction from eitherprime minister or cabinet. In Germany, the chan-cellor sets the guidelines but the constitution goeson to say that ‘each Federal Minister shall conductthe affairs of his department autonomously and onhis own responsibility’. So Germany operates min-isterial government within the framework ofChancellor democracy. Ministerial governmentemerges when, as in Germany, department headsare appointed for their knowledge of the field andare expected to use their professional experience to

shape their ministry’s policy. At least until the late1960s, the Netherlands was another example: ‘abanker or economist at Finance, a lawyer atJustice, a farmer or farmer’s son at Agriculture’(Andeweg and Irwin, 2002, p. 116).

Ministerial government also emerges in coali-tions where parties appoint their own people tohead particular ministries. In the Netherlands, theprime minister neither appoints, dismisses norreshuffles ministers. Cabinet members serve withbut certainly not under their ‘leader’. In these con-ditions the premier’s status is diminished, withministers owing more loyalty to the party thatappoints them than to the prime minister. Thechief executive is neither a chief nor an executivebut a skilled conciliator. Similarly, the Italian con-stitution enjoins collective responsibility of theCouncil of Ministers to parliament but this hasbeen notably lacking. Especially in the FirstRepublic (1947–92), interdepartmental coordina-tion of policy was notoriously weak as factionswithin parties – and not just parties themselves –came to own particular ministries.

Trends in parliamentary government

In assessing the balance of power within the par-liamentary executive, we must address trends inthe relationship between prime minister andcabinet and in the workings of cabinet itself. Inview of the fundamental differences between presi-dential and parliamentary government, we canimmediately dismiss the popular thesis that primeministers are becoming ‘presidential’ in theirpowers (Foley, 2000). Given the structural con-trasts between the two systems, such a metaphorshould be resisted.

However, the broad trend is undoubtedly forprime ministers, both in single-party and coalitiongovernments, to acquire more weight in relationto the cabinet. This tendency reflects three factors:increasing media focus on the premier, thegrowing international role of the chief executiveand the emerging need for policy coordination asgovernance becomes more complex (King, 1994a).

Two less publicized but equally significantchanges have taken place in the operation of thecabinet system. First, the expansion of the cabinethas been contained at around two dozen by intro-

THE POLITICAL EXECUTIVE 279

ducing the category of non-cabinet ministers. Tobelong to ‘the ministry’, to use a New Zealandphrase, no longer guarantees a seat round thecabinet table. In Canada, the same device of non-cabinet ministers has been used to contain acabinet which had reached the unwieldy size of 40by 1987.

DefinitionCabinet committees are small workgroups ofthe full cabinet, established to focus on specificareas such as the budget, legislation or overallstrategy. In addition to these standing commit-tees, prime ministers also set up ad hoc commit-tees of ministers to respond to specific issuessuch as labour disputes or terrorist acts. Cabinetcommittees are much closer to the point of deci-sion than the full cabinet.

Second, cabinet committees have emerged askey decision arenas in larger countries. These com-mittees developed during and after wars, reflectingthe volume and urgency of business, and thenobtained more formal status in peacetime. NewZealand has around 12 committees, including aninfluential one that addresses overall strategy. InAustralia, decisions made by cabinet committeescan only be reopened in full cabinet with theapproval of the prime minister.

So cabinet government, to the extent that it stillexists, has become government by the cabinetnetwork, with the real decisions merely confirmedin full cabinet. Indeed, even these committeesnow largely ratify decisions fixed up before themeeting through informal consultations usuallyled by the minister most directly concerned. Thegoverning capacity of prime ministers nowdepends on their ability to manipulate the entirenetwork of cabinet committees, not simply theincreasingly formal meetings of the full cabinet(Dunleavy, 2003).

Heads of state and parliamentary government

One hallmark of a parliamentary system is, inBagehot’s classic analysis, the distinction betweenthe ‘efficient’ and ‘dignified’ aspects of government(Bagehot, 1963). Where presidential systems

combine the head of state and the head of govern-ment in one person, parliamentary rule separatesthe two roles. Efficient leadership rests with thecabinet, premier and ministers but dignified orceremonial leadership lies with the head of state.Such worthy activity does take some pressure offprime ministers, creating more time for them toconcentrate on the political aspects of their job.

The position of head of state is either inherited(a monarchy) or elected (a presidency) (Box 15.4).At least in Europe and Asia, royal heads of stateremain surprisingly numerous. Half the countriesof Western Europe are constitutional monarchies,including Belgium, Denmark, the Netherlands,Spain and the United Kingdom. In some formerBritish colonies such as Canada, a governor-general acts as a stand-in for the monarch.

In a democratic era, monarchs are reluctant toenter the political arena. Yet royal influence canoccasionally be significant, especially in times oftransition. In the 1970s, King Juan Carlos helpedto steer Spain’s transition to democracy. Morerecently, the King of Belgium played a conciliatoryrole in his country’s long march to federal status,leading Senelle (1996, p. 281) to claim that

were it not for the monarchy as symbol of thecohesion of the kingdom and therefore thevisible incarnation of federal loyalty, the Belgianexperiment would be doomed to failure.

Monarchies aside, most presidents in parliamen-tary systems are elected. Sometimes the presidentis directly elected (e.g. Ireland); in other cases heor she is elected by parliament (e.g. Israel).Alternatively, a special electoral college is used,often comprising the national legislature plus rep-resentatives of regional or local government (e.g.Germany).

Elected presidents have a little more latitudethan monarchs in addressing national issues(though far less, of course, than presidents in apresidential system). Many presidents in parlia-mentary systems do occasionally use their publicposition to nudge the political agenda towardsproblems receiving inadequate attention or to pro-posals which run counter to the national interest.For example, in 2003 the Italian President CarloCiampi refused to sign into law a media bill which

280 GOVERNMENT AND POLICY

would have strengthened the commercial interestsof the country’s Prime Minister, Silvio Berlusconi.Such presidential decisions can earn respect aslong as they are rare and based on an assessment ofthe national interest. But presidents must becareful to speak for the nation as a whole.

Semi-presidential government

Presidential and parliamentary governmentprovide the pure models of the political executive.The semi-presidential executive draws on bothformats, combining an elected president with aprime minister and cabinet accountable to parlia-ment. So with semi-presidential government weenter more varied territory. The French politicalscientist Maurice Duverger (1980) provided theclassic definition:

A political regime is considered semi-presidentialif the constitution which established it combinesthree elements: (1) the president of the republicis elected by universal suffrage; (2) he possessesquite considerable powers; (3) he has oppositehim, however, a prime minister and ministerswho possess executive and governmental powerand can stay in office only if the parliament doesnot show its opposition to them.

Thus, the semi-presidential executive is a hybrid,seeking to marry within the executive the nationalfocus of an elected president with a prime ministerresponsive to the interests represented in theassembly (Figure 15.4). The president in such asystem usually has special responsibility for foreignaffairs, can appoint ministers (including the primeminister), initiate referendums, veto legislationand dissolve the assembly. In theory, the presidentcan offer leadership on foreign affairs while theprime minister addresses the intricacies ofdomestic politics through parliament.

The president in a semi-presidential executive isin rather than above politics. As a two-headedsystem, the semi-presidential executive creates adivision of authority within the executive and, forthat reason, an invitation to struggle between pres-ident and prime minister.

If the United States exemplifies the presidentialsystem, the French Fifth Republic (1958– ) pro-vides the model for the semi-presidential execu-tive. The 1958 constitution was designed toprovide stable governance in the context of a divi-sive war in Algeria and the legacy of the unstableFourth Republic, which experienced 23 primeministers in its 12-year life.

The new constitution created a presidency fitfor the dominating presence of its first occupant,General Charles de Gaulle. Regarding himself as

THE POLITICAL EXECUTIVE 281

Country Head of state Method of selection Tenure

Austria President Direct popular election 6 years

Canada British monarch* Heredity Life#

Germany President Election by a joint Bundestag and Land 5 years convention

India President Election by a college of federal and state 5 years assemblies

Sweden Monarch Heredity Life#

United Kingdom Monarch Heredity Life#

* Represented in Canada by a governor general appointed for five years by the monarch on a recommendation from Canada’s prime minister.# Or until abdication.

B OX 1 5 . 4

Selecting the head of state in some parliamentary democracies

national saviour, de Gaulle argued that ‘poweremanates directly from the people, which impliesthat the head of state, elected by the nation, isthe source and holder of that power’ (Knapp andWright, 2001, p. 53). In office, de Gaulle’s impe-rious style developed the office to – perhaps evenbeyond – its constitutional limits.

DefinitionSemi-presidential government, sometimescalled the dual executive or premier-presiden-tialism, combines an elected president per-forming political tasks with a prime minister whoheads a cabinet accountable to parliament.Theprime minister, usually appointed by the presi-dent, is responsible for day-to-day domestic gov-ernment but the president retains an oversightrole, responsibility for foreign affairs and canusually take emergency powers.The Fifth FrenchRepublic is the leading example (Shugart andCarey, 1992).

Following a constitutional amendment in 1962,the president has been directly elected, fully estab-lishing the semi-presidential format (Elgie,1999a). The effect of the amendment, it is argued,was to create a powerful presidency where previ-ously there had just been a powerful president. Ina further amendment in 2000, the presidentialterm was reduced from a renewable seven years toa renewable five years.

The constitution certainly grants the Frenchpresident extensive powers. He (and as in the USAall so far have been men)

� is guarantor of national independence and theconstitution

� can take emergency powers (not invoked since1961)

� heads the armed forces and chairs the maindefence committee

� negotiates treaties� can call referendums (nine so far) � appoints some senior judges and civil servants � presides over the Council of Ministers � can dissolve the assembly (but not veto legisla-

tion)� formally appoints the prime minister, in prac-

tice from the party winning assembly elections.

In pursuing these roles, the president is sup-ported by an influential personal staff in theÉlysée Palace. So far, all five presidents havesought to govern in expansive style, seeking topilot the ship of state rather than just to arbitrateconflicts emerging from below decks.

What of prime ministers in France’s semi-presi-dential executive? Their main concern is domesticaffairs, casually dismissed by de Gaulle as ‘theprice of milk’. Appointed by the president butaccountable to parliament, the prime minister’stask is rarely simple. He or she (Edith Cresson wasPM 1991–92) directs the day-to-day work of thegovernment, operating within the president’s styleand tone. Since the government remains account-able to parliament, much of the prime minister’swork focuses on managing the National Assembly.The ability of the assembly to force the primeminister and the Council of Ministers to resignafter a vote of censure provides the parliamentarycomponent of the semi-presidential executive.

282 GOVERNMENT AND POLICY

Figure 15.4 Semi-presidential government

Voters

PresidentLegislaturecan dissolve

chooses andcan dismiss

Prime Minister and cabinet

Note: In semi-presidential government, the relationships betweeninstitutions varies more by country than is the case with pure parliamentary or presidential government.

accountableto

can bringdown

elect

THE POLITICAL EXECUTIVE 283

Just how different is modern France? The case forFrench exceptionalism can be stated in three words:the French Revolution. Of course, the shock waves of1789 reverberated throughout Europe as the destruc-tion of absolute monarchy laid the groundwork for theemergence of modern, secular nation-states. But therevolution created a distinctive ethos within Franceitself. Like other states built on revolution, notably theUnited States, France is an ideal as well as a country.But in France the ideals remained contested.

The legacy of the revolution is not just a disputedcreed of liberty, equality and fraternity. It is also awidespread belief that the state can and shouldimplement its ideas rationally, even against opposi-tion from a diverse and sometimes antagonisticsociety.The result, it is claimed, is a state-centredcountry characterized by haughty bureaucrats, exten-sive regulation, limited pluralism and a mass politicalculture which combines dependence on the statewith hostility towards it.

Yet French uniqueness certainly declined, and possiblydisappeared, in the second half of the twentiethcentury.The country became more modern, urbanand industrial, resulting in a more conventional andconsolidated democracy.The country’s retreat fromempire left France, like Britain, as a middle-rankingpower with a new base in the European Union.

France’s governing architecture of a semi-presidentialexecutive remains distinctive but it works: policies are

formed, decisions are reached. Even the judiciary hasacquired some weight.‘The revolutionary impulse isexhausted’, concludes Hayward (1994, p. 32). Francetoday seems to be a normal country, preoccupied withthe workaday question of maintaining economicgrowth, full employment and sound governmentfinances.

So contemporary France is certainly no longer a revo-lutionary society. But to portray France as just anotherdemocracy is perhaps to go too far. Inherited tradi-tions still condition the way France approaches its typ-ically European combination of a cosseted workforce,subsidized farmers and considerable but unequalaffluence. Public discourse still tends to assume thatthe state must be capable of both creating new jobsand protecting the rights of existing workers.Dirigisme (state direction), suggests Wright (1997), hasevolved rather than disappeared: the government isstill prepared to intervene in the economy, albeit in amarket-conforming way, when it perceives nationalinterests at stake.

In France, sovereignty is still presented as a cardinalvirtue, with the result that globalization is seen as athreat as much as an opportunity. American influ-ences, in particular, are strongly resisted. Even if l’ex-ception française is a myth, it is a legend whichremains important to the French way of politics.

Population: 59m.Gross domestic product per head:

$22,600.Form of government: a unitary democ-

ratic republic, headed by an electedpresident.The current Fifth Republicwas established in 1958.

Legislature: the lower chamber, theNational Assembly, contains 577directly elected members.The 321members of the Senate are indirectlyelected, a third at a time, through localgovernment for a long nine-year term.The legislature is weaker than in manydemocracies.

Executive: France’s semi-presidentialexecutive combines a strong presi-dent, directly elected for five years,with a prime minister who leads aCouncil of Ministers accountable tothe assembly.

Judiciary: French law is still based onthe Napoleonic Codes (1804–11).TheConstitutional Court has grown in sig-nificance during the Fifth Republic.

Electoral system: a two-ballot system isused for both presidential andassembly elections, with a majorityvote needed for victory on the firstround.The last experiment with PRwas in 1986 for the National Assembly.

Party system: left–right conflictremains a principle of French politics,though parties themselves have beenless stable. But the exceptional presi-dential election of 2002 failed todeliver the customary choice betweensafe candidates of the left and theright. Rather, Jacques Chirac of thecentre right was reelected in a run-offwith Jean-Marie Le Pen, leader of thefar right Front National. On this occa-sion, the socialist candidate wasdefeated in the first round.

Profile F R A N C E

Further reading: Knapp and Wright (2001), Schain(2004), Stevens (2003).

DefinitionCohabitation occurs in a semi-presidential exec-utive when president and prime minister aredrawn from different political camps.This situa-tion has occurred three times in the French FifthRepublic. Cohabitation intensifies competitionbetween the two principals and places the presi-dent in the awkward position of leading both thenation and the opposition.

The crucial relationship in the semi-presidentialexecutive is between the president, on the onehand, and prime minister and assembly, on theother. While the constitution may give control offoreign affairs to the president and reservedomestic policy to the prime minister, an interde-pendent world does not permit such pigeon-holing. France’s relationship with the EU, forinstance, encompasses both foreign and domesticaffairs, complicating the decision-making process.Before one EU summit, Germany’s chancellorinsisted on meeting the French president andprime minister together, to speed negotiations.

Presidents and prime ministers therefore need towork in harmony, a task made easier when theparty in the Elysée Palace also has a majority inthe assembly. This has been the case for most ofthe Fifth Republic. Indeed, the reduction of thepresident’s term to five years was partly an attemptto coordinate presidential and parliamentary elec-tion terms, limiting the likelihood of cohabitation.

When cohabitation does occur, as it has donethree times since 1986, presidential power tends toshrink. In these circumstances, prime ministersassert their constitutional duty to ‘determine anddirect the policy of the nation’. Crucially, though,cohabitation has not led to a crisis of the regime.The Fifth Republic has delivered the stability thatits architects intended. Just as the United Statescopes with power divided between the WhiteHouse and Congress, so the French experienceconfirms that the semi-presidential executive canprovide stable government even when presidentand prime minister are drawn from different polit-ical blocs.

Semi-presidential government has held particularappeal to European countries facing internationaldifficulties. France adopted a semi-presidential

solution in 1962 following serious unrest in itsAlgerian colony. Finland found a semi-presidentialsystem helpful in managing its sensitive relation-ship with its large Russian neighbour. And a dualexecutive also proved attractive to CentralEuropean states in the immediate aftermath ofcommunism’s collapse.

As international pressures recede, however, so thepresident’s star in this format of government tendsto wane. In 2000, France reduced the president’stenure from seven years to five. In the same year,Finland modified its constitution to strengthen the parliamentary element (Nousiainen, 2001).Further, some post-communist countries, such asHungary, have moved away from semi-presiden-tialism into a parliamentary form. The semi-presi-dential executive is more than a transitional device,but the system has not threatened the preeminenceof parliamentary government in Europe.

The executive in new democracies

The new democracies established in the finaldecades of the twentieth century have varied intheir choice of presidential or parliamentary gov-ernment. In general, post-military Latin Americahas retained presidential systems despite someserious academic prodding in the direction of par-liamentary regimes (Linz, 1990). South America’scontinuing bias to presidential rule, confirmed bya Brazilian referendum in favour of the form in1993, reflects a fear of social instability and vio-lence in the absence of strong political authority.Presidential rule does not suffice to resolve theunderlying tensions of unequal societies but, giventhe continent’s strongman tradition, the presiden-tial option is considered the safer choice.

By contrast, most post-communist democraticstates in central Europe have settled on parliamen-tary government, following a flirtation with semi-presidential systems. In the early years ofpost-communism, presidential authority was con-siderable, particularly where the initial incumbentswere endowed with the moral authority resultingfrom leading the revolt against communism, aswith Lech Walesa in Poland and Vaclav Havel inthe Czech Republic. As with General de Gaulle inFrance, these post-communist leaders could offer a

284 GOVERNMENT AND POLICY

symbol of stability in difficult times. Fitzmaurice(1998, p. 65) writes,

There was a need for a presidency that could, inthis extraordinary period of political turbulence,represent a centre of stability in shifting andoften unstable parliamentary situations. Therewas also a need for a moral symbol in difficulttimes and for a respected national representativeto the outside world who could be an asset onthe European and world stage, especially in thebattle for membership of the IMF, Council ofEurope, NATO and the EU.

But as these new European democracies havestabilized, so the parliamentary dimension hasgained ground. Thus Malovà and Haughton(2002) judge that

most Central and Eastern European constitu-tions now have a weak head of state, even wherethe president is directly elected, who playsmostly a symbolic role and intervenes in politicsonly in explicitly defined cases to limit institu-tional conflicts.

Even in Poland, where the president is directlyelected (Box 15.5), the new constitution of 1997softened the president’s veto powers. Millard(2003, p. 35) concludes that following thesereforms Poland ‘could no longer be viewed as asemi-presidential system’.

In post-communist settings, government itself –rather than the society it seeks to control – isdeemed to be the danger. Such democracies havesought to contain rather than to concentrate polit-ical power but have taken the parliamentary routeto this objective. They have adopted proportionalrepresentation in order to reflect social and polit-ical divisions within parliament. The coalitiongovernments which result, though rarely ascoherent or stable as those in Western Europe, dorepresent some sharing of political authority, atleast compared to the communist era.

So, although the experience of post-communistcentral Europe confirms that the dividing linebetween semi-presidential and parliamentary gov-ernment is not always clear, it also demonstratesthat parliamentary government remains Europe’snatural democratic form.

The executive in authoritarian states

In authoritarian states, formal executive structures– the executive, legislature and the judiciary – areless well-developed than in democracies. The topoffice may consist of a presidency (as in manycivilian regimes) or a ruling council (as in manymilitary regimes) but the central feature of thenon-democratic executive is its lack of institution-alization. The leader seeks to focus power onhimself and his supporters, not to distribute itamong institutions. Jackson and Rosberg’s idea of

THE POLITICAL EXECUTIVE 285

Directly Nominate PM Emergency Authority over Introduce elected? and dissolve powers? foreign and bills?

assembly? defence policy?

Czech Republic No Yes No No No

Hungary No Yes Yes Yes Yes

Poland Yes Yes No Yes Yes

Slovakia Yes Yes No No Yes

Source: Adapted from Baylis (1996). See also White et al. (2003).

B OX 1 5 . 5

Presidential powers in some new Eastern European democracies

personal rule (1982), developed in the context ofAfrican politics, travels widely through the non-democratic world. Politics takes precedence overgovernment and personalities matter more thaninstitutions: a feast of presidents but a famine ofpresidential systems.

DefinitionPersonal rule is defined by Jackson and Rosberg(1982, p. 19) as ‘a system of relations linking rulersnot with the “public” or even with the ruled butwith patrons, associates, clients, supporters andrivals who constitute the “system”.The system is“structured” not by institutions but by the politi-cians themselves.The fact that it is ultimatelydependent upon persons rather than institutionsis its essential vulnerability.’

The formula of weak institutions plus strongpolitics is ultimately unconvincing. It results incharacteristic ailments, including struggles oversuccession, insufficient emphasis on policy, poorgovernance and even a danger of regime collapse.In particular, the lack of a succession procedure(excepting hereditary monarchies) can create aconflict among potential inheritors not just afterthe leader’s exit but also in the run-up to it.Authoritarian leaders keep their job for just aslong as they can ward off their rivals. They mustmount a constant threat watch and be prepared toneuter those who are becoming too strong. Politicscomes before policies.

The price of defeat, furthermore, is high; politicscan be a matter of life and death. When anAmerican president leaves office, he can retire to hislibrary to write his memoirs; yet ousted dictatorsrisk a harsher fate. By necessity, the governing styleof non-democratic rulers inclines to the ruthless.

We will use post-colonial Africa, the MiddleEast and post-communist countries in central Asiato illustrate these themes, before turning to theexecutive under totalitarian rule.

Africa

Until the era of democratization in the 1990s,post-colonial Africa illustrated the weakness ofgoverning institutions and the importance of per-sonal leadership. Leaders were adept at using the

coercive and financial resources of the regime toreward their friends and punish their enemies. AsSandbrook (1985) wrote of Mobutu Sese Sekoduring his long tenure as president of Zaire:

No potential challenger is permitted to gain apower base. Mobutu’s officials know that theirjobs depend solely on the president’s discretion.Frequently, Mobutu fires cabinet ministers,often without explanation. Everyone is kept offbalance. Everyone must vie for his patronage.

In post-colonial Africa, as in most authoritarianstates, personal rule was far from absolute.Inadequately accountable in a constitutional sense,many personal rulers were highly constrained byother political actors. These included the military,leaders of ethnic groups, landowners, the businessclass, the bureaucracy, multinational companiesand even factions in the leader’s own court. Tosurvive, leaders had to distribute the perks ofoffice so as to maintain a viable coalition ofsupport drawn from these groups. If they madetoo many false moves in the patronage game, theycould find themselves replaced by a more skilfulplayer. These domestic imperatives explain whypresidents ignored the preaching of internationalagencies in favour of a competitive marketeconomy. Presidents preferred to invest in theirown political future rather than long-term eco-nomic development.

Middle East

In the Middle East, personal rule remains central toauthoritarian rule (Bill and Springborg, 2000). Inmany of these oil-rich kingdoms, shahs, sheikhs andsultans continue to rule in traditional patriarchalstyle. ‘Ruling’ rather than ‘governing’ is the appro-priate term. In Saudi Arabia, advancement withinthe ruling family depends not on merit but oncloseness to the family’s network of advisers, friends,flatterers and guards. Public and private are notsharply distinguished; each forms part of the ruler’ssphere. Again, government posts are not secure butare occupied on good behaviour, as demonstratedby unswerving loyalty to the ruler’s interests.

Such systems of personal rule have survived forcenturies, limiting the development of strong gov-

286 GOVERNMENT AND POLICY

ernment institutions. Leaving succession difficul-ties to one side, the paradox of Middle Easternpolitics is that personal rule itself constitutes astable system, providing an exception to thegeneral theme of instability in the authoritarianexecutive. It remains to be seen whetherSeptember 11, 2001 has inspired America toengage with a project of liberalization and democ-ratization in the Middle East which could seri-ously threaten this highly traditional order.

Central Asia

Personal rule also characterizes the post-commu-nist states of central Asia such as Kyrgyzstan andUzbekistan. While central Europe has moved in ademocratic direction, many successor republics tothe Soviet Union have seen the rise of authori-tarian regimes with strong, personalized and barelyaccountable presidents. In these impoverished andmainly agricultural republics, where democracyhas never flourished and experience of indepen-dent statehood is limited, rulers are concernedwith power and voters with their daily struggle.Neither group cares greatly for structures of gov-ernment.

In the central Asian republic of Uzbekistan, forexample, the presidency is strongly personalized:

power resides as much in the person of the presi-dent as in the office. The Uzbek presidency isnot just a formal power position; it is also thecenter of an extensive informal network ofregionally-based, patron–client ties. The presi-dent is, in effect, the chief patron.

(Easter, 1997)

The totalitarian executive

We might expect the institutions of government tohave played a central role in the totalitariansystems of fascism and communism. After all,these were political regimes par excellence, seekingto lead and transform society. But again, personalrule dominated, at least under fascism.

In fascist thinking, the personal authority of theruler rather than the institutions of state and partyprovided the driving force. The state was idolizedmore than institutionalized and the party was

exploited by Hitler and Mussolini as a device forobtaining power. Once power was achieved, stateand party merged as personal vehicles of thesupreme ruler. The leader defined the interests ofthe regime and duties owed to the dictator tookpriority over any obligations to state or party. InNazi Germany, for example, the notorious SS(Schutzstaffel, protection units) began as a partysecurity force but underwent a huge expansionafter Hitler came to power, becoming the personalinstrument of the Führer (Brooker, 2000, p. 136).

Nazi Germany, less so Mussolini’s Italy, was apolitical, or even a revolutionary, regime in whichpower bases were established informally withHitler acting as an arbiter of conflict among thebarons beneath. Power was neither contained ininstitutions nor exercised through bureaucraticrules; had it been, the Führer’s personal ascen-dancy would have been threatened.

The situation differed substantially in commu-nist states. Here, certainly, power was more regu-larized. However, the party was the driving force,offering political direction to the state. Of course,most communist states did have a clear structureof government. This resembled the parliamentaryform (Figure 15.5). A presidium (equivalent to acabinet) was headed by a chairperson (prime min-ister). This presidium was, in turn, an innersteering body of a larger Council of Ministerswhich was itself formally ‘elected’ by the parlia-ment. An honorific post of state president alsoexisted.

THE POLITICAL EXECUTIVE 287

Figure 15.5 Typical executive structure in communist states

Presidium of theCouncil of Ministers

Supreme Soviet

Council of Ministers

Voters

But in practice the ruling communist partydominated the formal institutions of the state; thekey post was general secretary of the party, notchairman of the presidium. The party secretaryoften confirmed his supremacy by taking a statepost, whether prime minister or president.However, power remained rooted in the party; ifthe top leader lost his position as party secretary,he was politically doomed.

China offers a partial exception to this picture ofthe communist executive. Chinese politics hasalways been more fluid and personal than in othercommunist states; some past leaders have notoccupied any formal positions at all, whether inthe party or the government. As is customary incommunist countries, the state is led by the party.But the party itself is divided into factions basedon personality and patronage more than policy. Itis these factions within the party which providethe glue of politics, binding the actors together ina predictable but non-institutionalized way.

In China’s factional environment, the top figurerequires cunning and patronage to prosper.Indeed, China confirms that even this massivecommunist state can be ruled in a highly personalway. As Saich (2004, p. 83) observes,

Personal power and relations with powerful indi-viduals are decisive throughout the Chinesepolitical system. While this may decline asreforms become more insitutionalized, mostChinese recognize that the best way to surviveand flourish is to develop personal relationships(guanxi) with a powerful political patron.

Authoritarian rule and despotism

The personal basis of rule in many non-democ-ratic regimes poses an important question: aresuch states are always likely to degenerate intodespotism and even barbarism? Are the murderousexcesses of twentieth-century dictators an inherentrisk of failing to subject power to constitutionallimits?

Certainly, some communist states fell victim tobrutal tyranny, despite – or because of? – thestrong organization provided by the communistparty. The Soviet Union under Joseph Stalin wasan extreme but important example. After Lenin’s

death in 1924, Stalin slowly acquired ascendancyover the party. He methodically picked off theother party leaders, becoming undisputed ruler ofthe Soviet Union by 1929 and remaining so untilhis death in 1953. Having risen to power throughthe party, he eventually reshaped it into one of hispersonal instruments, alongside the political policeand the state bureaucracy.

Stalin’s methods were fierce indeed:

In the Great Terror of 1937–38, the politicalpolice arrested some five million people, exe-cuted over 800,000 of them and incarceratedmost of the remainder in prisons or labourcamps. Among the victims were a large majorityof the party’s Central Committee and more thanhalf the delegates to the most recent PartyCongress. The Terror destroyed the party as anindependent political entity; it was now offi-cially described as a party of Lenin-Stalin fol-lowers.

(Brooker, 2000, p. 138)

One-man rule and extreme leadership cults werefeatures of many other communist states.Examples include Mao Zedong in China, Castroin Cuba, Ho Chi Minh in Vietnam, Tito inYugoslavia, Ceaucescu in Romania and Kim IlSung in North Korea. And such cults of person-ality were often linked with extreme brutality,albeit rarely on the scale of Stalin’s Russia.

But not all totalitarian, and still less all authori-tarian, regimes descended into barbarity. After all,Stalin drew not just on communist ideology andparty organization but also on Russia’s highlyautocratic history. Elsewhere, communist rule wasless severe. In Eastern Europe, rulers in late com-munism sought to soften the harshness of Sovietdomination. Some national leaders were anony-mous party functionaries. Janos Kadar, the harm-less leader of Hungary between 1958 and 1988,rode the tramcar to work each day, unrecognizedby his people. The communist experience suggeststhat personal despotism is far from an inevitableconsequence of authoritarian rule.

In any case, many non-communist authoritarianrulers have lacked the means of social controlneeded to indulge in Stalinist terror. The experi-ence of black Africa and the Middle East is that

288 GOVERNMENT AND POLICY

personal rule is usually far from despotic. It can bea balanced and stable, if rarely progressive, form ofgovernance.

Key reading

Next step: Lijphart (1992) remains an excellentcollection on parliamentary and presidential gov-ernment.

Much recent work on executive institutions hasfocused on presidential government (includingLatin America): for instance, Linz and Valenzuela(1994), Mainwaring and Shugart (1997b) andShugart and Carey (1992). On semi-presidential

government, the original source is Duverger(1980); Elgie (1999b) provides a more recent per-spective. On the American presidency, see Rose(2000) and the classic study by Neustadt (1991).Comparative but contrasting studies of parliamen-tary government include Laver and Shepsle (1994)and Müller and Strøm (2000a). For comparativestudies of prime ministers see Wright et al. (2000);Rose (2001) examines the British prime ministerin an interdependent world. The executive in post-communist Europe is covered in White et al.(2003). On personal rule, Jackson and Rosberg(1982) is the classic analysis, using African exam-ples; Bill and Springborg (2000) examine theMiddle East. Brooker (1995) assesses twentieth-century dictatorships.

THE POLITICAL EXECUTIVE 289

Evolution 291Recruitment 292Organization 293Accountability 299New public management 301Bureaucracy in new democracies 303Bureaucracy in authoritarian states 304Key reading 307

The bureaucracy is the state’s engine room. Itconsists of permanent salaried officials

employed by the state to advise on, and carry out,the policies of the political executive. The bureau-cracy is indispensable to modern government yet,lacking the legitimacy of election, it has alwaysaroused controversy. The central strands of thisdebate are responsiveness and efficiency. Onresponsiveness, the question is: how can civil ser-vants be made accountable to the politicians theynotionally serve? And on effectiveness, the issue is:how in the absence of a competitive market canwe ensure that the bureaucracy does its job withefficiency and economy? This theme of value formoney became particularly prominent in the eraof lean government in the 1990s.

Delimiting the bureaucracy raises tricky issues ofdefinition. Reflecting the complexity of moderngovernment, public employees have a range ofemployment relationships with the state (Figure16.1). The broadest term is the public sector, alsocalled the public service or public administration.

This covers all employees whose salary comesdirectly or indirectly from the public purse.However, the public sector includes several areasnot normally counted as part of the civil service,such as teachers and the armed forces. Civil ser-vants proper are usually defined as employees whoare

� paid directly by the national exchequer� subject to the state’s conditions of service

(including access to its pension scheme)� engaged in shaping or more commonly imple-

menting government decisions.

Just to complicate matters, a few countries suchas Germany do extend civil service status toteachers, even though this group works at oneremove from the government.

DefinitionBureaucracy is, literally, rule by officials.Theword ‘bureau’ comes from the Old French termburel, meaning the cloth used to cover a writingdesk or bureau.The second half of the wordcomes from the Greek kratos, meaning rule, justas in democracy.Today, the bureaucracy refers tothe salaried officials who conduct the detailedbusiness of public administration, advising onand applying policy decisions.

The number of civil servants with a direct policy-advising role, the group of special political interest,is no more than a few thousand. This is becausemost civil servants work in the field, applyingpolicy away from the decision-making centre.Traditionally, the ‘mandarins’ (to use the oldChinese term for high-level bureaucrats) formed aspecial grade in the civil service, often filled byable graduates recruited straight from university.However, some countries have now introducedmore flexible practices at this highest level, makingmore use of short-term contracts and open recruit-

Chapter 16

The bureaucracy

Figure 16.1 Delimiting the bureaucracy

The public sector

The civil service

High level policy advisers

290

ment. Examples include the Senior ExecutiveServices established in the United States (1978),Australia (1984) and New Zealand (1988).

While the word ‘bureaucracy’ is often useddescriptively as a synonym for the civil service, it isalso employed in a more abstract way as a modelfor organizing public administration. The analysisof bureaucracy presented by the German sociolo-gist Max Weber (1864–1920) is preeminent here;it is an important model to examine. Weber con-ceived of bureaucracy as a structured hierarchy inwhich salaried officials reached rational decisionsby applying explicit rules to the facts before them.Specifically, Weber’s model contains the followingfeatures:

� Bureaucracy involves a carefully defined divi-sion of tasks.

� Authority is impersonal, vested in the rules thatgovern official business. Decisions are reachedby methodically applying rules to particularcases; private motives are irrelevant.

� People are recruited to serve in the bureaucracybased on proven or at least potential compe-tence.

� Officials who perform their duties competentlyhave secure jobs and salaries. Competent offi-cials can expect promotion according toseniority or merit.

� The bureaucracy is a disciplined hierarchy inwhich officials are subject to the authority oftheir superior.

Weber’s central claim was that bureaucracy madeadministration more efficient and rational; hebelieved that it was the means by which modernindustrial efficiency could be brought to bear oncivil affairs. To quote Weber himself:

the fully developed bureaucratic apparatus com-pares with other organizations exactly as doesthe machine with non-mechanical modes ofproduction. Precision, speed, clarity, knowledgeof files, continuity, discretion, unity, strict sub-ordination, reduction of friction and of materialand personal costs – these are raised to theoptimum in the strictly bureaucratic administra-tion.

(Gerth and Mills, 1948)

For Weber, the ideal bureaucracy was a finepiece of administrative machinery. But like manymodern devices, bureaucracy brought the risk ofdominating its supposed masters. Weber’s contri-bution was therefore to pose the question of therelationship between bureaucracy and democracy,a concern that inspired much discussion in thetwentieth century.

Evolution

To appreciate why Weber considered modernbureaucracy so efficient, we must consider whatpreceded it. As with other aspects of government,the precursors varied between Europe and theNew World. In Europe, clerical servants were orig-inally agents of the royal household, serving underthe personal instruction of the ruling monarch(Raadschelders and Rutgers, 1996). Many featuresof modern bureaucracies – regular salaries, pen-sions, open recruitment – arose from a successfulattempt to overcome this idea of public employ-ment as personal service to the monarch. Indeed,the evolution of royal households into twentieth-century Weberian bureaucracies was a massivetransformation, intimately linked to the rise of themodern state itself. It was a transition from patri-archy to bureaucracy, a story of the depersonaliza-tion of administration as the royal household wasslowly converted into public service (Hyden,1997, p. 243). Today, we take the features ofbureaucratic organization for granted but forWeber the form was strikingly new, a modern phe-nomenon to be both admired and feared.

In the New World, however, civil service devel-opment was more pragmatic. Lacking the longstate tradition of West Europe, public administra-tion was initially considered a routine applicationof political directives. In the United States, theoriginal philosophy was that almost every citizenwas qualified for almost every public job; indeed, aprofessional civil service was considered somewhatundemocratic (Christensen and Peters, 1999, p.100). This populist theory conveniently under-pinned the spoils system, a term deriving from thephrase ‘to the victor, the spoils’. Spoils meant thatsuccessful candidates, including newly-electedpresidents, were expected to distribute government

THE BUREAUCRACY 291

jobs to those with the foresight to support thewinning candidate. The spoils system continued atleast until 1883 when the Pendleton Act created aCivil Service Commission to recruit and regulatefederal employees. So where a merit-based bureau-cracy had emerged in Europe in reaction tomonarchy, in the USA a professional administra-tion supplanted spoils.

In the twentieth century, bureaucracies reachedtheir zenith. The depression and two world warsvastly increased government intervention insociety. The welfare state, completed in Europe inthe decades following the Second World War,required a massive Weberian bureaucracy to dis-tribute grants, allowances and pensions in accor-dance with complex eligibility conditions set bypoliticians. By 1980, public employmentaccounted for almost a third of the total work-force in Britain and Scandinavia, though much ofthe expansion had occurred at local level.

However, the final quarter of the twentiethcentury witnessed declining faith in governmentand, more to the point, deteriorating publicfinances. Seizing on this fiscal crisis, right-wingpoliticians such as Ronald Reagan and MargaretThatcher called for, and to an extent delivered, notjust a reduced role for the state but also a changingstyle of bureaucratic operation away from strictWeberian guidelines. As a result, the contempo-rary emphasis is on management, results and effi-ciency. At the start of the twenty-first century,modern civil servants therefore face conflictingexpectations. At one and the same time, they areexpected to show flexibility but also to abide byrules; to deliver results but also to work within setprocedures; and to act decisively while also con-sulting widely.

Recruitment

Recruitment to bureaucracies has evolved intandem with the development of the civil serviceitself. The shift from patrimonial to Weberianbureaucracies was a transition from recruitment bypersonal links with the ruler to open selection onmerit. Jobs became available, at least in theory, tothe whole population. Even though these reformsoccurred in most democracies as long ago as the

late nineteenth century, recruitment to the civilservice remains an important theme. Selectionmethods and employee profiles are scrutinizedmore carefully than in the private sector. Further,what counts as ‘merit’ still varies between coun-tries, revealing subtle contrasts in conceptions of acivil servant’s role.

Britain exemplifies a unified (or generalist) tra-dition. Indeed, it pushes the cult of the amateur toextremes. Administration is seen as the art ofjudgement, born of intelligence and honed byexperience. Specialist knowledge should be soughtby bureaucrats but then treated with scepticism;experts should be on tap but not on top.Recruiters look for general ability, not technicalexpertise. All-round ability should enable suc-cessful candidates to acquire or at least interpretwhatever technical knowledge is needed; the keyrequirement is not knowledge but the ability tolearn. For the same reason, a good administratorshould be able to serve in a variety of departmentsand will be more rounded for doing so.

DefinitionIn a unified (or generalist) bureaucracy, recruit-ment is to the civil service as a whole, not to aspecific job within it. Administrative work is con-ceived as requiring intelligence, education andappropriate personal skills but not technicalknowledge. By contrast, a departmental (or spe-cialist) approach recruits people with technicalbackgrounds to a specific department or job.Unified bureaucracies are career-based whiledepartmental civil services are job-based.

An alternative method of pursuing the unifiedapproach is to recruit to a corps (body) of civil ser-vants rather than to a specific job in a ministry.The French civil service, for example, recruits civilservice through competitive examinations to suchbodies as the Diplomatic Corps or the FinanceInspectorate. Certainly, these corps offer their ownspecialized post-entry training but the seeminglytechnical basis of this induction is rather mis-leading. Although civil service recruitment is intheory to a specific corps, it is in reality as much toan elite which encompasses both public andprivate realms. Even within the civil service over athird of corps members are working away from

292 GOVERNMENT AND POLICY

their home corps at any one time. At its highestlevels, the French bureaucracy is clearly generalistalbeit with corps that provide more of a bow tospecialized training than is offered in Britain.

Some unified civil services stress one particularform of technical expertise: law. In manyEuropean countries with a codified law tradition,a legal training is common among higher bureau-crats. Germany is a good illustration. Over 60 percent of top German civil servants are lawyers,compared to just 20 per cent in the United States.The German model has influenced other coun-tries, notably Japan, where the recruitment base isnarrower still. Most of those who pass through the‘dragon gate’ examination for recruitment to high-level positions in the Japanese civil service aregraduates of just one department: TokyoUniversity’s law school. Where law is less centralto politics, however, the dominance of legaltraining has declined as other degrees, such as eco-nomics, have achieved recognition. For instancesince 1945, the percentage of law graduatesamong civil servants working in Norway’s centralministries has fallen from a half to a fifth(Christensen and Egeberg, 1997).

DefinitionThe theory of representative bureaucracyclaims that a civil service recruited from all sec-tions of society will produce policies that areresponsive to the public and, in that sense,democratic (Meier, 1993, p. 1). Passive represen-tation exists when the demographic profile ofthe bureaucracy matches that of the population.Active representation occurs when civil ser-vants take the same decisions as would be madeby the represented – that is, the public.

In a departmental (specialist) system, recruitersfollow a different philosophy from Britain’s gener-alist approach or the French corps. They look forspecialist experts for individual departments, withmore movement in and out of the civil service at avariety of levels. The Finance Ministry will recruiteconomists and the Department of Health willemploy staff with medical training. Recruitment isto particular posts, not to an elite civil service or acorps. This emphasis on a specific job is commonin countries with a weak state in which the admin-

istration lacks the status produced by centuries ofservice to pre-democratic rulers. The UnitedStates, New Zealand and the Netherlands areexamples. In the Netherlands, each departmentsets its own recruitment standards, normallyrequiring training or expertise in its own area.Once appointed, mobility within the civil serviceis limited; staff who remain in public serviceusually stay in the same department for theirentire career (Andeweg and Irwin, 2002, p. 176).The notion of recruiting talented young graduatesto an elite, unified civil service is weak or non-existent.

Organization

Here we examine the detailed organization ofcentral government activity, looking at how thecogs of the administrative machine are arranged.Although structures – and labels – vary bycountry, the key distinction is between three kindsof organization (Box 16.1). The first and mostimportant unit is the department (sometimescalled ministry) where policy is made. In nearly allcountries, a dozen or so established departmentsform the stable core of central government, cov-ering such areas as finance, defence and foreignaffairs. A second unit comprises the divisions, sec-tions or bureaus into which departments aredivided. These are the operating units of govern-ment – the powerhouses where expertise residesand in which detailed policy is both formed andapplied (to complicate matters, these divisions aresometimes called ‘departments’ in countries wherethe larger unit is termed a ‘ministry’). Forexample, departments of transport have sectionscovering road, rail and air travel. And the thirdunit is the non-departmental body operating atone or more removes from ministries. These semi-detached organizations combine public fundingwith operational autonomy; they are growing innumber and significance.

Departments and divisions

Government departments (or ministries) form thecentrepiece of modern bureaucracies. Here we findthe bodies pursuing the traditional tasks of gov-

THE BUREAUCRACY 293

ernment: for example, finance, defence andforeign affairs. The United States has 15 depart-ments, each headed by a Secretary of Stateappointed by the president. The Netherlands has13; Canada, always prone to political inflation,has over twenty. New Zealand leads the field with

around forty departments, though a single min-ister may take charge of several departments.

Most countries follow a similar sequence inintroducing departments. The first to be estab-lished are those performing the core functions ofthe state: finance, law and order, defence andforeign affairs. These ministries are often asancient as the state itself. In the United States, forinstance (see Figure 16.2), the Department ofState and the Treasury each date from 1789. At alater stage countries add extra ministries to dealwith new functions. Initially these are usually agri-culture (1862 in the USA) and commerce andlabour (1903 and 1913), followed later in thetwentieth century by welfare departments dealingwith social security, education, health andhousing.

Reflecting Weber’s principles, departments areusually organized in a clear hierarchy. A single min-ister sits at the pinnacle of the organization albeitoften supported in large departments by juniorministers with responsibilities for specific divisions.A senior civil servant, often called the Secretary, isresponsible for administration and for forming thecrucial bridge between political and bureaucraticlevels (in Japan this person is called the vice-min-ister, reflecting the high status of top bureaucratsthere). Table 16.1 shows the elaborate structure ofthe German Economics Ministry, a fine illustrationof Weber’s quasi-military chain of command.

Departments are typically arranged into severaldivisions, each responsible for an aspect of

294 GOVERNMENT AND POLICY

Government department or ministryAn administrative unit over which a minister exer-cises direct management control. Usually structuredas a formal hierarchy and often established bystatute.

Division, section or bureauAn operating unit of a department, responsible tothe minister but often with considerable indepen-dence in practice (especially in the USA, where manysuch divisions are called ‘agencies’).

Non-departmental public bodyOperates at one or more removes from the govern-ment, in an attempt to provide management flexi-bility and political independence. Sometimes called‘quangos’, a term of American origin that originallymeant ‘a quasi non-governmental organization’ butwhich is often now taken to mean a ‘quasi-govern-mental organization’.

B OX 1 6 . 1

The organization of government:departments, divisions and agencies

Figure 16.2 Founding of current cabinet level departments in the USA

1750 1800 1850 1900 1950 2000

Trea

sury

(178

9)St

ate

(178

9)

Inte

rio

r (18

49)

Ag

ricu

ltu

re (1

862)

Just

ice

(187

0)

Co

mm

erce

(190

3)

Lab

or (

1913

)

Def

ense

(194

7)H

ealt

h a

nd

Hu

man

Se

rvic

es (1

953)

Ho

usi

ng

an

d U

rban

D

evel

op

men

t (1

965)

Tran

spo

rtat

ion

(196

6)

Ener

gy

(197

7)Ed

uca

tio

n (1

979)

Vete

ran

s A

ffai

rs (1

988)

Ho

mel

and

Sec

uri

ty (2

002)

THE BUREAUCRACY 295

The social background of senior civil servants is invari-ably unrepresentative of the general population. In theWestern world, the typical high-level civil servant is amale graduate, from an urban background and from amiddle- or upper-class family that was itself active inpublic affairs (Aberbach et al., 1981, p. 80).These find-ings have disturbed advocates of a ‘representativebureaucracy’.This term was introduced by Kingsley(1944), an American scholar who claimed that themiddle-class composition of Britain’s civil serviceexploded the myth of its supposedly neutral bureau-cracy. By implication, the same argument could beapplied to other countries. But should we be concernedabout the social skew of public servants? And, if so,should recruitment on merit be modified by positivediscrimination in favour of under-represented groupssuch as ethnic minorities and women?

The case forThree arguments support the thesis that a bureaucracyshould reflect the social profile of the population:

� Civil servants whose work involves direct contactwith specific groups will be better at the job if theyalso belong to that category. A shared language isthe most obvious example but the point can perhapsbe extended to ethnicity and gender.

� A civil service balanced between particular groups,such as religions or regions, may encourage stabilityin divided societies such as Northern Ireland.

� Democracy is said to involve government by, and notjust for, the people. A representative civil service,involving participation by all major groups in society,will enhance the acceptability of decisions. Positivediscrimination in favour of under-represented groupsis the only effective solution to the problem. But itwill only be needed for a short period because, oncea representative bureaucracy is established, it willmaintain itself naturally as minorities notice the rolemodels in post and become more likely to apply forjobs themselves.

The case againstThe principle of recruitment on merit is fundamental topublic administration and should not be abandoned infavour of social engineering.The public interest is bestserved by selecting the best people for the job, irrespec-

tive of their background.The correct solution to under-representation is not positive discrimination butimproving the qualifications of the excluded minorities.In any case, positive discrimination creates two prob-lems of its own. First, those denied jobs because theybelong to the ‘wrong’ social category are naturallyresentful. Second, those who are accepted just becausethey come from the ‘right’ social background are placedin an awkward position. In any case, it is far from clearthat the narrow social background of senior civil ser-vants does produce the prejudice against the left thatKingsley claimed to detect. Norwegian studies showthat the attitudes of civil servants depend more on theirpost, level and department than on their social or edu-cational background:‘where you stand depends onwhere you sit’ (Laegreid and Olsen, 1978). And wherecivil servants are allowed to pursue political careers,they frequently join parties of the left. In France, a thirdof the senior public servants elected to the FrenchNational Assembly in 1978 were members of theSocialist Party. Generally, the dominant ideology amongtop bureaucrats seems to be centrist, not conservative(Aberbach et al., 1981). Broadening the social base ofrecruits would not, in practice, change bureaucraticdecisions.

AssessmentThe arguments for a representative public service didlead to affirmative action programmes in some coun-tries, notably the United States, in the 1970s and 1980s.Considerable efforts were made to ensure that the staffprofile matched that of the wider population. Canadiangovernments, concerned since the 1960s to improverecruitment from French-speakers, also extended theirrecruitment efforts. However, such schemes neverachieved the same popularity in Europe, perhapsbecause they would have involved accepting the inad-equacy of the constitutional requirement of neutralityimposed on some civil services there. Even in NorthAmerica, attitudes were ambivalent, reflecting a tensionbetween social engineering and merit-based recruit-ment. In any event, affirmative action schemes lostmomentum in the more conservative 1990s.

Further reading: Birch (1972), Kingsley (1944), Meier (1993),van der Meer and Roborgh (1996).

DEBATE

A REPRESENTATIVE BUREAUCRACY?

the organization’s work. Thus an EducationDepartment might have separate divisions forprimary, secondary and higher education – and, inpractice, for many other aspects of its work.Divisions are the operating units of departments,the sections within which the work gets done.They are the workhorses of government, the storeof its experience and, in practice, the site wheremany important decisions are reached. In short,divisions within ministries are the engine-rooms ofgovernment.

In many democracies, such divisions or bureausacquire added importance because they are par-tially autonomous from their parent department.The extreme example is the USA, whose bureau-cracy is the great exception to the principle ofhierarchy in departments. Even in their formalstructure, American departments are more likemultinational corporations, containing many divi-sions jostling within a single shell. The depart-ments are merely the wrapping round a collection

of disparate divisions and it is these bureaus whichform the main operating units of the federal gov-ernment. For example, within the Department ofHealth and Human Services it is divisions such asthe Centers for Medicare and Medicaid Serviceswhich administer specific federal welfare schemes.Reporting formally to the president, bureau chiefsspend much of their time ensuring their opera-tional independence from the White House.Congress, not the president, creates and fundsbureaus. And what Congress gives it can (andoccasionally does) take away. The autonomy ofbureaus within American departments is a majorand often underestimated reason why Americanpresidents experience such difficulty in imposingtheir will on Washington’s complex politicalprocess.

The contrast between America’s fragmentedbureaucracy and the German system of hierar-chical departments could hardly be more extreme.Even in Germany, however, it would be wrong tosuppose that working practices correspond exactlyto the organization chart implied in Table 16.1,with its Weberian image of information movingsmoothly up and down the administrativepyramid. In practice, the 2,000 divisions (‘sec-tions’) of the German federal ministries possess aconcentration of expertise that enables them toblock or at least circumvent changes proposedfrom on high. In most other democracies, too,divisions within departments possess their ownethos derived from long experience with their area.This ‘house view’ can breed a natural cynicismtowards the latest political initiative and the topminister may need to circumvent divisional resis-tance by seeking advice from political advisers.

Non-departmental public bodies

The defining feature of non-departmental publicbodies is that in theory they operate at one removefrom government departments, with a formal rela-tionship of at least semi-independence. Such insti-tutions occupy an ambivalent position, createdand funded by the government, but in contrast todivisions within a department they are free fromday-to-day ministerial control. Once appointed bythe government, the members of such bodies areexpected to operate with considerable autonomy.

296 GOVERNMENT AND POLICY

Table 16.1 The structure of Germany’s Ministryof Economics and Labour

Position Number of Service levelpositions

The minister 1 –Parliamentarystate secretaries 3 –Administrative state secretaries 3 Political officialsDepartment heads 11 Political officialsSubdepartment heads 34 Higher serviceSection heads 150 Higher serviceSection assistants 460 Higher serviceCaseworkers 615 Elevated serviceClerical/secretarial 822 Intermediate

serviceMessengers etc. 288 Lower-level

service

Note: In Germany,‘political officials’ are tenured civil servants whomust be transferred to another job of suitable status if not retainedby their minister.

Source: Conradt (2001),Table 8.1, updated from the department’swebsite at http://www.bmwi.de.

In the United States, where such agencies areprominent, leading examples include the CentralIntelligence Agency (CIA), the EnvironmentalProtection Agency (EPA) and the NationalAeronautics and Space Administration (NASA).

Scandinavian countries, notably Sweden, havealso long relied on non-departmental bodies toimplement policy set by the ministry. In Sweden,government departments are small and ministersexercise no authority over the implementation ofpolicy. Rather, a network of independent and rela-tively autonomous central boards and agencies –the Social Welfare Board, the National LabourMarket Board, the National Pensions Board, etc. –executes policy directives (Arter, 1999, p. 154).

Non-departmental public bodies are establishedfor a range of reasons. These include: to provideprotection from political influence, to operatewith more flexibility than would be acceptable fora division of a ministry, to acknowledge the pro-fessional status of staff employed in them orsimply as a response to short-term pressure to ‘dosomething’ about a problem. Throughout thedemocratic world, non-departmental bodies aregrowing in number, complicating not just the aca-demic task of mapping government but also thepractical job of ensuring that the government as a

whole acts in a coherent manner. Indeed, moderngovernance cannot be understood without delvingdeeper into the undergrowth of these organiza-tions.

The classification of non-departmental bodies inBox 16.2 distinguishes between state-owned enter-prises, other agencies operating separately fromdepartments and non-statutory bodies created bydepartments for specific, often short-term,reasons. In many established democracies in theEnglish-speaking world, the balance between theseformats changed considerably in and after the1980s. Specifically, privatization finished off manystate-owned enterprises while demand for agenciesto provide services (including regulation)increased. As a result, the shape of the publicsector underwent considerable alteration.

A state-owned enterprise (SOE) is – or perhapswe should say ‘was’ – a government corporationestablished by law to sell goods and services, withno private shareholders. Examples includeCanada’s Crown Corporations, the US PostalService and Britain’s nationalized industries.Although such bodies were supposed to operate atarm’s length from government, political interfer-ence often led to overstaffing and underinvest-ment. In Britain, most government corporations

THE BUREAUCRACY 297

Type Definition Examples

State-owned enterprise A government corporation established Most countries: mail service (SOE) by statute to trade goods or services New Zealand: Coal Corporation of NZ

USA: Amtrak (a passenger rail network)

Other statutory Statutory bodies operating separately from Sweden: Social Welfare Boardagencies (and sometimes independently of ) USA: Federal Election Commission

government departments. Most agencies UK: Food Standards Agencyeither supply a public service directly orregulate an aspect of society

Non-statutory Bodies set up by departments to offer Such bodies are often temporary (e.g.organization specialized services or advice advisory committees on mad cow

disease). In many countries, stable examples include scientific research councils and bodies for distributing public funds to the arts

B OX 1 6 . 2

Types of non-departmental public body

were sold back to the private sector duringConservative rule between 1979 and 1997, apolicy that has now been copied by other coun-tries. Indeed, the SOEs that remain in publichands in Britain, such as the Post Office, facegrowing competition in international marketsfrom private companies advantaged by theirfreedom to raise money from the market forinvestment.

Yet far from the privatization earthquakedestroying the role of the state in the economy, ithas simply reshaped the landscape of regulation.Ownership and control have given way to supervi-sion. Today, the most important non-depart-mental bodies are not government corporationsbut agencies. These are public organizations oper-ating separately from departments and charged(normally by statute) either with delivering publicservices or with regulating an aspect of social lifewhere a public interest is held to be at stake.Services provided by agencies include intelligence-gathering, management of public buildings andthe provision of driving licences. According to theschool of new public management (p. 301), spe-cialist agencies, operating in a businesslike mannerfree from detailed political interference, should bemore efficient at supplying services than a divisionof a government department.

The other form of agency is the regulatory body.Examples include bodies to regulate naturalmonopolies (e.g. water supply), adoption, broad-casting, elections, food standards and nuclearenergy. Regulatory agencies are increasing innearly all established democracies, partly tobalance risks which cannot be well-judged by theprivate sector. For example, weighing the benefitsof introducing a new drug against the danger ofside effects is perhaps a task for public-mindedexperts rather than for self-interested drug compa-nies or even busy politicians (Boyne, 2003).Britain now has well over 100 regulatory bodies,employing over 20,000 people and costing around£1 billion ($1.5 billion) per year. The media entre-preneur Rupert Murdoch even claims that regula-tion represents socialism’s comeback trail; his viewis that ‘socialism is alive and well, and living inregulatory agencies’.

The hope, at least partly reflected in reality, isthat regulatory agencies will act as a buffer

between the government and the regulatedbodies, reducing the excessive intervention whichheld back many government corporations.However, the danger is that such agencies willcome to serve the interests of those they super-vise. Regulatory capture has been widelyobserved, not least with state utility commissionsin the United States (Wilson, 1989). Even if reg-ulators avoid capture, they will still want to per-petuate their own existence, a bias which leadsthem to over-value the need for public regula-tion.

DefinitionRegulatory capture arises when public agenciescreated to regulate a particular industry come toserve the interests of those they supervise.

Reflecting their importance, government corpo-rations and agencies are usually established by aspecific statute (i.e. law). However, ministers alsoset up (and occasionally dissolve) non-statutoryorganizations to offer advice or provide an execu-tive function in specialized areas of activity.Whether a particular function is handled by astatutory or non-statutory body varies by countrybut non-statutory examples might include scien-tific advisory panels, research funding committees,arts councils and training boards. Although muchof their work is routine, non-statutory bodies stillattract recurring if sometimes unjust criticism. Incontrast to bodies established under law, theyreport to the sponsoring minister, not to theassembly. Membership tends to be seen as politicalpatronage and accountability is regarded asopaque and intermittent. Criticism is especiallysharp in countries such as Britain where the mythof parliamentary sovereignty still conditions polit-ical discourse. But even in Britain many non-statutory bodies survived a crusading culllaunched by Mrs Thatcher in the 1980s. Even ifmany non-statutory bodies do outlast their useful-ness, the device seems to be indispensable in pro-viding ministers with a flow of information andadvice on specialized topics. Charting the non-departmental public bodies in any establisheddemocracy confirms the complexity of contempo-rary governance and gives the lie to any simpleclaims of a diminished role for the state.

298 GOVERNMENT AND POLICY

Accountability

How should civil servants be rendered accountablefor their actions? Given its detailed knowledge, itspermanence, its scale and its control of policyimplementation, the bureaucracy is bound to bemore than a mechanical conduit for politicaldirectives. Senior public employees – departmentsecretaries, heads of divisions, chairs of non-departmental public bodies – are invariably in aposition to influence policy. The ‘problem’ of con-trolling bureaucratic ‘power’ in a democracy was aparticular concern of Weber’s (Gerth and Mills,1948, pp. 232–5). He identified the danger ofpublic ‘servants’ coming to dominate their elected‘masters’. Indeed, Weber argued that

under normal conditions, the power position ofa fully developed bureaucracy is always over-whelming. The ‘political master’ finds himself inthe position of the ‘dilettante’ who stands oppo-site the ‘expert,’ facing the trained official whostands within the management of administra-tion.

As Weber himself realized, his model of civil servants reporting to a single minister in a self-contained department provided inadequateaccountability in practice. Today, more emphasis isplaced on a looser but also more flexible view ofaccountability. High-level civil servants areincreasingly subject to ‘multiple accountabilities’ –for example, to the prime minister and the financeministry and to agreements with other organiza-tions (international and domestic) as well as to the ministers in their own department. Polidano(1998, p. 35) argues that in a complex environ-ment, where most policy-making involves coordi-nation between several organizations, ‘bureaucratscan be prevented from complying with ministerialdirections, however legitimate those directionsmay be. Multiple accountabilities are aninescapable part of the reality of government’. For such reasons, many analysts now accept that senior administrators should learn and indeed be taught ‘political craft’ – the ability to seepolicy options in a wider political context (Goetz,1997).

So the issue today is not so much preventing

bureaucrats from influencing policy as ensuringsome measure of accountability for the decisionswhich public servants help to shape. Box 16.3 dis-tinguishes two main forms of accountability:internal controls within the civil service (includingdirection by the minister) and the increasing rangeof accountabilities to external bodies such as thelegislature and the judiciary. We discuss each typeseparately.

Internal controls

The traditional form of bureaucratic account-ability is, of course, to the minister in charge ofthe department. Ministers direct; public servantsexecute. Although no longer sufficient, hierar-chical control by a minister remains an essentialpart of bureaucratic accountability. However, inpractice, the capacity of ministers to exert suchcontrol is conditioned by two other factors: thereach of political appointments into the bureau-cracy and the use ministers make of personaladvisers.

In theory, the greater the number of politicalappointments to a department, the easier it is toensure political accountability. Recognizing thatsenior bureaucrats should possess political craft,many established democracies now tend to staffimportant ministries with politically loyal andsympathetic civil servants. This practice, long

THE BUREAUCRACY 299

Internal controls

Ministerial directionFormal regulationCompetition between departments Professional standards

External scrutiny by

Legislature and judiciary OmbudsmenInterest groups and the mass media

Sources: Polidano (1998), Stone (1995).

B OX 1 6 . 3

Forms of bureaucratic accountability

familiar in Germany and Finland, is spreading toother Western democracies. Increasingly, politi-cians want civil servants who are, in MrsThatcher’s famous phrase, ‘one of us’. As Page andWright (1999, p. 278) write,

there is a common trend among administrativesystems which stress the norm of civil serviceneutrality to appoint, either as civil servants oradvisers, people in whom one has trust.Increasing political influence in senior appoint-ments suggests the possibility that membershipof a ‘neutral’ civil service is decreasing as a guideto trust among political elites.

Yet it is far from clear that in practice ministerialcontrol does increase with the number of politicalappointments. In the United States, an incomingpresident appoints around 3,000 people, a taskwhich itself becomes an administrative distractionfor a new president. Ministers who want to getthings done might be well-advised to trust theircivil servants rather than to seek ever-closer polit-ical control.

Political accountability of the bureaucracy canalso be aided by providing ministers with personaladvisory staff. Because such gurus are not part ofthe department’s permanent staff, they can act astheir minister’s eyes and ears, reporting back onissues which might otherwise be lost in the officialhierarchy. The Executive Office of the Presidentand the White House Office of the Americanpresidency are the fullest expression of thisapproach. These offices form a counter-bureau-cracy within the political system and one that ismore likely to be ideologically or politically driventhan the formal bureaucracy. Again, however, thissword has two edges. Political advisers may help toassert political control but they create their ownproblems of accountability. Advisers are neitherelected nor vetted by Congress. The danger is thatthey are too dependent on their patron, preferringto offer blandishments and flattery rather thanhome truths.

A preferable system, perhaps, is the Frenchcabinet (not to be confused with the cabinetswhich form the apex of the government in parlia-mentary systems). A French cabinet is a group ofabout 15 to 20 people who form the minister’s

personal advisory staff and work directly under hisor her control. Cabinets provide the minister withideas and help in liaising with the department,other ministries, the party and the constituency.However, most cabinet members come from thecivil service and return to it after a few years withtheir minister. Thus the French system offers amethod of securing political advisers who are morethan mere yes-men.

In addition to formal control by their minister,public officials are subject to increasing scrutinyby an army of regulators within government: the‘waste watchers, quality police and sleaze busters’as Hood et al. (1999) term them. Auditors inspectthe books, standard-setters check performance,funders assess the outcomes achieved and assortedinspectors monitor everything from employmenttrends to recycling rates. Contemporary ideologymay preach a bonfire of regulations but modernpractice is an overdose of inspection, sometimesbreeding cynicism among officials and occasion-ally distracting from the main business at hand. Toillustrate the growth of oversight, Hood et al.(1999, p. 42) project ten regulating bodies foreach UK government department by the end ofthe twenty-first century should the rate of regula-tory expansion observed in the final quarter of thetwentieth century continue unchecked.

A more informal, but perhaps more effective,internal device for regulating the bureaucracy iscompetition between departments. Spendingministries must compete against each other formoney, with the finance ministry acting asumpire and cashier. In addition, establisheddemocracies rely on the professional standards ofcivil servants, particularly in bureaucracies with astrong technical emphasis. In Norway andSweden, for instance, many civil servants work inspecialist directorates covering areas such as engi-neering, medicine and railways. These directoratesgive expression to professional expertise and are atleast partially independent of ministries. In thesesmall Scandinavian countries, trust betweenpoliticians and bureaucrats remains high and thesystem usually functions smoothly. In the main,senior civil servants respond to political signalswithout explicit direction – a nice result but noteasily replicated elsewhere (Christensen andPeters, 1999).

300 GOVERNMENT AND POLICY

External scrutiny

External scrutiny is an expanding form ofaccountability for public servants. Traditionally,bureaucrats could easily escape both political andpublic scrutiny when, as in Britain, ministersalone were considered responsible to parliamentfor the actions of their officials. Civil servantscould and did hide under their minister’s skirts.Fortunately, perhaps, the British stress on theanonymity of higher civil servants was nevermatched in other liberal democracies. In theUnited States, bureaucrats are more forthcomingin their congressional appearances. But even insecretive Britain the rise of legislative committeesof scrutiny has added a new dimension to bureau-cratic accountability. Slowly and shyly, and notbefore time, public officials are becoming willingto account in public for their work.

DefinitionAn ombudsman is a public official appointed bythe legislature to investigate allegations of mal-administration in the public sector.These watch-dogs originated in Scandinavia but they havebeen emulated elsewhere though often withmore restricted jurisdiction and resources.

As with other areas of politics, the judiciary isalso growing in importance as an arena in whichthe bureaucracy can be called to account. Judgesare increasingly willing to use administrative lawas a device for influencing bureaucratic procedures(p. 221). A more recent addition to the mecha-nisms of external scrutiny is the ombudsman. Thiswatchdog was first introduced in Sweden and thenemulated in New Zealand, followed later by otherEuropean countries. Assigned to investigate com-plaints of maladministration, ombudsmen musthave strong powers of investigation if they are tosucceed. So far, governments outside Scandinaviahave proved reluctant to grant this facility. Finally,interest groups and the mass media also provideincreasing external checks on bureaucraticbungling. A vigorous mass media can also act as aselective check on the bureaucracy: regular televi-sion programmes, for example, now specialize inexposing public scandal and bureaucratic inepti-tude. However, oversight tends to be selective and,

in the case of the media, short-term. Exposésrarely lead to structural reform; the specific casemay be resolved but the complacency of thebureaucracy often continues.

New public management

‘Government is not the solution to the problem;government is the problem.’ This famous declara-tion by Ronald Reagan is one inspiration behindthe new public management (NPM), a creedwhich swept through the Anglo-American worldof public administration in the final decades of thetwentieth century. NPM represents a powerful cri-tique of Weber’s ideas about bureaucracy. It hasattracted many specialists who do not share theideological perspective of Ronald Reagan, it isspoken of warmly by international bodies such asthe Oganization for Economic Cooperation andDevelopment and it has led to radical change inthe public sectors of Australia, Canada, the UnitedKingdom and especially New Zealand.

The best way to approach NPM is to considerOsborne and Gaebler’s Reinventing Government(1992), an exuberant statement of the newapproach. Subtitled ‘How the EntrepreneurialSpirit is Transforming the Public Sector’, thisAmerican best-seller outlined ten principles whichgovernment agencies should adopt to enhancetheir effectiveness (Box 16.4). Where Weber’smodel of bureaucracy was based on ideas of effi-ciency drawn from the Prussian army, Osborneand Gaebler are inspired by the freewheelingworld of American business.

The authors cite with enthusiasm several exam-ples of public sector organizations which have fol-lowed their tips. One is the California parksdepartment that allowed managers to spend theirbudget on whatever they needed, without seekingapproval for individual items of expenditure.Another is the public convention centre whichformed a joint venture with private firms to bringin well-known entertainment acts, with each sidesharing both the risk and the profit. The under-lying theme in such anecdotes is the gains achiev-able by giving public servants the flexibility tomanage by results (that is, ‘managerialism’). Andthe significance of this, in turn, is the break it rep-

THE BUREAUCRACY 301

resents with Weber’s view that the job of a bureau-crat is simply to apply fixed rules to cases. For itssupporters, NPM is public administration for thetwenty-first century; Weber’s model is dismissed ashistory. Public administration has been displacedby public management.

While Osborne and Gaebler provide a convert’shandbook, Hood (1996, p. 271) offers a more dis-passionate and comparative perspective (Box16.5). Hood shows that NPM has penetrated fur-thest in Anglo-American countries andScandinavia, where the public sector is mostamenable to political control. By contrast, coun-tries with a strong state and high-prestige bureau-cracy, for example Germany, Japan and Spain,have made little progress in implementing the newphilosophy. In these countries, senior civil servantscontinue to guard the public interest; their task isto apply codified law to specific cases. Given suchtraditions, the managerial ethos of economy, effi-ciency and effectiveness will not easily prosper. Aparticular problem is that the status and duties of

civil servants are entrenched in extensive legalcodes, making radical change impossible withoutlegislation.

Within the Anglo-American countries, NewZealand proved to be a testing-ground for NPM.In the 1980s and 1990s, successive governments –first Labour and then National – revolutionizedthe structure, management and role of the publicsector. A remarkable coalition (perhaps even con-spiracy) of economic theorists in the Treasury,senior politicians from both major parties andbusiness leaders came together to ram throughunpopular but far from ineffective reforms. Oneparticular feature of the ‘New Zealand model’ isits massive use of contracts (Boston et al., 1995).This goes far beyond the standard fare of usingprivate firms to supply local services such asgarbage collection. It extends to engaging privatesuppliers in sensitive areas such as debt collection.By such means, the Department of Transportreduced its direct employees from around 5,000staff in 1986 to fewer than fifty in the mid-1990s,an astonishing transformation. In addition, con-tracts are widely used within New Zealand’s publicsector to govern the relationships between pur-chasers (for example the Transport Department)and providers (for example Transit New Zealand,responsible for roading, and the Civil AviationAuthority, charged with air safety and security).‘Contractualism’ within the public sector is an

302 GOVERNMENT AND POLICY

� Promote competition between service providers.� Empower citizens by pushing control out of the

bureaucracy into the community.� Measure performance, focusing not on inputs but

on outcomes.� Be driven by goals, not rules and regulations.� Redefine clients as customers and offer them

choices – between schools, between training pro-grammes, between housing options.

� Prevent problems before they emerge, rather thanoffering services afterwards.

� Earn money rather than simply spend it.� Decentralize authority and embrace participatory

management.� Prefer market mechanisms to bureaucratic ones.� Catalyse all sectors – public, private and voluntary

– into solving community problems.

Source: Adapted from Osborne and Gaebler (1992).

B OX 1 6 . 4

Steer, don’t row! Osborne andGaebler’s ten principles forimproving the effectiveness ofgovernment agencies

� Managers are given more discretion but are heldresponsible for results.

� Explicit targets are set and used to assess results.� Resources are allocated according to results.� Departments are ‘unbundled’ into more indepen-

dent operating units.� More work is contracted out to the private sector.� More flexibility is allowed in recruiting and

retaining staff.� Costs are cut in an effort to achieve more with less.

Source: Adapted from Hood (1996).

B OX 1 6 . 5

Components of new public management

additional step, and a more direct challenge toWeber’s model, than simply contracting out ser-vices to the private sector.

What lessons can be learned from NewZealand’s ambitious innovations in public admin-istration? Mulgan (1997, p. 146) offers a balancedassessment. He concludes that

the recent reorganization of the public servicehas led to greater clarity of government func-tions and to increased efficiencies in the provi-sion of certain services to the public. At thesame time, it has been expensive in the amountof resources consumed by the reform processitself and also in the added problems of coordi-nation caused by the greatly increased numberof individual public agencies.

The rise of NPM and the contract culture is onereason why the accountability of public officialshas become more complex. When something goeswrong with a service provided by an agency oper-ating under contract to government, who shouldtake the blame: the supplier or the department? InBritain, parliament has traditionally held ministersto account for all the actions carried out in theirname. As The Times wrote in 1977, ‘the constitu-tional position is both crystal clear and entirelysufficient. Officials propose. Ministers dispose.Officials execute.’ Yet by 1994 most British civilservants were working in one of about 100 semi-independent agencies (O’Toole and Chapman,1995). In theory, the minister sets the policy andthe agency carries it out. But when a politicalstorm blows up – when convicts escape fromprison or the Child Support Agency pursuesabsent fathers too zealously – it is still ministerswho are hauled before parliament. Knowing this,ministers are inclined to interfere with operationalmatters, thus contradicting the original purpose ofthe reform. Agency managers discover that theyare not free to manage after all, with damage tomorale.

The complexities of accountability in a reformedcivil service lead some critics to suggest that ‘ahuge hole now exists in the operation of Britishdemocracy’ (Campbell and Wilson, 1995, p. 287).Public servants are becoming more responsivedownwards, to their users, and also more open to

scrutiny from alternative political authorities, suchas parliamentary committees. Probably thesedevelopments represent a change in accountabilityrather than a decline. Control is melting awayfrom the minister’s office to a diffuse set of agen-cies and their clients. Weber’s hierarchy of controlbased on direct provision by departments is givingway to a looser network based on persuasionrather than order-giving. Governance is replacinggovernment. For previously centralized countries,notably Britain and New Zealand, the politicalimplications of NPM are profound. Members ofparliament wedded to the idea of ministerialaccountability to the assembly are not pleased todiscover that their cherished myth of sovereigntyhas shrivelled under pressure from the complexrealities of modern policy-making.

Bureaucracy in new democracies

A common legacy of an authoritarian regime is anover-powerful, unaccountable and corrupt bureau-cracy. Overcoming this difficult inheritance byestablishing the supremacy of elected over bureau-cratic authority is an important component ofdemocratic consolidation. In particular, the task isto move the bureaucracy away from the highlypolitical mode of operation under the old ordertowards a more professional Weberian model,where appointments are based on merit and cor-ruption is contained. Only when these challenginglong-term goals have been achieved does it makesense to contemplate applying the fads and fash-ions of new public management to the publicsector. A bureaucracy based on ‘old-fashioned’Weberian ideas remains a sensible and evendemanding aspiration for a new democracy evenas it comes to be seen as inadequate for a consoli-dated democracy (Verheijen, 1998).

Spain’s transition to democracy illustrates thedifficulties of bringing a post-authoritarianbureaucracy to heel. Under the old regime domi-nated by General Franco, ‘the bureaucracy hadbeen a central linchpin of the highly centralized,backward-looking dictatorship and was accord-ingly well-represented within the power elite’(Heywood, 1995, p. 130). Especially in the dicta-torship’s early years, the dominant test of appoint-

THE BUREAUCRACY 303

ment as a public official was loyalty to the regime.In operation the bureaucracy remained overstaffedand corrupt – albeit with some modernization asFranco’s rule began to soften before the general’sdeath in 1975. Even after the old order collapsed,administrative restructuring remained a low pri-ority amid the drama of democratic transition. Itwas not until the crushing electoral victory of theSocialists in 1982 that major reforms of a nowdemoralized service were attempted. The Socialistsinitiated such obvious changes as preventing thesame person from occupying more than onepublic post; they sought also to reduce the signifi-cance of cuerpos (corps of officials) by focusingrecruitment on specific jobs in particular min-istries. But these changes met with only partialsuccess. They were followed by further reformslater in the 1980s aimed at inducing a more man-agerial and less legalistic outlook among publicservants. Yet again the outcome was mixed. As lateas 1992, Socialist Prime Minister Felipe Gonzalezcould describe the continuing inertia of the publicservice as the greatest frustration of his premier-ship. Continuing public distrust of state offices isone of the few weaknesses in Spain’s generally suc-cessful transition to democracy. The Spanish expe-rience reveals in particular that adapting anauthoritarian bureaucracy to a democratic envi-ronment is a long-term task.

The difficulties of establishing an efficient publicadministration are even greater in post-communistcountries. After all, the collapse of communism wasnot just the end of a dictatorship; it was also the dis-integration of an all-pervasive, over-extended andover-politicized bureaucracy. Initially, regime changeled to administrative chaos. For example, many civilservants only received their salary on an irregularbasis, forcing them to find ways of supplementingtheir income. A common response was to exploitthe massive bank of formal regulations inheritedfrom the communist era. An official stamp, forexample to authorize a new business, remained avaluable commodity. One result of post-commu-nism, suggests Crawford (1996, p. 105), was simplythat the price of bribes went up. And a distinctivefeature of post-communist societies was, of course,the vast inheritance of public ownership.Bureaucrats found that they could manipulate pri-vatization or exploit the monopoly position of

newly privatized companies for their own benefit.So post-communist bureaucracies remained in a pre-Weberian stage – they did not operate by the consis-tent application of formal regulations. As with theroyal households of premodern Europe, a public jobwas valued as an opportunity to tax as much as asource of a regular salary. Ironically, public officialsfound themselves perfectly positioned to exploit thetransition to a ‘market economy’.

DefinitionCivil Service Acts specify procedures forrecruiting, training, promoting, paying and dis-missing bureaucrats. In post-communist states,such laws affirm bureaucratic professionalismagainst a backdrop of communist party control.

Some post-communist states, particularly theCzech Republic, Hungary, Poland and Slovenia,have now made considerable progress in over-coming these problems. In 1992, for instance,Hungary became the first country in Central andEastern Europe to introduce a Civil Service Act(Baka, 1998). Such legislation is a sensitive issuethroughout the post-communist world because itbreaks with the communist practice of politicalcontrol of appointments. Civil service laws arepart of an attempt to strengthen the distinctionbetween political and administrative roles, pushingthe bureaucracy in a Weberian direction. However,in the least democratic post-communist countries,such battles have often been lost. In Bulgaria, forinstance, ‘the civil service has traditionally beenconsidered to be a tool of a particular politicalparty rather than a servant of society as a whole.The civil service is perceived not only as a tool togovern but also as a weapon against the politicalopponent’ (Nikolova, 1998). Only a dreamer, sug-gests Stainov (1993), could imagine ‘a picturewhere, when a government [in Bulgaria] ischanged, the civil servant remains firmly in hisplace without caring too much about the politicalsubject in power’.

Bureaucracy in authoritarian states

The bureaucracy, like the military, is usually amore powerful force in non-democratic regimes

304 GOVERNMENT AND POLICY

THE BUREAUCRACY 305

Contained in a series of islands the size of Montana,and lacking all major natural resources, Japan’s 127mpeople have built the second largest economy in theworld. From the ashes of defeat in the Second WorldWar, Japan by the early 1990s had become the world’slargest creditor nation and donor of economic aid.How was this transformation achieved?

Specific historical factors were part of the answer.After the war, American aid, an undervalued yen,cheap oil and the procurement boom caused by theKorean War all contributed to economic recovery.YetJapan took remarkable advantage of these favourablecircumstances, focusing its initial efforts on heavyindustry but eventually becoming the world’s leadingproducer and exporter of sophisticated industrial andconsumer goods.

‘Japan, Inc.’ was a popular (and populist) explanationof the country’s success. According to this interpreta-tion, the ethnically homogenous Japanese weredriven not just by memories of wartime hunger butalso by a shared desire to catch up with the West. Amore sophisticated explanation was institutional.Although the country’s postwar constitution was anAmerican-imposed liberal democracy, in practice thepolitical system was dominated by business.Thisenabled long-term investment to proceed byrepressing any popular demands for rapid increases indomestic consumption, creating the Japaneseparadox of ‘a rich country with poor people’.

The Japanese civil service is accorded high status,attracts able recruits through open competition and

motivates them with the thought of plum post-retire-ment jobs in the private sector. As Johnson (1987, p.68) writes, senior bureaucrats form part of ‘the eco-nomic general staff, which is itself legitimated by itsmeritocratic character’.The bureaucracy certainlyplayed a major role in postwar reconstruction. It wasintertwined with the dominant Liberal DemocraticParty (conservative in all but name) and big business.These groups formed a ruling elite though one that isnow declining in coherence.

The professional economic bureaucracy, and in partic-ular the Ministry of International Trade and Industry(MITI), was a key force behind Japan’s success. Aspostwar reconstruction began, MITI targeted specificgrowth industries such as cameras which wereshielded from overseas competition until theybecame competitive. MITI operated mainly throughpersuasion, thus reducing the risk of major mistakes.

In the 1990s, Japan’s economy suffered prolongedasset deflation and even the once-dominant LDP wasforced into coalition. State-led deflation painted thebureaucracy in a harsher light than did state-ledgrowth. Corruption scandals made large companieswary of hiring retired bureaucrats. But in the earlierpostwar decades, Japan was a preeminent example ofhow a small, meritocratic bureaucracy, operatinglargely on the basis of persuasion, can guide eco-nomic development within a predominantly marketeconomy.

Population: 127m.Gross domestic product per head:

$28,700.Form of government: unitary parlia-

mentary state with a ceremonialemperor.

Legislature (‘Diet’): the 480 members ofthe House of Representatives areelected for a four-year term.Thesmaller upper house, the House ofCouncillors, is less significant.

Executive: an orthodox parliamentaryexecutive, with a cabinet and primeminister accountable to the Diet.

Senior bureaucrats play a major role inpolicy-making.

Judiciary: the 15-member SupremeCourt possesses the power of judicialreview under the 1946 constitutionbut has proved to be unassertive.

Electoral system: under the additionalmember system introduced in 1996,300 members of the lower house areelected in single-member constituen-cies while the other 200 are elected byproportional representation. Althoughthe new system was designed toreduce corruption and increase policy

debate, turnout in 1996 fell sevenpoints to 60 per cent.

Party system: postwar Japanese politicshas been dominated by the conserva-tive Liberal Democratic Party (LDP). Itmonopolized power from its forma-tion in 1955 until 1993 and has partici-pated in coalition rule since 1994. In2003, the LDP won 244 seats, withmajor opposition coming from thereformist Democratic Party of Japan(177) seats).

Profile J A PA N

Further reading: Johnson (1987), Koh (1989), Rosenbluthand Thies (2000).

than it is in democracies. By definition, institu-tions of representation – elections, competitiveparties and freely organized interest groups – areweak in authoritarian regimes, leaving more roomfor agencies of the state to prosper. A dictator candispense with elections or even with legislaturesbut he cannot rule without bureaucrats to imple-ment his will. But the bureaucracy can be morethan a dictator’s service agency, not least in devel-oping countries. Often in conjunction with themilitary, it can itself become a leading politicalforce, claiming that its technical expertise andability to resist popular pressures is the only routeto long-term economic development. This claimmay have initial merit but eventually bureaucra-cies in non-democratic regimes are prone tobecome bloated, over-politicized and inefficient,acting as a drag on rather than a stimulus tofurther progress. In the long run, bureaucraticregimes, like military governments, become partof the problem rather than the solution.

The bureaucracy has undoubtedly played a pos-itive role in most authoritarian regimes that haveexperienced rapid economic growth. In the 1950sand 1960s, for instance, the bureaucracy helpedto foster economic modernization in severalMiddle East regimes. In conjunction with themilitary and a strong national leader such asAbdul Nasser (President of Egypt, 1956–70),modernizing bureaucracies were able to initiatestate-sponsored development even against theopposition of conservative landowners. In similarvein, O’Donnell (1973) coined the term ‘bureau-cratic authoritarianism’ to describe LatinAmerican countries such as Brazil and Argentinain which bureaucratic technocrats, protected by arepressive military government, imposed a moremodern economy against opposition from somesocial groups. Many of the high-performingeconomies of East Asia, such as Indonesia andMalaysia, provide more recent examples of thecontribution the bureaucracy can make to devel-opment in authoritarian settings.

But instances of the bureaucracy instigating suc-cessful modernization are the exception ratherthan the rule. More often, the bureaucracy hastended to inhibit rather than encourage economicdevelopment. The experience of sub-SaharanAfrica following independence provides a more

sobering and representative assessment of the roleof bureaucracy in non-democratic regimes. Aftercolonial rulers departed, authoritarian leaders usedtheir control over public appointments as a polit-ical reward, overwhelming the delicate distinctionbetween politics and administration. Their cavalierapproach to public appointments was com-pounded by chronic unemployment which led toexcess labour, especially among the newly edu-cated, being absorbed into the administration as away of buying off trouble. The result was uncon-trolled expansion of the civil service; by the early1990s, public employment accounted for mostnon-agricultural employment in Africa (Smith,1996, p. 221). Once appointed, public employeesfound that ties of kinship meant that they wereduty-bound to use their privileged positions toreward their families and ethnic group, producingfurther expansion of the civil service. The resultwas a fat bureaucracy incapable of acting as aneffective instrument for development. Rather, theexpanding administrative ‘class’ extracted resourcesfrom society for its own benefit, in that sense con-tinuing rather than replacing the colonial model.Only towards the end of the twentieth century,under pressure from international agencies, wereattempts made to rein in the public sector throughan emphasis on building administrative capacity(Turner and Hulme, 1997, p. 90).

DefinitionAdministrative capacity refers to the bureaucracy’s ability to address social problemsthrough effective management and implementation of public policy.

Even where bureaucratic-led development hassucceeded, the formula often outlasts its useful-ness. As several East Asian states began to discoverat the end of the twentieth century, public admin-istrators are more effective at building an indus-trial economy than at continuing to manage itonce it becomes mature and open to internationalcompetition. In Indonesia, for example, the Asianfinancial crisis of the late 1990s exposed theextent to which investment patterns had been dis-torted by ‘crony capitalism’, with access to capitaldepending on official contacts rather than theanticipated rate of return.

306 GOVERNMENT AND POLICY

The position of the bureaucracy in totalitariansystems in some ways echoes its role in authori-tarian regimes. But one key difference marks outadministration in the communist world in partic-ular: its sheer scale. The size of the bureaucracyunder communism flowed from the totalitariancharacter of its guiding ideology. To achieve itstheoretical mission of building a new society, theparty had to control all aspects of development,both economic and social, through the state. Mostobviously, the private sector disappeared and theeconomy became an aspect of state administra-tion. In the extreme case of the Soviet Union vir-tually every farm, factory and shop formed part ofthe bureaucracy. The shop assistant, the butcher,the electrician – all were employees of the Unionof Soviet Socialist Republics. This required onearmy of administrators to do the work andanother to provide coordination. The SovietUnion became the most bureaucratic state theworld had ever seen.

DefinitionThe nomenclatura (Russian for a list of names)was a panel of trusted people from which rulingcommunist parties appointed to posts in thebureaucracy.

In addition, communist bureaucracies wereintensely political, with the ruling party pene-trating deeply into the administration. Indeed, theessence of communist rule lay in combiningbureaucratic and political rule in one giganticsystem. The party regarded the bureaucracy asboth indispensable and potentially unreliable – asa force which through its control of implementa-tion might one day dominate its political masters(Lewin, 1997). Hence, the party sought to domi-nate the bureaucracy in the same way that it con-trolled the armed forces: by controlling all majorappointments. It achieved this goal through thenomenklatura, a mechanism that provided a pow-erful incentive for the ambitious to gain and retaina sound reputation within the party. The nomen-klatura system continues to this day in China,where the list is now said to contain over eightmillion names (Manion, 2000, p. 434).

Fascist regimes provide both similarities andcontrasts with the communist approach to bureau-

cracy. Like communist states, fascism was an ide-ology of mobilization, seeking to place the entireresources of the society at the service of an expan-sionist state. Although in Germany Hitler sought‘war in peacetime’, unlike communist leadersHitler was not interested in how the administra-tion should achieve his demanding objectives.Instead, the ‘leadership of men’ was regarded assuperior to sterile rules and bureaucratic proce-dures. Hitler applauded the personal form of rulepractised in Germany’s annexed territories, wherelocal commanders wielded complete power. TheFührer himself was a poor administrator, prefer-ring highly informal decision-making whenindeed he could be persuaded to take decisions atall. As Mommsen (1997, p. 75) writes, ‘the Nazidictatorship did not so much expand govern-mental prerogatives through bureaucratic means asprogressively undermine hitherto effective publicinstitutions through arbitrary use of power’.Caplan (1988, pp. 322–3) argues that underHitler the ‘subversion of the civil service waspiecemeal and ad hoc, the effect of incompetence,impatience and neglect rather than the pursuit of aclear alternative’. The profoundly non-bureau-cratic character of Nazi rule – described by Caplanas ‘government without administration’ – contrastsdeeply with communist practice.

Key reading

Next step: Peters (2000) is a clear and compara-tive introduction to bureaucracy.

Heady (1996) is an alternative to Peters, whilePage (1992) is a comparative study adopting aWeberian approach. Bekke et al. (1996) is a livelyedited collection. Osborne and Gaebler (1992) isan enthusiast’s account of new public manage-ment. Verheijen and Coombes (1998) assessNPM’s impact in Europe, both East and West,while Minogue et al. (1999) focus more on devel-oping countries. For bureaucracy in particularcountries and regions, see Chandler (2000) for arange of consolidated democracies, Fesler andKettl (1996) for the United States, Campbell andWilson (1995) for Britain viewed comparatively,Boston (1995) for New Zealand and Page and

THE BUREAUCRACY 307

Wright (1999) for Europe. For Japan, Johnson(1987) is an influential interpretation. Kershawand Lewin (1997) contains useful chapters onbureaucracy in the contrasting dictatorships of

Stalin and Hitler. With an emphasis on the rela-tionship between administration and develop-ment, Turner and Hulme (1997) portray the roleof bureaucracy in authoritarian regimes.

308 GOVERNMENT AND POLICY

Stages of the policy process 309Public policy in established democracies 315Public policy in new democracies 319Public policy in authoritarian states 322Key reading 324

The task of policy analysis is to understand whatgovernments do, how they do it and what dif-

ference it makes (Dye, 2001). Where orthodoxpolitical science examines the organization of thepolitical factory, policy analysis examines theproducts emerging from it. So the focus is on thecontent, instruments, impact and evaluation ofpublic policy. The prime emphasis is downstream,on the implementation and results of policy,rather than upstream, on the origins of policy inits institutional sources. Policy analysis is con-cerned with improving the quality and efficacy ofpublic policy, giving the subject a distinctly prac-tical air. Policy analysts want to know whether andwhy a policy is working and how else it might bepursued.

DefinitionA policy is a broader notion than a decision. At aminimum, a policy covers a bundle of decisions.More generally, it reflects an intention to decidein a particular way in the future. According toColebatch (1998), policies are expected to showcoherence (policy as strategy), hierarchy (policyas instructions to staff ) and instrumentality(policy as purpose). In fact, many policies arelittle more than aspiration (policy as window-dressing).

Stages of the policy process

In approaching the study of public policy, it ishelpful to distinguish the five stages of the policyprocess shown in Figure 17.1. These divisions are

more analytical than chronological, meaning thatin the real world they often overlap. It is, ofcourse, possible to find examples of policy that doclosely follow the sequence shown in the figure.An oft-cited example is Britain’s Clean Air Act1956, a successful attempt to use smoke abate-ment to eliminate unhealthy winter smogs inLondon. An issue emerged, a solution was pro-posed and the problem duly went away(Sanderson, 1961). Although no contemporarypolicy analyst would regard this example as typicalof the rather uncertain nature of most policy, abrief review of the stages approach will nonethelesshelp to introduce the distinct focus of theapproach.

Chapter 17

The policy process

Figure 17.1 Stages of the policy process

INITIATIONdeciding to make a

decision

FORMULATIONdeveloping concrete

proposals

IMPLEMENTATIONputting the policy into

practice

EVALUATIONappraising impact and

success

REVIEWcontinue, revise or

terminate?

309

Initiation

Why did governments expand welfare policy forthe first three decades after 1945 and then reduceit thereafter? Why did many Western governmentstake companies into public ownership after thewar and then start selling them back to the privatesector in the 1980s and 1990s? These are ques-tions about policy initiation – about the decisionto make (or, almost as important, to change)policy in a particular area. As these examples illus-trate, the policy agenda often shows considerablesimilarity across countries. This resemblancereflects some mixture of shared problems, interna-tional competition and explicit lesson-drawingacross countries.

Control over policy initiation would be anaccomplished form of power. At least withindemocracies, however, the agenda cannot be con-trolled by a single group. Rather, policy proposalsemerge from what Kingdon (1984) calls the‘policy primeval soup’ in uncertain and unpre-dictable ways. Whether particular issues and solu-tions are extracted from the soup depends on theopening of policy windows (such as the opportu-nities for innovation created by the election of anew government). These openings soon close: thecycle of attention to a particular issue is short, andboth political debate and the public mood con-tinue to evolve. So Kingdon suggests that policyentrepreneurs are not confined to social move-ments (see p. 130). A similar role needs to be per-formed in orthodox policy-making in order toensure that a given proposal is put forward whenits moment arrives.

Beyond the political process itself, we can iden-tify three general influences on policy initiation:science, technology and the media. Science isclearly one factor driving policy agendas. Thecurrent concern with global warming, for instance,began primarily with scientific assessments of itsfuture implications. Rather unusually, the scienceforewarned us that a problem was on its way.

However, we should note that governments com-mission as well as consume science. Ministers’power over the production of knowledge can helpthem to keep uncomfortable issues off the policyagenda. The British government, for instance, wasunduly slow in commissioning research into mad

cow disease which swept through the country’sherds (and into the brains of a few of its people) inthe 1990s. In this case, manipulation of the researchprocess proved fatal. Thus scientific research cannotbe seen as wholly external to politics.

Second, the application of science – that is, tech-nology – also influences policy initiation.Governments have an incentive to encourage newtechnology: it stimulates economic competitive-ness and its prompt introduction allows a countryto steal a march on other states. But many newtechnologies involve regulation. Designs fornuclear power stations must be assessed for safety;frequencies for mobile telephones must be allo-cated; new drugs must be examined for sideeffects; and the effects of allowing geneticallyaltered plants, fish and animals into the environ-ment must be considered. In all these cases tech-nological innovation forces policy initiation fromgovernment. The danger is that governments willapprove those ‘advances’ which offer clear short-term gain at the cost of wider long-term risk.Nuclear energy is arguably an example.

Third, the ability of the media to highlightissues means they are also a significant, if some-times overstated, influence on policy initiation. Asingle and perhaps unusual incident is often takenup by one media outlet and then, through packjournalism, amplified by other channels untilpoliticians are forced to respond to what has sud-denly become a serious social problem. Commontopics of such media-induced moral panicsinclude asylum seekers, crime waves, drug crazes,food scares, gang wars, infectious diseases, maddogs, teenage hooligans and welfare scroungers(Henshel, 1990). These balloons of concern oftencollapse as quickly as they are inflated; only in afew cases are the media turning their searchlighton a problem where public policy does indeedneed to be initiated.

Formulation

Once a decision has been taken to address a spe-cific problem, policy must be formulated.Deciding to decide is one thing; deciding what todecide is much more difficult. The coalition whichformed behind the vague idea that somethingmust be done soon dissolves when it comes to

310 GOVERNMENT AND POLICY

concrete proposals producing losers as well aswinners. Formulating precise legislative andadministrative proposals in response to diffusesupport for action is a task requiring good politicalcraft.

In analysing policy formulation, analysts havedeveloped two models: the rational or synopticmodel associated with Simon (1983) and theincremental model developed by Lindblom(1979). These accounts form an important part ofthe policy analysis tradition and we will considerthem in more detail.

The key contrast between the two models is this.The rational model views policy as emerging froma systematic search for the most efficient means ofachieving defined goals. By contrast, the incre-mental model sees policy as resulting from a com-promise between actors who have goals which areill-defined or even contradictory (Box 17.1). Putdifferently, where the rational model seeks the bestpolicy in theory, an incremental framework seeksout a practical policy acceptable to all the interestsinvolved.

An example will clarify this contrast. Supposewe are in charge of education and we have decidedto initiate a policy to improve students’ perfor-

mance. If we were to adopt a rational approach,we would first specify the outcomes sought, suchas the proportion of students achieving a givenlevel of qualifications. Then we would considerthe most efficient means of maximizing that goal:should we invest in new schools, improved facili-ties, more teachers or some combination thereof?

An incremental approach, however, starts from adifferent point. Here we would begin (and end)with regular consultation with all the organizedinterests: teacher unions, parents’ associations,educational researchers. We would hope that fromthese discussions a consensus would emerge onhow extra resources should be spent. The long-term goals of this expenditure might not be mea-sured or even specified but we would assume thata policy acceptable to all is unlikely to be disas-trous. Such an approach is policy-making by evo-lution, not revolution; an ‘increment’ is literally asmall change to an existing sequence.

Of these two approaches, the rational model isclearly more demanding, requiring detailedanalysis rather than careful politics. Specifically, itrequires policy-makers to:

� rank all their values � formulate specific options � check all the results of choosing each option

against each value� select the alternative that achieves most values.

DefinitionCost–benefit analysis involves giving a mone-tary value (positive or negative) to every conse-quence of choosing each option and thenselecting the option with the highest net benefit.Thus, the efficiency gain from adding a newrunway to an airport can be netted off againstthe additional noise pollution for local residents.The technique is transparent but time-con-suming.

This is an unrealistic counsel of perfection. Itrequires policy-makers to foresee the unforeseeableand measure the unmeasurable. Even so, tech-niques such as cost–benefit analysis (CBA) havedeveloped in an attempt to implement aspects ofthe rational model (Boardman et al., 2000).Seeking to analyse the costs and benefits associated

THE POLICY PROCESS 311

Rational model Incremental model

Goals are set before Goals and means are means are considered considered together

A good policy is the A good policy is one on most appropriate to which all main actors can achieve explicit goals agree

Analysis is Analysis is selective; the comprehensive; all object is acceptable effects of all options policy, not the best policyare addressed

Theory is heavily used Comparison with similar problems is heavily used

Sources: Adapted from Lindblom (1959, p. 81) and Parsons (1995,p. 285).

B OX 1 7 . 1

Rational and incremental modelsof policy-making

with each possible decision does have strengths,particularly when one from a small set of optionsneeds to be chosen.

CBA brings submerged assumptions to thesurface and can benefit those interests whichwould otherwise lack political clout. It contributesto transparent policy-making by forcing decision-makers to account for policies where costs exceedbenefits. In other words CBA can help to keeppolicy-makers honest. For such reasons, CBA isnow formally applied to any regulatory proposalin the USA expected to have an impact on theeconomy exceeding $100m.

However, CBA, and with it the rational modelof policy formulation, also has weaknesses. Itunderplays soft factors such as fairness and thequality of life. It is also cumbersome and expensiveand in the real political world its conclusions areoften sidestepped. Just as the ‘rational’ model isnot always more rational than an incrementalapproach, so too a CBA of CBA would not alwaysyield positive results.

The incremental model was developed byLindblom (1959) as part of a reaction against therational model. Lindblom’s starting point is thatpolicy is continually remade in a series of minoradjustments, rather than as a result of a single,comprehensive plan. Incrementalism representswhat Lindblom calls the science of muddlingthrough. This approach may not lead to achievinggrand objectives but, by taking one step at a time,it does at least avoid making huge mistakes. Inincremental policy-making, what matters is notthat those involved should agree on objectives butthat agreement should be reached on the desir-ability of following a particular policy, even whenobjectives differ. Hence policy emerges from,rather than precedes, negotiation with interestedgroups.

A broadly comparable view of policy-makingemerges from Kingdon’s rather alarmingly titledgarbage can model of decision-making (1984).The garbage can model imagines that issues andpolicy-makers mix at random within an organiza-tion, just as different forms of trash blend haphaz-ardly in a garbage can. In a university, for instance,a committee containing one group of staff mayhappen to address one aspect of a particularproblem, with some implications possibly picked

up later by a completely separate group within theinstitution. Policy-making is partial, fluid, chaoticand incomplete, exhibiting no more rationalitythan is found in the arrangement of waste in adustbin (Cohen et al., 1972).

Although Kingdon was mainly concerned withdecision-making within organizations, his perspec-tive also has relevance to the political system as awhole. Clearly, his model suggests that real policy-making is far removed from the rigours of ratio-nality. At best, some problems will be partlyaddressed some of the time (Bendor et al., 2001).

As Lindblom (1977, 1990) himself came to rec-ognize, incremental policy formulation deals withexisting problems rather than with avoiding futureones. It is politically safe but unadventurous;public policy becomes remedial rather than innov-ative. But the threat of ecological disaster, forinstance, has arisen precisely from human failureto consider the long-term, cumulative impact ofindustry upon the environment. So neither incre-mentalism nor decision-making via the garbagecan will provide an adequate response to complexglobal issues. The trick is to find ways of dealingwith such problems which do not fall victim tothe shortcoming of the rational model, namelydemanding more of policy-makers than they canpossibly deliver.

Implementation

After a policy has been set, it must be put intoeffect. An obvious point, of course, except thatmuch political science stopped at the point wheregovernment reached a decision, ignoring themyriad difficulties which arise in policy execution.Probably the main achievement of policy analysishas been to direct attention to these problems ofimplementation.

Even today, putting the policy into practice issometimes still regarded as a technical matter ofadministration. Turning a blind eye to implemen-tation issues can be politically convenient for min-isters but is always dangerous and can bedamaging. For example, the British government’sfailure to prevent mad cow disease from crossingthe species barrier to humans in the late 1980s is aclassic instance of implementation failure. Officialcommittees instructed abattoirs to remove infec-

312 GOVERNMENT AND POLICY

tive material (such as the spinal cord) from slaugh-tered cows but initially took no special steps toensure these regulations were carried out carefully.As a result of incompetence in slaughterhouses,the disease agent continued to enter the humanfood chain.

Again reflecting the standard conventions ofpolitical science, the traditional view of implemen-tation adopted a top-down approach (Hogwoodand Gunn, 1984). The question posed was theclassical problem of bureaucracy: how to ensurepolitical direction of unruly public servants.Elected ministers had to be able to secure compli-ance from departments and agencies already com-mitted to pet projects of their own. Withoutvigilance from on high, sound policies would behijacked by lower-level officials obsessed withexisting procedures, thus diluting the impact ofnew initiatives.

DefinitionA top-down approach conceives the task ofpolicy implementation as ensuring that policyexecution delivers the outputs and outcomesspecified by the policy-makers. By contrast, abottom-up approach considers that the role ofthose who execute policy in reshaping broadobjectives to fit local and changing circum-stances should be both recognized and wel-comed.

But this top-down approach to implementationfocuses excessively on control and complianceissues. Like the rational model of policy-makingfrom which it springs, it may be an unrealistic andeven counterproductive approach. The plausiblesuggestion emerging from a bottom-up perspec-tive is that policy-makers should seek to engagerather than control those who translate policy intopractice. Writers in this tradition, such as Hill andHupe (2002), ask: what if circumstances havechanged since the policy was formulated? Andwhat if the policy itself is poorly designed?

Many policy analysts now suggest that objectivesare more likely to be met if those who executepolicy are given flexibility over its application andtherefore its content. At street level (the pointwhere policy is delivered), policy emerges frominteraction between local bureaucrats and affected

groups. Here at the sharp end, goals can often bebest achieved by adapting them to local condi-tions. If the policy is left unmodified, its fate willbe that of the mighty dragon in the Chineseproverb: no match for the neighbourhood snake.

For instance, education, health care and policingmust surely differ between the rural countrysideand multicultural areas in the inner city. Further,local implementers will often be the only peoplewith full knowledge of how policies interact, andsometimes contradict, each other. They will alsoknow the significant actors in the locality and willbe able to seek their support. When the Americanpolitician Tip O’Neill said, ‘all politics is local’, hecould well have had policy implementation inmind.

So a bottom-up approach reflects an incrementalview of policy-making in which implementation isseen as policy-making by other means. Thisapproach reflects the contemporary emphasis ongovernance, with its stress on the many stake-holders involved in the policy process.

Evaluation

Just as policy analysis has increased awareness ofthe importance of implementing policy, so too hasit sharpened our focus on evaluation. The job ofpolicy evaluation is to work out whether a policyhas achieved its goals. Like the recipe for rabbitstew which begins ‘first catch your rabbit’, thisneatly sidesteps initial difficulties. As we have seen,the motives behind a policy are often multiple,unclear and even contradictory. This ‘mushiness ofgoals’, to use Fesler and Kettl’s phrase (1996, p.287), means that policy-makers’ intent is often apoor benchmark for evaluation. Given that mostpolicy is made incrementally, identifying precisetargets is a difficult task.

The question of evaluation was often ignorednot just in traditional political analysis but also bygovernments themselves. Sweden is a typicalexample. In the postwar decades, a succession ofSocial Democratic governments concentrated onbuilding a universal welfare state without evenconceiving of a need to evaluate the effectivenesswith which services were delivered by anexpanding bureaucracy. In France and Germany,and other continental countries where bureau-

THE POLICY PROCESS 313

cratic tasks are interpreted as the legalistic applica-tion of rules to cases, the issue of policy evaluationstill barely surfaces.

Yet without some evaluation of policy, govern-ments will fail to learn the lessons of experience.In the United States, Jimmy Carter (President,1977–81) did insist that at least 1 per cent of thefunds allocated to any project should be devotedto evaluation. The vast number of reports requireddid not noticeably improve policy effectiveness. Itwas not until a wave of public sector reform in the1990s that evaluation began to return to the fore.To take the USA again, the GovernmentPerformance and Results Act 1993 now requireseach agency to perform an annual programmeevaluation which is then intended to be used bythe government to revise plans and budgets.Similarly, the Labour government elected inBritain in 1997 claimed a new pragmatic concernwith evidence-based policy: what matters is whatworks (Sanderson, 2002). In some other democra-cies, too, public officials began to think, often forthe first time, about whether policies wereachieving their goals and at what cost.

DefinitionPolicy outputs are what government does;policy outcomes are what governmentachieves. Outcomes are the activity; outcomesare the effects, both intended and unforeseen.Outputs are measured easily enough: so manynew prisons built or a specified increase in thestate pension. Outcomes are harder to ascertain:a reduction in recidivism or in the number ofelderly people living in poverty.

Evaluation studies distinguish between outputs(what government does) and outcomes (what gov-ernment achieves, including unintended conse-quences). The link between the two is oftentenuous. In 1966, for example, the US govern-ment published the results of the ColemanReport, a massive sociological study of Americansecondary schools. This renowned study foundthat outputs such as teachers’ salaries and educa-tional expenditure had little effect on the ultimateoutcome of education: namely, children’s learning.The main influence on pupils’ educational successwas the family background of the child and its

peers, not public spending. Children from alower-class background were likely to underper-form even if they were placed in a well-resourcedschool. The Coleman Report illustrates a pointfamiliar to most policy-makers: outcomes resistchange, even when resources are devoted toaltering them.

Just as policy implementation in accordancewith the top-down model is an unrealistic goal, sojudging policy effectiveness against specific objec-tives is an implausibly scientific approach to evalu-ation. A more bottom-up, incremental view is thatevaluation should simply gather in the opinions ofall the stakeholders affected by the policy, yieldinga qualitative narrative rather than a barrage ofimplausibly precise statistics. As Parsons (1995, p.567) describes this approach,

evaluation has to be predicated on wide and fullcollaboration of all programme stakeholders:agents (funders, implementers); beneficiaries(target groups, potential adopters); and thosewho are excluded (‘victims’).

In such a naturalistic evaluation, the varyingobjectives of different interests are welcomed.They are not dismissed as a barrier to objectivescrutiny of policy. This is a more bottom-up –indeed incremental – approach because the stake-holders might agree on the success of a policy eventhough they judge it against different standards.

Review

Once a policy has been evaluated, or even if it hasnot, the three possibilities are: to continue, torevise or to terminate. Most policies, or at least thefunctions associated with them, continue withonly minor revisions. Once a role for governmentis established, it tends to continue. But the agencycharged with performing the function does changeover time. In the United States, for instance, 426separate agencies were established between 1946and 1997 but a majority of these had been termi-nated by the end of the period. The agent ofchange was usually an alteration in the partisancolour of the administration (Lewis, 2002). So theobservation that there is nothing so permanent asa temporary government organization appears to

314 GOVERNMENT AND POLICY

be wide of the mark. Functions continue but theagencies performing them can change.

Yet even if agency termination is surprisinglycommon, the intriguing question remains: why ispolicy termination so rare? Why does governmentas a whole seemingly prefer to adopt new functionsthan to drop old ones? Bardach (1976) suggests fivereasons for the difficulties of policy termination:

� Policies are designed to last a long time.� Policy termination brings conflicts which leave

too much blood on the floor.� No one wants to admit the policy was a bad

idea.� Policy termination may affect other pro-

grammes and interests.� Politics rewards innovation rather than tidy

housekeeping.

Public policy in established democracies

While policy can be usefully analysed into itscomponent parts, it is also helpful to take abroader view. In this section, we seek to chart themajor shifts in the policy agenda of Western states,transformations which reflect evolving concep-tions of the state itself yet which, like other aspectsof policy analysis, remain understated in descrip-tions of government institutions.

Broadly, we can divide the history of public policyin what are now the established democracies intothree phases: the nightwatchman or liberal state ofthe nineteenth century and earlier, the welfare stateof the later twentieth century and the emerging reg-ulatory state of the twenty-first century (Box 17.2).Such a scheme fits the experience of NorthernEuropean states, including the UK, with fair accu-racy. However, the history of the USA, which neverdeveloped a welfare state but which was one of thefirst democracies to introduce independent regula-tory agencies, is exceptional once more.

The nightwatchman state

The nightwatchman state was a minimal opera-tion, concentrating on maintaining law and order,protecting private property and extracting suffi-

cient resources to allow rulers to pursue theirforeign policy. The state apparatus remainedpoorly developed, with a limited bureaucracy andlocal administration largely the responsibility ofprovincial notables. The nightwatchman role wasan early form taken by the post-feudal but pre-democratic state in much of Europe, notably theUnited Kingdom, up to the nineteenth century; itreflected a liberal philosophy of non-intervention.

The nightwatchman metaphor comes from JohnLocke (1632–1704), an English philosopher wholaid the early foundations of liberal thinking.Locke considered the sole function of governmentto be that of protecting natural, God-given rightsto life, liberty and property. In Locke’s view, citi-zens should merely be provided with order, protec-tion and the means of enforcing contracts.

In the United States, a country built to a liberaldesign, Thomas Jefferson (1743–1826) expressedthe nightwatchman conception when he wrote:‘that government is best which governs least’. Thenightwatchman state had no interest in publicwelfare: ‘the drunkard in the gutter is just wherehe ought to be’, commented the American sociol-ogist William Sumner (1840–1910).

THE POLICY PROCESS 315

Type of state Domestic agenda Period

Nightwatchman Maintains law Nineteenth state and order and century and

protects private earlierproperty

Welfare state Provides Second half minimum of twentieth welfare to all century,citizens particularly

in Western Europe

Regulatory state Sets rules and Final standards decades of

the twentieth century and later

B OX 1 7 . 2

The changing agenda of theWestern state

In America and beyond, this liberal philosophyof clear but limited individual rights reflected thestruggle for religious toleration. However, limitingthe state to a night watchman role also provedhighly congenial to the emerging business class.Indeed, the free-market doctrine of laissez faire (‘toallow to do’) was perhaps the most significantelement in the nightwatchman construct. AsOpello and Rosow (1999, p. 97) comment,

The liberal state, then, is in one respect aminimal state; that is, it is deliberately struc-tured not to be itself a threat to the ‘naturalright’ of property ownership, which is the ulti-mate justification for the dominant position ofthe bourgeoisie within the state.

The welfare state

The welfare state, which reached its zenith inWestern and especially Northern Europe in the1960s and 1970s, was clearly based on a moreexpansive and positive view of the state’s role.However, the culmination of the welfare state in thepostwar decades reflected a long evolution. Even as

industrialization proceeded in the nineteenthcentury, the nightwatchman state had been drawninto a measure of economic and environmental reg-ulation. In Britain, for example, the Factory Act1833 had created a framework for inspecting facto-ries while the Factory Act 1847 limited the workingday to a maximum of ten hours.

However, the real origins of collective welfareprovision lay in Germany before the First WorldWar. Under the rule of Otto von Bismarck(German Chancellor, 1871–90), Germany hadpioneered social insurance schemes which sharedrisks such as accident and illness, at least forindustrial workers. Building on this innovativeGerman model, the period from the 1920s sawthe gradual extension of collective welfare in mostdemocracies to more areas of life (e.g. pensionsand family allowances) and to more groups in thepopulation (e.g. rural people and dependants ofindustrial workers). By the 1970s, coverage for themain aspects of welfare extended to virtually theentire population in most democracies (Table17.1).

By then, European democracies had become‘welfare states’ in a triple sense. First, the state took

316 GOVERNMENT AND POLICY

Table 17.1 Introduction of social insurance to some democracies

Industrial Health Pensions Unemployment Family accident benefit allowances

Australia 1902 1945 1909 1945 1941

Austria 1887 1888 1927 1920 1921

Canada 1930 1971 1927 1940 1944

Denmark 1898 1892 1891 1907 1952

Finland 1895 1963 1937 1917 1948

France 1898 1898 1895 1905 1932

Germany 1871 1883 1889 1927 1954

Netherlands 1901 1929 1913 1916 1940

New Zealand 1900 1938 1898 1938 1926

Norway 1894 1909 1936 1906 1946

Sweden 1901 1891 1913 1934 1947

* Innovator in bold.

Source: Adapted from Pierson (1998), table 4.1.

a prime role in ensuring the provision of aminimum standard of welfare to all its citizens,supplanting the ad hoc provision throughchurches and charities in the nightwatchman state.Second, providing welfare became the main func-tion of public administration, consuming thelion’s share of both taxpayers’ revenue and officials’time. Third, welfare rights became an expressionof social citizenship, a notion which sought toextend the scope of democracy itself.

The concept of social citizenship was developedby the British sociologist T. H. Marshall(1893–1981). In an influential account written inthe aftermath of war, Marshall (1950, pp. 16–19)defined social citizenship as covering ‘the wholerange from the right to a modicum of economicwelfare and security to the right to share to the fullin the social heritage and to live the life of a civi-lized being according to the standards prevailingin society’. Influenced by such considerations,several democracies – for instance, France – beganto embed a statement of welfare rights in theirconstitution.

This development of social citizenship contrastssharply with the nightwatchman state in whichthose who received public welfare through thepoor laws were denied the vote. Where the night-watchman state gave priority to liberty, the welfarestate was premised on equality.

Of course, democracies continued to vary bothin their levels of welfare support and in the ideo-logical support offered. Several factors help toaccount for variations in the share of nationalproduct devoted to public welfare:

� Wealthier countries spend a higher proportionof national income, and not just a higheramount, on welfare (Wilensky, 1984).

� Unitary states (such as Britain) tend to behigher spenders than federations (such asCanada). Unitary status permits nationalwelfare schemes to be set up and administered,often through local government.

� Welfare spending is higher in countries whereparties of the left predominate (e.g. the SocialDemocrats in Sweden).

� Countries such as Austria in which Catholicparties are a major governing force also tend tobe generous providers. Catholic traditions

favour social support, particularly when admin-istered through the church.

� Low spenders include countries where collectiveprovision is still seen as ‘state welfare’ ratherthan a ‘welfare state’. The USA is still seen as anexample of this ‘residual’ or ‘liberal’ welfarestate despite an expansion of governmentresponsibility in the New Deal (1933–39), andheavy contemporary support for the elderly(Esping-Andersen, 1990).

DefinitionsIn a public pay-as-you-go pension scheme, thepensions of retirees are paid directly by contribu-tions from current workers.This system, whichforms an important plank of pensions provisionin much of continental Europe, becomes unsus-tainable as the population ages and the depen-dency ratio (the number of pensioners dividedby the number of contributors) deteriorates.Thestate will need to take corrective action.

By contrast, in a funded pension system,usually operated through private or occupa-tional pension providers, individual contributionsare invested on behalf of the individual so that apot of capital accumulates from which a retire-ment income can be drawn. Funded pensionsshould require little call on the public purse.Funded pensions predominate in, for example,the Netherlands, the United Kingdom and theUSA.

The 1980s witnessed the first real setbacks forthe welfare state. The underlying problem wasfinancial: as the average age of the populationincreased, so the total cost of pensions, medicalcare and support services went up – and was pro-jected to continue growing well into the twenty-first century. At the same time the workingpopulation, which shoulders the burden ofsupport for the elderly, declined in number.Financial projections looked particularly bleak inthose countries, such as Germany, which com-bined a greying population with a pay-as-you-gopensions scheme. But even in the USA projectionsof future welfare spending were daunting. Giventhe expanding cost of its Social Security andMedicare schemes, one journalist was even movedto describe the American government as ‘a giant

THE POLICY PROCESS 317

insurance company with a small defence affiliate’(i.e. its military forces) (Plender, 2003).

International economic pressures also impingedon thinking about welfare. If the cost of onecountry’s welfare system was higher than all therest, the international competitiveness of itseconomy might be endangered. Pierson (1998)suggests that the move to a more open interna-tional economy at least ‘curtailed opportunities forthe further development of national welfare states’.Certainly the oil crises of the 1970s, marking atransition to slower growth in the establisheddemocracies, were a catalyst to rethinking first thefeasibility and then the desirability of state-pro-vided welfare from cradle to grave.

Such problems led to some retrenchment of thewelfare state in the 1980s and 1990s. Benefitswere marginally reduced, eligibility rules weretightened, charges were introduced for servicessuch as medical treatment, few new commitmentswere taken on and the state made an effort torevive the old caring agencies, such as charities andthe church.

But the welfare state experienced a correctionrather than a crisis. On the whole, taxpayers’revolts did not materialize and the basic structuresof the welfare state remained in place. In practice,the generous welfare states found in Scandinaviadid not seem to suffer a dramatic loss of economiccompetitiveness. At least in the Nordic world,Kuhnle (2000, p. 226) argues that welfare provi-sion helped to maintain political legitimacythrough difficult times:

the Nordic experience in the 1990s shows thatthe universal and comprehensive welfare statecan be a vital shock absorber which stabilizes theeconomy so that the economy can recuperatefast and well.

So Marshall’s notion of social citizenship may betouted less often in the twenty-first century butboth its cash value and its public popularityremain substantial (Esping-Andersen, 2002).

The regulatory state

Although the crisis of the welfare state may havebeen overplayed, the final decades of the twentieth

century did witness a fundamental shift in theagenda and focus of public policy in many estab-lished democracies, especially in Europe. In socialwelfare, service delivery was increasingly contractedout to private agencies; in the economy, publicindustries were privatized (Feigenbaum et al., 2003).

Since one motive here was to render economiesmore competitive, this transition is sometimes dis-cussed as a move from a welfare state to a competi-tion state (Cerny, 1990). But it is not clear that thefrontiers of the state have been decisively modified;indeed, newly privatized monopolies must be regu-lated and public oversight has become more intensein such sectors as education, the environment,employment, scientific research and consumer pro-tection. Indeed, smart regulation – allocating radiofrequencies to mobile telephony, encouragingbroadband internet use, developing digital televi-sion – can enhance growth in advanced economies.

For want of a happier term we will describe thisshift in the policy approach as a transition to aregulatory state (Box 17.2). Such a state uses rules,standards and other public statements as majorpolicy instruments, rather than relying on directprovision of goods and services. More than inmost democracies, regulation has always been aleading mode of governance in the USA; thenotion of a regulatory state suggests that Americanpractice is now diffusing to other establisheddemocracies in Europe and elsewhere.

The most striking evidence for the retreat of thestate from direct provision comes from the influen-tial policy of privatization followed by Britain’sConservative government under Mrs Thatcher inthe 1980s. The speed and thoroughness with whichthe United Kingdom government ‘sold off thefamily silver’ (to quote the disapproving words offormer Conservative prime minister HaroldMacmillan) attracted international interest (Table17.2). Just as nationalization had been seen by thepostwar Labour government as a way of intro-ducing public control and rational planning to keyindustrial sectors, so privatization reflected an ideo-logically charged desire to ‘roll back the frontiers ofthe state’.

The key point, though, was that creating privatemonopolies – as with telephones, gas and electricity– required the creation of new offices of regulationat least until competition became established. As

318 GOVERNMENT AND POLICY

Majone (1996, p. 2) notes, ‘in Britain and else-where, the privatization of the public utilities hasbeen followed by price regulation . . . the lastfifteen years have been a period less of deregulationthan of intense regulatory reform’. The outcome isas much a regulatory state as a rolled-back state.

In analysing the rise of the regulatory state inBritain, Moran (2003) identifies three majordimensions. These are:

� regulation of privatized industries where com-petition remains weak (e.g. water supply)

� external supervision of previously self-regulatinginstitutions such as universities and financialmarkets

� social regulation in such areas as equal opportu-nities, health and safety and food standards.

This new style of governance through regulationdoes, however, bring forth new problems. Perhapsthe most important of these is the issue of democ-ratic legitimacy. Most new regulatory bodiesoperate not as divisions of ministries but as agen-cies operating at one remove from the centres ofpolitical – and therefore democratic – power.Similar attempts to depoliticize regulation can beobserved in the independence given to centralbanks to set interest rates, as now applies in bothBritain and the eurozone. In the European Union,also, the capacity of the unelected Commission toissue enforceable regulations represents its majorlever of influence.

Majone (1996) suggests two reasons in defenceof this trend away from political control, eventhough it runs counter to orthodox democraticthinking. First, autonomous agencies can adoptmore consistent, credible and long-term policiesthan is feasible for elected politicians who remainsubject to short-term pressures from the voters. Ifan unelected central bank can control inflationmore effectively than an elected government, theargument goes, surely the bank should be giventhe job?

Second, delegating political authority to special-ized professional regulators is most appropriatefor issues that are technical (e.g. telecommunica-tions) rather than redistributive (e.g. taxation) innature. The level of taxation is a political questionfor which there is no technical answer but theremay well be a single best way of regulating mobiletelephony. Clearly, as the regulatory state matures,so we may expect more discussion about itsuncertain relationship with traditional ideas ofdemocracy.

Public policy in new democracies

‘Rebuilding the ship at sea’ is an apt metaphor forthe problems confronting new democracies ingeneral and post-communist democracies in par-ticular (Elster et al., 1998). The policy challenge isprecisely to design new institutions that restruc-ture the role of the state in society. In establisheddemocracies, institutions process new policies; innew democracies, the policy is to develop newinstitutions.

THE POLICY PROCESS 319

Table 17.2 Case-by-case privatization in theUnited Kingdom, 1980s

Year Company Proceeds privatized (£ million)

1981 British Aerospace 149

1981 Cable & Wireless* 224

1982 Britoil 549

1983 BP* 565

1984 Enterprise Oil 392

1984 Jaguar Cars 294

1984 British Telecom* 3,916

1985 British Aerospace 551

1986 British Gas 5,434

1987 British Airways 900

1987 Rolls Royce 1,362

1987 British Airports Authority 1,281

1988 British Steel Corporation 2,482

1989 Water Authorities 5,240

* Privatized in stages: only the first tranche is shown.

Note: Only privatizations yielding over £100m are shown. Electricitydistribution was privatized in 1990 and electricity generation the fol-lowing year.

Source: Adapted from Prosser and Moran (1994), table 3.2.

On the one hand, the state must reduce itsinvolvement in some sectors, notably theeconomy, to allow room for a more competitiveprivate sector to develop. On the other hand, thestate must develop new modes of regulation: forexample, overseeing the functioning of markets,giving independent authority to the judiciary andcreating an efficient and professional bureaucracy.Thus the state in most new democracies must alterits entire shape, thinning out in some areas butbuilding up in others.

The scope of the policy changes needed to con-solidate a new democracy is far greater than thatinvolved in the transition from welfare to regula-tory states in established democracies. Further, thedifficulties are increased in the aftermath of transi-tion by the threatening combination of unrealisticexpectations and economic decline. If there is onelesson to be learnt from the policy experience ofthird-wave democracies, it is that freshly electedgovernments – and especially their army ofWestern advisers – vastly understated the size,complexity and likely duration of the task ofrecasting the state’s relationship with society.

Post-communist states provide the most dramaticexamples of the policy transformation required ofnew democracies. When the communist order col-lapsed, an entire method of organizing society wentwith it; far from springing a leak, the communistboat sank. The task facing new leaders was totransform societies fuelled by power into societiesbased on rules, a project that still continues.

Take the economy as an example. Under commu-nism, the state owned most major enterprises, with acentral plan offering coordination. Large enterprisesalso served as welfare providers, producing whatElster et al. (1998, p. 204) describe as a ‘tight cou-pling’ between the workplace and social provision:

Firms provided crèches, holiday homes, housing,health services, training and other welfare facili-ties for their staff. Continuous and lifelong par-ticipation in the production process was theproviso of collectivist protection.

This elaborate, inefficient but functioningnetwork could not be quickly replaced by marketmechanisms. Partly as a result of the post-commu-nist experience, it is now clear that market

economies must be constructed; the task is farmore complicated than merely allowing thecommand economy to fade away. A successfulprivate economy, as in the established democracies,is in fact an intricate public and even politicalachievement. It requires entrepreneurs to show ini-tiative, capital markets to provide resources forinvestment, consumers to spend money, courts toresolve disputes, bureaucrats to keep their fingersout of the pie and a government to act as anumpire rather than a player (see Box 17.3).

Initially, few of these conditions were met in anypost-communist state. The legal challenge ofestablishing property rights, as well as tax, compe-tition and bankruptcy laws, was considerable. Inaddition, state-owned enterprises had no experi-ence of marketing their products. To understandthe transformation required, Schmieding (1993, p.236) invites us to imagine

the problems that would befall a Western marketeconomy if all firms discovered one morning thattheir entire staff for sales, advertising, financeand legal matters had gone off on a three-yearholiday to Mars – together with all the lawyers,judges, bankers and public administrators.

A comparison of post-communist privatizationpolicy with that of the established democraciesdemonstrates the difficulties. In Britain, privatiza-tion of state-owned corporations had proceededon a case-by-case basis, with a sophisticated finan-cial sector available to underwrite the issue (that is,to buy any shares left unsold) and to offer adviceon pricing and marketing. Public monopolies weresold into an established private sector, with sweet-eners offered to small investors who were only toopleased to take advantage.

By contrast, in post-communist states, the objectwas not to sell into the private sector but to createa private sector by, in effect, disposing of an entireeconomy. Further, individual enterprises were oftenin poor condition and any commercial bankslacked both the sophistication and the access tocapital available in London’s financial district.

Difficulties abounded. What if the originalowners of the enterprise, whose property had orig-inally been confiscated by the communists,demanded restitution? What about the enterprises’

320 GOVERNMENT AND POLICY

welfare obligations? Why should investors trustgovernments to deliver on their promises? And,most important, who had money to buy?

In practice, post-communist states adopted arange of strategies in restructuring theireconomies. Several Eastern European countriessuch as Hungary and Czechoslovakia could drawon economic liberalization which had precededcommunism’s fall. Throughout the post-commu-nist world, many enterprises were acquired byexisting managers (‘spontaneous privatization’),whether by design, default or theft. Occasionally,organizations were sold to overseas companies, amethod of bringing in foreign expertise andcapital which nonetheless risked domestic unpop-ularity. Or, in many cases, strategic parts of theeconomy were simply left in public hands with theexpectation that performance would improve asmarket disciplines emerged, permitting a case-by-case solution later in the reform cycle.

DefinitionMass privatization refers to attempts by post-communist governments to sell off many state-owned enterprises at the same time, often byproviding the public with low-cost vouchersexchangeable for shares. By contrast, establisheddemocracies (notably Britain) practised case-by-case privatization in which public corporationswere sold one at a time.

Interestingly, many post-communist countries(excluding Hungary) also attempted mass orvoucher privatization at the early stage. Here,coupons which could be exchanged for shares inprivatized enterprises were offered to the public atlittle or no cost. Alternatively, vouchers wereplaced with investment companies that stoodbetween the investing citizen and the enterprise.Mass privatization succeeded in disposing ofmany enterprises at once, using a standard charterfor each enterprise, and contributed to publiceducation about investment; it achieved far more– and far faster – than the better-known case-by-case method pioneered in Britain. Given the tur-bulence of early post-communism, massprivatization can be viewed as a significant policyachievement.

But mass privatization led to mass disappoint-ment since many of the enterprises turned out tobe duds; the citizens of Eastern Europe gainedfewer rewards than did Mrs Thatcher’s army ofsmall shareholders in the UK. Also, voucherstended to fragment control over enterprises at atime when a few powerful shareholders mighthave been able to force through much-neededrestructuring; Hungary’s emphasis on ‘real owners’rather than voucher ownership was understand-able. In retrospect, the voucher method wassimply one step along a winding road to a func-tioning market economy. Lieberman et al. (1996,p. 9) offer an overall assessment:

Mass privatization achieved a great deal by cre-ating a critical mass of private companies onwhich other reforms could build. But consider-able effort would be needed to complete the pri-vatization process, largely through case-by-caseprivatization of large strategic enterprises.

THE POLICY PROCESS 321

Dimension Description

Privatization Creating privately-owned companies

Corporate Ensuring these companies are governance professionally managed, thus

encouraing external investment

Capital markets Developing banks and investment companies with capital to invest and the skill to invest it wisely

Regulatory Establishing a stable framework of policy regulation e.g. patent and copy-

right law, competition policy

Judiciary Ensuring that judges have the training, resources and autonomy to apply rules in a consistent way,resisting bribes and political pressure

Government Facilitating an arms-length relationship between companies and government

B OX 1 7 . 3

Building market institutions inpost-communist societies

Although post-communist states require mostrebuilding, the nature of the task is fundamentallysimilar in other new democracies with a strongstate tradition. In authoritarian Latin America, forexample, the state had not replaced the capitalistsector but had nurtured powerful corporations(and industrialists), producing a requirement forthe new democracy to engineer privatization andderegulation, and to open up protected companiesto domestic and international competition.

Lijphart and Waisman (1996a) note the mainsimilarities between privatization in EasternEurope and Latin America. In both regions,

� The transition occurred as a result of an eco-nomic as well as political crisis of the oldregime, including in the case of Argentina,hyperinflation and massive capital flight.

� The transition was carried out by the state itself,allowing the old elite to recycle itself as the newbourgeoisie.

� Privatization was contested, with oppositioncentred on those whose wealth was threatenedby liberalization.

When we turn to new democracies in thesmaller states of Africa and Latin America, we findthat the problems of institution-building are evengreater. The policy agenda here is about strength-ening the capacity of both the public sector andthe private sectors; the emphasis in established orpost-communist democracies on rebalancing therelationship between public and private is adubious prescription for new democracies in poorcountries where capacity of any kind is limited.

It is certainly true that the state has been a dom-inant economic and political force in these smallercountries, sometimes crowding out the privatesector. But the solution is not so much less govern-ment but a different kind of government,embedded in rule-following institutions ratherthan personal rule. Similarly, there is little point inadopting Western ideas of new public manage-ment within the bureaucracy; rather, the purposeshould be to build up an orthodox civil servicethat applies rules consistently and economically.Developing the market also requires enhancing thepublic infrastructure of transport, communicationand education.

In Africa, however, fragmentary attempts atcapacity-building have so far produced onlymeagre results. Governments often lack the abilityto implement their policies throughout the terri-tory. They lack numbers on the ground and mustoften rely on traditional local leaders who aretherefore able to veto the implementation of thereform agenda. A state that barely exists cannot beexpected to engage in serious and effective policy-making and implementation. It is of course true,as Chazan et al. note (1999, p. 344), that theunderlying problems of the African continentextend well beyond politics; the difficultiesinclude poor soil, an arid climate, tropical diseases,Aids and a labour surplus. As a result, even in aworld that is more democratic than ever before,millions of people continue to live lives of extremepoverty and degradation. If solutions are to behad, the comparative study of politics should helpto find them.

Public policy in authoritarian states

Studying public policy in non-democratic regimesconfirms the importance of distinguishingbetween different types of authoritarian govern-ment. At one extreme, many communist statesattempted a type of planning virtually withoutprecedent in history; every communist state for-mulated clearer national goals and targets than anydemocracy. The result was an often decisive andgenerally ruthless commitment to a single goal,notably industrialization. At the other extreme,many military and personal rulers show immenseconcern about their own prosperity but none at allfor their country’s, leading to a policy shortage.

If there is a general theme to the policy process innon-democracies (including even communiststates), it is this subservience of policy to politics. Inthe more competitive world in which authoritarianrulers now find themselves, such weaknesses areincreasingly exposed, adding to pressures for demo-cratic reform. In this section, we will review thecommunist experience before turning to the policyprocess in contemporary authoritarian regimes.

Communist states differed enormously fromestablished democracies in their policy goals andmeans. For one thing, the ruling party articulated

322 GOVERNMENT AND POLICY

clear objectives for society; for another, the statesought to implement these goals though detailedplanning. Yet planning eventually yielded eco-nomic stagnation and thus contributed to the col-lapse of communism in Eastern Europe and theSoviet Union. But why did planning fail? Toanswer these questions, we must first look at howplanning worked.

The Soviet Union is the clearest examplebecause it ran the most planned economy onearth. The Soviet Union was the land of The Plan.Gosplan, the State Planning Committee, drew upannual five-year plans which were given the statusof law once they had received political approval.Implementation was the responsibility of min-istries which controlled individual enterprisesthrough a complex administrative network.Detailed planning was forced by a commandeconomy. A factory could not buy its componentson the market when there was no market. Insteadarrangements had to be made for another factoryto manufacture the parts and deliver them on time– and that factory in turn had to be supplied withraw materials.

The flaw became obvious. For anything to goright everything had to go right – so inevitablysomething went wrong! Because the right compo-nents did not arrive at the right time, all sorts ofinformal and often illegal deals had to be fixed upto ensure that the arbitrary production quota wasmet. Further, the system was dominated by plan-ners and producers, rather than customers. Targetswere based on quantity, not quality. As a result,goods were shoddy when they were produced atall. Local managers were not empowered to useinitiative, even though they were well-aware ofwhat needed to be done.

The endemic weaknesses of an over-centralizedeconomy led to a black market. Shortages encour-aged corrupt swaps among individuals with accessto resources: good cuts of meat in exchange forcigarettes, train tickets for university places. Thosewith little to offer, notably old people, just waitedin line – for an average of three hours a day in theSoviet Union.

So was the planned economy an unmitigateddisaster? Not completely. Notably in the USSR, itdid prove successful at building the foundations ofindustrial development, albeit at a horrifying

human price. Heavy industry was the great successof the planned economy, both in communist stateswhich were undergoing industrialization for thefirst time and in those which were rebuilding after1945.

This success derived from the philosophy of thebig push. The resources needed to meet the goalwould be allocated to the priority area, whateverthe impact on other sectors. The objective of thebig push took precedence and other consequences(including a degraded environment in the case ofindustrialization) were ignored. Objectives deter-mined budgets rather than vice versa.

The big push was a deliberately blinkeredapproach which ignored the overall view but oftensucceeded in achieving its specific targets. Forinstance, Stalin’s drive to industrialize Russia trans-formed a society of peasants into a world indus-trial power within a generation. Later, China andCuba applied the big push to their own societies,again with spectacular results. The targets wereoften economic, such as steel production, butsometimes extended to social objectives such asadequate housing and health care.

Where, then, does the collapse of state planningleave the remaining ‘communist’ states, notablyChina? The PRC’s rulers have certainly reducedthe importance of central planning yet they canhardly be said to have created the conditions for amarket economy. Massive, and massively ineffi-cient, state-owned enterprises still pervade theeconomy, soaking up labour and serving as anindirect welfare state. The expansion of the non-state sector has stimulated continued growth butparty contacts still determine access to economicopportunities. To most eyes, the system is inher-ently corrupt; it is certainly technically inefficientin that political criteria distort economic deci-sions, leading to a huge misallocation of capital.Yet this idiosyncratic model is still deliveringgrowth in what remains a poor country. As long aseconomic growth continues, China’s nominallycommunist rulers may succeed in resolving thepolitical tensions induced by corruption andincreasing inequality. Yet judging by the experi-ence of communism in Eastern Europe and theSoviet Union, that same growth will eventuallydeliver a demand for more fundamental politicalreform.

THE POLICY PROCESS 323

Some non-communist authoritarian regimeshave also led economic development, notably inEast Asian states such as Malaysia and Singapore(see Japan profile, p. 305). But policy stagnation isa more familiar pattern under authoritarian rule.Policy inertia often results from a lack of motiva-tion among the ruling elite. Often, the key task fornon-elected rulers is to play off domestic politicalforces against each other so as to ensure the ruler’sown continuation in office, an art developed to itshighest level by the cautious ruling families of theMiddle East. Or the ruler may want to enrichhimself, his family and his ethnic group, a taskhardly conducive to orderly policy development.More commonly, patronage is the main politicalcurrency; the age-old game of building up andpaying down political debts again works againstcoherent policy, transparently applied. As Chazanet al. (1999, p. 171) note in their discussion ofAfrica,

patriarchal rule has tended to be conservative: itpropped up the existing order and did little topromote change. It required the exertion of agreat deal of energy just to maintain control.

Finally, rulers may simply lack the ability tomake coherent policy. This weakness wascommon enough among military governments.The generals may seize power in an honest

attempt to eliminate corruption and improvepolicy-making but they soon discover that gover-nance is more complicated than they had imag-ined. Eventually, they slink back to their lair, withlittle achieved.

Key reading

Next step: Stone (2001) introduces policyanalysis from a political perspective.

John (1998) and Parsons (1995) are alternatives toStone while Fischer (2003) adopts a critical butaccessible approach. Colebatch (1998) discussesthe policy concept while Sabatier (1999) reviewstheories of the policy process. On implementa-tion, the classic American study by Pressman andWildavsky (1973) can be supplemented by Hilland Hupe (2002). Rose (2004) is a comparativestudy of lesson-drawing in public policy. Pierson(1998, 2000) and Esping-Andersen (2002) areexcellent sources on the welfare state, whileMajone (1996) is outstanding on regulation.Elster et al. (1998) cover institutional design inpost-communist societies, while Lijphart andWaisman (1996b) draw comparisons with LatinAmerica. On privatization, see Wright (1994) forWestern Europe and Lieberman et al. (1997) forEastern Europe.

324 GOVERNMENT AND POLICY

P.1 New sections and subsections in this edition xP.2 Key features of the website

http://www.palgrave.com/politics/hague xi1.1 Nations and states 101.2 The language of power 121.3 Weber’s classification of authority 142.1 Declaration of the Rights and Duties of Man

and the Citizen (1789), Articles 1–6 202.2 Aspects of the growth of the Western state,

1789–1975 212.3 States from empires: waves of decolonization 222.4 Some key acronyms in the global order 272.5 The impact of IG0s on national politics:

winners and losers 272.6 Intergovernmental organizations and the world

economy 333.1 Forms of democracy 353.2 Aristotle’s characterization of democracy 363.3 Huntington’s three waves of democratization 404.1 Forms of authoritarian rule 525.1 The advantages of comparison 695.2 The difficulties of comparison 715.3 Levels of analysis in comparative politics 755.4 Major techniques in comparative politics 805.5 Some types of case study 815.6 Dependent, intervening and independent

variables 846.1 The clash of civilizations? 1037.1 Contemporary trends in mass communication 1087.2 Some tests used by j ournalists, to determine

newsworthiness 1137.3 Media effects in reinforcement, agenda-setting

and framing 1147.4 Measuring public opinion 1187.5 Limiting the damage: China and the intemet 1218.1 Patterns of political participation, by type of

regime 1228.2 Comparing social movements, political parties

and interest groups 1298.3 A century of revolutions 1379.1 Electoral systems: legislatures 1489.2 Methods for electing presidents: some examples 1579.3 Decline in party identification, 1970s–1990s 1589.4 Features of the electoral system and of

individuals which increase turnout 1609.5 The referendum, initiative and recall 16210.1 Waves of interest group formation in the

United States 16710.2 Protective and promotional groups 16810.3 Democracies grouped by the extent of

corporatism 18010.4 Examples of social organizations in contemporary

China 18411.1 Typical elements of a major political party in

Western Europe 186

11.2 Types of party organization 18711.3 Selecting candidates for legislative elections 18911.4 Selection of party leaders in established

democracies 19011.5 Major cleavages in the development of Western

European societies and their impact on parties 19411.6 Party systems in democracies 19511.7 Party families in post-communist Europe 20112.1 The standard format of constitutions 21012.2 Entrenching the constitution: some examples 21112.3 Institutions for judicial review: supreme courts

v. constitutional courts 21512.4 Methods of selecting judges 22012.5 Methods for judicial regulation of the

bureaucracy 22113.1 The allocation of functions in the Canadian

and German federations 22913.2 Financial transfers from the federal government

to states 23413.3 Methods for distributing power away from the

centre 23613.4 Exploring the status of local government 24013.5 Structures for local govermnent 24213.6 Typical powers of local authorities 24214.1 Types of parliamentary committee 25114.2 Functions of legislatures 25314.3 Resolving differences in the versions of a bill

passed by each chamber 25714.4 Stages of recruitment to the legislature 26014.5 A policy classification of assemblies 26615.1 Separate elections in the United States 27115.2 Comparing presidential powers in Brazil and

the USA 27215.3 Procedure for installing a government in

parliamentary systems where no party possesses a majority of seats 276

15.4 Selecting the head of state in some parliamentary democracies 281

15.5 Presidential powers in some new Eastern European democracies 285

16.1 The organization of government: departments, divisions and agencies 294

16.2 Types of non-departmental public body 29716.3 Forms of bureaucratic accountability 29916.4 Steer, don’t row! Osborne and Gaebler’s ten

principles for improving th effectiveness of government agencies 302

16.5 Components of new public management 30217.1 Rational and incremental models of

policy-making 31117.2 The changing agenda of the Western state 31517.3 Building market institutions in post-communist

societies 320

Appendix: list of boxes

325

■AAberbach, J. et al. (1981) Bureaucrats and Politicians (Cambridge,

MA: Harvard University Press).Adams, F. (2003) Deepening Democracy: Global Governance and

Political Reform in Latin America (New York: Praeger). Addison, J. and Schnabel, C. (eds) (2003) International

Handbook of Trade Unions (Cheltenham and Northampton,MA: Edward Elgar).

Ågh, A. (1996) ‘Democratic Parliamentarism in Hungary: TheFirst Parliament (1990–94) and the Entry of the SecondParliament’, in The New Parliaments of Central and EasternEurope, ed. D. Olson and P. Norton (London: Frank Cass) pp. 16–39.

Ågh, A. (1998) The Politics of Central Europe (London andThousand Oaks, CA: Sage).

Agüero, F. and Stark, J. (eds) (1998) Fault Lines of Democracy inPost-Transition Latin America (Coral Gables, FL: University ofMiami).

Almond, G. and Bingham Powell, G. (2000) Comparative PoliticsToday: A World View, 7th edn (New York: HarperCollins).

Almond, G. and Verba, S. (1963) The Civic Culture (Princeton,NJ: Princeton University Press).

Almond, G. and Verba, S. (eds) (1980) The Civic CultureRevisited (Princeton, NJ: Princeton University Press).

Altunisik, M. and Kavli, Ö. (2003) Turkey (London and NewYork: Routledge).

Ames, B. (2002) ‘Party Discipline in the Chamber of Deputies’,in Legislative Politics in Latin America, ed. S. Morgenstern andB. Nacif (Cambridge and New York: Cambridge UniversityPress) pp. 185–221.

Anderson, B. (1991) Imagined Communities: Reflections on theOrigins and Spread of Nationalism, revised edn (London andNew York: Verso).

Anderson, B. (1998) The Spectre of Comparisons: Nationalism,South-East Asia and the World (London and New York: Verso).

Anderson, C. and Ward, D. (1996) ‘Barometer Elections inComparative Perspective’, European Journal of Political Research(15) 447–60.

Anderson, M. (1997) Frontiers: Territory and State Formation inthe Modern World (Cambridge: Polity).

Anderson, P. (1974) Lineages of the Absolutist State (London:NLB).

Andeweg, R. and Irwin, G. (1993) Dutch Government and Politics(Basingstoke: Macmillan).

Andeweg, R. and Irwin, G. (2002) Governance and Politics of theNetherlands (Basingstoke and New York: Palgrave).

Andrews, J. and Momtinola, G. (2004) ‘Veto Players and theRule of Law in Emerging Democracies’, Comparative PoliticalStudies (37) 55–87.

Arblaster, A. (2002) Democracy, 3rd edn (Buckingham andPhiladelphia, PA: Open University Press).

Arel, D. (2001) ‘Political Stability in Multinational Democracies:Comparing Language Dynamics in Brussels, Montreal andBarcelona’, in Multinational Democracies, ed. A.-G. Gagnon

and J. Tully (Cambridge and New York: Cambridge UniversityPress) pp. 65–89.

Arendt, H. [1951] (1966 edn) The Origins of Totalitarianism(New York: Harcourt Brace).

Arendt, H. (1958) The Human Condition (Chicago, IL:University of Chicago Press).

Arendt, H. (1970) On Violence (London: Penguin).Argun B. (2003) Turkey in Germany: The Transitional Sphere of

Deutschkei (London and New York: Routledge). Aristotle (1962 edn) The Politics, trans. T. Sinclair

(Harmondsworth and Baltimore, MD: Penguin).Arter, D. (1999) Scandinavian Politics Today (Manchester:

Manchester University Press, and New York: St Martin’s Press).Asher, H. (1988b) Polling and the Public: What Every Citizen

Should Know (Washington, DC: Congressional Quarterly, Inc.).

■BBaaklini, A., Denoeux, G. and Springborg, R. (1999) Legislative

Politics in the Arab World: The Resurgence of DemocraticInstitutions (Boulder, CO and London: Lynne Rienner).

Baer, M. (1993) ‘Mexico’s Second Revolution: Pathways toLiberalization’, in Political and Economic Liberalization inMexico: At a Critical Juncture?, ed. R. Roett (Boulder, CO andLondon: Lynne Rienner) pp. 51–68.

Bagehot, W. [1867] (1963 edn) The English Constitution(London: Fontana).

Baka, A. (1998) ‘The Framework for Public Management inHungary’, in Innovations in Public Management: Perspectivesfrom East and West Europe, ed. T. Verheijen and D. Coombes(Cheltenham and Northampton, MA: Edward Elgar) pp.128–42.

Baker, B. (2002) Taking the Law into Their Own Hands: LawlessLaw Enforcers in Africa (Aldershot and Burlington, VT:Ashgate).

Bakvis, H. and Skogstad, G. (eds) (2002) Canadian Federalism:Performance, Effectiveness and Legitimacy (Don Mills, Ontarioand Oxford: Oxford University Press).

Balme, R. (1998) ‘The French Region as a Space for PublicPolicy’, in Regions in Europe, ed. P. Le Galès and C. Lequesne(London and New York: Routledge) pp. 181–98.

Barber, J. (1992) The Pulse of Politics: Electing Presidents in theMedia Age (New Brunswick, NY and London: TransactionBooks).

Bardach, E. (1976) ‘Policy Termination as a Political Process’,Policy Sciences (7) 123–31.

Barry, B. (1988) Sociologists, Economists, and Democracy (Chicago,IL: University of Chicago Press).

Bartolini, S. and Mair, P. (1990) Identity, Competition and ElectoralAvailability: The Stabilization of European Electorates 1885–1985(Cambridge and New York: Cambridge University Press).

Batt, J. (2003) ‘Defining Central and Eastern Europe’ inDevelopments in Central and East European Politics 3, ed. S.White S., J. Batt and P. Lewis (Basingstoke: PalgraveMacmillan) pp. 3–22.

References

To search these references by chapter, go to http://www.palgrave.com/politics/hague

——————

326

Bauböck, R. (1994) Transnational Citizenship (Cheltenham andNorthampton, MA: Edward Elgar).

Bayart, J.-F. (1993) The State in Africa: The Politics of the Belly(Harlow and New York: Longman).

Bayley, S. (1998) Labour Camp: The Failure of Style overSubstance (London: Batsford).

Baylis, T. (1996) ‘Presidents versus Prime Ministers: ShapingExecutive Authority in Eastern Europe’, World Politics (48)297–322.

Beattie, A. (2003) ‘Decoding the Fed: The Art of ReadingBetween the Lines’, Financial Times, 27 June.

Bednar, J., Eskridge, W. and Ferejohn, J. (2001) ‘A Political Theoryof Federalism’, in Constitutional Culture and Democratic Rule,ed. J. Ferejohn, J. Rakove and J. Riley (Cambridge and NewYork: Cambridge University Press) pp. 223–70.

Bekke, H., Perry, J. and Toonen, T. (1996) Civil Service Systems inComparative Perspective (Bloomington, IN: Indiana UniversityPress).

Bell, D. et al. (eds) (1995) Towards Illiberal Democracy in PacificAsia (London: Macmillan).

Bell, J. (1991) ‘Administrative Law’, in The BlackwellEncyclopaedia of Political Science, ed. V. Bogdanor (Oxford andCambridge, MA: Blackwell) pp. 9–12.

Bendor, J., Moe, T. and Shotts, K. (2001) ‘Recycling the GarbageCan: An Assessment of the Research Program’, AmericanPolitical Science Review (95) 169–90.

Bennett, W. Lance and Entman, R. (eds) (2001) MediatedPolitics: Communication in the Future of Democracy(Cambridge and New York: Cambridge University Press).

Bentley, A. (1908) The Process of Government (Chicago, IL:University of Chicago Press).

Bentley, T., Jupp, B. and Stedman Jones, D. (2000) Getting toGrips with Depoliticization (London: Demos) athttp://www.demos.co.uk.

Berger, S. (2002) ‘Social Partnership 1880–1989: The DeepHistorical Roots of Diverse Strategies’ in Policy Concertationand Social Partnership in Western Europe: Lessons for the 21stCentury, ed. S. Berger and H. Compston (New York andOxford: Berghahn) pp. 335–52.

Berger, S. and Compston, H. (eds) (2002) Policy Concertationand Social Partnership in Western Europe: Lessons for the 21stCentury (New York and Oxford: Berghahn).

Berglund, S. (1990) ‘Finland in a Comparative Perspective’, inFinnish Democracy, ed. J. Sunberg and S. Berglund (Jyväskylä:Finnish Political Science Association) pp. 11–25.

Bergman, T. (2000) ‘Sweden: When Minority Cabinets Are theRule and Majority Coalitions the Exception’, in CoalitionGovernments in Western Europe, ed. W. Müller and K. Strøm (Oxford and New York: Oxford University Press) pp.192–230.

Berrington, H. (1995) ‘Political Ethics: The Nolan Report’,Government and Opposition (30) 429–51.

Best, H. and Cotta, M. (2000) Parliamentary Representatives inEurope, 1848–2000 (Oxford and New York: OxfordUniversity Press).

Bevir, M. and Rhodes, R. (2002) ‘Interpretive Theory’, in Theoryand Methods in Political Science, 2nd edn, ed. D. Marsh andG. Stoker (Basingstoke and New York: Palgrave Macmillan)pp. 131–52.

Bill, J. and Springborg, R. (2000) Politics in the Middle East, 4thedn (New York: Longman).

Birch, A. (1972) Representation (London: Pall Mall).Blais, A. and Massicote, L. (2002) ‘Electoral Systems’, in

Comparing Democracies 2: New Challenges in the Study of

Elections and Voting, ed. L. LeDuc, R. Niemi and P. Norris(Thousand Oaks, CA and London: Sage) pp. 40–69.

Blais, A., Massicotte, L. and Dobrzynska, A. (1997) ‘DirectPresidential Elections: A World Summary’, Electoral Studies(16) 441–55.

Block, F. (1977) ‘The Ruling Class Does Not Rule’, SocialistRevolution (3) 6–28.

Blom-Hansen, J. (2000) ‘Still Corporatism in Scandinavia: ASurvey of Recent Empirical Findings’, Scandinavian PoliticalStudies (23) 157–81.

Blondel, J. (1973) Comparative Legislatures (Englewood Cliffs,NJ: Prentice-Hall).

Blondel, J. (1993) ‘Consensus Politics and Multiparty Systems’,paper presented at the Conference on ConsensualPolicymaking and Multiparty Politics, Australian NationalUniversity, 1993.

Boardman, A. et al. (2000) Cost–Benefit Analysis: Concepts andPractice, 2nd edn (Englewood Cliffs, NJ: Prentice-Hall).

Bodin, J. [1576] (1962 edn) The Six Books of a Commonweal, ed.K. McRae (Cambridge, MA: Harvard University Press).

Bogason, P. (1996) ‘The Fragmentation of Local Government inScandinavia’, European Journal of Political Research (30) 65–86.

Bogdanor, V. (ed.) (1991) The Blackwell Encyclopaedia of PoliticalScience (Oxford and Cambridge, MA: Blackwell).

Boli, J. and Thomas, G. (eds) (1999) Constructing World Culture:International Non-Governmental Organizations since 1875(Stanford, CA: Stanford University Press).

Bork, R. (2003) Coercing Virtue: The Worldwide Rule of Judges(Washington, DC: American Enterprise Institute).

Borón, A. (1998) ‘Faulty Democracies? A Reflection on theCapitalist “Fault Lines”’, in Fault Lines of Democracy in Post-Transition Latin America, ed. F. Agüero and J. Stark (CoralGables, FL: University of Miami) pp. 41–66.

Boston, J. (1995a) ‘Inherently Governmental Functions and theLimits to Contracting Out’, in The State Under Contract, ed. J.Boston (Wellington: Bridget Williams) pp. 78–111.

Boston, J. (ed.) (1995b) The State Under Contract (Wellington:Bridget Williams).

Boston, J. et al. (eds) (1995) Reshaping the State: New Zealand’sBureaucratic Revolution (Oxford and New York: OxfordUniversity Press).

Boulding, K. (1989) Three Faces of Power (Thousand Oaks, CAand London: Sage).

Bourgault, L. (1995) Mass Media in Sub-Saharan Africa(Bloomington, IN: Indiana University Press).

Bourne, A. (2003) ‘Regional Europe’, in European Union Politics,ed. M. Cini (Oxford and New York: Oxford University Press)pp. 278–93.

Bowler, S. and Donovan, T. (2002) ‘Democracy, Institutions andAttitudes about Citizen Influence on Government’, BritishJournal of Political Science (32) 371–90.

Bowler, S. and Farrell, D. (eds) (1992) Electoral Strategies andPolitical Marketing (London: Macmillan).

Bowler, S., Donovan, T. and Tolbert, C. (eds) (1998) Citizens asLegislators: Direct Democracy in the United States (Columbus,OH: Ohio State University Press).

Bowles, N. (1998) Government and Politics of the United States,2nd edn (Basingstoke: Macmillan).

Boyne, R. (2003) Risk (Milton Keynes: Open University Press). Bratton, M. (1997) ‘Deciphering Africa’s Divergent Transitions’,

Political Science Quarterly (112) 67–93.Bratton, M. (1998) ‘Second Elections in Africa’, Journal of

Democracy (9) 51–66.Bratton, M. and van de Walle, N. (1997) Democratic Experiments in

REFERENCES 327

Africa: Regime Transitions in Comparative Perspective (Cambridgeand New York: Cambridge University Press).

Braun, D. (ed.) (2000) Public Policy and Federalism (Aldershotand Burlington, VT: Ashgate).

Braun, D. et al. (2003) Fiscal Policies in Federal States (Aldershotand Burlington, VT: Ashgate).

Breslin, S. et al. (eds) (2002) New Regionlaism in the Global PoliticalEconomy: Theories and Cases (London and NY: Routledge).

Breuilly, J. (1993) Nationalism and the State, 2nd edn (New York:St Martin’s Press).

Brewer, P., Aday, S. and Gross, K. (2003) ‘Rallies All Round: TheDynamics of System Support’, in Framing Terrorism: The NewsMedia, the Government and the Public, ed. P. Norris, M. Kernand M. Just (London and NY: Routledge) pp. 229–54.

Bromley, C. and Curtice, J. (2002) ‘Where Have All the VotersGone?’, in British Social Attitudes: the 19th Report, ed. A. Parket al. (London and Thousand Oaks, CA: Sage) pp. 118–32.

Brooker, P. (1995) Twentieth-Century Dictatorships: TheIdeological One-Party States (Basingstoke: Macmillan).

Brooker, P. (2000) Non-Democratic Regimes: Theory, Governmentand Politics (Basingstoke: Macmillan, and New York: StMartin’s Press).

Brooks, S. (1996) Canadian Democracy: An Introduction, 2nd edn(Toronto and Oxford: Oxford University Press).

Broughton, D. (1995) Public Opinion Polling and Politics inBritain (Hemel Hempstead: Harvester Wheatsheaf ).

Broughton, D. (2002) ‘Participation and Voting’, inDevelopments in West European Politics 2, ed. P. Heywood, E.Jones and M. Rhodes (Basingstoke and New York: Palgrave)pp. 94–114.

Brown, A. (ed.) (2001) Contemporary Russian Politics: A Reader(Oxford and New York: Oxford University Press).

Brown, L. (1984) International Politics and the Middle East: OldRules, Dangerous Games (Princeton, NJ and London: PrincetonUniversity Press).

Bryce, J. (1921) Modern Democracies, Vol. 2 (New York:Macmillan).

Brzezinski, Z. (1997) ‘The New Challenge to Human Rights’,Journal of Democracy (8) 3–8.

Brzezinski Z. (2002) ‘Confronting Anti-American Grievances’,New York Times, 1 September.

Buckley, M. and Fawn, R. (eds) (2003) Global Responses toTerrorism: 9/11, Afghanistan and Beyond (London and NewYork: Routledge).

Budge, I. (1996) The New Challenge of Direct Democracy(Cambridge: Polity).

Bull, H. (1977) The Anarchical Society: A Study of Order in WorldPolitics (Basingstoke: Macmillan).

Burgess, M. (2000) Federalism and European Union: The Buildingof Europe, 1950–2000 (London and New York: Routledge).

Burgess, M. (2001) ‘Competing National Visions: Canada–Quebec Relations in a Comparative Perspective’, in Multi-national Democracies, ed. A.-G. Gagnon and J. Tully(Cambridge and New York: Cambridge University Press) pp.257–74.

Burgess, M. (2003) ‘Federalism and Federation’, in EuropeanUnion Politics, ed. M. Cini (Oxford and New York: OxfordUniversity Press) pp. 65–79.

Burgess, M. and Gagnon, A.-G. (eds) (1993) ComparativeFederalism and Federation: Competing Traditions and FutureDirections (Hemel Hempstead: Harvester Wheatsheaf ).

Burke, E. [1774] (1975 edn) ‘Speech to the Electors of Bristol’ inEdmund Burke on Government, Politics and Society, ed. B. Hill(London: Fontana) pp. 156–8.

Burnham, W. (1982) ‘The Constitution, Capitalism and theNeed for Rationalized Regulation’, in How Capitalistic Is theConstitution? ed. R. Goldwyn and W. Schambra (Washington,DC: American Enterprise Institute) pp. 72–87.

Burns, N. et al. (2001) The Private Roots of Public Action: Gender,Equality and Political Participation (Cambridge, MA: HarvardUniversity Press).

Butler, A. (2004) Contemporary South Africa (Basingstoke andNew York: Palgrave).

Butler, D. (1989) British General Elections Since 1945 (Oxfordand Cambridge, MA: Blackwell). [new ed?]

Butler, D. and Kavanagh D. (2002) The British General Electionof 2001 (Basingstoke and New York: Palgrave).

■CCamp, R. (2002) Politics in Mexico: The Democratic

Transformation, 4th edn (Oxford and New York: OxfordUniversity Press).

Campbell, A., Converse, P., Miller, A. et al. (1960) The AmericanVoter (New York: Wiley).

Campbell, C. and Wilson, G. (1995) The End of Whitehall: Deathof a Paradigm? (Oxford and Cambridge, MA: Blackwell).

Canel, E. (1992) ‘Democratization and the Decline of UrbanSocial Movements in Uruguay: A Political InstitutionalAccount’, in The Making of Social Movements in Latin America:Identity, Strategy and Democracy, ed. A. Escobar and S. Alvarez(Boulder, CO and Oxford: Westview) pp. 276–90.

Caplan, J. (1988) Government Without Administration: State andCivil Service in Weimar and Nazi Germany (Oxford: ClarendonPress).

Carey, J. (1998) Term Limits and Legislative Representation(Cambridge and New York: Cambridge University Press).

Carothers, T. (2002) ‘The End of the Transition Paradigm’,Journal of Democracy (13) 5–21.

Carty, H. (2002) ‘Canada’s Nineteenth Century Parties at theMillennium’, in Political Parties in Advanced IndustrialDemocracies, ed. P. Webb, D. Farrell and I. Holliday (Oxfordand New York: Oxford University Press) pp. 345–78.

Case, W. (1996) ‘Can the “Halfway House” Stand?Semidemocracy and Elite Theory in Three Southeast AsianCountries’, Comparative Politics (28) 437–64.

Centeno, M. and Silva, R (1998) ‘Introduction’, in The Politics ofExpertise in Latin America, ed. M. Centeno and P. Silva(Basingstoke: Macmillan, and New York: St Martin’s Press) pp.1–12.

Cerny, P. (1990) The Changing Architecture of Politics (ThousandOaks, CA and London: Sage).

Chalmers, D. (1990) Dilemmas of Latin AmericanDemocratization: Dealing with International Forces, Papers onLatin America, No. 18 (New York: Columbia UniversityInstitute of Latin American and Iberian Studies).

Chandler, J. (1993) Local Government in Liberal Democracies: AnIntroductory Survey (London and New York: Routledge).

Chandler, J. (ed.) (2000) Comparative Public Administration(London and New York: Routledge).

Chatterjee, D. and Scheid, D. (eds) (2003) Ethics and ForeignIntervention (Cambridge and New York: Cambridge UniversityPress).

Chazan, N. et al. (1999) Politics and Society in ContemporaryAfrica, 3rd edn (Boulder, CO: Lynne Rienner).

Chebabi, H. and Linz, J. (eds) (1998) Sultanistic Regimes(Baltimore, MD and London: Johns Hopkins UniversityPress).

328 REFERENCES

Cheibub, J. (2002) ‘Minority Governments, Deadlock Situationsand the Survival of Presidential Democracies’, ComparativePolitical Studies (35) 284–312.

Christensen, T. and Egeberg, M. (eds) (1997) Forvaltnings-kunnskap (Oslo: Aschehoug Tano).

Christensen, T. and Guy Peters, B. (1999) Structure, Culture andGovernance: A Comparison of Norway and the United States(Lanham, MD and Oxford: Rowman & Littlefeld).

Cigler, C. and Loomis, B. (eds) (2002) Interest Group Politics, 6thedn (Washington: Congressional Quarterly Press).

Clark S. (1995) State and Status: The Rise of the State andAristocratic Power in Western Europe (Montreal and Kingston:McGill-Queen’s University Press).

Close, D. (ed.) (1995) Legislatures and the New Democracies inLatin America (Boulder, CO and London: Lynne Rienner).

Clover, C. (2003) ‘Iraqi Jigsaw Confronts Makers ofConstitution’, Financial Times, 29 September, p. 8.

Cohen, M., March, J. and Olsen, J. (1972) ‘A Garbage CanModel of Organizational Choice’, Administrative ScienceQuarterly (17) 1–25.

Colebatch, H. (1998) Policy (Buckingham: Open UniversityPress).

Collier, D. (1991) ‘The Comparative Method: Two Decades ofChange’, in Comparative Dynamics: Global ResearchPerspectives, ed. D. Rustow and K. Erickson (New York:HarperCollins) pp. 7–31.

Conover, P. and Searing, D. (1994) ‘Democratic Citizenship andthe Study of Political Socialization’, in Developing Democracy:Essays in Honour of J. F. P. Blondel, ed. I. Budge and D. McKay(Thousand Oaks, CA and London: Sage) pp. 24–55.

Conradt, D. (2001) The German Polity, 7th edn (New York:Longman).

Constable, P. (2001) ‘Pakistan’s Predicament’, Journal ofDemocracy (12) 15–29.

Conway, M. (2001) ‘Political Participation in AmericanElections: Who Decides What?’, in America’s Choice 2000, ed.W. Crotty (Boulder, CO: Westview) pp. 79–94.

Cornelius, W. and Weldon, J. (2004) ‘Politics in Mexico’, inComparative Politics Today: A World View, 8th edn, ed. G.Almond et al. (New York and London: Pearson Longman) pp.466–519.

Cotta, M. (2000) ‘Defining Party and Government’, in TheNature of Party Government: A Comparative EuropeanPerspective, ed. J. Blondel and M. Cotta (Basingstoke:Macmillan) pp. 56–95.

Cottey, A. et al. (eds) (2003) Democratic Control of the Military inPost-Communist Europe (Basingstoke: Palgrave).

Cox, G. and Morgenstern, S. (2002) ‘Latin America’s Reactive Assemblies and Proactive Presidents’, in LegislativePolitics in Latin America, ed. S. Morgenstern and B. Nacif(Cambridge and New York: Cambridge University Press) pp.446–68.

Craske, M. (2001) ‘Another Mexican Earthquake: An Assessmentof the 2 July 2000 Elections’, Government and Opposition (36)27–47.

Crawford, K. (1996) East Central European Politics Today(Manchester: Manchester University Press and New York: StMartin’s Press).

Crawford, N. (2003) Argument and Change in World Politics:Ethics, Decolonization and Humanitarian Intervention(Cambridge and New York: Cambridge University Press).

Crick, B. (2000) In Defence of Politics, 5th edn (Harmondsworth:Penguin).

Crockett, D. [1834] (1987 edn) A Narrative of the Life of David

Crockett of the State of Tennessee, intro. J. Shackford and S.Folmsbee (Knoxville, TN: University of Tennessee Press).

Cronin, T. (1989) Direct Democracy: The Politics of Initiative,Referendum and Recall (Cambridge, MA and London: HarvardUniversity Press).

Crothers L. and Lockhart C. (eds) (2000) Culture and Politics: AReader (Basingstoke: Palgrave Macmillan).

Crouch, H. (1996) Government and Society in Malaysia (Cornell,NY: Cornell University Press).

Curran, J. and Seaton, J. (2003) Power Without Responsibility: ThePress and Broadcasting in Britain, 6th edn (London and NewYork: Methuen).

Cutler, L. (1980) ‘To Form a Government’, Foreign Affairs (59)126–43.

■DDahl, R. (1957) ‘The Concept of Power’, Behavioral Science (2)

201–15.Dahl, R. (1961) Who Governs? Democracy and Power in an

American City (New Haven, CT and London: Yale UniversityPress).

Dahl, R. (1989) Democracy and its Critics (New Haven, CT andLondon: Yale University Press).

Dahl, R. (1993) ‘Pluralism’, in The Oxford Companion to Politicsof the World, ed. J. Krieger (New York and Oxford: OxfordUniversity Press) pp. 704–7.

Dahl, R. (1998) On Democracy (New Haven, CT and London:Yale University Press).

Dahl, R. (1999) ‘Can International Organizations BeDemocratic?’, in Democracy’s Edges, ed. I. Shapiro and C.Hacker-Cordon (New Haven, CT and London: YaleUniversity Press) pp. 19–36.

Dahl, R. (2000) ‘A Democratic Paradox?’, Scandinavian PoliticalStudies (23) 246–51.

Dahl, R. (2001) How Democratic is the American Constitution?(New Haven, CT and London: Yale University Press).

Dahl, R., Shapiro, I. and Cheibub, J. (eds) (2003) The DemocracySourcebook (Cambridge, MA: MIT Press).

Dalton, R. (1994) The Green Rainbow: Environmental Groups inWestern Europe (New Haven, CT and London: Yale UniversityPress).

Dalton, R. (2002) ‘Political Cleavages, Issues and ElectoralChange’, in Comparing Democracies 2: New Challenges in theStudy of Elections and Voting, ed. L. LeDuc, R. Niemi and P.Norris (London and Thousand Oaks, CA: Sage) pp. 189–209.

Dalton, R. and Wattenberg, M. (2000) Politics Without Partisans:Political Change in Advanced Industrial Democracies (Oxfordand New York: Oxford University Press).

Dalton, R., McAllister, I. and Wattenberg, M. (2000) ‘TheConsequences of Partisan Dealignment’, in Parties WithoutPartisans, ed. R. Dalton and M. Wattenberg (Oxford and NewYork: Oxford University Press) pp. 37–63.

Davies, J. (1962) ‘Toward a Theory of Revolution’, AmericanSociological Review (27) 5–18.

Davies, N. (1996) Europe: A History (Oxford and New York:Oxford University Press).

Davis, J. (1998) Leadership Selection in Six Western Democracies(Westport, CT: Greenwood).

Day, S. and Shaw, J. (2002) ‘European Union Electoral Rightsand the Political Participation of Migrants in Host Polities’,International Journal of Population Geography (8) 183–99.

de Burea, G. and Wieler, J. (eds) (2002) The European Court ofJustice (Oxford and New York: Oxford University Press).

REFERENCES 329

de Swaan, A. (1973) Coalition Theories and Cabinet Formation(Amsterdam: Elsevier).

de Tocqueville, A. [1835] (1966 edn) Democracy in America (NewYork: Vintage Books). Also available on-line at the AmericanStudies website at the University of Virginia: http://xroads.virginia.edu.

de Tocqueville, A. [1856] (1954 edn) The Ancien Regime and theRevolution in France (London: Fontana).

Deegan, H. (2001) The Politics of the New South Africa: Apartheidand After (Harlow and New York: Longman).

Dehousse, R. (1998) The European Court of Justice (London:Macmillan).

della Porta, D., Kriesi, H. and Rucht, D. (eds) (1999) SocialMovements in a Globalizing World (London: Macmillan andNew York: St Martin’s Press).

Denemark, D. (2002) ‘Choosing MMP in New Zealand:Explaining the 1993 Electoral Reform’, in Mixed-MemberElectoral Systems: The Best of Both Worlds, ed. M. Shugart andM. Wattenberg (Oxford and New York: Oxford UniversityPress) pp. 70–95.

Derbyshire, J. and Derbyshire, I. (1999a) Encyclopedia of WorldPolitical Systems (London and Armonk, NY: M.E. Sharpe).

Derbyshire, J. and Derbyshire, I. (1999b) Political Systems of theWorld (Oxford: Helicon).

Diamond, L. (1992) ‘Economic Development and DemocracyReconsidered’, in Reexamining Democracy: Essays in Honor ofSeymour Martin Lipset, ed. G. Marks and L. Diamond(Thousand Oaks, CA and London: Sage) pp. 93–131.

Diamond, L. (1997) ‘Democracy in the Americas’, Annals of theAmerican Academy of Political and Social Sciences (550) 12–41.

Diamond, L. (1999) Developing Democracy: Toward Consolidation(Baltimore, MD and London: Johns Hopkins UniversityPress).

Diamond, L. and Lipset, S. (1995) ‘Legitimacy’, in TheEncyclopaedia of Democracy, ed. S. Lipset (London and NewYork: Routledge) pp. 747–51.

Diamond, L. and Plattner, M. (eds) (1996) The Global Resurgenceof Democracy, 2nd edn (Baltimore, MD and London: JohnHopkins University Press).

Diamond, L. and Plattner, M. (eds) (2001) The GlobalDivergence of Democracies (Baltimore, MD: Johns HopkinsUniversity Press).

Diamond, L. and Plattner, M. (eds) (2002) Democracy afterCommunism (Baltimore, MD and London: Johns HopkinsUniversity Press).

Diamond, L., Linz, J. and Lipset, S. (1995) ‘Introduction: WhatMakes For Democracy?’, in Politics in Developing Countries:Comparing Experiences with Democracy, ed. L. Diamond, J.Linz and S. Lipset (Boulder, CO and London: Lynne Rienner)pp. 1–33.

Diamond, L., Plattner, M. and Brumberg, D. (eds) (2003) Islamand Democracy in the Middle East (Baltimore, MD andLondon: Johns Hopkins University Press).

Dicey, A. [1885] (1959 edn) Introduction to the Study of the Lawof the Constitution, 10th edn (London: Macmillan).

Diehl, P. (2001) The Politics of Global Governance: InternationalOrganizations in an Interdependent World (Boulder, CO andLondon: Lynne Rienner).

Dittmer, L. and Gore, L. (2001) ‘China Builds a MarketCulture’, East Asia (19) 1–50.

Dodd, L. and Oppenheimer, B. (eds) (2001) CongressReconsidered, 7th edn (Washington, DC: CongressionalQuarterly Press).

Dogan, M. and Kazancigil, A. (eds) (1994) Comparing Nations:

Concepts, Strategies, Substance (Cambridge, MA and Oxford:Blackwell).

Dogan, M. and Pelassy, G. (1990) How to Compare Nations(Chatham, NJ: Chatham House).

Doherty, B. et al. (2003) ‘Explaining the Fuel Protests’, BritishJournal of Politics and International Relations (5) 1–23.

Domingo, P. and Sieder, R. (2001) ‘Conclusions’, in Rule of Lawin Latin America: The International Promotion of JudicialReform, ed. P. Domingo and R. Sieder (London: Institute ofLatin American Studies, University of London) pp. 142–64.

Domínguez, J. and Giraldo, J. (1996) ‘Parties, Institutions andMarket Reforms in Constructing Democracies’, inConstructing Democratic Governance: Mexico, Central Americaand the Caribbean in the 1990s, ed. J. Domínguez and A.Lowenthal (Baltimore, MD and London: Johns HopkinsUniversity Press) pp. 3–41.

Donovan, M. (1995) ‘The Politics of Electoral Reform in Italy’,International Political Science Review (16) 47–64.

Dooley, B. and Baron, S. (eds) (2001) The Politics of Informationin Early Modern Europe (London and New York: Longman).

Döring, H. (2001) ‘Parliamentary Agenda Control andLegislative Outcomes in Western Europe’, Legislative StudiesQuarterly (26) 145–65.

Döring, H. and Hallerberg, M. (eds) (2004) Patterns ofParliamentary Behaviour: Passage of Legislation Across WesternEurope (Aldershot and Burlington, VT: Ashgate).

Dorussen, H. and Taylor, M. (eds) (2002) Economic Voting(London and New York: Routledge).

Downs, A. (1957) An Economic Theory of Democracy (New York:Harper).

Downs, G., Roche, D. and Baroom, P. (1996) ‘Is the Good Newsabout Compliance Good News about Cooperation?’,International Organization (50) 379–406.

Drake, P. and Jaksic, I. (eds) (1989) The Struggle for Democracy inChile (Lincoln, NE and London: University of Nebraska Press).

Duchacek, I. (1973) Power Maps: The Comparative Politics ofConstitutions (Santa Barbara, CA: ABC Clio).

Duchacek, I. (1991) ‘Constitutions/Constitutionalism’, in TheBlackwell Encyclopaedia of Political Science, ed. V. Bogdanor(Oxford and Cambridge, MA: Blackwell) pp. 142–4.

Dunleavy, P. (2003) ‘Analysing Political Power’, in Developmentsin British Politics 7, ed. P. Dunleavy et al. (Basingstoke andNew York: Palgrave Macmillan) pp. 338–59.

Dunleavy, P. et al. (eds) (2003) Developments in British Politics 7(Basingstoke and New York: Palgrave Macmillan).

Duverger, M. [1954] (1970 edn) Political Parties (London:Methuen).

Duverger, M. (1980) ‘A New Political System Model: Semi-Presidential Government’, European Journal of PoliticalResearch (8) 165–87.

Dye, T. (2001) Understanding Public Policy, 10th edn(Englewood Cliffs, NJ: Prentice-Hall).

Dyson, K. (1980) The State Tradition in Western Europe: A Studyof an Idea and Institution (Oxford: Martin Robertson).

■EEaster, G. (1997) ‘Preference for Presidentialism: Postcommunist

Regime Change in Russia and the NIS’, World Politics (49)184–211.

Easton, D. (1965a) A Framework for Political Analysis(Englewood Cliffs, NJ: Prentice-Hall).

Easton, D. (1965b) A Systems Analysis of Political Life (New York:Wiley).

330 REFERENCES

Eberle, J. (1990) ‘Understanding the Revolutions in EasternEurope’, in Spring in Winter: The 1989 Revolutions, ed. G.Prins (Manchester: Manchester University Press) pp. 193–209.

Eckstein, H. (1998a) ‘Congruence Theory Explained’, in CanDemocracy Take Root in Post-Soviet Russia? Explorations inState–Society Relations, ed. H. Eckstein et al. (Lanham, MDand Oxford: Rowman & Littlefield) pp. 3–33.

Eckstein, H. (1998b) ‘Russia and the Conditions of Democracy’,in Can Democracy Take Root in Post-Soviet Russia? Explorationsin State–Society Relations, ed. H. Eckstein et al. (Lanham, MDand Oxford: Rowman & Littlefield) pp. 349–81.

Eckstein, H. et al. (eds) (1998) Can Democracy Take Root in Post-Soviet Russia? Explorations in State–Society Relations (Lanham,MD and Oxford: Rowman & Littlefield).

Edelman, M. (1964) The Symbolic Uses of Politics (Urbana, IL:University of Illinois Press).

Edelman, M. (1995) ‘Israel’, in The Global Expansion of JudicialPower, ed. C. Tate and T. Vallinder (New York and London:New York University Press) pp. 403–16.

Eigen, L. and Siegel, J. (1993) The Macmillan Dictionary ofPolitical Quotations (New York: Macmillan).

Eising, R. (2003) ‘Interest Groups and the European Union’ inEuropean Union Politics, ed. M. Cini (Oxford and New York:Oxford University Press) pp. 192–204.

Eising, R. and Cini, M. (2002) ‘Disintegration or Recon-figuration? Organized Interests in Western Europe’, inDevelopments in West European Politics 2, ed. P. Heywood, E.Jones and M. Rhodes (Basingstoke and New York: Palgrave)pp. 168–83.

Elazar, D. (1996) ‘From Statism to Federalism: A ParadigmShift’, International Political Science Review (17) 417–30.

Elazar, D. (1998) Constitutionalizing Globalization: ThePostmodern Revival of Confederal Arrangements (Lanham, MDand Oxford: Rowman & Littlefield).

Eley, G. and Suny, R. (1996) ‘From the Moment of SocialHistory to the Work of Cultural Representation’, in BecomingNational: A Reader, ed. G. Eley and R. Suny (Oxford and NewYork: Oxford University Press) pp. 3–37.

Elgie, R. (1995) Political Leadership in Liberal Democracies(Basingstoke: Macmillan).

Elgie, R. (1999a) ‘France’, in Semi-Presidentialism in Europe, ed.R. Elgie (Oxford and New York: Oxford University Press) pp.1–21.

Elgie, R. (ed.) (1999b) Semi-Presidentialism in Europe (Oxfordand New York: Oxford University Press).

Elgie, R. (2003) Political Institutions in Contemporary France(Oxford and New York: Oxford University Press).

Elster, J., Offe, C. and Preuss, U. (1998) Institutional Design inPost-Communist Societies: Rebuilding the Ship at Sea(Cambridge and New York: Cambridge University Press).

Emy, M. and Hughes, O. (1991) Australian Politics: Realities inConflict (South Melbourne: Macmillan).

Englebert, D., Tarango, S. and Carter, M. (2002) ‘Dismember-ment and Suffocation: A Contribution to the Debate on AfricanBoundaries’, Comparative Political Studies (35) 1093–1118.

Englebert, P. (2002) State Legitimacy and Development in Africa(Boulder, CO and London: Lynne Rienner).

Entman, R. (1993) ‘Framing: Toward Clarification of a FracturedParadigm’, Journal of Communications (43) 51–8.

Eriksen, E. (1990) ‘Towards the Post-Corporate State?’,Scandinavian Political Studies (13) 345–64.

Esping-Andersen, G. (1990) The Three Worlds of WelfareCapitalism (Oxford: Polity).

Esping-Andersen, G. (1996a) ‘After the Golden Age? Welfare State

Dilemmas in a Global Economy’, in Welfare States in Transition:National Adaptations in Global Economies, ed. G. Esping-Andersen (Thousand Oaks, CA and London: Sage) pp. 1–31.

Esping-Andersen, G. (ed.) (1996b) Welfare States in Transition:National Adaptations in Global Economies (Thousand Oaks,CA and London: Sage).

Esping-Andersen, G. (ed.) (2002) Why We Need a Welfare State(Oxford and New York: Oxford University Press).

Eulau, H. (1963) The Behavioral Persuasion in Politics (New York:Random House).

Evans, C. (2001) ‘Committees, Leaders and Message Politics’, inCongress Reconsidered, 7th edn, ed. L. Dodd and B.Oppenheimer (Washington, DC: Congressional QuarterlyPress) pp. 82–97.

Evans, J. (2003) Voters and Voting: An Introduction (London andThousand Oaks, CA: Sage).

Evans, P., Rueschemeyer, D. and Skocpol, T. (eds) (1985)Bringing The State Back In (Cambridge and New York:Cambridge University Press).

■FFairbanks, C. (2001) ‘Ten Years After the Soviet Breakup:

Disillusionment in the Caucasus and Central Asia’, Journal ofDemocracy (2001) 49–56.

Farnen, R. and Meloen, J. (eds) (2000) Democracy,Authoritarianism and Education (Basingstoke: Palgrave).

Farrell, D. (2001) Electoral Systems: A Comparative Introduction(London and New York: Palgrave).

Farrell, D. and Schmitt-Beck, R. (eds) (2002) Do PoliticalCampaigns Matter? Campaign Effects in Elections andReferendums (London and New York: Routledge).

Feigenbaum, H., Henig, J. and Hamnett, C. (2003) Shrinkingthe State: The Political Underpinnings of Privatization(Cambridge and New York: Cambridge University Press).

Feldman, N. (2000) After Jihad: America and the Struggle forIslamic Democracy (New York: Farrar, Straus and Giroux).

Feldman, O. (1993) Politics and the News Media in Japan (AnnArbor, MI: University of Michigan Press).

Fensmith, J. (2001) ‘The Political and Social Implications ofChina’s Accession to the World Trade Organization’, ChinaQuarterly (167) 573–91.

Ferejohn, J. Rakove, J. and Riley, J. (eds) (2001) ConstitutionalCulture and Democratic Rule (Cambridge and New York:Cambridge University Press).

Fernando J. and Heston A. (1997) ‘NGOs between States,Markets and Civil Society’ Annals of the American Academy ofPolitical and Social Sciences (554) 8–19.

Fesler, J. and Kettl, D. (1996) The Politics of the AdministrativeProcess, 2nd edn (Chatham, NJ: Chatham House).

Figgis, J. (1960) Political Thought from Gerson to Grotius,1414–1625 (New York: Harper Torchbooks).

Finer, S. [1962] (1988 edn) The Man on Horseback: The Role ofthe Military in Politics (Boulder, CO: Westview).

Finer S. (1975) ‘State- and Nation-building in Europe: The Roleof the Military’, in The Formation of National States in WesternEurope, ed. C. Tilly (Princeton, NJ: Princeton UniversityPress) pp. 84–163.

Finer, S. (1997) The History of Government from the Earliest Times,3 vols (Oxford and New York: Oxford University Press).

Finley, M. (1985) Democracy Ancient and Modern (London:Hogarth Press).

Fiorina, M. (1981) Retrospective Voting in American NationalElections (New Haven, CT: Yale University Press).

REFERENCES 331

Fischer, F. (2003) Reframing Public Policy: Discursive Practice andDeliberative Practices (Oxford and New York: OxfordUniversity Press).

Fishkin, J. (1991) Democracy and Deliberation: New Directions forDemocratic Reform (New Haven, CT and London: YaleUniversity Press).

Fitzmaurice, J. (1998) Politics and Government in the VisegradCountries: Poland, Hungary, the Czech Republic and Slovakia(London: Macmillan and New York: St Martin’s Press).

Flammang, J. et al. (1990) American Politics in a Changing World(Pacific Grove, CA: Brooks/Cole).

Flora, P. and Heidenheimer, A. (eds) (1981) The Development ofWelfare States in Europe and America (New York: Transaction).

Foley, M. (1999) ‘In Kiev They Fine a Journalist $1m and CutOff All the Phones’, The Times, April 2 1999, p. 45.

Foran, J. (ed.) (1997) Theorizing Revolutions (London and NewYork: Routledge).

Forsyth, M. (1989) Federalism and Nationalism (Leicester:Leicester University Press).

Forsyth, M. (1996) ‘The Political Theory of Federalism: TheRelevance of Classical Approaches’, in Federalizing Europe: TheCosts, Benefits and Preconditions of Federal Political Systems, ed.J. Hesse and V. Wright (Oxford and New York: OxfordUniversity Press) pp. 47–64.

Fortescue, J. (1714) On the Governance of the Kingdom of England(London: Bowyer).

Foucault, M. (1977) Discipline and Punish: The Birth of thePrison, trans. A. Sheridan (New York: Pantheon).

Frank, A. (1969) Capitalism and Underdevelopment in LatinAmerica (New York: Monthly Review Press).

Franklin, M. (1992) ‘The Decline of Cleavage Politics’, inElectoral Change: Responses to Evolving Social and AttitudinalStructures in Western Countries, ed. M. Franklin, T. Mackie andH. Valen (Cambridge and New York: Cambridge UniversityPress) pp. 383–405.

Franklin, M. (1996) ‘Electoral Participation’, in Elections andVoting in Global Perspective, ed. L. LeDuc, R. Niemi and P.Norris (Thousand Oaks, CA and London: Sage) pp. 216–35.

Franklin, M. (2002) ‘The Dynamics of Electoral Participation’ inComparing Democracies 2, ed. L. LeDuc, R. Niemi and P.Norris (London and Thousand Oaks, CA) pp. 148–68.

Franklin, M., Mackie, T., Valen, H. et al. (eds) (1992) ElectoralChange: Responses to Evolving Social and Attitudinal Structuresin Western Countries (Cambridge and New York: CambridgeUniversity Press).

Fransmyr, T. et al. (eds) The Quantifying Spirit in the EighteenthCentury (Berkeley, CA: University of California Press).

Fried, R. (1990) Nightmare in Red: The McCarthy Era inPerspective (New York and Oxford: Oxford University Press).

Friedrich, C. (1937) Constitutional Government and Politics (NewYork: Harper).

Friedrich, C. and Brzezinski, Z. (1965) Totalitarian Dictatorshipand Autocracy (New York: Praeger).

Frühling, H. (1998) ‘Judicial Reform and Democratization inLatin America’, in Fault Lines of Democracy in Post-TransitionLatin America, ed. F. Agüero and J. Stark (Coral Gables, FL:University of Miami) pp. 237–62.

■GGagnon, A.-G. and Tully, J. (eds) (2001) Multinational Democracies

(Cambridge and New York: Cambridge University Press). Gallagher, M. et al. (2000) Representative Government in Modern

Europe, 3rd edn (New York and London: McGraw-Hill).

Gallagher, M. and Uleri, P. (eds) (1996) The ReferendumExperience in Europe (Basingstoke: Macmillan).

Gamble A. (2003) ‘Remaking the Constitution’ in Developmentsin British Politics 7, ed. P. Dunleavy et al. (Basingstoke andNew York: Palgrave Macmillan) pp. 18–38.

Gamble, A. and Payne, A. (eds) (1996a) Regionalism and WorldOrder (Basingstoke: Macmillan).

Gamble, A. and Payne, A. (1996b) ‘The New Regionalism’, inRegionalism and World Order, ed. A. Gamble and A. Payne(Basingstoke: Macmillan) pp. 247–64.

Gardner, J. (2000) About Common Cause. Common Cause homepage, available at http://www.commoncause.org/about/fact.htm .

Gavin, N. and Sanders, D. (2003) ‘The Press and its Influenceon British Political Attitudes under New Labour’, PoliticalStudies (51) 573–91.

Gay, P. du (2000) In Praise of Bureaucracy: Weber, Organization,Ethics (London and Thousand Oaks, CA: Sage).

Geertz, C. [1973] (1993 edn) ‘Thick Description: Toward anInterpretative Theory of Culture’, in Interpretation of Cultures,ed. C. Geertz (London: Fontana) pp. 1–33.

Gellner, E. (1983) Nations and Nationalism (Oxford andCambridge, MA: Blackwell).

Gerth, H. and Mills, C. Wright (1948) From Max Weber(London: Routledge & Kegan Paul).

Gill, G. (2000) The Dynamics of Democratization: Elites, CivilSociety and the Transition Process (Basingstoke: Macmillan, andNew York: St Martin’s Press).

Gill, G. and Markwick, R. (2000) Russia’s Stillborn Democracy:From Gorbachev to Yeltsin (Oxford and New York: OxfordUniversity Press).

Gills, B., Rocamora, J. and Wilson, R. (eds) (1993) Low IntensityDemocracy: Political Power in the New World Order (Boulder,CO and London: Pluto).

Ginsberg, B. (1982) The Consequences of Consent (Reading, MA:Addison Wesley).

Ginsberg, B. (1986) The Captive Public: How Mass OpinionPromotes State Power (New York: Basic Books).

Giol, J. (1990) ‘By Consociationalism to a MajoritarianParliamentary System: The Rise and Decline of the SpanishCortes’, in Parliament and Democratic Consolidation inSouthern Europe, ed. U. Liebert and M. Cotta (London:Pinter) pp. 92–131.

Glaser, D. (2001) Politics and Society in South Africa (London andThousand Oaks, CA: Sage).

Gleason, A. (1995) Totalitarianism: The Inner History of the ColdWar (New York: Oxford University Press).

Glynn, C. et al. (1998) Public Opinion (Boulder, CO andOxford: Westview Press).

Goban-Klas, T. and Sasinka-Klas, T. (1992) ‘From Closed toOpen Communication Systems’, in Democracy and CivilSociety in Eastern Europe, ed. P. Lewis (London: Macmillanand New York: St Martin’s Press) pp. 76–90.

Goetz, K. (2003) ‘Government at the Centre’, in Developments inGerman Politics 3, ed. S. Padgett, W. Paterson and G. Smith(Basingstoke: Palgrave Macmillan) pp. 17–37.

Goetz, K. (1997) ‘Acquiring Political Craft: Training Grounds forTop Officials in the German Core Executive’, PublicAdministration (75) 753–75.

Goldsmith, M. (1996) ‘Normative Theories of LocalGovernment: A European Survey’, in Rethinking LocalDemocracy, ed. D. King and G. Stoker (Basingstoke and NewYork: Macmillan).

Goldstone, J. (1991) ‘An Analytical Framework’, in Revolutions of

332 REFERENCES

the Late Twentieth Century, ed. J. Goldstone, T. Gurr and F.Moshiri (Boulder, CO and Oxford: Westview) pp. 37–51.

Gomm, R. and Hammersley, M. (eds) (2000) Case Study Method:Key Issues, Key Texts (Thousand Oaks, CA and London: Sage).

Gong, T. (2002) ‘Dangerous Collusion: Corruption as aCollective Venture in Contemporary China’, Communist andPostsocialist Studies (35) 85–103.

Goodman, F. (2001) ‘The Supreme Court’s Federalism: Real orImagined?’, Annals of the American Academy of Political andSocial Science (574) 9–23.

Graber, D. (1997) Mass Media and American Politics, 5th edn(Washington, DC: CQ Press).

Green, D. (2002a) ‘Constructivist Comparative Politics:Foundations and Framework’, in Constructivism andComparative Politics, ed. D. Green (Armonk, NY and London:M.E. Sharpe) pp. 3–59.

Green, D. (ed.) (2002b) Constructivism and Comparative Politics(Armonk, NY and London: M.E. Sharpe).

Green, D. and Shapiro, I. (1994) Pathologies of Rational ChoiceTheories (New Haven, CT and London: Yale University Press).

Green, J. (1994) Dictionary of Cynical Quotations (London:Cassell).

Greenbaum, T. (1998) Handbook for Focus Group Research, 2ndedn (Thousand Oaks, CA and London: Sage).

Greenwood, J. (2003) Interest Representation in the EuropeanUnion (Basingstoke and New York: Palgrave Macmillan).

Griffin, R. (ed.) (2004) Fascism as a Totalitarian Movement(London: Frank Cass).

Griffith, J. (1997) The Politics of the Judiciary, 4th edn (London:Fontana).

Griffiths, I. (1995) The African Inheritance (London and NewYork: Routledge).

Grugel, J. (2002) Democratization: A Critical Introduction(Basingstoke: Palgrave Macmillan).

Guarnieri, C. and Pederzoli, P. (2002) The Power of Judges: AComparative Study of Courts and Democracy, trans. C. Thomas(Oxford and New York: Oxford University Press).

Guéhenno, J.-M. (1995) The End of the Nation-State (Min-neapolis, MN and London: University of Minnesota Press).

Guibernau M. (1999) Nations Without States: Political Communitiesin a Global Age (Cambridge and Malden, MA: Polity).

Gunther, R. and Mughan, A. (eds) (2000) Democracy and theMedia: A Comparative Perspective (Cambridge and New York:Cambridge University Press).

Gunther, R., Montero, J. and Linz, J. (eds) (2002) PoliticalParties: Old Concepts and New Challenges (Oxford and NewYork: Oxford University Press).

Gurr, T. (1980) Why Men Rebel (Princeton, NJ: PrincetonUniversity Press).

■HHabermas, J. (1978) Knowledge and Human Interests, 2nd edn,

trans. J. Shapiro (London: Heinemann Education). Hadenius, A. (ed.) (1997) Democracy’s Victory and Crisis (New

York and Cambridge: Cambridge University Press).Hagopian, F. (1998) ‘Democracy and Political Representation in

Latin America in the 1990s: Pause, Reorganization orDecline?’, in Fault Lines of Democracy in Post-Transition LatinAmerica, ed. F. Agüero and J. Stark (Coral Gables, FL:University of Miami) pp. 99–144.

Hagopian, F. (2004) ‘Brazil’ in Comparative Politics Today: AWorld View, 8th ed., ed. G. Almond et al. (New York andLondon: Pearson Longman) pp. 520–79.

Hall, B. and Biersteker, T. (2002) (eds) The Emergence of PrivateAuthority in Global Governance (Cambridge and New York:Cambridge University Press).

Hall, P. and Ikenberry, G. (1989) The State (Milton Keynes:Open University Press).

Halliday, F. (1999) Revolution and World Politics: The Rise andFall of the Sixth Great Power (Basingstoke: Macmillan).

Hamilton, A. [1788a] (1970 edn) The Federalist, No. 51, introW. Brock (London: Dent and New York: Dutton) pp. 263–7.

Hamilton, A. [1788a] (1970 edn) The Federalist, No. 70, introW. Brock (London: Dent and New York: Dutton) pp. 357–63.

Hamilton, A. [1788b] (1970 edn) The Federalist, No. 84, introW. Brock (London: Dent and New York: Dutton) pp. 436–45.

Hamilton, A. [1788c] (1970 edn) The Federalist, No. 45, introW. Brock (London: Dent and New York: Dutton) pp. 233–8.

Handelman, H. (1997) Mexican Politics: The Dynamics of Change(New York: St. Martin’s Press).

Hansen, M. (1991) The Athenian Democracy in the Age ofDemosthenes (Oxford and Cambridge, MA: Blackwell).

Harding, N. (ed.) (1984) The State in Socialist Society (London:Macmillan).

Hare, T. (1873) The Election of Representatives (London:Longmans).

Harriss, J. (1995) ‘A Time of Troubles: Problems ofHumanitarian Intervention in the 1990s’, in The Politics ofHumanitarian Intervention, ed. J. Harriss (London and NewYork: Pinter) pp. 1–15.

Harrop, M., and Miller, W. (1987) Elections and Voters: AComparative Introduction (Basingstoke: Macmillan).

Haugaard, M. (ed.) (2002b) Power: A Reader (Manchester andNew York: Manchester University Press).

Hay, C. (2002) Political Analysis: A Critical Introduction(Basingstoke and New York: Palgrave).

Hayek, F. (1960) The Constitution of Liberty (Chicago, IL:University of Chicago Press).

Haynes, J. (1998) Religion in Global Politics (Harlow and NewYork: Addison Wesley Longman).

Haynes, J. (2001) Democracy in the Developing World: Africa,Asia, Latin America and the Middle East (Cambridge: Polity).

Haynes, J. (2003) ‘Comparative Politics and “Globalization” ’,European Political Science (2) 17–26.

Hayward, J. (1994) ‘Ideological Change: The Exhaustion of theRevolutionary Impulse’, in Developments in French Politics, ed.P. Hall, J. Hayward and H. Machin (Basingstoke: Macmillan)pp. 15–32.

Hazan, R. (1997a) ‘The 1996 Election in Israel: Adopting PartyPrimaries’, Electoral Studies (16) 95–102.

Hazan, R. (1997b) ‘Legislative-Executive Relations in an Era ofAccelerated Reform: Reshaping Government in Israel’,Legislative Studies Quarterly (22) 329–50.

Hazan, R. (2002) ‘Candidate Selection’, in ComparingDemocracies 2: New Challenges in the Study of Elections andVoting, ed. L. LeDuc, R. Niemi and P. Norris (London andThousand Oaks, CA: Sage) pp. 108–26.

Heady, F. (1996) Public Administration: A ComparativePerspective, 5th edn (New York: Marcel Dekker).

Heater, D. (1999) What Is Citizenship? (Cambridge: Polity andMalden, MA: Blackwell).

Hechter, M. (2000) Containing Nationalism (Oxford and NewYork: Oxford University Press).

Heclo, H. (1974) Modern Social Policies in Britain and Sweden(New Haven, CT and London: Yale University Press).

Heclo, H. (1978) ‘Issue Networks and the ExecutiveEstablishment’, in The New American Political System, ed. A.

REFERENCES 333

King (Washington, DC: American Enterprise Institute) pp.87–124.

Held, D. (1996) Models of Democracy, 2nd edn (Cambridge:Polity).

Helms, D. (ed.) (2000) Institutions and Institutional Change in theFederal Republic of Germany (Basingstoke: Palgrave).

Henry, N. (2004) Public Administration and Public Affairs, 9thedn (Upper Saddle River, NJ: Pearson).

Henshel, R. (1990) Thinking About Social Problems (New York:Harcourt Brace Jovanovich).

Herbst, J. (2001) ‘Political Liberalization in Africa after 10 Years,’Comparative Politics (33) 357–75.

Herbst, S. (1998) Reading Public Opinion: How Political ActorsView The Political Process (Chicago and London: University ofChicago Press).

Hesse, J. and Wright, V. (eds) (1996) Federalizing Europe? TheCosts, Benefits and Preconditions of Federal Political Systems(Oxford and New York: Oxford University Press).

Hettne, B. (ed.) (1999) Globalism and the New Regionalism(Basingstoke: Macmillan).

Heywood, P. (1995) The Government and Politics of Spain(Basingstoke: Macmillan).

Hilderband, M. (2002) ‘Capacity Building’, in Handbook onDevelopment Policy and Management, ed. C. Kirkpatrick, R.Clarke and C. Polidano (Aldershot and Brookfield, VT:Edward Elgar) pp. 323–32.

Hill, L. (2002) ‘On the Rightness of Compelling Citizens to“Vote”: The Australian Case’, Political Studies (50) 80–101.

Hill, M. and Hupe, P. (2002) Implementing Public Policy:Governance in Theory and Practice (London and ThousandOaks, CA: Sage).

Hinsley, F. (1986) Sovereignty (Cambridge and New York:Cambridge University Press).

Hirschl, R. (2002) ‘The Political Origins of Judicial Empowermentthrough Constitutionalization: Lessons from Israel’sConstitutional Review’, Comparative Politics (33) 315–35.

Hirst, P. and Thompson, G. (1999) Globalization in Question:The International Economy and the Possibilities of Governance,2nd edn (Oxford and Cambridge, MA: Blackwell).

Hobbes, T. [1651] (1968 edn) Leviathan, ed. M. Oakeshott(Toronto: Crowell-Collier).

Hoffman, J. (1998) Sovereignty (Buckingham: Open UniversityPress).

Hoffman, S. (1995) ‘The Politics and Ethics of MilitaryIntervention’, Survival (37) 29–51.

Hogwood, B. and Gunn, L. (1984) Policy Analysis for the RealWorld (Oxford: Oxford University Press).

Holland, K. (1991) ‘Introduction’, in Judicial Activism inComparative Perspective, ed. K. Holland (Basingstoke:Macmillan) pp. 1–11.

Holliday, I. (2000) ‘Executives and Administrations’, inDevelopments in British Politics 6, ed. P. Dunleavy et al.(Basingstoke: Macmillan and New York: St Martin’s Press) pp.88–107.

Hollifield, J. and Jillson, C. (2000) ‘The DemocraticTransformation: Lessons and Prospects in Pathways toDemocracy’, in The Political Economy of DemocraticTransitions, ed. J. Hollifield and C. Jillson (London and NewYork: Routledge) pp. 3–20.

Holman, M. and Wallis, W. (2000) ‘In Office But Out of Power’,Financial Times Survey of Nigeria, 30 March, p. 1.

Holmstrom, B. (1995) ‘Sweden’, in The Global Expansion ofJudicial Power, ed. C. Tate and T. Vallinder (New York andLondon: New York University Press) pp. 345–68.

Holzgrefe, J. and Keohane, R. (eds) (2003) HumanitarianIntervention: Ethical, Legal and Political Dilemmas (Cambridgeand New York: Cambridge University Press).

Hood, C. (1996) ‘Exploring Variations in Public ManagementReform in the 1990s’, in Civil Service Systems in ComparativePerspective, ed. H. Bekke, J. Perry and T. Toonen(Bloomington, IN: Indiana University Press) pp. 268–87.

Hood, C. et al. (1999) Regulation Inside Government: Waste-Watchers, Quality Police and Sleaze-Busters (Oxford and NewYork: Oxford University Press).

Hooghe, L. and Marks, G. (2001) Multilevel Governance andEuropean Integration (Lanham, MD: Rowman & Littlefield).

Horowitz, D. (1977) The Courts and Social Policy (Washington,DC: Brookings).

Hosli, M., van Deemen, A. and Widgrén, M. (eds) (2002)Institutional Challenges in the European Union (London andNew York: Routledge).

Howe, H. (2001) Ambiguous Order: Military Forces in AfricanStates (Boulder, CO and London: Lynne Rienner).

Hrebenar, R. and Scott, R. (1997) Interest Group Politics inAmerica, 3rd edn (Englewood Cliffs, NJ: Prentice-Hall).

Huntington, S. (1991) The Third Wave: Democratization in theLate Twentieth Century (Norman, OK and London: Universityof Oklahoma Press).

Huntington, S. (1993) ‘Clash of Civilizations’, Foreign Affairs(72) 22–49.

Huntington, S. (1996) The Clash of Civilizations and the Makingof World Order (New York: Simon & Schuster).

Hurrell, A. (1995) ‘Explaining the Resurgence of Regionalism inWorld Politics’, Review of International Studies (21) 331–58.

Huth, P. and Allee, T. (2003) The Democratic Peace and TerritorialConflict in the Twentieth Century (Cambridge and New York:Cambridge University Press).

Hyden, G. (1997) ‘Democratization and Administration’, in Democracy’s Victory and Crisis, ed. A. Hadenius (Cambridgeand New York: Cambridge University Press) pp. 242–62.

■IIbarra, P. (2003) Social Movements and Democracy (London:

Palgrave Macmillan). Ignazi, P. (1992) ‘The Silent Counter-Revolution: Hypotheses on

the Emergence of Extreme Right-Wing Parties in Europe’,European Journal of Political Research (22) 3–35.

Ignazi, P. (2003) Extreme Right Parties in Western Europe (Oxfordand New York: Oxford University Press).

Illner, M. (1998) ‘Local Democratization: The Czech Republicafter 1989’, in Participation and Democracy: Comparisons andInterpretations, ed. D. Rueschmeyer, M. Rueschmeyer and B.Wittrock (Armonk, NY and London: M.E. Sharpe) pp.51–82.

Inglehart, R. (1971) ‘The Silent Revolution in Europe:Intergenerational Change in Post-Industrial Societies’,American Political Science Review (65) 991–1017.

Inglehart, R. (1990) Culture Shift in Advanced Industrial Society(Princeton, NJ: Princeton University Press).

Inglehart, R. (1997) Modernization and Postmodernization:Cultural, Economic and Social Change in 43 Societies(Princeton, NJ and London: Princeton University Press).

Inglehart, R. (1999) ‘Postmodernization Erodes Respect forAuthority, but Increases Support for Democracy’, in CriticalCitizens: Global Support for Democratic Governance, ed. P.Norris (Oxford and New York: Oxford University Press) pp.236–56.

334 REFERENCES

Inoguchi, T. (2002) ‘Broadening the Basis of Political Capital inJapan’, in Democracies in Flux: The Evolution of Social Capitalin Contemporary Society, ed. R. Putnam (Oxford and NewYork: Oxford University Press) pp. 359–92.

Inter-Parliamentary Union (2000) Women in NationalParliaments, IPU home page at: http://www.ipu.org/wmn-e/world.htm.

International Organization for Migration (IOM) (2003) AnnualReport (Geneva: IOM).

■JJackson, J. (1996) ‘Political Methodology: An Overview’, in A

New Handbook of Political Science, ed. R. Goodin and H.Klingemann (Oxford and New York: Oxford University Press)pp. 714–48.

Jackson, J. (1998) The World Trade Organization (Herndon, VAand London: Cassell).

Jackson, P. and Nexon, D. (2002) ‘Globalization, TheComparative Method, and Comparing Constructions’, inConstructivism and Comparative Politics, ed. D. Green(Armonk, NY and London: M.E. Sharpe) pp. 88–120.

Jackson, R. and Rosberg, C. (1982) Personal Rule in Black Africa:Prince, Autocrat, Prophet, Tyrant (Berkeley, CA: University ofCalifornia Press).

Jacob, H. et al. (1996) Courts, Law and Politics in ComparativePerspective (New Haven, CT and London: Yale University Press).

Jasiewicz, K. (2003) ‘Elections and Voting Behaviour’, inDevelopments in Central and East European Politics 3, ed. S.White, J. Batt and P. Lewis (Basingstoke and New York:Palgrave Macmillan) pp. 173–89.

Jay, A. (1996) The Oxford Dictionary of Political Quotations(Oxford and New York: Oxford University Press).

Jayanntha, D. (1991) Electoral Allegiance in Sri Lanka(Cambridge and New York: Cambridge University Press).

Jeffery, C. (2003) ‘Federalism and Territorial Politics’ inDevelopments in German Politics 3, ed. S. Padgett, W. Patersonand G. Smith (Basingstoke: Palgrave Macmillan) pp. 38–59.

Jenkins, J. (1983) ‘The Transformation of a Constituency into aMovement: Farmworker Organizing in California’, in SocialMovements of the Sixties and Seventies, ed. J. Freeman (Londonand New York: Longman) pp. 52–70.

John, P. (1998) Analysing Public Policy (London and New York:Continuum).

John, P. (2001) Local Governance in Western Europe (London andThousand Oaks, CA: Sage).

Johnson, C. (1987) Japan: Who Governs? The Rise of theDevelopmental State (New York and London: Norton).

Johnson, J. (2001) ‘Path Contingency in PostcommunistTransformations’, Comparative Politics (33) 253–74.

Jones, C. (1994) The Presidency in a Separated System(Washington, DC: Brookings Institution).

Jones, M. (1995a) ‘A Guide to the Electoral Systems of theAmericas’, Electoral Studies (14) 5–21.

Jones, M. (1995b) Electoral Laws and the Survival of PresidentialDemocracies (Notre Dame, IN: University of Notre DamePress).

Joyce, P. (2002) The Politics of Protest: Extra-Parliamentary Politicsin Britain since 1970 (London: Palgrave Macmillan).

Judge, D. and Earnshaw, D. (2003) The European Parliament(Basingstoke and New York: Palgrave Macmillan).

Jungar, A.-C. (2002) ‘A Case of a Surplus Majority Government:The Finnish Rainbow Coalition’, Scandinavian Political Studies(25) 57–83.

■KKahn, J. (2002) Federalism, Democratization and the Rule of Law

in Russia (Oxford and New York: Oxford University Press). Karatnycky, A. (2002) ‘Muslim Countries and the Democracy

Gap’, Journal of Democracy (13) 99–112. Karvonen, L. and Kuhnle, S. (eds) (2001) Party Systems and Voter

Alignments Revisited (London and New York: Routledge). Katz, R. (1997) Democracy and Elections (Oxford and New York:

Oxford University Press).Katz, R. (2002) ‘Reforming the Italian Electoral Law, 1993’, in

Mixed-Member Electoral Systems: The Best of Both Worlds, ed.M. Shugart and M. Wattenberg (Oxford and New York:Oxford University Press) pp. 96–122.

Katz, R. and Mair, P. (1995) ‘Changing Models of PartyOrganization and Party Democracy: The Emergence of theCartel Party’, Party Politics (1) 5–28.

Katzenstein, J. (1984) Corporatism and Change: Austria,Switzerland and the Politics of Industry (Ithaca, NY: CornellUniversity Press).

Kay, J. (2003) ‘Too Many Polls Are Apt to Harm a Democracy’,Financial Times, 14 August.

Kelsen, H. (1942) ‘Judicial Review of Legislation: A ComparativeStudy of the Austrian and the American Constitution’, Journalof Politics (4) 183–200.

Keman, H. (ed.) (1993) Comparative Politics: New Directions inTheory and Method (Amsterdam: VU University Press).

Kennedy, P. (1993) Preparing for the Twenty-First Century(London: HarperCollins).

Kennedy, P. (1994) Preparing for the Twenty First Century(London: Fontana).

Kennedy, P. and Roudometof, V. (eds) (2002) Communities AcrossBorders: New Immigrants and Transnational Cultures (Londonand New York: Routledge).

Kershaw, I. (2001) Hitler, 1936–1945: Nemesis (London:Penguin).

Kershaw, I. (2000) The Nazi Dictatorship: Problems andPerspectives of Interpretation, 4th edn (London: Arnold).

Kershaw, I. and Lewin, M. (eds) (1997) Stalinism and Nazism:Dictatorships in Comparison (Cambridge and New York:Cambridge University Press).

Khong, Y. (1992) Analogies at War: Korea, Munich, Dien Bien Phuand the Vietnam Decisions of 1965 (Princeton, NJ: PrincetonUniversity Press).

Kieh, G. and Agbese, P. (eds) (2003) The Military and Politics inAfrica: From Intervention to Democratic Control (Aldershot andBurlington, VT: Ashgate).

Kieh G. and Ogaba P. (eds) (2003) The Military and Politics inAfrica: From Intervention to Democratic Control (Aldershot:Ashgate).

King, A. (1981) ‘The Rise of the Career Politician in Britain –and its Consequences’, British Journal of Political Science (11)249–85.

King, A. (1994a) ‘Ministerial Autonomy in Britain’, in CabinetMinisters and Parliamentary Government, ed. M. Laver and K.Shepsle (Cambridge and New York: Cambridge UniversityPress) pp. 203–25.

King, A. (1994b) ‘“Chief Executives” in Western Europe’, inDeveloping Democracy: Comparative Research in Honour of J. F.P. Blondel, ed. I. Budge and D. McKay (Thousand Oaks, CAand London: Sage) pp. 150–64.

King, A. (ed.) (2002) Leaders’ Personalities and the Outcomes ofDemocratic Elections (Oxford and New York: OxfordUniversity Press).

REFERENCES 335

King, G., Keohane, R. and Verba S. (1994) Designing SocialInquiry: Scientific Inference in Qualitative Research (Princeton,NJ: Princeton University Press).

Kingdon, J. (1984) Agendas, Alternatives and Public Policy(Boston, MA: Little, Brown).

Kingsley, J. (1944) Representative Bureaucracy (Yellow Springs,OH: Antioch).

Kircheimer, O. (1966) ‘The Transformation of the WesternEuropean Party Systems’, in Political Parties and PoliticalDevelopment, ed. J. LaPalombara and M. Weiner (Princeton,NJ: Princeton University Press) pp. 177–200.

Kitschelt, H. et al. (1999) Postcommunist Party Systems:Competition, Representation and Inter-Party Competition(Cambridge and New York: Cambridge University Press).

Klapper, J. (1960) The Effects of Mass Communication (New Yorkand London: Free Press).

Knapp, A. and Wright, V. (2001) The Government and Politics ofFrance, 4th edn (London and New York: Routledge).

Knutsen, O. (1990) ‘Materialist and Postmaterialist Values andStructures in the Nordic Countries’, Comparative Politics (23)85–101.

Knutsen, O. (2001) ‘Social Class, Sector Employment andGender as Party Cleavages in the Scandinavian Countries: AComparative Longitudinal Study, 1970–95’, ScandinavianPolitical Studies (24) 311–50.

Kobach, K. (1997) ‘Direct Democracy and Swiss Isolationism’,West European Politics (20) 185–211.

Koh, B. (1989) Japan’s Administrative Elite (Berkeley, CA:University of California Press).

Kolstø, P. (2000) Political Construction Sites: Nation-Building inRussia and the Post-Soviet States (Boulder, CO: WestviewPress).

Koopmans, R. (1999) ‘Globalization or Still National Politics? AComparison of Protests Against the Gulf War in Germany,France and the Netherlands’, in Social Movements in aGlobalizing World, ed. D. della Porta, H. Kriesi and D. Rucht(London: Macmillan and New York: St Martin’s Press) pp.57–70.

Kopecky, P. (1995) ‘Developing Party Organizations in East-Central Europe: What Type of Party is Likely to Emerge?’,Party Politics (1) 515–34.

Kopecky, P. (2003) ‘Structures of Representation: The NewParliaments of Central and Eastern Europe’, in Developmentsin Central and East European Politics 3, ed. S. White, J. Battand P. Lewis (Basingstoke and New York: Palgrave Macmillan)pp. 133–52.

Kornhauser, W. (1959) The Politics of Mass Society (Glencoe, IL:Free Press).

Kostadinova, T. (2002) ‘Do Mixed Electoral Systems Matter? ACross-National Comparison of Their Effects in EasternEurope’, Electoral Studies (21) 23–34.

Kostiner, J. and Teitelbaum, J. (2000) ‘State Formation and theSaudi Monarchy’, in Middle East Monarchies: The Challenge ofModernity, ed. J. Kostiner (Boulder, CO and London: LynneRienner) .

Krasner, S. (1999) Sovereignty: Organized Hypocrisy (Princeton,NJ and Chichester: Princeton University Press).

Krasner, S. (ed.) (2001) Problematic Sovereignty: Contested Rulesand Political Possibilities (New York and Chichester: ColumbiaUniversity Press).

Krauthammer, C. (1990) ‘In Praise of Low Voter Turnout’, Time,21 May, p. 88.

Kudrle, R. and Marmor, T. (1981) ‘The Development of WelfareStates in North America’ in The Development of Welfare States

in Europe and America, ed. P. Flora and A. Heidenheimer(New Brunswick, NJ and London: Transaction) pp. 187–236.

Kuhn, T. [1970] (1996 edn) The Structure of ScientificRevolutions, 3rd edn (Chicago, IL: University of ChicagoPress).

Kuhnle, S. (2000) ‘The Scandinavian Welfare State in the 1990s:Challenged but Viable’, West European Politics (23) 209–28.

Kvavik, R. (1976) Interest Groups in Norwegian Politics (Oslo,Bergen and Tromsø: Universitetforlaget).

■LLachmann, R. (1997) ‘Agents of Revolution: Elite Conflicts and

Mass Mobilization from the Medici to the Yemen’, inTheorizing Revolutions, ed. J. Foran (London and New York:Routledge) pp. 73–101.

Laegreid, P. and Olsen, J. (1978) Byråkrati og Beslutninger(Bergen: Norwegian University Press).

Landes, R. (1995) The Canadian Polity: A ComparativeIntroduction (Scarborough, Ontario: Prentice-Hall Canada).

Landman, T. (2003) Issues and Methods in Comparative Politics:An Introduction, 2nd edn (London and New York: Routledge).

Lane, J.-E. and Ersson, S. (2002) Culture and Politics: AComparative Approach (Aldershot and Burlington, VT:Ashgate).

LaPalombara, J. (1974) Politics Within Nations (EnglewoodCliffs, NJ: Prentice-Hall).

Latouche, S. (1996) The Westernization of the World (Cambridgeand Cambridge, MA: Polity).

Laver, M. (1983) Invitation to Politics (Oxford: Martin Robertson).Laver, M. and Shepsle, K. (eds) (1994) Cabinet Ministers and

Parliamentary Government (Cambridge and New York:Cambridge University Press).

Lazarsfeld, P. and Merton, R. [1948] (1996 edn) ‘MassCommunication, Popular Taste and Organized Social Action’,in Media Studies: A Reader, ed. P. Marris and S. Thornham(Edinburgh: Edinburgh University Press) pp. 14–24.

Lechner, F. and Boli, J. (eds) (2003) The Globalization Reader(Oxford and Cambridge, MA: Blackwell).

LeDuc, L. (2002) ‘Referendums and Initiatives: The Politics ofDirect Democracy’, in Comparing Democracies 2: NewChallenges in the Study of Elections and Voting, ed. L. LeDuc, R.Niemi and P. Norris (London and Thousand Oaks, CA: Sage)pp. 70–87.

LeDuc, L., Niemi, R. and Norris, P. (2002a) ‘ComparingDemocratic Elections’, in Comparing Democracies 2: NewChallenges in the Study of Elections and Voting, ed. L. LeDuc, R.Niemi and P. Norris (Thousand Oaks, CA and London: Sage)pp. 1–39.

LeDuc, L., Niemi, R. and Norris, P. (eds) (2002b) ComparingDemocracies 2: New Challenges in the Study of Elections andVoting (Thousand Oaks, CA and London: Sage).

Lenin, V. [1902] (1963 edn) What Is To Be Done? (Oxford:Clarendon Press).

Lesch, A. (2003) ‘Politics in Egypt’, in Comparative Politics Today:A World View, 8th edn, ed. G. Almond et al. (New York:Longman) pp. 581–632.

Levitsky, S. (2000) ‘The “Normalization” of Argentine Politics’,Journal of Democracy (11) 56–69.

Levy, D. and Bruhn, K. (2001) Mexico: The Struggle forDemocratic Development (Berkeley, CA: University ofCalifornia Press).

Lewin, M. (1997) ‘Bureaucracy and the Stalinist State’, inStalinism and Nazism: Dictatorships in Comparison, ed. I.

336 REFERENCES

Kershaw and M. Lewin (Cambridge and New York:Cambridge University Press) pp. 53–74.

Lewis, B. (2003) The Crisis of Islam: Holy War and Unholy Terror(London: Weidenfeld & Nicolson).

Lewis, B. (2002) What Went Wrong? Western Impact and MiddleEastern Response (London: Weidenfeld & Nicolson).

Lewis, D. (2002) ‘The Politics of Agency Termination:Confronting the Myth of Agency Termination’, Journal ofPolitics (64) 89–1207.

Lewis, P. (ed.) (1996a) Party Structure and Organization in East-Central Europe (Aldershot and Brookfield, VT: Edward Elgar).

Lewis, P. (1996b) ‘Introduction and Theoretical Overview’, inParty Structure and Organization in East-Central Europe, ed. P.Lewis (Aldershot and Brookfield, VT: Edward Elgar) pp.1–19.

Lewis, P. (2000) Political Parties in Postcommunist Eastern Europe(London and New York: Routledge).

Lewis, P. (2003) ‘Political Parties’, in Developments in Central andEast European Politics 3, ed. S. White, J. Batt and P. Lewis(Basingstoke and New York: Palgrave Macmillan) pp. 153–72.

Liddle, R. (1996) ‘A Useful Fiction: Democratic Legitimation inNew Order Indonesia’, in The Politics of Elections in SoutheastAsia, ed. R. Taylor (Cambridge and New York: CambridgeUniversity Press) pp. 34–60.

Lieberman, I., Nestor, S. and Desai, R. (eds) (1997) BetweenState and Market: Mass Privatization in Transition Economies(Washington, DC: World Bank).

Liebert, U. and Cotta, M. (eds) (1990) Parliament andDemocratic Consolidation in Southern Europe (London: Pinter).

Lijphart, A. (1971) ‘Comparative Politics and the ComparativeMethod’, American Political Science Review (65) 682–93.

Lijphart, A. (1977) Democracy in Plural Societies: A ComparativeExploration (Berkeley, CA: University of California Press).

Lijphart, A. (1984) Democracies: Patterns of Majoritarian andConsensual Government in Twenty One Countries (New Haven,CT and London: Yale University Press).

Lijphart, A. (ed.) (1992) Parliamentary versus Presidential Govern-ment (Oxford and New York: Oxford University Press).

Lijphart, A. (1994) Electoral Systems and Party Systems (NewHaven, CT and London: Yale University Press).

Lijphart, A. (1997) ‘Unequal Participation: Democracy’sUnresolved Dilemma’, American Political Science Review (91)1–14.

Lijphart, A. (1999) Patterns of Democracy: Government Forms andPerformance in Thirty Six Countries (New Haven, CT andLondon: Yale University Press).

Lijphart, A. (2000) ‘The Future of Democracy: Reasons forPessimism but also Some Optimism’, Scandinavian PoliticalStudies (23) 265–72.

Lijphart, A. and Crepaz, M. (1991) ‘Corporatism and ConsensusDemocracy in Eighteen Countries: Conceptual and EmpiricalLinkages’, British Journal of Political Science (21) 235–46.

Lijphart, A. and Waisman, C. (1996a) ‘The Design ofDemocracies and Markets: Generalizing Across Regions’, inInstitutional Design in New Democracies: Eastern Europe andLatin America, ed. A. Lijphart and C. Waisman (Boulder, COand Oxford: Westview) pp. 235–48.

Lijphart, A. and Waisman, C. (eds) (1996b) Institutional Designin New Democracies: Eastern Europe and Latin America(Boulder, CO and Oxford: Westview).

Lindblom, C. (1959) ‘The Science of Muddling Through’, PublicAdministration (19) 78–88.

Lindblom, C. (1977) Politics and Markets (New York: BasicBooks).

Lindblom, C. (1979) ‘Still Muddling, Not Yet Through’, PublicAdministration Review (39) 517–26.

Lindblom, C. (1990) Inquiry and Change: The Troubled Attemptto Understand and Shape Society (New Haven, CT andLondon: Yale University Press).

Linz, J. (1970) ‘An Authoritarian Regime: The Case of Spain’, inMass Politics: Studies in Political Sociology, ed. E. Allard and S.Rokkan (New York: Free Press and London: Collier-Macmillan) pp. 251–83.

Linz, J. [1975] (2000 edn) Totalitarian and Authoritarian Regimes(Boulder, CO and London: Lynne Rienner).

Linz, J. (1978) ‘Crisis, Breakdown and Re-equilibration’, in TheBreakdown of Democratic Regimes, ed. J. Linz and A. Stepan(Baltimore, MD and London: Johns Hopkins University Press)pp. 1–124.

Linz, J. (1990) ‘The Perils of Presidentialism’, Journal ofDemocracy (1) 51–69.

Linz, J. and Stepan, A. (1996) Problems of Democratic Transitionand Consolidation: Southern Europe, South America andPostcommunist Europe (Baltimore, MD and London: JohnsHopkins University Press).

Linz, J. and Stepan, A. (eds) (1978) The Breakdown of DemocraticRegimes (Baltimore, MD and London: Johns HopkinsUniversity Press).

Linz, J. and Valenzuela, A. (eds) (1994) The Failure of PresidentialDemocracy (Baltimore, MD: Johns Hopkins University Press).

Lippman, W. (1922) Public Opinion (London: Allen & Unwin).Lipset, S. [1960] (1983 edn) Political Man (New York: Basic

Books).Lipset, S. (1990) Continental Divide: The Values and Institutions

of the United States and Canada (London and New York:Routledge).

Lipset, S. and Rokkan, S. (1967) ‘Cleavage Structures, PartySystems and Voter Alignments’, in Party Systems and VoterAlignments, ed. S. Lipset and S. Rokkan (New York andLondon: Free Press) pp. 1–65.

Lister, F. (1996) The European Union, the United Nations and theRevival of Confederal Governance (Westport, CT: GreenwoodPress).

Lively, J. (1991) ‘Sieyès, Emmanuel Joseph’ in The BlackwellEncyclopaedia of Political Thought, ed. D. Miller (Oxford andCambridge, MA: Blackwell) pp. 475–6.

Locke, J. [1690] (1965 edn) Two Treatises of Government (NewYork: New American Library).

Longley, L. and Davidson, R. (eds) (1998) The New Roles ofParliamentary Committees (London: Frank Cass).

Loughlin, J. (2001) Subnational Democracy in the EuropeanUnion: Challenges and Opportunities (Oxford and New York:Oxford University Press).

Lowi, T. (1969) The End of Liberalism (New York: Norton).Lowndes, V. (2002) ‘Institutionalism’, in Theory and Methods in

Political Science, 2nd edn, ed. D. Marsh and G. Stoker(Basingstoke and New York: Palgrave Macmillan) pp. 90–108.

Lubbers, M., Gijsberts, M. and Scheepers, P. (2002) ‘ExtremeRight-wing Voting in Western Europe’, European Journal ofPolitical Research (41) 345–78.

Lubman, S. (1999) Bird in a Cage: Legal Reform in China afterMao (Princeton, NJ: Princeton University Press).

Luce, E. and Sevastopulo, D. (2003) ‘Mother India Calls In HerAsset-Rich but Far-Flung Offspring’, Financial Times, 18January, p. 7.

Luckham, R. and White, G. (1996b) ‘Democratizing the South’,in Democratization in the South: The Jagged Wave, ed. R.

REFERENCES 337

Luckham and G. White (Manchester and New York:Manchester University Press) pp. 1–10.

Lukes, S. (1974) Power: A Radical View (London: Macmillan). Lukes, S. (ed.) (1986) Power (Oxford and Cambridge, MA: Basil

Blackwell).Luong, P. and Weinthal, E. (2001) ‘Prelude to the Resource Curse:

Explaining Oil and Gas Development Strategies in the SovietSuccessor States’, Comparative Political Studies (34) 367–99.

Lutz, J. and Lutz, B. (2004) Global Terrorism (London and NewYork: Routledge).

Lyon, D. (2003) Surveillance after September 11 (Cambridge andCambridge, MA: Polity).

■MMackenzie, W. (1958) Free Elections (London: Allen & Unwin). Maddex, R. (2000) Constitutions of the World, 2nd edn

(Washington, DC and London: CQ Press).Madeley, J. (ed.) (2003) Religion and Politics (Aldershot and

Burlington, VT: Ashgate). Madison J. [1781] (1970 edn) ‘The Federalist, No. 10’ in A.

Hamilton, J. Madison and J. Jay, The Federalist or, The NewConstitution (London: Dent and New York: Dutton). Firstpub. in the New York Packet, November 23.

Magyar, K. (1992) ‘Military Intervention and Withdrawal inAfrica: Problems and Perspectives’, in From Military to CivilianRule, ed. C. Danopoulos (London and New York: Routledge)pp. 230–48.

Mahler, G. (2003) Comparative Politics: An Institutional andCross-National Approach, 4th edn (Upper Suddle River, NJ:Pearson).

Maier, K. (2002) This House Has Fallen: Nigeria in Crisis(London and New York: Penguin).

Mainwaring, S. (1992) ‘Presidentialism in Latin America’, inParliamentary versus Presidential Government, ed. A. Lijphart(Oxford and New York: Oxford University Press) pp. 111–17.

Mainwaring, S. (1997) ‘Multipartism, Robust Federalism andPresidentialism in Brazil’, in Presidentialism and Democracy inLatin America, ed. S. Mainwaring and M. Shugart (Cambridgeand New York: Cambridge University Press) pp. 55–109.

Mainwaring, S. and Scully, T. (eds) (1995) Building DemocraticInstitutions: Party Systems in Latin America (Stanford, CA:Stanford University Press).

Mainwaring, S. and Shugart, M. (1997a) ‘Juan Linz,Presidentialism and Democracy: A Critical Appraisal’,Comparative Politics (29) 449–72.

Mainwaring, S. and Shugart, M. (eds) (1997b) Presidentialismand Democracy in Latin America (Cambridge and New York:Cambridge University Press).

Mair, P. (1994) ‘Party Organizations: From Civil Society to theState’, in How Parties Organize: Change and Adaptation inParty Organizations in Western Democracies, ed. R. Katz and P.Mair (Thousand Oaks, CA and London: Sage).

Mair, P. (1996) ‘Comparative Politics: An Overview’, in A NewHandbook of Political Science, ed. R. Goodin and H.Klingemann (Oxford and New York: Oxford University Press)pp. 309–35.

Mair, P. (ed.) (1990) The West European Party System (Oxford andNew York: Oxford University Press).

Mair, P. (2001) ‘The Freezing Hypothesis: An Evaluation’, inParty Systems and Voter Alignments Revisited, ed. L. Karvonenand S. Kuhnle (London and New York: Routledge) pp. 27–44.

Mair, P. and van Biezen, I. (2001) ‘Party Membership in Europe,1980–2000’, Party Politics (7) 5–22.

Majone, G. (1996) Regulating Europe (London and New York:Routledge).

Malovà, D. and Haughton, T. (2002) ‘Making Institutions inCentral and Eastern Europe, and the Impact of Europe’, WestEuropean Politics (25) 101–20.

Mamdani, M. (2001) When Victims Become Killers: Colonialism,Nativism and Genocide in Rwanda (Princeton, NJ: PrincetonUniversity Press).

Mandaville, P. (2003) Transnational Muslim Politics: Reimaginingthe Umma (London and New York: Routledge).

Manion, M. (2000) ‘Politics in China’, in Comparative PoliticsToday: A World View, 7th edn, ed. G. Almond et al. (NewYork: Longman) pp. 419–62.

Manion, M. (2004) ‘Politics in China’, in Comparative PoliticsToday: A World View, 8th edn, ed. G. Almond et al. (New Yorkand London: Pearson Longman) pp. 418–65.

Mann, M. (1986) The Sources of Social Power, Volume 1: A Historyof Power from the Beginning to AD 1780 (Cambridge and NewYork: Cambridge University Press).

Mann, M. (1997) ‘The Contradictions of ContinuousRevolution’, in Stalinism and Nazism: Dictatorships inComparison, ed. I. Kershaw and M. Lewin (Cambridge andNew York: Cambridge University Press) pp. 135–57.

Mann, T. and Orren, G. (eds) (1992) Media Polls in AmericanPolitics (Washington, DC: Brookings Institution).

Manuel, P. and Cammisa, A. (1999) Checks and Balances? How aParliamentary System Could Change American Politics (Boulder,CO: Westview Press).

Manza, J., Cook, F. and Page, B. (2002) Navigating PublicOpinion: Polls, Policy and the Future of American Democracy(Oxford and New York: Oxford University Press).

March, D. and Olsen, J. (1984) ‘The New Institutionalism:Organizational Factors in Political Life’, American PoliticalScience Review (78) 778–49.

March, J. and Olsen, J. (1996) ‘Institutional Perspectives onPolitical Institutions’ Governance (9) 247–64.

Marks, G. and Diamond, L. (eds) (1992) ReexaminingDemocracy: Essays in Honour of Seymour Martin Lipset(Thousand Oaks, CA and London: Sage).

Marongiu, A. (1968) Medieval Parliaments: A Comparative Study(London: Eyre & Spottiswoode).

Marshall, T. [1950] (1991 edn) Citizenship and Social Class andOther Essays (London: Pluto Press).

Martínez-Lara, J. (1996) Building Democracy in Brazil: ThePolitics of Constitutional Change, 1985–95 (Basingstoke:Macmillan and New York: St Martin’s Press).

Marvin, C. (1988) When Old Technologies Were New: Thinkingabout Communications in the Late Nineteenth Century (Oxfordand New York: Oxford University Press).

Mark, K. and Engels, F. [1848] (1967 edn) The CommunistManifesto, intro. G. Stedman Jones (Harmondsworth:Penguin).

Massicotte, L. (2001) ‘Legislative Unicameralism: A GlobalSurvey and a Few Case Studies’, Journal of Legislative Studies(7) 68–83.

Matthews, T. (1989) ‘Interest Groups’, in Politics in Australia, ed.R. Smith and L. Watson (Sydney: Allen & Unwin) pp.211–27.

May, R. (2000) ‘The Limits of Democracy in CommonwealthAfrica’, Commonwealth & Comparative Politics (38) 171–80.

Mayall, J. (1996) The New Interventionism, 1993–1994: UnitedNations Experience in Cambodia, Former Yugoslavia andSomalia (Cambridge and New York: Cambridge UniversityPress).

338 REFERENCES

Mayhew, D. (1991) Divided We Govern: Party Control,Lawmaking and Investigations, 1946–1990 (New Haven, CTand London: Yale University Press).

Mazey, S. and Richardson, J. (eds) (1998) Interest Intermediationin the EU (London and New York: Routledge).

Mazey, S. and Richardson, J. (2001) ‘Interest Groups and EUPolicy-Making: Organizational Logic and Venue Shopping’, inEuropean Union: Power and Policy-Making, 2nd edn, ed. J.Richardson (London and New York) pp. 217–38.

Mazmanian, D. and Sabatier, P. (1989) Implementation andPublic Policy (Lanham, MD and London: University Press ofAmerica).

McAllister, I. and Studlar, D. (2002) ‘Electoral Systems andWomen’s Representation: A Long-Term Perspective’,Representation (39) 3–14.

McCargo, D. (2002) Media and Politics in Pacific Asia (Londonand New York: Routledge).

McChesney, R. (1999) Rich Media, Poor Democracies (Urbana,IL: University of Illinois Press).

McCreery, D. (2002) ‘State and Society in Nineteenth CenturyGoiás’, in Studies in the Formation of the Nation State in LatinAmerica, ed. J. Dunkerley (London: Institute of LatinAmerican Studies) pp. 133–60.

McDonald, F. and McDonald, E. (1988) ‘Federalism in America:An Obituary’, in Requiem: Variations on Eighteenth CenturyThemes ed. E. McDonald and F. MacDonald (University ofKansas Press) pp. 195–206.

McEldowney, J. (2003) ‘Administrative Law’ in The Concise OxfordDictionary of Politics, 2nd edn, ed. I. McLean and A. Macmillan(Oxford and New York: Oxford University Press) p. 3.

McKay, D. (1997) American Politics and Society, 4th edn (Oxfordand Cambridge, MA: Blackwell).

McKay, D. (2001a) Designing Europe: Comparative Lessons Fromthe European Experience (Oxford and New York: OxfordUniversity Press).

McKay, D. (2001b) American Politics and Society, 5th edn(Oxford and Cambridge, MA: Blackwell).

McKenzie, R. (1955) British Political Parties (London:Heinemann).

Meadows, D. (1994) ‘If the World Were a Village of OneThousand People’, in Futures by Design: The Practice ofEcological Planning, ed. D. Aberley (Gabriola Island, BritishColumbia: New Society).

Meier, K. (1993) ‘Representative Bureaucracy: A Theoretical andEmpirical Exposition’, Research in Public Administration (2)1–35.

Melleuish, G. ‘The State in World History: Perspectives andProblems’ (2002) Australian Journal of Politics and History (48)322–35.

Mendelsohn, M. and Parkin, A. (eds) (2001) ReferendumDemocracy: Citizens, Elites and Deliberation in ReferendumCampaigns (Basingstoke: Palgrave).

Meny, Y. and Knapp, A. (1998) Government and Politics inWestern Europe: Britain, France, Italy, West Germany, 3rd edn(Oxford and New York: Oxford University Press).

Mernissi, F. (1993) Islam and Democracy: Fear of the ModernWorld (London: Virago).

Meyer, A. (1983) ‘Cultural Revolutions: The Uses of the Conceptof Culture in Comparative Communist Studies’, Studies inComparative Communism (16) 1–8.

Mezey, M. (1979) Comparative Legislatures (Durham, NC: DukeUniversity Press).

Michels, R. [1911] (1915 edn) Political Parties (New York: DoverPublications).

Migdal, J. (2001) State in Society: Studying how States and SocietiesTransform and Constitute One Another (Cambridge and NewYork: Cambridge University Press).

Milbrath, L. and Goel, M. (1977) Political Participation: Howand Why Do People Get Involved in Politics?, 2nd edn (Chicago,IL: Rand McNally).

Mill, J. [1861] (1991 edn) ‘Considerations on RepresentativeGovernment’, in Collected Works of John Stuart Mill, Vol. 19,ed. J. O’Grady and B. Robson (Toronto: University of TorontoPress and London: Routledge) pp. 371–577.

Mill, J. S. [1861] (1972 edn) Considerations on RepresentativeGovernment, ed. H. Acton (London: Dent).

Millard, F. (2003) ‘Poland’, in Developments in Central and EastEuropean Politics 3, ed. S. White, J. Batt and P. Lewis(Basingstoke: Palgrave Macmillan) pp. 23–40.

Miller, D. (1991) ‘Politics’, in Blackwell Encyclopaedia of PoliticalThought, ed. V. Bogdanor (Oxford and Cambridge, MA:Blackwell) pp. 390–1.

Miller, W. (1995) ‘Quantitative Methods’, in Theory and Methodsin Political Science, ed. D. Marsh and G. Stoker (Basingstokeand New York: Macmillan) pp. 154–72.

Mills, C. Wright (1956) The Power Elite (New York and Oxford:Oxford University Press).

Milton, A. (2000) The Rational Politician: Exploiting the Media inNew Democracies (Aldershot: Ashgate).

Minkenberg, M. (2001) ‘The Radical Right in Public Office:Agenda-setting and Policy Effects’, West European Politics (24)1–21.

Minogue, M., Polidano, C. and Hulne, D. (eds) (1999) Beyondthe New Public Management: Changing Ideas and Practices inGovernance (Cheltenham and Northampton, MA: EdwardElgar).

Mishra, S. (1994) ‘Party Organization and Policy Making in aChanging Environment: The Indian National Congress’, inHow Political Parties Work: Perspectives from Within, ed. K.Lawson (Westport, CT: Praeger) pp. 153–79.

Moe, T. (1980) The Organization of Interests (Chicago, IL:University of Chicago Press).

Mommsen, H. (1997) ‘Cumulative Radicalization andProgressive Self-Destruction as Structural Determinants of theNazi Dictatorship’, in Stalinism and Nazism: Dictatorships inComparison, ed. I. Kershaw and M. Lewin et al. (Cambridgeand New York: Cambridge University Press) pp. 75–87.

Momoh, A. and Adejumobi, S. (2002) The National Question inNigeria: Comparative Perspectives (Aldershot: Ashgate).

Montesquieu, C.-L. (1949) The Spirit of the Laws, ed. F.Neumann (New York: Hafner). [1748] (1949 edn).

Moran, M. (2003) The British Regulatory State: High Modernismand Hyper-Innovation (Oxford and New York: OxfordUniversity Press).

Morgenstern, S. and Nacif, B. (eds) (2002) Legislative Politics inLatin America (Cambridge and New York: CambridgeUniversity Press).

Morgenthau, H. (1966) ‘The Purpose of Political Science’, in ADesign for Political Science: Scope, Objectives and Methods, ed. J.Charlesworth (Philadelphia, PA: American Academy ofPolitical and Social Science) pp. 63–79.

Morlino, L. (1995) ‘Italy’s Civic Divide’, Journal of Democracy (6)173–7.

Morton, F. (1995) ‘The Living Constitution’, in IntroductoryReadings in Canadian Government and Politics, ed. R.Wagenberg (Mississauga, Ontario: Copp Clark) pp. 41–72.

Motyl, A. (2002) ‘Imagined Communities, Rational Choices,Invented Ethnies’, Comparative Politics (34) 233–50.

REFERENCES 339

Mouritzen, P. and Svara, J. (2002) Leadership at the Apex:Politicians and Administrators in Western Local Governments(Pittsburgh, PA: University of Pittsburgh Press).

Mughan, A. (2000) Media and the Presidentialization ofParliamentary Elections (Basingstoke: Palgrave).

Mulgan, R. (1997) Politics in New Zealand, 2nd edn (Auckland:Auckland University Press).

Mulgan, R. (2003) Holding Power to Account: Accountability inModern Democracies (Basingstoke: Palgrave Macmillan).

Müller, W. and Strøm, K. (1999a) ‘Party Behaviour andRepresentative Democracy’, in Policy, Office or Votes? HowPolitical Parties in Western Europe Make Hard Decisions, ed. W.Müller and K. Strøm (Cambridge and New York: CambridgeUniversity Press) pp. 279–310.

Müller, W. and Strøm, K. (1999b) Policy, Office, or Vote? HowPolitical Parties in Western Europe Make Hard Decisions(Cambridge and New York: Cambridge University Press).

Müller, W. and Strøm, K. (eds) (2000a) Coalition Governments inWestern Europe (Oxford and New York: Oxford UniversityPress).

Müller, W. and Strøm, K. (2000b) ‘Coalition Governance inWestern Europe’, in Coalition Governments in Western Europe,ed. W. Müller and K. Strøm (Oxford and New York: OxfordUniversity Press) pp. 559–92.

Müller-Rommel, F. and Poguntke, T. (eds) (2002) Green Partiesin National Governments (London: Frank Cass).

Munro, W. (1925) The Governments of Europe (New York:Macmillan).

Myers, A. (1975) Parliaments and Estates in Europe to 1789(London: Thames & Hudson).

Myers, D. and O’Connor, R. (1998) ‘Support for Coups inDemocratic Political Culture’, in Comparative Politics (30)193–212.

■NNacif, B. (2002) ‘Understanding Party Discipline in the Mexican

Chamber of Deputies: The Centralized Party Model’, inLegislative Politics in Latin America, ed. S. Morgenstern and B.Nacif (Cambridge and New York: Cambridge University Press)pp. 254–86.

Nadeau, R., Niemi, R. and Yoshinaka, A. (2002) ‘A Cross-National Analysis of Economic Voting: Taking Account of thePolitical Context across Time and Nations’, Electoral Studies(21) 403–23.

Neto, O. (2002) ‘Presidential Cabinets, Electoral Cycles andCoalition Discipline in Brazil’, in Legislative Politics in LatinAmerica, ed. S. Morgenstern and B. Nacif (Cambridge andNew York: Cambridge University Press) pp. 48–78.

Neustadt, R. (1991) Presidential Power and the Modern Presidents(New York: Free Press).

Neustadt, R. (2001) ‘The Weakening White House’, BritishJournal of Political Science (31) 1–11.

Newell, J. and Bull, M. (2002) ‘Italian Politics After the 2002General Election: Plus ça change, plus c’est la même chose’,Parliamentary Affairs (55) 626–42.

Newton, K. (1999) ‘Social and Political Trust in EstablishedDemocracies’, in Critical Citizens: Global Support forDemocratic Governance, ed. P. Norris (Oxford and New York:Oxford University Press) pp. 169–87.

Niblock, T. (2003) Saudi Arabia (London and New York:Routledge).

Nicolaidis, K. and Howse, R. (eds) (2001) The Federal Vision:Legitimacy and Levels of Governance in the United States and the

European Union (Oxford and New York: Oxford UniversityPress).

Nikolova, K. (1998) ‘The Framework for Public ManagementReform in Bulgaria: A View From the Inside’, in Innovations inPublic Management: Perspectives from East and West Europe, ed.T. Verheijen and D. Coombes et al. (Cheltenham andNorthampton, MA: Edward Elgar) pp. 59–102.

Norden, D. (1996) Military Rebellion in Argentina: BetweenCoups and Consolidation (Lincoln, NE and London: Universityof Nebraska Press).

Norpoth, H. (1996) ‘The Economy’, in Elections and Voting inGlobal Perspective, ed. L. LeDuc, R. Niemi and P. Norris(Thousand Oaks, CA and London: Sage) pp. 299–318.

Norris, P. (1995) Comparative Models of Political Recruitment(Bordeaux: ECPR Workshop on Political Recruitment).

Norris, P. (1996) ‘Legislative Recruitment’, in Elections and Votingin Global Perspective, ed. R. Niemi and P. Norris (ThousandOaks, CA and London: Sage) pp. 184–215.

Norris, P. (1999a) ‘The Growth of Critical Citizens and ItsConsequences’, in Critical Citizens: Global Support forDemocratic Governance, ed. P. Norris (Oxford and New York:Oxford University Press) pp. 257–72.

Norris, P. (1999b) ‘Who Surfs Cafe Europa? Virtual Democracyin the USA and Western Europe’, paper presented at theAmerican Political Science Association Annual Meeting,Atlanta, GA, 1–5 September.

Norris, P. (ed.) (1999c) Critical Citizens: Global Support forDemocratic Governance (Oxford and New York: OxfordUniversity Press).

Norris, P. (2000) A Virtuous Circle: Political Communication inPostindustrial Societies (Cambridge and New York: CambridgeUniversity Press).

Norris, P. (2002) Democratic Phoenix: Reinventing PoliticalActivism (Cambridge and New York: Cambridge UniversityPress) http://www.pippanorris.com.

Norris, P. (2004) Electoral Engineering: Voting Rules and PoliticalBehaviour (Cambridge and New York: Cambridge UniversityPress).

Norris, P., Kern, M. and Just, M. (eds) (2003) Framing Terrorism:The News Media, the Government and the Public (London andNew York: Routledge).

Norton, A. (1991) ‘Western European Local Government inComparative Perspective’, in Local Government in Europe:Trends and Developments, ed. R. Batley and G. Stoker(Basingstoke: Macmillan) pp. 21–40.

Norton, P. (1990a) Legislatures (Oxford and New York: OxfordUniversity Press).

Norton, P. (1990b) Parliaments in Western Europe (London: FrankCass).

Norton, P. (1997) ‘Parliamentary Oversight’, in Developments inBritish Politics 5, ed. P. Dunleavy et al. (Basingstoke:Macmillan, and New York: St Martin’s) pp. 155–76.

Norton, P. (ed.) (1998) Parliaments and Governments in WesternEurope (London and Portland, OR: Frank Cass).

Norton, P. (2001) The British Polity, 4th edn (Harlow and NewYork: Longman).

Norton, P. and Ahmed, N. (eds) (1999) Parliaments in Asia(London and Portland, OR: Frank Cass).

Nousiainen, J. (1994) ‘Finland: Ministerial Autonomy,Constitutional Collectivism and Party Oligarchy’, in CabinetMinisters and Parliamentary Government, ed. M. Laver and K.Shepsle (Cambridge and New York: Cambridge UniversityPress) pp. 88–105.

Nousiainen, (2001) ‘From Semi-Presidentialism to Parliamentary

340 REFERENCES

Government: Political and Constitutional Developments inFinland’, Scandinavian Political Studies (24) 95–109.

Nugent, N. (2003) The Government and Politics of the EuropeanUnion, 5th edn (Basingstoke and New York: PalgraveMacmillan).

Nurmi, H. and Nurmi, L. (2002) ‘The 2000 PresidentialElection in Finland’, Electoral Studies (21) 473–9.

Nye, J. (1990) Bound to Lead: the Changing Nature of America’sPower (New York: Basic Books).

Nye, J., Zelikow, P. and King, D. (eds) (1997) Why People Don’tTrust Government (Cambridge, MA: Harvard University Press).

Nye, N. (2002) The Paradox of American Power: Why the World’sOnly Superpower Can’t Go It Alone (Oxford and New York:Oxford University Press).

■OO’Brien, D. (1993) Storm Center: The Supreme Court in American

Politics (New York: Norton).O’Brien, K. (1990) Reform Without Liberalization: China’s

National People’s Congress and the Politics of Institutional Change (Cambridge and New York: Cambridge UniversityPress).

O’Donnell, G. (1973) Modernization and BureaucraticAuthoritarianism: Studies in South American Politics (Berkeley,CA: California University Press).

O’Donnell, G. (1994) ‘Delegative Democracy’, Journal ofDemocracy (5) 55–69.

O’Donnell, G. (1996) ‘Delegative Democracy’, in The GlobalResurgence of Democracy, 2nd edn, ed. L. Diamond and M.Plattner (Baltimore, MD and London: Johns HopkinsUniversity Press) pp. 94–110.

O’Donnell, G. and Schmitter, P. (1986a) ‘Tentative Conclusionsabout Uncertain Democracies’, in Transitions fromAuthoritarian Rule: Prospects for Democracy, ed. G. O’Donnell,P. Schmitter and L. Whitehead (Baltimore, MD and London:Johns Hopkins University Press) pp. 1–65.

O’Donnell, G. and Schmitter, P. (1986b) Transitions fromAuthoritarian Rule: Tentative Conclusions from UncertainDemocracies (Berkeley and Los Angeles, CA: University ofCalifornia Press).

O’Donnell, G., Schmitter, P. and Whitehead, L. (eds) (1986)Transitions from Authoritarian Rule (Baltimore, MD andLondon: Johns Hopkins University Press).

O’Sullivan, N. (1986) Fascism (London: Dent).O’Toole, B. and Chapman, R. (1995) ‘Parliamentary

Accountability’, in Next Steps: Improving Management inGovernment?, ed. B. O’Toole and G. Jordan (Aldershot andBrookfield, VT: Dartmouth) pp. 118–41.

Oksenberg, M. (2001) ‘The Issue of Sovereignty in the AsianHistorical Context’, in Problematic Sovereignty: Contested Rulesand Political Possibilities, ed. S. Krasner (New York andChichester: Columbia University Press) pp. 83–104.

Olson, D. (1994) Legislative Institutions: A Comparative View(New York: M.E. Sharpe).

Olson, M. (1968) The Logic of Collective Action: Public Goods andthe Theory of Groups (New York: Schocken Books).

Olson, M. (1982) The Rise and Decline of Nations (New Haven,CT and London: Yale University Press).

Olson, M. (2000) Power and Prosperity: Outgrowing Communistand Capitalist Dictatorships (New York: Basic Books).

Ong, W. (1982) Orality and Literacy: The Technologizing of theWord (London and New York: Methuen).

Opello, W. and Rosow, S. (1999) The Nation-State and Global

Order: A Historical Introduction to Contemporary Politics(Boulder, CO and London: Lynne Rienner).

Osborne, D. and Gaebler, T. (1992) Reinventing Government:How the Entrepreneurial Spirit Is Transforming the Public Sector(New York and London: Penguin).

Osiander, A. (2001) ‘Sovereignty, International Relations and theWestphalian Myth’, International Organization (55) 251–89.

Ostrogorski, M. (1902) Democracy and the Organisation ofPolitical Parties (London: Macmillan).

Owen, J. (2002) ‘The Foreign Imposition of DomesticInstitutions’ International Organization (50) 375–409.

Owen, R. (1993) ‘The Practice of Electoral Democracy in theArab East and North Africa: Some Lessons from Nearly aCentury’s Experience’ in Rules and Rights in the Middle East,ed. E. Goldberg et al. (Seattle, WA: University of WashingtonPress) pp. 17–40.

Owen, R. (2002) State, Power and Politics in the Middle East, 2ndedn (London and New York: Routledge).

■PPadgett, S. (2000) Organizing Democracy in Eastern Germany:

Interest Groups in Post-Communist Society (Cambridge andNew York: Cambridge University Press).

Page, E. (1992) Political Authority and Bureaucratic Power: AComparative Analysis, 2nd edn (Hemel Hempstead: HarvesterWheatsheaf ).

Page, E. and Goldsmith, M. (1987) Central and LocalGovernment Relations (London and Thousand Oaks, CA:Sage).

Page, E. and Wright, V. (eds) (1999) Bureaucratic Elites in WesternEuropean States: A Comparative Analysis of Top Officials(Oxford and New York: Oxford University Press).

Paine, T. [1791/2] (1984 edn) Rights of Man (Harmondsworth:Penguin).

Palan R. (2002) ‘Tax Havens and the Commercialization of StateSovereignty’ International Organization (56) 151–76.

Panebianco, A. (1988) Political Parties: Organization and Power(Cambridge and New York: Cambridge University Press).

Parkinson, J. (2001) ‘Who Knows Best? The Creation of theCitizen-Initiated Referendum in New Zealand’, Government &Opposition (36) 403–421.

Parry, G., Moyser, G. and Day, N. (1992) Political Participationand Democracy in Britain (Cambridge and New York:Cambridge University Press).

Parsons, T. (1967) ‘On the Concept of Political Power’, inSociological Theory and Modern Society, ed. T. Parsons (NewYork and London: Free Press) pp. 286–99.

Parsons, W. (1995) Public Policy: An Introduction to the Theoryand Practice of Policy Analysis (Brookfield, VT and Aldershot:Edward Elgar).

Partridge, H. (1998) Italian Politics Today (Manchester:Manchester University Press).

Patterson, S. and Mughan, A. (eds) (1999) Senates: Bicameralismin the Contemporary World (Columbus, OH: Ohio StateUniversity Press).

Paul, T. (ed.) (2004) The Nation-State in Question (Princeton, NJ:Princeton University Press).

Pennings, P., Keman, H. and Kleinnijenhuis, J. (1999) DoingResearch in Political Science: An Introduction to ComparativeMethods and Statistics (Thousand Oaks, CA and London:Sage).

Peregudov, S. (2001) ‘The Oligarchical Model of RussianCapitalism’, in Contemporary Russian Politics: A Reader, ed. A.

REFERENCES 341

Brown (Oxford and New York: Oxford University Press) pp.259–68.

Peretti T. (2001) In Defence of a Political Court (Princeton, NJ:Princeton University Press).

Perlmutter, A. (1981) Modern Authoritarianism (New Haven,CT: Yale University Press).

Perlmutter, A. (1997) Making the World Safe for Democracy: ACentury of Wilsonianism and its Totalitarian Challenges (ChapelHill, NC: University of North Carolina Press).

Peters, B. Guy (1998) Comparative Politics: Theory and Methods(London: Macmillan and New York University Press).

Peters, B. Guy (1999) Institutional Theory in Political Science: The‘New Institutionalism’ (London and New York: Pinter).

Peters, B. Guy (2000) The Politics of Bureaucracy, 5th edn (WhitePlains, New York: Longman).

Petersson, O. (1989) Maktens Natverk (Stockholm: Carlssons).Petracca, M. (1992) ‘The Rediscovery of Interest Group Politics’,

in The Politics of Interests: Interest Groups Transformed, ed. M.Petracca (Boulder, CO and Oxford: Westview) pp. 3–31.

Petro, N. (1995) The Rebirth of Russian Democracy: AnInterpretation of Political Culture (Cambridge, MA andLondon: Harvard University Press).

Pharr, S. and Putnam, R. (eds) (2000) Disaffected Democracies:What’s Troubling The Trilateral Countries? (Princeton, NJ:Princeton University Press).

Phillips, A. (1995) The Politics of Presence (Oxford and New York:Oxford University Press).

Pierre, J. and Peters, B. Guy (2000) Governance, Politics and theState (Basingstoke: Macmillan and New York: St Martin’s Press).

Pierson, C. (1998) Beyond the Welfare State? The New PoliticalEconomy of Welfare, 2nd edn (University Park: PennsylvaniaState University Press and Cambridge: Polity).

Pierson, C. (ed.) (2000) The New Politics of the Welfare State(Oxford and New York: Oxford University Press).

Pinkney, R. (1990) Right-Wing Military Government (London:Pinter).

Pion-Berlin, D. (ed.) (2001) Civil–Military Relations in LatinAmerica (Chapel Hill, NL: University of North CarolinaPress).

Plender, J. (2003) ‘Wheezing Through the Fiscal Fog’, FinancialTimes, 4 April, p. 17.

Poggi, G. (1978) The Development of the Modern State: ASociological Introduction (London: Hutchinson).

Poggi, G. (1990) The State: Its Nature, Development and Prospects(Cambridge and Cambridge, MA: Polity).

Poggi, G. (2001) Forms of Power (Oxford and Cambridge, MA:Blackwell).

Polidano, C. (1998) ‘Why Bureaucrats Can’t Always Do WhatMinisters Want: Multiple Accountabilities in WestminsterDemocracies’, Public Policy and Administration (13) 35–50.

Pomper, G. et al. (2001) The Election of 2000: Reports andInterpretations (Chatham, NJ: Chatham House).

Power, T. (1998) ‘The Pen Is Mightier than the Congress:Presidential Decree Power in Brazil’ in Executive DecreeAuthority, ed. J. Carey and M. Shugart (Cambridge and NewYork: Cambridge University Press) pp. 197–232.

Pressman, J. and Wildavsky, A. (1973) Implementation (Berkeley,CA: University of California Press).

Pridham, G. (1990) ‘Political Parties and Consolidation inSouthern Europe: Empirical and Theoretical Perspectives’, inParliament and Democratic Consolidation in Southern Europe,ed. U. Liebert and M. Cotta (London: Pinter) pp. 225–48.

Pridham, G. (ed.) (1995) Transitions to Democracy (Brookfield,VT and Aldershot: Dartmouth).

Prillaman, W. (2000) The Judiciary and Democratic Decay inLatin America: Declining Confidence in the Rule of Law(Westport, CT: Praeger).

Prosser, T. and Moran, M. (1994) ‘Privatization and RegulatoryChange: The Case of Great Britain’, in Privatization andRegulatory Change in Europe, ed. M. Moran and T. Prosser(Buckingham and Bristol, PA: Open University Press) pp.35–49.

Prunier, G. (1997) The Rwanda Crisis: History of a Genocide(London: Hurst).

Przeworski, A. (1991) Democracy and the Market: Political andEconomic Reforms in Eastern Europe and Latin America(Cambridge and New York: Cambridge University Press).

Przeworski, A. (1995) ‘The Role of Theory in ComparativePolitics’, World Politics (48) 16–21.

Przeworski, A. and Teune, H. (1970) The Logic of ComparativeInquiry (New York: Wiley).

Przeworski, A. et al. (2000) Democracy and Development(Cambridge and New York: Cambridge University Press).

Putnam, R. (1976) The Comparative Study of Political Elites(Englewood Cliffs, NJ: Prentice Hall).

Putnam, R. (1993) Making Democracy Work: Civic Traditions inModern Italy (Princeton, NJ: Princeton University Press).

Putnam, R. (1995) ‘Bowling Alone: America’s Declining SocialCapital’, Journal of Democracy (6) 65–78.

Putnam, R. (2000) Bowling Alone: The Collapse and Revival ofAmerican Community (New York: Simon & Schuster).

Putnam, R. (2001) ‘Civic Engagement in ContemporaryAmerica’, Government and Opposition (30) 135–56.

Putnam, R. (ed.) (2002) Democracies in Flux: The Evolution ofSocial Capital in Contemporary Society (Oxford and New York:Oxford University Press).

Putnam, R., Pharr, S. and Dalton, R. (2000) ‘What’s Troublingthe Trilateral Countries?’, in Disaffected Democracies: What’sTroubling the Trilateral Countries? ed. S. Pharr and R. Putnam(Princeton, NJ: Princeton University Press) pp. 3–30.

Putzel, J. (1997) ‘Why Has Democratization Been a WeakerImpulse in Indonesia and Malaysia than in the Philippines?’,in Democratization, ed. D. Potter et al. (Cambridge: Polity) pp.240–68.

Pye, L. (1985) Asian Power and Politics: The Cultural Dimensionsof Authority (Cambridge, MA: Harvard University Press).

Pye, L. (1995) ‘Political Culture’, in The Encyclopaedia ofDemocracy, ed. S. Lipset (London and New York: Routledge)pp. 965–9.

■QQualter, T. (1991) ‘Public Opinion’, in The Blackwell

Encyclopaedia of Political Science, ed. V. Bogdanor (Oxford andCambridge, MA: Blackwell) p. 511.

■RRaadschelders, J. and Rutgers, M. (1996) ‘The Evolution of Civil

Service Systems’, in Civil Service Systems in ComparativePerspective, ed. H. Bekke, J. Perry and T. Toonen(Bloomington, IN: Indiana University Press) pp. 67–99.

Ragin, C. (1994a) ‘Introduction to Qualitative ComparativeAnalysis’, in The Comparative Political Economy of the WelfareState, ed. T. Janoski and A. Hicks (New York and Cambridge:Cambridge University Press) pp. 299–319.

Ragin, C. (1994b) Constructing Social Research (Thousand Oaks,CA and London: Sage).

342 REFERENCES

Ragin, C., Berg-Schlosser, D. and de Meur, G. (1996) ‘PoliticalMethodology: Qualitative Methods’, in A New Handbook ofPolitical Science, ed. R. Goodin and H. Klingemann (Oxfordand New York: Oxford University Press) pp. 749–68.

Ramseyer, J. and Rasmusen, E. (2001) ‘Why Are Japanese Judgesso Conservative in Politically Charged Cases?’, AmericanPolitical Science Review (95) 331–44.

Ranney, A. (2004) ‘Politics in the United States’, in ComparativePolitics Today: A World View, 8th ed., ed. G. Almond et al.(New York and London: Pearson Longman) pp. 742–86.

Reed, S. and Thies, M. (2002) ‘The Causes of Electoral Reformin Japan’, in Mixed-Member Electoral Systems: The Best of BothWorlds, ed. M. Shugart and M. Wattenberg (Oxford and NewYork: Oxford University Press) pp. 152–172.

Remington, T. (2004) ‘Politics in Russia’, in Comparative PoliticsToday: A World View, 8th edn, ed. G. Almond et al. (New Yorkand London: Pearson Longman) pp. 366–417.

Remmer, K. (1989) Military Rule in Latin America (Boston, MA:Unwin Hyman).

Rhodes, R. (1994) ‘State-building Without a Bureaucracy: The Case of the United Kingdom’, in Developing Democracy:Essays in Honour of J. F. P. Blondel, ed. I. Budge and D. McKay (Thousand Oaks, CA and London: Sage) pp.165–88.

Rhodes, R. (1995) ‘The Institutional Approach’, in Theory andMethods in Political Science, ed. D. Marsh and G. Stoker(Basingstoke: Macmillan) pp. 42–57.

Rhodes, R. (1996) ‘The New Governance: Governing withoutGovernment’, Political Studies (44) 652–67.

Rich, R. (2001) ‘Bringing Democracy into International Law’,Journal of Democracy (12) 20–34.

Ridley, F. (1975) The Study of Government: Political Science andAdministration (London: Allen & Unwin).

Riesman, D. (1950) The Lonely Crowd (New Haven, CT: YaleUniversity Press).

Riker, W. (1962) The Theory of Political Coalitions (New Haven,CT and London: Yale University Press).

Riker, W. (1975) ‘Federalism’, in The Handbook of PoliticalScience, Vol. 5, ed. F. Greenstein and N. Polsby (Reading, MA:Addison-Wesley) pp. 93–172.

Riker, W. (1996) ‘European Federalism: The Lessons of PastExperience’, in Federalizing Europe? The Costs, Benefits andPreconditions of Federal Political Systems, ed. J. Hesse and V.Wright (Oxford and New York: Oxford University Press) pp.9–24.

Ripley, R. and Franklin, G. (1991) Congress, the Bureaucracy andPublic Policy, 5th edn (Homewood IL: The Dorsey Press).

Roberts, G. (2002) ‘The Politics of Legislative Membership inthe Federal Republic of Germany’, Government & Opposition(37) 231–49.

Robinson, M. and White, G. (eds) (1998) The DemocraticDevelopmental State: Political and Institutional Design (Oxfordand New York: Oxford University Press).

Rogers, E. (1971) Communication of Innovation: A Cross-culturalApproach (New York: Free Press).

Rokkan, S. (1970) Citizens, Elections, Parties (New York: McKay).Rose, R. (1989) Politics in England: Change and Persistence

(Basingstoke: Macmillan).Rose, R. (1991) ‘Comparing Forms of Comparative Analysis’,

Political Studies (39) 446–62.Rose, R. (2000) The Post-Modern President, 2nd edn (Chatham,

NJ: Chatham House).Rose, R. (2001) The Prime Minister in a Shrinking World

(Cambridge: Polity).

Rose, R. (2004) Learning Lessons in Comparative Public Policy(London and New York: Routledge).

Rose, R. and Munro, N. (2002) Elections Without Order: Russia’sChallenge to Vladimir Putin (Cambridge and New York:Cambridge University Press).

Rose, R. and Shin, D. (2001) ‘Democratization Backwards: TheProblem of Third-Wave Democracies’, in British Journal ofPolitical Science (31) 331–54.

Rosenau, J. (1992) ‘Governance, Order and Change in WorldPolitics’, in Governance without Government: Order and Changein World Politics, ed. J. Rosenau and E.-O. Czempiel(Cambridge and New York: Cambridge University Press) pp.3–6.

Rosenbluth, F. and Thies, M. (2000) ‘Politics in Japan’, inComparative Politics Today: A World View, 7th edn, ed. G.Almond et al. (New York: Longman) pp. 327–72.

Rothstein, B. (2002) ‘Sweden: Social Capital in the SocialDemocratic State’, in Democracies in Flux: The Evolution ofSocial Capital in Contemporary Society, ed. R. Putnam (Oxfordand New York: Oxford University Press) pp. 289–332.

Rousseau, D. (1994) ‘The Constitutional Judge: Master or Slave ofthe Constitution?’, in Constitutionalism, Identity, Difference andLegitimacy: Theoretical Perspectives, ed. M. Rosenfeld (Durham,NC and London: Duke University Press) pp. 261–83.

Roy, O. (1994) The Failure of Political Islam (London: I.B.Taurus).

Rudd, C. and Hayward, J. (2001) ‘Coverage of Post-war ElectionCampaigns: The Otago Daily Times’, paper presented at theNew Zealand Political Studies Conference, Massey University,December 2001.

Rueschmeyer, D., Rueschmeyer, M. and Wittrock, B. (eds)(1998) Participation and Democracy: Comparisons andInterpretations (Armonk, New York and London: M.E.Sharpe).

Rummel, R. (1997) Death by Government (New Brunswick, NJand London: Transaction Books).

Russell, M. (2000a) Reforming the House of Lords: Lessons fromOverseas (Oxford and New York: Oxford University Press).

Russell, M. (2000b) ‘A “More Democratic and Representative”Upper House? Some International Comparisons’,Representation (37) 131–8.

Russell, M. (2001) ‘What Are Second Chambers For?’,Parliamentary Affairs (54) 442–58.

■SSaalfeld, T. (2000) ‘Germany: Stable Parties, Chancellor

Democracy and the Art of Informal Settlement’, in CoalitionGovernments in Western Europe, ed. W. Müller and K. Strøm(Oxford and New York: Oxford University Press) pp. 32–85.

Sabatier, P. (1999) Theories of the Policy Process (Boulder, CO:Westview).

Sabatier, P. and Jenkins-Smith, H. (1993) Policy Change andLearning: An Advocacy Coalition Approach (Boulder, CO:Westview).

Saich, A. (2004) Governance and Politics of China, 2nd edn(Basingstoke and New York: Palgrave).

Saikal, A. (2003) Islam and the West: Conflict or Cooperation?(Basingstoke and New York: Palgrave Macmillan).

Sait, E. (1938) Political Institutions: A Preface (New York:Appleton-Century).

Sakwa, R. (2000) ‘Russia’s Permanent Uninterrupted Elections of1999–2000’, Journal of Communist and Transition Studies, (16)85–112.

REFERENCES 343

Sandbrook R. (1985) The politics of Africa’s Economic Stagnation(Cambrige and New York: Cambridge University Press).

Sanderson, I. (2002) ‘Evaluation, Policy Learning and Evidence-Based Policy Making’, Public Administration (80) 1–22.

Sanderson, J. (1961) ‘The National Smoke Abatement Societyand the Clean Air Act (1956)’, Political Studies (9) 236–53.

Sartori, G. (1976) Parties and Party Systems: A Framework forAnalysis (Cambridge and New York: Cambridge UniversityPress).

Sartori, G. (1993) ‘Totalitarianism: Model Mania and Learningfrom Error’, Journal of Theoretical Politics (5) 5–22.

Sartori, G. (1994) Comparative Constitutional Engineering: AnInquiry into Structures, Incentives and Outcomes (Basingstoke:Macmillan).

Sassen, S. (ed.) (2002) Global Networks, Linked Cities (Londonand New York: Routledge).

Scarrow, H. (1969) Comparative Political Analysis (New York:Harper & Row).

Scarrow, S. (2002a) ‘Party Decline in the Parties State? TheChanging Environment of German Politics’, in Political Partiesin Advanced Industrial Democracies, ed. P. Webb, D. Farrell andI. Holliday (Oxford and New York: Oxford University Press)pp. 77–106.

Scarrow. S. (2002b) ‘Germany: The Mixed-Member System as aPolitical Compromise’, in Mixed-Member Electoral Systems: TheBest of Both Worlds, ed. M. Shugart and M. Wattenberg(Oxford and New York: Oxford University Press) pp. 55–69.

Schabas, W. (2001) An Introduction to the International CriminalCourt (Cambridge and New York: Cambridge UniversityPress).

Schain, M. (2004) ‘Politics in France’, in Comparative PoliticsToday: A World View, 8th edn, ed. G. Almond et al. (New Yorkand London: Pearson Longman) pp. 206–59.

Schaller, T. and Williams, T. (2003) ‘The ContemporaryPresidency: Postpresidential Influence in the Postmodern Era’,Presidential Studies Quarterly (33) 188–200.

Scharpf, F. (1988) ‘The Joint Decision Trap: Lessons FromGerman Federalism and European Integration’, PublicAdministration (66) 239–78.

Schattschneider, E. (1942) Party Government (New York: Farrar& Reinhart).

Schatzberg, M. (2001) Political Legitimacy in Middle Africa:Father, Family, Food (Bloomington, IN: Indiana UniversityPress).

Schirazi, A. (1998) The Constitution of Iran: Politics and the Statein the Islamic Republic (London: L.-B. Taurus).

Schmidt, M. (2003) Political Institutions in the Federal Republic ofGermany (Oxford and New York: Oxford University Press).

Schmieding, H. (1993) Europe after Maastricht (London:Institute of Economic Affairs).

Schmieding, H. (1998) ‘From Plan to Market: On the Nature ofthe Transformation Crisis’, Weltwirtschaftliches Archiv (129)216–53.

Schmitter, P. and Lehmbruch, G. (eds) (1979) Trends TowardsCorporatist Intermediation (Thousand Oaks, CA and London:Sage).

Schöpflin, G. and Hosking, G. (eds) (1997) Myths andNationhood (London: Hurst).

Schubert, G. (1972) ‘Judicial Process and Behaviour During theSixties’, Political Science (5) 6–15.

Schumpeter, J. (1943) Capitalism, Socialism and Democracy(London: Allen & Unwin).

Scott, S. (1997) ‘Australia and International Institutions’, in NewDevelopments in Australian Politics, ed. B. Galligan, I.

McAllister and J. Ravenhill (South Melbourne: Macmillan)pp. 271–90.

Scully, R. (2003) ‘The European Parliament’, in European UnionPolitics, ed. M. Cini (Oxford and New York: OxfordUniversity Press) pp. 166–78.

Searing, D. (1994) Westminster’s World: Understanding PoliticalRoles (Cambridge, MA and London: Harvard UniversityPress).

Selden, M. (1995) China in Revolution: The Yenan Way Revisited(Armonk, New York: M.E. Sharpe).

Senelle, R. (1996) ‘The Reform of the Belgian State’, inFederalizing Europe? The Costs, Benefits and Preconditions ofFederal Political Systems, ed. J. Hesse and V. Wright (Oxfordand New York: Oxford University Press) pp. 266–324.

Shanks, C., Jackobson, H. and Kaplan, J. (1996) ‘Inertia andChange in the Constellation of IntergovernmentalOrganizations’, International Organization (50) 593–627.

Shapiro, M. [1964] (2002 edn) ‘Political Jurisprudence’, in OnLaw, Politics and Judicialization, ed. M. Shapiro and A. StoneSweet (Oxford and New York: Oxford Univeristy Press) pp.19–54.

Shapiro, M. (1990) ‘The Supreme Court from Early Burger toEarly Rehnquist’ in The New American Political System, 2ndedn, ed. A. King (Washington: American Enterprise Institute)pp. 47–86.

Shapiro, M. and Stone, A. (1994) ‘The New ConstitutionalPolitics of Europe’, Comparative Political Studies (26)397–420.

Shapiro, M. and Stone Sweet, A. (2002) On Law, Politics andJudicialization (Oxford and New York: Oxford UniversityPress).

Sharlet, R. (1997) ‘The Progress of Human Rights’, inDevelopments in Russian Politics 4, ed. S. White, A. Pravda andZ. Gitelman (Basingstoke: Macmillan) pp. 129–48.

Shively, W. (2002) Power and Choice: An Introduction to PoliticalScience, 8th edn (New York and London: McGraw-Hill).

Shugart, M. and Carey, J. (1992) Presidents and Assemblies:Constitutional Design and Electoral Dynamics (Cambridge andNew York: Cambridge University Press).

Shugart, M. and Wattenberg, M. (eds) (2000) Mixed-MemberElectoral Systems: The Best of Both Worlds (Oxford and NewYork: Oxford University Press).

Silva, P. (ed.) (2001) The Soldier and the State in South America(London: Palgrave).

Simeon, R. and Cameron, D. (2002) ‘IntergovernmentalRelations and Democracy: An Oxymoron If Ever There WasOne’, in Canadian Federalism: Performance, Effectiveness andLegitimacy, ed. H. Bakvis and G. Skogstad (Don Mills,Ontario and Oxford: Oxford University Press) pp. 278–95.

Simon, H. (1983) Reason in Human Affairs (Oxford andCambridge, MA: Blackwell).

Sjolin, M. (1993) Coalition Politics and Parliamentary Power(Lund: Lund University Press).

Skinner, Q. (1978) The Foundations of Modern Political Thought,Vol. 1 (Cambridge and New York: Cambridge UniversityPress).

Sklair, L. (2003) The Transnational Capitalist Class (Oxford andCambridge, MA: Blackwell).

Skocpol, T. (1979) States and Social Revolutions: A ComparativeAnalysis of France, Russia and China (Cambridge and NewYork: Cambridge University Press).

Slaughter, A. (1997) ‘The Real New World Order’, Foreign Affairs(76) pp. 183–94.

Smith, A. (1998) Nationalism and Modernism: A Critical Survey of

344 REFERENCES

Recent Theories of Nations and Nationalism (London and NewYork: Routledge).

Smith, B. (1996) Understanding Third World Politics (London:Macmillan).

Smith, G. (1989) Politics in Western Europe, 5th edn (Aldershot:Gower).

Smith, G. (1995) Federalism: The Multiethnic Challenge (Harlowand New York: Longman).

Smith, M. (1995) Pressure Politics (Manchester: Baseline Books). Smooha, S. (2002) ‘The Model of Ethnic Democracy’, Nations

and Nationalism (8) 475–503.Soley, L. (1998) Free Radio: Electronic Civil Disobedience

(Boulder, CO and Oxford: Westview).Solomon, P. and Foglesong, T. (2000) Courts and Transition in

Russia: The Challenge of Judicial Reform (Boulder, CO andOxford: Westview).

Somit, A. (1994) ‘. . . And Where We Came Out’, in TheVictorious Incumbent: A Threat to Democracy?, ed. A. Somit etal. (Aldershot and Brookfield, VT: Dartmouth) pp. 11–18.

Sørensen, G. (1997a) ‘An Analysis of Contemporary Statehood:Consequences for Conflict and Cooperation’, Review ofInternational Studies (23) 253–70.

Sørensen, G. (1997b) Democracy and Democratization, 2nd edn(Boulder, CO and Oxford: Westview).

Sørensen, G. (2004) The Transformation of the State: Beyond theMyth of Retreat (Basingstoke and New York: PalgraveMacmillan).

Spruyt, H. (1994) The Sovereign State and Its Competitors(Princeton, NJ: Princeton University Press).

Stainov, P. (1993) Durzhava i Chinovnichestvo (Sofia: SlavianskaBeseda).

Steen, A. (1995) Change of Regime and Political Recruitment: TheParliamentary Elites in the Baltic States (Bordeaux: ECPRWorkshop on Political Recruitment).

Steinmo, S. (2003) ‘The Evolution of Policy Ideas: Tax Policy inthe Twentieth Century’, British Journal of Politics andInternational Relations (5) 206–36.

Stepan, A. (2001) Arguing Comparative Politics (Oxford and NewYork: Oxford University Press).

Stevens, A. (2003) Government and Politics of France 3rd edn(Basingstoke and New York: Palgrave Macmillan).

Stoliarov, M. (2002) Federalism and the Dictatorship of Power inRussia (London and New York: Routledge).

Stone, B. (1995) ‘Administrative Accountability in the“Westminster” Democracies: Towards a New ConceptualFramework’, Governance (8) 505–26.

Stone, D. (2001) Policy Paradox: The Art of Political DecisionMaking, rev. edn (New York: Norton).

Stone Sweet, A. (2000) Governing with Judges: Constitutional Politicsin Europe (Oxford and New York: Oxford University Press).

Stone Sweet, A. (2002) ‘Constitutional Courts and ParliamentaryDemocracy’, West European Politics (25) 77–110.

Stouffer, S. (1966) Communism, Conformity and Civil Liberties(New York: Wiley).

Strange, S. (1994) States and Markets (London and New York:Pinter).

Strayer, J. (1965) Feudalism (Princeton, NJ: Princeton UniversityPress).

Street, J. (2001) Mass Media, Politics and Democracy (Basingstokeand New York: Palgrave).

Strom, K. (1990) Minority Government and Majority Rule(Cambridge and New York: Cambridge University Press).

Stubbs, R. and Underhill, R. (eds) (1999) Political Economy andthe Changing Global Order, 2nd edn (Basingstoke: Macmillan).

Sundberg, J. (2002) ‘The Scandinavian Party Model at the Crossroads’, in Political Parties in Advanced Industrial Democracies, ed. P. Webb, D. Farrell and I. Holliday(Oxford and New York: Oxford University Press) pp.181–216.

Sweating, D. (2003) ‘How Strong Is the Mayor of London?’,Policy and Politics (31) 465–78

■TTaagepera, R. and Recchia, S. (2002) ‘The Size of Second

Chambers and European Assemblies’, European Journal ofPolitical Research (41) 165–85.

Tálos, E. and Kittel B. (2002) }Austria in the 1990s: TheRoutine of Social Partnership in Question?’ in PolicyConcertation and Social Partnership: Lessons for the 21stCentury, ed. S. Berger and H. Compston (New York andOxford: Berghahn) pp. 35–50.

Taras, R. (2003) ‘Executive Leadership: Presidents andGovernments’, in Developments in Central and East EuropeanPolitics 3, ed. S. White S., J. Batt and P. Lewis (Basingstoke:Palgrave Macmillan) pp. 115–31.

Tarrow, S. (1998) Power in Movement: Social Movements andContentious Politics, 2nd edn (Cambridge and New York:Cambridge University Press).

Tate, C. and Vallinder, T. (eds) (1995) The Global Expansion ofJudicial Power (New York and London: New York UniversityPress).

Taylor, C. (1993) Reconciling the Solitudes: Essays on CanadianFederalism and Nationalism (Montreal and Kingston: McGill-Queen’s University Press).

Telhami, S. (2001) ‘The Road to Palestinian Sovereignty:Problematic Structures or Conventional Obstacles?’, inProblematic Sovereignty: Contested Rules and PoliticalPossibilities, ed. S. Krasner (New York and Chichester:Columbia University Press) pp. 301–22.

Tetlock, P. and Belkin, A. (1996) Counterfactual ThoughtExperiments in World Politics (Princeton, NJ and London:Princeton University Press).

Teune, H. (1995a) ‘Preface’, Annals of the American Academy ofPolitical and Social Sciences (540) 8–10.

Teune, H. (1995b) ‘Local Government and Democratic PoliticalDevelopment’, Annals of the American Academy of Political andSocial Sciences (540) 11–23.

Thatcher M. (1988) Speech to the College of Europe (London:Margaret Thatcher Foundation). http://www.margaretthatcher.org.

Thatcher, M. and Stone Sweet, A. (2002) ‘Theory and Practice ofDelegation to Non-Majoritarian Institutions’, West EuropeanPolitics (25) 1–22.

Thomas, B. (2003) ‘What the Poor Watch on TV’, Prospect(January) 46–51.

Thomas, C. (1997) ‘Globalization and the South’, inGlobalization and the South, ed. C. Thomas and P. Wilkin(Basingstoke: Macmillan) pp. 1–17.

Thomas, C. (ed.) (2001) Political Parties and Interest Groups:Shaping Democratic Goverance (Boulder CO and London:Lynne Rienner).

Thompson M. (2001) ‘Whatever Happened to “Asian Values”?’,Journal of Democracy (12) 154–65.

Tilly, C. (1975) ‘Reflections on the History of European State-Making’, in The Formation of National States in WesternEurope, ed. C. Tilly (Princeton, NJ: Princeton UniversityPress) pp. 3–83.

REFERENCES 345

Tilly, C. (1997) ‘Means and Ends of Comparison inMacrosociology’, Comparative Social Research (16) 43–53.

Tismaneanu, V. (ed.) (1995) Political Culture and Civil Society inRussia and the New States of Eurasia (New York: M.E. Sharpe).

Tivey, L. (ed.) (1981) The Nation-State: The Formation of ModernPolitics (London: Martin Robertson).

Tracey, M. (1998) The Decline and Fall of Public ServiceBroadcasting (New York and Oxford: Oxford University Press).

Trotsky, L. [1932/3] (1965 edn) The History of the RussianRevolution, Vol. 1, trans. M. Eastman (London: Gollancz) p.216.

Tsebelis, G. and Garrett, G. (1997) ‘Agenda Setting, Vetoes andthe European Union’s Co-Decision Procedure’, Journal ofLegislative Studies (3) 74–92.

Tsebelis, G. and Money, J. (1997) Bicameralism (Cambridge andNew York: Cambridge University Press).

Turner, F. (1995) ‘Reassessing Political Culture’, in Latin Americain Comparative Perspective: New Approaches to Methods andAnalysis, ed. P. Smith (Boulder, CO and Oxford: Westview)pp. 195–224.

Turner, M. and Hulme, D. (1997) Governance, Administrationand Development (London: Macmillan).

■UUlam, A. (1974) The Russian Political System (New York:

Random House).Uleri, P. (2002) ‘On Referendum Voting in Italy: Yes, No or

Non-Vote? How Italian Parties Learned to ControlReferendums’, European Journal of Political Research (41)863–883.

Umbach, M. (ed.) (2002) German Federalism: Past, Present,Future (Basingstoke: Macmillan).

UNHCR (United nations High Commission for Human Rights)(1966) United Nations Covenant on Civil and Political Rights(http://www.unhcr.ch).

US Congress (2002) Report of the Joint Enquiry into the TerroristAttacks of September 11, 2001 (Washington, DC: USCongress) http://usinfo.state.gov/topical/pol/terror.

■Vvan Biezen, I. (2003) Political Parties in New Democracies: Party

Organization in Southern and East–Central Europe (Basingstokeand New York: Palgrave Macmillan).

van Creveld, M. (1999) The Rise and Decline of the State(Cambridge and New York: Cambridge University Press).

van der Meer, F., and Roborgh, R. (1996) ‘Civil Servants andRepresentativeness’, in Civil Service Systems in ComparativePerspective, ed. H. Bekke, J. Perry and T. Toonen(Bloomington, IN: Indiana University Press) pp. 119–33.

van Deth, J. (2000) ‘Interesting but Irrelevant: Social Capital andthe Salience of Politics in Western Europe’, European Journal ofPolitical Research (37) 115–147.

van Eijk, R. (1997) ‘The United Nations and the Reconstructionof Collapsed States’, African Journal of International andComparative Law (9) 543–72.

Vanhanen, T. (1997) Prospects of Democracy: A Study of 172Countries (London and New York: Routledge).

Vatikiotis, M. (1995) Political Change in Southeast Asia: Trimmingthe Banyan Tree (London and New York: Routledge).

Verba, S. (1987) Elites and the Idea of Equality: A Comparison ofJapan, Sweden and the United States (Cambridge, MA andLondon: Harvard University Press).

Verba, S., Scholzman, K. and Brady, H. (1995) Voice andEquality: Civic Voluntarism in American Politics (Cambridge,MA and London: Harvard University Press).

Vereshchetin, V. (1996) ‘New Constitutions and the Old Problemof Relationship between International Law and National Law’,European Journal of International Law (7) 29–41.

Verheijen, T. (1998) ‘NPM Reforms and Other Western ReformStrategies’, in Innovations in Public Management: Perspectivesfrom East and West Europe, ed. T. Verheijen and D. Coombes(Cheltenham and Northampton, MA: Edward Elgar) pp.407–17.

Verheijen, T. and Coombes, D. (eds) (1998) Innovations in PublicManagement: Perspectives from East and West Europe(Cheltenham and Northampton, MA: Edward Elgar).

von Beyme, K. (2003) ‘Constitutional Engineering in Centraland Eastern Europe’, in von Beyme, K. (1985) Political Partiesin Western Democracies (Aldershot: Gower).

■WWachendorfer-Schmidt, U. (ed.) (2000) Federalism and Political

Performance (London and New York: Routledge).Wahlke, J. et al. (1962) The Legislative System (New York: Wiley).Walpole, S. (1881) The Electorate and the Legislature (London:

Macmillan). Wang, X. (2002) ‘The Postcommunist Personality: The Spectre

of China’s Market Reforms’, The China Journal (47) 1–17. Ware, A. (2002) The Direct Primary in the United States: Party

Institutionalization and Transformation in the American North(Cambridge and New York: Cambridge University Press).

Warwick, P. (1994) Government Survival in ParliamentaryDemocracies (Cambridge and New York: Cambridge UniversityPress).

Waters, M. (2000) Globalization, 2nd edn (London and NewYork: Routledge).

Watt, E. (1982) Authority (London: Croom Helm).Wattenberg, M. (2000) ‘The Decline of Party Mobilization’, in

Parties Without Partisans, ed. R. Dalton and M. Wattenberg(Oxford and New York: Oxford University Press) pp. 64–76.

Weaver, R. and Rockman, B. (eds) (1993) Do Institutions Matter?Government Capabilities in the United States and Abroad(Washington: The Brookings Institution).

Webb, P., Farrell, D. and Holliday, I. (eds) (2002) Political Partiesin Advanced Industrial Democracies (Oxford and New York:Oxford University Press).

Weber, M. [1923] (1946 edn) ‘The Social Psychology of theWorld Religions’ in From Max Weber: Essay in Sociology, ed.and trans. H. Gerth and C. Wright Mills (Oxford and NewYork: Oxford University Press) pp. 267–301.

Weber, M. [1922] (1957 edn) The Theory of Economic and SocialOrganization (Berkeley, CA: University of California Press).

Weber, M. [1918] (1990 edn) ‘The Advent of PlebiscitarianDemocracy’, in The West European Party System, ed. P. Mair(Oxford and New York: Oxford University Press) pp. 31–7.First pub. 1918.

Weiler, J. (1994) ‘A Quiet Revolution: The European Court ofJustice and its Interlocutors’, Comparative Political Studies (26)510–34.

Weiss, T. and Gordenker, C. (1996) ‘Pluralizing GlobalGovernance’, in Nongovernmental Organizations, the UnitedNations and Global Governance, ed. T. Weiss and C. Gordenker(Boulder, CO and London: Lynne Rienner) pp. 17–50.

Weissberg, R. (1998) Political Tolerance: Balancing Communityand Diversity (Thousand Oaks, CA and London: Sage).

346 REFERENCES

Weissberg, R. (2002) Polling, Policy and Public Opinion: The CaseAgainst Heeding ‘The Voice of the People’ (London and NewYork: Palgrave Macmillan).

Werbner, R. and Ranger, T. (eds) (1996) Postcolonial Identities inAfrica (London and Atlantic Highlands, NJ: Zed Books).

Wheare, K. (1951) Modern Constitutions (London and New York:Oxford University Press).

Wheare, K. (1963) Federal Government, 4th edn (Oxford andNew York: Oxford University Press).

Wheare, K. (1968) Legislatures (Oxford: Oxford UniversityPress).

Wheeler, N. (2002) Saving Strangers: Humanitarian Interventionin International Society (Oxford and New York: OxfordUniversity Press).

White, S., Batt, J. and Lewis, P. (eds) (2003) Developments inCentral and East European Politics 3 (Basingstoke: PalgraveMacmillan).

Whittaker, D. (2004) Terrorists and Terrorism in the ContemporaryWorld (London and New York: Routledge).

Wiarda, H. (ed.) (2004) Comparative Politics: Critical Concepts inPolitical Science, six vols. (London and New York: Routledge).

Wilensky, H. (1984) The Welfare State and Equality (Berkeley,CA: University of California Press).

Williams, C. (1995) ‘A Requiem for Canada?’, in Federalism: TheMultiethnic Challenge, ed. G. Smith (Harlow: Longman) pp.31–72.

Williams, R. (1962) Communications (Harmondsworth and NewYork: Penguin).

Williams, R. (1989) Television: Technology and Cultural Form,2nd edn (London and New York: Routledge).

Wilson, G. (2003) Business and Politics: A ComparativeIntroduction, 3rd edn (Basingstoke: Macmillan).

Wilson, G. (1990b) Interest Groups (Oxford and Cambridge,MA: Blackwell).

Wilson, J. (1989) Bureaucracy: What Government Agencies Do andHow They Do It (New York: Basic Books).

Wilson, J. and Dilulio, J. (2004) American Government:Institutions and Policies, 9th edn (Boston, MA: HoughtonMifflin).

Wilson, W. (1885) Congressional Government (Boston, MA:Houghton Mifflin).

Wiseman, J. (1985) ‘Introduction’ in Democracy and PoliticalChange in Sub-Saharan Africa, ed. J. Wiseman (London andNew York: Routledge) pp. 1–10.

Wiseman, J. (1996) The New Struggle for Democracy in Africa(Aldershot and Brookfield, VT: Avebury).

Wolinetz, S. (ed.) (1997) Political Parties (Aldershot andBrookfield, VT: Ashgate).

Wong, D. (2002) ‘Ready for Debate and Democracy’ FinancialTimes Survey of Singapore, April 12, p. 7.

World Bank (1997) World Development Report: The State in aChanging World (Oxford and New York: Oxford UniversityPress).

Wright, V. (ed.) (1994) Privatization in Western Europe: Pressures,Problems and Paradoxes (London and New York: Pinter).

Wright, V. (1997) ‘La Fin du Dirigisme?’, Modern andContemporary France (5) 151–5.

Wright, V., Guy Peters, B. and Rhodes, R. (eds) (2000)Administering the Summit: Administration of the Core Executivein Developed Countries (Basingstoke: Macmillan).

Wuthnow, R. (2002) ‘The United States: Bridging the Privilegedand the Marginalized?’, in Democracies in Flux: The Evolutionof Social Capital in Contemporary Society, ed. R. Putnam(Oxford and New York: Oxford University Press) pp. 59–102.

■X,■Y,■ZXia, M. (1999) ‘China’s National People’s Congress: Institutional

Transformation in the Process of Regime Transition(1978–98)’, in Parliaments in Asia, ed. P. Norton and N. Ahmed (London and Portland, OR: Frank Cass) pp.103–30.

Yayla, A. (2002) ‘Islam or Democracy’, in Islam, Civil Society andMarket Economy, ed. A. Yayla (Ankara: Liberte) pp. 1–18.

Yin, R. (2002) Case Study Research: Design and Methods, 3rd edn(Thousand Oaks, CA and London: Sage).

Young, L. and Cross, W. (2002) ‘The Rise of PlebiscitaryDemocracy in Canadian Political Parties’, Party Politics (8)673–99.

Zaslavsky, K. and Brym, J. (1978) ‘The Functions of Elections inthe USSR’, Soviet Studies (30) 362–71.

Zassoursky, I. (2003) Media and Power in Post-Soviet Russia(armonk, NY: M.E. Sharpe).

Zilliacus, K. (2001) ‘“New Politics” in Finland: The Greens andthe Left Wing in the 1990s’, West European Politics (24)27–54.

Zirakzadeh, C. (1997) Social Movements in Politics: AComparative Study (London and New York: Longman).

REFERENCES 347

Note: entries in bold indicate a definition;compare indicates a cross-reference to acontrasting entry.

■AAbacha, S. 61absolute deprivation 138absolute majority 148 accountability (bureaucratic) 299additional member system (AMS) 148administrative law 221, 221–3 advocacy coalition 174affirmative action 295Africa 25, 44, 47, 59, 61, 119–21, 127,

182, 306, 322 see also individual countries

agenda-setting 114–5agitation and propaganda 120 alternative vote (AV) 148, 150

compare run-off electoral systemsAlmond, G. 89–90Anderson, B. 9, 11, 12anti-globalization movement 34 apartheid (apartness) 213Arab world 62–6

see also authoritarian rule, Islam,personal rule

Arendt, H. 13 Aristotle 4, 36 Assassins 135Asia see individual countriesassemblies see legislaturesasymmetric federalism 229Atatürk, K. 63Athens, ancient 36–8 Australia 40, 134, 148, 150, 211, 216,

226, 255, 258, 275 Austria 178–80authoritarian rule 51–68, 52

communist states 52, 53–6economic development 57 fascist states 52interest groups 182–4 military governments 52, 58–62parties 201–2traditional 51–3, 52, 63–4compare constitutions, democracy

authority 13–15, 14compare power

■Bbacksliding 40 Bagehot, W. 265 Basic Law (Germany) 42 behaviouralism 78Belgium 127, 230, 280Bentham, J. 226Bentley, A. 177Berlusconi, S. 281 bicameral legislature

see also federalism‘big push’ 323Bill of Rights 210 bin Laden, O. 51, 134

see also September 11, Islam, politicalterror and violence

Bismarck, O. 316 Blackstone, W. 7block grants 234 Bodin, J. 7–8, 18Boutros-Ghali, B. 31 Brandeis, L. 235Brazil 271–4Bretton Woods 33Britain see United Kingdom British Broadcasting Corporation (BBC)

109British North America Act (Canada)

232Bryce, J. 251, 265 budget-making 257–8Bulgaria 304 Bundestag, Bundesrat see Germany

(legislature)bureaucracy 290, 290–308

accountability 299–301 authoritarian states 304–7 bureaucratic authoritarianism 306capacity-building 306, 322communist states 307fascist states 307generalist v. specialist 292new democracies 303–4 new public management (NPM)

301–3regulatory agencies 298, 318–9recruitment 292–3, 295 representative bureaucracy 295unified v. departmental 292Weber’s analysis 291

Burke, E. 254 Bush, G. W. 26, 63, 95, 155, 234, 260

■Ccabinet 278–80 cabinet committees 280cabinet (France) 300Canada 150, 170, 190–1, 199, 218,

229, 232, 235, 262, 275Cancún talks (2003) 34capacity building 306Carlos, J. 280career politician 262cartel party 192case study 79–82, 80

compare statistical analysiscase-by-case privatization 321

compare mass privatizationcatch-all party 187categorical grant 234 Cecil, R. 249chamber of commerce 170, 180charismatic authority 14

compare traditional and legal-rationalauthority

Charter of Rights and Freedoms (Canada)218

Chavez, C. 130China (People’s Republic) 66–8

authoritarian state 51 ‘big push’ 323 bureaucracy 307 Chinese Communist Party (CCP)

203–4Cultural Revolution 225 decentralization 67elections 165 executive 288Great Leap Forward 57interest groups 183–4internet 121 law 225 National People’s Congress

266–7participation 134policy process 323political culture 97 revolution 56, 137 Tiananmen Square 134

checks and balances 270Chile 60–1Christian Church 17–18Ciampi, C. 280citizenship 11, 317 citizens’ jury 118

compare opinion poll civic culture 89–90

Index

348

civil service 290see also bureaucracy

civil service act 304civil society and civic deficit 127

compare totalitarianismcivilizations, clash of (Huntington)

102–4class action 170clientelism 132

compare legal-rational authority coalition government 152–4, 275–8,

277compare single-party government

coat-tails 146cohabitation 284

compare coalitionsCold War 21, 24, 46, 53, 59, 60 collective goods 166Commission (EU) see European UnionCommittee of the Regions and Local

Communities (CoR) 239 Common Market see European Unioncommunist states 53–6

bureaucracy 307 constitutions 225 elections 165executive 287–8 interest groups 183–4 judiciary 225 legislatures 266–7 mass media 120–1 parties 202–3 planning 322–3 political culture 97 political participation 133–4subnational government 246 see also individual countries

comparative approach 69–85 case studies 79–82focused comparison 82–3interpretive approach 71–2, 72levels of analysis 75 most similar v. most different designs

82–3, 83selection bias 73–4, 74small-N problem 74 statistical analysis 83–5

competition state 318–19 compare welfare state, regulation

compulsory voting 161concurrent (simultaneous) elections 156concurrent (shared) powers 229 council-mayor system 242

compare mayor-council system conditionality 33confederation 229

compare federalismconference (mediation) committee 257 confidence vote 259

compare interpellationconstitutions and constitutional rule

210, 210–27amendment 211–12codified 210–11

entrenched 211–2 origins 212 structure of documents 210 compare authoritarian rulesee also individual countries

constitutional courts 214–16 compare supreme courtssee also judiciary

constructive vote of no confidence 278contracts 302cooperative federalism 233

compare dual federalismcorporatism 178, 178–81

compare pluralismcorps (body) of civil servants 292–3,

304correlation 84corruption 61, 68cost-benefit analysis (CBA) 311,

311–12Council of Ministers see European Unioncouncil manager system (local

government) 242compare mayor–council system

coup, military 58–60, 59see also military rule, revolutions

courts see judiciaryCrick, B. 4Crockett, D. 256 Cuba 165cumul des mandats 244Cultural Revolution (China) 225Czech Republic 23, 46, 125, 148,

200–1, 284–5, 304

■DDahl, R. 13, 36, 40, 49, 178Dar al-Islam (House of Islam) 12de Gaulle, C. 281–2 de Tocqueville, A. 81, 96, 126 dealignment 157–9, 158debt crisis 33–4 Declaration of the Rights of Man and the

Citizen (France, 1789) 20 Declaration of Human Rights (UN,

1948) 217decentralization 236

compare devolutiondecision trap (Scharpf ) 235decoding 105decolonization 22–3deconcentration 236

compare decentralizationdeliberation 254

compare representationdemocracy 36–50

‘Asian’ 48Athenian 36–8 direct 35, 36–8 and peace 136representative and liberal democracy,

35, 38–9semi-democracy, 35, 46–8

compare authoritarian rule see also democratization, established

democracies, new democraciesdemocratic centralism 203democratic consolidation 43democratic deficit 29democratization 39–46Denmark 109, 169, 180, 191

see also Scandinaviadependency ratio 317dependency theory 32 dependent variable 74developmental state 57, 305 devolution 236

compare decentralizationdiaspora 10, 12Dicey, A. 209Diplock, Lord 222–3direct democracy 35, 36–8

compare representative democracy direct election 155–6, 156, 249

compare indirect electiondistrict magnitude 151dominant party system 195–6 compare two-party and multi-party

systemsDowns, A. 159–60, 197dual executive see semi-presidential

executivedual federalism 231–4

compare cooperative federalismdual system (local government) 243

compare fused system Duverger, M. 153, 281

■EEast Germany see GermanyEastern Europe 53–6

see also post-communist states,individual countries

Easton, D. 4, 77–8Eckstein, H. 128education 95, 98Egypt 57, 132, 133, 306Eisenhower, D. 221 elections 145–65

in authoritarian states 164–5 concurrent 156in established democracies 145–60 direct and indirect 155–6, 156, 249 founding and second elections 163franchise 145–6 functions 145initiative 160–3, 162in new democracies 163–4 presidential 154–6 primary 190 recall and referendum 160–3, 162second-order 145 turnout 159–60 voting behaviour 156–9 see also individual countries

electoral democracy see semi-democracy

INDEX 349

electoral-professional party (Panebianco)188

electoral reform 150, 154electoral systems 124, 146–56, 148elite (caucus) party 187

compare mass and catch-all partieselite political culture 95, 98–100emergency debate (legislature) 259

compare interpellationempires 21–25, 51–3, 63 enabling authority (local government)

243England see United Kingdomepistemic community 174equalization grant 234 established (consolidated) democracy

35, 38–9 compare new democracy, authoritarian

ruleEuropean Coal and Steel Community

(ECSC) 29 European Committee of the Regions and

Local Communities (COR) 239European Convention on Human Rights

(ECHR) (1950) 217 European Court of Justice (ECJ) 216,

217European Economic Community (EEC)

29European Monetary System (EMS) 29European Parliament 254–5 European Regional Development Fund

(ERDF) 238 European Union (EU) 29, 176

democratic deficit 29 federal status 237 regions 238–9regulation 319treaties 29 compare North American Free Trade

Association, United States of Americaevaluation of policy 313–14executive 268, 268–89

authoritarian states 285–9 communist states 287–8 fascist states 287 new democracies 284–5 parliamentary government 274–81personal rule 285–7, 286post-communist states 284–5 presidential government 268–74 semi-presidential (dual) executive

281–4, 282

■Ffascist states 56–8, 97, 120, 183, 204,

246, 266, 307 compare communist states

Federal Republic of Germany (FRG) seeGermany

federalism 228–36, 229allocation of functions 228–9asymmetric 229

dual and cooperative 231–4 European Union 237executive 234 intergovernmental relations 234 legislatures 250strengths and weaknesses 235–6compare unitary government

feudalism 17–18, 247 Finer, S. 37, 212Finland 109, 277, 284

see also ScandinaviaFiorina, M. 158first past the post (plurality) electoral

system 146–50, 148flash party 193focused comparison 80, 82–3

compare case studyfocus group 117–19, 118

compare opinion pollformateur 277Foucault, M. 13founding election 163

compare second electionFox, V. 45framing (by media) 114–15France 283

bureaucracy 222, 292–3 cabinet 300constitution 19–20 electoral system 148, 155, 157executive 281–4 interest groups 169–70 judiciary 222 local government 244 nation 8, 20 revolution (1789) 10–11, 19, 38, 82,

135, 139–40, 283 subnational government 238unitary state 244

franchise 145–6Franco, General 56, 303

see also fascist states, Spain Frank, A. 32 Franklin, B. 231free rider 166free trade area (FTA) 231

compare federalismfreedom of the press 116Friedrich, C. 210Fujimori, A. 164 functional equivalence 72fused system (local government) 243,

243–4compare dual system

■GGaebler, T. 301–2 garbage can model 312Geertz, C. 80Gellner, E. 10general competence (local government)

240genocide 134

Germany 94 bureaucracy 293, 294–6 coalition government 276, 277constitution (Basic Law) 42 constructive vote of no confidence

278elections and electoral system 152–3 executive 279 federalism 229, 233–4, 250 head of state 281 interest groups 181 judiciary 215–16, 233 legislature 250, 257 parliament 250, 252, 257 parties 187–8, 191, 192, 199 political culture 96

gerrymandering 149Gingrich, N. 251gladiators and spectators (participation)

123globalization 25, 25–34, 110–12 Goebbels, J. 120Gorbachev, M. 56 Gosplan (Soviet Union) 323 governance 6, 242–3 government 4–5, 5see also bureaucracy, executive, limited,

local and regional government,federalism

grand coalition 277Greece 36–8 green parties 155, 170 Grotius, H. 31guanxi (personal connections) 68, 288Gurr, T. 138

■HHamilton, A. 269Havel, V. 284head of state 280–1Hitler, A. 58, 287 Hobbes, T. 5, 18Hoffman, S. 31 Holland see Netherlands humanitarian intervention 31 Hungary 23, 44, 46, 54, 152, 200, 264,

284, 304Huntington, S. 40, 102–4

■IIceland 11

see also Scandinaviaimplementation of policy 312–13 incremental policy-making 311–12

compare rational policy-making incumbency effect 262independent variable 83 India 23, 57, 196, 231 indirect election 155, 155–6

compare direct election industrial revolution 194

compare postindustrial revolution

350 INDEX

Inglehart, R. 93, 95initiative 162

compare recall institutions and institutionalization

75–7, 76interest groups 167–84

authoritarian states 182–4 channels of access 169–72 communist states 182–4 conditions of influence 175–7 corporatism 177–81, 178density of membership 175issue networks 172–4, 174new democracies 181–2 peak associations 169pluralism 177, 177–81post-communist states 181–2 protective and promotional 168compare social movements

intergovernmental organization (IGO)25–30, 26

intergovernmental relations 234 International Bank for Reconstruction

and Development (IBRD) seeWorld Bank

international community 31 International Court of Justice (ICJ) 223International Covenant on Civil and

Political Rights 226 International Criminal Court (ICC)

223international law 225–7, 226International Monetary Fund (IMF)

33–4international non-governmental

organization (NGO) 26, 30internet 6, 109, 111, 121 interpellation 259

compare confidence vote, emergencydebate

interpretive approach 72intervening variable 84investiture rules 276Iran 15, 64–5 Iraq 26, 62, 231 Iraq War (2003) 130iron law of oligarchy (Michels) 188iron triangles 172–4Islam 12, 61, 62–6, 102–4, 120, 152 issue networks 172–4, 174

compare subgovernments Italy 18, 58, 91–2, 154, 160–1, 198,

238, 257

■JJapan 154, 196, 220, 293, 305 J-curve theory (Davies) 138 Jefferson, T. 315 Johnson, L. B. 270joint decision trap (Scharpf ) 235judiciary 27–8

activism and restraint 216–20, 217globalization 27–8

independence and recruitment 220–1 new democracies 223–4review 212, 212–6see also constitutional courts, law

■KKaradzic, R. 227 Kelsen, H. 214Kennedy, J. F. 256, 270Khatami, M. 66Khomeini, R. 15, 137 Khrushchev, N. 56 Kingdon, J. 310, 312Kingsley, J. 295Kircheimer, O. 187Kitschelt, H. 200Kurds 11

■LLatin America

authoritarian rule 58–62 bureaucracy 306 decolonization 22democratization 43–4, 47 executive 271–4, 284 interest groups 181–2judiciary 223–4 legislatures 260, 263–4 military 58–62 policy process 322 social movements 130subnational government 245technocracy 99unitary government 236compare post-communist states see also individual countries

law 46, 223–7, 254–5, 293 legal-rational authority 14, 15

compare charismatic and traditionalauthority

legislatures Athenian 36authoritarian states 264 committees 250–2, 259 communist states 266–7decline 265 functions 252–9 globalization 28law-making 254–5 Mezey’s classification 266new democracies 263–4 post-communist states 264 recruitment 259–63 relations with executive 274–8 representation 252–3 scrutiny 258–9talking and working 251unicameral and bicameral 229, 249see also individual countries

legitimacy 15Lenin, V. 7, 55, 140 Lewis, B. 64

liberal democracy 35, 38–9see also democracy

Lindblom, C. 311–12Linz, J. 68 Lipset, S. 15, 44, 193list system (PR) 148

compare single transferable vote lobbying 171–2 local government 239–44

dual v. fused 243enabling authority 243functions 242 general competence 240relationship with central government

243–4status 239structures 241 ultra vires 240

Locke, J. 5, 19, 315 logics of appropriateness and of

consequences 76lonely crowd (Riesman) 126

compare social capital Lukes, S. 13Lula (Luis Inácia da Silva) 274Luther, M. 18

■MMadison, J. 38, 233, 247 mad cow disease 310, 312–13Mahathir, D. 34, 49Majone, G. 319majority electoral systems 148, 150, 155

compare plurality electoral systemmajority government 275Malaysia 34, 47 mandatory voting 161 Mandela, N. 213 Mao Zedong 67, 134 March, J. 76Marshall, T. 317, 318Marx, K. and Marxism 53–4, 79, 120,

140–1compare Lenin, V.see also communist states

mass media 106–16, 108authoritarian states 119–21 development of 106–8 impact on public 114–15internet 6, 109, 111, 121new democracies 115–16 reinforcement, agenda-setting and

framing 114–15 spin 112television v. newspapers 113–14trends 108–12 see also individual countries

mass party 187 compare elite (caucus) party

mass society (Kornhauser) 129compare social capital

mass privatization 321compare case-by-case privatization

INDEX 351

Mayhew, D. 270mayor–council system (local government)

242compare council manager system

McCarthy, J. 98 McKenzie, R. 188mechanical effect of electoral systems

(Duverger) 153media see mass mediamediation (conference) committee 251,

257medieval governance 17–18, 38Mexico 45, 131, 165, 183, 260Michels, R. 188 Michigan (party identification) model of

voting 156–7, 157microcosm theory of representation 253Middle East 62–6, 286–7

see also authoritarian rule, Islam,individual countries

migration 11Milbrath, L. 123military coup 58–60, 59military rule 58–62, 66, 132–3, 324military withdrawal 60–2 Mill, J. S. 10, 254 Milosevic, S. 226 minimum winning coalition (MWC)

277minority government 275mixed member electoral systems 148,

152, 189compare plurality system, proportional

representationMobutu, S. 286monarchy 51–3, 280–1 Montesquieu, C.-L. 102 Morgenthau, H. 12 most similar and most different designs

82–3, 83muddling through (Lindblom) 312 multilevel governance 228multinational corporation (MNC) 30–2multinational state 10, 10–11

compare nation-state, stateless nationmultiparty system 152–6, 195, 199–200 Murdoch, R. 109, 298Mussolini, B. 58, 246, 287

see also fascist states

■NNader, R. 175 Nasser, A. 57, 132, 133, 306 nation 8–12, 9, 106

compare statenation-state 10, 10–11

compare multinational state, statelessnation

National Rifle Association (NRA) 175National Socialist German Workers’ Party

(NSDAP, Nazi Parry) 207nationalism 8–10, 10, 194 navette (shuttle) 257

Netherlandsbureaucracy 293 elections 161 interest groups 168 judiciary 216 legislature 259 parliamentary system 275 parties 189 political culture 100subnational government 238

new democracies 35, 43–6, 47bureaucracy 303–4 elections 163–4 executive 284–5 interest groups 181–2 law 223–4legislature 263–4 participation 124–8 parties 200–1 policy process 319–22 political communication 115–16political culture 95–6 subnational government 244–5compare established democracies, semi-

democracynew politics 127–31 new public management (NPM) 301–3

compare bureaucracynew social movements (NSMs) 127–31newspapers 106–8New Zealand

bureaucracy 301, 302–3 constitution 211, 218 elections and electoral reform 154,

158, 159 female representation 125legislature 248 parties 190referendums 160 subnational government 238, 239

Nigeria 61 nightwatchman state 315–6 Nimby (not in my backyard) group 167 Nixon, R. 100, 271nomenklatura (list of names) 307non-governmental organization (NGO)

26, 30normal science (Kuhn) 81North American Free Trade Association

(NAFTA) 28, 45 Norway 95, 152, 158, 159, 238, 257,

275, 276, 293see also Scandinavia

Nuremberg tribunal 226

■OObasanjo, O. 61oil crises (1970s) 318oligarchy 188

see also traditional authoritarian rule Olsen, J. 76 Olson, M. 7, 180ombudsman 301

open and closed party lists 151opinion poll 117–19, 118

compare focus groupOsborne, D. 301–2 Ostrogorski, M. 185 Ottoman empire 22, 63 outlier 84oversight (scrutiny) function 253,

258–9oversize coalition 277

■PPaine, T. 38 Pakistan 66 palace politics 51–3 Palestine 12Panebianco, A. 186parliamentary government 274–81, 275

government formation and coalitions275–8

prime minister and cabinet 278–9compare presidential government

parliaments see legislaturesParsons, T. 12participation see political participation parties 185, 185–205

authoritarian states 201–5 communist states 53–6, 202–4 decline 157–9, 186, 191–3 electoral-professional (Panebianco)

188elite, mass and catch-all 187–8 functions 185 globalization 28new democracies 200–1 organization 186–9 post-communist states 200–1 protest 193social base 193–4 vote-maximizers 197

partisan dealignment 157–9, 158party competition 194–200party identification 156–7, 157

compare retrospective voting party systems 194, 194–200 patrimonial (patriarchal) authority 51,

324compare legal-rational authority

patron–client networks 131–2 pay-as-you-go pension scheme 317peak association 168

compare promotional and protectivegroups

Peel, R. 116 pensions 317People’s Republic of China (PRC) see

ChinaPerón, J. 60 Perot, R. 28 Persia see Iranpersonal rule (Jackson and Rosberg) 286Peru 164 Pinochet, A. 60, 132

352 INDEX

planned economy 323 pluralism 177, 177–8

compare corporatismplurality (first past the post) electoral

system 146–50, 148, 189compare proportional representation

Poland 181, 245, 285policy-connected coalition 277policy entrepreneur 130policy process 309, 309–24

authoritarian states 322–4 communist states 322–3 established democracies 315–19 evaluation 313–14 garbage can model 312implementation 311–12new democracies 319–22 outputs v. outcomes 314post-communist states 320–2 rational (synoptic) v. incremental

311, 311–12stages 309–15termination 314–15

polis (city community) 36political communication 106–21, 130

see also mass mediapolitical culture 89–104

authoritarian states 97–8 civic culture 89–90 communist states 97–8 elite political culture 98–100 new democracies 96–7 political trust and social capital 90–3post-communist states 96see also postmaterialism, civilizations,

Islam political exclusion 124political executive see executivepolitical learning see political

socialisationpolitical participation 122–41, 123

authoritarian states 131–4 communist states 133–4 established democracies 122–4 new democracies 124–8 political violence and terror 134–6 regimented 133

political parties see partiespolitical recruitment 189–91 political socialization 100, 100–2 political system 77–8political terror and violence 134political trust 90–3politics 3–4 post-colonial states 21–25post-communist states 22–3

bureaucracy 304 culture 96–7democratization 43–6executive 284–5 interest groups 181–2 law 224–5mass media 115–16 parties 200–1

policy process 320–2political participation 124–8subnational government 244–5see also individual countries

postindustrial revolution 194 compare industrial revolution

postmaterialism 93–5, 95power 12, 12–13

compare authoritypresidential government 268–74, 269

compare parliamentary governmentsee also executive

pressure groups see interest groups primary election 156 prime minister 278–80

see also parliamentary governmentprivatization 297–8, 318–19 promotional and protective groups 168propaganda 120Proposition 13 (California, 1978) 160proportional representation (PR) 148,

150–2, 189compare plurality electoral system

protest parties 193psychological effect of electoral systems

(Duverger) 153public opinion 116–9, 117public policy see policy process public relations 171–2public sector 290

see also bureaucracy Putin, V. 47, 49, 115, 128, 182 Putnam, R. 90–3, 123, 191

■Qquango 294quotas 124, 295

■Rrainbow coalition 277rational policy-making 311–12 compare incremental policy-makingReagan, R. 21, 270, 271, 301 recall 162

compare initiativerecruitment

judicial 220–1 legislative 259–63party 189–91

referendum 160–3, 162compare initiative, recall

Reformation 18regions (between states) 28–9, 46 regions (within states) 238–9, 244–5regression analysis 83–4regulation 297–8, 318–19regulatory capture 298reinforcement (mass media) 114–15relative deprivation 137–41, 138religion and churches 51–3, 62–6,

102–4, 130, 181, 240see also Islam

representation 252–3 compare deliberation

residual powers (federalism) 229 retrospective voting 158

compare party identification revenue-sharing 234 revolution 136–41, 137

China 137 Eastern Europe 137 France 82, 135, 139–40 industrial 53, 194, 323 national 194post-industrial 93–5, 194 Russia 53, 137, 140–1 theories 137–9Turkey 137

Rhodes, R. 6, 72, 75risk management 298 Rokkan, S. 193Roosevelt, F. 108Rummel, R. 136run-off electoral systems 148, 150, 155

compare alternative voteRussia 23, 47, 53–6, 115, 125, 128,

140–1, 163, 182, 222 see also Soviet Union

Rwanda 136, 226

■SSalazar, A. 56

see also fascist states Salinas, R. 45 sample survey 117–19, 118

compare focus groupSARS virus 26–7Sartori, G. 185, 210Saud, I. 63–4Saudi Arabia 63–4, 286 Scandinavia

bureaucracy 297 interest groups 169, 180–1 internet 109legislatures 252 newspapers 113parties 191, 194political culture 93unitary government 236welfare state 318 see also Denmark, Finland, Iceland,

Norway, Sweden scatterplot 83 Schattschneider, E. 179, 189Schumpeter, J. 39, 41 Schwarzenegger, A. 162scrutiny (oversight) 258–9Seattle, battle in (1999) 34second ballot 148, 150 second chambers see legislatures second election 163

compare founding election second-order election 145 select committee 251

compare standing committee

INDEX 353

selection bias 73–4, 74selective perception, interpretation and

recall 114 self-determination 9semi-competitive election 47, 164–5 semi-democracy 35, 46–9, 47

see also democracysemi-presidential (dual) executive

281–4, 282compare parliamentary and presidential

government September 11 (2001) 35, 46, 51, 62–6,

89, 102–4, 134–6, 233, 235, 287see also civilizations (clash of ), Islam,

political terror, Saudi Arabia Serbia 11shuttle (navette) 257

compare mediation committeeSièyes, A. 249 Simon, H. 311Singapore 48, 202 single currency see European Unionsingle-party government 274–6

compare coalitionssingle transferable vote (STV) 148

compare list system Skinner, Q. 19Skocpol, T. 74, 79, 139small-N problem 74social capital 90–3, 92, 125–6social movements 127, 127–31

compare interest groups, political parties

Somalia 31South Africa 195, 213, 220 sovereignty 7–8, 8, 31 Soviet Union (USSR) 53–6, 165,

202–3, 246, 288, 323see also communist states, Russia

Spain 96, 151, 223, 238, 303–4spectators and gladiators (participation)

123spin doctor 112spoils 291–2

compare civil service actspurious correlation 84 Stalin, J. 56, 288 standing committee 251

compare select committeestare decisis (stand by decisions made)

215state 7, 7–8, 17–34, 78–9, 135–6

see also authoritarian rule, communiststates, competition state, establisheddemocracy, fascist states, federalism,multinational state, new democracy,nightwatchman state, post-colonialstates, post-communist states, semi-democracy, totalitarian and welfarestate

stateless nation 10, 10–12compare nation-state, multinational

statestate-owned enterprise (SOE) 68, 297

statistical analysis 83–5compare focused comparisons, case

studiesstatutory agency 297–8 structural adjustment programme 33subgovernment 172–4

compare issue network subnational government 238–9, 244–6 subsidiarity 233successor party 200Sumner, W. 315 Sundberg, J. 194 supreme court 210–21

compare constitutional courtsSweden

bureaucracy 300, 301 constitutional acts 211electoral system 148, 152folkrörelser 129, 131 head of state 281internet 109investiture rule 276judiciary 217legislature 252, 259local government 240onbudsman 301 parties 195 referendums 160, 162trust 92 see also Scandinavia

Switzerland 8, 162synoptic (‘rational’) policy-making

310–12compare incremental policy-making

systems analysis 77–8

■Ttechnocracy 99

compare democracyterm limits 81, 156, 260, 262territory 7, 19–20Teune, H. 239 terror 134–6

see also September 11 (2001)Thatcher, M. 21, 300theocracy 64, 64–6

see also Islam‘thick description’ (Geertz) 80thresholds (electoral systems) 152 Thugs 135Tiananmen Square massacre 134 totalitarian states 53, 133–4, 246, 266,

287–8compare civil society, liberal democracysee also authoritarian states

trade unions 167, 168, 170–1 traditional authority 14

compare charismatic and legal-rationalauthority

traditional authoritarian rule 51–3see authoritarian rule

transmission model of communication105

Truman, H. 270 trust 90–3Turkey 62–3, 103, 152 turnout 126, 159–60

compare compulsory voting Tuvalu 248 two-ballot electoral system 148, 150,

155compare alternative vote (AV)

two-party system 195, 196–9 tyranny 39

compare traditional authoritarian rule

■UUganda 224 Ukraine 244 ultra vires 240

compare general competenceunicameral legislature 248–9

compare bicameral legislatureUnion of Soviet Socialist Republics

(USSR) see Soviet Unionunitary government 236–9

compare federalismUnited Kingdom 261

bureaucracy 222, 292, 303 constitution 209 democracy 41–2devolution 238elections 148, 149 executive 275 interest groups 167, 174, 179 head of state 281 judiciary 218, 222 legislature 252, 255, 258, 261 local government 243 mass media 106–8, 109 party system 196 policy process 318–19

United Nations 8, 10, 26, 30, 217see also humanitarian intervention

United States of America 147 administrative law 222bureaucracy 291, 300 Congress 170, 247, 249, 251–2, 256,

258constitution 19, 210–12 elections and voting 148, 155, 156–9,

190federalism 230–1, 235 interest groups 170 judiciary 214–15, 218, 200, 235 liberal democracy 40–41local government 240participation 126parties 156–7, 187, 190, 196 pluralism 179 political culture 90–2 political participation 126presidential system 269–71, 300 revolution 19spoils 291–2see also September 11 (2001)

354 INDEX

upper chambers see legislaturesUzbekistan 287

■Vvanguard party 55 Verba S. 79–80voting behaviour 156–9

■WWalesa, L. 284 Walpole, S. 257war and the state 18, 20

Warren, E. 221 Watergate 100 waves of democratization (Huntington)

39–43Weber, M. 7, 14–15, 51, 291, 299welfare state 20–1, 316–8 Western Europe see European Union,

individual countriesWest Germany see Germany Westphalia, Peace of (1648) 18Williams, R. 105Wilson, W. 251women in politics 123–5world (profiled) 24World Bank (International Bank for

Reconstruction and Development)33–4

World Trade Center see September 11(2001)

World Trade Organization (WTO) 6,33, 68

world wide web 6, 109, 111, 121

■YYeltsin, B. 47

■ZZealots 135

INDEX 355