key texts in human geography -...

Key Texts in Human Geography

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page i

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page ii

Key Texts in Human Geography

Edited by

Phil Hubbard, Rob Kitchin and Gill Valentine

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page iii

Editorial arrangement © Phil Hubbard, Rob Kitchin and Gill Valentine 2008

Chapter 1 © Bo LenntropChapter 2 © Michael F. GoodchildChapter 3 © Martin CharltonChapter 4 © Ron JohnstonChapter 5 © Andy WoodChapter 6 © David Seamon and Jacob SowersChapter 7 © Tim CresswellChapter 8 © Noel CastreeChapter 9 © Martin PhillipsChapter 10 © Nick PhelpsChapter 11 © Susan HansonChapter 12 © David GilbertChapter 13 © Satish KumarChapter 14 © Jonathan Beaverstock

Chapter 15 © Keith Woodward and John PaulJones III

Chapter 16 © Claudio MincaChapter 17 © Neil CoeChapter 18 © Nick SpeddingChapter 19 © Robyn LonghurstChapter 20 © John PicklesChapter 21 © Phil HubbardChapter 22 © Jo SharpChapter 23 © Philip KellyChapter 24 © Sarah DyerChapter 25 © Alan LathamChapter 26 © Ben Anderson

First published 2008

Apart from any fair dealing for the purposes of research or private study, or criticism orreview, as permitted under the Copyright, Designs and Patents Act, 1988, this publicationmay be reproduced, stored or transmitted in any form, or by any means, only with theprior permission in writing of the publishers, or in the case of reprographic reproduction,in accordance with the terms of licences issued by the Copyright Licensing Agency.Enquiries concerning reproduction outside those terms should be sent to the publishers.

SAGE Publications Ltd1 Oliver’s Yard 55 City RoadLondon EC1Y 1SP

SAGE Publications Inc.2455 Teller RoadThousand Oaks, California 91320

SAGE Publications India Pvt LtdB 1/I 1 Mohan Cooperative Industrial AreaMathura RoadNew Delhi 110 044

SAGE Publications Asia-Pacific Pte Ltd33 Pekin Street #02-01 Far East Square Singapore 048763

Library of Congress Control Number 2007940824

British Library Cataloguing in Publication data

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

ISBN 978-1-4129-2260-9ISBN 978-1-4129-2261-6 (pbk)

Typeset by C&M Digitals (P) Ltd., Chennai, IndiaPrinted in Great Britain by TJ International Ltd, Padstow, CornwallPrinted on paper from sustainable resources

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page iv

Contents

List of Contributors viiAcknowledgements ixList of Figures and Tables xi

Editors’ Introduction xiii

1 Innovation Diffusion as Spatial Process (1953): Törsten Hägerstrand 1Bo Lenntrop

2 Theoretical Geography (1962): William Bunge 9Michael F. Goodchild

3 Locational Analysis in Human Geography (1965): Peter Haggett 17Martin Charlton

4 Explanation in Geography (1969): David Harvey 25Ron Johnston

5 Conflict, Power and Politics in the City (1973): Kevin Cox 33Andy Wood

6 Place and Placelessness (1976): Edward Relph 43David Seamon and Jacob Sowers

7 Space and Place (1977): Yi-Fu Tuan 53Tim Cresswell

8 The Limits to Capital (1982): David Harvey 61Noel Castree

9 Uneven Development (1984): Neil Smith 71Martin Phillips

10 Spatial Divisions of Labour (1984): Doreen Massey 83Nick Phelps

11 Geography and Gender (1984): Women and Geography Study Group 91Susan Hanson

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page v

12 Social Formation and Symbolic Landscape (1984): Denis Cosgrove 99David Gilbert

13 Capitalist World Development (1986): Stuart Corbridge 109Satish Kumar

14 Global Shift (1986): Peter Dicken 117Jonathan Beaverstock

15 The Condition of Postmodernity (1989): David Harvey 125Keith Woodward and John Paul Jones III

16 Postmodern Geographies (1989): Edward Soja 135Claudio Minca

17 The Capitalist Imperative (1989): Michael Storper and Richard Walker 145Neil Coe

18 The Geographical Tradition (1992): David Livingstone 153Nick Spedding

19 Feminism and Geography (1993): Gillian Rose 163Robyn Longhurst

20 Geographical Imaginations (1994): Derek Gregory 171John Pickles

21 Geographies of Exclusion (1995): David Sibley 179Phil Hubbard

22 Critical Geopolitics (1996): Gearóid Ó’Tuathail 189Jo Sharp

23 Logics of Dislocation (1996): Trevor J. Barnes 197Philip Kelly

24 Hybrid Geographies (2002): Sarah Whatmore 207Sarah Dyer

25 Cities (2002): Ash Amin and Nigel Thrift 215Alan Latham

26 For Space (2005): Doreen Massey 225Ben Anderson

Index 235

viÿÿCONTENTS

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page vi

Contributors

Ben Anderson is Lecturer in Human Geography, University of Durham, UK

Jonathan Beaverstock is Professor of Economic Geography, NottinghamUniversity, UK

Noel Castree is Professor of Human Geography, University of Manchester, UK

Martin Charlton is Senior Research Fellow, National Centre for Geocomputation,National University of Ireland, Maynooth

Neil Coe is Senior Lecturer in Human Geography, University of Manchester, UK

Tim Cresswell is Professor of Human Geography, Royal Holloway, University ofLondon, UK

Sarah Dyer is Lecturer in Human Geography, Oxford University, UK

David Gilbert is Professor of Urban and Historical Geography, Royal Holloway,University of London, UK

Michael F. Goodchild is Professor of Geography, University of California, SantaBarbara, US

Susan Hanson is Professor of Geography, Clark University, US

Phil Hubbard is Professor of Urban Social Geography, Loughborough University, UK

Ron Johnston is Professor of Human Geography, Bristol University, UK

John Paul Jones III is Professor of Geography, University of Arizona, US

Philip Kelly is Associate Professor of Geography, York University, Canada

Satish Kumar is Lecturer in Human Geography, Queen’s University Belfast, UK

Alan Latham is Lecturer in Geography, University College London, UK

Bo Lenntrop is Emeritus Professor, Department of Geography, Stockholm University,Sweden

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page vii

viiiÿÿCONTRIBUTORS

Robyn Longhurst is Professor of Geography, University of Waikato, New Zealand

Claudio Minca is Professor of Human Geography, Royal Holloway, University ofLondon, UK

Nick Phelps is Reader, Bartlett School, University College London, UK

Martin Phillips is Reader in Social and Cultural Geography, University of Leicester, UK

John Pickles is Earl N. Phillips Distinguished Professor of International Studies,University of North Carolina, US

David Seamon is Professor of Architecture, Kansas State University, US

Jo Sharp is Senior Lecturer in Human Geography, University of Glasgow, UK

Jacob Sowers is a doctoral student in the Geography programme, Kansas StateUniversity, US

Nick Spedding is Lecturer of Geography, University of Aberdeen, UK

Andy Wood is Associate Professor of Human Geography, University of Kentucky, US

Keith Woodward is Lecturer in Human Geography, University of Exeter, UK

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page viii

Acknowledgements

The editors want to express their gratitude to all the contributors for responding sopositively to their invitation to contribute to this volume, and for working to tight deadlines.We also wish to thank Robert Rojek for his patience and encouragement whilst webought this project to completion.

Among our contributors, Susan Hanson wishes to acknowledge discussions withSophie Bowlby and Megan Cope on Geography and Gender. Ron Johnston wishes tothank Les Hepple, Tony Hoare, Kelvyn Jones, Charles Pattie and Eric Pawson for valu-able discussions of his essay and comments on draft versions. Nick Phelps would like tothank Doreen Massey, Philip Cooke and Mick Dunford for providing further backgroundon the radical economic geographical scene of the late 1970s and early 1980s. SatishKumar acknowledges assistance from Niall Majury, Nuala Johnson, Diarmid Finneganand Harjit Singh in rehearsing the account of capitalism, neo-liberalism, and dialecticsthat underpins his chapter, and also thanks David Zou for assistance in bibliographiccompilation. Phil Hubbard also wishes to thank all members of the LoughboroughUniversity Department of Geography reading group for sharing their thoughts on DavidSibley’s Geographies of Exclusion.

© 26.1 ‘Ceci n’est pas I’espace’ Massey, Doreen. For Space. SAGE, London 2005.

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page ix

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page x

List of Figures and Tables

Figures

8.1 The primary circuit of capital accumulation 65

9.1 Senses of first and second nature in uneven development 75

26.1 ‘Ceci n’est pas I’espace’ 225

Tables

9.1 Smith’s logico-historical construction of the production of nature 739.2 The production of scale within capitalist production 77

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/17/2008 1:00 PM Page xi

Hubbard-Prelims:00-Hubbard-Prelims.qxp 4/15/2008 7:20 PM Page xii

Editors’ Introduction

Phil Hubbard, Rob Kitchin and Gill Valentine

Why key texts?

Geography, like all academic disciplines, isnever static, with geographers always seekingto either extend and consolidate particularways of thinking and doing or to develop newunderstandings of the unfolding relationshipbetween people, place and environment. Farfrom being a discipline preoccupied with themere accumulation of facts about the world,geography is a discipline where our under-standing of the world is constantly beingevaluated in the light of new ideas and think-ing, with empirical projects always informedby notions that some forms of knowledge andways of knowing may be more productive orvalid than others. Empirical studies of whatappears to be happening in particular contextsthus build up into wider theoretical accountsthat, in turn, drive new explorations of howpeople, place and environment are entwined incomplex and relational geographies. Withoutthis diversity of thought and sense of progres-sion – i.e. the idea that we are moving towardsa more productive understanding of the waythe world works – geography would long agohave become an intellectual backwater, ratherthan the vibrant, vital and varied discipline thatmany currently believe it to be.

This book is based on the premise that textsplay a crucial role in this story of disciplinarydevelopment. More specifically, it works withthe assumption that particular texts can be readand interpreted as symptomatic (and perhapstotemic) of key transitions in the ways that we

think, practise and write geography.The widestpossible definition of a geography text mightinclude conference papers, journal articles,book chapters, literature reviews, workingpapers, online articles, monographs, studenttextbooks, dictionaries, encyclopedias, readers,gazeteers, maps, and atlases. All of these may, indifferent ways, presage important shifts in theway geography is conducted. Yet in this volumewe want to focus on one particular kind oftext: the book. More specifically we want tohone in one type of book – the authoredmonograph.

Despite some concern in the discipline thatpublishers are less and less willing to publishmonographs, preferring instead to publishstudent-oriented texts like this one, we focushere on authored monographs for two princi-pal reasons. First, while most monographs areempirical – in the sense they seek to describeor map a particular aspect of the world – manyalso make important theoretical statementsabout the way that geographical knowledgesshould be constructed and disseminated. Formany, this is the acid test of a geographicresearch monograph: for a book to make acontribution that furthers the discipline, itneeds to spell out possible routes towards amore relevant, ethical or viable geography byadvocating a particular approach to its subjectmatter. As such, monographs seek to transformgeographical thinking and praxis through asustained engagement with, and exploration of,a set of theoretical ideas as well as the detailingof particular empirical ‘facts’. They are often

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xiii

books with Big Ideas and Big Ambitions, andif their thesis gains currency they become keyreference works which are mined, re-workedand critiqued by subsequent generations.Authored texts thus become foils that stimulatenew ideas and thinking. This is not to say thatresearch articles, chapters, edited books, etc.do not make similarly important contributionsto debates concerning disciplinary progress.Indeed, papers in journals may often stimulateimportant changes in the disciplinary land-scape by providing rapid dissemination ofresearch findings. Yet we would argue thatauthored books often become the key mile-stones in disciplinary histories in ways thatarticles rarely do because they allow authors toconnect disparate empirical and theoreticalelements to develop a wider, more systematicand rigorous argument about the way theworld works.

Second, we focus on authored monographsbecause students are often referred to these ‘keytexts’, encouraged to engage with them inorder to understand particular modes ofthought and the history of the discipline and toreflect on the ideas contained within themwith respect to shaping their own geographicalthinking and praxis. Many courses on the his-tories and philosophies of geography are in factstories in which key authored texts are givendue prominence, with these key works deemedto have punctuated geography’s histories. Fromthe perspective of the present, a retrospectivereading of these works is often encouraged as away of understanding how we got to wherewe are today. Making oneself familiar with keytexts is part of any geographical education – formany educators, ‘thinking geographically’ issomething that only emerges from criticalreading and re-reading of geography’s ‘ur-texts’. Contemporary libraries, we note, areoften all-too-ready to dispose of older booksto make way for new tomes, but most retainthose volumes which educators suggest are‘classics’ which students will always need toreturn to.

This brief discussion indicates that thisvolume is necessarily restrictive in its defini-tion of what a ‘key text’ is. Not only do weignore papers, chapters, edited collections,readers and conference presentations, we alsodisregard a number of important student-oriented textbooks. Ron Johnston (2006) arguesthat textbooks are particularly important ininstitutionalizing particular approaches togeography, given that they often proclaim tobe ‘objective’ or authoritative introductionsto the discipline. Although we would notnecessarily disagree, given that such volumesare written to be accessible to as wide anaudience as possible, we feel there is littleneed to provide a guide to such texts. Nor dowe consider some of the important historiesof the discipline (Johnston’s, 2005, Geographyand Geographers – now in its fifth edition –being a prime example, alongside Peet’s,1998, Modern Geographical Thought, or Cloke,Philo and Sadler’s, 1991, Approaching HumanGeography) for the same reason. Other booksthat have gone through multiple iterations,and have hence been integral in policingthe boundaries of the discipline, are alsoprecluded from consideration here (e.g. theDictionary of Human Geography, now in itsfifth edition, the Companion Encyclopedia ofHuman Geography, now in its second, theInternational Encyclopedia of Human Geography,and so on).

By focusing on specific authored texts weare accordingly not trying to suggest othertypes of text are insignificant in shaping geo-graphical thought. Yet we feel that, by theirvery nature, the type of books we focus onhere were often written for an academic peeraudience rather than a student one. For theuninitiated, many appear remarkably dense,use difficult language and work throughcomplex theory and unfamiliar examples,and are not accessible in the same way that atextbook might be. Many talk to debates andsocial-economic contexts that have longsince disappeared or are only just coming

xivÿÿEDITORS’ INTRODUCTION

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xiv

into being, or refer to events or processes thatwould only be known to those who havegrown up in particular contexts. In a sense,this is precisely why lecturers are often keento refer their students to these texts – theywant to challenge them, encourage them todevelop their skills of critical reading andappreciate the impact of particular thinkersand ways of thinking on the practices ofgeography.

This given, Key Texts provides an introduc-tion to 26 books that we argue have made asignificant impact on the theoretical underpin-nings and praxis of human geography in the last50 or so years. The book’s ambition is two-fold.First, it aims to serve as a primer for students,introducing them to specific monographs,exploring the nuance of the authors’ argumentsand explaining why they should take the timeand trouble to engage with the text itself ratherthan summaries provided in textbooks. Tothat end, each of the entries in this volume isan interpretive essay that highlights: the posi-tionality and biography of the author(s); thesignificance of the text in relation to the geo-graphical debates and issues current at the timeof writing; the book’s main arguments andsources of evidence; its initial impacts andreception; how the book was subsequently cri-tiqued, evaluated and incorporated into thegeographical imagination; and how the bookchanged – and continues to influence – thepractices of geography.

Secondly, the book seeks to contribute toongoing debates over the production of geo-graphic knowledge by posing some importantquestions of what constitutes a ‘key text’. Itis of course crucial to ask why some booksbecome privileged, and to consider how disci-plinary histories become written around keytexts as well as key thinkers (see Hubbard et al.,2004). In recent years scholars interested in thehistory of geographical knowledge productionhave come to argue that geographical endeav-our occurs in a highly diversified landscape,shaped by issues such as educational training,

personality and location, friendships and colle-giality, disciplinary gatekeeping and access todisciplinary networks, prevalent trends andvested interests, and wider debates on therelevance and value of the academy and thefunding of higher education. In other words,it has become recognized that geographicscholarship is shaped by multiple factors, somepersonal, cultural and social, and some thatare more political and economic in nature(Barnes, 2002). While the academy is a place ofcollegiality and collaboration, it can also be acompetitive environment with most acade-mics working both for themselves and theirinstitutions as they seek to acquire kudos,funding and intellectual respect. In the UK,for example, departments are in competitionwith one another under the influence ofa Research Assessment Exercise which isfocused on research outputs. The influence ofRAE culture on the shape and form of insti-tutional geography is still to become clear, yetthe potential to be identified by one’s institu-tion as a ‘research inactive’ academic createsimmense pressures to work in particularways, and to work to identified assessmentcriteria.

These diverse factors shape what kinds ofideas and praxis become mainstream, and, inturn, influence who become recognized asthe key thinkers in a discipline (though, asthe exchange in Environment and Planning A37: 161–187, illustrates, there are certain dan-gers in trying to name those who are mostinfluential in the discipline). However, asdebates about English language and Anglo -American dominance in the production ofgeographic knowledge have highlighted,knowledge production has both a history anda geography, with some scholars located inkey centres, others on the periphery (seeBerg and Kearns, 1998; Garcia-Ramon, 2003;Kitchin, 2005; Paasi, 2005). As such, it isimportant to acknowledge that the produc-tion of geoographical knowledge is messy,contingent, relational and political, meaning

EDITORS’ INTRODUCTIONÿÿxv

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xv

that any history of the discipline needs to bewritten in ways that are cognizant of suchpolitics. Key Texts is no exception.

The authors of the entries in this volumewere not asked to explicitly address suchissues, but often raise questions of authority,privilege and hierarchy in their contributions.As many acknowledge, it is seldom the casethat a book becomes significant because ofsheer good fortune. Rather, a book catchesand helps further promote the zeitgeist of aparticular moment, crystallizing the authors’thoughts at the time, and often many otherpeoples’, working within and outside thediscipline. Indeed, it is likely that many con-ference papers and journal articles precededthe publication of a ‘key text’, especially givenhow long books take to write. And it is oftenthe case that other related or similar booksappeared at roughly the same time. What dis-tinguishes a ‘key text’ – a book that was takenup or ‘enrolled’ within dominant disciplinarynetworks – is that it said something significantthat had widespread appeal and which chal-lenged its readers to think differently aboutthe world. It simply did not repeat argumentsemerging within the journal literature, itextended these, amplified them and illustratedthem with rich empirical material. Of course,other books might have been saying similarthings, but were perhaps saying them lesswell, with less conviction or were promotingslightly different viewpoints. And while thebook’s content is the crucial factor determin-ing its reception and uptake, there is nodenying that issues of authorship and author-ity are also significant. In short, it matters whowrote the book. Some books are highly antic-ipated given their author’s existing reputation;others emanate from unheralded sources butbecome best-selling works. Most, however,are published to indifference, and neverachieve anything more than modest sales:the ability of an author to promote theirbooks through their other activities and

networking can be vital in ensuring a bookhas a shelf-life.

What is clear is that some books emergeto become ‘classics’ within the discipline. Theauthors of such books may (reluctantly orotherwise) become gatekeepers within thediscipline – recognized as ‘key thinkers’ – inthe sense both they and their books are heldup as promoting a particular way of doinggeography. As such, many of the entries inthis volume highlight interesting debatesabout the politics of geographical knowledgeproduction. So too does our choice of keytexts. In making the difficult choices we havemade about which books are worthy of con-sideration, we realize that we are not simplyreflecting established knowledges, we areactively perpetuating particular value claimsabout whose views matter, and which booksshould be read. That given, we are certainthat the books we exclude will be as signifi-cant for some readers as the books weinclude – and we hope that these exclusionsare interrogated as meaningfully and produc-tively as was the case for the companionwork for this text – Key Thinkers on Space andPlace (see especially the review forum inEnvironment and Planning A 37: 161–187).Given the controversy that our selection willundoubtedly excite in some quarters, it ishence necessary to spend some little timeoutlining the criteria for selection that wehave employed in this volume.

Which key texts?

When drawing up a list of some of the mostimportant texts in human geography, we areforced to make some difficult decisions as towhat we understand the boundaries of humangeography to be. Indeed, even if we are happyto exclude key texts in physical geography(the subject of a volume yet to come?), thereare certainly books on the relation of the

xviÿÿEDITORS’ INTRODUCTION

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xvi

physical and human world that have beensignificant in changing the purview of geog-raphers and their understanding of what thesubject matter of the discipline might be.There is also the vexed question of what dis-tinguishes a geographical text from otherkinds of text, given many key interventions indebates over space and place have been madeby those who do not identify as geographersor claim to be writing for a geographicalaudience. The boundaries between humangeography and planning, urban studies, history,anthropology, sociology, philosophy and areastudies have often been highly porous andgeographical thinking has certainly borrowedand benefited significantly from texts writtenby those located in other disciplines.

Given these fluid and indistinct bound-aries, our first criterion for selection was toconsider only books written by people whoself-identified as geographers, and were writ-ing, first and foremost, for a geographicalaudience. This is to take a narrow view of thediscipline, perhaps, but is in keeping with oneof the most widely accepted (if hoary) defin-itions of geography: that is, geography is whatgeographers do. Secondly, we limited ourchoices to books published in English in thelast 50 years. While this is a somewhat arbi-trary cut-off date, there are good reasons forsupposing that students will be most fre-quently steered towards these texts: manyrecent, student-friendly histories of the disci-pline tend to start with the post-war shiftfrom regional description to a theoreticallyinclined spatial science tradition (i.e. what iscommonly referred to as the ‘quantitative rev-olution’); students are most likely to besteered to books whose ideas still have cur-rency in contemporary debates (and thesetend to be the most recently published); manyuniversity libraries simply do not have anextensive catalogue of books dating back tothe pre-World War Two years, and studentsundertaking courses in Anglophone countries

often lack an advanced proficiency in lan-guages other than English that would preventthem from engaging with non-English texts.

Within these broad parameters, we still faceddifficult decisions about what constituted a ‘keytext’ and which ‘key texts’ to include. One strat-egy to aid our selection might have been toconsult the citation rates for different books (i.e.the number of times a book has been referredto in other books and articles). There is sometradition of using citation analysis to identifythe ‘weavers and makers’ of human geography(e.g. Bodman, 1991), and online databases andsearch engines (Google ScholarTM or the ISIIndexes) certainly facilitate such analyses. Yetnot all such analyses are robust and reliable, andwe should remain mindful that not all citationsare favourable. Equally, some books appear tobe more cited than read (a charge made in atleast one of the chapters in this volume), andself-citation can often inflate the apparentimportance of a given text. Suffice to say, mostof the books in this volume are well-cited(some more so than others), but not allof the most cited books in geography areincluded here.

Another way of honing in on the key textswithin the discipline might have been to selectthe best-selling texts. However, for a variety ofreasons, sales might not be a good indicator ofsignificance. As we noted above, textbooks,especially those designed for ‘introduction tohuman geography’ courses, tend to have signif-icantly more sales than research monographs.This is because monographs are generally tar-geted at the author’s peer group rather thanstudent masses, and so might have limited sales.They might, however, have hundreds morecitations than a textbook which sold in farhigher numbers (suggesting that they swaymore influence). Further, geography books sel-dom (if ever) break into the best-seller charts,with few geographers ever having adoptedthe position of a genuine ‘public intellectual’(see Ward, 2006; Castree, 2006 on public

EDITORS’ INTRODUCTIONÿÿxvii

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xvii

geographies). Some texts may well achieve salesbeyond geography, reaching student and acad-emic audiences in cognate disciplines, but veryfew break through to ‘discerning’ public audi-ences in the same way that, for example,historical or archaeological books currently doin the UK. This does not mean that the keytexts we profile have not sold well, with somefeatured here having gone to multiple editionsand repeat print runs. Irrespective of this, wewould claim that the acid test of a key text isnot its popularity or citationality, but itslongevity; that is to say that their impact is bestmeasured in terms of their influence on subse-quent texts.

While mindful of citations, sales, andlongevity, we chose to narrow our selectionfurther by consulting with colleagues fromacross human geography as to what booksthey felt merited inclusion based on theirexperience as researchers and teachers. Fromthat extended list we whittled the booksdown to the publisher’s limit of approxi-mately 25 (given page length constraints).Here, we tried to provide a regular temporalspacing of books, including texts publishedwithin each decade; to include texts that wereimportant within specific sub-disciplinesas well as human geography as a whole; andto include texts that engaged with and pro-moted the many ‘-ologies’ and ‘-isms’ thathave permeated recent geographical thought.We also took the decision to try to includesome quite recent books that we feel havethe potential to become ‘key texts’ giventheir initial reception and how quickly theirideas have permeated the discipline.

This then is not a random selection ofbooks. It is a set of books that we believe areworth reading, either individually or collec-tively. Each book has made an importantintervention not just within a given sub-discipline (e.g. urban, rural, social, economic,political, historical or cultural geography) butshaped the wider practices and imaginations

of human geography. Indeed, if one were tocritically read all the books included in thisvolume, one would have a very good grasp ofgeographical theory and practice over thepast 50 years. It is nonetheless a subjectivelyderived list and we would in no way claimit is the list of the most influential booksin human geography. Other geographerswould have of course drawn up their ownlists and may be distraught to find some oftheir favourite texts excluded here (possiblyincluding their own books!) They will nodoubt suggest that our list bears the imprintof our own particular exposure to geographythrough Anglo-American traditions, our ownresearch interests and expertise, our own peernetworks – all of which might have impingedon our judgement as to which books haveexercised most influence on geographicalthought. This is unavoidable, not least becauseany book published prior to 1990 pre-datesour professional experience in the discipline(indeed the earliest books were written beforeall three of us were born!). But, for all its flaws,we feel that the 26 entries in this book dosome kind of justice to the diversity of humangeography practised in the last 50 years, witheach having ‘pushed the envelope’ intellectu-ally, methodologically and philosophically,shaping the landscape of human geography aswe see it today.

How to use this book

The most important thing to stress about thisbook is that it is not intended to substitute foran engagement with the text itself. While eachentry provides a synopsis of the book in hand,it is necessarily brief, and often glosses over thenuance of the book’s argument in the interestsof identifying its essential arguments. Thebook is designed to be a primer, to be readalongside the text, and seeks to encourage a crit-ically informed engagement and exploration

xviiiÿÿEDITORS’ INTRODUCTION

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xviii

of the intricacies of each book. Each ofour entries thus provides useful backgroundcontext that might help the reader understandthe situation within which a particular textwas written (i.e. the wider social and politicalconditions that prevailed at the time, as wellas the disciplinary preoccupations whichprompted the authorship of that volume). Italso considers how the book was received bythe wider academic community, noting theway that the book was reviewed and howpeople reacted to the ideas being forwarded.In many cases, the entry also documents howpeople engaged with the ideas and took themforward in different ways. Accordingly, eachentry considers the way in which the bookaffected and shaped the geographies that suc-ceeded it, through a critical appraisal of thebook’s key thematic concerns, its particularapproach and espousal of specific philosophiesof geographic knowledge production.

Those asked to write chapters for this vol-ume were asked to do so because we felt theywould be able to offer a critical, reflective andbalanced assessment of the text in question.Inevitably, many of our authors have writtenabout a book that has profoundly shapedtheir own life as a geographer, perhaps influ-encing their own approach to the discipline.Some are explicit about this, and provide ahighly personalized account of how the bookinfluenced them; others are less forthcoming,and instead try to produce an account inwhich their own opinions are harder to dis-cern. But in either case it is highly unlikelythat the authors have produced an unbiasedinterpretation, with most likely to be predis-posed towards (or, occasionally, against) thebook they are considering. This is unavoid-able: as we have stressed, there is no suchthing as an objective assessment, and there isno one who is in a position to ultimatelydetermine the value of a text. What thereforeneeds to be remembered is that each of ourentries comes with a point of view that you,

and your tutors or colleagues, might notshare.

Our hope is thus that this book will provea useful companion for students seeking toengage with geography’s rich histories (andnot a crib that will dispense with the needto actually read each book in question) andmight provide a useful template for how stu-dents might critically engage with texts ingeneral (indeed a useful exercise is for stu-dents to undertake critical reviews of textsnot included in this volume). The bibliogra-phy that closes each entry consequentlyoffers numerous departure points for furtherexplorations of the text and its author, andthroughout we include frequent cross-references to other entries in this volume (aswell as the companion text, Key Thinkers onSpace and Place). As we suggested above, whenwe commissioned people to write the entriesin this book, we chose people who we feltmight have a close knowledge or perhapseven affinity for the book in question. Eachis also knowledgable about the place of thebook in the wider disciplinary landscape andin the ‘geographical tradition’. Yet we implorestudents not to take their views for granted,as their summation is not necessarily the onethat other geographers might make. Perhapstheir reading contradicts your own, or comesto different conclusions. Any one of thebooks featured here is open to multiplereadings, and sometimes even readings thatthe author never intended. Such is the poly-semous nature of text. In the final analysis,we hope that this book provokes people toread and re-read these texts, subject them todiscussion and interrogation, and form theirown situated interpretations. Perhaps, intime, these engagements might even stimu-late the production of new key texts!Whatever, we hope Key Texts is a useful andstimulating text that is much, much morethan a lesson in geographical navel-gazingand nostalgia.

EDITORS’ INTRODUCTIONÿÿxix

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xix

xxÿÿEDITORS’ INTRODUCTION

Secondary sources and referencesBarnes, T. (2002) ‘Performing economic geography: two men, two books and a cast of

thousands’, Environment and Planning A 34: 487–512.Berg, L.D. and Kearns, R.A. (1998) ‘America unlimited’, Environment and Planning D:

Society and Space 16: 128–132.Bodman, A. (1991) ‘Weavers of influence: the structure of contemporary geographic

research’, Transactions, Institute of British Geographers 16: 21–37.Castree, N. (2006) ‘Geography’s new public intellectuals’, Antipode 38: 396–412.Garcia-Ramon, M.D. (2003) ‘Globalization and international geography: the questions of

languages and scholarly traditions’, Progress in Human Geography 27: 1–5.Hubbard, P., Kitchin, R.M. and Valentine, G. (2004) Key Thinkers on Space and Place.

London: Sage.Johnston, R.J. (2006) ‘The politics of changing human geography’s agenda: textbooks and

the representation of increasing diversity’, Transactions, Institute of British Geographers31: 286–303.

Kitchin, R. (2005) ‘Disrupting and destabilizing Anglo-American English-language hege-mony in geography’, Social & Cultural Geography 6: 1–15.

Paasi, A. (2005) ‘Globalization, academic capitalism and the uneven geographies ofinternational journal publishing spaces’, Environment and Planning A 37: 769–789.

Review Forum: ‘Key Thinkers on Space and Place’, Environment and Planning A 37: 161–187.Ward, K. (2006) ‘Geography and public policy: towards public geographies’, Progress in

Human Geography 30: 495–503.

Hubbard-Introduction:Hubbard-Introduction.qxd 4/14/2008 7:43 PM Page xx

INNOVATION DIFFUSION AS SPATIAL PROCESS(1953): TÖRSTEN HÄGERSTRAND

Bo Lenntrop

The diffusion of innovations – the origin anddissemination of cultural novelties – is an area ofstudy which concerns all sciences dealing withhuman activity, including, not least of all, culturaland economic geography. (Hägerstrand, 1953: 1)

Introduction

It is difficult to grasp the importance ofTörsten Hägerstrand’s key work on innovationdiffusion – his doctoral thesis from 1953 –without an appreciation of the historical con-text in which the work was conceived andprepared and the fact that it was first translatedinto English by Allan Pred some 14 yearsafter submission. Most notable for setting outtheories of spatial diffusion and adoption,Hägerstrand’s early research also contains thekey to the later development of his ideas. Oneexample is time geography, which became for-malized in the 1960s, but whose conceptualroots were to be found in Hägerstrand’s writ-ing in the 1940s and 1950s.

Hägerstrand arrived at Lund University inthe late 1930s. His interest was directed, moreor less by chance, toward migration and hestarted to work on a project intended to chartthe entire demographic development of aconsiderable geographical area of Swedenfrom 1840 to 1940. Building an impressivecollection of data, this enormous undertakingleft Hägerstrand with a profound empiricalunderstanding of demographic development.

At the same time he also developed a deepertheoretical proficiency, setting in motion hisparticular geographical worldview. More par-ticularly, the systematic collection and analysisof data on the life courses of a population overa century contributed to the germination andgrowth of a foundational idea; namely theimportance of analysing spatial processes.

Over time Hägerstrand absorbed impor-tant theoretic ideas and trends from beyondSwedish geography. These did not emanatefrom the regional perspective then dominantin university teaching, with Hägerstrand stat-ing ‘lectures in regional geography wereabominably boring … Geography appearednot as a realm of ideas or a perspective on theworld but as an endless array of ency-clopaedic data’ (Hägerstrand, 1983: 244).Rather they came to his attention through achance acquaintance. His future wife, Britt,was then working for the ethnologist SigridSvensson who was conducting research andpublishing books on diffusion processes,and one of his school colleagues had a burn-ing interest in numerical analyses and indeveloping computers and was an early visitorto the US.

Having gained a sound knowledge ofdemography and ideas on diffusion coursesand simulation models, the foundations of hisdoctoral thesis had fallen into place. In manyrespects this work was to mark a decisivebreak with the then dominant tradition ofregional studies. Hägerstrand’s principal aim

1

Hubbard-Ch-01:01-Hubbard-Ch-01.qxd 4/14/2008 7:40 PM Page 1

in his doctoral thesis was not to present abroad regional description of an area, butinstead to investigate and illuminate a prob-lem. That the material concerned a specificarea, argued Hägerstrand, was a regrettablenecessity and not a methodological finessewhich in itself marked a stand againstregional geography.

Innovation diffusion: the Swedish version

For obvious reasons a doctoral thesis writtenin Swedish does not reach a large internationalreadership. Nevertheless, it is informative tocomment on its immediate reception, 14 yearsbefore it was first published in English. Thefirst academic review of Hägerstrand’s doctoralthesis was published in 1953 – in the sameyear he was awarded his doctorate. Thereviewer was Edgar Kant, previously professorin Tartu in Estonia, but who had by then spentmany years in Lund at the Department ofGeography. Kant was widely read in interna-tional scholarly literature and he was also thefirst opponent at Hägerstrand’s disputation.

Kant opens his review with a discussion ofthe respective pros and cons of research spe-cialization. Kant was of the opinion that thedisadvantages associated with specialization‘begin to become apparent when large lacu-nae arise leaving poorly-lit areas in borderzones, as monadnocks of the total ignorance’(Kant, 1953: 221). It is useful to recite Kant’sconcluding comments in order to relateexactly why he saw Hägerstrand’s thesis aspioneering.

The author has, to a noteworthy extent,utilized new methods and establishednew links to neighbouring disciplines.This must present itself as innovative tothose who perceive geography as indissol-ubly bound to traditional methods ofinvestigation and research subjects, such

as landscape analysis, which have as theironly or primary task ascertaining interac-tions between man and nature. … It maytranspire that the author’s longest expedi-tions into the unknown border- andtwilight-zone have been but excursionsleaving many areas as yet largely unex-plored. Those who follow in his path can,however, draw benefit from his pioneer-ing work and fashion new riches. (Kant,1953: 225)

In his doctoral thesis Hägerstrand investi-gated the changing extent of propagation ofcultural artefacts. He did this by selecting sixspecific indicators; three for agriculture (statesubsidies for improving pastures, control ofbovine tuberculosis, soil mapping) and threemore general indicators (postal money trans-fer, automobiles, telephones). The choice ofindicators was guided by the need that theseshould be localizable to coordinates and thattheir development over time could be fol-lowed with a very high degree of precision.Furthermore, it was necessary that the indi-cators had been adopted by a sizeableproportion of the population.

The next stage in the investigation was toestablish reduction bases for the indicators.This was necessary because it would havebeen meaningless to work with absolutenumbers of acceptors. Hägerstrand carefullyanalysed the population (or demographic)development in the region and the locationand size of each of the respective homesteadsand residential apartments in order to con-struct the ‘reduction bases’ against which thenumber of acceptors should be matched.

An important section of the thesis, encom-passing nearly 100 pages, then deals with theactual diffusion process. Hägerstrand’s pri-mary objective here is not to identify specificdetails in this diffusion process, but rather togeneralize about the characteristics thatcould serve as the basis for subsequent oper-ationalization in models. However, it was

2ÿÿBO LENNTROP


important to observe certain characteristicsas these generated variations in the diffusionprocess. A significant degree of the discrep-ancy between the indicators is based on thelevel of state intervention, for example inpromoting the introduction of controls onbovine tuberculosis. Car purchase was princi-pally a matter of private decision making,even if state legislation played a certain role.The introduction and diffusion of the tele-phone in the region of investigation wasaffected by the manner and the speed withwhich the electricity network was developed.Consequently, the diffusion of the six indica-tors reveals different courses and show howthey are influenced to lesser or greaterextents by planning and policy at nationaland regional levels. The larger part of thisdiscussion is of more general interest, and isnot confined to its historical or regionalcontext.

With the help of this detailed empiricalknowledge, Hägerstrand then formulated aseries of experimental, stochastic models toshow how innovations spread within a popula-tion. The first model was very simple and issteered entirely by chance and provides a picturethat most closely approximates the manner inwhich a rumour spreads through a population.Hägerstrand therefore focuses on how differ-ences in acceptance and how unevenly spreadinformation could be modelled. Concerningthe indicators for agriculture, he examines how,for example, farm size can influence the propen-sity to accept an innovation. In his studies,Hägerstrand identifies the importance of theproximity and thus comes to deal with privateinformation diffusion and the question of howthis should be modelled. He examines thechorological characteristics of information andhow migration and telephone data can be uti-lized to describe the extent of the range ofprivate information. In this particular casestudy he finds that migration data provides thebest approximation as the telephone networkwas incomplete at the time of the study.

Furthermore, there were zonal boundaries thatacted to deform the contact field.

The resulting models operated with a real,coordinate based population. The diffusion ofan innovation in a population is determined byconstructing a so-called mean information field(a concept still outlined in many standard text-books on human geography), which illustrateshow the probability of making contact withanother individual decreases with increased dis-tance. The empirical basis for the MIFscomprising 5 × 5 cells (each cell being 1 squarekilometre) is grounded on migration data. Thematrix illustrates the probability of a contactbeing made from the central cell to one ormore of the surrounding cells. The probabilitiesare cumulated (from 1 to 10,000) and each cellis attributed an interval proportional to theprobability. The matrix is centred on an indi-vidual having knowledge of the innovation inquestion. By drawing a random number onedecides which cell (interval) is met. The matrixis used in the manner of a floating grid – i.e. itmoves over the fixed population and is centredover those individuals who have knowledge ofthe innovation in each generation and who areprepared to spread this knowledge. In this man-ner the innovation is continually diffused overtime to new generations and gives rise to spa-tial patterns of acceptors, which are randomlydetermined but always within the given proba-bility intervals. Even when the rules of thegame and the probabilities remain unchanged,the results of different simulations are often verydivergent as a result of the stochastic factor.

In these models Hägerstrand experimentedwith different forms of physical and socialbarriers. The propensity to accept an innova-tion was modelled (for example, a personmust be ‘hit’ two or three times before accept-ing) and the degree of correspondence withthe empirically ascer t ained patterns becameincreasingly close. As a result of the degree ofcorrespondence between the actual and themodelled courses, Hägerstrand was able toconclude that the key elements deciding the

TÖRSTEN HÄGERSTRANDÿÿ3


course of innovation diffusion had beencaptured. It is worth noting that all of thesecomprehensive modelling experiments werecarried out by hand as computers had not asyet been developed. Each of the thousandupon thousand of random numbers was col-lated from tables in a strict order.

The first international commentary onHägerstrand’s doctoral thesis was written byJohn Leighly and published in the GeographicalReview in 1954. Leighly, then Professor atBerkeley, was well acquainted with Swedishgeography, which he had considered for severalyears (Leighly, 1952). Leighly emphasized hisadmiration of the scope and precision of theempirical work, not least concerning the maprepresentation: ‘His “relative” mapping usesrefinements (isarithms at numerical intervalsgiven by geometric progressions, interpolationof isarithms by logarithmic intervals) that makeit exemplary’ (Leighly, 1954: 440). However,and in a similar manner to many othercommentators, Leighly viewed Hägerstrand’sinterpretation of innovation diffusion and thefollowing operational modelling as the ‘culmi-nation’ of his work. Leighly (1954: 441)concludes by commenting that anyone doingresearch in this area ‘cannot afford to ignoreHägerstrand’s methods and conclusions’.

While the attention surroundingHägerstrand’s doctoral thesis waned in theperiod following this appraisal, the same can-not be said regarding interest in his theoreticalideas and methodological approach. Aspectsof his principal research were gradually dis-seminated via lectures, conferences and minorpublications. Shortly prior to the completionof his doctoral thesis, Hägerstrand laid out hisresearch field in an article entitled ‘The prop-agation of innovation waves’ (Hägerstrand,1952). By 1967 he had published threearticles on migration, two on diffusion,and one on simulation. A very widelyread and often cited article is ‘A MonteCarlo approach to diffusion’ (1965), which

directly deals with models and is publishedin four journals and has even been trans-lated into Japanese.

Innovation diffusion: theEnglish version

Although Hägerstrand’s research had becomerelatively well known on the internationalscene, the broader understanding of hisresearch remained fragmentary. Accordingly,both Gilbert White in Chicago and AllanPred at Berkeley argued that his doctoral the-sis should be translated into English. Predtook on this task as he was both fluent inSwedish and well versed in the specific areaof research; two attributes which Hägerstrandargued were prerequisites for the successfultranslation of his work. Pred also wrote aPostscript in which he introduced TorstenHägerstrand providing a detailed backgroundto the research, and how the field of researchhad developed.

The book was well received. However, thefact that 14 years had passed since the publica-tion of the original Swedish edition is at leastpartly evidenced by the content of a numberof more critical commentaries. Quantitativegeography had further developed and wasquickly establishing an important set of newmethodologies and theories (see Chapters 2, 3and 4). One particular line of developmentconcerned the statistical and objective com-parison of patterns in time and space, whichmade it possible to elucidate more preciselydifferences and similarities in these patterns. Inhis review, Gunnar Olsson pointed out that‘the evaluation of the model was based onmore intuition and visual inspection thanobjective statistical testing’ (Olsson, 1969: 310).None of the reviews published during the1950s had raised this issue.

Simulation had become a popular modusoperandi within the discipline, and against this

4ÿÿBO LENNTROP


background Olsson had somewhat criticallystated: ‘In conclusion, most of those who nowexperiment with Monte Carlo models seemto have missed the point. Thus it is rare thatthe relation between theory and model is asexplicit as in Hägerstrand’s own work’ (Olsson,1969: 311). This observation further supportsthe proficiency with which Hägerstrand hadsucceeded in fusing thorough empirical mate-rial with a well-developed command oftheory. This is what gives his work its credibil-ity and strength.

Richard L. Morrill, himself a former visitingfellow at Lund who also had experience ofsimulations, argued that although Hägerstrand’searlier work had been disseminated relativelywidely, the opportunity now existed for awider audience to discover that this was byno means a curious approach, but insteadconstituted a theoretical framework foundedon an unusually rigorous empirical grounding.Morrill (1969) went to particular lengths toemphasize the linkages between spatial pat-terns and individual behaviour, and argued thisrepresented a revolutionizing insight and apowerful break with tradition within thediscipline. In a more anecdotal vein, wemight also note that he concluded hisreview in Economic Geography by protestingagainst the ‘outrageous price’ of the book(then $16).

L.J. Evenden (1969) forwarded a compre-hensive and positive review in the journal SocialForces. He argued that: ‘… Hägerstrand allows aglimpse of a fertile imagination and, simultane-ously, teaches the useful lessons that patience isa research virtue, that theory is fundamentallyabout something, and that reliable theory in socialscience stays close to this something’ (Evenden,1969: 9). Evenden further argued that evidencethat Hägerstrand’s research had left a lastinglegacy was to be found in the wide range offields in which Monte Carlo simulation hadsubsequently been applied. Simulation was wellknown and widely used at the time of the

publication of the English edition, so there wasno question of its subject matter being receivedas breaking news. Evenden emphasized that thework represented a combination of a carefulfactual analysis and the best of traditionaland contemporary geographical scholarship,and contended that Hägerstrand’s doctoralthesis would long be considered a classicwork in the development of geographicaltheory.

The economist Harvey Leibenstein, thenbased at Harvard University, reviewed the bookin the Journal of Economic Literature. He main-tained that it was of interest for economistswho were interested in diffusion processes andlocation theory, as well as modulated process-ing of innovation diffusion. Leibenstein furthercontended that it was difficult to assess thevalue of the models’ predictions, in spite of allthe empirical information. He also found theapproach altogether mechanical – a critiquewhich one can at least partly agree upon – ifone did not take into consideration the factthat the translated thesis was 14 years old. Afurther point made by Leibenstein was that ithad been a mistake to translate the thesis in itsentirety. The meticulous account of empiri-cal material was, he deemed, utterly boringand would probably act to ‘… repel mostreaders’ (Leibenstein, 1969: 11). It is as wellto add that no such critique has been forth-coming from any of the geographers whohave reviewed the book; instead they haveviewed this detail as an extra bonus – stimu-lating for many, previously unknown formost readers.

As time goes by

While Hägerstrand’s research on diffusionwas largely well known prior to the publica-tion of the English version, it did open thepossibility for a deeper understanding of hisresearch. At first it was widely read and acted



to inspire many researchers, but one gains theimpression that it fairly quickly became awork that was deferentially cited for its volu-minous length and that the main elementsof his research were largely disseminatedthrough his articles. Another contributingreason was that Hägerstrand did not furtherdevelop his diffusion research. He abandonedresearch on diffusion courses at a relativelyearly stage in his career. However, two-thirdsof the references to his work in academicjournals concern this field of research.The English version of his doctoral thesis,with almost 400 citations, is by far themost cited.

There is still much to be gained from read-ing Hägerstrand’s doctoral thesis, particularlythe manner in which he constructed hismodels. It was his models and the use ofMonte Carlo-simulation that had the greatestimpact and these resulted in citations and ref-erences and acted as the models for manysubsequent variants. In a number of articlesone can find abridged discussions concerningfields of private information and the influ-ence of proximity in the diffusion of ideas. Areading of Hägerstrand’s doctoral thesis does,however, provide much more – his reasoningon propagation courses, on planning and pol-itics relating to diffusion processes, etc. Thereader will also perceive the consequentialinterplay between the empirical and thetheoretical.

Without an awareness of the doctoralthesis, it would be easy to disregard the impor-tance of empirical data for advanced ideaconstruction and pioneering research. Theimportance of compiling a base of knowl -edge from fundamental empirical work,with methodological and theoretical thinkingbeing intimately connected, is perhaps themost important lesson to be learned fromHägerstrand’s research on diffusion. Assuredly,for the contemporary readers, familiar withshorter journal articles and student-friendly

texts, Hägerstrand’s doctoral thesis is dauntinglycomprehensive. However, it is well worthreading as a classical scientific work and onewhich is also a tour de force in terms of map anddiagram presentation.

In his doctoral thesis Hägerstrand workedwith models that were operationalized byhand, but all along these were envisaged withfuture computer technology in mind. He wascognisant of the important role computertechnology would play, not least for geogra-phers, and in an article (in Swedish) publishedin the mid-1950s, he pointed out areas ofapplication for coordinate based informationand how this could be analysed. This laterinspired Hägerstrand’s early involvement withareas of application such as coordinate basedreal estate registers and the development ofcomputer mapping; a forerunner to GIS.

Viewed after a great number of years,Hägerstrand’s writing from the 1940s and 50sretains its vitality and relevance. His book waspresented in Progress in Human Geography asone of the discipline’s classic works (1992),with comments by Andrew Cliff and AllanPred and an authorial response by Hägerstrandhimself. His work in the field of diffusionresearch has also been cited in The Dictionaryof Human Geography where it is referred to atlength.

In his portrait of Hägerstrand, Flowerdew(2004) sketches a broader picture of hisresearch that does not confine itself to diffu-sion. However, he names this area of researchas one of Hägerstrand’s more important con-tributions, arguing that it included much inthe way of new thinking and that it greatlyinfluenced the wider research community. Thesame spirit, but expressed in other shades ofmeaning, is conveyed in the image forwardedby many researchers of Hägerstrand as one ofthe founders of modern human geography(Progress in Human Geography, 2005). For inter-ested readers a newly published bibliography isavailable (Lenntrop, 2004).

6ÿÿBO LENNTROP


Conclusion

At an early stage in his research Hägerstrandemphasized proximity – in both geographicaland social networks – as a crucial factor inaccessing and spreading ideas (not leastemphasized in time-geography). Through hiscontacts, which also partly arose by chance –he gained, as we have seen examples of above,new building blocks for his research. Absorbingnew impulses and ideas from outside is,however, insufficient without being able tochoose and then fuse together new and olderideas into new combinations and constructs.Hägerstrand had both the ability and a worldview with exactly these qualities. It was by nomeans a statistical understanding but a dynamicone, which after years of development also hadroots reaching long back in time.

A further trademark of Hägerstrand’sresearch was a focus on processes. In his doc-toral thesis he rejected the enterprise of

comparing conditions at different momentsin time and instead endeavoured to directlyelucidate the processes that formed differentspatial patterns. He grasped reality as inter-acting processes in which events and actionsof one moment conditioned events andactions of the next and not as variablesrelated to each other (Hägerstrand, 1992).This reasoning came to influence the entireconstruction of time-geography and its asso-ciated conceptual apparatus.

Despite the copious amounts of time andenergy that Hägerstrand invested in collectingand interpreting empirical material, he alwaysretained the notion of generalization and theaim of bringing forth the underlying factorsrather than losing one’s way in the surfacedetails. This is a trademark for his entire bodyof research and can easily be traced in hiswork. With the aid of a few concrete exampleshe was able to show fundamental relations orpaths of development.


Secondary sources and referencesEvenden, L.J. (1969) ‘Review of Innovation Diffusion as a Spatial Process by Törsten

Hägerstrand’, Social Forces 47 (3): 356–357.Flowerdew, R. (2004) ‘Törsten Hägerstrand’, in P. Hubbard, R. Kitchin and G. Valentine

(eds) Key Thinkers on Space and Place. London: Sage, pp. 149–154.Hägerstrand, T. (1952) ‘The propagation of innovation waves’, Lund Studies in Geography,

Ser B. Human Geography No. 4.Hägerstrand, T. (1953) Innovationsförloppet ut korologisk synpunkt. Akad.avh. Meddelande

från Lunds Universitets Geografiska Institution. Avhandlingar XXV. Lund 1953. Hägerstrand, T. (1967) Innovation Diffusion as a Spatial Process. Translation and

Postscript by Allan Pred. Chicago and London: University of Chicago Press.Hägerstrand, T. (1983) ‘In search for the sources of concepts’, in A. Buttimer (ed.) The

Practice of Geography. Harlow: Longman, pp. 238–256.Hägerstrand, T. (1992) ‘Author’s response. In Classics in Human Geography revisited’,

Progress in Human Geography 16 (4): 543–544.Kant, E. (1953) Recension av Törsten Hägerstrand Innovationsförloppet ur korologisk

synpunkt. Svensk Geografisk Årsbok, Årg. 29: 222–225.Leibenstein, H. (1969) ‘Review of Innovation Diffusion as a Spatial Process by Törsten

Hägerstrand; Allan Pred’, Journal of Economic Literature 7 (4): 1213–1214.Leighly, J. (1952) ‘Population and settlement: Some recent Swedish studies’, Geographical

Review 42 (1): 134–137.


8ÿÿBO LENNTROP

Leighly, J. (1954) ‘Innovation and area under geographical record’, Geographical Review44 (3): 439–441.

Lenntrop, B. (2004) ‘Publications by Törsten Hägerstrand 1938–2004’, GeografiskaAnnaler, Serie B, Human Geography, 86 B (4). Updated version on www.keg.lu.se/Publikationer/Törsten Hägerstrand.

Morrill, R.L. (1969) ‘Review of Innovation Diffusion as a Spatial Process by TörstenHägerstrand’, Economic Geography 45 (3): 356–357.

Olsson, G. (1969) ‘Review of Innovation Diffusion as a Spatial Process by TörstenHägerstrand; Allan Pred’, Geographical Review 59 (2): 309–311.

Progress in Human Geography (2005) ‘Makers of modern human geography – TörstenHägerstrand (1916–2004)’, Progress in Human Geography 29 (3): 327–349.


THEORETICAL GEOGRAPHY (1962):WILLIAM BUNGE

Michael F. Goodchild

All geographers have essentially the sameproblem, to make sense out of the globe.(Bunge, 1966: xvii)

Introduction

Theoretical Geography first appeared in 1962,and in an expanded second edition in 1966(which is the volume cited here). To Cox(2001: 71) it is ‘perhaps the seminal text ofthe spatial-quantitative revolution. Certainlyin terms of laying out the philosophical pre-suppositions of that movement it had nopeer.’ But Cox goes further:

It was also the spatial-quantitative revolu-tion that gave impetus to conceptualprecision in the field … it was the prospectof measurement, of operationalization insome piece of empirical research, thathelped us discover the value of a carefulspecification of our concepts and anexamination of their consistency … So, ifwe want to see where we have come from,what our intellectual debts are, there arefew better places to start than TheoreticalGeography. (Cox, 2001: 71)

The rationale for the book, laid out in theIntroduction, is that geography is a science;that every science is defined by its domain ofknowledge, which for geography is the Earthas the home of humanity; that every sciencehas both a factual or empirical side and a the-oretical side; and that ‘there are many books

on geographic facts and none on theory’(Bunge, 1966: x). Moreover, given ‘the authoris a theoretical geographer’ (Bunge, 1966: x) –theory is what he does. Theory ‘must meetcertain standards including clarity, simplicity,generality, and accuracy’ (Bunge, 1966: 2), andonly through theory can we discover ‘thesepatterns, these morphological laws … so that(our) planet, Earth, fills (our) consciousnesswith its symmetry and ordered beauty’(Bunge, 1966: xvi).

The book and its contents

The medieval Christian idea of a geometri-cally perfect world centered on Jerusalem(Harley and Woodward, 1987) had beenabandoned in the Renaissance, and replacedwith a view that the Earth’s surface was infi-nitely complex. So, although science hadmade great strides in explaining other aspectsof the cosmos, at the same time, by apparentlyrejecting religious teaching, it had led to aconfusing view of human behavior, and ofour immediate surroundings. In Bunge’sview, it was only through geographic theorythat we could return our concepts of theworld to an emotionally satisfying and reas-suring simplicity.

For Bunge, there were simple laws to bediscovered about the Earth’s surface, and par-ticularly about the patterns of phenomenafound there. They would be found for bothhuman and physical phenomena, and because

2


the study of one could inform the other itmade sense for departments of geography tohouse both physical geographers and humangeographers. The key to understandingpatterns was geometry, which allowed theprecise description of pattern and, through itstheorems, allowed the researcher to reasonabout pattern and thence to theory.

The book opens with a chapter on themethodology of geography which makes theauthor’s position clear. Two examples are usedto focus attention on what will be the recur-ring geometric theme of the book. Regionalgeography, long dominated by descriptivework that emphasized the unique propertiesof places, could be put firmly within a scien-tific paradigm if one focused on devisingobjective and replicable methods for definingregions, and concentrated on statistical evi-dence instead of subjective impression. As asecond example, the shifting river courses ofthe lower Mississippi bore a striking geomet-ric resemblance to the historic developmentof north-south highways in the Seattle area,and similar explanations could be found ifthe geomorphological process of buildingnatural levees were compared with the socialprocess of commercial development alongarterial roads.

The second chapter, titled ‘Metacartography,’takes another icon of traditional geography,the map, and goes to great lengths to argue itsessential mathematical nature, and the degreeto which such core cartographic concepts asprojection, overlay, and generalization can beexpressed in the mathematical frameworks ofgeometry and set theory. A chapter on shapediscusses potential measures of this somewhatelusive property, while subsequent chapterscover sampling, topology, and geodesics. A fullchapter on central place theory reflects theearly 1960’s focus on this as a crucial elementof geography’s ‘theoretical turn’ (see Kelly,Chapter 23, this volume). The book hints atmany of the major developments in quantita-tive geography and geographic information

science over the coming decades, though ofcourse it is impossible to find anticipation oftoday’s emphasis on computation and com-puter networks.

But there is far more to the book than is sug-gested by this quick summary, as one mightinfer from the commentary from Cox quotedabove. The book needs to be understood withinthe context of debates that raged in geographyat that time, and in the context of the muchwider intellectual movement of which it is inmany ways the opening statement. It also needsto be understood within the ageless desire ofgeographers to rank among the respected disci-plines of the academy.

Debates within geography

To Bunge, the most important intellectualdebate in geography in the late 1950s and early1960s was that between the nomothetic and idio-graphic perspectives on science. Although theiractual positions were much more nuanced, geo-graphers often identify the perspectives with thenames of Schaefer (1953) and Hartshorne(1959) respectively. Bunge had been stronglyinfluenced by Schaefer, having been drawn tohis work by William Garrison while a student atthe University of Washington, and it is no acci-dent that the first citation in the book is tohis paper.

The nomothetic position holds that resultsin science are of value only when they aregeneral, applying equally at all locations inspace and time. For example, Mendeleev’sPeriodic Table would be far less significant ifthe elements behaved differently during leapyears, and Newton’s Laws of Motion wouldbe far less significant if they applied only inthe State of Minnesota. In other words,empirical science proceeds by abstracting andgeneralizing away from the space-time con-text in which all observations are made; andscientific results are applied by re-insertingthem into context.

10ÿÿMICHAEL F. GOODCHILD


For Bunge science was inherently nomo-thetic, and geography was a science. Its successwould be measured strictly by the number ofgeneral principles that it discovered, within itsdefined domain of knowledge. These wouldbe principles governing the patterns thatemerge on the Earth’s surface as a result ofsocial and physical processes, ranging from thedendritic patterns of stream channels to theregular spacing of settlements. The purpose ofgeography was to discover these principles,and the chapters of his book laid the founda -tion, by outlining a methodology and providingexamples.

Geographers are clearly going to find thisnomothetic position difficult, since the descrip-tion of variation over the surface of the planetis so much part of the intellectual traditionof the discipline, and much would be lost ifdescription were justified only if it led to gen-eral principles. So while physicists mightmarginalize mere description, geographers mustnecessarily treat it with more respect. The idio-graphic position holds that description hasinherent value, particularly if it is conductedaccording to scientific principles – for example,using generally agreed terms, and methods thatare described in sufficient detail to be replicableby others. But Bunge and others rejected thisposition, arguing that an idiographic geographywould always be a marginal science.

To Bunge, then, it was important that geog-raphy establish its credentials as a science. Thefirst edition appeared in 1962, at the very outset of what later became known as the quan-titative revolution in geography (Berry, 1993).Bunge was one of a group of graduate stu-dents who had been inspired at the Universityof Washington in the late 1950s by WilliamGarrison, Edward Ullman, and DonaldHudson. The group included Duane Marble,Brian Berry, Michael Dacey, Richard Morrill,John Nystuen, Arthur Getis, and Waldo Tobler.As these fresh PhDs gained positions in influ-ential universities in the US, their approachrapidly caught the imagination of others,

including Leslie King, Leslie Curry, MauriceYeates, Peter Haggett and Richard Chorley inthe UK, among others. Further key textsappeared, including Haggett’s Locational Analysisin Human Geography (1966) (see Charlton,Chapter 3, this volume), Chorley’s and Haggett’sModels in Geography (1967), and King’s StatisticalAnalysis in Geography (1969). New journalswere established, positions created in almost alldepartments, and by the end of the decade itcould truly be said that geography had beentransformed.

But it would be oversimplifying to suggestthat this transformation merely altered thebalance between nomothetic and idiographicvisions of science. As Cox argued in thepassage quoted earlier, Bunge also con-tributed to a transformation of geographicpractice that included all aspects of the scien-tific enterprise. It came at a critical point,when computers were becoming available asengines of analysis and modelling. His bookmakes a passionate case for mathematics – thelingua franca of science – with its well-definedand widely accepted terms and its formalreasoning, and for particular applications ofgeometry to the analysis of geographic pat-tern. It argues the merits of quantification,and of the formal reasoning of statisticalinference; moreover, as we have seen, it is atpains to elaborate the relationship betweenmaps, the traditional research tool of the geo-grapher, and mathematics, the pervasive toolof science.

Reading the book again after 40 years, oneis struck with the sense of excitement of theearly quantifiers, and by a rhetorical style thatclearly emulates the sciences and is rarelyencountered in geography today. In theintroduction to the second edition, publishedin 1966, Bunge writes:

the discovery of a great number of pat-terns of location during the fall of 1962was helpful. We now know why politicalunits have their overall shape, why rivers

WILLIAM BUNGEÿÿ11


are dendritic, and why alluvial fans andBurgess’s rings are such close spatialcousins. Also, the problem of the con-struction of the isometric surface is muchbetter understood. (Bunge, 1962: xiv)

Nevertheless the author cannot resistcontinuing:

To see region construction, one of the lastpreserves of the non or anti-mathematicalgeographers, crumble away before theever growing appetite of the computingmachines is a little unnerving even for ahard case quantifier. (Bunge, 1962: xiv)

The revolution was well on the way tovictory.

Counter-revolution

Where as Theoretical Geography may indeedhave been the ‘shot heard round the world,’ andwhile it provided a general outline of how tomake the pitchforks, guns, and guillotines thatwould be needed before the revolutionariescould prevail, much of its content does notstand up well to detailed analysis. While itargues strongly for mathematics, there is pre-cious little actual mathematics in the book, andit fell to others such as Alan Wilson (1970) toprovide the rigorous theorizing. It is easy to seeparallels between the behavior of floating mag-nets in a dish of water and the behavior ofvendors positioning themselves on a featurelessplain (Bunge, 1962: 283). But on more carefulanalysis it is obvious that the laws of magneticforce do not have precise analogs in the laws ofentrepreneurial behavior. Bunge was (and stillis) the passionate visionary, but it took otherswith a deeper knowledge of mathematics anda patient commitment to rigor to begin tobuild the peer-reviewed literature of quantita-tive geography. Moreover, it was not longbefore the counter-revolution began.

One of the first salvos was fired by Sack(1972) (see also Sack, 1973; Bunge, 1979).Certainly one could find consistent patternsacross the Earth’s surface – the geometricform of the river meander, Horton’s laws ofstream number (Horton, 1945), the regularspacing of settlements in Southern Germany –but geometric pattern in and of itself couldnot be regarded as explanation, and could nottherefore provide the sense of satisfaction thatcomes from knowing how a pattern iscaused. Spatial properties such as latitude, dis-tance, or direction could never be said totruly explain anything. Sack saw Bunge as theprimary exponent of a confusion betweengeometric pattern and explanation, andattacked his notion that it was an interest inspace and spatial properties that defined thediscipline.

This issue remains an important one today.The spatial tradition is still a key part of thediscipline, though only one of many in a plu-ralistic field (Pattison, 1964). At the sametime, few would claim today, as Bunge seemsto have done in 1962, that geography some-how has or should have a monopoly on allthings spatial. Sack points out that it is notdistance itself that explains human behavior,but the costs or other impediments that resultfrom spatial separation, such as transport costor the time taken to travel. But if these areapproximated by mathematical functions ofdistance this argument seems to be hair-splitting – to all intents and purposes it is dis-tance that explains why people have morefriends nearby than far away, even in theInternet age, and why fresh milk is morelikely to come from nearby cows.

Moreover the floating magnet examplepoints to another fundamental problem, thefact that many different processes can leadto similar geometric forms. Thus it can bedifficult to deduce general statements aboutprocess from the patterns found on theEarth’s surface, particularly if one cannotobserve how those patterns change through



time. The spatial clustering of cases of adisease provides a compelling example. In thetraditional re-telling of the story, the mapmade by Dr John Snow of the cholera out-break in London in 1854 showed a clearclustering around the pump that he had sus-pected of providing cholera-carrying water,and thus of causing the outbreak (see, forexample, Longley et al., 2005). But a clusterin space can always arise through two verydifferent processes. In a first-order process thelikelihood of an event is a function of loca-tion, and events are more likely in somelocations than others, in this case near thepump. In a second-order process events affecteach other, and the presence of one eventmakes others more likely in the immediatevicinity; an example in epidemiology wouldbe contagion, in which a disease is passeddirectly from one individual to another.Under the hypothesis of a contagious process,which at the time was how cholera was gen-erally believed to have been transmitted, aninitial carrier of the disease who happened tolive near the pump could have caused thesame spatial pattern. Only by observing thepattern’s development through time couldone acquire evidence to resolve between thetwo hypotheses.

The book’s lasting impact

It was problems and issues such as these thatled geographers in the 1970s to question thevalue of Bunge’s heavy reliance on the spatialapproach, and to look elsewhere for moreproductive sources of explanation. Many oth-ers remained firmly committed, however, andcontinued to refine the methods and modelsthat Bunge had pioneered. Computing pro-vided an additional impetus, as it becamepossible to bring much more powerful toolsto bear on geographic data. By the 1980sgeographic information systems (GIS) hadbecome a focus for much of this work, and

many of Bunge’s ideas had been imple-mented and successfully applied. The rathersimplistic ideas of central place theory hadevolved into location-allocation, his discus-sion of optimal routing and geodesics hadblossomed into corridor location, and hisprimitive ideas of spatial similarity hadspawned metrics of spatial autocorrelationand the fields of spatial statistics andgeostatistics.

Today, Bunge’s ideas of a geographygrounded in geometry are alive and well, butnot for quite the reasons he suggested. Asmany have argued (see, for example, Laudan,1996), science is not only about the ratheremotional topic of explanation, but also aboutprediction, design, and other useful humanactivities. Goodchild and Janelle (2004) list sixarguments for a spatial approach in the socialsciences (and note that all six apply to bothspace and time):

• Integration. A spatial approach allowsinformation to be placed in context, andallows information from different sourcesto be correlated. These are often presentedas the most compelling functions of GIS.

• Cross-sectional analysis. As Bungeargued, spatial patterns can often provideinsight into process, particularly if pat-terns are observed through time.

• Spatial theory. A spatial theory can bedefined as one that includes such spatialproperties as location, distance, or direc-tion in its structure. Central place theory,theories of spatial diffusion, and theoriesof spatial ecology all have this property.

• Place-based analysis. These are recentlydeveloped methods of analysis that recog-nize the fundamental variability of theEarth’s surface, and produce maps of resultsrather than single summary statistics.The geographically weighted regressiondevised by Fotheringham et al. (2004) is agood example.

• Prediction, design, and policy formula-tion. As noted earlier, the application of

WILLIAM BUNGEÿÿ13


general results requires that they be placedback into a space-time context. To predictthe effects of a change in oil prices on theeconomy of California, for example, it isnecessary to take general results on howeconomies respond to oil prices and toapply them in the specific space-time con-text of California.

• Information storage and retrieval.Space and time provide a powerful way oforganizing knowledge, and of retrievinginformation from today’s vast stores of dig-ital data (see, for example, the AlexandriaDigital Library, www.alexandria.ucsb.edu,and the overview of geoportals by Maguireand Longley, 2005).

And what of the author himself? WilliamBunge is a very imposing person, two metersin height and powerfully built, and with enor-mous energy. Like many others, he wasstrongly engaged by the political turmoil of the1960s, and frustrated by the slow, reflectivepace of academic life and what he perceived asits lack of direct influence and action. He felta strong urge to demonstrate the practicalimportance of the ideas he espoused inTheoretical Geography. As a faculty member atWayne State University, he spent several yearsimplementing his ideas in the ghettos ofDetroit (Bunge, 1971), while continuing toparticipate in the meetings of the MichiganInter-University Community of MathematicalGeographers (MICMOG), which were heldclose to the point of minimum aggregate travelof the University of Michigan, Michigan StateUniversity, and Wayne State University (in ahigh school in Brighton, MI). The DetroitExpedition became a seminal exercise inworking with disadvantaged communities,applying many of the ideas of the early quan-tifiers towards the improvement of the humancondition. Detailed maps of human depriva-tion were created based on observation, andthe group worked aggressively towards theadoption of better school districting plans that

implemented many of the ideas of locationtheory.

Bunge took a strong stance against theVietnam War, and became increasingly disil-lusioned with both US society and USacademe. His positions led to increasing fric-tion both on and off campus, and in the early1970s he left the US for Canada, where hetook short-term positions at the Universityof Western Ontario and York University. Thegraduate seminar he gave at the University ofWestern Ontario was well received by manyof the students, but his openly expressed dis-gust with the political positions of some ofhis colleagues made it impossible to renewhis contract. He drove taxis for a time inToronto, and eventually settled in small-townQuebec. In 1988 Bunge’s Nuclear War Atlasappeared (Bunge, 1988), a powerful applica-tion of spatial techniques to draw attention tothe impossible consequences of the use ofthermonuclear devices. The evolution andpoliticization of his thinking is clearly evi-dent in his 1979 retrospective on TheoreticalGeography that was published in the Annals ofthe Association of American Geographers to markthe AAG’s 75th Anniversary (Bunge, 1979).The passion is there, more strongly than ever,as is his commitment to scientific method,empirical observation, and sound theory, butit is directed at different targets.

Conclusion

As Macmillan argued in another commentary(Macmillan, 2001), despite all the critiquesof quantification and positivism that haveappeared in the past three decades, Bunge’sTheoretical Geography remains as ‘a major land-mark in the history of geographic thought’(1966: 74). It appeared ‘on the cusp betweenthe old world and the new’ (ibid.: 74), betweenthe old analog world of crude, imprecise toolsand the modern world of abundant data and



powerful techniques of analysis, visualiza-tion, and simulation. Reading the book onegets a sense of what an effort it must havebeen to prepare the maps, tables, and sum-mary statistics using the pens, slide-rules,hand calculators, and log tables of 1962, andhow much more one could have achievedtoday. Yet virtually all of the major areas oftoday’s spatial perspective are there, in a

book largely written while Bunge was agraduate student. There can be few periodsin the history of any discipline when agroup of people created quite the intellec-tual ferment that must have existed ingeography at the University of Washingtonin the late 1950s and early 1960s. TheoreticalGeography captured that ferment in a waythat is still meaningful today.

WILLIAM BUNGEÿÿ15

Secondary sources and referencesBerry, B.J.L. (1993) ‘Geography’s quantitative revolution: initial conditions, 1954–1960. A

personal memoir’, Urban Geography 14 (5): 434–441.Bunge, W. (1962) Theoretical Geography (1st edn). Lund Studies in Geography Series

C: General and Mathematical Geography. Lund, Sweden: Gleerup.Bunge, W. (1966) Theoretical Geography (1st edn). Lund Studies in Geography Series

C: General and Mathematical Geography. Lund, Sweden: Gleerup.Bunge, W. (1966) Theoretical Geography (second edition) Lund Studies in Geography

Series C: General and Mathematical Geography, No. 1. Lund, Sweden: Gleerup.Bunge, W. (1971) Fitzgerald; Geography of a Revolution. Cambridge, MA: Schenkman.Bunge, W. (1973) ‘Commentary: spatial prediction’, Annals of the Association of

American Geographers 63 (4): 566–568.Bunge, W. (1979) ‘Perspective on Theoretical Geography ’, Annals of the Association of

American Geographers 69: 169–174.Bunge, W. (1988) The Nuclear War Atlas. New York: Blackwell.Chorly, R.J. and Haggen, P. (eds). (1967) Models in Geography. London: Methuen.Cox, K.R. (2001) ‘Classics in human geography revisited: Bunge, W., Theoretical

Geography. Commentary 1’, Progress in Human Geography 25 (1): 71–73.Fotheringham, A.S., Brunsdon, C. and Charlton, M. (2004) Geographically Weighted

Regression. Hoboken, NJ: Wiley.Goodchild, M.F. and Janelle, D.G. (2004) ‘Thinking spatially in the social sciences’, In

M.F. Goodchild and D.G. Janelle, editors, Spatially Integrated Social Science. New York:Oxford University Press, pp. 3–22.

Haggett, P. (1966) Locational Analysis in Human Geography. London: St Martin’s Press.Harley, J.B. and Woodward, D. (1987) The History of Cartography. Volume 1:

Cartography in Prehistoric, Ancient, and Medieval Europe and the Mediterranean.Chicago: University of Chicago Press.

Hartshorne, R. (1959) Perspective on the Nature of Geography. Chicago: Rand McNally.Horton, R.E. (1945) ‘Erosional development of streams and their drainage basins:

hydrophysical approach to quantitative morphology’, Geological Society of AmericaBulletin 56: 275–370.

King, L.J. (1969) Statistical Analysis in Geography. Englewood Cliffs, NJ: Prentice-Hall.Laudan, L. (1996) Beyond Positivism and Relativism: Theory, Method, and Evidence.

Boulder, CO: Westview Press.


Longley, P.A., Goodchild, M.F., Maguire, D.J. and Rhind, D.W. (2005) GeographicInformation Systems and Science. New York: Wiley.

Macmillan, W. (2001) ‘Commentary 2: geography as geometry’, Progress in HumanGeography 25 (1): 73–75.

Maguire, D.J. and Longley, P.A. (2005) ‘The emergence of geoportals and their role inspatial data infrastructures’, Computers, Environment and Urban Systems 29: 3–14.

Pattison, W.D. (1964) ‘The four traditions of geography’, Journal of Geography 63 (5):211–216.

Sack, R.D. (1972) ‘Geography, geometry and prediction’, Annals of the Association ofAmerican Geographers 62: 61–78.

Sack, R.D. (1973) ‘Comment in reply’, Annals of the Association of AmericanGeographers 63 (4): 568–569.

Schaefer, F. (1953) ‘Exceptionalism in geography: a methodological examination’, Annalsof the Association of American Geographers 43: 226–249.

Wilson, A.G. (1970) Entropy in Urban and Regional Modelling. London: Pion.



LOCATIONAL ANALYSIS IN HUMAN GEOGRAPHY(1965): PETER HAGGETT

Martin Charlton

Locational analysis is concerned with the needto look for pattern and order in geography ...with the locational systems we study and themodels we create to describe them, and withthe types of explanation we use in makingsense of our findings. (Haggett, 1965: 1–2)

Introduction

By 1945 most universities in Britain possesseda geography department. However, whilstmany disciplines taught at university had astrong and rigorous intellectual coherence,particularly the sciences, geography was notone of them. Sydney Wooldridge and GordonEast, Professors of Geography at King’sCollege and Birkbeck College London, hadattempted to suggest some new directions forthe discipline in The Spirit and Purpose ofGeography (Wooldridge and East, 1951). Muchresearch, however, was regional in nature, andhuman geography concerned itself with theidentification of regions, their classificationand description. Papers with titles such asHorticultural Developments in East Yorkshire, UrbanRegions of St Albans, or Egypt’s PopulationProblem were the rule rather than the excep-tion. It has been suggested the Second WorldWar was won with Science; that said, the tech-nological leaps which had been made between1939 and 1946 were not mirrored by equiva-lent progress in geography. By comparisonwith mathematics or operations research,

human geography’s largely descriptive orienta-tion was hardly scientific in its approach tounderstanding problems, and in particular, theproblems of a world attempting to regain somesemblance of normality after a conflict whichhad been global in nature.

There was, however, a scientific interest ingeographical problems outside the discipline –as early as 1934 a small group of sociologistsmet in the Adelphi Hotel, Philadelphia, to dis-cuss the statistical problems of using data forcensus tracts. Gehlke and Biehl (1934) andNeprash (1934) were, in this sense, way aheadof geographers. It would be another 30 yearsbefore geographers started to attack such issueseffectively. Indeed, Trevor Barnes (2003) sug-gests that 1955 marks the date when one couldpoint to the emergence of what has becomeknown as spatial science in the US with activ-ity at the Universities of Washington and Iowa,and subsequently, Chicago, Northwestern,Michigan, and Ohio. Barnes (2003) also sug-gests that the roots of this activity go back toearlier traditions, notably the work of vonThünen, Alfred Weber, and August Lösch. Allthree were searching for rationalism. Lösch’stext, The Economics of Location, appeared in1954. Barnes (2003) comments that Lösch’sinfluence was strongest on those viewinggeography as nomothetic (law-seeking) loca-tional analysis rather than an idiographic(descriptive) view of regional economies. By1960 a formidable range of graduates hademerged in US universities, including Brian

3


Berry, William Bunge, Michael Dacey, ArthurGetis, Duane Marble, Richard Morrill, JohnNystuen and Waldo Tobler. Armed with anarray of techniques and the computer, theywere ready to change human geography. Theyhad Science on their side.

On the other side of the Atlantic was TörstenHägerstrand, Professor of Geography at theUniversity of Lund. He too was interested inusing the power of mathematics to modelinnovation diffusion. Innovation Diffusion asSpatial Process appeared in 1953, but was notpublished in English until 1967. The tool thatHägerstrand is credited with introducing to thesocial scientists is that of Monte Carlo simula-tion (Evenden, 1969). By contrast, little wastaking place in Britain, but things were about tochange. In 1957 Peter Haggett was appointedUniversity Demonstrator in Geography at theUniversity of Cambridge. Aware of the devel-opments taking place in both the USA andSweden he began to put together some lecturenotes for undergraduates taking geography atCambridge. As often happens with such collec-tions a suggestion was made that the materialmight usefully be turned into a book. Haggettcredits his colleague and co-author on manybooks and papers, Richard Chorley, with theidea. Haggett notes ruefully in his preface‘the manuscript was written painfully slowly’.Elsewhere there was little obvious movetowards a theoretical geography: the contentsof the Transactions of the Institute of BritishGeographers (IBG) changed little in the decadeafter the Swedish publication of Hägerstrand’svolume.

In 1965, Haggett’s Locational Analysis inHuman Geography was published. An intellec-tual tour-de-force, it was unlike any other Britishtext on geography. Later that year many IBGmembers must have been astounded on open-ing the December issue of Transactions of theInstitute of British Geographers to find StanGregory’s report on a symposium held atBristol University on regression techniques ingeography (Gregory, 1965), with a further

20 pages from Chorley and Haggett on trendsurface mapping in geographical research(Chorley and Haggett, 1965). If those acrossthe Atlantic needed evidence that humangeography in the ivory towers of British acad-eme was at last shifting out of the doldrums,this was it.

The book and its arguments

The published text stands at just over 300 pages,divided into two main sections preceded by achapter entitled Assumptions. Part One, Models ofLocational Structure, includes five chapters eachwith a single-word title: Movement, Networks,Nodes, Hierarchies, and Surfaces. Part Two,Methods in Locational Analysis, has four chapters,Collecting, Description, Region-Building, andTesting. Without the context set by the FirstPart, the Second Part would lack the theoreti-cal structure which holds it all together.

In Part One, Haggett brings togethermuch of the theoretical work in geographywhich had appeared in the literature from themiddle of the nineteenth century onwards.He examines briefly geographers’ search fororder and their attempts to make sense of theworld. He considers the nature of geography‘a thorn in the side of school and universityadministrators’ (1965: 9). The emergence ofregional science created a linkage betweeneconomics and geography which was thecatalyst for an interdisciplinary approach tolocational problems which was ‘immenselyproductive’. One of the activities whichemerges is model-building – ‘to codify whathas gone before and excite fresh enquiry’.The models which Haggett examines in PartOne are for the most part normative – that isthey describe some ideal state of spatial orga-nization, although they are generally basedon observations of human spatial behaviour.There are four principal theorists fromwhose work he draws: Johann von Thünen,Walter Christaller, Alfred Weber, and August

18ÿÿMARTIN CHARLTON


Lösch – all representing German geographi-cal traditions. He later draws on more recentwork from the United States: Michael Dacey,William Garrison, William Bunge, WalterIsard and Brian Berry are all heavily cited.There are apparently fewer figureheads fromEurope and elswehere, although TörstenHägerstrand’s work is dealt with in somedetail.

Christaller’s Central Place model appearsin Movement (Chapter 2), Nodes (4) andHierarchies (5). Weber’s work is also cited almostexclusively in Hierarchies (5). Losch’s contribu-tions appear in all chapters in Part One. Amongthe 20th century commentators, Movements (2)has additional material from Hägerstrand, Isardand Bunge, Networks (3) cites Garrison andBunge. Nodes (4) features work by Berry andHierarchies (5) draws heavily on Isard’s work.Von Thünen’s work is covered extensively inSurfaces (6) with additional material from Sauer,Berry and Bunge. What Haggett achieves inPart One is to demonstrate the relationshipsbetween the individual components of loca-tional analysis.

In Movement (2), we shift from movementalong paths, and the associated idea of a least-cost path, through the modelling of interactionprocesses with gravity models, continuousmovement across a field (which looks forwardto Tomlin’s map algebra), movement in zones,and finally to the temporal aspect of movementin which Hägerstrand’s diffusion modelsare explained. Chapter 3, Networks, considersroutes and routing. Some of the concepts – theleast-cost path between two or more settle-ments – has diffused well outside geography.There are references to graph theory in math-ematics but the connection with Dijkstra’sshortest-path algorithm is tantalizingly absent,and discussion of concepts such as route densityand network change.

Chapters 4 and 5, Nodes and Hierarchies,deal with settlements and their interrelation-ships. We start with models of settlementpattern, examing the deviations from regularity

due to agglomeration, variations in resourcelocation, and temporal change. Haggett exam-ines the relationship between the size of a townand its rank in a list of towns in a region – therank/size relationship and its various models.This is expanded into a discussion of therelationship between the size and spacing ofsettlements and what the appropriate modelsmight be, and with respect to hierachies, theirfunction and potentional distortions frommodels due to agglomeration or resourcelocalization.

The final chapter considers Surfaces. Modelsmay be of continuous processes, such as popu-lation density, or discrete processes, whichare illustrated by considering land use zones.Thünian models of minimum-movement aregiven an extensive treatment – Haggett alsolinks Weberian models for locating points inspace with von Thünen’s problem of locatingareas in space. The models may be subject todistortion – the discussion leads us throughBurgess’ and Hoyt’s models of city structure toa consideration of what happens when peopledo not behave as the models might suggest: the‘spectre of sub-optimal behaviour’. The chap-ter and the first part of the book end with thesuggestion that models are not always appro-priate, Haggett arguing that the modelbuilding should follow satisficer rather thanoptimizer principles. As Haggett notes ruefully,‘Most of our existing models have been withus far too long’ (1965: 182).

In contrast, Part Two is informed moreexclusively by geographical research of thetime and seeks to place some logic on geo-graphical methods – ‘the ways in whichgeographical information can be gathered,measured, classified, and described ... in orderthat our existing concepts may be criticallyexamined’ (1965: 185). Haggett starts withthe amassing of evidence and proceedstowards the testing of hypotheses in thesefour chapters.

These chapters are not a cookbook oftechniques borrowed from other disciplines,

PETER HAGGETTÿÿ19


but the reader is assumed to have some basicstatistical knowledge. What becomes clear onreading these chapters are the strides madesince the early 1960s in the science of han-dling geographical data. In 1965 techniquesand methods in quantitative geography werein their infancy or had yet to be developed,and the computing context, which we takefor granted, was absent. Haggett hence startsthe section with a discussion of issues of datacollection. Many of the foci are still pertinenttoday – where do we find data? what sort ofdata might we find? and what might its char-acteristics be? Haggett moves gently from theabstract (what is a population?) towardsthe operational (how do we recognize oursample and the areas to which it relates?) Heconsiders various approaches to sampling,illustrating the discussion with various differ-ent sample designs. He also considers thegeographical element – noting that the datawe collect are often for irregular or inconve-nienent or incompatiable spatial units – andprovides examples of what would now referto as areal aggregation or interpolation.

Sound advice given in many introductoryquantitative analysis courses is that an analystshould explore data long before any modellingis carried out. Haggett’s journey continues withDescription. His approach is that the task is to‘describe briefly and accurately locational pat-terns revealed by data collection’ (1965: 211).The basic tools are cartographic and statistical.Mapping data is the starting point describing‘absolute location’. In many ways the discus-sion seems somewhat basic but Haggett waswriting before the development of inte r -polation techniques or geostatistics andrepresenting several variables was as difficultthen as it is now. Haggett considers some basictechniques before examining principal compo-nents analysis, albeit rather briefly. Again, wemust remember that carrying out such ananalysis was a major computational task. Thegeometer returns as Haggett then describes sta-tistical indices for shapes, points, and networks.

In spite of the dominance of the region ingeographical writing, much of Chapter 9 onRegion-building will be unfamiliar to today’sstudents. The underlying questions are thoseof creating regions grouping them. Forexample, what sort of characteristics doregions have? Consideration of this questionleads into set theory, boundary superimposi-tion and a collection of quantitative methodsbased on geometry. An important problem inareal geography is the modifiable areal unitproblem which occurs when data are aggre-gated at different spatial scales, producingvery different distributions, depending onhow the areas are defined and mapped.Haggett calls this the scale problem and con-siders it at length, noting that geographershad not made much contribution to itsanalysis. He suggests methods to tackle itbased on variance decomposition, filter map-ping, and trend surface modelling.

Today’s student would recognize much ofthe final chapter – Testing. In Haggett’s viewtesting ‘provides an appropriate check on thetheoretical excesses of Part One’ and presentsan ‘enragingly simplified’ guide to someappropriate statistical methods. The first sec-tion deals with the models and origins ofhypotheses in human geography and thesecond is a tour of some of the statisticalmethods then used in geography. The lastsection on testing via analogues is concernedwith simulation methods, and Haggett tanta-lizes the reader with an implicit suggestionthat there might be an axiomatic approachto geography. He concludes with the obser-vation: ‘In the long run the quality ofgeography in this century will be judged lessby its sophisticated techniques or its exhaus-tive detail, than by the strength of its logicalreasoning’. A second edition of LocationalAnalysis appeared in 1977 – it was consider-ably revised and was almost double thelength of the original.Andrew Cliff and AllanFrey were co-authors for this volume. It wasalso published as two separate paperbacks,



entitled ‘Locational Models’ and ‘LocationalMethods’ respectively (Haggett et al., 1977).

How the book was received

One of the first to comment publicly aboutLocational Analysis was Emrys Jones, a geog-rapher at Queen’s University, Belfast, in TheGeographical Journal in 1966. Jones begins bywarning readers of the some of the worstaspects of 1950’s sociology: pseudo-mathematics,borrowed jargon, and quantification mania.Noting that recent developments in geogra-phy seemed to lack a coherent framework hepraises Haggett for providing a frameworkfor the new techniques and aims of geo-graphical study. Jones summarizes theachievement of the book: ‘Formerly disparateelements, like bits of a jigsaw puzzle, haveclicked into place, and we are now con-fronted with a complete framework based onlogical premises to enable us to describelocation elements in an exciting way. This isbrilliantly done’ (Jones, 1966).

Peter Gould took a similar view in hisreview for the readers of the GeographicalReview (Gould, 1967). His enthusiasm is palpa-ble: ‘here, at last, is a truly exciting workin human geography, splendidly organized,chock-full of ideas’; towards the end of hisreview he comments: ‘taken as a whole thisbook not only gives a structure and coherenceto the work of the last ten years … but it showsthe continuity of such efforts with the best ofthe earlier traditions’. John Kolars (Kolars,1967) presented a more sober review inEconomic Geography in July 1967. He saw the‘value of this book is as a reference for thecognoscenti and as a guide for the catechu-men’. Kolars describes the book as ‘a synthesisof such proportions that it assumes a signifi-cance of its own’. Indeed, Haggett was to usethe term synthesis in a later title of his own.Kolars also notes that for many sources, tech-niques and problems Haggett has drawn from

both physical and human geography, and goeson to suggest that by dropping the Human fromits title allows it to express more clearly thespirit and purpose of geography as a disciplinein the physical and social sciences.

The impact: what happened next?

The next few years saw an explosion of activityexploring the ideas so elegantly broughttogether in Locational Analysis. Several notabletexts appeared, including Brian Berry andDuane Marble’s (1968) Spatial Analysis, DavidHarvey’s (1969) Explanation in Geography, andRon Abler et al.’s (1971) Spatial Organization.New journals devoted to the spatial analysiswere launched on both sides of the Atlantic,including Geographical Analysis (USA) andEnvironment and Planning (UK). A number ofpractitioners came together to create a forumfor the exchange of ideas. The QuantitativeMethods Study Group was incorporated bythe Institute of British Geographers in 1969and ‘saw growth, expansion and widespreadacceptance of quantitative methodology’(Gregory, 1983). The QMSG launched a seriesof undergraduate texts in 1975, the CATMOG(Concepts and Techniques in ModernGeography) series, edited by Peter Taylor. Therewere parallel developments in Europe – the firstof what became a bi-annual colloquium washeld in Strasbourg in 1978. The meetingbrought together invited representatives fromBritain, France, Austria, and Germany and hascontinued to stimulate much collaborativeresearch. Moreover, the ideas from Geographystarted to diffuse into other disciplines, notablyplanning (which was in a growth phase duringBritain’s postwar reconstruction) as plannersfound themselves confronted with texts such asModels in Planning (Lee, 1973) and Urban andRegional Models in Geography and Planning(Wilson, 1974).

Other disciplines were raided. Location-allocation modelling (Rushton et al., 1973)

PETER HAGGETTÿÿ21


had its origins in Operations Research, whilegeographer Alan Wilson introduced methodsfrom statistical mechanics (Wilson, 1969) andpure mathematics (Wilson, 1981) into thearsenal of techniques. Ideas and approachesfrom statistics were borrowed too. The socialsciences took on wholesale the ideas of clas-sical statistical inference and used them asthe engines of increasingly complex softwarepackages such as SPSS, OSIRIS and SAS.Indeed SPSS proclaims itself as the StatisticalPackage for the Social Sciences. However,much had already taken place in sociology.For example, Social Area Analysis (Shevkyand Bell, 1955) led to widescale experimen-tation with multivariate techniques such asfactor analysis, discriminant analysis, canoni-cal correlation analysis and cluster analysis.Gregory (1983) suggests with the benefit ofhindsight we may regard some of the publi-cations from the period as premature. However,the combination of enthusiasm and excite-ment is evident from the final words to anintroductory text on Factor Analysis: ‘grabsome census data for your town, a factoranalysis program and a large computer andrun the data through the program … tryingall the options’ (Goddard and Kirby, 1976).

There were a few lone voices of caution –geographical data tended not to have thedesirable properties that form the underpin-ning assumptions of classical statistics. PeterGould was one – in his statistics inferens paperhe sounded some warning bells (Gould,1970). Others, having failed to change theworld with large electronic computers andmatrix inversion, tried radically differentapproaches. David Harvey argued that geog-raphers had been worshipping at the wrongaltar, and that there were richer opportunitieselsewhere (Harvey, 1973). Social Justice and theCity was a revelation for many as much asLocational Analysis had been a decade earlier.It is perhaps worthwhile considering anobservation of the statistician Jan de Leeuw

on science: ‘Science is, presumably, cumula-tive. This means that we all stand, to useNewton’s beautiful phrase, “on the shouldersof giants”’. It also means, fortunately, that westand on top of a lot of miscellaneous stuffput together by thousands of midgets. If wewant to study a scientific problem we do thisin the historical context, and we do not startfrom scratch. This is one of the peculiarthings about the social sciences. They doseem to accumulate knowledge, there arevery few giants, and every once in a whilethe midgets destroy the heaps’ (de Leeuw,1994). Indeed, Harvey was to describe thetotality of the output from quantitativeresearch as a ‘hill of beans’. The debatebetween spatial scientists and social theoristscontinues today and there are those forwhom spatial analysis is at best an irrelevanceto the mainstream of the discipline (seeCloke et al., 1991, amongst many).

A new technology appeared during the1980s, that of Geographical InformationSystems. The marriage of techniques fromcomputational geometry, database systems, andcomputer science saw the emergence of arange of software packages for handling spatialdata. Spatial analysis as practised with GIS dealswith geometrical operations on spatial data.On both sides of the Atlantic governmentsponsored research initiatives were funded inthe form of the ESRC’s Regional ResearchLaboratory network and the National Centerfor Geographic Information and Analysis inthe USA, accompanied by journals such as theInternational Journal of Geographical InformationSystems and Transactions in GIS. The rise of GIShas been paralleled by a return to the problemsof statistical modelling with spatial data.Practitioners have not only included geogra-phers but also statisticians.

Today, spatial data handling software isubiquitous, and appears in a wide variety offorms. Dozens of airlines have on-line ticketordering systems which will provide the



traveller with the quickest or cheapest ticketoption. Planning rail or road journeys via theInternet uses the same underlying models andmethods. Satellite navigation systems are stan-dard items in high street electronics stores.Geometry and geography make interestingbedfellows and computational geometry putsthe progeny in the hands of the public. Thechallenging problems in locational analysisremain challenging – and there is no shortageof researchers willing to tackle them.

Conclusion

Haggett’s book stands as an early interventionin geography’s attempt to restyle itself as a spa-tial science, and remains an obligatory point ofpassage for many geographers working onspatial data analysis, GIS and geocomputation.Trevor Barnes’ summation of the book is thusas follows:

On the surface, its limpid prose, elegantdiagrams, and mathematical equationsare not the sort of thing to get peopleriled up to change the world. But fromrecent interviews I’ve carried out in con-nection with a project to write about thehistory of the quantitative revolution ingeography during the 1950s and 1960s,and economic geography in particular, itis clear that Haggett’s book had exactlythat effect. Moreover, its influence wasnot just confined to within the academy,persuading people merely to write a dif-ferent type of academic paper. (Barnes,2002: 10)

The material impacts of Haggett’s book onhis contemporaries and subsequent generationswere then, in many cases, massive. As such,there is little doubt that Locational Analysis isone of human geography’s Key Texts – onewhose legacies continue to unfold.

PETER HAGGETTÿÿ23

Secondary sources and referencesAlber, R., Adams, J.S. and Gould, P. (1971) Spatial Organisation: The Geographer’s View

of the World, Englewood Cliffs, NJ: Prentice-Hall.Barnes, T.J. (2002), ‘Critical notes on economic geography from an ageing radical’,

Acme 1 (1): 8–20. http://www.acme-journal.org/vol1/barnes.pdfBarnes, T.J. (2003) ‘The place of locational analysis: a selective and interpretive history’,

Progress in Human Geography 27 (1): 69–85.Berry, B.J.L. and Marble, D. (1968) Spatial Analysis: A Reader in Statistical Geography.

Englewood Cliffs: Prentice-Hall.Bunge, W. (1962) ‘Theoretical geography’, Lund Studies in Geography, series C, General

and Mathematical Geography 1.Chorley, R.J. and Haggett, P. (1965) ‘Trend-surface mapping in geographical research’,

Transactions of the Institute of British Geographers 37: 47–67.Cloke, P., Philo, C. and Sadler, D. (1991) Approaching Human Geography: An Introduction

to Contemporary Theoretical Debates. London: Paul Chapman.De Leeuw, J. (1994) Statistics and the Sciences. Unpublished manuscript, UCLA

Statistics Program.Evenden, J.Y. (1969) ‘Review: Innovation diffusion as a spatial process’ by Törsten

Hägerstrand, Social Forces 47 (3): 356–357.Gehlke, C.E. and Biehl, K. (1934) ‘Certain effects of grouping upon the size of the correlation

coefficient in census tract material’, Journal of the American Statistical Association 29,Supplement: 169–170.


Goddard, J.B. and Kirby, A. (1976) An Introduction to Factor Analysis. Norwich: GeoAbstracts Ltd.

Gould, P. (1967) ‘Review: Locational analysis in human geography’, GeographicalReview 57 (2): 292–294.

Gould, P. (1970) ‘Is statistix inferens the geographical name for a wild goose’, EconomicGeography 46, Supplement: 439–448.

Gregory, S. (1965) ‘Regression techniques in geography’, Transactions of the Institute ofBritish Geographers 37: ix–xi.

Gregory, S. (1983) ‘Quantitative geography: the British experience and the role of theInstitute’, Transactions of the Institute of British Geographers, New Series 8: 80–89.

Hägerstrand, T. (1967) Innovation Diffusion as a Spatial Process. Chicago: University ofChicago Press.

Haggett, P. (1965) Locational Analysis in Human Geography. London: Edward Arnold.Haggett, P., Cliff, A.D. and Frey, A. (1977) Locational Analysis in Human Geography (2nd

edn). New York: Wiley.Harvey, D.W. (1969) Explanation in Geography. London: Edward Arnold.Harvey, D.W. (1973) Social Justice and the City. London: Edward Arnold.Jones, E. (1966) ‘Review: Location in geography: a statistical approach’, The

Geographical Journal 132 (2): 267–268.Kolars, J. (1967) ‘Review: Locational analysis in human geography’, Economic

Geography 43 (3): 276–277.Lee, C. (1973) Models in Planning: An Introduction to the use of Quantitative Models in

Planning. Oxford: Pergamon.Neprash, J.A. (1934) ‘Some problems in the correlation of spatially distributed variables’,

Journal of the American Statistical Association 29, Supplement: 167–168.Rushton, G., Goodchild, M.F. and Ostresh, L.M. (1973) Computer Programs for Location-

Allocation Problems. Monograph No. 6, Department of Geography, University of lowa.Shevky, E. and Bell, W. (1955) Social Area Analysis: Theory, Illustrative Application and

Computational Procedures. Stanford: Stanford University Press.Wilson, A.G. (1969) Entropy in Urban and Regional Modelling. London: Centre for

Environmental Studies.Wilson, A.G. (1974) Urban and Regional Models in Geography and Planning. London:

Wiley.Wilson, A.G. (1981) Catastrophe Theory and Bifurcation: Applications to Urban and

Regional Systems. London: Croom Helm.Wooldridge, S.W. and East, W.G. (1951) The Spirit and Purpose of Geography. London:

Hutchinson.



EXPLANATION IN GEOGRAPHY (1969):DAVID HARVEY

Ron Johnston

By our theories you shall know us. (Harvey,1969: 486)

Introduction

Although the discipline of geography hasalways been characterized by flux, the 1960s isrecognized as a particular turbulent decade: byits end, the discipline incorporated practicesvery different from those deployed 10 yearsearlier. Particularly important here was the dis-content with disciplinary practices which somegeographers felt after having served withscholars from other disciplines in the AmericanOffice of Strategic Services during World WarTwo (Barnes and Farrish, 2006). Subsequentlyin the mid-1950s a number of US geographers –notably a cohesive group of faculty and gradu-ate students at the University of Washington,Seattle – began to promote a very differentvision of geography (bridging both physicaland human arms of the discipline) based on the‘scientific methods’ deployed by physicists and,in the social sciences, economists. By the early1960s, this group was rapidly attracting adher-ents in a number of major US graduateschools, and in 1963 one observer-participantclaimed that the ‘revolution’ promulgated fromSeattle was successfully over.

Almost contemporaneously, a similar ‘revolu-tion’ was taking shape in the UK, based atthe University of Cambridge. Two recentlyappointed faculty members – Dick Chorley andPeter Haggett – were attracted to the ‘scientificmodel’ (which Chorley had encountered and

adopted while a geology graduate student in theUS), and began to teach related material, mainlystatistical methods, in their introductory practi-cal classes for geography undergraduates. One ofthe demonstrators in the first year of that class(in 1960) was a postgraduate student working inhistorical geography – David Harvey (Harvey’s1962, PhD thesis was on the historical geogra-phy of the Kentish hop industry – see Harvey,1963 – and this stimulated his subsequent inter-est in the processes involved in evolving spatialpatterns: Harvey, 1967).

Although the shift in geographical practicethese two groups were advancing is oftenreferred to as the discipline’s ‘quantitativerevolution’ (on which, see Johnston andSidaway, 2004), it involved much more thanjust applying mathematical and statisticaltechniques to geographical data. It was fun-damentally a ‘theoretical revolution’, whichchanged the entire mind-set of how researchwas to be undertaken and new knowledgepresented. Words such as ‘theory’, ‘model’,‘hypothesis’ and ‘law’ become common in thegeographical lexicon, as researchers strove toproduce knowledge that was cumulative (inthe sense that it built on earlier research notonly to better explain the world but also tomodify or improve it).

When a discipline is experiencing suchmajor change, new teaching materials areneeded to introduce students to (and justify)‘revolutionary’ practices. Introductory text-books usually lag behind such changes giventhat publishers have to be convinced that thereis a viable market for ‘revolutionary tracts’!

4


This was certainly the case with the 1960s‘new geography’. The first books clearly enun-ciating the ‘scientific approach’ appeared in1965 (Haggett’s Locational Analysis in HumanGeography – this was very much a publisher’sgamble on the shape of the discipline’s future(see Charlton, Chapter 3 this volume) – andChorley and Haggett’s edited volume Frontiersin Geographical Teaching: Bunge’s (1962) mono-graph on Theoretical Geography had only alimited circulation, although a revised editionin 1966 attracted wider attention – seeGoodchild, Chapter 2 this volume). A spate oftexts oriented towards the ‘revolution’ onlyemerged some five years later. This includedDavid Harvey’s Explanation in Geography, theproduct of almost a decade teaching under-graduates at the University of Bristol aboutthe ‘new’ scientific basis to geographical work,building on discussions with colleagues inSweden and the US (where he spent consid-erable time during the decade). As he puts it inthe Preface, writing the book was part of hislearning experience as he developed teachingmaterials – he ‘wrote this book in order toeducate myself. I sought to publish it becauseI feel sure there are many geographers, bothyoung and old, who are in a similar state ofignorance to that which I was in before Icommenced to write’ (Harvey, 1969: v).

The book and its argument

Harvey’s book – like that of the other‘rev olutionaries’ he joined – reflected a deepdissatisfaction with geographical practices expe-rienced when he was an undergraduate and inhis formative years as a researcher. For Harvey,quantification was necessary but far from suffi-cient: measurement was a requisite tool, butmuch more important was for human and phys-ical geographers to deploy the ‘fantastic power’of the scientific model. Hence he explored –and wrote a book (he termed it an ‘interimreport’) about – ‘the ways in which geographi-cal understanding and knowledge can be

acquired and the standards of rational argumentand inference that are necessary to ensure thatthe process is reasonable’ (Harvey, 1969: viii).This exploration took him into a literature pre-viously almost entirely ignored by geographers,so that although he included plentiful referencesto a small number of geographers (especiallythose instrumental in fomenting the ‘revolu-tion’) his ‘Author index’ indicates how heavilyhe drew on philosophers of science – Ackoff,Braithwaite, Carnap, Churchman, Hempel,Kuhn, and Nagel; Popper is also cited, though toa lesser extent (and his ideas regarding falsifica-tion are very summarily dismissed) and evenEinstein and Russell (although neither Ayernor Wittgenstein) are mentioned. In addition,a significant number of mathematicians andstatisticians such as Anscombe, Blalock, Fishburn,Fisher, Krumbein, Sneath, and Sokal are alsocited.

From the outset, Harvey’s book focusesexplicitly on explanation. After a brief intro-ductory chapter setting out his main concernas methodology rather than philosophies –i.e. on how geographers should produceexplanations – Explanation in Geographymoved to a discussion of what explanationmeans. Harvey defined it as ‘making an unex-pected outcome an expected outcome, ofmaking a curious event seem natural or nor-mal’ (Harvey, 1969: 13) because it can beshown to be generated by similar processesand in similar conditions as previous eventsof the same type. Harvey’s concern was with‘rational explanation’, statements verifiableby others because the procedures involved intheir production can be repeated and/or areopen to scrutiny. It is a ‘formal procedure …[at the] hard inner core of methodology’(Harvey, 1969: 23).

Following that brief framework-settingintroduction (of only 26 pages), the remainderof the book comprised five main sectionsdealing with: explanation in geography; theo-ries, laws and models; languages; models fordescription in geography; and models forexplanation in geography. At the outset he

26ÿÿRON JOHNSTON


contrasted (in an oft-reprinted diagram) theinductive or ‘Baconian’ path to explanationwith his preferred deductive route, which pro-ceeds through the establishment of a modelrepresenting the researcher’s image of theworld, the derivation of hypotheses regardingsome aspect of that image, testing the hypoth-esis’s validity, and the formulation of theoriesand laws synthesizing the knowledge gained –from which revised models can be derived.

Four terms/concepts stood out in thelexicon Harvey associated with this route toexplanation. Hypotheses were presented as logi-cally consistent ‘controlled speculations’. Theycan never be tested absolutely – conclusionsare always provisional – but Harvey indicatedthat they guide the production of scientificknowledge in a rigorous (and hence replicable)way. The key outcomes of testing hypothesesare laws and theories. Laws are sometimes pre-sented as universal truths – statements whosevalidity is constrained by neither time norspace. It is never possible to reach such conclu-sions, but scientists act as if they can, using theirfindings as statements of the current ‘conven-tional wisdom’ encapsulating what we alreadyknow and providing the foundations for fur-ther scientific exploration. They are rigorouslyproduced but ultimately provisional conclu-sions representing the contemporary state ofknowledge. For geography (especially humangeography, although much physical geographyat the time – such as the Davisian model oflandscape evolution – similarly lacked rigorousunderpinnings), they contrasted sharply withthe ‘explanatory sketches’ traditionally offeredas accounts of ‘observed reality’. The produc-tion of laws in geography, according to Harvey,involved searching for ‘hidden order withinchaos’. Because the results of such searches arelikely to be provisional/tentative, Harvey sug-gested that geographers might prefer theconcept of ‘law-like statements’, general claimsthat are both ‘reasonable with respect to expe-rience and consistent with respect to eachother’ – a coherent body of knowledge corre-sponding with the observed ‘reality’.

Theories are systems of linked statementsabout defined subject matter. These may beentirely closed systems – as with Euclideangeometry; they may be sets of deductive state-ments derived from accepted axioms; or theymay be less-formally stated ‘sketches’. Theyare not just speculative ideas – as sometimesimplied in vernacular uses of the term: any-one can fabricate such a ‘system of apparentwisdom in the folly of hypothetical delusion’(Harvey, 1969: 97: here, Harvey is quotingJames Hutton via Chorley). Scientific theoriestake ‘such speculations and transform ... themfrom badly understood and uncomfortableintrusions upon our powers of “pure” objective description into highly articulatesystems of statements of enormous explana-tory power’ (Harvey, 1969: 87–88).

Various types of theory extend along acontinuum from highly formalized, internally-closed sets of statements (as with many formsof mathematics and logic), through sets ofstatements which are only partial (eitherbecause their primitive terms, or axioms –the assumptions on which they are built – areincomplete or because the deductions fromthose foundations are not fully elaborated), towhat Harvey terms ‘non-formal theories …statements made with theoretical intention,but for which no theoretical language hasbeen developed’ (Harvey, 1969: 98). The lasttype ‘scarcely conform in any respect to thestandards of scientific theory’ (Harvey, 1969:130), and are characteristic of previous ‘theo-retical’ work within geography. Geographershad to move forward, Harvey argued, eitherby deriving theories from axiomatic state-ments – more likely to be feasible in physicalgeography, which can deduce, for example,landscape-producing processes and the likelyresultant forms from physical laws – or, inhuman geography, by generating assumptionsabout human behaviour from which state-ments about spatial patterns can be deduced.

Whatever the theory’s origin, establishing itsempirical status involves moving to the final keyword in the new lexicon Harvey employed – the

DAVID HARVEYÿÿ27


model. This has a diversity of meanings in bothpopular and scientific language; for Harvey’s‘new geography’ a model was a representationof a theory – i.e. an outcome of a series of law-like statements. Such representations becamethe source for hypotheses, leading to tests of atheory’s empirical validity.

An example of a (human) geographical the-ory illustrating these fundamental concepts iscentral place theory: an idea about the spatialorganization of settlement hierarchies thatunderpinned a great deal of geographical workin the 1950s–1960s. Harvey showed that it wasderived from a set of fundamental economicpostulates (assumed laws) about consumer andprovider behaviour (i.e. profit-maximizationfor providers and minimization of travelcosts for consumers) and the nature of thegoods/services being supplied/demanded.These are linked in a single theory fromwhich it is possible to deduce the spatialarrangement of service centres. Models couldbe derived showing the expected morphologyof that spatial arrangement in different con-texts, which had common features such as thehexagonal arrangement of centres in nestinghierarchies. Specific hypotheses could then betested in particular empirical situations.

For Harvey, the most important of theseconcepts was theory: without theories ‘theexplanation and cognitive description ofgeographic events is inconceivable’ (Harvey,1969: 169). But how could such theories beexpressed? The language to be deployed wasthat of mathematics – ‘the language of sci-ence’ – within which he concentrated ontwo sub-fields: geometry as the language ofspatial form (geography being defined as thestudy of ‘objects and events in space’ –Harvey, 1969: 191); and probability as the lan-guage of chance, necessarily used because ‘theworld is governed by immutable chanceprocesses’ (Harvey, 1969: 260) so that preciseprediction is rarely possible, especially sogiven the extent of our ignorance aboutthose processes. A scientific geography wouldnot be a deterministic geography, therefore,

but rather comprise probabilistic statementsof likely explanations (of the ‘hidden orderwithin chaos’) – hence Harvey’s advocacy ofstatistics as crucial in evaluating hypotheses.

Given the key components of the scientificmethod and its language, as applied to geogra-phy (both human and physical; Harvey saw nodifference between the two in their method-ological structure), Harvey thus dedicated twosections of the book to modelling in geogra-phy – descriptive and explanatory. These are, ineffect, chapters about methods: the first dealswith measurement and how one portrays theworld – how information is collected, classifiedand displayed; the second with procedures fortesting hypotheses of cause and effect.

In the concluding chapter, Harvey summa-rized 480 pages of detailed material as ‘somerough and ready guidelines for the conduct ofempirical research in geography’, presentingthe tools that might be used when we ‘have topin down our speculations, separate fact fromfancy, science from science fiction’ (Harvey,1969: 481). That is done scientifically by pro-ducing ‘an adequate corpus of geographictheory’ – coherent statements about aspects ofthe world, validated by geographers’ adoptionof the protocols of scientific method. ForHarvey geography in the 1960s lacked such aclear identity and sense of direction, hence theclarion call in the final pages of his text:

Without theory we cannot hope for con-trolled, consistent and rational explanationof events. Without theory we can scarcelyclaim to know our own identity. … theoryconstruction on a broad and imaginativescale must be our first priority in thecoming decade. … Perhaps the slogan weshould pin upon our study walls for the1970s ought to read: ‘By our theories youshall know us’. (Harvey, 1969: 486)

The book and its consequences

Explanation is a long and detailed book. It is notdifficult to read once the basic concepts are

28ÿÿRON JOHNSTON


appreciated, but it is not presented as atextbook – or at least not as such books are nowpresented (i.e. there are suggested readings at theend of each chapter, but no boxes or otherdevices designed to focus on key ideas and/orexemplars). Harvey never intended to write sucha textbook: as he expressed it, the book was ‘writ-ten for anyone who cares to read it and foranyone who cares to use it in whatever way orways they find congenial or useful. … The soonerwe stop writing for ‘an’ audience … or for ‘thebeginning graduate student’ the better off we willbe’ (Harvey, 1971: 323). His goal was neverthelessthat of other contemporary textbook writers: tointroduce the methods geographers (should)deploy to produce knowledge, beginning withthe key methodological protocols and then set-ting out detailed procedures. It is thus similar toother contemporary volumes, notably Abler et al.(1971), which began with discussions of ‘scientificmethod’ and research procedures but then, in amuch more ‘student-friendly’ way, illustratedthese with detailed (human) geographicalexamples; Haggett (1965) put procedures afterexamples – they became separate books in thesecond edition – but said very little about‘scientific method’. Earlier books, such asGregory (1963), were entirely about procedures(interestingly, Gregory’s book is not referencedin Explanation).

Explanation’s accessibility – at least toHarvey’s peers – was testified by reviews atthe time. Among the main reviews are thoseby Douglas Amedeo (1971), for whom itwas clear that ‘geographers need this kind ofbook, and it would be a profound disserviceto the field to provoke them into ignoringit’; Stan Gregory (1970) who called it ‘animportant book, which is certain to have aconsiderable influence upon the developmentof geography over the next decade’; and JulianWolpert (1971) who termed it ‘enormouslyinstructive’, suggesting Harvey ‘has con-tributed to the discipline a rigorously conciseoverview of the accumulated “science”which converts current research and teachingefforts … [providing] geographically-relevant

access to the mainstream of the philosophy,methodology, and language of science’. Thesereviews depicted the book as important,especially as it was the first full-length state-ment of what might be involved ingeographers adopting the methodology (andunderlying philosophy) of the natural sci-ences – what geographers later referredto with increasing frequency as positivism (aterm not in Explanation’s index!).

But Explanation’s impact was probably lessthan its originality and depth deserved, giventhat it was the first full statement of what a‘scientific geography’ should look like. Forexample, Explanation is unlikely to have beenwidely used as a textbook – certainly not forundergraduates – for three main reasons. Thefirst is its density and presentation. Many inthe early 1970s will have been directed to it assupplementary reading for the increasinglypopular courses on the history and philosophyof geography – but it was too detailed to bedeployed as the main text for a course ongeographical methods (Harvey’s course atBristol was a rarity within undergraduate pro-grammes at the time and the book appearedmore than a decade before taught postgradu-ate courses were instituted in many Britishgeography departments: it may have attractedmore attention in US graduate schools, butcompeted with more ‘accessible’ books such asAbler et al., 1971). Some undergraduates willhave been invited to read particular partsas the basis for tutorial/seminar discussions;others may have either discovered it serendip-itously on a library/bookshop shelf or beendirected towards it, and become absorbed byits arguments. But for most students (and alsoacademics intrigued by and attracted to thechanges in their discipline) its main role was asa reference text, something to be used whenseeking detailed material about, for example,the key concepts of theory, law, hypothesis andmodel. Notably, it offered few examples northe case studies that are so often the key tostudent appreciation of a set of ideas, and cer-tainly its chapters on procedures did not take

DAVID HARVEYÿÿ29


the ‘how-to’ form typical of introductory text-books (for example, Amedeo and Golledge,1975, was a more ‘user-friendly’ introductionto the protocols of the scientfic method).

Secondly, although original in its depth andcatholicity Explanation was not entirely novel.‘The shock of the new’ had hit geographerssome years before with the publication of bookssuch as Chorley and Haggett (1965), Haggett(1965) and Chorley and Haggett (1967). Thesewere much more instrumental in bringing thefomenting revolution in geographical practicesto academic geographers’ attention (especiallyin the UK). For those converted, however,Explanation provided the detailed expositionneeded to make them fully aware of the com-plexity of the practices they were planning toadopt. But, as Harvey made clear, the book wasnot about philosophy but rather methodswithin a particular philosophy, and although heclearly addressed many philosophical issues, itwas left to later authors (notably Gregory, 1978)to explore issues of epistemology and ontologymore fully, linking geographers to a much widerset of philosophical debates.

Thirdly, by the time Explanation appearedthe practices that it advanced were beingstrongly contested – not least by David Harveyhimself! He hints at this in his Preface:

Compared to my situation five years ago Inow feel much more learned and wise, butrelative to what I still have to learn I feelmore ignorant than ever. Indeed, sincecompleting this manuscript in June 1968 Ihave changed several opinions and I canalready identify errors and shortcomingsin the analysis. (Harvey, 1969: viii)

Thus, when he responded to by far thelongest review/critique to appear – byStephen Gale (1971), who thought it ambi-tious and stimulating, but deficient as bothtextbook and reference volume – Harvey hadmoved away, not from science per se but fromthat particular scientific philosophy and formof scientific method. As detailed in a numberof essays in the early 1970s (reprinted in

Harvey, 1973), he turned to Marxism as hissource of theoretical inspiration, claiming thescientific method he had previously advancedwas ideologically infused in that it sustainedthe political status quo and, however sophis-ticated its descriptions of the world, wasunable to appreciate the underlying processeswhich produced them (see Harvey, 1974).Harvey certainly did not abandon theory:rather, he moved his theoretical stance,adopting a new set of protocols and proce-dures. As Peet (1998) put it, with ExplanationHarvey showed geographers they neededtheory, but almost immediately realized it wasthe wrong theory: his clarion call accordinglyremained intact – but the theories did not.

The book’s legacy

Although there was a clear ‘revolution’ inDavid Harvey’s own approach to human geog-raphy almost immediately after Explanation waspublished, quantitative work remains a strongstrand within the discipline – increasinglysophisticated in both its methods and in thecomputational, statistical and GIS technologieson which it now depends. But much of thatwork has increasingly distanced itself from itspositivist foundations – particularly with regardto the search for law-like statements. As dis-cussed in recent essays (such as Fotheringham,2006), most quantitative work involves the rig-orous interrogation of large, spatially structureddata sets with the goal of finding order withina highly complex world (but never implyingthat such order is fixed for eternity). Such workseeks to accrue ‘sufficient evidence on whichto base a judgement about reality that mostreasonable people would find acceptable’(Fotheringham, 2006: 241), in contexts whereunderstanding calls for the deployment oflarge, aggregate data sets. For such enterprises,much of the detail in Explanation is irrelevantgiven that many of the methods it promoteshave been superseded. Indeed, while Explanationwas one of the foundation stones in the creation

30ÿÿRON JOHNSTON


of a ‘new theoretical geography’, the superstruc-ture has since been substantially reconstructed insubsequent decades.

In retrospect, it is clear that Explanationwas written on the cusp of a major change inHarvey’s own work in the early 1970s (asHarvey, 2002, 2006, documents; see also Castree,Chapter 8, this volume). In spite of, and perhapsbecause of, that he remains one of geography’smost influential scholars. He has also –uniquely – been the subject of two critical eval-uations. The first (Paterson, 1985) covered theExplanation years and the first decade ofHarvey’s Marxist explorations, but gave less thanone-third of its space to the former and, whiledocumenting the major discontinuity in histhought, gave little attention to what might beidentified as a major continuity – ‘the impor-tance of general theory’. (This misrepresentsHarvey for whom theory is the continuity, not‘general theory’, whatever that might be.) Thesecond (Castree and Gregory, 2006) appearedsome 35 years after the switch in Harvey’s the-oretical orientation and, given his productivityand the seminal nature of much that he has pub-lished since, not surprisingly pays relatively littleattention to the first decade or so or Harvey’scareer. Trevor Barnes puts that ‘first life’ into itssocial context and stresses its continuities withHarvey’s post-1970 output – commitmentsto the discipline and practice of geography, topolitics (i.e. the application of geographicalknowledge), and ‘perhaps most germane…, totheory’ (Castree and Gregory, 2006: 42); in thesame volume, Derek Gregory unpacks thelacunae in Explanation – the black-box systemdiagrams that expressed an ignorance ofprocess and the sterility of mathematical lan-guage, while Eric Sheppard identifies severalmore continuities, notably a concern for spaceand time – and space–time. But it is Harvey

the Marxist who gets the bulk of the attention(as in Castree, 2004).

Conclusion

To paraphrase a statement made in anothercontext, Explanation is undoubtedly now ‘morerevered than read’ – and may always have beenso. As Castree (2004: 181) puts it, Explanation‘gave Harvey’s generation of geographers aheavyweight justification and manifesto fortheir project’, aligned ‘the discipline with theso-called “real” sciences like physics and, forsome geographers, boosted the discipline’s self-image’ – although many human geographerssought status within the social sciences ratherthan suffering from ‘physics envy’. But Harveyabandoned his generation – or many of them –for an alternative project, to which a newgeneration of converts was attracted. Indeed,according to his autobiographical essay, insome ways he abandoned the first project longbefore he completed it – having a ‘lust to wan-der and diverge, to challenge authority, to getoff the beaten path of knowledge into some-thing different, to explore the wild recesses ofthe imagination as well as of the world’(Harvey, 2002: 167). He did finish it, however,but responded to Stephen Gale’s (1971) reviewby saying that he was at a disadvantage becauseGale had read the book and ‘I have never readit. What is more, I have no intention of doing sonow’. Explanation was behind him, but remainsa permanent and potent reminder of a crucialtime in geography’s turbulent recent history; assuch, it should be read not just as a pioneeringand influential exploration of ‘scientificmethod’ and its philosophical underpinningsbut also one of the first substantive geographi-cal engagements with social science.

DAVID HARVEYÿÿ31

Secondary sources and referencesAbler, R., Adams, J. and Gould, P. (1971) Spatial Organization: The Geographer’s View

of the World. Englewood Cliffs, NJ: Prentice-Hall.Amedeo, D. (1971) ‘Review of Explanation in Geography by David Harvey’, Geographical

Review 61: 147–149.


32ÿÿRON JOHNSTON

Amedeo, D. and Golledge, R.G. (1975) An Introduction to Scientific Reasoning inGeography. New York: John Wiley.

Barnes, T.J. and Farrish, M. (2006) ‘Between regions: science, militarism, and Americangeography from World War to Cold War’, Annals of the Association of AmericanGeographers 96: 807–826.

Bunge, W. (1962) Theoretical Geography (second edition 1966). Lund: C.W.K. Gleerup.Castree, N. (2004) ‘David Harvey’, in P. Hubbard, R. Kitchin and G. Valentine (eds) Key

Thinkers on Space and Place. London: Sage, 181–188.Castree, N. and Gregory, D. (eds) (2006) David Harvey: a Critical Reader. Oxford:

Blackwell Publishing.Chorley, R.J. and Haggett, P. (eds) (1965) Frontiers in Geographical Teaching. London:

Methuen.Chorley, R.J. and Haggett, P. (eds) (1967) Models in Geography. London: Methuen.Fotheringham, A.S. (2006) ‘Quantification, evidence and positivism’, in S. Aitken and

G. Valentine (eds) Approaches to Human Geography. London: Sage, 237–250.Gale, S. (1971) ‘On the heterodoxy of explanation: a review of David Harvey’s

Explanation in Geography’, Geographical Analysis 3: 285–322.Gregory, D. (1978) Ideology, Science and Human Geography. London: Hutchinson.Gregory, S. (1963) Statistical Methods and the Geographer. London: Longman.Gregory. S. (1970) ‘Review of Explanation in Geography by David Harvey’, The

Geographical Journal 136: 303.Haggett, P. (1965) Locational Analysis in Human Geography. London: Edward Arnold.Harvey, D. (1963) Explanation in Geography. Arnold: London.Harvey, D. (1963) ‘Locational change in the Kentish hop industry and the analysis of land

use patterns’, Transactions, Institute of British Geographers 323: 123–140.Harvey, D. (1967) ‘Models of the evolution of spatial patterns in human geography’, in

R.J. Chorley and P. Haggett (eds) Models in Geography. London: Methuen, 549–608.Harvey, D. (1969) Explanation in Geography. Arnold: London.Harvey, D. (1971) ‘On obfuscation in geography: a comment of Gale’s heterodoxy’,

Geographical Analysis 3: 323–330.Harvey, D. (1973) Social Justice and the City. London: Edward Arnold.Harvey, D. (1974) ‘What kind of geography for what kind of public policy’, Transactions,

Institute of British Geographers 63: 18–24.Harvey, D. (2002) ‘Memories and desires’, in P.R. Gould and F.R. Pitts (eds)

Geographical Voices: Fourteen Autobiographical Essays. Syracuse: SyracuseUniversity Press, 149–188.

Harvey, D. (2006) ‘Memories and desires’, in S. Aitken and G. Valentine, (eds)Approaches to Human Geography. London: Sage, 184–190.

Johnston, R.J. and Sidaway, J.D. (2004) Geography and Geographers: Anglo-AmericanHuman Geography since 1945 (sixth edition). London: Arnold.

Paterson, J.L. (1985) David Harvey’s Geography. London: Croom Helm.Peet, R. (1998) Modern Geographical Thought. Oxford: Blackwell.Wolpert, J. (1971) ‘Review of Explanation in Geography by David Harvey’, Annals of the

Association of American Geographers 61: 180–181.


CONFLICT, POWER AND POLITICS IN THECITY (1973): KEVIN COX

Andy Wood

The American city is in a state of crisis. Themelting pot of yesterday has become aPandora’s box of troubles – flight to the suburbs, ghetto poverty, racial conflict, inade-quacies in public provision are fodder forcontemporary urban politics. Conflict hasbecome endemic in the metropolitan areas:between the ‘turfs’ of social groups, betweensuburbs and central city, and between neigh-borhoods and the city itself. This book isconcerned with the geography of these conflicts.(Cox, 1973: 1)

Introduction

Kevin Cox’s third book Conflict, Power andPolitics in the City: A Geographic View, hadthe misfortune to be published in 1973 –the same year as David Harvey’s SocialJustice and the City. Citation counts indicatethe tremendous significance of the latterwork but if Harvey’s overshadowed Cox’sbook the latter nevertheless remains one ofthe key texts from the early 1970s that puturban geography at the forefront of disci-plinary trends (with Ley, 1974 and Ward,1971 also fitting that bill). While humangeography has, in the main, moved on fromthe intellectual ideas that underpin Conflict,Power and Politics in the City the bookstands as an important transitional docu-ment that has remarkable contemporaryresonance even though the theoretical tidehas turned.

The combination of Preface and Conclusionindicates that Conflict, Power and Politics in theCity was written to meet three important goals.The first was to advance a rigorous, academicaccount that would help to explain the urbanproblems that were increasingly apparent inUS cities such as Cleveland, Los Angeles,Newark and New York in the late 1960s andearly 1970s. In fact rioting and violent politicalprotest have a long history associated with thecity but the urban conflagrations of the late1960s threatened to thoroughly disrupt theidea that the US was a nation of growingwealth and prosperity for all. It is indeednotable that Conflict, Power and Politics in theCity was published in McGraw-Hill’s ProblemsSeries in Geography. Earlier volumes hadaddressed atmospheric pollution (Bach, 1972),poverty (Morrill and Wohlenberg, 1971) andsocial well-being (Smith, 1973). The series gaveCox the opportunity to develop an academictext that was explicitly focused on contempo-rary urban problems while avoiding the moreparticular focus of Harold Rose’s 1971 bookThe Black Ghetto in the same series.

A second goal of Cox’s book was, as hesuggests (Cox, 1973: 1), to publicize the pol-icy implications of urban analysis, with thefinal chapter devoted to a lengthy discussionof what he saw as the necessary governmen-tal actions required to address the conflictsexamined in his book. The third objective wasone of developing a specifically geographicalaccount of urban conflicts and the processesthat underpin them. In this respect Cox’s

5


immediate academic environment seems tohave been significant. Kevin Cox had grownup near Warwick in England and had studiedGeography at Cambridge. He then traveled tothe University of Illinois where he completeda Master’s degree and his doctorate. In 1965the editor of the McGraw-Hill series – NedTaaffe – had recruited Cox to Ohio State aspart of a strategy of hiring what were tobecome key figures in the discipline, includingLarry Brown, Emilio Casetti, George Demko,Howard Gauthier, Reginald Golledge, LeslieKing and John Rayner (Barnes, 2004). Theintellectual coherence and direction to thedepartment was provided by an uncompro-mising focus on quantitative geography andthe prosecution of the spatial science tradi-tion. Trevor Barnes has written that duringthe 1960s ‘perhaps more than at any otherplace, OSU (Ohio State University) self-consciously remade itself into a site of quan-titative geography’ (Barnes, 2004: 582). Withoutwanting to claim that these biographical detailsdetermined Cox’s position it is clear that thisimmediate environment had significant influ-ence on the nature and direction of Cox’sventure.

Conflict, Power and Politics in theCity: the book and its arguments

The book, which runs to a trim 133 pages, isorganized into five chapters. Chapter Oneintroduces the basic concepts that providethe foundation for Cox’s framework. ChapterTwo, the shortest, addresses the territorialorganization of metropolitan areas and estab-lishes the context for the two meatierchapters that follow. The first of these(Chapter Three) examines the relationshipbetween ‘metropolitan fragmentation andurban conflict,’ focusing largely on fiscal ormonetary disparities between central city andsuburb. Chapter Four then switches geo-graphical scales to examine conflict within

the city. The concluding chapter, as notedabove, addresses the policy implications ofthe foregoing analysis.

Chapter One establishes the theoreticalbasis for the book and it is worth spendingsome time outlining the basic argument.Following the well-worn tradition of the spa-tial science that defined his department, Coxstarts out with a set of simplifying assumptionsabout the nature of economic and politicallife. Cox initially asks the reader to assume apurely private economy comprising individualdecision-making units in the form of house-holds, firms and other organizations, each ofwhich has a ‘utility-function’ that specifiespreferences for different goods and services or‘commodity bundles.’ Each unit allocatesresources in trying to maximize its utilityby securing its preferred mix of goods andservices. Although the terms may seem quitepeculiar now, the basic argument is a veryfamiliar one for the time.

Where Cox departs from the conventionalview, however, is through the complication thatindividual utilities are ‘not independent of theresource allocations of others’ (Cox, 1973: 2).The thoroughly social nature of urban lifeproves to be a key theme in Cox’s account. Thefact that an individual’s utility is influenced bythe resource allocation of others generateswhat Cox terms ‘externality’ or ‘spillovereffects.’ Externality effects can be either posi-tive or negative depending upon the nature ofwhat they provide. Improvements that ‘beau-tify’ or improve a neighborhood – such asparks or ‘good’ schools – are seen as generat -ing positive externalities to property ownerswithin that neighborhood. In contrast, theintroduction of ‘noxious’ sources or facilitiessuch as sewage farms (water-treatment plants)or homeless shelters generate externalities thatare likely to be perceived as negative by thoseaffected by them. Given the fundamentalassumption that individual decision-makingunits will always maximize their own utility,Cox argues that, for society as a whole, negative

34ÿÿANDY WOOD


externalities or indirect costs tend to be‘overproduced’ while positive externalities orindirect benefits tend to be ‘underproduced.’Accordingly, externality effects ‘detract fromoverall social welfare and pose serious problemsfor society as a whole as well as for the individ-uals that make up that society’ (Cox, 1973: 3).

While externality effects are pervasive,Cox insists they are not spatially random.Instead typical externalities are seen asexplicitly geographical in that the intensity ofexternality effects varies with location rela-tive to the source of the externality. Forexample, those that live adjacent to a park areseen to gain greater benefit than those atsome distance from it. Likewise, those next toa refuse plant or power station are more likelyto be affected by pollution and noise thanthose further away. Furthermore, externalityeffects in the city are magnified by the rela-tive proximity of decision-makers to oneanother and thus to the externalities thatresult from their decision-making. In short,externality effects – whether negative or posi-tive – are hard to avoid in the urban context.Cox attributes much of the variation in localenvironmental quality within the city toexternality effects with the basis of locationalconflict seen as rooted in the conflictbetween the attempt to maximize individualutility and the externality generating effectsof other proximate decision-makers.

Employing a simple symmetry, Cox arguesthat if locational conflicts are the productof externality effects then the resolution ofthose conflicts must involve coordinating theresource allocations that produce these out-comes. To that end there are two basiccoordinating mechanisms or strategies. Thefirst involves the private coordination of activ-ities in which individual decision-makers seekto mitigate the impact of negative externali-ties. The second requires public interventionthrough the centralized, collective decision-making apparatus of government. In the caseof private coordination Cox identifies two

alternative courses of action available todecision-makers. The first – which is essentiallyan ‘exit’ option – involves relocation. The sec-ond is to address immediate externality effectsby bargaining in situ with those that producethem. Cox highlights a range of hypotheticaloutcomes resulting from the two privatestrategies and the conceptual model provides aclear and robust basis for examining alternativecourses of action. Cox argues that solutionsinvolving relocation predominate or, insimple terms, residents generally prefer to exitrather than to bargain. Rather than producinga locational solution maximizing the utility ofall, the result is the very familiar metropolitanpattern in which the city is organized or seg-regated into discrete neighborhoods, each ofwhich is relatively homogenous in terms ofincome and race. Cox (1973: 9) argues thatsuch patterns of stratification reflect the closerelationship between income, race and thetype and nature of externality effects.

The product of such strategies is clearly notoptimal given the basic contradiction betweenindividual utility maximization and the gen-eral welfare of society. For Cox the means foraddressing this tension is found not in theindividual coordinating efforts of private indi-viduals but rather in the coercive and binding,collective decision-making power of govern-ment. Governments are seen as particularlyimportant in providing ‘public goods’ – goodsthat, following Hirschman (1970), are nonex-clusive and equally available. These public goods,which would not be produced by privatemeans, play an important coordinating rolewith respect to externalities. Public health andenvironmental legislation, for example, isdesigned to limit some of the most egregiousnegative externalities, while other forms ofregulation, such as zoning, seek to eliminateexternalities resulting from the juxtapositionof incompatible land uses.

Whereas private forms of coordinationgenerate a de facto pattern of spatially discreteneighborhoods, Cox argues that public

KEVIN COXÿÿ35


coordination gives rise to a de jure pattern ofjurisdictional spaces. Again coordination isseen to mitigate but not eliminate negativeexternality effects. The locational conflictsthat result are neatly divided into thosebetween jurisdictions and those within juris-dictions. Exclusionary zoning on the part ofUS suburbs provides a very clear example ofthe way in which the actions of one govern-mental unit can have detrimental impacts onothers, not least those subsequently excludedfrom suburban housing, such as the low-income populations of the central city.A second type of locational conflict derivesfrom public coordination within jurisdictionsas larger institutions impose externalities onthose subordinate to them. The impurenature of certain public goods provides a raftof opportunities for groups of residents toseek to maximize positive externalities suchas ‘good schools’ and minimize the impact ofnegative externalities.

In the final section of Chapter One, Cox(1973: 14) establishes a ‘relatively simple modelof locational conflict in an urban context.’ Asindicated above, the core of the argument isthat decision-making units – whether house-holds, firms or other organizations – allocateresources to maximize their utility. The variousexternalities that result generate locationalconflict between proximate decision-makers.Locational conflicts can be resolved throughrelocation or through private bargaining,although these solutions tend to be ‘subopti-mal.’ Accordingly, some measure of collectivecontrol over property rights is required toensure more optimal solutions. One of themost interesting and persuasive aspects of thecore argument is its independence of any oneparticular scale of analysis. As Cox (1973: 15)argues, ‘the collectivization of property rightsand the vesting of them in a superordinateauthority … creates new decision-makingunits on a new and larger geographical scale.These units in turn create externalities forother units at the same geographic scale or at

smaller geographic scales and a new cycle oflocational conflict and conflict resolution isgenerated.’

The remainder of the book seeks to applythe conceptual model to conflicts within met-ropolitan areas. Chapter Two establishes thebroad context and examines both the way inwhich populations are localized into neigh-borhoods and jurisdictions as well as howthese entities are organized territorially withinthe city. In short, Cox argues that there are defacto and de jure forms of territorial organiza-tion. The de facto organization of the city isone based on neighborhoods which tendtowards homogeneity in terms of income, eth-nicity and race. The de jure organization thentends to regularize or institutionalize such seg-regation through jurisdictional fragmentationon the one hand and the central city/suburbandivide on the other. Yet segregation into dis-crete neighborhoods cannot guarantee controlof externalities, meaning that securing theprovision of positive externalities commonlydepends upon what Cox terms the ‘respecifi-cation of property rights’ in the form of thecoercive power of governmental authority.The de facto and de jure forms of organizationare routinely coupled and in Chapter TwoCox argues that the power to secure positiveexternalities while avoiding the negative isrelated to the extent to whether neighbor-hoods are ‘organized’ or not. Yet this is not thesole determinant of which locations get whatwith Cox suggesting, as a segue into ChapterThree, that interjurisdictional differences andrelationships are also vitally important.

Chapter Three focuses specifically on whatCox terms the central city–suburban fiscaldisparities problem. In some ways this is apeculiarly American phenomenon giventhe way in which these spatial disparities areinstitutionalized by the fragmentation of thecity into jurisdictions that have considerablelatitude in funding and decision-making.Over the past three decades a considerableamount of effort has been spent examining

36ÿÿANDY WOOD


the nature and extent of divisions betweenUS central cities and their suburbs and possi-ble ways of addressing such disparities.Potential solutions include mechanisms suchas fiscal transfer, metropolitan consolidationand other forms of regional governance.Cox’s original account exposes the powerfuleconomic logic that has generated and thenreproduced city–suburb inequalities. Thecontext here was the relative impoverishmentof the central city, the aging of its housingstock, the fiscal problems created through thesuburbanization of business and the selectivesuburbanization of wealthier white residents,otherwise known as ‘white flight.’

One of the highlights of Chapter Three isCox’s quantitative evaluation of disparities,examining the extent and nature of their vari-ability over space. Cox uses factor analysis toidentify the metropolitan areas that exhibit themost egregious imbalances between ‘need’ ordemand for public services and available taxresources. The analysis indicates ‘the fiscal dis-parities problem is largely a problem of the USNortheast and Midwest’ (Cox, 1973: 46). Coxattributes this to national migration patternsimposing costs or negative externalities onNorthern central cities but also, and perhapsmore significantly, to the locational arrange-ments of individuals and households withinmetropolitan areas. In brief, Cox argues wealth-ier residents are able to capitalize on the positiveexternalities of metropolitan living while escap-ing many of the negative externalities throughsuburban residence. The exclusive nature of thesuburbs is maintained by a number of mecha-nisms. These include exclusionary zoningpractices in the form of minimum lot sizes andzoning for single-family residences. We can seethese practices as a form of NIMBYism (the‘not in my back yard’ syndrome) in which thosewith wealth and resources that depend uponcertain ‘less desirable’ land-uses, such ascommercial and industrial developments,are nevertheless able to exclude them fromtheir particular, immediate neighborhood.

Discrimination within housing and employmentmarkets also plays a significant sorting rolewhile the general disparity in property valuesbetween city and suburb further serves toexclude low-income residents. As Cox (1973:59) suggests, ‘those who can meet the require-ments of the suburban filter suburbanize. Thosewho cannot meet the requirements stay in thecity.’While there are problems here in terms ofthe apparent universal nature of such prefer-ences – think of the way in whichgentrification marked a return to the centralcity by those with economic assets (Ley, 1996) –the selective nature of suburbanization remainsa powerful force underpinning contemporarysocio-spatial disparities within US metropolitanareas.

Chapter Four shifts the scale of analysis toexamine locational conflicts within local juris-dictions. This is an important complement toChapter Three for, as Cox suggests, the elimi-nation of metropolitan fragmentation wouldnot necessarily eliminate inequities in publicallocation and thus the basis for locationalconflict. Variation in educational provision –especially within central cities – is once againseen as a critical source of externalities thatreinforce existing urban inequalities. One ofthe most interesting aspects of Chapter Fouris the way in which Cox makes clear therelationship between political power andinequalities in resource provision and theallocation of public goods. His discussion chal-lenges the pluralist view that each vote carriesequal weight in determining policy outcomes.Towards the end of Chapter Four, Cox exam-ines in more detail two groups that exercisepolitical leverage that greatly outweighs theirelectoral significance. The discussion ofmiddle-class households reaffirms the generalnotion that those with economic assets arebetter able to secure preferable politicaloutcomes. However, the discussion of ‘thedowntown business elite’ hints that certaininterests have an even more vital stake insecuring positive flows of resources and value

KEVIN COXÿÿ37


as a consequence of the immobility of theirinvestments. Those familiar with debates onthe local dependence of certain business andproperty-owners can clearly trace a line ofargument from Conflict, Power and Politics in theCity to the much cited 1988 Annals of theAssociation of American Geographers paper, whereCox develops an argument about the geo-graphic fixity of particular local interests in aglobal era (Cox and Mair, 1988). The sugges-tion that urban inequalities are the product ofthe relationship between economic assets andpolitical power has indeed been a vital one forsubsequent developments in urban geographyand urban studies more broadly.

Chapter Five examines the policy impli-cations of Cox’s model of metropolitaninequalities and the conflicts they generate.His is an explicitly redistributive argument andin this sense a liberal claim for social change incontrast to the more radical strategies favoredby later Marxist accounts. Cox argues socialinjustice is the product of two basic conditions.The first is the spatial organization of theurban political system such that the benefitsderived from the public allocation of resources‘depends very much on location relative to …job opportunities, schools, freeways, and theboundaries of different municipalities and ofschool catchment areas’ (Cox, 1973: 106). Butinjustice is also seen as the product of a secondbasic inequality in the allocation of privateresources, with those with economic powerable to lobby and influence more effectivelythan those without. In short, Cox argues spatialinequity in public provision is ‘strongly corre-lated with social inequity’ (Cox, 1973: 126) andwhile his discussion of policy options isdivided into those that effect changes in spatialorganization and those that focus on the socialallocation of income and wealth it is clear thatthe favored solution is some combination of achange in metropolitan form – such as metro-politan integration – combined with aneffective redistribution of income through tax-ation. In his estimation, greater social and

spatial equalization of resources should serve tomitigate or even eliminate negative externali-ties, undermine segregation and thus producea more just and equitable city.

Evaluating Conflict, Power andPolitics in the City

The strengths and limits of Conflict, Power andPolitics in the City can be examined at twodifferent levels. The first is in terms of thebook’s immediate reception; the second itsmore lasting impact. The initial reception ofthe book was decidedly positive. MurrayAustin, in the Annals of the Association ofAmerican Geographers, referred to it as ‘a majorcontribution to the study of urban problems’(Austin, 1973: 389); Melvin Albaum (1973) sawit as ‘an important volume’ in the McGraw-Hill series, while Norman Walzer describedit as ‘an illuminating analysis’ (Walzer, 1973:476). These initial reviews pointed to anumber of clear strengths. The first is theunabashedly conceptual nature of Cox’s con-tribution (Ironside, 1976). Conflict, Power andPolitics in the City provided possibly the firstspecifically geographical framework forexamining the city and its politics. In doingso it builds upon a set of key concepts andideas in which geographical notions such asproximity, location, accessibility and scale arecentral. Over the years there have been manycontributions to the geography literature thathave sought to ‘add’ geography to an existingmix of concepts and claims. Such work hastended to spatialize ideas and theories fromother disciplines rather than develop the geo-grapher’s conceptual toolbox. In contrast, thefoundation of Cox’s argument is a set of geo-graphical concepts that are developed into anoverarching framework that insists on thecentrality of space in producing and resolvingconflict in the city.

Second, it is apparent that the book blendsconceptual rigour with a rich and varied

38ÿÿANDY WOOD


array of empirical illustrations. These aredrawn overwhelmingly from the US withone or two complementary examples fromthe UK (see below for the limits to such astrategy). But this is by no means a book thatargues by empirical example. Instead Coxdraws on a wide range of materials, includingacademic case studies, government reportssuch as Senate and Congressional records,and newspaper sources to present a range ofcases that illustrate the broader conceptualframework and the geographical basis ofurban politics. The examples are generallyroutine rather than exceptional but thatserves as an even more effective illustration ofthe general purchase of Cox’s framework. Athird notable contribution of the text is thatit clearly establishes the policy implications ofthe analysis.

The book also had a more lasting influenceon a number of subsequent literatures. The firstwas on locational politics and the study of con-flicts arising from externalities (O’Loughlinand Munski, 1979; Ley and Mercer, 1980).These themes have been of particular interestto urban geographers and the study of negativeexternalities and NIMBYism has proved to beespecially enduring (Lake, 1987). A second setof geographical studies examined questions ofinequality in access to public facilities and inthe distribution of resources across the city (fora review see DeVerteuil, 2000). Thirdly, Cox’sbook was influential – largely beyond geogra-phy this time – in the study of the politicalactivity of neighborhood groups (Rich, 1980).

While the book has a number of very pos-itive attributes, with the benefit of hindsight,it is possible to identify four key limitationsto Cox’s text. Hindsight always provides per-spective and many of the limitations havebeen brought into sharper relief by subse-quent developments in human geography.The first – and arguably the most significant –is the book’s reliance on an economisticpublic choice model centered on the rational,utility-maximizing individual. For Cox,

individual residents necessarily maximize thereturn on their investments, calculate thecosts and returns to alternative options andmechanistically follow the course given bythe most rational return. In 1973 this was thedominant view in economics and Cox sim-ply applied what he saw as ‘intellectuallyintriguing’ work to the question of locationalallocation and conflict in the city. The limita-tions of such an approach are now wellknown – including its assumptions about therational and self-interested nature of individ-ual action – and even the initial reviews hadstarted to question the viability of such anapproach. As time has gone on the approachlooks increasingly archaic and at a time whencities are once more rising up the academicagenda the social theories that underpinthe normative claims of geographers such asAsh Amin (2006), David Harvey (2000) andNigel Thrift (2005) could not be moredifferent from the neoclassical model thatformed the foundation for Cox’s account.

If the first limitation derives from thebook’s commitment to what became a quiteproblematic theoretical framework then thesecond reflects the book’s rather limitedrange. Again we should not see Conflict, Powerand Politics in the City in isolation given thatmany contemporaneous texts also focused onthe question of ‘who gets what, where’(Castells, 1977 is the most prominent com-pany here). From this perspective urbanconflict is seen as conflict over the distribu-tion of resources as well as the relativebalance between costs and benefits in pro-ducing use values. In retrospect, Kevin Coxwould see this as a major limitation and histurn to Marxism clearly affirmed the limita-tions of an approach that ignored questionsof production and the generation of surplus.For those who were influenced by Marx, thesocial welfare question of ‘who gets what’was displaced in the 1970s and early 1980s bya concern with how the ‘what’ is produced inthe first place and by an analysis of the social

KEVIN COXÿÿ39


relations forged in the sphere of production.For Cox himself, this shift involved less arejection of the significance of the ‘urban’and its politics and more an attempt to linkurban conflicts and outcomes to class rela-tionships rather than to those manifest in theneighborhoods, school districts and residentsassociations that are central to Conflict, Powerand Politics in the City.

A third limitation similarly reflects thebook’s limited terms of reference. The essenceof Conflict, Power and Politics in the City is anattractive but ultimately abstract account ofthe mechanisms producing conflict in the city.But of course the city is a particular one – theUS city. The nature of the McGraw-Hill seriesin combination with Cox’s own interestslargely dictated the geographical limits of thebook, but the failing here is less the focus on asingle country – and, as noted above, there areexamples drawn from Britain that offset thatparticular critique – and more that the bookfailed to recognize the particularities of theUS case. The dominance of private propertyand the intense fragmentation of governmen-tal authority are widespread conditions forsure, but they are not universal – Havana is notGuangzhou which is not Detroit – and in thisrather obvious sense the book perhaps failedto talk to as wide an audience as possible.

Fourth, while the book is resolutely geo-graphical the geographies are very clearlydefined by notions of fixity, territory andbounded space. In some ways this is an artifactof Cox’s focus on the neighborhood as the pri-mary locus of social life and the claim thatneighborhoods tend to be homogenous withrespect to income and race. With this assump-tion Cox avoids questions relating to thedynamics of the household and especially gen-der relations (Drake and Horton, 1983), as wellas the influence of the national and thetransnational on the city. The 1970s marked areversal in the long-term demographic trendof a decreasing proportion of foreign-born inthe US population and in cities like Miami,

Los Angeles and New York the role of thetransnational came to play a much moreprominent role in the urban imaginary. Thiswas partly the consequence of migration butalso, perhaps, about the increasing integrationof these cities into regional and global flowsof investment, trade and ideas. While ‘globaliza-tion’ had yet to emerge as a significant academicconcern, Cox’s emphasis on the territorial, thebounded and the fixed has seemed to lose itsresonance in a globalizing world in which thefocus has shifted to the mobile, the mutable andthe unbound.

Conclusion

Although Cox’s book was never as widelycited as many of the other classic texts in thiscollection, the influence of Conflict, Power andPolitics in the City endured well beyond theinitial flurry of positive reviews. Reading ittoday it remains a simple, elegant andyet rigorous argument about the nature ofconflict in the US city. Yet this is not a bookthat serves as a model statement of any oneparticular view, approach or era. Indeed itsenduring legacy might best be seen as atransitional work between a spatial sciencetradition characterized by parsimony andsimplicity and a much more politicized viewof the world in which political influence andpower are closely tied to economic assets andinterests.

DeVerteuil (2000: 59) argues that Cox’swork was ‘instrumental in recasting facilitylocation away from neoclassical, quantitativemodels normatively concerned with spatialoutcomes and toward a more conflictual,political and socially embedded analyticalframework concerned with spatial processes.’While DeVerteuil’s review is focused on oneaspect of Conflict, Power and Politics in the Cityhis assessment of Cox’s contribution could befruitfully applied to a range of additional issuesand themes. Indeed, the most enduring legacy

40ÿÿANDY WOOD


of all was precisely this focus on the politicalnature of locational conflict and Cox’s empha-sis on the significance of government – orwhat we would later term ‘the state’ – in shap-ing the city. Subsequent work, not least hisown, would develop these ideas in terms of thepolitical-economic dynamics of the city butCox’s book insists that the city is the productof forces that derive their power from privateproperty on the one hand and governmentalor state authority on the other. These twosources of power, and the relationship betweenthem, would form the basis for a great deal ofsubsequent work on the nature and dynamicsof the city. Cox has subsequently made a sub-stantial direct contribution to this field through

his work on the ‘new urban politics’ (see Cox,1993), its relationship to globalization (Cox,1995) as well as his work on the geographicspecificity of politics and political conflict (Coxand Mair, 1988; Cox, 1998). Much of this workhas continued to emphasize the relationshipbetween state power and the realization ofeconomic interests in the city. While the rejec-tion of what had come before was by no meansas complete or dramatic as Harvey’s book ofthe same year (1973) the central focus on therelationship between private property and theexercise of political power surely marked awelcome turn away from avowedly apoliticallocation theory towards a thoroughly politi-cized urban geography.

KEVIN COXÿÿ41

Secondary sources and referencesAlbaum, M. (1973) ‘Conflict, Power and Politics in the City’ (book review), Professional

Geographer 26 (3): 339.Amin, A. (2006) ‘The good city’, Urban Studies 43 (5–6): 1009–1023.Austin, C.M. (1973) ‘Conflict, Power and Politics in the City’ (book review), Annals of the

Association of American Geographers 63 (3): 389–390.Bach, W. (1972) Atmospheric Pollution. New York: McGraw-Hill.Barnes, T.J. (2004) ‘Placing ideas: genus loci, heterotopia and geography’s quantitative

revolution’, Progress in Human Geography 28 (5): 565–595.Castells, M. (1977) The Urban Question: A Marxist Approach. London: Edward Arnold. Cox, K.R. (1973) Conflict, Power, and Politics in the City: A Geographic View. New York:

McGraw-Hill.Cox, K.R. (1993) ‘The local and the global in the new urban politics – a critical view’,

Environment and Planning D: Society and Space 11 (4): 433–448.Cox, K.R. (1995) ‘Globalization, competition and the politics of local economic development’,

Urban Studies 32 (2): 213–224.Cox, K.R. (1998) ‘Spaces of dependence, spaces of engagement and the politics of

scale, or: looking for local politics’, Political Geography 17 (1): 1–23.Cox, K.R. and Mair, A. (1988) ‘Locality and community in the politics of local

economic development’, Annals of the Association of American Geographers78 (2): 307–325.

DeVerteuil, G. (2000) ‘Reconsidering the legacy of urban public facility location theory inhuman geography’, Progress in Human Geography 24 (1): 47–69.

Drake, C. and Horton, J. (1983) ‘Comment on editorial essay: sexist bias in political geog-raphy’, Political Geography Quarterly 2 (4): 329–337.

Harvey, D. (1973) Social Justice and the City. London: Edward Arnold.Harvey, D. (2000) Spaces of Hope. Berkeley: University of California Press.Hirschman, A. (1970) Exit, Voice and Loyalty. Cambridge: Harvard University Press.


42ÿÿANDY WOOD

Ironside, R.G. (1976) ‘Conflict, Power and Politics in the City’ (book review),Geographical Analysis 8 (1): 104–109.

Lake, R. (1987) Resolving Locational Conflict. New Brunswick, NJ: Center for UrbanPolicy Research.

Ley, D. (1974) The Inner City Ghetto as Frontier Outpost. Washington, DC: Associationof American Geographers.

Ley, D. (1996) The New Middle Class and the Remaking of the Central City. Oxford:Oxford University Press.

Ley, D. and Mercer, J. (1980) ‘Locational conflict and the politics of consumption’,Economic Geography 56 (2): 89–109.

Morrill, R. and Wohlenberg, E. (1971) The Geography of Poverty in the United States.New York: McGraw-Hill.

O’Loughlin, J. and Munski, D.C. (1979) ‘Housing rehabilitation in the inner city –comparison of 2 neighborhoods in New Orleans’, Economic Geography 55 (1): 52–70.

Rich, R.C. (1980) ‘A political-economy approach to the study of neighborhood organiza-tions’, American Journal of Political Science 24 (4): 559–592.

Smith, D. (1973) The Geography of Social Well-being in the United States. New York:McGraw-Hill.

Thrift, N. (2005) ‘But malice aforethought: cities and the natural history of hatred’,Transactions of the Institute of British Geographers 30 (2): 133–150.

Walzer, N. (1973) ‘Conflict, Power and Politics in the City’ (book review), Journal ofRegional Science 13 (3): 473–477.

Ward, D. (1971) Cities and Immigrants. New York: Oxford University Press.


PLACE AND PLACELESSNESS (1976):EDWARD RELPH

David Seamon and Jacob Sowers

A deep human need exists for associations withsignificant places. If we choose to ignore thatneed, and to allow the forces of placelessnessto continue unchallenged, then the future canonly hold an environment in which placessimply do not matter. If, on the other hand,we choose to respond to that need and to tran-scend placelessness, then the potential existsfor the development of an environment in whichplaces are for [people], reflecting and enhanc-ing the variety of human experience. Whichof these two possibilities is most probable,or whether there are other possibilities, is farfrom certain. But one thing at least is clear –whether the world we live in has a placelessgeography or a geography of significant places,the responsibility for it is ours alone. (Relph,1976: 147)

Introduction

Geographers have long spoken of the impor-tance of place as the unique focus distinguishinggeography from other disciplines. Astronomyhas the heavens, History has time, andGeography has place. A major question thatgeographers must sooner or later ask, however,is ‘What exactly is place?’ Is it merely a synonymfor location, or a unique ensemble of nature andculture, or could it be something more?

Beginning in the early 1970s, geographerssuch as Yi-Fu Tuan (1974), Anne Buttimer(1976), and Edward Relph (1976, 1981,

1993) grew dissatisfied with what they feltwas a philosophically and experientiallyanemic definition of place. These thinkers,sometimes called ‘humanistic geographers,’probed place as it plays an integral rolein human experience. One influential resultof this new approach was Edward Relph’sPlace and Placelessness, a book that continuesto have significant conceptual and practicalimpact today, both inside and outsidegeography.

In the early 1970s, Relph was a doctoralstudent at the University of Toronto, work-ing on his dissertation concerning therelationship between Canadian nationalidentity and the symbolic landscapes of theCanadian Shield, especially those representedby lakes and forests (Relph, 1996). As his pro-ject progressed, he became dissatisfied withthe lack of philosophical sophistication givento the definition of place. Relph found thissupposed conceptual pillar of the disciplineto be superficial and incomplete, especially interms of the importance of place in ordinaryhuman life. How could one study placeattachment, sense of place, or place identitywithout a clear understanding of the depthand complexity of place as it is experiencedand fashioned by real people in real places?Eventually, Relph scrapped his CanadianShield study and shifted focus to a broaderlook at the nature and meaning of place as itplays an integral part in the lives of humanbeings.

6


A phenomenology of place and space

Published in 1976, Place and Placelessnessis a substantive revision of Relph’s 1973University of Toronto doctoral dissertation inGeography. As he emphasizes at the startof the book, his research method is ‘a phe-nomenology of place’ (Relph, 1976: 4–7).Phenomenology is the interpretive study ofhuman experience. The aim is to examineand to clarify human situations, events,meanings, and experiences as they are knownin everyday life but typically unnoticedbeneath the level of conscious awareness(Seamon, 2000). One of phenomenology’sgreat strengths is seeking out what is obviousbut unquestioned and thereby questioning it.To uncover the obvious, we must step backfrom any taken-for-granted attitudes andassumptions, whether in the realm of every-day experience or in the realm of conceptualperspectives and explanations, including thescientific. In Place and Placelessness, Relphsteps back to call into question the taken-for-granted nature of place and its significance asan inescapable dimension of human life andexperience.

Relph begins Place and Placelessness with areview of space and its relationship to place. Heargues that space is not a void or an isometricplane or a kind of container that holds places.Instead, he contends that, to study the relation-ship of space to a more experientially-basedunderstanding of place, then space too must beexplored in terms of how people experience it.Although Relph says that there are countlesstypes and intensities of spatial experience, hedelineates a heuristic structure grounded in ‘acontinuum that has direct experience at oneextreme and abstract thought at the other…’(Relph, 1976: 9). On one hand, he identifiesmodes of spatial experience that are instinctive,bodily, and immediate – for example, what hecalls pragmatic space, perceptual space, and exis-tential space. On the other hand, he identifies

modes of spatial experience that are morecerebral, ideal, and intangible – for example,planning space, cognitive space, and abstractspace. Relph describes how each of thesemodes of space-as-experienced has varyingintensities in everyday life. For example, exis-tential space – the particular taken-for-grantedenvironmental and spatial constitution of one’severyday world grounded in culture and socialstructure – can be experienced in a highlyself-conscious way such as when one is over-whelmed by the beauty and sacredness of aGothic cathedral; or in a tacit, unself-consciousway as one sits in the office day after day pay-ing little attention to his or her surroundings.

Although the spatial modes that Relphidentifies may each play a particular role ineveryday experience, Relph emphasizes thatin reality these modes are not mutually exclu-sive but all part and parcel of human spatialexperience as a lived, indivisible whole. Forexample, he explains that cognitive concep-tions of space understood through maps mayhelp to form our perceptual knowledge,which in turn may color our day-to-day spa-tial encounters as we move through real-worldplaces. Though a radical idea in the 1970s,Relph’s conclusion that space is heteroge-neous and infused with many different liveddimensions is largely taken for granted in geo-graphical studies today as researchers speak ofsuch spatial modes as sacred space, genderedspace, commodified space, and the like.

One of Relph’s central accomplishments inPlace and Placelessness is his preserving an inti-mate conceptual engagement between spaceand place. Many geographers speak of bothconcepts but ultimately treat the two as sepa-rate or give few indications as to how they arerelated existentially and conceptually. ForRelph, the unique quality of place is its powerto order and to focus human intentions, expe-riences, and actions spatially. Relph thus seesspace and place as dialectically structured inhuman environmental experience, since our

44ÿÿDAVID SEAMON AND JACOB SOWERS


understanding of space is related to the placeswe inhabit, which in turn derive meaningfrom their spatial context.

Depth of place

A central reason for Relph’s exhaustive studyof place is his firmly held belief that suchunderstanding might contribute to the main-tenance and restoration of existing places andthe making of new places (also see Relph,1981, 1993). He argues that, without a thor-ough understanding of place as it has humansignificance, one would find it difficult todescribe why a particular place is special andimpossible to know how to repair existingplaces in need of mending. In short, beforewe can properly prescribe, we must first learnhow to accurately describe – a central aim ofphenomenological research.

In examining place in depth, Relph focuseson people’s identity of and with place. By theidentity of a place, he refers to its ‘persistentsameness and unity which allows that [place]to be differentiated from others’ (Relph, 1976:45). Relph describes this persistent identity interms of three components: (1) the place’sphysical setting; (2) its activities, situations, andevents; and (3) the individual and groupmeanings created through people’s experi-ences and intentions in regard to that place.

Relph emphasizes, however, that placeidentity defined in this threefold way is notsufficiently pivotal or deep existentiallybecause, most essentially, places are ‘signifi-cant centres of our immediate experiences ofthe world’ (Relph, 1976: 141). If places are tobe more thoroughly understood, one needs alanguage whereby we can identify particularplace experiences in terms of the intensity ofmeaning and intention that a person andplace hold for each other. For Relph, thecrux of this lived intensity is identity withplace, which he defines through the concept

of insideness – the degree of attachment,involvement, and concern that a person orgroup has for a particular place.

Insideness and outsideness

Relph’s elucidation of insideness is perhapshis most original contribution to the under-standing of place because he effectivelydemonstrates that this concept is the corelived structure of place as it has meaningin human life. If a person feels inside a place,he or she is here rather than there, safe ratherthan threatened, enclosed rather than exposed,at ease rather than stressed. Relph suggeststhat the more profoundly inside a place aperson feels, the stronger will be his or heridentity with that place.

On the other hand, a person can be sepa-rate or alienated from place, and this mode ofplace experience is what Relph calls outside-ness. Here, people feel some sort of liveddivision or separation between themselvesand world – for example, the feeling ofhomesickness in a new place. The crucialphenomenological point is that outsidenessand insideness constitute a fundamentaldialectic in human life and that, throughvarying combinations and intensities of out-sideness and insideness, different places takeon different identities for different individu-als and groups, with human experiencetaking on different qualities of feeling, mean-ing, ambience, and action.

The strongest sense of place experience iswhat Relph calls existential insideness – a situationof deep, unself-conscious immersion in placeand the experience most people know whenthey are at home in their own community andregion. The opposite of existential insideness iswhat he labels existential outsideness – a sense ofstrangeness and alienation, such as that often feltby newcomers to a place or by people who,having been away from their birth place, return

EDWARD RELPHÿÿ45


to feel strangers because the place is no longerwhat it was when they knew it earlier.

In his book, Relph discusses seven modesof insideness and outsideness (no doubt thereare more) grounded in various levels of expe-riential involvement and meaning. The valueof these modes, particularly for self-awareness,is that they apply to specific place experiencesyet provide a conceptual structure in whichto understand those experiences in broader,more explicit terms.

Placelessness

In the last half of the book, Relph examinesways in which places may be experiencedauthentically or inauthentically (terms borrowedfrom phenomenological and existential philos-ophy). An authentic sense of place is ‘a directand genuine experience of the entire complexof the identity of places – not mediated anddistorted through a series of quite arbitrarysocial and intellectual fashions about howthat experience should be, nor followingstereotyped conventions’ (Relph, 1976: 64).

Individuals and groups may create a sense ofplace either unself-consciously or deliberately.Thus, because of constant use, a nondescripturban neighborhood can be as authentic aplace as Hellenic Athens or the Gothic cathe-drals – the latter both examples, for Relph, ofplaces generated consciously. Relph arguesthat, in our modern era, an authentic sense ofplace is being gradually overshadowed by a lessauthentic attitude that he called placelessness:‘the casual eradication of distinctive places andthe making of standardized landscapes thatresults from an insensitivity to the significanceof place’ (Relph, 1976: Preface).

Relph suggests that, in general, placelessnessarises from kitsch – an uncritical acceptanceof mass values, or technique – the overridingconcern with efficiency as an end in itself.The overall impact of these two forces, whichmanifest through such processes as mass

communication, mass culture, and centralauthority, is the ‘undermining of place forboth individuals and cultures, and the casualreplacement of the diverse and significantplaces of the world with anonymous spacesand exchangeable environments’ (Relph,1976: 143).

Influence of Place and Placelessness

Since Relph’s book was published, there hasbeen a spate of popular studies on the natureof place. In addition, thinkers from a broadrange of conceptual perspectives – from posi-tivist and neo-Marxist to post-structuralist andsocial-constructivist – have drawn on the ideaof place, though understanding it in differentways and using it for different theoretical andpractical ends (Creswell, 2004; Seamon, 2000).

Scholarly interest in Place and Placelessnesshas steadily increased over the years. Accordingto citation indices in the sciences, social sci-ences, and the arts and humanities, the bookhas been referenced in scholarly journals atotal of 357 times from 1977 to 2005. In thefirst 10 years, there was an average of some 12citations per year; since then, references havesteadily increased to 36 entries in 2004.Geographers have cited the book most since1989 (142 entries), though scholars in envi-ronmental studies also demonstrate stronginterest (118 entries). In addition, the book hasbeen cited by researchers in psychology (43times), sociology (42), urban studies (30), plan-ning (21), health (10), and anthropology (9).

To provide the reader with an indicationof how Relph’s ideas in Place and Placelessnesshave been used as a major conceptual moor-ing point by other researchers, we highlightthree examples – one book, one article, andone dissertation (for a more extensive list, seeSeamon, 2000). Published two years afterRelph’s book, geographer David Seamon’s AGeography of the Lifeworld is the first majorstudy to draw on Relph’s notion of insideness



and to demonstrate how it could beextended phenomenologically to examine atopic that Seamon calls everyday environmentalexperience – the sum total of people’s firsthandinvolvements with the geographical world inwhich they live (Seamon, 1979: 15–16).Seamon considers how, through experienceddimensions like body, feelings, and thinking,the quality of insideness is expressed geo-graphically and environmentally. Seamon’swork illustrates how Relph’s phenomenol-ogy of place offers a field of conceptualclarity from which other researchers mightembark on their own phenomenologicalexplorations.

A second study illustrating the conceptualpotential of Place and Placelessness is landscapearchitect V. Frank Chaffin’s research, whichfocuses on Isle Brevelle, a 200-year-old rivercommunity on the Cane River of Louisiana’sNatchitoches Parish (Chaffin, 1989). Throughan interpretive reading of the region’s historyand geology, in-depth interviewing of resi-dents, and his own personal encounters withIsle Brevelle’s landscape while canoeing on theCane River, Chaffin aims to reach an empatheticinsideness with this place – in other words,he attempts to find ways to be open toand thereby to understand more deeply IsleBrevelle’s unique sense of place. One centralaspect of Chaffin’s encounter with this place ishis unexpected realization that the Cane Riveris not an edge that separates its two banks but,rather, a seam that gathers the two sidestogether as one community and one place.

A third study using themes from Place andPlacelessnesss for conceptual mooring is psy-chologist Louise Million’s dissertation (Million,1992), which examines phenomenologicallythe experience of five rural Canadian familiesforced to leave their ranches because of theconstruction of a reservoir dam in southernAlberta. Drawing on Relph’s modes of inside-ness and outsideness, Million identifies thecentral lived qualities of what she calls involun-tary displacement – the families’ experience of

forced relocation and resettlement. Making useof in-depth interviews with the five families,she demonstrates how place is prior to invol-untary displacement, with the result that thisexperience can be understood existentially as aforced journey marked by eight stages –(1) becoming uneasy, (2) struggling to stay,(3) having to accept, (4) securing a settlement,(5) searching for the new, (6) starting over,(7) unsettling reminders, and (8) wanting to resettle.In delineating the lived stages in the processof losing place and attempting to resettle,Million’s study demonstrates how Relph’smodes of insideness and outsideness can beused developmentally to examine placeexperience and identity as they strengthen,weaken, or remain more or less continuousover time.

Criticisms of Place and Placelessness

Broadly, one finds three major criticisms ofPlace and Placelessness: that it is essentialist; outof touch with what places really are today;and structured around simplistic dualismsthat misrepresent and limit the range of placeexperience, particularly the possibility of a‘global sense of place’ (Massey, 1997: 323).The essentialist claim has been brought forthespecially by Marxists (e.g., Peet, 1998: 63)and social constructivists (e.g., Cresswell,2004: 26, 30–33), who argue that Relphpresupposes and claims an invariant and uni-versal human condition that will be revealedonly when all ‘non-essentials,’ includinghistorical, cultural, and personal qualities,are stripped away, leaving behind someinescapable core of human experience. Thesecritics point out that, in focusing on theexperience of place as a foundational exis-tential quality and structure, Relph ignoresspecific temporal, social, and individual cir-cumstances that shape particular places andparticular individuals’ and groups’ experienceof them.

EDWARD RELPHÿÿ47


This criticism misunderstands the basic phe-nomenological recognition that there aredifferent dimensions of human experience andexistence that all must be incorporated in a thor-ough understanding of human and societalphenomena. These dimensions include: (a)one’s unique personal situation – e.g., one’sgender, physical and intellectual endowments,degree of ableness, and personal likes and dis-likes; (b) one’s unique historical, social, andcultural situation – e.g., the era and geographi-cal locale in which one lives, his or hereconomic and political circumstances, and hisor her educational, religious, and societal back-ground; and (c) one’s situation as a typicalhuman being who sustains and reflects a typi-cal human world – e.g., Relph’s claim thatplace is an integral lived structure in humanexperience.

What is exciting and dynamic about Relph’sbroad conclusions regarding place – whichfirst of all relate to dimension (c) – is theirpotential as starting points for more specificphenomenological investigations of (a) and (b)as exemplified, for example, in the real-worldstudies of Chaffin and Million highlightedabove. Chaffin’s research demonstrates howRelph’s broad principles can inform and directphenomenological research focusing on thesocial and cultural dimensions of one specificplace – the Cain River community. Similarly,in her study of the displaced Alberta ranchers,Million illustrates how Relph’s general princi-ples and conclusions can guide empiricalresearch in regard to specific individuals andfamilies in a specific place, time, and situation.In turn, Chaffin and Million’s more groundeddiscoveries clarify and amplify Relph’s broaderclaims.

In short, Relph’s phenomenology of placepoints toward a conceptual and methodolog-ical reciprocity between the general and thespecific, between the foundational and the par-ticular, between the conceptual and the lived.This convincing ‘fit’ among levels is a hall-mark of the best phenomenology.

A lack of conceptual sophistication?

In a commentary written for Place andPlacelessness’ 20th anniversary, Relph (1996)suggested that, in hindsight, another majorweaknesses of Place and Placelessness was itslack of conceptual sophistication, particularlyits straightforward use of dialectical oppositesas a way to conceptualize place experience –insideness/outsideness, place/placelessness,authenticity/inauthenticity, and so forth.One result is that critics have often misun-derstood Relph’s point of view, claiming hefavored places over placelessness, insidenessover outsideness, authentic over inauthenticplaces, rootedness over mobility, and place asa static, bounded site over place as a dynamic,globally-connected process (Cresswell, 2004;Massey, 1997; Peet, 1998).

If, however, one reads the book carefully anddraws on his or her own personal experiencesof place for evidence and clarification, he or sherealizes the extraordinary coverage and flexibil-ity of Relph’s conceptual structure. Especiallythrough the continuum of insideness and out-sideness, he provides a language that allows fora precise designation of the particular experi-ence of a particular person or group in relationto the particular place in which they find them-selves. Relph also provides a terminology fordescribing how and why the same place can beexperienced differently by different individuals(e.g., the long-time resident vs. the newcomervs. the researcher who studies the place) or how,over time, the same person can experience thesame place differently at different times (e.g.,the home and community that suddenly seemso different when one’s significant other dies).

As Relph’s book strikingly demonstrates, amajor strength of phenomenological insights istheir provision of a conceptual language thatallows one to separate from taken-for-grantedeveryday experience – the lifeworld as it is calledphenomenologically (Buttimer, 1976; Seamon,1979, 2000). Too often, researchers lose sightof the need to move outside lifeworld



descriptions and terminology, and the result isconfusion or murkiness as to the exact phe-nomenon they are attempting to understand.

For example, in feminist and cultural-studies research that focuses on negative andtraumatic images of place (e.g., Rose, 1993:53–55), an emphasis is sometimes given tohow family violence generates homes wherefamily members feel victimized and insecure.Too often, the post-structural and social-constructivist conclusion is to call into ques-tion the entire concept of home and placeand to suggest that they might be nostalgic,essentialist notions that need vigorous societaland political modification – perhaps evensubstitution – in postmodern society.

Relph’s modes of insideness and outside-ness point to an alternative understanding.The problem is not home and place but aconceptual conflation for which Relph’s lan-guage provides a simple corrective: thevictim’s experience should not be interpretedas a lack of at-homeness but, rather, as onemode of existential outsideness, which inregard to one’s most intimate place – thehome – is particularly undermining andpotentially life-shattering.

Relph’s notion of existential outsidenessallows us to keep the experiences of homeand violation distinct. Through his lived lan-guage of place, we can say more exactly thatdomestic violence, whether in regard towomen or men, is a situation where a placethat typically fosters the strongest kind ofexistential insideness has become, paradoxi-cally, a place of overwhelming existentialoutsideness. The lived result must be pro-foundly destructive.

The short-term phenomenological ques-tion is how these victims can be helped toregain existential insideness. The longer-termquestion is what qualities and forces in oursociety lead to a situation where the existentialinsideness of home and at-homeness devolvesinto hurtfulness and despair. Something isdeeply wrong, and one cause of the problem

may be the very problem itself – i.e., the grow-ing disruption and disintegration of places andinsideness at many different scales of experi-ence, from home to neighborhood to city tonation (Fullilove, 2004; Relph, 1993).

How today to have insideness and placewhen change is constant, society is diverse,and so many of the traditional ‘truths’ nolonger make sense is one of the crucial ques-tions of our age. Place and Placelessness offersno clear answer, but it does provide an inno-vative language for thinking about thequestion.

Dwelling and journey

Another concern that some critics voicedregarding Place and Placelessness is that itfavors home, center, and dwelling over hori-zon, periphery, and journey (Cresswell, 2004;Massey, 1997; Peet, 1998). As Relph (1996)says in his 20th-anniversary commentary, hewas accused of emphasizing the positivequalities of place and ignoring or minimizingnegative qualities – e.g., the possibility thatplace can generate parochialism, xenophobia,and narrow-mindedness (also see Relph,2000). Again, a close reading of the bookreveals a flexibility of expression – a recogni-tion that an excess of place can lead to aprovincialism and callousness for outsidersjust as an excess of journey can lead to a lossof identity or an impartial relativity thatallows for commitment to nothing. Thebroader point is that, in the book’s liveddialectics (center/horizon, place/placelessness,and so forth), there is a wonderful resilience ofconceptual interrelationship that is anotherhallmark of the best phenomenology.

In his 20th-anniversary commentary, Relph(1996) also points out that some criticsmistakenly read the book as a nostalgic paeanto pre-modern times and places (e.g., Peet,1998). How could the kind of authentic placesthat he emphasized exist in our postmodern

EDWARD RELPHÿÿ49


times of cyberspace, continuous technologicalchange, human diversity, and geographical andsocial mobility?

This criticism, of course, ignores a centralconclusion of Place and Placelessness: thatregardless of the historical time or the geo-graphical, technological, and social situation,people will always need place because having andidentifying with place are integral to whatand who we are as human beings (Casey,1993; Malpas, 1999). From this point of view,the argument that postmodern society, throughtechnological and cultural correctives, cannow ignore place is questionable existentiallyand potentially devastating practically, whetherin terms of policy, design, or popular under-standing (Relph, 1993).

Instead, the crucial question that boththeory and practice should ask is how a‘progressive’ sense of place and insidenesscan be made even in the context of our rela-tivist, constantly-changing postmodernworld (Cresswell, 2004; Horan, 2000; Massey,1997). Twenty years ago, Relph was one of thefirst thinkers to broach this question, which heexplores in greater detail in his later RationalLandscapes and Humanistic Geography (Relph,1981). Today, due to Relph’s penetrating insightsand the work of a small coterie of thinkers andpractitioners like Christopher Alexander (2002–2005), Mindy Fullilove (2004), Bill Hillier(1996), Thomas Horan (2000), Daniel Kemmis(1995), and Robert Mugerauer (1994), wehave the start of an answer to this questionphrased in a phenomenological language(Seamon, 2004) that interprets place in a wayconsiderably different from the post-structural,social-constructivist, and neo-Marxistperspectives that currently dominate academicdiscourse on place (Cresswell, 2004).

In spite of the dramatic societal and envi-ronmental changes that our world faces today,place continues to be significant both as avigorous conceptual structure as well as anirrevocable part of everyday human life(Horan, 2000). This is not to suggest that theworld must or could return to a set of distinctplaces all different, unconnected, and more orless unaware of each other. In today’s globally-linked society, place independence is in manyways impossible (Cresswell, 2004; Relph,2000). More so, the importance of place andlocality must be balanced with an awarenessof and connections to other places and globalneeds (Massey, 1997). The point is that anempathetic and compassionate understandingof the worlds beyond our own places may bebest grounded in a love of a particular placeto which I myself belong. In this way, we mayrecognize that what we need in our everydayworld has parallels in the worlds of others(Relph, 1981, 1993).

Conclusion

Place and Placelessness is a remarkable demon-stration of the potential conceptual andpractical power of place, which, by its verynature, gathers worlds spatially and environ-mentally, marking out centers of humanaction, intention, and meaning that, in turn,help make place (Casey, 1993; Malpas, 1999).In many ways, the continuing dissolution ofplaces and insideness in the world helps toexplain the escalating erosion of civility andcivilization, in the West and elsewhere.Relph’s Place and Placelessness first pointed tothis dilemma some 30 years ago and is todaymore relevant than ever.


Secondary sources and referencesAlexander, C. (2002–2005) The Nature of Order (4 volumes). Berkeley, California: Center

for Environmental Structure.Buttimer, A. (1976) ‘Grasping the dynamism of lifeworld’, Annals of the Association of

American Geographers 66: 277–292.


EDWARD RELPHÿÿ51

Casey, E. (1993) Getting Back into Place: Toward a Renewed Understanding of thePlace-World. Bloomington: Indiana University Press.

Chaffin, V.F. (1989) ‘Dwelling and rhythm: The Isle Brevelle as a landscape of home’,Landscape Journal 7: 96–106.

Cresswell, T. (2004) Place: A Short Introduction. London: Blackwell.Fullilove, M. (2004) Root Shock. New York: Ballantine.Hillier, B. (1996) Space is the Machine. Cambridge: Cambridge University Press.Horan, T.A. (2000) Digital Places: Building Our City of Bits. Washington, DC: Urban Land

Institute.Kemmis, D. (1995) The Good City and the Good Life. New York: Houghton Mifflin.Malpas, J.E. (1999) Place and Experience: A Philosophical Topography. Cambridge:

Cambridge University Press.Massey, D. (1997) ‘A global sense of place’, in T. Barnes and D. Gregory (eds) Reading

Human Geography. London: Arnold, pp. 315–323.Million, M.L. (1992) ‘It Was Home’: A Phenomenology of Place and Involuntary

Displacement as Illustrated by the Forced Dislocation of Five Southern AlbertaFamilies in the Oldman River Dam Flood Area. Doctoral dissertation, SaybrookInstitute Graduate School and Research Center, San Francisco, California.

Mugerauer, R. (1994) Interpretations on Behalf of Place: Environmental Displacementsand Alternative Responses. Albany, NY: State University of New York Press.

Peet, R. (1998) Modern Geographic Thought. Oxford: Blackwell.Relph, E. (1976) Place and Placelessness. London: Pion.Relph, E. (1981) Rational Landscapes and Humanistic Geography. New York: Barnes &

Noble.Relph, E. (1993) ‘Modernity and the Reclamation of Place’, in D. Seamon, (ed.) Dwelling,

Seeing, and Designing: Toward a Phenomenological Ecology. Albany, NY: SUNYPress, pp. 25–40.

Relph, E. (1996) ‘Reflections on Place and Placelessness’, Environmental andArchitectural Phenomenology Newsletter 7 (3): 14–16 [special issue on the twentiethanniversary of the publication of Place and Placelessness; includes commentaries byMargaret Boschetti, Louise Million, Douglas Patterson, and David Seamon].

Relph, E. (2000) ‘Author’s Response: Place and Placelessness in a New Context [Classics inHuman Geography Revisited, Place and Placelessness]’, Progress in Human Geography24 (4): 613–619 [includes commentaries by John R. Gold and Mathis Stock].

Rose, G. (1993) Feminism and Geography. Cambridge: Polity.Seamon, D. (1979) A Geography of the Lifeworld: Movement, Rest, and Encounter. New

York: St. Martin’s.Seamon, D. (2000) ‘A way of seeing people and place: Phenomenology in environment-

behavior research’, in S. Wapner, J. Demick, T. Yamamoto, and H. Minami (eds)Theoretical Perspectives in Environment-Behavior Research. New York: Plenum,pp. 157–178.

Seamon, D. (2004) ‘Grasping the dynamism of urban place: Contributions from the workof Christopher Alexander, Bill Hillier, and Daniel Kemmis’, in T. Mels (ed.) ReanimatingPlaces. Burlington, VT: Ashgate, pp. 123–145.

Tuan, Y.-F. (1974) Topophilia: A Study of Environmental Perceptions, Attitudes, andValues. Englewood Cliffs, NJ: Prentice-Hall.


SPACE AND PLACE (1977): YI-FU TUAN

Tim Cresswell

Abstract knowledge about a place can beacquired in short order if one is diligent … Butthe ‘feel’ of a place takes longer to acquire.(Tuan, 1977: 183)

Introduction

I distinctly recall reading Yi-Fu Tuan’s Spaceand Place as an undergraduate at UniversityCollege London in 1985. I was taking a sec-ond year class called Humanistic Geography(changed a few years later to CulturalGeography – a sign of the times). It had beeneight years since it was published and it hadalready become a classic text – a worthyfollow-on to Tuan’s 1974 book, Topophilia. Iwas immediately captured by it. It was likenothing I had ever read in a geography class.It was full of ideas but engagingly written. Itwas not obviously geography (as I under-stood it at the time) but yet everything aboutit seemed central to how geography could oreven should be. It didn’t have long sectionson methodology or a review of recent litera-ture. There was nothing pedantic about it. Itjust seemed to jump straight in and get onwith thinking about some difficult questions.

I had no idea then how significant Space andPlace would become to human geographyand to me personally. By 1985 the idea ofhumanistic geography was a part of everyundergraduate degree. It is fair to say that ithad ceased to be revolutionary and hadbecome simply a part of the way significant

parts of human geography were conducted.In 1977, when Space and Place was published,Humanistic Geography had been on mostpeople’s intellectual horizons for only aroundfive years. The first book I ever bought for ageography course was Ley and Samuels’ editedcollection Humanistic Geography: Prospects andProblems, published in 1978. Edited collectionssuch as this rely on a number of authors whocan subscribe to the book’s central idea andthe willingness of a publisher to believe thereis a market for it. That book, therefore, markeda degree of acceptance that something calledHumanistic Geography had arrived. Topophiliahad been published in 1974 and in the sameyear Tuan had published a paper called Spaceand Place: Humanistic Perspective. The other keytexts in my 1985 course included work byEdward Relph (1976), Anne Buttimer andDavid Seamon (1980), David Ley (1974),Donald Meinig (1979) and J.B. Jackson (1980).Central to the work of all of these scholars was(and for many still is) the question of howpeople create a meaningful world and mean-ingful lives in the world. The notion of placewas central to this endeavour.

The decade before Space and Place was pub-lished had seen human geography becomesomething less than human. Spatial science, thequantitative revolution, and logical positivismhad all sought to treat the world, and the peo-ple in it, as objects rather than subjects. Theidea of rational people making rational choicesin a rational world had held sway (see Ableret al., 1971 and Haggett, 1965). Words like

7


54ÿÿTIM CRESSWELL

‘location’, ‘spatial patterns’, ‘distance’ and‘space’ had been central in the pages of booksand journals. Human geography had operatedmore or less as a pseudo-science. Humanisticgeography was, in part, a critique of this wayof thinking about the world and the humaninhabitation of it. The final pages of Space andPlace make this critique clear. Although thebook does not engage spatial science directlyit does frequently compare the richness ofan experiential perspective to more scientificand dehumanized approaches. ‘What we can-not say in an acceptable scientific language wetend to deny or forget. A geographer speaks asthough his knowledge of space and place werederived exclusively from books, maps, aerialphotographs, and structured field surveys.He [sic] writes as though people were endowedwith mind and vision but no other sense withwhich to apprehend the world and find mean-ing in it. He [sic] and the architect-planner tendto assume familiarity – the fact that we are ori-ented in space and home in place – ratherthen describe and try to understand what“being-in-the-world” is truly like’ (1977:200–201).And later, ‘The simple being, a con-venient postulate of science and deliberatepaper figure of propaganda, is only too easy forthe man [sic] in the street – that is, most of us –to accept’ (1977: 203). This ‘simple being’possibly refers to the abstract ‘rational man’ ofspatial science and economics: the man whoweighs up all options before making a rationalchoice about what to do next. Such a view ofhumanity has no place for meaning.

But Space and Place is not first and foremosta critique of spatial science. Its message is, forthe most part, a positive one which asks us asgeographers to be more aware of the ways inwhich we inhabit and experience the world –to increase, as Tuan puts it in the final lineof the book, our ‘burden of awareness’.Central to this awareness is the concept ofplace. The humanistic conception of placedescribes a way of relating to the world. Justas human geographers now are constantly

learning from theory and philosophy, oftenfrom continental Europe, so in the 1970sgeographers looked to theories and philoso-phies for inspiration. The ideas that becamemost central to the humanistic endeavourwere phenomenology and existentialism.These philosophies, put very simply, insistedthat people had the burden of making theirown meaning in the world through theirown actions. In addition, the productionof meaning arose from a process known as‘intentionality’. Intentionality described therelationship between human consciousnessand the objects that people were consciousof. It is impossible to be conscious, phenom-enology insisted, without being conscious ofsomething. It is in this ofness that meaningwas produced. Humanistic geographers, ingeneral, but particularly Tuan, did not spenda great deal of time repeating the ideas andwords of the philosophers who inspiredthem. Their task was not to explain thethoughts of others but to use these thoughtsin the production of new knowledge. Tuandoes, briefly, nod to the work of PaulRicoeur in the early pages of Space and Placein order to underline the importance of theword ‘experience’ to his book.

Experience is directed to the external world.Seeing and thinking clearly reach out beyondthe self. Feeling is more ambiguous. As PaulRicoeur put it, ‘Feeling is … without doubtintentional: it is a feeling of “something” – thelovable, the hateful [for instance]. But it is avery strange intentionality which on the onehand designates qualities felt on things, on per-sons, on the world, and on the other handmanifests and reveals the way in which the selfis inwardly affected’. In feeling ‘an intentionand an affection coincide in the same experi-ence’ (Tuan quoting Ricoeur, 1977: 9).

Ideas such as ‘experience’ and ‘feeling’were, by and large, not in the vocabulary ofhuman geographers in the early 1970s.Spatial scientists were not very interested inhow people related to the world through


YI-FU TUANÿÿ55

experience. Theirs was a world of simplepeople, meaning Tuan’s ideas were revolu-tionary stuff. While the spatial scientistswanted to understand the world and treatedpeople as part of that world (just like rocks, orcars or ice but with the magic ingredient ofrationality added), Tuan, in Space and Place,focuses on the relationship between peopleand the world through the realm of experi-ence. Tuan writes ‘[t]he given cannot beknown in itself. What can be known is a real-ity that is a construct of experience, acreation of feeling and thought’ (Tuan, 1977:9). The focus of Space and Place is thus howwe, as humans, are in-the-world – how werelate to our environment and make it intoplace.

The experience, as both thought and feeling,referred to in the book’s subtitle, is central tothe main argument of the book – the differen-tiation of space and place. What experiencedoes is transform a relatively abstract notion ofspace into a relatively lived and meaningfulnotion of place. While space was the favouredobject of the spatial scientist (and is still thefavoured object of social theorists) it is the wayspace becomes endowed with human meaningand is transformed into place that lies at theheart of humanistic geography. ‘What begins asundifferentiated space becomes place as we getto know it better and endow it with value …the ideas ‘space’ and ‘place’ require each otherfor definition. From the security and stability ofplace we are aware of the openness, freedomand threat of space, and vice versa. Furthermore,if we think of space as that which allowsmovement, then place is pause; each pause inmovement makes it possible for location to betransformed into place’ (Tuan, 1977: 6). This isthe most important contribution of Space andPlace to human geography – the distinctionbetween an abstract realm of space and anexperienced and felt world of place. Tuanforced geographers to stop taking these wordsfor granted and to consider what they mightmean for being human.

In addition to the space/place distinctionwhich lies at the heart of the book there aremany additional themes developed by Tuanthat have since become central to humangeography. Indeed, re-reading Space and Placerecently, I was struck by how Tuan foreshad-ows and informs much of the most excitingwork in the contemporary social sciences andhumanities. While the notions of space andplace Tuan developed have become common-sense (the ultimate compliment), notions ofexperience, the sensual, the emotional and theembodied are all the objects of much heateddebate in geography departments today.Non-representational theory, for instance, asan examination of the pre-conceptual, mirrorsTuan’s early musings on the centrality of expe-rience and emotion (see Thrift, 2004; Lorimer,2005). Back in 1977 Tuan was keen to reflecton the multiplicity of senses (not just sight) tounderstand our experience of the world in afuller way. ‘An object or place’, Tuan writes,‘achieves concrete reality when our experi-ence of it is total, that is, through all the sensesas well as with the active and reflective mind’(1977: 18). Chapter Four is devoted tothe ways in which space and place are under-stood in thoroughly embodied ways. Perhapsas a reaction to the thoroughly disembodiednotion of rational economic man in spatialscience, Tuan asserts that ‘Man and worlddenote complex ideas. At this point, we alsoneed to look at simpler ideas abstracted fromman and world, namely, body and space,remembering however that the one not onlyoccupies the other but commands and ordersit through intention. Body is “lived body” andspace is humanly constructed space’ (1977: 35).The importance of the lived body is particu-larly underlined in the chapter on ‘SpatialAbility, Knowledge and Place’ (Chapter Six),Here Tuan prefigures recent work on thehabitual and the pre-cognitive as he writes:‘There are many occasions on which weperform complex acts without the help ofmental or material plans. Human fingers are


56ÿÿTIM CRESSWELL

exceptionally dexterous. A professional typist’sfingers fly over the machine; all we see is blurof movement. Such speed and accuracy sug-gest that the typist knows the keyboard in thesense that he can envisage where all the lettersare. But he cannot; he has difficulty recallingthe positions of the letters that his fingersknow so well’ (1977: 68). Indeed, Tuan arguesthat one of the ways space is turned into placeis through kinaesthetic familiarity – the habit-ual ability to move through it unthinkingly. ToTuan, though, the focus on experience is aholistic one. Alongside the habitual and pre-cognitive lie the processes of cognition that areable to reflect on and share ways of being inplace. The importance of the visual is empha-sized in Chapter Twelve (Visibility: the Creationof Place) in which he argues: ‘Place can bedefined in a variety of ways. Among them is this:place is whatever stable object captures ourattention. As we look at a panoramic scene oureyes pause at points of interest. Each pause istime enough to create an image of place thatlooms large momentarily in our view’ (Tuan,1977: 161). Place, then, is a subtle mixtureof both fairly self-aware reflection and theworld of habitual action. We may know a placeprofoundly by participating in it in an unself-conscious way but we may also relate to placesthrough the construction of something visible.This is how monuments, for instance, tradi-tionally work. Place-making from the smallestcorner of a favourite room to the nation andthe globe happens both though practice andthrough a more symbolic register.

Tuan also pays considerable attention tothe problem of time in Space and Place. Againthis prefigures more recent assertions of theneed to consider time alongside space inhuman geography (May and Thrift, 2001).Both space and place, he argues, have impor-tant temporal dimensions. The awareness oftime and space comes together in the body.‘We have a sense of space because we canmove and of time because, as biologicalbeings, we undergo recurrent phases of

tension and ease’ (Tuan, 1977: 118). Temporalityis often described spatially. Thus we talk ofthings being forward or backward in time aswell as space. Here is now. There is then.There is half an hour away. Observationssuch as these open up a plethora of diverseobservations about how we humans inhabittime and space. Dancers, for instance, ‘moveforward, sideways, and even backward withease. Music and dance free people from thedemands of purposeful goal-directed life,allowing them to live briefly in what ErwinStraus calls “presentic” unoriented space’(1977, 128–129). Place too, Tuan tells us, hasto be understood temporally. Place can beunderstood as a pause in time as well as space.A true sense of place takes time to establish.‘If we see the world as process’ Tuan tells us‘we should not be able to develop any senseof place’ (1977: 179). Mobility, in Space andPlace, is part of the world of space and is anti-thetical to place and the kinds of attachmentplace necessitates. Finally, place can be a wayof making time visible. We get to know aversion of the past through the ways placesare made to represent memories. This caninclude everything from the family photoswe place on our mantelpiece to the statues ofheroes that appear in city squares.

The division between space and placeconstructed by Tuan became a taken-for-granted distinction in human geography bythe early 1980s. This distinction, more or lessin the form Tuan suggested, is still taught tostudents across the English-speaking world.Space and Place is still widely used throughoutthe world, frequently cited as the founda-tional text of notions of place and admiredacross many disciplines and beyond the acad-emic world entirely. My book, In Place/Out ofPlace (1996), would have been unthinkablewithout Tuan’s inspiration. In that book Itook the notion of place as a field of care andcentre of meaning and developed it througha consideration of the exclusions that suchattachments foster and what that might mean


for people, practices and things that weredeemed ‘out of place’. This was typical of theway many younger geographers, during the1990s, took the idea of place developed byTuan and began to insert it into a world ofcultural politics that Tuan mostly ignored.Benjamin Forest, for instance, revealed howplace became an important symbol for newforms of gay identity in West Hollywood(1995) and Karen Till has shown how placeis implicated in the construction and contes-tation of public memory in Berlin (2005).In each of these works Tuanian place wasboth adopted and adapted to consider issuesof contestation and power as well as themore humanistic notions of meaning andexperience.

Geographies of place, 30 years on fromTuan’s book, share his concern for meaning,belonging and experience but also includereflections on power, exclusion and socialdifference that are largely missing from Spaceand Place. To some this has gone so far thatthe original interest in individual experiencehas been subsumed by wider social issues. Anumber of books have built on the humanis-tic perspective advocated by Tuan (Entrikin,1991; Sack, 1997; Adams et al., 2001). Sack,for instance, has built on Tuan’s notion ofplace by considering the ways place as a cen-tre of human meaning ties together worldsthat are normally held apart – the worlds ofnature, meaning and society. In addition, hehas developed an ethical and moral frame-work for human action based on a Tuaniannotion of place (1997, 2003). Tom Mels’edited collection, Re-animating Place (2004)attempts to recapture some of the humanisticinsights of Tuan, along with Anne Buttimer,through an engagement with the rhythms ofplace as a grounded form of human experi-ence. More importantly perhaps many ofTuan’s ideas have travelled into geographieswhich are not explicitly humanistic andindeed to other disciplines entirely. Thenotion of place as a meaningful centre of

human experience, for instance, has beendeveloped by philosophers such as Casey(1998) and Malpas (1999).

Geographers and others, with backgroundsin Marxism, feminism and post-structuralism,have all developed critiques of Tuan’s notion ofplace in particular (Rose, 1993; Massey, 1997;Harvey, 1993). Place, they argue, is far toovague a notion. It ignores power relations andforces of exclusion that work through a con-tested association of location, meaning andpractice. Perhaps the most well-known critiqueof humanistic conceptions of place in generalis Massey’s development of the idea of a ‘pro-gressive sense of place’ (also called a ‘global’ or‘extrovert’ sense of place). Massey takes broadlyhumanistic discussions of place to task forinsisting on a bounded sense of identityattached to place that is rooted in history inmore or less linear ways (Massey, 1997). Places,she argues, are rather the product of multiplemobilities intersecting. These are open tochange and associated with multiple, ratherthan singular, identities and histories. Places, assuch, do not have clearly defined boundaries.Such a view is very different from Tuan’s asser-tion that a world of process is antithetical to theconstruction of any kind of sense of place.Others prefer to use the notion of space in amore subtle way, suggesting a variety of figura-tive and literal spaces including social space,third space and lived space. Many have beeninspired by the French urban theorist HenriLefebvre, whose book The Production of Space(1991) introduced a notion of space whichis socially produced, meaningful and lived.Abstract space, the kind of space discussed byTuan, becomes just one version of space inLefebvre. Lefebvre writes of ‘representations ofspace’ as the conceptual space of planners,politicians and social engineers. This closelymirrors Tuan’s outline of space as a moreabstract realm. But Lefebvre also writes of ‘rep-resentational spaces’ – ‘space as directly livedthrough its associated images and symbols’(Lefebvre, 1991: 39). This is inhabited space

YI-FU TUANÿÿ57


58ÿÿTIM CRESSWELL

and, as such, it is very close to Tuan’s definitionof place. Given the multiple definition of spacein Lefebvre it is no surprise that place does notreally feature in the work of Soja who, likeLefebvre, develops notions of space as bothmeaningful and practised, thus obviating theneed for discussions of place (Soja, 1996).

Conclusion

As a way of doing geography Space and Place,like much of Tuan’s work, is hard to pin down.One reason for this is that Tuan does notundertake ‘research’ in the usual sense of theword. He does not use any ‘method’ that islikely to appear in a methodology class. Thereis no evidence of ethnography, semiotics, sta-tistical analysis or even in-depth archival work.The impression I get from reading Tuan’swork is of someone who reads a lot of what isoften called secondary material and makesconnections between them that would other-wise have remained unnoticed. Many of Tuan’sobservations are assertions of common-sense.

Often they are supported by a range ofanthropological literature, classic sociologytexts, poetry, novels and snippets of psychology.It is almost as if Tuan hires an army of creativeacademics, writers and philosophers to do theresearch for him. The examples Tuan uses areas diverse as his sources but rarely includereflections on modern culture – particularlypopular culture. We are more likely to encounterindigenous groups of Africa seen through theeyes of anthropologists or the plans of anancient Chinese city than anything recogniz-ably here and now. Tuan writes in a veryaccessible reflective way. It is, in many ways, amodel of writing for a large educated audi-ence that stretches far beyond the confines ofa particular set of theoretical interests or,indeed, a particular discipline. It was this kindof writing that so engaged me back in 1985. Ayear later I was thinking about doing a PhDand was advised to contact the geographerswhose work I most admired. I wrote to Yi-Fuand received a charming letter back. I went toMadison, Wisconsin to start my academiccareer.

Secondary sources and referencesAbler, R., Adams, J. and Gould, P. (1971) Spatial Organization: The Geographer’s View

of the World. London: Prentice-Hall.Adams, P., Hoelscher, S. and Till, K. (eds) (2001) Textures of Place: Exploring Humanist

Geographies. Minneapolis: University of Minnesota Press.Buttimer, A. and Seamon, D. (1980) The Human Experience of Space and Place.

New York: St Martin’s Press.Casey, E. (1998) The Fate of Place: A Philosophical History. Berkeley, CA: University of

California Press.Cresswell, T. (1996) In Place/Out of Place: Geography, Ideology and Transgression.

Minneapolis: University of Minnesota Press.Entrikin, J.N. (1991) The Betweenness of Place: Towards a Geography of Modernity.

Baltimore, MD: Johns Hopkins University Press.Forest, B. (1995) ‘West Hollywood as Symbol: The Significance of Space in the Construction

of a Gay Identity’, Environment and Planning D: Society and Space 13: 133–157.Haggett, P. (1965) Locational Analysis in Human Geography. London: Edward Arnold.Harvey, D. (1993) ‘From space to place and back again’, in J. Bird, B. Curtis, T. Putnam,

G. Robertson and L. Tickner (eds) Mapping the Futures. London: Routledge, pp. 3–29.


YI-FU TUANÿÿ59

Jackson, J.B. (1980) The Necessity for Ruins. Amherst: University of MassachusettsPress.

Lefebvre, H. (1991) The Production of Space. Oxford: Blackwell.Ley, D. (1974) The Black Inner City as Frontier Outpost. Washington, DC: AAG.Ley, D. and Samuels, M. (eds) (1978) Humanistic Geography: Prospects and Problems.

London: Croom Helm.Lorimer, H. (2005) ‘Cultural geography: the busyness of being more-than-

representational’, Progress in Human Geography 29 (1): 83–94.Malpas, J.E. (1999) Place and Experience: A Philosophical Topography. Cambridge:

Cambridge University Press.May, J. and Thrift, N. (eds) (2001) Timespace: Geographies of Temporality. London:

Routledge.Massey, D. (1997) ‘A global sense of place’, in T. Barnes and D.G. Gregory (eds) Reading

Human Geography. London: Arnold, pp. 315–323.Meinig, D. (ed.) (1979) The Interpretation of Ordinary Landscapes. Oxford: Oxford

University Press.Mels, T. (2004) Re-Animating Place: A Geography of Rhythms. London: Ashgate.Relph, E. (1976) Place and Placelessness. London: Pion.Rose, G. (1993) Feminism and Geography: The Limits of Geographical Knowledge.

Cambridge: Polity.Thrift, N. (2004) ‘Summoning life’, in P. Cloke, P. Crang and M. Goodwin (eds)

Envisioning Human Geographies. London: Arnold, pp. 81–103.Sack, R. (1997) Homo Geographicus. Baltimore, MD: Johns Hopkins University Press.Sack, R. (2003) A Geographical Guide to the Real and the Good. New York: Routledge.Soja, E. (1996) Thirdspace: Expanding the Geographical Imagination. Oxford: Blackwell.Till, K. (2005) The New Berlin: Memory, Politics, Place. Minneapolis: University of

Minnesota Press.Tuan, Y.-F. (1974) ‘Space and Place: Humanistic Perspective’, Progress in Geography 6:

211–252.Tuan, Y.-F. (1974) Topophilia: A Study of Environmental Perceptions, Attitudes and

Values. Englewood, NJ: Prentice-Hall.Tuan, Y.-F. (1977) Space and Place: The Perspective of Experience. Minneapolis:

University of Minnesota Press.


THE LIMITS TO CAPITAL (1982): DAVID HARVEY

Noel Castree

The aim is … to create frameworks of under-standing, an elaborated conceptual apparatus,with which to grasp the most significantrelationships at work within the intricatedynamics of social transformation. (Harvey,1982: 450–451)

Introduction

The Limits to Capital was first published in1982. It is a book with epic ambitions. It aimsto explain how a raft of geographical phe-nomena – such as city-regions, nation statesand transportation networks within andbetween them – are integral to the function-ing of the world’s dominant economic system(capitalism). Harvey is arguably the mostfamous contemporary geographer, known forhis politically engaged scholarship beyond, asmuch as within, the field of professional geog-raphy. Author of over a dozen major works(including The Condition of Postmodernity: seeWoodward and Jones, Chapter 15, this volume),The Limits to Capital is Harvey’s ‘favourite text’(2001: 10).

The book, which was nearly a decade inthe making, was written for two audiences.As the leading Marxist geographer of hisday, Harvey hoped that The Limits to Capitalwould influence research and teaching amongwhat, at the time, was a small group of like-minded geographers (such as his formerstudents Neil Smith and Richard Walker). Asa geographical Marxist, Harvey also wrote

Limits to Capital in the hope of persuading afairly large community of academic Marxistsoutside geography to take geographical ques-tions more seriously than they had previously.In short, The Limits to Capital was intended tobe a paradigmatic contribution: nothing lessthan a modern geographical equivalent ofKarl Marx’s magisterial work Capital (uponwhich Harvey’s book is in large measurebased). The one important contrast withMarx’s own writing is that Harvey’s book waslargely an academic contribution, a significantfact to which I will return later.

I will structure my account of The Limitsto Capital as follows. In the next section I willset the book in its original context andexplain the significance of its status as alargely academic work. I will then attempt tosummarize the book’s arguments for thebenefit of student readers who, with fewexceptions, would not get far if they tried toread The Limits of Capital unaided. In thepenultimate section I will consider the book’simpact within and beyond professional geog-raphy between 1982 and the date of its firstreissue (1999). My brief concluding sectionrelates The Limits to Capital to the immediatecircumstances of our time and likely futurescenarios for humanity worldwide.

Marxism, geography, capitalism

‘Internalist’ approaches to intellectual changein academic disciplines focus on debates

8


within those disciplines. Such approachespresume that academia is relatively isolatedfrom the wider society, and that intellectualprogress emerges from the to-and-fro of crit-icism internal to the academic community.By contrast, ‘externalist’ approaches situatechanging academic research in a wider soci-etal context – a context seen to exert amore-or-less direct influence on what acade-mics choose to do. The Limits to Capital mustbe understood in both its disciplinary andwider societal contexts, as I will now explain,starting with the former.

When he wrote The Limits to Capital,Harvey was a Professor of Geography at theJohns Hopkins University, in the east coast UScity of Baltimore. Prior to this he had been aLecturer in Geography at Bristol University,England. In Bristol, Harvey was exposed to theso-called ‘new Geography’ being pioneered bythe likes of Richard Chorley (at CambridgeUniversity) and Peter Haggett (a BristolUniversity professor). In their view, Geographyshould become a ‘spatial science’: one whosespecial role within the family of sciences was torigorously describe and explain the location ofphenomena on the earth’s surface. Inspired bythe ‘new Geography’, Harvey wrote the defin-itive geography book on scientific method(Explanation in Geography, 1969; see Johnston,Chapter 4, this volume). This book gave furtherimpetus to spatial scientists’ preoccupationwith hypothesis testing, the search for spatialregularities and laws, the development of gen-eral models and theories, and the use ofstatistical techniques.

However, by 1969 a decade of scientificgeography was already proving too much forHarvey. The reason was a gap between whatwas going on inside the discipline of geogra-phy and what was going on in the widerworld. The disciplinary and societal contextscollided, and Harvey wanted to realign themby transforming professional geography fromwithin. The late 60s and early 70s were tur-bulent years. The post-war economic boom

in the West came to an end with crippling oilprice hikes; the US civil rights movementwas in full flow; the modern environmentalmovement came into existence; the ‘imperi-alist’ war in Vietnam rumbled on until theUS was obliged to withdraw its troops; dissi-dents struggled against French and Britishcolonial powers in Africa and beyond; therewere the famous student riots and strikes inParis in May 1968; feminism made wavesthroughout the so-called ‘developed’ world;and trade unions were at loggerheads withnational governments across the Westernworld. The solidities of the post-1945 eraseemed suddenly to be melting into air. AsHarvey (2001: 5–6) noted retrospectively, ‘Iturned in my magnum opus to the publishersin May 1968, only to find myself acutelyembarrassed by the change in political tem-perature … I realised I had to rethink a lot ofthings I had taken for granted …’.

In Social Justice and the City and a stringof now classic essays, Harvey (1973) becameknown as an uncompromising critic of spatialscience and a pioneer of Marxist geography.He called for nothing less than a ‘revolutionin geographic thought’ in which a Marxianresearch agenda would replace the spatialscience paradigm. In Harvey’s view, this para-digm was profoundly problematic. In essence,its claims to ‘objectivity’ and ‘neutrality’ con-cealed its complicity with a profoundly unjustsocial order that operated in the interests ofpower elites nationally and globally. ‘Scientific’studies by geographers of migration patternsin space and time, of traffic congestion, or ofhow to manage water resources dealt withsymptoms not causes in Harvey’s estimation.They remained blind to the true nature andorigins of myriad economic, social and envi-ronmental problems. What is more, for Harveyspatial science wrongly presumed that acade-mic research could and should be impartialwhen, in fact, it was nothing more than anon-neutral part of the reality it sought to com-prehend. In Harvey’s view, university research

62ÿÿNOEL CASTREE


and teaching did not merely ‘re-present’ truthsexisting ‘out there’ in the wider world. Instead,they helped to constitute that wider world inso-far as academic ‘expertise’ was used to shapestudent, governmental, business and publicbehaviour.

Marx was a famous – for some notorious –nineteenth-century philosopher, politicaleconomist, journalist, pamphleteer and labourmovement leader. Dismayed by the appallingconditions in which growing numbers ofworking-class people were forced to live andwork, he sought to understand how and whyso few people get rich while so many remainpoor. In his later writings, such as Capital vol-ume one, Marx argued that capitalism was amajor cause of social inequality (and environ-mental degradation) in the Victorian world.The appeal of Marx’s work to Harvey wasarguably two-fold. First, Marx’s focus on capi-talism offered a way to talk about deep causesrather than the surface problems – the kind ofproblems that so many spatial scientists studiedand sought to ameliorate through public pol-icy prescriptions. Secondly, Marx’s later worktook an ‘activist’ view of knowledge in whichto study the world is to change the world, giventhe right conditions. Marx saw ‘knowledge’ asthe polite world for competing ideologies thatserve specific societal interests. Marxism wasintended to be the ideology of the workingclass: a set of explanatory tools that can helpthis class realize its own true interests by creat-ing a more egalitarian mode of production (asocialist one).

In Social Justice and subsequent essayspublished in the 1970s, Harvey did indeedprecipitate something like the revolution ingeographic thought he wished for. His workespecially appealed to a younger generationof geographers who had been politicized bythe turbulence of the post-1968 period. Butironically, Harvey’s radicalization of geogra-phy (and more particularly the human side,since physical geographers carried on muchas before) did not have any wider societal

impacts. In part this was because Marxistgeography fast became an academic move-ment, one that conformed to the professionalstandards of Western academia. Even Harvey,socialized as he was into the norms of eliteuniversities, tended to focus his writing activ-ities on high-level, difficult works like TheLimits to Capital – works unlikely to attract awide readership. Secondly, even if the earlyMarxist geographers like Harvey had con-nected their work more vigorously to workers’movements, the prospects for achieving amore socialist society had diminished by thelate 70s.

The optimism of the late 60s gave way, inless than 10 years, to despair within andbeyond the West. More authoritarian govern-ments (e.g. those of Margaret Thatcher)would not tolerate worker demands for bet-ter pay and conditions; meanwhile, globaleconomic restructuring led to job losses inmany highly unionized countries and jobgains in areas with little or no history oflabour organizing. In short, by the timeHarvey handed over the manuscript of TheLimits to Capital to his publisher, academicgeography had been radicalized in a left-wing direction but the wider world seemedmore conservative than it had been whenHarvey made his initial turn to Marxism adecade earlier.

The Limits to Capital explained

The Limits to Capital is a text about thehorrors and glories of the capitalist world inwhich we live. Given how insinuated intomost aspects of people’s lives capitalism nowis, Harvey’s book has a universality of rele-vance that most books by academics cannotequal. To appreciate the distinctiveness of thetext we need to start with some commentsabout theory, about the kind of ‘geography’that has long interested David Harvey, andabout something called ‘dialectics’.

DAVID HARVEYÿÿ63


The Limits to Capital is a work of theory. Insimple terms, to theorize is to engage in aprocess of abstract, rather than concrete,argumentation. To ‘abstract’ means to ‘takefrom’: abstraction thus involves pulling some-thing out of its context of operation in orderto understand how it works. Theorists thusdo mentally what clinical researchers do sur-gically. They dissect a whole (e.g. a socialsystem) in order to inspect the parts so thatthe particular function of these parts andtheir inter-relationships can be understood.Generally speaking, and to use cartographicimagery, ‘theorists’ therefore try to representsocial (and/or biophysical) reality in terms ofa conceptual map that highlights the keytopographical features, their relative position-ing and their relationships.

So much for theory. What about geogra-phy? There are three things to say here. Thefirst is that Marx and most of his intellectualsuccessors did not pay much attention to geo-graphical issues. This meant that most Marxistsunderstood capitalism in non-geographical or‘aspatial’ ways: as if it existed on the head of apin without the need for material landscapesof production, transportation, consumptionand social reproduction. Secondly, this inatten-tion to geography was linked to a falseequation between the sorts of things that geo-graphers study and the realm of ‘facts’. Fordecades, non-geographers (and even a fewgeographers!) have assumed that geographycannot be a theoretical discipline because thesorts of things it studies are ‘contingent’. Acontingent occurrence – like the constructionof a new airport or the incidence of famine –will almost always have causes that can berepresented theoretically. For instance, a rela-tionship of gender inequality between menand women can be described and explainedconceptually (as an outcome of ‘patriarchal’social relations). But this relationship willmanifest itself in, and produce, differentcontingent forms – for instance, particularheterosexual arrangements in the home,

specific workplace dynamics and so on.Accordingly, the contingent forms are con-nected to but do not in themselves constitutethe ‘underlying’ processes generating them.Contingencies must be studied empirically,and because they often vary in time and spacegeographers are chief among those whoinvestigate them.

Harvey dissents from this view. Since SocialJustice he has argued that geographical phe-nomena can be theorized. This is the same assaying that these phenomena have a constitu-tive role to play in the fundamental processesthat give rise to them in the first place. Here,then, there is no distinction between ‘process’and ‘outcome’ because the latter makes theformer flesh and, once it exists, may affectthe subsequent operations of the process inquestion. This brings me to a third and finalobservation about the kind of ‘geography’that interests Harvey. Going back to his doc-toral thesis on the Kent hop industry, Harveyhad long been fascinated with geographicaldifference: with the rich specificities of peopleand place. However, as Harvey shows in TheLimits to Capital and elsewhere, we can makesome general observations about these other-wise unique geographies. More specifically,Harvey argues that there are some ‘signaturegeographies’ – or characteristic patterns –associated with capitalism’s past, present and(so long as it exists) future. These include thegrowth and decline of city-regions housedwithin nation states and the construction ofelaborate transportation networks to moreclosely connect them.

Lastly, a word about dialectics. Dialecticsis both an analytical tool (an epistemologyand method) and a broad description of howsocial and biophysical systems work (anontology). In both cases contradiction is a keyidea. In the ontological sense, a dialecticalapproach says that some social and biophysi-cal systems contain opposing tendencies, thecollision of which leads to change in (andeven the dissolution of) those systems. In the

64ÿÿNOEL CASTREE


conceptual and methodological sense, adialectical approach obliges the theorist toapply their mental scalpel in a specific way.Faced with a large complex system (like cap-italism), the theorist makes a series of mental‘cuts’ that reveal one set of processes; subse-quent cuts reveal other processes that canthen be shown to both arise from andyet contradict those identified earlier in theanalysis. This means that when one readsthe work of a dialectical thinker like Harveythe beginning of the analysis only reallymakes sense at the end of the book or essayin question.

Marx’s original argument

Drawing upon Capital and other late worksby Marx, Harvey theorizes capitalism as aprocess rather than a thing. This is entirelyappropriate because the capitalist mode ofproduction is really about circulation. As Marxargued, capitalism can be described as shownin Figure 8.1.

Here owners of the means of production(capitalists) advance a sum of money to pur-chase three things: (i) labour power (LP:people who must sell their skills and time fora wage in order to live), (ii) inputs (e.g. parts,raw materials), and (iii) infrastructure (e.g.premises, machines) [ii and iii = MP]. Thesethings are then deployed in the production(P) of new commodities (C*) which are soldin order to recoup the original sum ofmoney advanced plus an increment.

In essence, capitalism has one over-ridinglogic: profit. It is thus a system in motionbecause capitalists are constantly aiming tomake returns on their investments duringeach round of commodity production. WhatMarx called ‘capital’ was precisely this processof ceaseless accumulation which assumes dif-ferent physical forms between the beginningand end of each production cycle. As part ofthis process, firms are compelled (unless they

enjoy monopolies) to innovate technically andorganizationally. This is because they mustcompete for market share with rival firms intheir sector. Clearly, this has implications forwaged workers because their employers willseek to alter salaries and benefits, hours ofwork, number of employees, and much morebesides. In sum, capitalists as a class engage ina constant struggle with workers to extractprofits because, for Marx, workers have thespecial quality of producing more economicvalue than they themselves are normallypaid to do their jobs. This struggle produceswinners and losers at any one time. Someemployees fare well and others do not;some firms go under for lack of ‘competi-tiveness’, while others prosper by laying-offworkers and substituting them with ‘smartmachinery’.

If, then, accumulation, inter-firm competi-tion, innovation and class struggle are allinterconnected elements of capitalism it is inessence a contradictory mode of production.This makes it a crisis-prone system preciselybecause its central features are profoundlyinconsistent. For instance, as wage workerscapitalists see people as a cost of productionto be minimized and made more efficient.But as consumers capitalists see people asspenders of money without whom they can-not realize profits. Here, then, is one ofseveral contradictions: in this case betweenthe points of production and the points ofcommodity sale, because most people fulfil

DAVID HARVEYÿÿ65

Figure 8.1 The primary circuit of capitalaccumulation

MP

LP

P...C*...M...+DM...C


a double function in capitalist societiesas earners and spenders of money. Marxand his acolytes hoped that when ‘over-accumulation crises’ occurred – as Marxthought they inevitably would from time-to-time – the working class as a whole wouldrebel against the capitalist system. In Westerncountries this has never come to pass, thoughit did occur in Russia in 1917 led by Vladimir Lenin (founder of the now defunctSoviet Union). This raises the importantquestion of how capitalism’s contradictionsare managed, by whom, and with whatspecific consequences.

Harvey’s contribution

The title of The Limits to Capital says muchabout the book’s contents, as one mightexpect. The ‘limits’ referred to by Harvey aretwo-fold. The first are those of Marx’s origi-nal arguments, especially their inattention togeographical issues. The second are the limitsof capitalism as an economic system. TheLimits to Capital seeks both to extend Marx’stheory of how capitalism works and, in sodoing, to more comprehensively identifycapitalism’s internal contradictions. As apurely theoretical book, The Limits to Capitalis intended to specify the ‘basic laws ofmotion’ of capitalism past, present and future.In other words, it deliberately does not offera close empirical demonstration of how cap-italism works ‘on the ground’ in real timesand places.

The first seven chapters rehearse the argu-ments sketched in the previous sub-section.Harvey recapitulates Marx’s propositions insome conceptual detail, leading to what hecalls ‘the first cut theory of crisis’. However, itis the six chapters of the second half of TheLimits that capture the book’s true originality.As with the early chapters, crisis is the keyorganizing them. The crisis tendencies Marxidentified clearly pose a major problem for

the capitalist mode of production. Potentially,they can precipitate social revolutions far andwide. Even when they do not, these tenden-cies can pitch whole societies into chaos –with even the state badly affected as it triesbravely to manage the problems for bothcapitalists and workers alike (remember thatthe state, as an institution, depends upon rev-enues generated within a healthy capitalisteconomy [in the form of taxes, for example]).The question then arises: how can crisesof over-accumulation be avoided? Harveyaddresses this question, in the first instance,with reference to the financial system.

Finance capital is, in basic terms, credit: it ismoney lent with a view to receiving interestfrom the lender. Though lending institutionsexisted long before capitalism ever did, theyhave a pivotal role to play in capitalist societies.This is so in two senses. First, a portion of thesurplus generated in the ‘primary circuit ofcapital’ (Figure 8.1, above) can go into short-or long-term savings by firms (and workerstoo save money, of course). Banks, buildingsocieties and other savings institutions can thenlend this money to new firms or to statebodies in order to finance particular invest-ments. Secondly, a portion of the surplusgenerated in the primary circuit can bediverted directly to investment. For instance,successful firms in the US can buy stocks-and-shares in new business ventures in ‘emergingeconomies’ like Bulgaria. In sum, financialinstitutions oil the wheels of capitalist com-merce because they divert large amounts ofcapital from the primary circuit into futureopportunities for commodity production,transportation and consumption. How, then,does this relate to the possible avoidance ofeconomic crisis? On the one side, the financialcircuit of capital provides what Harvey calls a‘temporal fix’. On the other side, it is central towhat Harvey calls a ‘spatial fix’. Let me takeeach in turn even though, in reality, they aretwo sides of the same coin. This requires a briefdiscussion of the role of the built environment.

66ÿÿNOEL CASTREE


For the primary circuit of capital to functionsuccessfully a portion of it must be composedof four kinds of built environments: namely,those of production (e.g. factories), consump-tion (e.g. shopping malls), reproduction (i.e.homes) and distribution/communication (e.g.airports, rail systems). These built environmentsare, if you like, the arteries that make theprocess of capital accumulation possible. Thereare two unique things about them. First, theyare extremely expensive to construct (forinstance, a new international airport can easilycost over one billion US dollars). Secondly,these costs mean that few lenders acting alonecan afford to finance them. This is where thefinancial system as a whole comes to the res-cue. By pooling the revenues of countless firmsand investors, financial institutions (and, forthat matter, many national governments) can‘switch’ large amounts of capital out of imme-diate production to pay for large-scale andlong-lasting infrastructural investments. Giventhat the financial returns for investors arepotentially uncertain and very long-term, suchinstitutions use fictitious capital to incentivizethese investors. As Harvey reminds readers inThe Limits to Capital, fictitious capital is ineffect a set of promissory notes that makeclaims on a future share of profits from the pri-mary circuit.

What all this means, as Harvey showsin Chapters Eight, Nine and Ten of TheLimits, is that the financial circuit of capital isat once autonomous from and yet tethered tothe primary circuit. In facilitating capitalswitching out of immediate commodity pro-duction, financial institutions can enable themuch-needed production of the four types ofbuilt environment listed above. In what waysdoes this switching constitute a ‘temporal fix’for capitalism’s crisis tendencies? Harveyargues that tying-up surplus capital (andlabour) in the construction, use and mainte-nance of new built environments is a way ofdeferring crisis. It postpones for many years(potentially at least) the proverbial chickens

coming home to roost. This is because thingslike road networks and factories only realizeprofits (if they do at all) in the medium-to-long term, but rarely ever in the short-term.However, none of this eliminates crises ten-dencies in the capitalist system. Harvey’s‘second cut’ crisis theory suggests that,eventually, new investments made via thefinancial system will fail to produce thewealth necessary to pay interest on thoseinvestments. At that point, a long-deferredcrisis of more-or-less general proportionscould engulf the capitalist world.

Let us now turn to the ‘spatial fix’, whichis central to the closing chapters of The Limitsto Capital. For commodity production tooccur, it is obviously necessary in most casesthat workers and employers be proximate. Itis equally necessary that both parties havebuilt environments of the four kinds men-tioned in order to undertake necessaryactivities locally and to connect with thewider capitalist world. Over time, a particulareconomic activity, or ensemble of activities,comes to characterize a particular town, cityor city-region and its inhabitants. Theseplaces, Harvey argues, have a ‘structuredcoherence’ or internal integrity. Cross-classalliances can develop to protect local interestsand investments against competing places inthe global economy. However, according toHarvey there is a problem. Given capitalism’schronic crisis tendencies, it frequently becomesfunctional to switch large amounts of capitalout of ‘developed’ towns and city-regions andinto less developed ones or ones that havefallen on hard times. This switching can involveinvesting in new built environments abroad inthe hope that they support new productiveactivities that will, in time, pay interest on theoriginal investments. This constitutes a ‘spatialfix’ for the obvious reason that geographicalexpansion is here being used to find an outletfor excess capital.

Harvey’s ultimate conclusion is that spatialfixes (like temporal ones) can never prevent

DAVID HARVEYÿÿ67


capitalist economic crises, only defer theirultimate occurrence. Indeed, he argues thatthey simply widen the landscape across whichcapitalism’s inner contradictions operate. It isprecisely in allowing these growth poles toemerge that formerly profitable places sud-denly face geographical competition – whichcan produce local, regional and national crisesas former ‘winners’ become present-day‘losers’. In short, uneven geographical developmentis an intrinsic, non-accidental part of capitalistlife. Given this contradictory logic, Harveyends The Limits to Capital on an apocalypticnote – his ‘third cut’ theory of crisis. Eventually,he argues, spatial and temporal fixes togetherwill no longer provide a safety valve for thesystem as a whole. At that point, there is likelyto be a mad scramble (as he argued happenedin the late 1930s) among capitalist countries toavoid suffering the worst consequences of eco-nomic crisis. (For a summary account of TheLimits to Capital see Harvey, 1985.)

Evaluating The Limits to Capital: fromthe first to the second edition

Needless to say, the previous section barelydoes justice to the arguments of The Limits toCapital. How then does one evaluate a booklike The Limits? The answer is by no meansobvious. The Limits to Capital is widelyknown within and beyond geography butnot, in my view, widely read, despite a reissuein 1999. Even though the book is highlycited by left-wing social scientists, this doesnot mean they have read the book in detail,let alone understood its intricate arguments.Does this mean that the book should beregarded as ‘a failure’ – undeserving of achapter-length treatment in a book on keytexts? My own view is that The Limits is atowering achievement. The sheer difficulty ofunderstanding the book is not, I submit, anegative sign of its quality or its importance.I say this for the following reasons.

First, one needs to look at who preciselyhas read, understood and been influenced byThe Limits of Capital, rather than worry aboutwho has not. As I noted in the introduction,Harvey wanted The Limits to Capital to be afoundational text for both Marxist geographyand geographical Marxism. In that he hassurely succeeded. In the years between thebook’s original publication and its reissue,Marxist geography became not only thedominant left-wing approach in Anglophonehuman geography, it also helped make left-wing geography as such a central elementof Anglophone geography tout court. This wasno mere coincidence. The Limits to Capitalspecified a new research agenda of such pro-fundity that it attracted some of the bestminds in human geography throughout the1980s. Without it, key Marxist books like TheGolden Age Illusion by David Rigby andMichael Webber (1996) could not have beenwritten. Likewise, without the credibility thatThe Limits to Capital lent Marxist geography atlarge, people like Neil Smith, Don Mitchell,Kevin Cox and Richard Walker could not havegone on to assume important positions inworld-leading geography departments. Outsidegeography, albeit belatedly, key Marxists likeFredric Jameson and the sociologist BobJessop began to read Harvey’s magnum opusclosely, once the best-selling The Condition ofPostmodernity had alerted them to his prodi-gious talents. That a special issue of theleft-wing academic journal Antipode was pub-lished to mark The Limits to Capital’s twentiethanniversary says much about the book’s impor-tance to Marxists within and outside academicgeography (Brenner et al., 2004).

Secondly, a book’s significance can also bemeasured by who its critics are as much as itsfans. Such is the range of topics covered inHarvey’s book and such is the depth of treat-ment each one receives, that it has become atouchstone for leading human geographersof a post- or non-Marxist persuasion. A goodexample of this is Money/Space by the British

68ÿÿNOEL CASTREE


geographers Andrew Leyshon and NigelThrift (1997). This book was a formativecontribution to the geographical study ofmoney in the modern world. It lays out abroad research agenda for the interrogationof where, why and in what specific formsmoney moves, rests and moves again. Eventhough the book’s authors were not Marxists,their text took Harvey’s arguments in TheLimits to Capital as a key point of departure(see Leyshon, 2004). For them, Harvey’sunderstanding of money missed, amongother things, its cultural importance byemphasizing its economic functions. This isone of several topic areas where human geo-graphers have developed new lines of inquirythrough a critical engagement with Harvey’sstrongly Marxist position.

These two positive points having beenmade, one still cannot escape the fact thatoutside academia Harvey’s book is almosttotally unknown. This would seem to be anindictment of Harvey’s own professed desireto change the world by altering people’sunderstanding of their daily lives. However, inHarvey’s defence, The Limits was not intendedto be read by a wider, non-academicaudience. Why, then, did he write the book inthe way he did? A cynical response to thisquestion might suggest that Harvey was slav-ishly playing the academic game. In otherwords, Marxist or not, he was simply follow-ing the well-worn path of previoussuccessful academics: write a ‘major book’ inorder to gain academic credibility. However,this is surely too cynical a view. After all,Harvey’s academic reputation was alreadysecure by the late 1970s, at least withingeography. This suggests an alternative expla-nation for the book’s density. Quite simply,capitalism itself is so complicated that only abook of equal complexity could do justice toits various (il)logics. Having written thebook, it was then up to Harvey himself andother Marxist geographers to translate itsinsights into more intelligible forms for

workers’ groups, trade unions and the like.That such acts of translation never occurredon a large scale is no indictment ofThe Limitsto Capital. Instead, it says much about the pre-vailing political mood in the West through the1980s and 90s. Outside academia Marxismbecame something of a dirty word duringthese years.

Conclusion

It may seem strange that The Limits to Capitalwas reissued in 1999. In the West, and moregenerally, the last 25 years have been oneswhen conservative and neo-liberal agendashave dominated politics, business and publicdebate. The ‘death of communism’ in the late1980s (when the former USSR and theEastern Bloc dissolved) seemed to indicatethe ‘superiority’ of capitalism over socialism.In this context, Marxist theory was not only‘off-message’ in the wider society; morepointedly, it seemed to have had its argu-ments confounded by capitalism’s remarkableability to prevent economic crises becominganti-capitalist revolutions. Additionally, fromthe early 1990s Marxism lost its previousacademic dominance within and outsideAnglophone human geography. A new gener-ation of academics chose to make their nameson the backs of non-Marxist theorists likeJacques Derrida, Michel Foucault, BrunoLatour, Paul Virilio and many others besides.By 1999, Harvey was seen by many youngerleft-wing academics as a bit of a ‘dinosaur’ – anold-fashioned ‘meta-theorist’ who was inatten-tive to the complexity and contingency ofsocial life. Given these facts, reprinting TheLimits to Capital may have seemed an act of folly.

I say seemed deliberately. When leading left-wing publisher Verso chose to introduce TheLimits to Capital to a new generation of acade-mics and activists it did so for good reason.Since the late 1990s, something of the revolu-tionary fervour of the late 60s/early 70s has

DAVID HARVEYÿÿ69


reappeared in public life worldwide. The often-violent protests against the World TradeOrganization, the Make Poverty History cam-paigns, the armed struggle of the Zapatistas insouthern Mexico are three of many possibleexamples of present-day discontent with themanifest injustices of life in a capitalist world.In this new context the book arguably has agreater relevance than at any time since it waspublished. Nonetheless, it will never be aneasy book to read and digest. If Marxists likeHarvey are really to change the world for thebetter, there needs to be more engagement

of the public and activists than heretofore(see Castree, 2006). This will be no easy taskgiven how closeted in the ‘ivory tower’ mostpresent-day academics are – even politicalacademics like Harvey himself. Meanwhile,within academia itself, there are precious fewMarxists of my own generation who can con-tinue the tradition of spatialized Marxism thatHarvey helped to inaugurate. In the years ahead,we may well confront the paradox that Marxismhas few expositors left in the university world,while a large audience receptive to Marxist ideasexists out there in the wider world.

70ÿÿNOEL CASTREE

Secondary sources and referencesBrenner, N., Castree, N. and Essletzbichler, J. (eds) (2004) ‘David Harvey’s The Limits

to Capital two decades on’, Antipode 36 (3): 401–549. Castree, N. (2006) ‘Geography’s new public intellectuals?’, Antipode 38 (2): 396–412. Harvey, D. (1969) Explanation in Geography. London: Arnold. Harvey, D. (1973) Social Justice and the City. London: Arnold. Reissued in 1988. Oxford:

Blackwell. Harvey, D. (1982) The Limits to Capital (Oxford: Blackwell). Reissued in 1999 and 2007

(London: Verso). Harvey, D. (1985) ‘The geopolitics of capitalism’, in D. Gregory and J. Urry (eds) Social

Relations and Spatial Structures. London: Macmillan, pp. 163–185.Harvey, D. (2001) Spaces of Capital. Edinburgh: Edinburgh University Press. Leyshon, A. (2004) ‘The Limits to Capital and geographies of money’, Antipode 36 (3):

461–469.Leyshon, A. and Thrift, N. (1997) Money/Space. London: Routledge. Rigby, D. and Webber, M. (1996) The Golden Age Illusion. New York: Guilford.


UNEVEN DEVELOPMENT (1984): NEIL SMITH

Martin Phillips

At the very least, uneven development is the geographic expression of the contra-dictions of capital … The historic mission ofcapital is the development of the forces ofproduction via which the geographical equal-ization of conditions and levels of productionbecome possible. The production of nature isthe basic condition for this equalization, butequalization is continually frustrated by the differentiation of geographic space. (Smith,1984: 152)

Introduction

In introducing Neil Smith’s (1984) UnevenDevelopment, I will begin with something ofan admission. As a Master’s student with aninterest in Marxist ideas of political economyand uneven development, I bought this booksoon after its publication in 1984 but strug-gled to really ‘get into’ the book, preferringinstead David Harvey’s (1982) The Limits toCapital for its steady and accumulating linesof argument (see Castree, Chapter 8 this vol-ume). However, I returned to Neil Smith’sbook in later years, particularly when prepar-ing to write a book on society and nature(Phillips and Mighall, 2000), at a time whenI had become very interested in other writ-ings by Neil Smith on gentrification (e.g.Smith, 1996). Re-reading the book after somany years, I began to think about the dif-ferent ways in which I engaged with it, andwhether this might say something about thebook itself or simply says something aboutmy changing interests and ability to tackle

what was, for me at least, initially quite adifficult book. Though I do not wish to dis-count the latter, I am going to suggest thatmy experiences reveal something significantabout the character of the text and the variedinterests of its author, arguing that there areconnections between my reading of the bookand its wider reception over the years.

The text and its author

As a reader I initially approached UnevenDevelopment with two quite different interests inmind, namely political economy and society–nature relations. In re-reading the book it wasre-assuring to find that these two concernsappear to have been those of its author as well.In its preface, for instance, Neil Smith statesthat the book ‘represents the meeting of twotypes of intellectual investigation’, namely aconcern with ‘renovating the terribly archaicconception of nature that dominates westernthought’ and an examination of the process ofuneven development which emerged out ofbeing ‘fascinated with the process of gentrifi-cation’ and a belief that this ‘was itself aproduct of spatially more universal, if quitespecific, forces operating at different scales’(Smith, 1984: vii).

Smith argues that his book seeks to integratethese two concerns, suggesting they have ‘untilrecently enjoyed little serious cross fertilisation’(Smith, 1984: ix). Accordingly, the book mightbe read as two books within one, with Chapters1 and 2 being concerned about nature and

9


introducing the conception of socially pro-duced nature, while the majority of the book(Chapters 3 to 5) focuses on the spatiality ofeconomic development and the concept of thesocial production of space. However, such abifurcated reading misses a clear line of argu-ment running through the book that acts tolink the two sections: the idea that nature andspace are socially produced. The argument,clearly outlined in the book’s introduction, isthat the latter is very much predicated on theformer, and that misunderstandings of theformer have hindered understandings not onlyof nature but also of space.

Smith begins Uneven Development by high-lighting the complex and contradictory waysin which nature has been defined, beforearguing that these views enact one of twoconceptualizations, namely either depictingnature as that which is external, a ‘realm of extrahuman objects and processes existing outsidesociety’ (Smith, 1984: 2) or portraying natureas that which is universal (i.e. nature as a char-acteristic or inherent state). Smith suggeststhis dualism can be seen in the philosophicalwritings of Kant, although he argues it hasboth earlier roots and more recent manifesta-tions. With respect to the latter, he focuses ontwo particular ‘modes of expressing and con-ceptualising nature’ (Smith, 1984: 3), whichhe identifies as ‘the scientific’ and ‘the poetic’.

In discussing science and nature, Smithfocuses particularly on the work of Bacon andNewton, suggesting they both utilized andpromulgated concepts of external and univer-sal nature. The latter sense of nature is implied,for instance, in notions of a ‘scientificmethod’, which Smith argues ‘dictates anabsolute abstraction both from the social con-text of the objects under scrutiny and fromthe social context of the scientific activityitself ’ (Smith, 1984: 4). By way of illustration,Smith comments on Newton’s ‘explanation’of an apple falling due to gravity:

When he watched the apple fall, Newtondid not ask about the social forces and

events that led to the planting of theapple trees … Nor did he ask aboutthe domestication of fruit trees that gavethe apple its form. He asked, rather, aboutthe ‘natural’ events defined in abstractionfrom its social context. (Smith, 1984: 4)

With reference to the poetic, Smith focuseson the concept of wilderness as expressed inNorth America by landscape artists such asCole, Church and Durant and writers such asEmerson and Thoreau. Smith suggests that thepoetic concept of nature presented in suchworks, although appearing quite differentfrom, and at times quite antagonistic to, thescientific perspectives, still relied on the twindualism of external and universal nature. Inparticular, Smith emphasizes that the ‘wilder-ness’, from the period of nineteenth-centuryromanticism onwards, has been seen both as aplace to escape from society and as a placewhere people can attain some transcendental/universal union with nature.

Smith argued that this poetic vision ofnature is ideological, which he defines as aviewpoint that is ‘an inverted, truncated, dis-turbed reflection of reality … rooted in …the practical experience of a given social class’(Smith, 1984: 15). In particular he suggeststhat the universal concept of nature is used to‘invest certain behaviours with the status ofnatural events’ whereby they are seen to beuniversal in both space and time: a norm tobe enacted everywhere and unchangingthroughout the past and into the future. ForSmith both elements of the dualistic conceptof nature are, however, bourgeois constructs,derived from and serving to justify capitalisticinterests.

Having outlined the dualistic concept ofnature, Smith turns his attention to undermin-ing it by highlighting how nature is ‘sociallyproduced’. For Smith the work of Karl Marxprovides key insights for the development of anon-ideological concept of nature, even thoughhe admits this concept is not ‘laid out com-pletely or succinctly’ (Smith, 1984: 34) in the

72ÿÿMARTIN PHILLIPS


writing of this nineteenth-century self-styled‘political economist’. Nonetheless, Smith sug-gests that a non-dualistic account of nature canbe discerned in Marx’s writing, one that is bothlogical and historical in character. Here Smithemphasizes the concepts of abstract and concretetheory mentioned by Marx (subsequently elab-orated in discussions of realism by the likes ofSayer, 1984), suggesting that nature should beseen as socially produced both in a general orabstract sense, and also more specifically, or con-cretely, as through time it becomes increasinglyproduced ‘for exchange’ and ‘by capitalist pro-duction’ (see Table 9.1).

Smith argues that in the most abstractsense, nature can be seen as socially producedin the dual sense that people, in order to live,have to work on nature and, in the process,are affected by nature. Smith suggests thattransformation of nature through work is a‘natural activity’ in the sense of being ‘aneternal nature-imposed necessity’ (Smith,1984: 35). In other words, production is anatural activity in the universal sense thathuman life would not be possible without.Furthermore, the activity can be seen toinvolve nature as ‘external’ in that, to use the

highly gendered language in which Marxwrote, ‘man … opposes himself to nature …setting in motion arms and legs, head andhand, the natural forces of his own body, inorder to appropriate nature’s production in aform adapted to his own wants’ (Marx, 1976:177; quoted in Smith, 1984: 36). However,within this quote one can see Marx bothemploying the notion of external nature andalso subverting it through reference to a uni-versal construction in which humans areplaced in nature. Furthermore, as Smith high-lights, Marx also subverted universal notionsof nature by suggesting that in the act ofappropriating nature through production,humans not only change external nature butalso their own nature, changing their physiol-ogy through the development of tools, theirmaterial needs, their consciousness and theirsocial relationships with each other. Societyand nature are, Smith argues, brought intounity through the natural activity of humanlabour on nature, which in turn creates, orproduces, new forms of human and non-human nature.

For Smith, the social production of nature iscentred around what Marx termed ‘use-value’

NEIL SMITHÿÿ73

Level of abstraction Production of nature characteristics

Production in general Production is a natural activity in the universalistic sense in that(Most abstract) without it there would be no life. Production involves the appropriation

of external nature utilizing the natural capacities of humans but inengaging in this activity both human and non-human nature istransformed.

Production for exchange Productive activity is orientated towards market exchange and there isextension in the scale of the production of nature that involves amovement towards second nature.

Capitalist production Production occurs through the wage–labour relationship in which (Most historically/spatially people who have become separated from the means of productionspecific) (i.e. land, equipment, resources, finance) sell their labour power to

capitalists who own the means of production and who extract aprofit by paying labour less than it produces in value. Capital requirescontinual production to extract profit and there is hence an inherentdrive to consume nature. First nature becomes transformed into second nature.

Table 9.1 Smith’s logico-historical construction of the production of nature


in that human labour on nature is focused onsatisfying basic, natural, needs such as suste-nance and shelter. He also adds that whilst thisabstract concept of the social production ofnature offers some important insights, particu-larly in relation to challenging some of theprecepts underpinning the dualistic construc-tions of nature previously identified, it is still of‘fairly limited’ value (Smith, 1984: 38). What isneeded, he suggests, is to examine more ‘con-crete’, historically specific forms of humanactivity and the production of nature. Inparticular, Smith draws attention to the emer-gence of ‘production for exchange’ and‘capitalist production’. The former relates tothe emergence of market exchange societieswhereby production is determined not by theuse-value a particular product may have butmore by the amount of other products, ormoney, that might be exchanged for it. Smithargues that this historically emergent form ofproduction represents an extension of the scaleof the production of nature such that ‘[h]umanbeings produce not only the immediate natureof their existence, but produce the entire soci-etal nature of their existence’ (Smith, 1984:44). As with production in general, Smith dis-cusses societal nature in terms of object ofproduction, human characteristics, reproduc-tion processes and human consciousness. Hesuggests that, amongst other things, one cansee the use of non-human nature being deter-mined by exchange as opposed to use value,growing separation between producers andthe products of their labour, deepening genderdivisions in labour, and an increased differen-tiation of mental activity from manual labour,such that nature itself becomes the subject ofmental contemplation as well as productivelabour. Smith (1984: 45) suggests that it iswithin this mode of production that a ‘cleav-age is created between nature and society’ – adivision which he elaborates in relation to theemergence of ‘second nature’.

Smith argues that the concept of secondnature, which he traces back to the work of

Greek and Roman philosophers such as Platoand Cicero, both emerged with, and shedslight on, exchange economies. Smith uses theterms ‘first’ and ‘second’ nature to refer topristine and humanly modified nature respec-tively. As such, a wild landscape, plant oranimal might be described as an element offirst nature, while a domesticated landscape –an agrarian landscape of cultivated fields, acultivated garden plant or a carefully breddomesticated animal – might be described asinstances of second nature in that they are theproduct of quite intensive human labouralthough still widely viewed as being naturalbecause their existence is dependent on non-human forces and processes.

This notion of first and second nature hasbeen widely employed in some more recentdiscussions of society and nature relations (e.g.see Castree and Braun, 2001). Smith, however,utilizes the terms in at least two further sensesthat differentiate and intertwine with eachother in a manner akin to a triple helix (seeFigure 9.1). First, he uses such terms to refer tonon-human and human entities respectively, orwhat might conventionally be described as ‘thenatural’ and ‘the human’. Such a terminology,however, effectively enacts the nature as exter-nal dualism of which Smith is critical, and byutilizing the notion of first and secondnature, Smith is able to make reference tohuman/non-human differences without imply-ing that the human is non-natural and also tohighlight how notions of nature as universalcome to be applied to human relations, such asthe commodity relations of market society.Finally, Smith uses the terminology of first andsecond nature to refer to a distinction betweenmateriality and abstraction that he sees as cen-tral to understanding capitalist production. Heargues, for instance, that while within produc-tion for exchange, ‘the difference between firstand second nature is simply the differencebetween the non-human and the humanlycreated worlds’ (the second sense of first andsecond natures discussed above), within



capitalist production ‘the distinction now isbetween a first nature that is concrete andmaterial, the nature of use values in general, anda second nature which is abstract, and derivativeof the abstraction from use-value which isinherent in exchange-value’ (Smith, 1984: 55).

Smith then outlines a classic Marxistaccount of capitalist production in whichlabour has become divorced from the meansof production and hence has to sell itself tocapitalists who extract surplus value (orprofit) by paying the workers less in valuethan they produce. There is therefore a con-tinual drive to produce, for without this therecan be no extraction of surplus value. In turn,this leads to new productions of naturewhich become dominated by second naturein all three of the senses implied by Smith. So,for instance, Smith writes ‘capital stalks theearth in search of material resources’, trans-forming nature into an ‘appendage of theproduction process’ (Smith, 1984: 49) and leav-ing ‘no original relation with nature unaltered,no living thing unaffected’ (Smith, 1984: xiv).Moreover, the historically emergent and highly

abstract exchange-value relations of capitalismbecome naturalized as ‘second-nature’ andcome to produce not only non-human naturebut also human nature.Capitalist production is,Smith argues, now ‘squarely at the centre ofnature’ (Smith, 1984: 65).

From Chapter 3 of the book, Smith turnshis attention to the issues of space and theconcept of uneven development. He starts byconsidering the ‘production of space’, a notionwhich he notes is ‘logical corollary of theproduction of nature’ (Smith, 1984: 66),although he adds that in this second part ofthe book he is keen to develop this conceptindependently from his discussion of the pro-duction of nature, before seeking to drawinglinks between the two productions. This hav-ing been said, it is clear that Smith adoptsvery similar lines of argument, outlining whathe sees are key problems in existing theoriesbefore drawing on Marxist concepts todevelop an understanding which, in his view,rectifies many of these problems.

The initial problem identified by Smithis the presence of ‘deeply engrained and

NEIL SMITHÿÿ75

Figure 9.1 Senses of first and second nature in uneven development

Pristine Nature

First Natures

Second Natures

Human Nature

Abstract Nature

ConcreteNature

Non-HumanNature

Humanly ModifiedNature


commonly held prejudices concerningspace’, whereby space is seen as ‘a field ofactivity or as a container’ and where there is‘a rather mechanical integration of space andsociety’ (Smith, 1984: ix). This understanding,Smith argues, embodies elements of twoconcepts of space known formally as absoluteand relative space, elements which, as with thescientific and poetic concepts of nature, existin a tangled inter-relationship. Furthermore,as with nature, Smith argues that these con-cepts emerged in association with changes inthe modes of human productive activityrelated to the rise of market exchange andcapitalism. He suggests that the concept ofabsolute, or Newtonian space, whereby space isseen as ‘a universal receptacle in whichobjects and events occur’ and which therebyacts as ‘a frame of reference, a co-ordinatesystem … within which all reality exists’(Smith, 1984: 68), appeared in associationwith the practices such as cadastral surveyingand the rise of commodity exchange whichfostered abstraction from both use and fromthe immediacy of place. Similarly, mathemat-ical concepts of relativistic space associatedwith, for instance, Einstein’s theory of relativ-ity, are seen by Smith to be expressions thatrelate to the development of capitalist pro-duction; and also the associated developmentof ‘second nature’ whereby society came tobe viewed as separate from nature andnotions of distinct natural and social spacesemerge abstracted from each other and fromlocations in physical, Newtonian space.

Smith goes on to argue that not only havethese concepts of space emerged out of pro-duction for exchange and capitalist production,but they have been integral to the workings ofthese modes of production. Focusing primarilyon capitalist production and drawing heavily onideas associated with Marx and his ‘labour the-ory of value’, Smith suggests that capitalistproduction involves the production of bothabsolute and relative space. Absolute space, heargues, is significant in that concrete labour

processes require a particularity of spatial attrib-utes: particular people, raw materials, andtechnologies combine in specific locations toproduce commodities. However, within capi-talism this particularity is always situated in –and produced out of – wider relations: forinstance, the product being produced will beoffered up for sale to a diversity of purchasers ina wide range of locations, and the people whoproduce it will be ‘hired’ in competition withother people. Smith further notes tensionsbetween the capitalist production of thesespaces, suggesting that while there is a generalhistorical movement towards relative space,epitomized in the globalization of capitalistproduction and the increasing mobility of cap-ital, absolute space is not eliminated but insteadmoves from spaces of ‘first nature’ to spaces of‘second nature’. In other words, whilst capital-ism increasingly relativizes nature throughvaluation via exchange values, it also createsnew particularities of space by, for instance, stri-ating space through social relations such asproperty and state territories, and through thefixity required by some forms of investment ofcapital (a point which is quite central to Smith’s‘rent-gap’ theory of gentrification – see Smith,1996). Within capitalist production there are,Smith argues, inherently contradictory ten-sions between geographical differentiationand equaliza tion, tensions whose resolution, hesuggests, provide the basis for understandingthe basis of uneven development.

For Smith there is, however, one furtherissue which needs to be addressed in order to‘completely specify’ (Smith, 1984: 131) theprocesses underlying uneven development, andthat is the issue of scale. Once again the focusfor Smith is production; hence UnevenDevelopment involves not only consideration ofthe production of nature and the productionof space, but also requires consideration of the‘production of scale’, with this last constructionitself very much being produced out of thefirst two creations. Drawing on his discussionof the production of space, for instance, Smith



suggests that the dialectic of differentiationand equalization of capitalist production isbound up with concepts of absolute and rela-tive space and first and second natures,producing three ‘primary’ scales of spatiality:the urban, the nation-state and the global.Importantly, while these terms have been rou-tinely employed within geographical analysis,Smith notes the arguments of Taylor (1982)concerning the need to view them not as pre-given, natural entities, but to reflect on theirrole in the social production of space. ForSmith the existence of these scales stems prin-cipally from the role they play in relation tothe processes of differentiation and equaliza-tion. In particular, he suggests that the urbanand the global scales respectively ‘represent theconsummate geographical expression of thecontradictory tendencies toward differentia-tion and equalization’ (Smith, 1984: 142),while nation-states are viewed as ‘differenti-ated absolute spaces’ created as a result of‘historical deals, compromises and wars’ (seeTable 9.2).

Neil Smith ends Uneven Development byhighlighting how his abstract analysis mighthelp illuminate aspects of the uneven develop-ment of the world in the 1970s and early1980s, plus he discusses some of the omissionsof his analysis and its inherently politicaldimension. For Smith this period witnessed arestructuring of geographical space which was‘more dramatic than any before’, and involvedchanges such as ‘[d]eindustrialization andregional decline, gentrification and extramet-ropolitan growth, the industrialization of theThird World and a new international divisionof labour, intensified nationalism and a newgeopolitics of war’. The range of issues, theexplicit call for recognition of the politicaldimensions of academic study and the exten-sive utilization of ideas derived from the workof Marx can all be seen to situate UnevenDevelopment and its author in the emergenceof radical/Marxist geography: indeed Peet(1998: 93) cites the book as ‘symptomaticof the best of Marxist geography’ whileCastree (2000: 269) argues that it ‘was a

NEIL SMITHÿÿ77

Spatial scale Dominant spatial tendency Key features

Global Equalization Universalization of the wage–labour relation through bothpolitical and economic integration encouragesequalization. However, differentiation does occur,principally through the differential value of labour powerwhich results in an international division of labour.

Nation – Not directly conditioned by capital differentiation andequalization, but rather is the product of series of politicaldeals, compromises and conflicts. Key impetus comesfrom the competition between different capitalistorganizations, with the nation–state often acting to protectthe collective interests of capital within its boundaries, plusacting to defend territories militarily and to regulate andguarantee the maintenance of the working class.

Urban Differentiation Manifestation of the centralization of capital which createsurban space as an ‘absolute space of production’. Extentdetermined by extent of local labour market and limit ofdaily commute within which equalization works. However,differentiation predominates through ground–rent systemwhich sorts competing land uses into different areas.

Source: Based on Smith (1984)

Table 9.2 The production of scale within capitalist production


major contribution to the development andconsolidation of Marxist geography as aleading postpositivist paradigm’.

Smith wrote Uneven Development soon aftercompleting a doctorate under the supervisionof the renowned Marxist geographer DavidHarvey at Johns Hopkins University, Baltimore,having previously studied at the University ofSt Andrews, Scotland. Corbridge (1987) arguesthat Uneven Development exhibits a clear debt toHarvey, particularly his Limits to Capital (seeCastree, Chapter 8 this volume). Smith himself,as noted earlier, explicitly argued that his bookdrew upon ideas developed in his doctoratewhich focused on one aspect of the restructur-ing of geographical space mentioned in UnevenDevelopment, namely gentrification, wherebycentral areas of cities were seen to be beingredeveloped for and in some cases by middleclass residents. By the time of publicationof Uneven Development, Smith had alreadypublished several articles which sought todemonstrate the link between gentrificationand the movement of capital associated withuneven development (e.g. Smith, 1982) andsubsequently Smith has developed an extensivecorpus of work on gentrification, much ofwhich continued to stress how gentrificationcould be seen as the product of spatially moregeneral processes associated with uneven devel-opment (see especially Smith, 1996). Prior toUneven Development Smith had also writtenon Marxist’s conceptualization of nature (Smithand O’Keefe, 1980) and on the theory andpractices employed in geography (Smith(1979). Subsequent to its publication Smithhas continued to explore both strands of work(e.g. Smith, 1987, 1992b) as well as continu-ing arguments relating to the production ofscale (Smith, 1992a).

Reception and evaluation

Uneven Development emerged at a time ofconsiderable debate about the purpose and

form of geography, and given this context itis not unsurprising that much of the com-mentary on the book has focused on thecontribution that it made to the establish-ment of a Marxist perspective and the degreeto which it exhibited some of the perceivedstrengths and weaknesses of this approach(e.g. Corbridge, 1987; Sack, 1987; D. Smith,1986; Castree, 2000; Swyngedouw, 2000). Inretrospect, Smith’s book might be seen toenact what now is often described as a ‘mod-ernist’ approach or attitude, whereby theoryis seen to revolve around the production ofconcepts which ‘correspond to central phe-nomena or processes around which theworld is seen to operate’ (Phillips, 2005: 56).Uneven Development visibly aspired to be‘grand theory’, and as such can be subjectedto criticism for being both totalizing andinsensitive to difference, although, as Castree(1995: 30) notes, the Marxism of Smith doestemper its explanatory-diagnostic impulse byrecognizing that knowledge is itself producedwithin historically, and spatially specific, con-texts, and hence ‘embodies a moment ofself-reflectivity concerning its claims to truthand reason’. For Castree, however, this self-reflective impulse is not taken far enough andhence the work of Smith still enacts an overlymodernist attitude.

Smith has, in part, recognized some defi-ciencies in his analysis, recognizing that genderrelations may have dynamics which cannot beread directly off the dynamics of production asimplied in Uneven Development. He has alsosubsequently bequeathed greater agency tosocial identity and subjectivity (Smith, 1996a;see also see Marston, 2000). This is not to say,however, that Uneven Development has beenconsigned to gather dust on people’s shelves,to be pulled out only as an illustration of anow unfashionable approach. Quite the oppo-site, as the 1990s saw the book re-published(Smith, 1990) and become widely cited,albeit arguably more for some of the termsand concepts it invoked rather than for its



over-arching, and complexly inter-related,theories of uneven development. In particular,Smith’s concepts of the production of natureand scale have become the focus of consider-able research and debate (see for instanceCastree, 1995, 2002; Marston, 2000). Onceagain, not all of the ensuing commentary hasfully endorsed the arguments of UnevenDevelopment. Castree (1995: 20), for instance,has argued that whilst the book represented akey intervention in the debate about capital-ism and nature, Smith’s concept of theproduction of nature seems to imply that cap-italism ‘determines every aspect of nature’,leaving no room for any ‘agency of nature’,whereby entities and processes might be con-ditioned by something other than humanlabour, either in general or by the particulari-ties of capitalist production. To return to thelanguage of Smith, perhaps first natures havenot been totally erased by capitalist secondnature. Similarly, with respect to the produc-tion of scale, questions have been raised bothabout the degree to which capitalist produc-tion is the sole producer of scale, and indeedwhether scale itself exists (see Marston, 2005;Smith, 2003).

More than 20 years after its publication,Uneven Development is still clearly the subjectof interest and debate. Indeed Castree (2000:269) has suggested that the ‘power and rele-vance’ of the book appear to have increasedrather than diminished in the years since itsinitial publication. Developments such asgenetic engineering and cloning, for instance,can be viewed as indicators of theproduction of humanly produced ‘secondnatures’, whilst globalization can be viewed asa movement into relativistic space which hasnot only continued but accelerated since the1980s. Furthermore, as Smith (2000) notes,globalization raises the issue of scale quitedirectly, not least in relation to the powerand significance of the nation-state. UnevenDevelopment is hence quite clearly a book ofgreat contemporary relevance, although, as my

own initial encounters with the text testify, itis not necessarily a very easy book to get togrips with. Not only does it challenge manyestablished perceptions of how the world isstructured – for instance, in terms of distinctrealms of society and nature – but, as I havetried to illustrate here, uses terms in multipleand relational ways. So for instance, Smith notonly differentiates first and second natures inconjunction with his abstract/concrete dis-tinctions (Figure 9.1), but in turn links theseto both use- and exchange-value and absoluteand relative space distinctions. Hence, in mar-ket exchange societies production is seen asincreasingly orientated towards production forexchange and placed in the abstract frame-work of absolute space, although retaininga grounding in the concrete materiality offirst nature. In capitalist production societies,nature itself became increasingly a secondnature product in relation to exchange valua-tion, whilst production in general is increasinglylocated in relativistic space.

By contrast to such ‘total theory’ where‘everything hangs together with everything else’(Castree, 2000: 267), subsequent researchers havetended to select particular components of thebook for discussion and research, most notablythe concepts of ‘production of nature’ and ‘pro-duction of scale’, without really endorsing, orarguably perhaps even being aware of, thebook’s overall perspective. The production ofnature, space and scale have come to be phrasesof great significance in contemporary humangeography, although studies associated withthem have been informed by a number of per-spectives, such as actor-network theory andsocial/cultural constructionism, that might beseen as quite antithetical to the Marxist theo-rization advanced by Smith. There have beensome claims that there may be scope for somerapprochement between these approaches (seeCastree, 2002), as well as some calls to reflect ofwhat might have been lost within recent appro-priations of Uneven Development. Swyngedouw(2000: 268), for instance, suggests that the book

NEIL SMITHÿÿ79



is distinctive and ‘even more powerful todaythan at the time of its writing’ in its willingnessto ‘grapple with “big issues”’. Castree (1995: 27)similarly suggests that there are still importantcontributions to be made by a Marxism which,whilst taking seriously ‘limited discursive

worlds’ and local environmental struggles andproblems, still sees value in building ‘a largerproject which situates those problems and joinstheir antagonism into a more global view ofand struggle over nature’s (and society’s) creativedestruction’.

Secondary sources and referencesCastree, N. (1995) ‘The nature of produced nature: materiality and knowledge construc-

tion in Marxism’, Antipode 27: 12–48.Castree, N. (2000) ‘Classics in human geography revisited’, Progress in Human

Geography 24: 268–271.Castree, N. (2002) ‘False antethesis? Marxism, nature and actor-networks’, Antipode 34:

111–146.Castree, N. and Braun, B. (2001) Social Nature: Theory, Practice and Politics. Oxford:

Basil Blackwell.Corbridge, S. (1987) ‘Review: Neil Smith, Uneven development: nature, capital and the

production of space’, Antipode 19: 85–87.Harvey, D. (1982) The Limits to Capital. Oxford: Basil Blackwell.Marston, S. (2000) ‘The social construction of scale’, Progress in Human Geography 24:

219–242.Marston, S., Jones, J.P. III and Woodward, K. (2005) ‘Human geography without scale’,

Transactions, Institute of British Geographers 30: 416–432.Marx, K. (1976) Capital: A Critique of Political Economy, volume 1. Harmondsworth:

Penguin.Peet, R. (1998) Modern Geographical Thought. Oxford: Blackwell.Phillips, M. (2005) ‘Philosophical arguments in human geography’, in M. Phillips (ed.)

Contested Worlds: An Introduction to Human Geography. London: Ashgate.Phillips, M. and Mighall, T. (2000) Society and Exploitation Through Nature. Harlow:

Prentice Hall.Sack, D. (1987) ‘Review: Uneven development: nature, capital and the production of

space by Neil Smith’, Geographical Review 77: 130–132.Sayer, A. (1984) Method in Social Science: A Realist Approach. London: Hutchinson.Smith, D. (1986) ‘Review: Uneven development: nature, capital and the production of

space by Neil Smith’, Transactions of the Institute of British Geographers, new series11: 253–254.

Smith, N. (1979) ‘Geography, science and post-positivist modes of explanation’,Progress in Human Geography 3: 356–383.

Smith, N. (1982) ‘Gentrification and uneven development’, Economic Geography 58: 139–155.Smith, N. (1984) Uneven Development: Nature, Capital and the Production of Space.

Oxford: Basil Blackwell.Smith, N. (1987) ‘Academic war over the field of geography’: the elimination of geog-

raphy at Harvard, 1947–1951, Annals of the Association of American Geographers77: 155–172.


NEIL SMITHÿÿ81

Smith, N. (1990) Uneven Development: Nature, Capital and the Production of Space(2nd edition). Oxford: Basil Blackwell.

Smith, N. (1992a) ‘Contours of a spatialized politics: homeless vehicles and the produc-tion of geographical scale’, Social Text 33: 55–81.

Smith, N. (1992b) ‘History and philosophy of geography: real wars, theory wars’.Progress in Human Geography 16: 257–271.

Smith, N. (1996) The New Urban Frontier: Gentrification and the Revanchist City.London: Routledge.

Smith, N. (1996a) ‘Spaces of vulnerability: the space of flows and the politics of scale’,Critique of Anthropology 16: 63–77.

Smith, N. (2000) ‘Socializing culture, redicalizing the social?’, Social and CulturalGeography, 25–28.

Smith, N. and O’Keefe, P. (1980) ‘Geography, Marx and the concept of nature’, Antipode12: 30–39.

Smith, R.G. (2003) ‘World city topologies’, Progress in Human Geography 27: 561–582.Swyngedouw, E. (2000) ‘Classics in human geography revisited’, Progress in Human

Geography 24: 266–268.Taylor, P.J. (1982) ‘A materialist framework for political geography’, Transactions, Institute

of British Geographers 7: 15–34.


SPATIAL DIVISIONS OF LABOUR (1984):DOREEN MASSEY

Nick Phelps

Space can be … conceptualised as the productof the stretched-out, intersecting and articulat-ing social relations of the economy. (Massey,1984: 2)

Introduction

At the outset of this chapter we can say that,like all books (academic or literary), SpatialDivisions of Labour is one ‘which requiresevaluation at different levels’ (Smith, 1986:189). It is curious to think that it is at oneand the same time deeply immersed in thespecific concerns of industrial location – aterm now rarely used in mainstream eco-nomic geography – and yet punctuated byprofound observations on the philosophy ofhuman geography. It is at once a critique ofexisting location theory and economic geo-graphical work and yet also a new approachto understanding uneven economic devel-opment. It is a theoretical tour de force andyet also a detailed empirical exploration ofthat theoretical approach. Among a group oflike-minded scholars its importance wasimmediate and prompted a major academicresearch programme – the Changing Urbanand Regional Systems (CURS) initiative –funded by the UK government’s Social ScienceResearch Council (now the Economic andSocial Research Council). Yet it has also hada profound longer-term impact in shapingnew academic research, presenting newpoints of departure for work on geography

and gender, geometries of power as well asprefiguring interest in, for example, labourmovement strategies and the institutional andcultural basis of capitalism. It is these multipleimpacts that have ensured Spatial Divisionsof Labour’s stature and its enduring appeal.Moreover, in all of this, something of theauthor’s character is revealed right from thefirst until the very last page.

After having read geography at OxfordUniversity (1963–1966), Doreen Massey wenton to study for an MA at University ofPennsylvania, Philadelphia (1971–1972) as aconscious decision to immerse herself in posi-tivist social science in general, and neoclassicaleconomics, location theory and regional sci-ence in particular. In the years following, whileworking at the Centre for EnvironmentalStudies, Massey was to develop the variousstrands of a critique of established positivistand behavioural location theory (Massey, 1977, 1978). This was also a time of profoundrestructuring of the UK economy – a periodsometimes referred to as de-industrialization.As Massey herself noted, the book was writtenat a moment that allowed reflection on majortransformations occurring since the 1960s. It isthe ‘deep’ understanding of these processes ofrestructuring in the UK, their industrial speci-ficity (Massey and Meegan, 1982) and diverseimpacts on cities and regions (Massey andMeegan, 1979, 1978) that forms the empiricalsubstance of Spatial Divisions of Labour. In thisregard, and ‘as an analysis of employmentchanges in post-war Britain’, it is ‘full of

10


84ÿÿNICK PHELPS

fascinating insights’ (Smith, 1986: 189).However, the book today may appear slightlycurious with its many references to a branch ofhuman geography – industrial geography –now largely disregarded. In this and in otherrespects, Spatial Divisions of Labour is, as we havecome to realize through the work of TrevorBarnes (2004), most definitely a product of aparticular time and place. And yet, throughits clarity, force of argument and novelty, SpatialDivisions of Labour has gained sufficient longevityto be used and cited in markedly different cir-cumstances (sometimes in a way that DoreenMassey herself might have guarded against). Assuch, while Doreen Massey had already estab-lished herself as a leading and highly innovativeeconomic geographer when Spatial Divisions ofLabour was published in 1984, its publication(and subsequent reissue in 1995) cemented herreputation in the discipline as a whole.


From the start, Massey’s Spatial Divisions ofLabour was much more than an industrialgeography of the UK. As she argued, ‘thestudy of industry and production is not just amatter of “the economic’’ but rather thateconomic relations and phenomena arethemselves constituted on a wider field ofsocial, political and ideological relations’(Massey, 1984: 7, original emphasis). As such,Spatial Divisions of Labour prefigured recentinstitutional and cultural ‘turns’ in economicgeography. And, as David Smith noted at thetime, ‘the book goes well beyond industrialand employment matters [to offer] a majormethodological redefinition of human geog-raphy’ (Smith, 1986: 190). It both set out todevelop a new theory of uneven geographi-cal development at the same time as itexemplified those concerns empirically. Whatwe have is ‘a densely argued book thatconstantly swings from abstract theory to

finely-honed empirical examples, culminat-ing in a philosophical claim about the verynature of social life’ (Clark, 1985: 291). Ifat some points Spatial Divisions of Labour isgrounded in the specific concerns of indus-trial geography, at others it is intervening inthe most abstract concerns of the philosophyof human geography. At the heart of SpatialDivisions of Labour is a desire, as Massey laterclarified, to reconceptualize space as ‘theproduct of the stretched-out, intersecting andarticulating social relations of the economy’(Massey, 1995: 2).

As I have already alluded to briefly, SpatialDivisions of Labour was steeped in a thoroughunderstanding of the turmoil of the UK econ-omy during the 1960s and 1970s. Somewhatunusually for a book that offers a major re-conceptualization of a sub-disciplinary field, itis also firmly grounded in a rich empiricalknowledge of its subject. In fact, as DoreenMassey has since recounted, the inclusion oflargely empirical chapters – albeit that they arecentral to her project of critique and recon-struction and to her dialectical movementbetween theory and evidence – was seen atthe time to have been something of a poten-tial weakness (for example, it was suggested tothe author at the time that the empiricalexamples in the book might come to beregarded as ideal types and hinder the widerimpact of the book). These largely empiricalelements focused on, for example, the evolv-ing spatial structures associated with theelectronics, clothing and footwear and serviceindustries in the UK, upon the class and gen-der relations in coalfield communities (ofsouth Wales, the northeast of England, centralScotland, and Cornwall), and upon broaderpatterns of uneven development, class forma-tion and politics in the UK as a whole.Nevertheless it was the theoretical content ofSpatial Divisions of Labour and other works byradical geographers that at the time elevatedgeography among related disciplines such as


DOREEN MASSEYÿÿ85

sociology, planning and even economics –with the majority of those involved in theCURS initiative, for example, being drawnfrom these disciplines outside of geography(Cooke, 2008).

In the manner of its creation, SpatialDivisions of Labour also poses some interestingquestions of much contemporary academicwork in human geography. Recent debates inacademic journals have lamented the detach-ment of human geographical research fromand its lack of impact upon policy and poli-tics in the UK in particular (see, for example,Dorling and Shaw, 2002; Martin, 2001, 2002;Massey, 2001, 2002). Yet even at the time andcertainly by today’s standards, Spatial Divisionsof Labour is very unusual in being forged fromthe author’s life-long engagement with pol-icy and politics – a theme that was to formthe title of a subsequent volume co-edited byMassey (Massey and Meegan, 1985).

The Academy – as a central institution ofcivilizations – has several important func-tions. At times it is the centre of the pursuitof abstract knowledge far removed fromeveryday concerns – the subject of HermanHesse’s novel The Glass Bead Game. At othertimes it has been the centre of societal cri-tique and resistance. Rarely, it seems, isacademia involved in both critique andreconstruction. Massey’s Spatial Divisions ofLabour is one such exception and it owes somuch to the author’s close involvement withpractical political and policy concerns. Theseconcerns continue today in the guise of, forexample, her involvement with the thinktank ‘Catalyst’ and questions of regional pol-icy (Amin et al., 2003). Prior to the writingof Spatial Divisions of Labour, Massey had beenclosely involved with the Greater LondonCouncil’s London Manufacturing Strategy devel-oped during the late 1970s and early 1980s.Indeed, there is a case for arguing that theclarity, power and persuasiveness of this mostinscrutable of academic books owes much to

the rigour required explaining concepts inarenas and for audiences outside academia.Hence the book speaks to the process notonly of understanding industrial restructur-ing on the ground in London and elsewherebut also to the immediate and pressing needto find practical policy solutions to the resul-tant social ills. While Spatial Divisions ofLabour and the locality studies that ensuedunder the CURS initiative came at a propi-tious time in relation to theoretical andmethodological developments in geographyand elsewhere in the social sciences, it wasdevelopments in the wider world that theywere most in tune with (Massey, 1995).

Spatial Divisions of Labour sought to con-ceptualize the geography of employment interms of the social relations of production. Tothis end it advanced an approach centred onthree more concrete conceptual innovations:‘place in economic structure’ – the relationof a particular industry or economic sector toother (in Marxist parlance) departments; ‘theorganisational structure of capital’ – a termused to capture dimensions of ownership andcapital concentration apparent in a particularsector of the economy; and the various ‘spa-tial structures’ that may characterize differentindustries at different times. It is the latterconceptual category that Spatial Divisions ofLabour made most famous when identifyingthe ‘locationally-concentrated’, ‘cloning’ and‘part-process’ spatial structures that tended tobe associated with particular industries atparticular times and in particular settings.These spatial structures were closely relatedto – though not determined by – the exi-gencies of different labour processes. Theywere also considered by Massey as merelyexamples that did not exhaust the range ofexisting or potential spatial structures. In thisrespect then, we see Massey’s interest in anddetermination to ‘formulate an approach toconceptualising … the arguably endlessadaptability and flexibility of capital’ (Massey,


86ÿÿNICK PHELPS

1984: 69). However, such was the novelty andpower of these categories that they have beenand, to some extent, continue to embodyrather static reference points or archetypesfor researchers.

Reception and impacts

It is easy to forget that a book of the statureof Spatial Division of Labour neverthelessalso attracted no small measure of detailedevaluation and critique at the time. Despitegeneral approval from most reviewers, a slewof more or less important criticisms were lev-elled at the book. Massey’s bold attempt tobridge the nomothetic–idiographic episte-mological divide in human geography wasitself the single largest cause of criticism.Massey herself reflected years later in thesecond edition how the book was criticizedboth as Marxist and yet not Marxist enough(Massey, 1991: 297). So for John Lovering,for example, Massey’s imposition of Wright’sabstractions on class onto concrete situationsproduced an economistic feel to the analysis(Lovering, 1986). However, the weight ofcriticism at the time fell squarely on the sideof the book being too concerned with theuniqueness of place and hence not beingprincipled enough in Marxist or realist terms.Smith (1986: 180) argued that ‘at the heart ofthe analysis is the uniqueness of place’, withMassey viewing ‘causal relations’ as enablingrather than determining (see Massey, 1995: 4).In the opening chapters Massey was con-cerned to avoid an essentialist Marxist accountof uneven development, developing her crit-icisms of ideas that had surfaced and werepopular at the time, arguing that ‘it is notenough to point to the stage reached in thedevelopment of capitalist relations in order tounderstand the complexity of spatial struc-tures’ (Massey, 1984: 81). Although not anexplicit target of Massey’s critique in thisrespect, Nigel Thrift was able to observe how

‘one of the oddest omissions in the book isany mention of the work of David Harvey’(Thrift, 1986: 148) – odd since David Harveywas and remains the major Marxist figure inhuman geography (see Woodward and Jones,Chapter 15 this volume). However, Harveywas not to stay silent on the matter for long,lamenting the extent to which Spatial Divisionsof Labour was ‘laden down with the rhetoricof contingency, place, and specificity’ to theextent that Marxian categories became inert(Harvey, 1987: 373). This line of argumentwas pursued in various guises by several othercommentators, notably those working in therealist tradition that had also become popularat the time. For Gordon Clark (1985: 292)what seemed lacking was a ‘macro sense ofthe structures of the regional system from aneconomic perspective’. Warde (1985) hadsuggested that Spatial Divisions of Labour was‘highly indeterminate’ in its analysis, whileCochrane identified how ‘there do not seemto be any general rules – or necessary rela-tions – at all’ such that ‘we are left with aprocess of infinite regress, in which the modelis presented, only to recede into the distanceevery time we try to use it’ (Cochrane, 1987:361). This was not the intention of the bookalthough it was something that Massey subse-quently acknowledged as a potential dangerin approaches that attempt to deal seriouslywith industrial and locational specificity(Massey, 1995: 320).

Perhaps the most immediate impact of thebook was to set in train a major UK researchprogramme – the CURS initiative (Cooke,1989; Harloe et al., 1990) – the machinationsof which also condensed many of the aboveconcerns. As Lovering notes ‘in the threehundred pages devoted to the “specificity ofplace” there is no concise or complete con-ceptualization of the locality’ (Lovering, 1986:71). The search for such a conceptualizationoccupied much of the research effort at thetime both within and without the variouslocality studies. As we have already noted,


DOREEN MASSEYÿÿ87

Massey studied during a time of transitionbetween the long-dominant regional geogra-phy tradition and the rapid succession ofnomothetic approaches that characterize the1960s and 1970s. These twin influences areapparent in Spatial Divisions of Labour. Masseythe person, and Spatial Divisions of Labour thebook, were to become icons within Marxisthuman geography and yet David Smith wasable to articulate a sense of the familiar not-ing how ‘there is more than a hint of arevitalized regional geography’ (Smith, 1986:190). The theme was taken up by Loveringwho suggested that ‘the case studies alsoshow how information produced by empiri-cal studies of a quite traditional kind may begiven a new meaning when situated in a morecoherent theoretical framework’ (Lovering,1986: 70). For others the attempt to navigatethe conceptual straits between nometheticaland idiographic approaches was less satisfac-tory since the locality studies that SpatialDivisions of Labour spawned ‘will do littlemore than repeat the empiricist locality stud-ies of an earlier generation which examinedindividual places for their own sake’ (Smith,1988: 62).

Both Spatial Divisions of Labour and thelocality studies commissioned in its wakeappeared at a time of rapid development oftheoretical and philosophical approaches inhuman geography that included not onlyradical/Marxist approaches that had surfacedby the late 1970s but also in particular therealist and structurationist approaches thathad become popular by the 1980s. Thus itwas that most of the sympathetic critique andreconceptualization of locality studies camefrom a realist perspective, culminating inDuncan’s (1988) ‘three levels of locality’effects and in Sayer’s (1989) defence of the‘new regional geography’ as distinct from theolder regional geography tradition.

The tension generated by the dominantnomothetic emphasis on structural processeson the one hand and the idiographic emphasis

on the uniqueness of place on the otherhand is one that begs important questionsof the historical transformation of places.Massey herself likened the process to thelayering of rounds of investment whereby‘spatial structures of different kinds can beviewed historically … as evolving in a suc-cession in which each is superimposed upon,and combined with, the effects of the spatialstructures which came before’ (Massey, 1984:118). The localities research that followedwas to occupy much of its research effortwith such ‘rules of transformation’. In doingso, one inappropriate metaphor achieved aprominence and illustrates clearly the poten-tial for unintended consequences to flow evenfrom the clearest of academic expositions. Thenotion of layers of investment is now inti-mately and erroneously associated with a‘geological metaphor’ originally drawn byWarde (1985). Massey herself felt it necessaryto point out ‘geology is not an appropriate wayof envisaging the layering process I had inmind … surely the notion of the combinationof layers is very ungeological’ (Massey, 1995:231, original emphasis). An alternative, moreaccurate metaphor of the dealing of hands ofcards was suggested by Gregory (1989).

Critiques and reappraisals

Inevitably, a book as ambitious and wide-ranging as Spatial Divisions of Labour waslikely to neglect quite specific aspects of thephenomena it was concerned with. Severaldetailed limitations of the book were high-lighted by reviewers at the time. Thrift (1986)and Warde (1985), for instance, highlightedthe limited attention paid to the role of thereproduction of labour in the book. This issomething that Massey conceded in the sec-ond edition, noting the book ‘never meant toencompass uneven development in everysphere of life’ (Massey, 1995: 334). SpatialDivisions of Labour focused on the spatial


88ÿÿNICK PHELPS

structures produced from the technical divi-sions of labour internal to companies. In thisrespect Massey was careful to note that the‘locationally-concentrated’, ‘cloning’ and ‘part-process’ spatial structures concentrated on inthe book were not in any sense exhaustive ofpossibilities. Yet Spatial Divisions of Labourclearly had an emphasis on the latter two spa-tial structures and partly as a function of thistended to underplay the inter-firm linkagesor social division of labour often associatedwith the single plant spatial structure. Thiswas to become a major focus of academicenquiry during the late 1980s and early1990s as academics rediscovered the phenom-ena of industrial districts and agglomeration.One of the major figures involved in thisenterprise – Michael Storper – was to raise thiscriticism (Storper, 1986) (see also Coe, Chapter17 this volume). It is a limitation that DoreenMassey readily conceded, noting how ‘extend-ing the … analysis … to all the various levelsof economic relations, especially includingthose between firms, would give a much fuller,and more complex and … possibly contradic-tory, picture of spatial divisions of labour’(Massey, 1995: 339).

Spatial Divisions of Labour also arguablybelongs to an increasingly rare corpus of aca-demic geographical writing in which ideasare expressed with clarity. As was noted atthe time (Smith, 1986; Thrift, 1986), SpatialDivisions of Labour is a very well-writtenbook that is free from jargon. The clarity ofits expression notwithstanding, it is also truethat the book is at times an uncomfortableread. It is uncomfortable for a conceptualrigour that exposes the laziness in precedingacademic research in the fields of locationalanalysis, regional economic development,and even some strands of Marxist economicgeography. What comes across repeatedlyis an insistence and a determination in thepresentation and working through of keyissues that avoid easy or fallacious solutions.

As one might expect, such a book and itsstyle are the product of a lively and deter-mined mind. So, David Smith, for example,was able to observe how ‘Some … will beintrigued by the imagery of class conflict(“battle”, “attack”, “assault”, “weapons”) – as ifMassey was demonstrating that a confronta-tional stance, like the struggle of labour itself,is not a male preserve’ (Smith, 1986: 190).Ought we to be surprised that the steelyqualities needed for Doreen Massey to assertherself as an academic leader of immensestature within what was and, to an extent,remains a male-dominated profession shouldappear at moments on the page and in per-son? These are qualities that every academicwould surely recognize in their engagementwith the sometimes alienating machinery ofacademic life, although they are, along withaltogether less admirable traits, ones rarelyexplicitly acknowledged in the academiccommunity.

It is also salutary to remember that SpatialDivisions of Labour’s impact at the time was,unlike geography’s quantitative revolution(Barnes, 2004), not borne of a deliberate orconcerted effort on the part of the authorto bend the academic machinery in itsservice – the sort of effort that has nowbecome perhaps too readily embraced in thepractice of geography. Whilst Doreen Masseyassumed a very influential position as a resultof its publication, she was unable to claimany special efforts on her part to garnerimpact for Spatial Divisions of Labour. If any-thing, her concerted efforts to promoteprogressive geographical thought has comeoutside academia in the realm of policy andpolitics, since Doreen Massey has been a reg-ular contributor to periodicals such asMarxism Today and Soundings. Instead, thesuccess of Spatial Divisions of Labour andthe CURS initiative owe much to the sheervariety and geographical scope of the associ-ations that Doreen Massey had struck up


DOREEN MASSEYÿÿ89

across several overlapping academic and pol-icy groupings as well as the sympathy ofestablished geographical figures such asMichael Wise to the aspirations of radicalgeographers.

The power of Doreen Massey’s SpatialDivisions of Labour rests on its desire toaddress the fundamental schism in humangeography. It navigates a course between anidiographic tradition – the description of

place – on the one side and the nomotheticoutlook – of the study of process – on theother. Its beauty and power lies in theauthor’s calm, clarity of purpose and expres-sion and tenacity in navigating these waters.It is the middle ground – this tensionbetween two rather distinct approaches inhuman geography – that arguably informsmuch of the most interesting work withinhuman geography which followed.

Secondary sources and referencesAmin, A., Massey D. and Thrift, N. (2003) ‘Decentering the nation: A radical approach to

regional inequality’, Catalyst Paper 8. London: Catalyst.Barnes, T. (2004) ‘Placing ideas: genius loci, heterotopia and geography’s quantitative

revolution’, Progress in Human Geography 28: 565–595.Clark, G.L. (1985) ‘Review of Spatial Divisions of Labour’, Economic Geography 61:

290–292.Cochrane, A. (1987) ‘What a difference the place makes: the new structuralism of local-

ity’, Antipode 19: 354–363.Cooke, P. (ed.) (1989) Localities: The Changing Face of Urban Britain. London: Unwin

Hyman. Cooke, P. (2008) ‘Locality debates’, in R. Kitchin and N. Thrift (eds), The International

Encyclopedia of Human Geography. Oxford: Elsevier.Dorling, D. and Shaw, M. (2002) ‘Geographies of the agenda: public policy, the discipline

and its (re)’turns’, Progress in Human Geography 26: 629–646.Duncan, S.S. (1988) ‘What is locality?’, in R. Peet and N. Thrift (eds), New Models in

Geography. London: Allen & Unwin.Gregory, D. (1989) ‘Areal differentiation and post-modern human geography’, in

D. Gregory and R. Walford (eds), Horizons in Human Geography. London: Macmillan.Harloe, M., Pickvance, C. and Urry, J. (1990) (eds). Place, Policy and Politics: Do

Localities Matter? London: Unwin Hyman.Harvey, D. (1987) ‘Three myths in search of a reality in urban studies’, Environment &

Planning D, Society & Space 5: 367–376.Lovering, J. (1986) ‘Book review: Spatial Divisions of Labour’, Urban Studies 23: 70–71.Martin, R. (2001) ‘Geography and public policy: the case of the missing agenda’,

Progress in Human Geography 25: 189–209.Martin, R. (2002) ‘A geography for policy or a policy for geography? A response to

Dorling and Shaw’, Progress in Human Geography 26: 642–644.Massey, D.B. (1977) ‘Towards a critique of industrial location theory’, in R. Peet (ed.),

Radical Geography. London: Methuen.Massey, D.B. (1978) ‘A critical evaluation of industrial location theory’, in F.E.I. Hamilton

and J.C.R. Linge (eds), Spatial Analysis, Industry and the Industrial Environment:Industrial Systems. Chichester: Wiley.


90ÿÿNICK PHELPS

Massey, D.B. (1979) ‘In what sense a regional problem?’, Regional Studies 13: 233–243.Massey, D.B. (1984) Spatial Divisions of Labour: Social Structures and the Geography

of Production. Basingstoke: Macmillan.Massey, D.B. (1991) ‘The political place of locality studies’, Environment & Planning A

23: 267–281, reprinted in Massey, D.B. (1994) Space, Place and Gender. Cambridge:Polity Press.

Massey, D.B. (1995) Spatial Divisions of Labour: Social Structures and the Geographyof Production (2nd edition). Basingstoke: Macmillan.

Massey, D.B. (2001) ‘Geography on the agenda’, Progress in Human Geography25: 5–17.

Massey, D.B. (2002) ‘Geography, policy and politics: a response to Dorling, and Shaw’,Progress in Human Geography 26: 645–646.

Massey, D.B. and Meegan, R. (1978) ‘Industrial restructuring versus the cities’, UrbanStudies 15: 273–289.

Massey, D. and Meegan, R. (1979) ‘The geography of industrial reorganisation: the spa-tial effects of the restructuring of the electrical engineering sector under the IndustrialReorganisation Corporation’, Progress in Planning 10 (3): 155–237.

Massey, D.B. and Meegan, R. (1982) The Anatomy of Job Loss: The How Why andWhere of Employment Decline. London: Methuen.

Massey, D.B. and Meegan, R. (1985) Politics and Method: Contrasting Studies inIndustrial Geography. London: Methuen.

Sayer, A. (1989) ‘The ‘new’ regional geography and problems of narrative’, Environment& Planning D, Society and Space 7: 253–276.

Smith, D.M. (1986) Review of Spatial Divisions of Labour, Regional Studies 20:189–190.

Smith, N. (1987) ‘Dangers of the empirical turn: some comments on the CURS initiative’,Antipode 19: 59–68.

Storper, M. (1986) ‘Review of Spatial Divisions of Labour’, Progress in HumanGeography 455–457.

Thrift, N. (1986) ‘Review of Spatial Divisions of Labour’, International Journal of Urbanand Regional Research 10: 147–149.

Warde, A. (1985) ‘Spatial change, politics and the division of labour’ in D. Gregory andJ. Urry (eds), Social Relations and Spatial Structures. Basingstoke: Macmillan,pp. 190–212.


GEOGRAPHY AND GENDER (1984): WOMEN ANDGEOGRAPHY STUDY GROUP

Susan Hanson

What we argue for … is not … an increase inthe number of studies of women per se ingeography, but an entirely different approach togeography as a whole. Consequently, we consider that the implications of gender in thestudy of geography are at least as important asthe implications of any other social or economic factor which transforms society andspace. (WGSG, 1984: 21)

Introduction

Imagine a world outside your classroom inwhich women’s subordinate status is vividlyevident through gender differences in educational attainment (years of formal edu-cation), employment (job type, hours ofemployment, wages), and political power(presence/absence in governmental decision-making positions). Imagine a world insideacademia in which geography journalsand university geography curricula are practically devoid of the words ‘gender’ and‘women’ and in which all instructionalmaterials implicitly, if not explicitly, assumethat all salient geographical actors are male.Imagine university geography teachingstaffs, graduate student cohorts, and national geography conferences composed almostentirely of white men. Such was the world ofthe early 1980s in which nine members1 ofthe Women and Geography Study Group collectively conceived and drafted Geography

and Gender: An Introduction to FeministGeography (1984).

The text and its authors

Unlike virtually all of the other key texts inthis primer, Geography and Gender was writ-ten explicitly with undergraduates in mind;the writing is refreshingly clear and engaging.In view of the book’s overall message andintent, however, it is clear that the intendedaudience extended well beyond undergradu-ates, to encompass the discipline as a whole.The main message is that it is important tounderstand how gender helps to structuregeographies of inequality if that inequality isever to be erased or at the very least signifi-cantly reduced. Juxtaposing gender (sociallyconstructed difference) vs sex (biological dif-ference), the authors understand genderrelations to be malleable precisely becausethey are socially constructed and thereforesubject to change. They recognize that suchchange – i.e., the kind of change in genderrelations that will yield greater gender equal-ity – will demand far more than simplymaking women visible in geography; it willrequire nothing less than ‘an entirely differentapproach to geography as a whole’ (WGSG,1984: 21). In other words, these nine authorswell understood the radical challenge that fem-inist geography posed to the entire discipline aswell as to students.

11


The book joins an interest in explainingthe value of a feminist approach in geographywith a concern to make clear the tenuousposition of women in UK geography. Thediscipline was decidedly masculine (only 10%of all full-time lecturing staff in the 50 UKgeography departments for which they hadsurvey data were women) and masculinist, inthe sense that women were simply invisiblein geography texts and other teaching mate-rials. As I have argued elsewhere for theAmerican case (Hanson, 2004), a look at thediscipline’s history reveals a close connectionbetween the social identities of its practition-ers and the topics deemed worthy of researchattention, the research approaches used, thenature of data collected, and so on. Geographyand Gender was the first book-length treat-ment of feminist geography, and it provideda glimpse not only of the gender of geogra-phy in the UK at the time but also of howgeographical thinking might contribute tounderstandings of women’s subordination.

The book is divided into three unequalparts. In the first, the authors lay out their argu-ments for studying feminist geography anddescribe four main approaches to understand-ing the unequal power relations that undergirdgender-based inequality: radical feminism,socialist feminism, Marxist approaches, andphenomenological and humanistic approaches.The second part, which comprises the heart ofthe book, contains four chapters, each onefocused on demonstrating the power of a fem-inist geographic analysis to produce newunderstandings in a core area of geographicknowledge: (1) urban spatial structure; (2)women’s employment, industrial location, andregional change; (3) access to facilities; and (4)women and development. In the third part ofthe book the authors turn to the status ofwomen in UK geography and the ways inwhich feminism should change approaches toteaching and research. Before considering thebook’s reception and impact, let us first lookmore closely at what these authors were saying.

In Part One, after making persuasivearguments about women’s invisibility ingeography and the need to overhaul geo-graphic theory so as to understand whywomen remain in subordinate positions, theauthors make clear their preference for social-ist feminism over the alternatives described.Socialist feminists recognize that male familymembers and employers alike currently exploitwomen’s unpaid work in the home, but theyalso believe that the state has a key role to playin effecting positive change for women, forexample by making education and employ-ment opportunities equal, by enabling womento live free of male violence, and by support-ing services that allow women and men tocombine family and work responsibilities.Feminists seek to change the world, andsocialist feminists advocate women and menuniting to pressure the state to enact policiesthat support gender equality.

The four chapters in Part Two take up top-ics of longstanding interest to geographersand demonstrate how feminist insights chal-lenge traditional theories. Each incorporatesthe relatively small amount of feministresearch that had been done by the early1980s, and the empirical material used ineach chapter is decidedly UK-based. Thechapter on urban spatial structure asks studentsto think about the implications of the spatialseparation of home and work that accompa-nied industrialization. The authors trace outsome of the historical ways in which the sep-aration of home and work emerged from thedivision of labor between female domesticworkers and male waged workers. Afterdescribing some of the progressive changes inliving patterns initiated during World War II(e.g., housing with communal kitchens andchild care), they describe the distinctly patri-archal dimensions of post-WWII urbanplanning and invite students to imagine whata non-sexist city might look like. The chap-ter on women’s employment, industrial location,and regional change links changes in women’s

92ÿÿSUSAN HANSON


employment patterns (increased labor forceparticipation, increased gender-based segre-gation in the labor market, and distinctgeographic patterning in women’s work) tolarger processes of economic and regionalchange within the UK. The authors pointout that a failure to disaggregate employmentpatterns by gender obscures some of the keyprocesses that actually drive regional change,such as the distinct geographical variation inthe demand for female labor of differenttypes, a variation that was poorly understoodat the time.

The third chapter in Part Two focuses onaccess to facilities. Through a case study ofwomen’s access to health facilities, the authorsshow that access – the availability of services –entails far more than just the ability to traversespace (travel); access also involves the hours ofoperation and the quality of the service at afacility vis-à-vis women’s needs and constraints.Women and development is the focus of thefourth chapter in Part Two. Noting that, likeother theories in geography, development the-ory has neglected women, the authors suggestthat development policies affect women differ-ently from men, that migration streams aregendered in different ways on different conti-nents, and that young, unmarried women arebeing exploited by capitalists locating certaintypes of industries in developing countries.

Part Three, Feminism and Methods ofTeaching and Research in Geography, docu-ments the disadvantaged status of women inUK geography departments, raises questionsabout differential treatment of male andfemale students by university teachers (how-ever inadvertent such treatment may be),offers ideas for student projects, and discussesthe ways in which feminist ideas shape theresearch process. In this last vein, the authorshave, throughout the substantive chapters ofPart Two, made suggestions about the need foradopting feminist-inspired methodologicalapproaches, such as supplementing quantita-tive surveys about health with in-depth

interviews with women to discern the degreeto which their health needs are being met or,in the context of understanding women’semployment, listening to what women have tosay about their everyday lives. Part Three alsoincludes a plea to pay attention to the genderbasis of male activities. The authors concludetheir book by reminding readers that the goalof feminist geography is to change genderrelations and that we can attain this goal onlyif we fundamentally change our theoreticaland empirical approaches to research.

As I re-read the book recently, I was struckby how the core message – pay attention togender if you really want to understand geo-graphic processes – still resonates. Yet in 1984,when Geography and Gender first appeared,this message was revolutionary! The discipli-nary norm was to ignore gender withabandon. Undoubtedly, this key text played akey role in creating some of the changes wehave seen in geography over the past 20 years,e.g., the increased presence of women in aca-demic geography, the recognition that genderthoroughly infuses geographic processes, andthe now-rather-large and rapidly growingbody of literature in feminist geography.Particularly as I re-read the problem-focusedfeminist analyses in Part Two on employ-ment, access, and development, I was struckby how vastly much more we now knowabout these problems than we did in the early1980s.

Although some of Geography and Gender’scentral arguments do transcend time andplace, the book is also very much a productof a particular time and a particular place. Inthis regard, I consider, first, some of the waysin which the authors’ feminist argumentsreflect the feminism of the time, and, second,some of the ways in which the book is verymuch rooted in the UK. The strong socialistfeminist stance that infuses the text and thevery concept of feminist geography espoused(‘feminist geographers are concerned withthe structure of social and spatial relations

WOMEN AND GEOGRAPHY STUDY GROUPÿÿ93


that contributes to women’s oppression’)signifies dominant concerns of feminism in the1970s and early 1980s (WGSG, 1984: 134).The focus on uniting women as women, agroup whose strongly shared gender identitywas assumed, pre-dates concerns that arose inthe mid-1980s about profound differencesamong women’s experiences and positionsand the problems such diversity poses forfeminist theory and politics. The concept ofpatriarchy, which figures prominently inGeography and Gender and which was verymuch on the minds of feminist theoristsat the time (see, for example, Foord andGregson, 1986), rarely appears in contemporaryfeminist analyses; patriarchy as an explanatoryframework in the ‘grand theory’ tradition hasby-and-large fallen prey to post-structuralistviews, which locate power in contextually vari-able discourses and practices rather than ingrand structures.

The nine authors are all British, and theirexperiences, reflected in the theories andempirical examples selected for inclusion inGeography and Gender, were largely rooted inevents and debates that were unfolding in theUK at the time. One might argue, too, thatthe particular approach to feminism theyespoused was linked to their being situated inthe urban UK. Certainly the topics that formthe focus of the chapters in Part Two, whilehaving wide currency then and now, reflectthe authors’ embeddedness within the UKand within the UK geography of the early1980s. The chapter on women’s employment,industrial location, and regional change, forexample, draws women’s employment intothe then-raging debates about industrialrestructuring and the massive economicdisruptions that were accompanying thatrestructuring. Similarly, the chapter on accesstakes up a long-standing theme in humangeography, one that, with its inherentconcern for equity, was especially relevantin a UK undergoing profound socialtransformations.

We are all parochial, in the sense that wetend to write about what we know, and whatwe know is bound to the particular placesand times in which we find ourselves.Geography and Gender was written with stu-dents, and mainly British students, in mind;although publishers are always asking text-book authors to write for a global studentaudience, the place from which an authorwrites inevitably pokes through to declareitself again and again throughout a text. Thisplace declaration may be particularly strikingto a non-local reader. As I was readingGeography and Gender with my American ori-gins and perspective, I was often askingmyself, ‘why didn’t they seem to know aboutx, or why didn’t they mention y,’ where xand y were publications by American femi-nist geographers available when this bookwas written. Examples are the six-article spe-cial feature on women, published in TheProfessional Geographer in 1982 and Mazeyand Lee’s (1983) Her Space, Her Place, also abook aimed at undergraduates.


Upon its publication, Geography and Genderwas welcomed as providing a much-neededintroduction to feminist geography; the factof the book – its very appearance – was her-alded as much as was the contribution of itsmessage. Those who reviewed the book werefor the most part laudatory. Sallie Marston,reviewing for Urban Resources, a journalaimed at educational issues, praised theauthors for attempting ‘to lay out the foun-dation for a truly feminist geographic theory’(Marston, 1986: 60). In an appreciativereview appearing in Contemporary Issues inGeography and Education, Linda Peake (1989)hailed the book’s challenges to the traditionaldivisions within the discipline (i.e., thosebetween the economic, social, political, andso on). Risa Palm (1986) in Progress in Human

94ÿÿSUSAN HANSON


Geography, and Ron Johnston (1985) inBritish Book News, while generally positiveabout the book’s contributions, also foundfault: Palm was critical of ‘insufficient docu-mentation for a number of claims’ (Palm,1986: 466), and Johnston complained that ‘inplaces the geography is not as explicit as itcould have been’ (Johnston, 1985: 19). Theinitial reception within geography was on thewhole, however, strongly supportive of theseauthors’ efforts.

A thoughtful review by someone who wasnot familiar with feminist work in geographyand was a self-proclaimed lapsed geographer,appeared in Feminist Review (Neligan, 1985).Although grateful to this group of feministgeographers for taking on the job of providingstudents with an introduction to feministgeography and appreciative of their challengesto geography traditions, Annie Neligan wasdisappointed in the (in her opinion) elemen-tary level of the analysis. Of the discussionabout zoning and urban spatial structure, shesays, ‘I wanted the writers to go on to discussthe effects of this zoning on women’s lives andhow these designs have perpetuated the divi-sions of labor that they reflect. They becomeone of the factors which confine women topart-time work, to taking low-paid servicejobs that can be done near home. They add tothe stresses of combining a job with collectingthe children from school, so that home work-ing becomes not an attractive but the onlypossible option’ (Neligan, 1985: 115). Shebelieves that in the section on women anddevelopment the authors have not adequatelyilluminated how ‘international capital com-bine[s] with the patriarchy of traditionalfamily life to produce the micro-chip assem-bly lines worked by young, Southeast Asianwomen, and how … this feed[s] back into thelocation of investment’ (Neligan, 1985: 115). Ifind these laments fascinating because in herarticulation of what she finds missing inGeography and Gender Neligan is also puttingher finger on just a small portion of what we

have learned about gender and geographysince 1984.

Impacts and effects

The impacts of Geography and Gender can stillbe felt throughout the discipline, perhaps mostpalpably for students in geography curricula.In 1984, as now, there was considerable debateas to whether feminist geography should betaught in stand-alone courses or infused intocourses throughout the curriculum. Theauthors of Geography and Gender advocatedpursuit of both approaches, and my sense isthat both have taken hold, not only inAnglophone countries but in many othercountries as well. In the US, students havecome to expect their instructors to incorpo-rate feminist perspectives and gender into arange of human geography courses, and theycomplain if these are absent. Thanks in part toGeography and Gender, feminist analyses havebecome part of the geographical imaginationthat engage geography students worldwide; itis no longer acceptable to study space andplace with gender absent.

Because Geography and Gender is rooted inthe context of the UK in the early 1980s, ithas sparked lively and sometimes acrimoniousdebate among feminist geographers, especiallyin Britain, about the appropriate scope andfocus of feminist geography. In view of thedepth of the book’s challenges to traditionalgeography, such debate is entirely fitting.Indeed, one might argue that feminist geogra-phy, and Geography and Gender in articulatingthe cutting edge of feminist geographyin 1984, was a key initiator of the broadercritical assessment of geographical theory(especially its devotion to the status quo andresistance to change) that followed. Althoughthe book’s authors did not explicitly recognizethe importance of diversity among womenor the situatedness of a researcher, their work – byproviding a coherent, cogent, and compelling



extended argument for the necessity of femi-nist geography – laid the groundwork for thepost-structural approaches that are nowwidely practiced in the discipline. This out-come is ironic in that some of the authorshave resisted post-structuralism because ofthe difficulties it poses for political action.Nevertheless, a major reason for the lastinginfluence of Geography and Gender is, nodoubt, that the authors went on to havedistinguished careers in which they havecontinued to contribute to the developmentof feminist geography.

Indicative of this lasting influence, in 2001Geography and Gender was the subject of a‘Classics in Human Geography Revisited’focus in Progress in Human Geography; briefcommentaries by Robyn Longhurst andSusan Smith were accompanied by a collec-tive response from the authors, in which theyreflected on their 1984 book in light of sub-sequent developments. Both commentariesare laudatory, offering sympathetic critiquesof what seems, from today’s vantage point, tobe oddly missing from the book.

Robyn Longhurst praises Geography andGender as pathbreaking in three ways: provid-ing a book-length treatment of gender in thefield, undertaking a gendered analysis of sev-eral key topics in geography, and, in itscollective authorship, challenging ‘the politicsof academic authorship’ (Longhurst 2001:253). She also calls the book visionary forforeshadowing several issues that have sincebeen greatly illuminated by feminist geogra-phers: work on men and masculinities; theneed for feminist thinking to transcend disci-plinary sub-fields, and the infusion of criticaltheorizing. As issues that, from a contempo-rary perspective, seem jarring, Longhurstidentifies the authors’ treatment of sex andgender as if they are separable and their deci-sion to speak in one collective voice, asquestionable given this ignores differenceswithin feminism (in their response, the authors’rejoinder that they did recognize political

differences among feminists in their descrip-tion of the different approaches to feminism).

Susan Smith reflects on the book’s contri-bution: ‘while geographies without womenstill appear in the literature, the much greatersensitivity to gender and sexuality in the disci-pline’s major journals is probably as good amarker as any of what this book helpedachieve’ (Smith, 2001: 255). She commentsthat, from today’s vantage point, the emphasisin Geography and Gender on gender differences(and their social construction) to the neglectof differences/diversity among women seemslike a shortcoming. Were they writing today,the authors would pay more attention toemotion, love, caring work and would recog-nize that gender is not a stable category butone that is performed differently in differentcontexts. Noting that, in 2000, only 6% ofProfessors of Geography in UK were women,Smith concludes, ‘Things have moved on since1984, but not so far that this classic text onGeography and Gender has lost its cutting edge’(Smith, 2001: 257).

In their response to these commentaries,the authors’ assessment of their book 15 yearsafter publication is instructive. Above all, theystill believe that ‘the book’s central point,namely that understanding gender makes adifference to our geographical imaginations,remains as relevant today as it did then’(Smith, 2001: 257). Although the authorsnow recognize that their decision to writewith a collective voice did suppress the diver-sity in their life experiences and subjectpositions, they note that the decision itselfcame from a desire to emphasize that allknowledge production is collaborative and tostress their common ground as women(along with one man) in geography. To them,the most significant shifts in feminist scholar-ship since the mid-1980s have been ‘theepistemological and methodological implica-tions of recognizing diversity’ and the ‘newtheorization of sex and gender [that] have madespace for researching and reconceptualizing the

96ÿÿSUSAN HANSON


body’ (Smith, 2001: 258). As they furtherexplain regarding this last point, ‘Issues aboutsexuality and embodiment [in the early1980s] seemed only to cement women firmlyinto the “nature” side of the nature-culturebinary’ (Smith, 2001: 258). The authors arealso now more circumspect about their abil-ity to effect change, and this is the case for atleast two reasons: the imposition of neoliberalpolicies and the difficulty of uniting peopleagainst gender oppression in the face of themany dimensions of diversity among women(and men).

In closing, the authors of Geography andGender observe that, although more womenare now present, academic geography in theUK is still dominantly white, male, andunderpinned (perhaps now more than ever)by a masculinist, competitive ideology. They

bemoan the absence of feminist principles inUK academic life. Although it seems fair togive these authors the last word on their ownbook, I would not end on such a pessimisticnote. The very presence of more women andmore feminists in geography, and increasinglyin positions of power within geography andwithin individual academic institutions – notonly in the UK but in countries around theworld – signals hope for the kinds of positivechange that the authors of Geography andGender articulated so well in 1984.

Note

1 Sophie Bowlby, Jo Foord, Eleanore Kofman,Jane Lethbridge, Jane Lewis, Linda McDowell,Janet Momsen, John Silk, and JacquelineTivers.


Secondary sources and referencesFoord, J. and Gregson, N. (1986) ‘Patriarchy: Towards a reconceptualisation’, Antipode

18 (2): 186–211.Hanson, S. ( 2004) ‘Who are ‘we’? An important question for geography’s future’, Annals

of the Association of American Geographers 94 (4): 715–722.Johnston, R.J. (1985) ‘Review of Geography and Gender’, British Book News, January

1985: 18–19.Longhurst, R. (2001) ‘Commentary on Geography and Gender’, Progress in Human

Geography 25 (2): 253–255.Marston, S. (1986) ‘Review of Geography and Gender’, Urban Resources 3 (2): 60–62.Mazey, M.E. and Lee, D. (1983) Her Space, Her Place: A Geography of Women.

Washington, D.C.: Association of American Geographers, Resource Publications inGeography.

Neligan, A. (1985) ‘Review of ‘Geography and Gender: An Introduction to FeministGeography’, Feminist Review 20: 113–118.

Palm, R. (1986) ‘Review of Geography and Gender ’, Progress in Human Geography 10(3): 466–467.

Peake, L. (1989) ‘Review of Geography and Gender ’, Contemporary Issues inGeography and Education: The Journal of the Association for CurriculumDevelopment 3 (1): 87–90.

Smith, S. (2001) ‘Commentary on Geography and Gender’, Progress in HumanGeography 25 (2): 255–257.

Women and Geography Study Group of the IBG (1984) Geography and Gender: AnIntroduction to Feminist Geography. Harlow, Longmans Group.


SOCIAL FORMATION AND SYMBOLIC LANDSCAPE(1984): DENIS COSGROVE

David Gilbert

The Palladian country house and its enclosedparkland of sweeping lawns, artisticallygrouped trees and serpentine lakes offers asynthesis of motifs owing their origins to arange of sources: late renaissance Italy, classi-cal humanism, the literary pastoral and theseventeenth- century painters in Rome. Thefinest of these ‘landscapes’, the parks at Stowe,Stourhead, Castle Howard, Chatsworth,Blenheim or The Leasowes, have come to beregarded as representative almost of the verycharacter of the English countryside. From arather different perspective they represent thevictory of a new concept of landownership, bestidentified by that favourite eighteenth centuryword, property. The ideology of English park-land landscape may perhaps best be introducedby examining the design and iconography ofone example, among the earliest and best-pre served of all, the garden at Rousham inOxfordshire. (Cosgrove, 1984: 199)

Introduction

Rousham House stands about 12 miles northof Oxford in the English Midlands, on thewest bank of the River Cherwell. The housewas built in 1635, but Rousham is bestknown for its landscaped garden, designed byWilliam Kent between 1737 and 1740.Although open to the public, visitingRousham can be a little daunting. Unlikemany other country houses, Rousham doesn’thave a gift shop, nor does it have glossy

posters or interactive displays about itshistory and design. No dogs and no childrenunder 15 are admitted. The garden at Roushamprovides one of the key examples in DenisCosgrove’s Social Formation and SymbolicLandscape, first published in 1984, and at firstsight some students seem to find the bookequally daunting. While the book is elegantlywritten, flicking through it can be somewhatintimidating. This is a book with a big subject –the history of the idea of landscape in theWest – that, unlike most geographical texts, isunafraid to consider the long sweep of his-tory. The illustrations, of fine-art landscapepaintings from the Renaissance to the mid-twentieth century, or of the buildings andplans of the sixteenth-century Venetian archi-tect Andrea Palladio, give an indication of theseemingly difficult materials (certainly unfa-miliar to many geography students) thatCosgrove uses in the book. Like Rousham,Social Formation and Symbolic Landscape is awork that demands and repays serious con-sideration and detailed thought. And, likeRousham, it is also a work of carefully crafteddetails but with a clear overall design, a singlecentral argument that has been influential inand beyond cultural geography.

Looking back, Social Formation andSymbolic Landscape is important because itestab lished a new politicized notion oflandscape as a fundamental concept in humangeography. Cosgrove’s approach to the idea oflandscape is summarized in an often-quotedpassage at the start of the book:

12


100ÿÿDAVID GILBERT

The landscape idea represents a way ofseeing – a way in which some Europeanshave represented to themselves and toothers the world about them and theirrelationships with it, and through whichthey have commented on social relations.Landscape is a way of seeing that has itsown history, but a history that can beunderstood only as part of a wider historyof economy and society; that has itsown assumptions and consequences, butassumptions and consequences whoseorigins and implications extend wellbeyond the use and perception of land;that has its own techniques of expression,but techniques which it shares with otherareas of cultural practice. (Cosgrove,1984: 1)

It is worth looking closely at this passage.Landscape is, as Cosgrove (1984: 13) putsit, ‘an imprecise and ambiguous concept’without a single clearly agreed meaning. Ineveryday language, we may talk about land-scapes in the physical world – ‘the landscapesof New Zealand’ – or we may think aboutlandscape as a particular form of representa-tion, often in art – ‘the landscapes of JohnConstable’. Cosgrove’s book is about both, ashe argues that landscape art is one particularform of a more general ‘way of seeing’ theworld.

In this context, the phrase, ‘way of seeing’,has a specific meaning. It refers to theapproaches of Marxist art historians of the1970s and 1980s, and specifically to the workof the British cultural critic, John Berger,who presented a BBC TV programmeand wrote a subsequent book called Waysof Seeing (Berger, 1972) (though Berger him-self took the terminology from the earlytwentieth-century art historian ErwinPanofsky). This approach argued that artcould only be understood properly and fullyin relation to wider contexts of social andeconomic power. Art is therefore not simplyan individual creative activity, undertaken by

lone artists. Who is able to produce art, whatis defined as art, and how it is seen and used,are all questions that cannot be answeredwithout reference to broader power relations.Indeed, this approach argued forcefully thatthe very category of art as a separate mode ofhuman activity depended on particular formsof social and economic organization. Putmost simply, the central argument of SocialFormation and Symbolic Landscape is that thisunderstanding of the relationship betweensocial order and art is also directly applicableto the study of landscape.

The ordering of the two main terms inthe title is an indication of the influence ofMarxism on Cosgrove’s thinking. ‘SocialFormation’ is given precedence over thesymbolism of landscape. Cosgrove was atpains to distance his work from approachesthat treated ‘the landscape way of seeing in avacuum outside of the context of a real his-torical world of productive human relations’(Cosgrove, 1984: 2). However, if Cosgrove sethimself against approaches to landscapes thatinterpret them with abstract references totheir beauty, or as the work of individualgeniuses, the terminology of the title alsoindicates his dissatisfaction with cruder vari-eties of Marxism. In some versions ofMarxism, cultural products like literature,painting, poetry or drama were regarded assecondary, superficial aspects of human activ-ity, to be understood as determined by themore fundamental economic organization ofa society. In using the term ‘social formation’rather than ‘mode of production’ whichwas more commonly associated with thisdeterministic Marxism, Cosgrove positionedhimself alongside influential British Marxistsof the period, such as the historian EdwardThompson and the cultural critic RaymondWilliams. Such work rejected simplisticnotions of an economic base and culturalsuperstructure, and argued culture was not aby-product of more fundamental conflictsbetween social classes over economic resources,


DENIS COSGROVEÿÿ101

but was ‘an active force in the reproductionand change of social relations’ (Cosgrove,1984: 57).

Indeed, Social Formation and SymbolicLandscape can be seen as a developmentand extension of the arguments of one ofRaymond Williams’ best-known works, TheCountry and the City, published in 1973.Williams explored the representation of therural and the urban in English literature,arguing both that literature reflected funda-mental power structures, but also that cultureworked as an active force in society that ben-efited the capitalist classes. Williams focuseddirectly on the power of geography in cul-ture, particularly on the significance ofthe representation of the countryside and thecity; Cosgrove extended this to consider thelandscape both as visual art and as a directmanipulation of the physical environment.

Reviewing the book in 1987, JamesDuncan argued that it was the book’s sophis-ticated Marxist framework that gave it ‘thepotential to shift the direction of a subfield’(Duncan, 1987: 309). Later reviews have alsohighlighted the significance of the centralargument. Don Mitchell, reflecting on theinfluence of the book on the development ofcultural geography, stressed that without thisstrong argument situating the landscape wayof seeing in the ‘wider history of economyand society’, the book would have been‘an accomplished and compelling syntheticdescription of changes in the nature of land-scape representation … but would have donenothing to explain those changes’ (Mitchell,1999: 505). However, in his new introductionto the 1998 edition, Cosgrove stepped backfrom this strong explanatory framework, andfollowing a shift made by many in HumanGeography and beyond, exchanged his focuson capitalism and class relations for a moregeneralized and diffuse concern with thedevelopment of modernity. Perspectives suchas feminism and post-colonial studies havebroadened debates about the relations between

social power and cultural forms since the firstpublication of Social Formation and SymbolicLandscape (see Phillips, 2005; Pratt, 2005). Inreading the book more than 20 years after itspublication, we need to ask whether it is toobound into the intellectual culture and debatesof its own period to have much relevance now,or whether it should be treated as an importantstarting point for later debates about landscapein particular and culture more generally.

The book and its author

Denis Cosgrove is probably best known as oneof the main figures in what came to be knownas the ‘New Cultural Geography’ (see Lilley,2004 for more details of Cosgrove’s career).Alongside others such as Peter Jackson andJames Duncan, Cosgrove argued for a radi-calized cultural geography, often in directcontra-distinction to an earlier established tra-dition associated with the work and legacy ofthe American cultural geographer, Carl Sauer(see Cosgrove, 1993; Cosgrove and Jackson,1987). In Social Formation and Symbolic Landscape,Cosgrove did point to the importance of earlytwentieth-century cultural geographers such asSauer and Paul Vidal de la Blache in develop-ing perspectives that emphasized landscapes asmaterial products of human societies. However,Cosgrove, Duncan, Jackson and others criti-cized the way that cultural geography, certainlyas taught in universities in the US in the1970s and 1980s, had reduced this legacy to aset of rather dull empirical techniques, ofteninvolving the detailed analysis of surviving ‘tra-ditional’ landscapes (for discussions of the‘New Cultural Geography’ and its relationshipwith Sauer’s ‘Berkeley School’ see the debatesin Price and Lewis, 1993a, 1993b; Cosgrove,1993; Duncan, 1993; Jackson, 1993). The ‘NewCultural Geography’ combined Cosgrove andDuncan’s focus on the politics of landscapewith another strand of work that concentratedon the cultural politics and spatialities of



identity. In the 1980s, the influence of Marxistslike Raymond Williams remained strong; PeterJackson’s landmark 1989 textbook Maps ofMeaning, while extending its scope beyondclass to address the cultural geographies of raceand gender, remained firmly grounded inWilliams’ cultural Marxism. By the 1990s,however, this new tradition had become muchmore fluid in its understanding of the nature ofsocial power and its relations with culture – notleast in the work of those academics mostclosely associated with the beginnings of the‘New Cultural Geography’.

Social Formation and Symbolic Landscape wasDenis Cosgrove’s first book, written while hewas a Lecturer at Loughborough University.During his time in the East Midlands, Cosgrovedeveloped a close intellectual relationshipwith Stephen Daniels of the University ofNottingham. The kind of materialist and polit-ical readings of landscapes and landscape artpropounded in Social Formation and SymbolicLandscape became characteristic of an importantstrand of East Midlands Geography in the1980s and 1990s. Perhaps the best-known workin this tradition is the collection co-edited byCosgrove and Daniels in 1988, The Iconographyof Landscape. The collection brought authorsfrom art history, theology and literature togetherwith historical geographers in studies rangingfrom sixteenth-century Italy to twentieth-century Canada, all concerned to decipher thesocial power of various forms of landscapeimagery. Although the introductory essay isquite short, it develops significantly the argu-ments in Social Formation and Symbolic Landscape.There is a much stronger emphasis on themethod required to interpret landscapes andlandscape art, advocating Panofsky’s ‘deepiconography’. Pictures, and indeed actual land-scapes, were to be treated as encoded ‘texts’,closely examined to reveal not just the mean-ings that had been consciously put there by anartist, landowner or landscape designer. Thismethod of iconography was also a way ofrevealing intrinsic meanings about social power

structures that had shaped the making of thelandscape.

However, beyond his influence on the disci-pline, Denis Cosgrove has also been that ratherrare beast, a British geographer with significantstanding across the humanities, and also beyondthe English-speaking world. As such, whileSocial Formation and Symbolic Landscape is nowregarded within Geography as a period piece,important in its time and significant forredirecting the sub-discipline of cultural geog-raphy, it has had a much more active andlasting presence in disciplines such as arthistory and landscape history, where it hasretained currency for the significance of itscentral argument, but more for its detailedcase-studies of the landscape idea in Italy,North America and England. Some chaptersdraw upon Cosgrove’s doctoral work, a muchmore tightly focused project on the landscapesof post-renaissance Italy, particularly in Veniceand the Veneto in the sixteenth century. Thiswork culminated in the book The PalladianLandscape, published in 1993. Significantly,both this book and Social Formation andSymbolic Landscape have been translated intoItalian.

Reading Social Formation andSymbolic Landscape

It is probably best to approach Social Formationand Symbolic Landscape through the commentaryand reflections provided by Denis Cosgrovein his introduction to the 1998 edition. Thisis not a straightforward summary of the argu-ments, but places the book in the context ofsubsequent developments in cultural geogra-phy and landscape studies. Social Formationand Symbolic Landscape was also includedrecently in the ‘classics in human geography’series in the journal Progress in HumanGeography. Here Cosgrove replied to com-mentaries on the book by Lawrence Bergand James Duncan, summarizing some of the



key points from his 1998 introduction (Berget al., 2005) A third route into the arguments ofthe book is to look at Cosgrove’s article‘Prospect, perspective and the evolution ofthe landscape idea’, published in 1985. Thisarticle overlaps with parts of the book, but con-centrates on one aspect of its argument,considering how the ‘invention’ of moderntechniques of perspective in the Italian renais-sance contributed to the exercise of power overspace through the landscape ‘way of seeing’.

Social Formation and Symbolic Landscape canbe treated as having two parts. The first part,consisting of the introduction and chapterson ‘the idea of landscape’ and ‘landscape andsocial formation’, is best thought of as a longessay expounding the theoretical themes ofthe book, while Chapters Three to Nineform detailed historical studies. The intro-duction itself is (unsurprisingly) the bestplace to start, but there are also sections thatrepay close attention in the second chapter.A key passage (‘landscape and perspective’,pp. 20–27) considers the power of perspec-tive. In 1435, the Florentine architect Albertipublished Della Pittura (or On Painting) inwhich he set out the principles of ‘linear’ or‘single point’ perspective. These principleshave formed the basis of what we mightdescribe as a ‘realist’ representation of three-dimensional space, and were dominant inWestern art between the renaissance and thelate-nineteenth century. To our eyes earlymedieval paintings appear flat and distorted,while later paintings appear much more life-like or photographic in quality. These laterpaintings appear more real because they useconverging lines to focus on a single point,the eye of the painter and the eye of theimagined observer.

Cosgrove connects Alberti’s ‘way of seeing’to the power structures of renaissance Italy,and particularly to the emergence of newforms of capitalist organization in the city-states of Tuscany. His argument is that artisticperspective, like parallel developments in

accountancy, navigation, surveying, mappingand the science of artillery, helped form whathe describes as the figure of the bourgeoisindividual, the powerful man placed at thecentre of a world that he saw, ownedand ordered. Cosgrove quotes Leonardo daVinci’s comment that the use of perspective‘transforms the mind of the painter into thelikeness of the divine mind, for with a freehand he can produce different beings, ani-mals, plants, fruits, landscapes, open fields,abysses and fearful places’ (Cosgrove, 1985:52). However, Cosgrove takes this further tosuggest that in this distanced, often aerialview of landscape, sometimes described asthe ‘sovereign eye’, it is ultimately not theindividual painter that takes the role of God,but those who own and control the land.

In the remainder of his chapter on the‘idea of landscape’, Cosgrove extends thisargument about power and ways of seeinglandscape to include modern geography asit developed as a discipline from the late-nineteenth century onwards: ‘in some respectsgeography’s concept of landscape may beregarded as the formalising of a world-viewfirst developed in painting and the arts into asystematic body of knowledge claiming thevalidity of a science’ (Cosgrove, 1984: 27). Ina key section that has strong resonances forcurrent debates in the discipline about repre-sentation (see Rose et al., 2003), Cosgroveargues that Geography has suffered from a‘visual bias’ which has meant that it has beenobsessed with surveying, mapping and repre-senting landscapes, rather than interrogatingthem as the constructions of particularsocial formations. He praises the HumanisticGeography of the 1970s and early 1980s foremphasizing the emotions and sensationsof human beings living within landscapes,but argues that this cannot be enough. Thetask for the critical geographer is to ‘tracethe history of the landscape way of seeingand controlling the world’ (Cosgrove,1984: 38).



The following chapter (‘Landscape andSocial Formation: Theoretical Considerations’)starts this task with a very broad discussion ofthe long-term history of social and economicdevelopment in the West. It’s a difficult chapter,which centres on contemporary debates aboutthe transition from feudalism to capitalism. Thekey points are: firstly, that this transition in thefundamental economic, social and politicalorganization of Western societies was a long-term process; secondly, that it took place atdifferent rates and in different forms in differ-ent countries; and thirdly that struggles overthe ownership and control of land were centralto the transition. These ideas are rather moreeasily understood in the historical case studiesthat follow in the second part of the book.These chapters consider, in turn, the relation-ships between social formations and landscapein renaissance Italy, in sixteenth-century Veniceand the Veneto region, and in North Americaand England during the seventeenth and eigh-teenth centuries. Two final chapters explore thelandscape way of seeing in the nineteenth andtwentieth centuries. Each chapter has a broadlysimilar form, discussing first the dominantforms of landscape art, architecture and designin the period, before drawing out their basis inbroader social conflicts and their use withinthose struggles.

It’s useful to go back to Rousham foran example of this method in action.Cosgrove gives a detailed description of thelandscape garden created by William Kent bythe Cherwell (Cosgrove 1984: 199–206). Likemore conventional art historical accounts ofthe garden, Cosgrove identifies the influenceof painting on Kent’s work, particularly thework of Claude Lorrain and Nicholas Poussin.Kent created a ‘Claudian landscape’ in ‘thereality of lawns, trees and controlled viewswithin and beyond the garden’ (Cosgrove,1984: 201). Kent’s garden was designed to beseen and experienced in a carefully con-structed circuit that combined details such asstatues and ornamental bridges in the gardenwith views out into the English countryside

beyond. However, as well as examining thedetails of the garden, Cosgrove also interpretsthe designs at Rousham as part of widerdevelopments in the cultural politics of land-scape. The English landowning class used artand landscape design, together with certain lit-erary forms, to express its position and power.The phrase ‘commanding view’ is often usedloosely today to indicate a wide panoramafrom a high point in the landscape; Cosgroveshows that the constructed views of the greatEnglish parks were literally both commandingand commanded.

However, Cosgrove goes further than thisrather blunt assertion that Rousham was alandscape of power, and much of the lastingvalue of the case study comes from the morenuanced way that he places the garden in theprecise context of the cultural politics of theperiod. Rousham’s constructed ‘natural’ land-scape, with its irregular shapes and serpentinelines, was a direct challenge to earlier formaland geometric landscape gardening as atVersailles, and as such could be treated as acelebration of English bourgeois ‘liberty’ asagainst continental absolutism. The gardenalso combined classical elements (filteredthrough Palladian landscape conventionsdeveloped in the sixteenth-century Veneto)with views of a supposedly unadornedEnglish countryside. This combination validatedthe taste and sensibility of the landowner, butalso placed the beauty of the English landscapewithin a longer narrative of the rise of Westerncivilization. What Cosgrove achieves is a deepreading of the landscape at Rousham that givesdue attention to artistic creativity and culturaltraditions, but also to the economic and socialmilieu within which they worked.

Critiquing Social Formation andSymbolic Landscape

All of the historical chapters of Social Formationand Symbolic Landscape combine subtle read-ings of the politics of landscape with an



underlying Marxian framework that givesexplanatory priority to changing class relations.While the detailed case studies retainconsiderable force and interest, criticisms ofthe book, including Denis Cosgrove’s ownreflections, concern this underlying argu-ment. At one level, this is part of broaderdevelopments in human geography and insocial theory more generally that have chal-lenged Marxist frameworks of understanding,whether in their more overtly deterministforms, or the more flexible, humanistictradition underpinning Social Formation andSymbolic Landscape. Rather than rehearsethese general critiques here, it is better tothink about how they translated into morespecific challenges to the arguments put for-ward by Cosgrove in 1984.

These challenges question the ultimatepriority given to class in Social Formationand Symbolic Landscape. Subsequent work,particularly that drawing on feminist andpost-colonial perspectives, has argued that alandscape way of seeing cannot be understoodwithout references to other dimensions ofsocial power, particularly those associated withgender and race. This challenge was madeforcefully by Gillian Rose in Feminism andGeography (Rose, 1993: particularly 86–112).One of Cosgrove’s strongest claims is thatlandscape is a ‘visual ideology’, taking a specificand partial world-view, and presenting it as anatural and necessary way of seeing. This class-bound view of the world erased the actualeconomic exploitation that made it possible,particularly of the rural working class. Roseextends this argument to claim not only that asimilar erasure of the inequalities of gendertakes place in the landscape way of seeing, butalso that Cosgrove’s account itself reinforcesthis through the almost complete absence ofconsideration of gender in the text. AsCosgrove acknowledges in his introduction tothe 1998 edition, the artists and landownersdiscussed in Social Formation and SymbolicLandscape are not just all male, but also ‘theyappear and communicate to us as eyes, largely

disconnected from any other corporeal orsensual aspects of their being’ (Cosgrove, 1998:xviii). Put another way, while all of Cosgrove’sintellectual effort goes into revealing how theunacknowledged power of social class struc-tured the work of these men, he has nothingto say about the impact of their masculinity.

This is a serious omission, openly recog-nized by Cosgrove in 1998. Although SocialFormation and Symbolic Landscape was writtenbefore the expansion of feminist thinking thattook place in Geography in the late-1980s, theimportance of gender relations in ‘ways of see-ing’ was already an important area of debate inthe social sciences and humanities more gen-erally. Berger’s Ways of Seeing (1972) includeda substantial section on the power relations inthe visual representation of women by and formen. At about the same time, Laura Mulveypublished her classic essay on the power of the‘male gaze’ in cinema (Mulvey, 1975). Evenworking within the confines of a Marxistapproach, such issues could and should havebeen woven into the argument of SocialFormation and Symbolic Landscape, as the prop-erty relations discussed were clearly gendered.Gillian Rose, however, takes her critique fur-ther, arguing not merely that Social Formationand Symbolic Landscape ignores the materialexploitation of women and their exclusionfrom the making of the landscape way of see-ing, but also that Cosgrove fundamentallymisreads the idea of the sovereign eye. ForRose, the central feature of this view oflandscape is that it is a dominant masculinegaze. Cosgrove in response is more cautiousabout this critique, resisting the claim that ‘thelandscape idea inevitably constructs genderedlandscapes as the passive, feminized objects ofa rapacious and voyeuristic male gaze’ whileacknowledging the need for more attentionbeing devoted to the role of sexual desire inlandscape representation (Cosgrove, 1998:xviii).

A second criticism of Social Formation andSymbolic Landscape is also connected to its cen-tral emphasis on class relations and property,



and relates to post-colonial theory. While thebook gives close attention to the significanceof the ‘New World’ of North America in thedevelopment of the European landscape idea,there is no consideration of the way thatEurope’s position within wider economic,political and cultural networks shaped ideasabout landscape. This is in part about anotherform of erasure; for example, the greatEnglish estates discussed in the book were asmuch the product of imperial expansion andslavery abroad as they were of class powerwithin England. Social Formation and SymbolicLandscape does not consider the export andhybridization of European landscape ideas inother environments as a part of the imperialproject. And the book also fails to considerthe ways that European landscapes werealtered by the flow into Europe of newimages and knowledge of ‘exotic’ environ-ments, and particularly by the cultivation inEurope of plants from beyond its boundaries.

A final possible criticism of Social Formationand Symbolic Landscape is emphasized byCosgrove himself (Cosgrove, 1998: xx–xxi;Cosgrove and Duncan, 2005: 481–482). Theoverall narrative of capitalist developmentused as the basis for the book suggests thatby the twentieth century land and land-ownership had ceased to be a central arena ofsocial and economic conflict. As a consequence,landscape was seen as of waning significanceboth in art, and more generally as a politicaland moral issue. Cosgrove’s turn from thisparticular reading of the history of capitalismtowards a more diffuse and complex concernfor the making of the modern world calls thisclaim into question at a theoretical level. But,as Cosgrove suggests, this argument was alsoa misreading of the significance of landscaperepresentation in modern times. By focusingon some schools of landscape art in the mid-twentieth century, he was able to argue thatthis was a form in decline, an increasinglyirrelevant backwater in the development ofmodern art. This seriously underplayed the

importance of landscape art in the develop-ment of the modern nation-state, whichhas subsequently been an important focusfor cultural geographers (see Daniels, 1993;Matless, 1998). Furthermore, while Cosgroverecognized the work in the 1970s and1980s of some landscape artists like RichardLong, he did not foresee the emergence ofnew kinds of landscape art that have dis-rupted the traditional form of the gaze, andwhich crucially have responded to newforms of environmental and identity politics(Nash, 1996).

Conclusion

Social Formation and Symbolic Landscape remainsan important read today not so much becauseof its importance in the development of cul-tural geography, but because its central theme –the power of landscape and the landscape wayof seeing – still matters. Few geographers,including Cosgrove himself, now accept theparticular formulation of social power pre-sented in the book, and its omissions areserious (although the way that it opened up,for example, discussion of the ‘sovereigneye’ in Geography made it easier for otherswith feminist or post-colonial perspectivesto develop the argument). Despite its over-whelming concern with landownership andclass relations, Social Formation and SymbolicLandscape was subtle enough to includetantalizing passages about the significanceof embodied experience and of the impactof myths and collective memory, both ofwhich have been important themes inrecent cultural geography. The challengeinspired by reading Social Formation andSymbolic Landscape today is to imagine anew formulation that connects the widerhistory of economies, societies, nations andempires to the study of landscape, as a wayboth of seeing the world and of being inthe world.



Secondary sources and referencesBerg, L., Cosgrove, D. and Duncan, J. (2005) ‘Commentaries and author’s response in

“Classics in human geography revisited”: Social Formation and Symbolic Landscape,’Progress in Human Geography 29: 475–482.

Berger, J. (1972) Ways of Seeing. London: Penguin.Cosgrove, D. (1983) ‘Towards a radical cultural geography: problems of theory’, Antipode

15: 1–11.Cosgrove, D. (1984) Social Formation and Symbolic Landscape. Beckenham, Kent:

Croom Helm.Cosgrove, D. (1985) ‘Prospect, perspective and the evolution of the landscape ideal’,

Transactions of the Institute of British Geographers, New Series, 10: 45–62.Cosgrove, D. (1993) ‘On “the reinvention of cultural geography” by Price and Lewis’,

Annals of the Association of American Geographers 83: 515–517.Cosgrove, D. (1994) ‘Contested global visions: One-World, Whole-Earth, and the Apollo

Space Photographs’, Annals of the Association of American Geographers 84:270–294.

Cosgrove, D. (1998) Social Formation and Symbolic Landscape (second edition) withnew introduction. Madison, WI: University of Wisconsin Press.

Cosgrove, D. and Daniels, S. (1988) The Iconography of Landscape. Cambridge:Cambridge University Press.

Cosgrove, D. and Jackson, P. (1987), ‘New directions in cultural geography’, Area 9: 95–101.

Daniels, S. (1993) Fields of Vision: Landscape Imagery and National Identity in Englandand the United States. Princeton: Princeton University Press.

Duncan, J. (1987) Review of Social Formation and Symbolic Landscape’, Annals of theAssociation of American Geographers 77: 309–311.

Duncan, J. (1993) ‘On “the reinvention of cultural geography” by Price and Lewis’, Annalsof the Association of American Geographers 83: 517–519.

Jackson, P. (1989) Maps of Meaning: An Introduction to Cultural Geography. London:Unwin Hyman.

Jackson, P. (1993) ‘On “the reinvention of cultural geography” by Price and Lewis’,Annals of the Association of American Geographers 83: 519–520.

Lilley, K. (2004) ‘Denis Cosgrove’, in P. Hubbard, R. Kitchin and G. Valentine (eds), KeyThinkers on Space and Place. London: Sage.

Matless, D. (1998) Landscape and Englishness. London: Reaktion. Mitchell, D. (1999) Review of Social Formation and Symbolic Landscape (second edi-

tion). Transactions of the Institute of British Geographers 24: 505–506.Mulvey, L. (1975) ‘Visual pleasure and narrative cinema’, Screen 16 (3): 6–18.Nash, C. (1996) ‘Reclaiming vision: Looking at landscape and the body’, Gender, Place

and Culture 3: 149–169.Phillips, R. (2005) ‘Colonialism and postcolonialism’, in P. Cloke, P. Crang and

M. Goodwin (eds), Introducing Human Geographies. London: Hodder Arnold.Pratt, G. (2005) ‘Masculinity-femininity’, in P. Cloke, P. Crang and M. Goodwin (eds),

Introducing Human Geographies. London: Hodder Arnold.Price, M. and Lewis, M. (1993a) ‘The reinvention of cultural geography’, Annals of the

Association of American Geographers 83: 1–17.


Price, M. and Lewis, M. (1993b) ‘Reply: On reading cultural geography’, Annals of theAssociation of American Geographers 83: 520–522.

Rose, G. (1993) Feminism and Geography: The Limits of Geographical Knowledge.Cambridge: Polity.

Rose, G., Matless, D., Driver, F., Ryan, J. and Crang, M. (2003) ‘Intervention roundtable:Geographical knowledge and visual practices’, Antipode 35: 212–243.

Williams, R. (1973) The Country and the City. London: Chatto & Windus.



CAPITALIST WORLD DEVELOPMENT (1986):STUART CORBRIDGE

Satish Kumar

Accounts of development and under develop-ment must be sensitive not only to the dynamicof the world system, but also to the relations ofproduction and their conditions of existencethat are the characteristics of particular coun-tries and development sectors. (Corbridge,1986: 154)

Introduction

Geographers have long documented howprocesses of uneven development have servedto separate the haves from the have-nots.Global capitalism and neo-liberal reforms haveworked to strengthen such divisions. InCapitalist World Development, Stuart Corbridge,a geographer and a development studies spe-cialist who has worked extensively in India,sought to document the spatial imprints ofuneven development and also provide a cri-tique of radical geographical theories thatsought to explain such processes. As such, thebook needs to be understood as a product ofa particular time, as a reaction to the Marxisttheories of development that dominated geo-graphical scholarship and the social sciences atthe time of its writing. Corbridge thus wrotethe book as a critique of radical developmentgeography and its emphasis on epistemologyand ideology rather than real-world experi-ences. In such Marxist accounts, the State isreified and economistic explanations only helpto essentialize conclusions about the develop-ing world. He calls for a dialogue between

Marxists and non-Marxists to provide arealistic assessment of underdevelopment andprovides a clear exposition of the intellectualorigins and arguments of key radical devel-opmental theories. In so doing, he astutelyclarifies complex arguments and concepts.The result is a book that is neither pro-capitalist nor anti-Left. It is an attempt to gobeyond trading dogmas and determinism inexplaining the persistence of underdevelopmentin the world, providing a focused introduc-tion to the world of radical developmentgeography.

Radical geography emerged as a reactionto growing social inequalities in the West,including concern over environmental issues,racism and global inequality. There was grow-ing unease with Western foreign policy,particularly the Vietnam fiasco, and concernfor the plight of the developing world. Whilesome radical geographers shied away fromMarxist theory, others actively engaged withits ideology and doctrine. Throughout the1970s and early 1980s debate continuedabout the form and nature that radical geog-raphy should take – for example, scientificversus critical Marxism, theoretical or histor-ical materialism, and structuralism versushumanism (see Duncan and Ley, 1982, for anextended analysis). In time, a materialistontology came to dominate Marxist geogra-phy, one in which capitalism came to beviewed as the source of all worldly evils – notleast inequality and unevenness of development.This marriage of Marxism and space produced

13


110ÿÿSATISH KUMAR

exciting and compelling explanations of globalinjustice by highlighting how the persistence ofuneven development is as important to capital-ism as the direct exploitation of labour (seeCastree, Chapter 8 this volume; Phelps, Chapter10 this volume).

The book’s key arguments

Capitalist World Development marked anattempt to challenge this dominant Marxistmode of an analysis, building on and extend-ing critiques such as those developed bySlater (1976) and deSouza and Porter (1976).These authors, like Corbridge, asserted thatneo-Marxist geography as then formulatedwas incapable of resolving developmentalproblems and by the middle of the 1980sscholars like Booth (1985) and Corbridgewere attempting to understand the nature ofimpasse in Marxist approaches to develop-ment and underdevelopment. In particular,Capitalist World Development critiqued thedogmatism and essentialism associated withMarxist development theories, such as theprimacy of ‘structures’ and the transcendentalqualities of ‘capital’, calling for a change ofthe ‘explanadum’ (that which requires expla-nation) to provide a solution to the impasse indevelopment studies. In doing so it sought tomove beyond the limitations of neo-Marxistand Marxists approaches without necessarilycondemning their transformatory ambitions.Corbridge showed how questions of debtand demography are influenced by not justMarxist theory and ideology, but also withthe material reality of the Third World.

Capitalist World Development starts with twobasic questions, namely, ‘What are the conse-quences of adopting a radical perspective ondifferential development?’ and ‘What are theconsequences of adopting a view of eventswhich emphasize a necessary conflict ofinterest between metropolitan capitalism andthe development of the periphery?’ The

answers were spelt out by Corbridge asfollows:

• Capitalism cannot or will not promotedevelopment of the Third World.

• Capitalism alone is ultimately responsiblefor the world’s demographic and environ-mental ills.

• Capitalism is incapable of promoting inde-pendent industrialization of the South.

• The fundamental divide shaping the cap-italist world system is between the Northand South.

In providing these answers Corbridge shiftedthe focus away from structuralism to empha-size the human face of structural constraintsin world capitalist development. As such,overcoming the tendency to oppositionism,determinism, spatial over-aggregation, andepistemological confrontation was critical toCorbridge (1986: 9–13) in setting in motiona new agenda for development studies.

Oppositionism relates to a confrontationalapproach to a given argument, which becomescounter-productive when it refuses to acknowl -edge logical and relevant arguments. Forexample, the term ‘capitalism’ has been radical-ized to provide a negative connotation, namelythat of promoting underdevelopment in theThird World. This is all the more true whenwe evaluate the perspective that populationgrowth rates are irrelevant to patterns of differ-ential development. In fact, population growthmay not necessarily cause underdevelopment(an argument illustrated by the growth of theeconomies of India and China in the twenty-first century). This can also be substantiatedwhen we address the deterministic claim ofradical development geography that capitalismis incapable of promoting industrialization ofthe Third World. Current world develop-ments do not bear out this proposition. ThusCorbridge’s (1986) claim that capitalismis capable of promoting industrialization ofthe periphery is particularly relevant today,


especially when we see the emergence ofthe newly industrialized economies (NIEs) ofSingapore, Malaysia, South Korea, Taiwan andHong Kong and newly industrializing coun-tries (NICs) of China and India.

Determinism was identified by Corbridge asanother major problem in the radical approachto development geography. This relates tocapitalist models of development and under -devel opment. Here capitalism can only pro moteunderdevelopment in the periphery as it islargely based on an exploitation of both pro-ductive and unproductive resources. Thesuggested way to overcome this exploitation,according to the Marxian analysis, is to disen-gage economically with the developed world.

The third critique developed by Corbridgewas that of spatial over-aggregation wherebythe capitalist system is conceived as being oneof north versus south or core versus periphery.Such forms of over-aggregation have tended tonormalize and ignore the variety of capitalistforms of development and underdevelopment.The different shades of capitalism visible todayrelate to the diverse forms that neo-liberalismhas articulated itself in space. Indeed, Corbridge’scritique anticipates and predates the new ideasabout the constantly transforming forms ofcapitalism current in the literature (see Peckand Tickell, 2002).

The fourth critique concerned the epistemo-logical confrontation resorted to by radicaldevel opment geographers. The lack of engage-ment between Marxists and non-Marxists hasarguably resulted in a ‘dialogue of the deaf ’(Chisholm, 1982: 11; Corbridge, 1986: 12). Here,there is a clear disjuncture between theory, ide-ology and actual practice. The privileging andover-emphasis on economistic explanation rein-forced essentialist and dogmatic conclusions,which were far removed from ground reality.Such a position has resulted in the failure of theMarxists and neo-Marxists to acknowledge thenegative role that any given State can play inpreventing development, particularly throughcorruption and reinforcing elitism. They have

also ignored the potential of endogenous growthin the periphery. Corbridge asserts that dismiss-ing an argument as less important for notsubscribing to the radical viewpoint does dis-service to academic engagement. The bookemphatically states that a non-progressive viewof capitalism dominated radical geography and a‘dialogue of the deaf ’ is of no use to the poorin the developing world whose future dependson the policies guided and informed by theo-ries of development and underdevelopment(Corbridge, 1986: 13). By evaluating the role ofcapitalism in fostering development and under-development, Corbridge offered a way out ofthe deterministic mode of thinking and sug-gests that there are no general laws of capitalistdevelopment which can become the basis forexplaining the persistence of differential unevendevelopment in the world. Accordingly, the cen-tral question should be how far capitalism can beheld responsible for underdevelopment in theworld?

To support his argument, Corbridge pro-vides a brilliant critique of Harvey’s (1974)paper on ‘Population, resources and the ideol-ogy of science’. While accepting Harvey’sincisive critique of Malthusian and evolution-ary ideas of population growth, Corbridgeargues Harvey’s relentless epistemologicaldogmatism did not develop non-Malthusianperspectives on the population-developmentdebates. Failure to engage with practical popu-lation problems and their politics of control inthe developing world are clearly important and,according to Corbridge, were not acknowl-edged by the radical development geographers(Corbridge, 1986: 87–103).

The discussion on neo-Malthusian issuesis still topical and relevant today, particularlyin the context of sustainable development. The recent emergence of India and China asemerging world economic powers puts to restall speculative ideas and ideologies of cata-strophism inherent in Malthusian accounts.Indeed as Corbridge notes, ‘rapid populationgrowth at times exerts a secular influence upon

STUART CORBRIDGEÿÿ111


economic growth’ (1986: 84). This isconfirmed by recent analyses emerging fromSouth Asia, which overturn the old adage that‘the potency of population growth is a nega-tive factor in India’s economic performance’(see Corbridge, 1986: 90). The argument hereis that demographic transition in selectiveregions actually aided the resurgence of eco-nomic growth in India. Despite the populationpressure, India’s economic growth has beenconsistently based on specific structural adjust-ments and liberalization of the economy. Thus,Corbridge’s (1986) question whether it ispossible (and/or desirable) to curb populationgrowth to promote a wider restructuringof the economy is answered in the affirmativein the case of China and India. However, whatis important to bear in mind is that boththese countries have an established populationmomentum and this does not come to a haltovernight. So, despite the one child per familypolicy in China and demographic transitionin southern Indian states, population growthhas not come to a grinding halt. Therefore,Corbridge’s (1986: 100) assertion that ‘takentogether these accounts leave us in little doubtthat a decline in fertility can be secured inadvance of radical structural changes, as themoderniser claims’, is generally vindicated bycurrent experiences.

The issue of climate change, also discussedin the book, has also become more topicalnow than ever and Corbridge refocuses theterms of debate across both the Left and Rightof the political spectrum. The idea that theorigins of environmental maladies are trace-able to the inefficiencies of capitalism is adeeply contested argument, riddled with ide-ological rhetoric, and remains a potent sourcefor debate among the radicals. Globalizationhas advanced a replication of productionforms and exchanges between the developedand the developing world, thereby overturn-ing the notion that less advanced nations areincapable of securing any lasting competitiveadvantage in the world of manufacturing,

services and trade. Indeed, while the existenceof capitalism is not disputed, the TNC-ledindustrialization espoused by the neo-Marxists for the developing world is passé incurrent globalized context. Today the terms ofreference have shifted from the dirigismic state toan entrepreneurial state and finally to a ‘pater-nal state’. Notions that industrialization in theThird World is incompatible with capitalism(Corbridge, 1986: 131) now appear very dated,given nation states are vying with emergentmarkets to seek a fairer share of the global pie.

The book makes an important assertionthat ideas of stagnation had no roots inMarxism and, in this context, the claimsmade by radical geographers in the contextof the periphery appear rather dubious (Peet,1975; Blaut, 1974). The persistence of under-development in the periphery can thereforebe apportioned to the internal contradictionsin the Third World, such as policy blundersand fiscal mismanagement, as much due tostructural constraints as to development. Inline with this thought the moral critique ofcapitalism needs to be retained, particularlyin the context of poverty and inequality inthe Third World (Kumar, 2004).

In making such arguments, Corbridge pro-vided a strong critique of neo-Marxists’ lack ofappreciation of the role of government in theThird World in assisting capitalist development.In a way the conclusion that Corbridge statesstill holds true, namely that ‘accounts ofdevelopment and underdevelopment must besensitive not only to the dynamics of the worldsystem, but also to the relations of productionand their conditions of existence that are thecharacteristics of particular countries anddevelopment sectors’ (Corbridge, 1986: 154).New forms of unequal exchange and asymme-tries of power are being orchestrated despitethe establishment of WTO. The global eco-nomic system remains embedded, therebyperpetuating inequality and disparity. In a sense,Corbridge’s critique anticipated contemporarycalls to analyse issues of neo-liberalism,

112ÿÿSATISH KUMAR



which was more contextual and relational inits form and functioning, rather than amonolithic concept. He emphasized nationstates as intrinsically empowered to deter-mine the course of development within theirboundaries. This of course is subject to qual-ification, in that functional democracy makesthis possible wherever accountability isdevolved into the hands of the citizens.

Critiques and legacy

Some of the critiques of Corbridge’s workseemed largely guided by prejudice rather thanfactual argument (Booth, 1993). Indeed, somegeographers who have reviewed this book havebeen highly critical (see Progress in HumanGeography, 2005). For instance, virulent out-bursts labelling the book as ‘centre-right wingpolemic writing in praise of monopoly capital-ism and against radicals’ (Blaut, 1989: 102)did not necessarily offer a fruitful engagementwith the book’s argument, but certainlyexposed the deep sense of insecurity in radicalscholarship. Blaut states ‘Corbridge adds little tothis standard argument of conservative socialscience … Radical geography in fact ratherbadly needs constructive criticism, but youwon’t find it in this volume’. Peet (1988: 190)stated, ‘in brief, radical development geographyis not criticised in terms of its own content butindirectly via various radical social theorieswhose connection with geography is unprovenand largely unexamined’. Further, he adds, ‘Thisis not a critique of radical development geogra-phy, which occupies no more than five percentof the text. On the few occasions geography ismentioned, simple caricatures are drawn, whichreveal more about Corbridge than their objects’(Peet, 1988: 191). Again, Johns (1990: 180)states, ‘the book’s title is a bit misleading. Thetext fails to engage either the literature orthe reality of Capitalist World Development; nor isit a sustained critique of radical developmentgeography’.

Gaile (1988), on the other hand, stated [t]his book is pro-scholarship and is not pro-capitalist’. Likewise Brookfield (1987: 119) notes,‘that modern currents in non-Marxist devel-opment writings are inadequately treated …[but] the writing style is very lucid and theauthor has taken great care in his efforts toclarify complex and sometimes turgid argu-ments for his readers’. In presenting a detailedand critical analysis, the book reflects theextent of Corbridge’s empathy with the devel-oping world.

While the book never got its full dues onpublication, it is clear that it helped move devel-opment studies beyond its impasse by outliningkey themes in post-Marxist theory in the wakeof Perestroika and its related Glasnost in theSoviet Union. Corbridge has since engagedwith issues of world debt, the power and role ofthe state (see Corbridge and Harriss, 2000,Reinventing India, and Corbridge et al., 2005.Seeing the State). He has hence provided acoherent and important contribution to thecritical debates of world capitalist developmentwhich is contingent and reflective of real issuesin the developing world. Thus, in the currentcontext his book may appear eclectic and crit-ical of his contemporaries, but it cannot befaulted for being reductionist and dogmatic.

For a current readership it will be moreappropriate to test his ideas in the wake ofthe recent development and transformationsin the peripheral economies such as India andChina. One has also to bear in mind that cap-italism is constantly evolving and changing tosuit the imperatives of world development.This book does much to translate difficult andobtuse ideas and concepts for non-specialists.It sets the terms of critical engagement, whichhas been followed by others in the radicalgenre. Corbridge asserts that one needs togo beyond dogmas and determinism todeconstruct the view that capitalism and its lawsof motion are not fixed in time and space.Here, the exigencies of time and space areconstantly mediated in the peripheries by


114ÿÿSATISH KUMAR

conditions of population, economic growth,state policies, governance, etc. and thereforeare not controlled by the inexorable tide ofthe ‘grand world system’. The book accord-ingly highlights the internal incoherence ofmany of the radical theories adapted by radi-cal development geographers such as Blaut(1970, 1974), Santos (1974), Slater (1976), andPeet (1975), to suggest an epistemologicaldenouncement of development geographers istantamount to throwing the baby out with thebath water and at no point does Corbridgesuggest such a course of action. As far as is dis-cernible, he calls for a realistic revision of therhetoric that ideology supersedes theorybased on evidence from the field. Indeed ‘aradical perspective need not be so pessimistic’(Corbridge, 1993: 213).

The emerging consensus regarding devel-opment in the periphery from both thenon-Marxists and those on the ‘Left’ is thattraditional ideas of capitalism suggest thespontaneous diffusion of development fromcore to the peripheries (see Baran’s thesis,1957), yet what we observe today is not aspontaneous diffusion but a protracted strug-gle, enforced by the post-colonial statethrough instruments of protectionism,import substitution industrialization and thegradual dismantling of the ‘dirigisme’ phaseof development. The case in point is that

diffusion of industrial development fromthe core to the periphery has not meant anall-round general development of the peo-ple in the developing world. Indeed, thevery sustainability and effectiveness of theseforms of development is highly suspect(Patnaik, 2006).

Conclusion

Regardless of reviewer responses, CapitalistWorld Development has become a standardreference in teaching modules dealing withissues of development across the spectrum ofsocial sciences. Corbridge’s critical insightspresent a necessary counterbalance to theperspectives provided by radical developmenttheorists. The relevance that the book mighthave in the future is difficult to predict.Suffice to say that capitalism as an entity hasnot dissipated over time, neither has under-development or poverty disappeared over theedges of global-local horizons. As long as weponder and grapple with the crisis of accu-mulation in both the core and peripheries,as long as we engage with the impossibilityof a singular approach to capitalism, of theimpasse in development studies, CapitalistWorld Development will remind us of the needto avoid the ‘dialogue of the deaf ’ at all times.

Secondary sources and referencesBaran, P.A. (1957) The Political Economy of Growth. New York: Monthly Review Press. Blaut, J. (1970) ‘Geographic models of imperialism’, Antipode 2: 65–85.Blaut, J. (1974) ‘The ghetto as an internal neo-colony’, Antipode 6: 37–41. Blaut, J. (1989) ‘Review of Capitalist World Development: A critique of radical development

geography’, The Professional Geographer 41: 102–103.Booth, D. (1985) ‘Marxism and development sociology: Interpreting the impasse’, World

Development 13 (7): 761–787.Booth, D. (1993) ‘Development research: From impasse to a new agenda’, in

F. Schuurman (ed.) Beyond the Impasse: New Directions in Development Theory.London: Zed Books, pp. 49–76.



Brookfield, H. (1987) ‘Review of Capitalist World Development: A critique of radicaldevelopment geography’, The Journal of Development Studies 24 (1): 118–119.

Chisholm, M. (1982) Modern World Development: A Geographical Perspective. London:Hutchinson.

Corbridge, S. (1986) Capitalist World Development: A Critique of Radical DevelopmentGeography. London: Macmillan.

Corbridge, S. (1993) ‘Marxisms, modernities and moralities: development praxis and theclaims of distant strangers’, Environment & Planning D: Society and Space 11: 449–472.

Corbridge, S. and Harriss, J. (2000) Reinventing India: Liberalisation, Hindu Nationalismand Popular Democracy. Cambridge: Polity Press.

Corbridge, S. (2005) ‘Classics in human geography revisited, Capitalist world develop-ment: a critique of radical development geography’, Progress in Human Geography,29 (5): 601–608. London: Macmillan.

Corbridge, S., Williams, G., Srivastava, M. and Veron, R. (2005) Seeing the State:Governance and Governmentality in India. Cambridge: Cambridge University Press.

deSouza, A.R. and Porter, P.W. (1976) ‘Development geography and radical–liberaldialogue’, Antipode 8 (3): 94–102.

Duncan, J. and Ley, D. (1982) ‘Structural Marxism and human geography: A criticalassessment’, Annals of the Association of American Geographers 72 (1): 30–59.

Gaile, G.L. (1988) ‘Review of capitalist world development: A critique of radical develop-ment geography’ Political Geography Quarterly 7 (4): 373–377.

Harvey, D. (1974) ‘Population, resources, and the ideology of science’, EconomicGeography 50: 256–277.

Johns, M. (1990) ‘Review of Capitalist World Development: A critique of radical develop-ment geography’, Antipode 22 (3): 177–180.

Kumar, M.S. (2004) ‘Rhetoric of globalisation and reforms: A case of India (1991–2003)’,in C.G. Krishnaswamy, S.R. Keshava and R.M. Tirlapur (eds), Better Expression onGlobalisation. Bangalore: Bangalore University Press, pp. 23–52.

Patnaik, P. (2006) ‘Diffusion of development’, Economic and Political Weekly, May 6:1766–1772.

Peck, J. and Tickell, A. (2002) ‘Neoliberalising space’, Antipode 34 (3): 380–404. Peet, R. (1975) ‘Inequality and poverty: A Marxist-geographic theory’, Annals of the

Association of American Geographers 65: 564–571.Peet, R. (1988) ‘Review of ‘Capitalist World Development: A critique of radical develop-

ment geograpshy’, Economic Geography 64 (2): 190–192.Santos, M. (1974) ‘Geography, Marxism and underdevelopment’, Antipode 6: 1–9.Slater, D. (1976) ‘Anglo-Saxon geography and the study of underdevelopment: Critical

notes on the emergence of a new tendency’, Antipode 8 (3): 88–93.


GLOBAL SHIFT (1986): PETER DICKEN

Jonathan Beaverstock

As its title suggests, the perspective of thisbook is global. It aims to describe and toexplain the massive shifts which have beenoccurring in the world’s manufacturing industryand to examine the impact of such large-scalechanges on countries and localities across theglobe. … Ultimately, the main thread whichbinds the various parts of the book together isthat of the effects of global industrial change.(Dicken, 1986: i)

Introduction

Peter Dicken, Emeritus Professor at theSchool of Environment and Development,the University of Manchester, has been one ofthe major ‘movers and shakers’ in debates inboth economic geography and globalization.He has been at Manchester for over fourdecades, was awarded his Personal Chair in1988 and has held distinguished positions atuniversities in Australia, Canada, Hong Kong,Mexico, Singapore and the United States. In1999 he was invited to become a Fellow ofthe Swedish Collegium for Advanced Studyin the Social Sciences and was awardedthe Royal Geographical Society-Instituteof British Geographer’s Victoria Medal in2001, followed by an Honorary Doctorate inPhilosophy from the University of Uppsala,Sweden in 2002. Ongoing editorial positionson international journals such as Competitionand Change, Journal of Economic Geography,Global Networks, Progress in Human Geographyand Review of International Political Economy

provide a continuing mark of his esteem inthe discipline (see Beaverstock, 2004).

Peter Dicken experienced a neo-classicaleconomic geography upbringing in the late1960s, spawning such ‘classics’ as Location inSpace: A Theoretical Approach to EconomicGeography (authored with Peter Lloyd in 1972).This book became one of the most significanttexts of the period using classical and neo-classical theorists such as Christaller, Isard andLosch to explain the role of locational analysisin explaining the spatial organization ofregional economic development. It was duringthe early 1970s, however, that Dicken began toquestion the orthodoxy of neo-classical loca-tional theory and the deductive approachesthat followed in the wake of the ‘quantitativerevolution’ in economic geography. Influencedby the behavioural backlash to the ‘quantita-tive revolution’ in economic and humangeography, Dicken became seduced by man-agement research in behavioural science andorganizational studies (now termed, broadly,‘Strategy’). Dicken’s inductive research frame-work paved the way for his pioneeringapproach to studying the role of businessdecision-making processes in shaping the globaldistribution of economic activity (Yeung andPeck, 2003). The study of multinational enter-prise (and corporations [MNCs]) very quicklybecame the central tenet of Dicken’s workduring the 1970s and early 1980s in a contextof rapid MNC international restructuring anda concomitant deindustrialization processwhich wrought havoc with Western manufac-turing industries, not least in the United

14


Kingdom’s industrial heartlands in the Northand Midlands. Dicken’s highly influential writ-ings of these times (e.g. Dicken, 1971, 1976,1977, 1980; Lloyd and Dicken, 1972/76),quickly became the forerunners to arguably hismost significant contribution to economicgeography: explaining international economicchange through the global behaviour and strat-egy of transnational corporations (TNCs)in a rapidly changing world. In essence, thisresearch became the foundation for Dicken’sseminal work, Global Shift: Industrial Change ina Turbulent World (1986).

Perhaps the most interesting analysis of thegenesis of Global Shift comes from the authorhimself. He noted that the book took abouttwo years to complete after being started inabout 1984, and, ‘at that time, “globalization”,as either a focus for geographical research or asa popular topic, barely existed’ (Dicken, 2004a:513). Dicken was convinced that in order tounderstand industrial restructuring and territo-rial development at the regional scale, one hadto seek explanations from what was going onat the world scale, particularly through an eval-uation of the organizational strategies of TNCs.This was very true for Dicken’s own ‘local’work on textiles and engineering in the NorthWest of England, an area being systematicallydeindustrialized and influenced significantly byworld events, for example, competition fromthe Newly Industrializing Countries (NICs),the rise of new ‘enabling’ technologies andchanging MNC/TNC behaviour as firmssought low-cost locations for production. Notsurprisingly then, Global Shift was about man-ufacturing industries – hence the subtitleIndustrial Change in a Turbulent World. AsDicken (2004a: 514) himself noted, ‘GlobalShift was essentially a book about the globaltransformation of manufacturing industriesand its effects on employment’. The mostintriguing aspect of Dicken’s thinking in thewriting of Global Shift was his decision to notuse the word ‘Geography’ in the title or to

make any claim to be a geographer on his part.As he subsequently lamented:

… [a]lthough the book was fundamentallygeographical, I took the decision at theoutset not to use the word ‘geography’ inthe title nor even to divulge my identity asa geographer. In some ways I now feel alittle ashamed of having done that. On theother hand, at that time – and, to someextent, this is still the case – most peoplewould not have taken such a book writtenby a ‘mere geographer’ very seriously. Soit proved. Sad though it may seem, I haveno doubt that part of the reason the bookcame to be used and accepted across awide range of social science disciplines, asa research and teaching book, was thatit was approached without disciplinarypreconceptions. (Dicken, 2004a: 514)

Dicken’s subsequent work has elaboratedmany of the themes in Global Shift, where the‘big picture’ of worldwide economic restruc-turing is often used as a gateway to explainindustrial change, and regional and territorial(re)development. Three major examples cometo mind. First, Global Shift’s corporation-basedapproach to economic restructuring has beenused by Dicken as a forerunner for a signifi-cant corpus of conceptual and empirical workon ‘webs on enterprise’ or business networkswithin and between transnational corpora-tions (see, for example, Dicken and Thrift,1992). Second, Global Shift was the catalyst forresearch on Japan in terms of the organiza-tional strategies of the soga shosha (tradingcompanies) and patterns of foreign directinvestment in Western Europe and the USA(see for example, Dicken and Miyamachi,1998). Third, Dicken’s latter-day research the-orizing the significance of Global ProductionNetworks in the world has its roots very muchin the Global Shift discourse (see, for example,Coe et al., 2004). In fact, the influenceof Global Shift is omnipresent in many of

118ÿÿJONATHAN BEAVERSTOCK


Dicken’s writings on globalization andterritorial re(development) (see, for example,Dicken, 2004b).


The beauty of Global Shift, which helpedestablish it as a key text in human geography,is its clarity of argument and rigorous inter-rogation of rich, in-depth empirical sourcesand examples taken from the economic geo-graphies of everyday corporate life pre-themid 1980s. Dicken argued that the rapidtransformations of industrial activity sincethe end of the Second World War was facili-tated by three major forces:

• the growth, internationalization andorganizational strategies of TNCsworldwide;

• the role of nation-states and national gov-ernments through trade, investment,regional development and macroeconomicpolicy;

• the revolutionary impact of enablingtechnologies in transport, communica-tion, production, organization andinternationalization.

Combined, these three ‘eclectic’ explanatoryforces were stitched together by Dicken to pre-sent a series of very persuasive perspectives thatoutlined why we should now think in terms ofan interconnected and interdependent globaleconomy rather than a merely inter-nationaleconomy (Dicken, 1986). Part One – ‘Patternsand Processes of Global Industrial Change’ – isnot just about economic change as orchestratedby the geoeconomic power of TNCs, but alsothe role of the nation-state through instrumentsof macroeconomic policy, FDI incentives, tradepolicies (for example, GATT), the power ofregional blocs (for example, the EEC), Japan, andimportantly, the NICs (Newly IndustrializedCountries).

In Part Two of the text, ‘The Picture ofDifferent Industries’, Dicken brings forwardthe theoretical writings in Part One toexplain international restructuring and terri-torial development in specific case studysectors: textile and clothing; iron and steel;motor vehicles; and electronics. Each chaptercarefully explains the major processes andpatterns of industrial restructuring and employ -ment change in both a detailed and non-technical fashion. The book is completedwith a series of chapters discussing the‘Stresses and strains of global industrialchange’ by reviewing the costs and benefits ofTNCs on host countries and the effects ofglobal economic change on the Western eco-nomics, the NICs and the Third World.

Initial reactions to the publication of GlobalShift during the late 1980s were, on balance,very welcoming. Book reviews championed itsinternational perspective, student-user friendlyapproach (particularly for the novice economicgeographer) and uncomplicated, straight-talking conceptual explanations, drawn primarilyfrom writers such as John Dunning, StephenHymer and Raymond Vernon (see forexample reviews by Jones, 1986 and Wise,1987). Krumme (1987: 132) lauded its eclecticapproach in introducing, ‘contemporary inter-national dimensions to modern economicgeography’, but was slightly concerned thatDicken had underplayed the significance ofinternational trade and had not used theopportunity in the case study chapters to focuson natural resource industries such as oil, tin,copper and global agricultural MNCs. Sharp(1987: 649) commented that the book was‘ambitious’ and drew the reader’s attentiontowards shortfalls in Dicken’s appraisal of therole of technological change in driving restruc-turing within MNCs. But perhaps the moststinging critique came from Richard Peet’s(1988) review in Progress in Human Geography.Peet suggested that the book’s ‘deliberate’eclectic approach was its major downfall

PETER DICKENÿÿ119


because it had no significant underlying struc-tural theory to explain its eclectic elements andtheir inter-relationships: ‘[S]o this useful book,on a topic of vital significance in the first andthird worlds, lacks a theory of social structureand reaches no general conclusions’ (Peet,1988: 152).

The significance of Global Shift

Peter can be accredited with puttingglobalization on the agenda in economicgeography. (Yeung and Peck, 2003: 2)

Global Shift … is one of human geogra-phy’s minuscule number of ambassadorialtexts. (Olds, 2004: 510)

In 2004, the journal Progress in HumanGeography earmarked Global Shift as one of its‘Classics in human geography revisited’ withcritical evaluations by Kris Olds (2004) andRay Hudson (2004) (with a response fromPeter Dicken himself, as noted above). Olds(2004) recognized that the very quick successof Global Shift was due to three main factors:timing; content; and style. As for timing, Oldsnoted that Global Shift was written at a timewhen international economic restructuringon a world scale was having significantimpacts in local and regional economies, par-ticularly in the North where debates aboutthe New International Division of Labour(NIDL) (Frobel et al., 1980), branch planteconomies and foreign direct investment(FDI) and deindustrailization were verymuch in vogue. What Global Shift achieved inits content was a weaving together, possiblyfor the first time, of many of the majordebates about the NIDL, TNCs, FDI, etc. ina context of international restructuring on aworld scale. In terms of style, I would supportOlds (2004: 509) claim, ‘that the level ofabstraction that Dicken employs plays a crit-ical role in the book’s attractiveness toprofessors (and especially students)’. Global

Shift’s well-balanced and informative approachbetween theory (of the internationalizationof TNCs) and empirics (a case study approach),both allied to the role of the nation-state, hasproduced a textbook which is highly-relevantand provides ‘joined-up thinking’ whenexplaining local and regional economicchange born from a world scale perspective.

For Hudson (2004) the attractiveness andsignificance of Global Shift was evidenced infive key characteristics. First, it focused on themajor actors that facilitated global economicchange; nation-states, TNCs and enablingtechnology. Second, it displayed an overtly‘geographical’ approach to explaining eco-nomic change on a world scale, teasing outthe subtleties of uneven development. Third,it explained how firms’ organizational formsand strategies (in different industrial sectorsof the economy) generated multifaceted geo-graphies of production, which, when combined,illustrated very clearly that processes of eco-nomic globalization were complex, unevenand highly interrelated. Fourth, it highlightedthe ‘footloose’ tensions between TNCs andthe nation-state as firm strategy continued toseek low cost locations for production. Fifth,it was an extremely informative textbook,drawing upon a wide range of ideas and datain tabular, figure and map form.

In the edited text Remaking the GlobalEconomy, Jamie Peck and Henry Wai-chungYeung (2003) pay homage to Peter Dickenand particularly the first edition of GlobalShift. They argue that ‘Dicken’s most signifi-cant intervention of the 1980s was thepublication of his first single-authored bookGlobal Shift … [which] … can claim to beone of the pioneering globalization texts’(Yeung and Peck, 2004: 9). Both of theseauthors note that the main contributions ofGlobal Shift emanated from three majordebates and issues at the time of writing thebook: firstly, the NIDL and associated worldindustrial transformation; secondly, the glob-alization discourse; and thirdly international



business, drawn from management science. Inessence, Dicken brought a critical geographicaleye to the writing on the NIDL, followingsomewhat in the wake of Massey’s (1984)Spatial Division of Labour (see Phelps, Chapter10 this volume). Global Shift brought muchclarity and rich empirical firm- and sectoral-specific studies to illustrate how and why thecorporate strategies of TNCs reproduced thegeographies of the NIDL. Moreover, Dickenwas able to combine the Fordism and post-Fordism discourses with a subtle explanationof the organizational strategies of TNCs toidentify the major process and patterns ofglobal production on a world scale in the1980s.

Yeung and Peck (2003) accordingly recog-nize that Dicken’s Global Shift has made asignificant contribution to the globalizationdiscourse/debate. Dicken was at the forefrontof putting the ‘global’ into the agenda of eco-nomic geography by studying TNCs, theirinternational strategies, and their effects onemployment and territorial development.Global Shift transcended the local and regionalscales in both concept and empirical practiceand thereby was at the forefront of implicitlyadvocating the global-local discourse throughits passionate process-led explanations ofuneven global economic change broughtabout by such actors as TNCs. The immensevalue of Global Shift as a globalization text isevident by the fact that ‘… [i]t is also one ofthe few geographical studies cited in comple-mentary works on globalization by otherprominent social scientists [such as] Hirst andThompson, 1996; Held et al., 1999’ (Yeungand Peck, 2003: 13). Global Shift followed inthe footsteps of a number of key texts onTNCs (e.g. Taylor and Thrift, 1982, 1986).But, according to Yeung and Peck, this iswhere the similarities between Global Shiftand its competitors ended because GlobalShift fully engaged with the major writersfrom the management school traditionof international business and MNC/TNC

strategy (e.g. Dunning, 1988). Dicken’s skillwas his ability to explore the growth, inter-nationalization and differentiated productionsystems of TNCs through a geographer’s dis-tinctive gaze through the lens of internationalbusiness theory.

Conclusion

The longevity of Global Shift (1986) is registeredin the fact it has already gone to three other edi-tions: Global Shift: The Internationalization ofEconomic Activity (1992); Global Shift: Transformingthe World Economy (1998) and Global Shift:Reshaping the Global Economic Map (2003). Whilsteach edition has added significant value fromthe previous one in terms of building theoreti-cal understanding and empirical quality, thespine of the meta-narrative remains just as inno-vative in the 2003 edition as it did in 1986. The1992 edition brought services into the equationwith a dedicated chapter focused on their inter-nationalization and significance in making theworld economy. In addition, in Dicken’s appraisalof the processes of economic change, moreemphasis was given to the network of relation-ships that exist within and between TNCs. Bythe 1998 and 2003 editions, Dicken had refinedand embellished more the processes of globalshift and in doing so produced one of the clas-sic explanations accounting for the growth,shape, internationalization and network organi-zational form of TNCs in the contemporaryworld economy. The 2003 edition saw the addi-tion of a new chapter on the distributionindustries; in January 2007, Sage published the5th Edition of the text: Global Shift: Mapping theChanging Contours of the World Economy, withcompanion resources available at the websitewww.sagepub.co.uk/dicken.

Notwithstanding the initial critical maulingsfrom Peet (1988) and others, Global Shift hasstood the test of time and been instrumental inthe development of many key strands of eco-nomic geography since its initial publication.

PETER DICKENÿÿ121


Four, of many, come to mind. First, Global Shiftshaped the way geographers and other socialscientists embraced the globalization discourse.One cannot underestimate the role the bookplayed in moving this debate/process forwardinto the 1990s. Second, Global Shift very clearlyillustrated the dynamic and changing role ofNICs in the world economy, particularly Asiantiger economies, including Japan. Third, GlobalShift lauded the ‘actor’ approach (the TNC) inunderstanding economic change in the worldeconomy. Dicken’s close analysis of interna-tional business systems, combined with a spatialperspective, has provided the forerunner forcontemporary work on such topics as: produc-tion (and commodity) chains, organizational

networks and firm–buyer relationships. Fourth,linked to the third, Global Shift has fosteredrecent work on Global Production Networksand relationality in economy (see for example,Coe et al., 2004; Dicken et al., 2001). But, morethan anything else, Global Shift has encourageda global sense of space and place whenexplaining the global-local economic nexus inunderstanding contemporary maps of worldactivity. In fact, Nigel Thrift’s strap-line for therecent edition sums up Global Shift’s influenceon geography, ‘Global Shift just keeps on get-ting better. There is no other source thatgives you the full story of globalization insuch a fluent and authoritative way. Not justrecommended but essential’.


Secondary sources and referencesBeaverstock, J.V. (2004) ‘Peter Dicken’, in P. Hubbard, R. Kitchin and G. Valentine (eds),

Key Thinkers on Space and Place. London: Sage, pp. 108–112.Coe, N., Hess, M., Yeung, H.W.C., Dicken, P. and Henderson, J. (2004) ‘Globalizing

regional development: A global production networks perspective’, Transactions of theInstitute of British Geographers 29 (4): 468–484.

Dicken, P. (1971) ‘Some aspects of the decision-making behaviour of business organi-zations’, Economic Geography 47 (3): 426–437.

Dicken, P. (1976) ‘The multiplant business enterprise and geographical space’, RegionalStudies 10 (4): 401–412.

Dicken, P. (1977) ‘A note on locational theory and the large business enterprise’, Area9 (2): 139–143.

Dicken, P. (1980) ‘Foreign direct investment in European manufacturing industry’,Geoforum 11 (2): 289–313.

Dicken, P. (1986) Global Shift: Industrial Change in a Turbulent World. London: Harper & Row.Dicken, P. (1992) Global Shift: The Internationalization of Economic Activity. London:

Paul Chapman Publishing.Dicken, P. (1998) Global Shift: Transforming the World Economy. London: Paul Chapman

Publishing. Dicken, P. (2003) Global Shift: Reshaping the Global Economic Map in the 21st Century.

London: Sage.Dicken, P. (2004a) ‘Author’s response’, Progress in Human Geography 28 (4): 513–515.Dicken, P. (2004b) ‘Geographers and globalization: (yet) another missed boat’,

Transactions of the Institute of British Geographers 29 (1): 5–26.Dicken, P. and Miyamachi, Y. (1998) ‘From noodles to satellites: The changing geogra-

phy of Japanese sogo shosha’, Transactions of the Institute of British Geographers23 (1): 55–78.


Dicken, P. and Thrift, N. (1992) ‘The organization of production and the production oforganization: why business enterprise matter in the study of geographical internation-alisation’, Transactions of the Institute of British Geographers 17 (1): 101–128.

Dicken, P., Kelly, P.F., Olds, K. and Yeung, H.W. (2001) ‘Chains and networks, territoriesand scales: towards a relational framework for analysing the global economy’, GlobalNetworks 1 (1): 99–123.

Dunning, J.H. (1988) Explaining International Production. London: Unwin Hyman.Frobel, F., Heinrichs, J. and Kreye, O. (1980) The New International Division of Labour.

Cambridge: Cambridge University Press. Hudson, R. (2004) ‘Commentary 2’, Progress in Human Geography 28 (4): 511–513.Jones, P. (1986) ‘Book review: Global Shift’, Geography 71 (4): 377.Krumme, G. (1987) ‘Book review: Global Shift’, Environment and Planning A 19 (1): 132–133.Lloyd, P.E. and Dicken, P. (1972 and 1976) Location in Space (1st and 2nd editions).

New York: Harper & Row.Massey, D. (1984) The Spatial Division of Labour. Basingstoke: Macmillan.Olds, K. (2004) ‘Commentary 1’, Progress in Human Geography 28 (4): 507–511.Peck, J. and Yeung, H.W.C. (eds) (2003) Remaking the Global Economy. London: Sage.Peet, R. (1988) ‘Book review: Global Shift’, Progress in Human Geography 12 (1): 151–152.Sharp, M. (1987) ‘Book review: Global Shift’, International Affairs 63 (4): 649.Taylor, M. and Thrift, N. (eds) (1982) The Geography of Multinationals. London: Croom

Helm.Taylor, M. and Thrift, N. (eds) (1986) Multinationals and the Restructuring of the World

Economy. London: Croom Helm.Wise, M.J. (1987) ‘Book review: Global Shift’, The Geographical Journal 153 (2):

274–275.Yeung, H.W.C. and Peck, J. (2003) ‘Making global connections: A geographer’s per-

spective’, in J. Peck and H. Wai-chung Yeung (eds) (2003) Remaking the GlobalEconomy. London: Sage, pp. 3–23.

PETER DICKENÿÿ123


THE CONDITION OF POSTMODERNITY (1989):DAVID HARVEY

Keith Woodward and John Paul Jones III

The experience of time and space haschanged, the confidence in the associationbetween scientific and moral judgements hascollapsed, aesthetics has triumphed overethics as a prime focus of social and intellec-tual concern, images dominate narratives,ephemerality and fragmentation take prece-dence over eternal truths and unified politics,and explanations have shifted from the realmof material and political-economic ground-ings towards a consideration of autonomouscultural and political practices. (Harvey,1989: 328)

Introduction

David Harvey’s (1989) The Condition ofPostmodernity: An Enquiry into the Origins ofCultural Change is more than a key text ingeography: its popularity and significance isunmatched outside of the discipline. A fewminutes with Google™ Scholar will affirmthat no academic book discussed in this vol-ume has been as widely cited as Condition ofPostmodernity. For less systematic evidence,students need only visit the offices of profes-sors who work in other fields; they’ll oftenfind that Condition of Postmodernity is theonly book by a geographer on their shelves.The forcefulness of Harvey’s argumentled Terry Eagleton, the renowned Marxistliterary theorist, to offer the followingassessment, printed on the back cover ofCondition:

Devastating. The most brilliant study ofpostmodernity to date. David Harveycuts beneath the theoretical debatesabout postmodernist culture to reveal thesocial and economic basis of this appar-ently free-floating phenomenon. Afterreading this book, those who fashionablyscorn the idea of a ‘total’ critique hadbetter think again. (Eagleton, 1989: np)

But the book itself is only part of the storyof its popularity. Another is that strange con-juncture of intellectual thought, culturaltrends, economic transformations, and politi-cal developments that in the 1980s came tobe known as ‘postmodernism.’ It is hard forthose whose intellectual awakening came inthe late 1990s or later to have a sense of thatera – of the immediacy of opportunities anddangers it seemed to present – but considerthis: for several hundred years something thatcame to be called ‘modernity’ developedapace. And then, like tracking the changingtemperature of time itself, there emerged awidespread feeling that modernity’s cher-ished moorings – a faith in human rationalityand logical communication, in economic,political, and social progress, in science, tech-nology, and aesthetic coherence, and in justand ethical systems of valuation and judg-ment – were being unhinged to such anextent that the world, especially the West, wasentering a new era. Though Harvey draws onhis always-keen geographical imagination inanalyzing postmodernism, his account of

15


126ÿÿKEITH WOODWARD AND JOHN PAUL JONES III

these shifts goes to issues much larger thanthe discipline of geography, and this helpsexplain why his book has been so widelyread. Condition of Postmodernity touched achord to which academics of many stripeswere attuned.

This essay on Condition of Postmodernity cov-ers its argument, impact, and critical reception.But before we move forward, there are threepreliminaries to address. First, Condition ofPostmodernity is unlike the books by Harveythat immediately preceded (Harvey, 1982 – seeCastree, Chapter 8 this volume) and followed(Harvey, 1996) it, both of which he nearlyabandoned in frustration, in that it nearly‘wrote itself.’ Harvey reports that the writingcame so easily that it ‘poured out lickety-split’(Harvey, 2002: 180). Harvey’s previous workon the urbanization of capital, on the history ofSecond Empire Paris and modern Baltimore,and on space and time within dialectical mate-rialism were foundational for Condition ofPostmodernity. The book amplifies an analysis ina 1987 essay he published in the radical geog-raphy journal, Antipode, in which he argued, inline with an earlier essay by Fredric Jameson(1984), that ‘post-modernity is nothing morethan the cultural clothing of flexible accumu-lation’ (Harvey, 1987: 279). That piece endedwith a challenge befitting Harvey’s intellectualdebt to Marxism: ‘A critical appraisal … of thecultural practices of postmodernity … appearsas one small but necessary preparatory steptowards the reconstitution of a movement ofglobal opposition to a plainly sick and troubledcapitalist hegemony’ (Harvey, 1987: 283). It wasHarvey’s friend and long-time editor, JohnDavey, who persuaded him to make thatappraisal.

A second thing to know is that Conditionof Postmodernity – along with Jameson’sdenser but similarly minded Postmodernism, or,the Cultural Logic of Late Capitalism (1991) –greatly contributed to making its objectof analysis passé. This is not to say thatthere were not already strong critics of

postmodernism, and perhaps even a sense ofintellectual and cultural exhaustion, by 1989.As Harvey (1989: ix) noted in his Preface:‘When even the developers tell an architectlike Moshe Safdie that they are tired of it,then can [The philosophers] be far behind?’Harvey’s own strategy was to historicize, locate,and explain postmodernism, and there are fewthings more disabling for a movement thatfancies detached moorings. After Condition,as Eagleton noted, the foundations of anapparently free-floating phenomenon wereestablished. Wind out of sails, the ship wasgrounded; postmodernism’s themes live on,but under different banners.

Third, in our opinion Condition ofPostmodernity should be read in conjunctionwith the text that followed it, Justice, Natureand the Geography of Difference (1996). A muchmisunderstood book, Justice complements itspredecessor: by specifying a suite of ontolog-ical questions that lay dormant in Condition ofPostmodernity; by laying out in clear detail adialectical analytic that underwrites Harvey’sapproach to explanation; and by respondingto critics of Condition of Postmodernity byaddressing its widely acknowledged sublima-tion of gendered and raced social relations.Harvey’s Justice also presages the currentinterest by geographers and others in ethicsand responsibility, and on how to theorizethe relationship between culture (social life)and the natural environment. Whereas thebook we describe here is largely critique,Justice helps readers understand more fullyhow that critique is grounded, while respond-ing to lapses in some of Condition ofPostmodernity’s arguments.

In writing about postmodernism, Harveyonce affirmed Pierre Bourdieu’s injunctionthat, ‘Every established order tends to producethe naturalization of its own arbitrariness’(quoted in Harvey, 1987: 279). For all itsuncertainty, multiplicity, and disorderliness, theage of postmodernism was very nearly one ofan ‘established order.’ It was Harvey’s mission


DAVID HARVEYÿÿ127

to unmask that naturalization. In the process,Condition of Postmodernity became part of thetime–space conjunction it analyzed, furthernaturalizing the book as, in Eagleton’s terms,devastating and brilliant.

The argument

For many reasons, it is important to readCondition of Postmodernity as a text devotedprincipally to the critique of a system ratherthan the promotion of a coherent alternative.While it is undeniable that Harvey neverstrays from the project of spatializing thepolitical economy, a move that spans almostthe entirety of his oeuvre, here the positive,revolutionary, and utopian aspects of Marxismare sent to the background, as something thatsits behind, but nevertheless frames, the topicat hand. True to its title, the book provides acritical analysis of economic and cultural con-ditions specific to and definitive of the lastquarter of the twentieth century. Importantly,Harvey recognizes that a double meaninghides within the idea of ‘condition’: it signalsthe state of actual, existing things surroundingus and making up the world but, at the sametime, it also indicates the historical tendenciesdriving global processes. Put another way,‘condition’ implies both the condition of thingsand that which conditions things. It is within thisdouble formulation that surfaces the begin-nings of the ontological development thatwill come to maturity in Justice, Nature andthe Geography of Difference: a condition is, atonce, a state of Being and a process ofBecoming.

Historically, the ‘condition of postmoder-nity’ is said to have developed out of (or tohave been a break with) the vast collection ofWestern philosophical, artistic, and scientifictheories that developed during the periodknown as ‘modernism.’ Though beginningwith the Enlightenment, this historical eragained ground through the establishment of

scientific positivism; the growth, spread, andtechno-practical coherences of industrialcapitalism; and the development of thedemocratic state form. These were in no waydiscrete historical events or processes; rather,each informed the other. Moreover, theyhelped map out a human-centered worldaimed at the development of free andautonomous human agents: rational eco-nomic citizens naturally embracing science,capitalism, and democracy. By the mid-twentieth century, however, the ideals ofmodernity had been pushed into crisis by theincreasingly glaring inequalities that accom-panied the development of capitalism and bythe ever-greater alienation fostered by theviolence and devastation of the two WorldWars. Artists, philosophers, and even scientistsincreasingly turned to fragmented, alienated,and relativistic representations of the world,revealing a growing dissatisfaction withappeals to the foundationalism that had beenthe cornerstone of modernist thinking.

Postmodernity met this discontent withseveral accounts of difference, positionality,and situatedness that appeared to ring thedeath-knell for aging visions of a worldrooted in essentialism, totalization, and uni-versality. One of the key moments in thistransformation was Jean-François Lyotard’sThe Postmodern Condition (1984) – fromwhich Harvey’s book derived its name and towhich it serves as a response. Here, Lyotardcalled for a rejection of the ‘grand narratives’of modernity, two of which were especiallysuspect in the postmodern critique: theassumed total autonomy of the individual(liberalist humanism, free market entrepre-neurialism), and the linear deterministicprogression of history (Marxist socialism, sci-entific progress). Thinkers such as MichelFoucault, Jacques Derrida, and Lyotard arguedthat such notions did not reflect any necessitywithin reality or the ‘nature’ of things somuch as the influence of power and dis-course in the ways we know and understand



the world (Dixon and Jones, 2004). Likewise,language, politics, and even identities becamematters not of universals, but of particularity,contingency, and difference.

Given this apparent break with the foun-dations of modernity, what possibilities remainfor the collective politics that Harvey andother Marxists find necessary for undermin-ing capitalism? His solution was to hold hisground, reanalyzing the relation betweenmodernity and postmodernity. He concludesthat the latter, in fact, does not represent abreak with the former, but rather its contin-uation, with changes marking adjustments totransformations in capitalist production andconsumption. For Harvey, modernity is insep-arable from the processes and institutionsdevoted to the accumulation of capital andthe utilization of labor, reaching its point ofinflection with the advent in 1914 of Fordism.Initiated by Henry Ford’s introduction of the‘five dollar, eight hour day as recompensefor workers manning the automated car-assembly line he had established’ (Harvey,1989: 125), Fordism was sealed in the post-World War II era as a social compact amongcapitalists, labor unions, and the social welfarestate. The macroeconomics of Fordism wasglobalized under the Bretton Woods agree-ment of 1944, which ‘turned the dollar intothe world’s reserve currency and tied theworld’s economic development firmly intoUS fiscal and monetary policy’ (Harvey, 1989:136). Accompanying this agreement was theopening of global markets to American cor-porate interests, and eventually Fordismbegan to spread throughout the globe.

By the mid-1960s, however, a number ofnational and regional markets had arisen tochallenge ‘United States hegemony withinFordism to the point where the BretonWoods agreement cracked and the dollar wasdevalued’ (Harvey, 1989: 141). Drawing fromhis earlier theories established in Limits toCapital (1982; see Castree, Chapter 8 thisvolume), Harvey points to the unraveling of

Fordism in the 1960s and 1970s: a system toorigid and contradictory to put off crises ofover-accumulation, it was inexorably beingsupplanted by a new, post-Fordist or flexibleform of accumulation. Flexibility was sectoralinsofar as capital was moved to invest in ser-vice industries; it was technical in the shifttoward more fluid labor agreements and out-sourcing arrangements; and it was geographicalin capitalism’s ever demanding need to ‘spa-tially fix’ its crises by mobilizing in ways thatlower costs, open new markets, and increaseprofits. Flexibility emphasized greater adven-turism on the part of the capitalist throughthe production of mobile, short-lived com-modities while, for the worker, whose ownlabor is sold as a commodity, this meant newforms of exploitation as promises of futureemployment were increasingly broken, whichin turn fostered increased transience and‘nomadism’ within the laboring class.

Critically, Harvey argues that, as the dis-tances and times it took to accumulate capitaland circulate commodities shrunk, our expe-rience of space and time similarly compressed.What is more, postmodernity’s rise at thisjuncture – as an intellectual, architectural,artistic, and cultural movement – was notcoincidental, for the sea-change called post-modernism is the direct result of theseexperiential dislocations. So, while previousrepresentations of postmodernity might haveargued that the moment was fundamentallythe product of cultural transformations (fromwhich economic changes, like the rise ofentertainment industries or the growth ofgentrification, then followed), Harvey’s analy-sis of the post-Fordist political economyturned this formulation upon its head.Making culture the shadow of economicprocesses, he explained in no uncertainterms that: ‘the emphasis upon ephemerality,collage, fragmentation, and dispersal in philo-sophical and social thought mimics theconditions of flexible accumulation’ (Harvey,1989: 302, our emphasis). Harvey illustrates this


DAVID HARVEYÿÿ129

causal reversal by examining key componentsof Western culture, drawing upon: (a) therecent history of the American city-scape,where he assesses several exemplary postmod-ern urban designs, including the spectacle-producing Disneyfication of Baltimore Harbor;and (b) the loss of depth, meaning, and historyin art and aesthetics, echoing a widespreademphasis on ‘the values and virtues of instan-taneity … and of disposability’ (Harvey, 1989:286) as capitalism moves from Fordist massproduction to flexibility.

Perhaps the most jarring aspect of Harvey’sargument is the suggestion that a great vari-ety of developments in recent progressivepolitics – such as feminism, anti-racism, andqueer activism – by virtue of their emphasesupon an apparently relativistic politics ofpositionality, proceed in the spirit of theserecent processes of capitalist development.Envisioning a postmodern identitarian poli-tics that shares more commonalities withpost-Fordist capitalism than with Marxistanti-capitalism (Harvey, 1989: 65), he suggeststhat postmodern strategies and argumentationdrawn upon by identity politics after the cul-tural turn may be only apparently progressive:

… postmodernism, with its emphasisupon the ephemerality of jouissance, itsinsistence upon the impenetrability of theother, its concentration upon the textrather than the work, its penchant fordeconstruction bordering on nihilism,its preference for aesthetics over ethics,takes matters too far. It takes thembeyond the point where any coherentpolitics are left, while that wing of it thatseeks a shameless accommodation withthe market puts it firmly in the tracks ofan entrepreneurial culture that is the hall-mark of reactionary neoconservativism.(Harvey, 1989: 116)

Thus, while politics of positionality mayseem progressive, Harvey asserts that suchfragmented strategies are in fact openings for,

if not inspired by, the equally fragmentedpractices of accumulation and productionin contemporary capitalism and, importantly,their attendant transformations of the spacesthat we daily encounter. In so arguing,Harvey makes the spaces and processesof post-Fordist capitalism the conditions forculturally inflected politics: ‘Aesthetic andcultural practices are particularly susceptibleto the changing experience of space and timeprecisely because they entail the constructionof spatial representations and artifacts out ofthe flow of human experience. They alwaysbroker between Being and Becoming’ (Harvey,1989: 327).

The impact of the Condition ofPostmodernity

As mentioned at the outset of this chapter,Condition of Postmodernity was widely readin many disciplines, and it has had a lastingimpact. In geography, the book’s most lastingmark ensued from Harvey’s efforts to con-nect economic and cultural analysis. Formany reasons that well predate the publica-tion of Condition of Postmodernity, economicand cultural geography had largely developedindependently of one another since the post-war period. The former grew in sophis ticationunder the dual influence of both spatialscientific and Marxist theories (see Barnes,1996; see Kelly, Chapter 23 this volume)while the latter was either practiced as naveempiricism (e.g., cultural geographies ofhousetypes and the like) or developed inspi-ration from humanistic geography (e.g., Tuan,1977; see Cresswell, Chapter 7 this volume).Now, it is true that there is a history of link-ing economic and cultural phenomena incritical theory: one need only point to thebase-superstructure model in Marxism, or tothe efforts of thinkers such as E. P. Thompson,Raymond Williams, Theodore Adorno, andStuart Hall, among others. But in geography



at the time of Condition’s publication, therewere only a handful of geographers whowere attempting, as Harvey was, to bringtogether the traditions of cultural interpreta-tion and political economic analysis. Amongthe notables in this period was Denis Cosgrove,whose Social Formation and Symbolic Landscape(1984 – see Gilbert, Chapter 12 this volume)brilliantly wove together political, economic,and cultural analysis. And Harvey too was anearly weaver of political economy and cul-tural interpretation in geography: one needonly look at his now classic analysis of thepolitical symbolism of the Basilica of Sacré-Coeur in Paris (Harvey, 1979).

In any event, the more general point is thathowever much Harvey would be later criti-cized for what were seen as one-sidedreadings of cultural texts (see below), and, forthat matter, however much one might dis-agree with his assessment that economicchange drives cultural responses, Condition ofPostmodernity stands as a model effort to bringtogether two subfields largely separated dueto their different objects of analysis (paintingsversus factories) and the theories and method-ologies typically brought to bear on them. Interms of analytic strategies, then, part ofCondition’s value comes from illustrating theconnections between economy and culture.

Since its publication, economic and cul-tural geography have become much moreclosely aligned if not integrated, and while itis impossible to thread causality back to thebook’s appearance, Condition of Postmodernitymust nonetheless be given its due. Today,many economic geographers actively embracethe subfield’s ‘cultural turn’ by investigating –usually in dialectical fashion – the intersec-tion between cultural forms and political-economic processes, often analyzing texts andvisual media with the tools of content anddiscourse analysis favored by cultural geogra-phers. Although quite different in theirapproach to the causal relations involved, wefind this sort of analysis exemplified in the

work of: Linda McDowell, who integratescultural analysis, economic geography, andfeminism in Capital Culture: Gender at Workin the City (1997); Nigel Thrift, who honesin on capitalism’s own ‘cultural circuit’ inKnowing Capitalism (2005); and Allen Scott,whose The Cultural Economy of Cities (2000)examines the economic bases of the cultureindustries in Los Angeles and other worldcities. This is not to say that Condition isresponsible for the growth of representa-tional or discursive analyses of the economyduring the last decade of the twentieth cen-tury, for Harvey’s own stance with respect to‘cultural political economy’, as it is some-times called, is clear enough. It is, rather, topoint to the fact that however contentiouseconomic geography’s cultural turn hasbeen, Condition, by virtue of its integration ofeconomy and culture, opened up new terri-tories in geography. Whether Harvey wouldapprove of the circuitous routes taken to getto the intersection is, however, a differentmatter.

Critical reactions

Of course, Condition of Postmodernity was notuniformly welcomed in geography, nor forthat matter was it praised in all quarters ofsocial and cultural theory. This is to beexpected: geography, as this volume shows, isa highly differentiated and often contentiousterrain, and many of the lines of conflictwithin the field are refractions of similardebates within theory more generally. Atworst, there were geographers and otherswho, in claiming Harvey to be a postmod-ernist, demonstrated that they never read thebook, much less its back cover. At its best, atthe time of Condition of Postmodernity’sappearance, two of the most important intel-lectual fault lines in critical geography (andelsewhere) were between what was perceivedto be an overly structuralist, totalizing, and


economistic Marxism (Duncan and Ley,1982), on the one hand, and poststructuralismand feminism, on the other hand. These lasttwo – which are related to but not sub-sumed by the more substantive tensionsbetween economic and cultural geography,as discussed above – were sometimes overlap-ping critical injunctions that together provideda tense intellectual field for the reception of thebook, and nowhere is this better illustratedthan in two lengthy essays, one by RosalynDeutsche (1991) and one by Doreen Massey(1991), both of which drew on poststructural-ist feminism to offer withering critiques.

Deutsche began her critique by accusingHarvey of relying on a masculinist and ocu-larcentric epistemology that unreflexivelyprofesses confidence in the ability to clearlygrasp causal connections free of any compli-cations that might be introduced by theviewer’s social positionality. This ‘totalizing’view, she maintained, underlies Harvey’sdeployment of a rigid Marxist analytic aimedat taming an unruly postmodernism filledwith difference and possibility. It also explainshis failure to recognize any limits in his per-spective, as well as his lack of acknowledgementof both feminist work on postmodernism(not in any way an easy combination: seeNicholson, 1990) and feminist representa-tional theory, particularly as it circulatedwithin the domain of art. As for the latter, sheoffered a stinging criticism of Harvey’s read-ing of the photographic self-portraits ofartist Cindy Sherman: where Harvey saw inSherman’s many disguises evidence for adepthless postmodern fetishism, Deutscheread them as critical commentaries on mod-ernist artistic theories and their emphases onthe individual artist and ‘his’ invocation ofuniversal truths; for Deutsche, Sherman’s por-traits disrupted such authorial notions bypointing to the social construction of meaningin artistic representations, while simultane-ously questioning the ‘reality effect’ ofdocumentary photography. On this point

Deutsche’s poststructuralism departs fromHarvey’s geographic materialism, tappinginto what had become a seemingly endlesspoint-counterpoint between materiality andrepresentation in the study of socio-spatiallife (Dixon and Jones, 2004). As Deutschehad it:

Reality and representation mutually implyeach other. This does not mean, as it isfrequently held, that no reality exists orthat it is unknowable, but only that nofounding presence, no objective source,or privileged ground of meaning ensuresa truth lurking behind representationsand independent of subjects. Nor is thestress on representation a desertion of thefield of politics … any claim to knowdirectly a truth outside representationemerges as an authoritarian form of rep-resentation employed in the battles toname reality. (Deutsche, 1991: 21)

Massey’s (1991) critique echoes elementsfound in Deutsche’s while redoubling onHarvey’s limited engagement with feministanalyses of patriarchy. She begins by quotinga now famous question posed by NancyHartsock (1987): ‘Why is it, exactly at themoment when so many of us who have beensilenced [under modernism] begin to demandthe right to name ourselves, to act as subjectsrather than objects of history, that just thenthe concept of subjecthood becomes [underpostmodernism] “problematic”?’ (quoted inMassey, 1991: 33, brackets added). She locatesthe rise of the postmodern not, as Harveydoes, within the coordinates of time–spacecompression, but in two opposing trends:progressive political activity marshaled arounddifference ‘in fields such as feminist studies,ethnic studies, and Third World studies’(Massey, 1991: 34), and the competitivejostling for position among career mindedacademics. On balance, however, Masseyoffers a more hopeful, feminist reading ofpostmodernism, while at the same time

DAVID HARVEYÿÿ131


affirming a commitment to retaining ‘strongaspects of what characterises the modernistproject, most particularly its commitment tochange, hopefully progressive’ (Massey, 1991:52). Nonetheless, modernity cannot be let offthe hook:

… the experience of modernism/modernityas it is customarily recorded, the produc-tion of what are customarily assumed to beits major cultural artefacts, and even its cus-tomary definition, are all constructed onand are constructive of particular forms ofgender relations and definitions of mas-culinity and of what it means to be awoman. This is not (‘just’) to say that mod-ernism was or is patriarchal (this wouldhardly be news, nor differentiate it frommany other periods in history); it is to saythat it is not possible fully to understandmodernism without taking account of this.To return more directly to Harvey, mod-ernism is about more than a particulararticulation of the power relations of time,space, and money. (Massey, 1991: 49)

The following year, Harvey (1992) respondedto both Deutsche and Massey (but, signifi-cantly, writing in the radical geographyjournal, Antipode, and not in Environment andPlanning D: Society and Space, the journal inwhich their critiques appeared). He began byacknowledging his regret for not integratingmore feminist work into Condition, notingthat, had he done so, the argument wouldhave been strengthened rather than diluted.But Harvey largely stuck to his guns, employ-ing his own differencing strategy wherebyDeutsche’s and Massey’s analyses were partic-ularized as emerging from one type offeminism, and not the one that suited his the-ory. In brief, Harvey has much more incommon with the socialist feminism ofNancy Hartsock (1987) than with the post-structuralist feminism of Deutsche, and hisresponse is at pains to point out the dangers

of what he perceives as a relativistic feminismthat, while addressing difference, lacks thestrong evaluative criteria to distinguish betweensocially and politically important axes ofidentification and insignificant ones: ‘Lackingany sense of the commonalities which definedifference, Deutsche is forced to adopt anundifferentiated, homogenizing, a-historicaland in the end purely idealist notion of dif-ference …’ (Harvey, 1992: 310). The chargeof idealism, moreover, runs through Harvey’sdismissal of Deutsche’s poststructuralism:

Deutsche may find my view that thereare social processes at work which arereal unduly limiting, but I find … theview that all understanding is preconsti-tuted, not with reference to a materialworld of social processes but with refer-ence to media images and discoursesabout that world, not only even morelimiting but downright reactionary sinceit leaves us helpless victims of discoursedeterminism. (Harvey, 1992: 316)

Harvey was even less sympathetic toMassey’s essay, accusing her in an obviouslyangry response of invoking a ‘flexible femi-nism’ in what he saw as a personal andopportunistic attack based on the fallaciousassumption (technically, a circumstantial adhominem) that ‘whatever the male gaze lightsupon is bound to be given an exclusivelymasculine and therefore sexist reading’(Harvey, 1992: 317).

Conclusion

In this chapter we have discussed the argu-ments in, impact of, and responses to Conditionof Postmodernity. Each has circulated aroundissues of content; that is, they pivot on Harvey’sanalysis of modernity’s passage to postmoder-nity, flexible accumulation’s role in time–spacecompression, the role of class relative to other



DAVID HARVEYÿÿ133

aspects of social difference, and the proper the-orization of economy and culture. Anotherway to approach the book is through its ‘modeof explanation’: how does Harvey’s analysiswork? To answer that question we turn in thisconclusion to the topic of ontology: theoriesof what the world is like, how it operates, andhow as a consequence we might understandand explain it.

At the outset, it is important to empha-size that discussions of ontology are neverfar removed from the politics of researchmore generally, and this is nowhere moreapparent than in critical geography, whichis currently engaged in a spirited debateover the status of ‘leftist’ thinking withinthe field. On the one hand, these debatesare about differences in and commitmentsto leftist thought with respect to the the-ory and praxis needed to move forward,that is, to confront and potentially overturncontemporary capitalism. On the otherhand, they express deep-seated differencesin ontology, and in particular betweenthe dialectical approach of critical realismand various strains of anti-essentialistpoststructuralism.

Critical realists (Sayer, 1992) hold thatevents are caused through the operation ofnecessary and contingent forces. These aretheorized as embedded within a depth ontol-ogy, wherein more general causal forces (e.g.,capitalism) are said to work in conjunctionwith contingent and contextually boundones (e.g., local political culture, the particu-lars of context-specific gender relations).Harvey’s book offers an example of a depthontology, as time–space compression is posi-tioned as a mediating mechanism betweenthe underlying structural forces of capitalismand the resulting, surface level cultural formsof intellectual trends, art and architecture, andpolitics. Condition of Postmodernity, whilenot explicitly referencing critical realism,demonstrates its explanatory power, and

while critical realism is not by necessityMarxist, it nevertheless complements Harvey’shistorical-geographic materialist analysisbecause of its focus on internal rather thanexternal relations among social phenomena(Sayer, 1992).

By contrast, poststructuralists reject thenotion of the structuring systems that char-acterize depth ontology. Either the world is‘overdetermined’ (Gibson-Graham, 1996), amodel in which causal forces are so mutuallyco-constituted that they are theoreticallyinseparable, or reality is itself always onceremoved through thought and language, andthus by the processes of categorization thatname and organize the world. In both forms,poststructuralists in the 1990s came to argueon behalf of theoretical agnosticism withrespect to ontology, preferring instead to pri-oritize epistemology, that is, how we come toknow the world rather than what makes upthe world. Quite simply, these authors – ofwhom Deutsche (1991) is a good example –maintain that epistemology ‘trumps’ ontology(Dixon and Jones, 2004). It should be clearthat Harvey has little patience for this groupof theorists.

More recently, however, numerous poststruc-turalists using various versions of affect theory,actor-network theory, non-representationaltheory, and site ontology, have reasserted theimportance of materialist ontological analysis(see, for example, Geoforum, 2004). AlthoughHarvey will be bemused to see himselfdescribed in parallel with these trends, henevertheless shares an affinity with them byvirtue of his longstanding dedication to anontology framed by historical-geographicalmaterialism. In this sense, to the extent that oneoutcome of the post-postmodern era is a vitaland renewed engagement with ontology, thenone of this book’s lasting legacies will havebeen its showcasing of the limits of a post-modernism preoccupied by epistemology anddiscourse.


Secondary sources and referencesBarnes, T. (1996) Logics of Dislocation: Models, Metaphors and Meanings of Economic

Space. New York: Guilford.Cosgrove, D. (1984) Social Formation and Symbolic Landscape. London: Croom Helm.Deutsche, R. (1991) ‘Boys town’, Environment and Planning D: Society and Space 9: 5–30.Dixon, D.P. and Jones III, J.P. (2004) ‘Poststructuralism’, in A Companion to Cultural

Geography, James D. Duncan, Nuala Johnson, and Richard Schein (eds), pp. 79–107.Oxford: Blackwell, pp. 79–107.

Duncan, J. and Ley, D. (1982) ‘Structural Marxism and human geography: a critical assess-ment’, Annals of the Association of American Geographers 72: 30–59.

Geoforum (2004) 25 (6): 675–764.Gibson-Graham, J.K. (1996) The End of Capitalism (as we knew it): A Feminist Critique of

Political Economy. Cambridge: Blackwell.Hartsock, N. (1987) Money, Sex, and Power: Toward a Feminist Historical Materialism.

Evanston: Northwestern University Press.Harvey, D. (1979) ‘Monument and myth’, Annals of the Association of American

Geographers 69: 362–381.Harvey, D. (1982) The Limits to Capital. Oxford: Blackwell.Harvey, D. (1987) ‘Flexible accumulation through urbanization: reflections on “post-

modernism” in the American city’, Antipode 19: 260–286.Harvey, D. (1989) The Condition of Postmodernity: An Enquiry into the Origins of Cultural

Change. Oxford: Blackwell.Harvey, D. (1992) ‘Postmodern morality plays’, Antipode 24: 300–326.Harvey, D. (1996) Justice, Nature and the Geography of Difference. Oxford: Blackwell.Harvey, D. (2002) ‘Memories and desire’, in P. Gould and F.R. Pitts (eds), Geographical

Voices. Syracuse: Syracuse University Press, pp. 149–188.Jameson, F. (1984) ‘Postmodernism, or, the cultural logic of late capitalism’, New Left

Review 146: 53–92.Jameson, F. (1991) Postmodernism, or, the Cultural Logic of Late Capitalism. Durham:

Duke University Press.Lyotard, J.F. (1984) The Postmodern Condition: A Report on Knowledge (Geoff Bennington

and Brian Massumi, trs). Minneapolis: University of Minnesota Press.Massey, D. (1991) ‘Flexible sexism’, Environment and Planning D: Society and Space 9:

31–57.McDowell, L. (1997) Capital Culture: Gender at Work in the City. Oxford: Blackwell.Nicholson, L.J. (ed) (1990) Feminism/Postmodernism. New York: Routledge.Sayer, A. (1992) Method in Social Science: A Realist Approach (2nd edition). New York:

Routledge.Scott, A.J. (2000) The Cultural Economy of Cities. London: Sage.Thrift, N.J. (2005) Knowing Capitalism. London: Sage.Tuan, Y.F. (1977) Space and Place. Minneapolis: University of Minnesota.



POSTMODERN GEOGRAPHIES (1989):EDWARD SOJA

Claudio Minca

For at least the past century, time andhistory have occupied a privileged position in thepractical and theoretical consciousness ofWestern Marxist and critical social science. […]Today, however, it might be space more thantime that hides consequences from us, the ‘mak-ing of geography’ that provides the mostrevealing tactical and theoretical world. Thisis the insistent promise and promise of post-modern geographies. (Soja, 1989: 1)

Introduction

‘All geographers should read this book’, urgedMichael Dear in his review of PostmodernGeographies that appeared on the Annals of theAssociation of American Geographers in 1991. Andif not all, then certainly a majority of the geo-graphers writing in those years on therelationship between geography and socialtheory did indeed read – and engage with –Ed Soja’s book. Postmodern Geographies,together with David Harvey’s The Condition ofPostmodernity published that same year (seeWoodward and Jones, Chapter 15 this vol-ume), is a work that has exercised a profoundinfluence on a whole generation of criticalhuman geographers (in English-speakingacademia but also well beyond) and whoseimpact on the discipline is visible even wellover a decade after its release. It is enough tothink of the widespread adoption in geogra-phy of the work of Foucault and Lefebvre –two key reference points in Soja’s call for acritical human geography – or to the currency

gained by the concept of socio-spatial dialectic torealize how pervasive this influence has been.Soja was one of the first writers to ‘import’into geography what he referred to as the newFrench spatial school, and his book was a disci-plinary milestone in opening a dialogue withcritical social theory.

The year 1989 was the last of a vibrantand, in many ways, revolutionary intellectualdecade for human geography and for the ‘spa-tial turn’ in the social sciences more broadly.The publication of Postmodern Geographiescame at a critical juncture in geography’srecent history: a moment in which ‘postmod-ern’ perspectives were affirming themselvesbut had not yet been completely metabolized bya geographical community that was becoming –at least in the English-speaking world –increasingly ‘critical’. The book’s seductivetitle perfectly captured the spirit of the times,while the subtitle (‘The Reassertion of Spacein Critical Social Theory’) spoke to readersoutside of the discipline, helping make thiswork a classic not only in geography but alsoin what we can refer to by way of short-handas ‘postmodern studies’.

Postmodern Geographies was, in many ways,something entirely new. Michael Dear’s(1991: 652) review of the book argued that‘In his reconstruction of Los Angeles [Soja]invents a new way of writing about the city’,concluding that ‘within geography, I wouldcompare Soja’s achievement favourably withthe other pivotal texts of recent decades:Haggett’s Locational Analysis and HumanGeography (1965); Harvey’s Social Justice and

16


136ÿÿCLAUDIO MINCA

the City (1973), and Gregory’s Ideology, Scienceand Human Geography (1978)’. But the book’snovelty – and daring – made it also theobject of critique or, in some cases, a pointedsilence. Postmodern Geographies was, in fact,curiously ignored by many key geographicaljournals, especially those based in the UK,where very few reviews of the book appeared(but see Dear, 1991; Eflin, 1990; Rose, 1991;Smith, 1991; from outside the discipline see,for example, MacLaughlin, 1994; Resch,1992). Such silence was surprising, especiallyfor someone like myself who at the time wasengaged in discussions about the book withcolleagues in Italy and beyond, and enthusi-astically leading successive generations ofstudents through Soja’s work.

The reason why I mention the somewhatintricate academic trajectories of PostmodernGeographies within Anglophone geography –even before saying anything much about thebook itself – has to do with the emphasis thatI intend to place on the importance of thevolume (even for those who completelyrejected Soja’s ‘postmodern’ vision) in thechanging cartographies of the discipline inthat tumultuous and incredibly prolific endof the decade, when postmodernism seemedto embody at once a blessing and a curse forthe future of geography. It is also importantto note that many of the reviews of PostmodernGeographies considered it together withHarvey’s The Condition of Postmodernity (seeWoodward and Jones, Chapter 15 this volume;see for example Friedland, 1992; Marden, 1992;Massey, 1991; McDowell, 1992; Warf, 1991) –although quite different in spirit, the two workswere somehow perceived as marking a newopening in the reflection on postmodernismand postmodern geography.

Key arguments

The intellectual ambition and the theoreticalbreath of Postmodern Geographies was simply

astonishing. After having contributed formany years to some of the most importantdebates on Marxism and social theory ingeography (see for example Soja, 1980),Postmodern Geographies launched Soja on anew and provocative research agenda. In hisattempt to re-establish a critical spatial per-spective in/on contemporary social theory,Soja highlighted what he perceived as one ofthe endemic problems marking the greatbulk of social inquiry from the end of thenineteenth century on: that is, the privileging oftime over space in the understanding of the socio-spatial fabric. Soja’s target then, from theopening pages of the book, was the pervasivehistoricism of modern social science that had,for far too long, neglected the importance ofthe ‘spatial’, allowing ‘time’ to dominate allinterpretations of the workings of the ‘social’.He suggested starting anew with a fundamen-tal rethinking of the ontological foundationsof modern European social science, in orderto inaugurate a critical human geographybased on new spatial ontologies and focusedon what he described as ‘the socio-spatialdialectic’. This grand enterprise appealed totheorists such as Lefebvre and Foucault, butalso to Berger, Jameson, Giddens and manyothers, in order to argue for a geographicalmaterialism centred on space, time, andbeing, conceived as a potential way of over-coming the ‘stranglehold’ of historicism thatprivileged time over space; a way to force arecognition of the deep ‘spatiality of sociallife’ and its profound political implications(Warf, 1991: 101).

The relevance for geography of the messagethat Soja’s book sent to the rest of the socialsciences can hardly be overrated. According toSoja (1989: 1), the answer that geographycould offer to this problematic privileging oftime over space was, in fact, the ‘insistentpremise and promise of postmodern geogra-phies’. By linking geography and the spatialwith the postmodern, Soja attempted not onlyto dismantle the implicit privileging of


EDWARD SOJAÿÿ137

geometric narratives of space in the past century,but also to argue that a new geographical wayof understanding society and its relations wasneeded – and that the return of philosophyand critical theory into the realm of humangeography was the key to this revolution.

Postmodern Geographies was considered, evenby its most forceful critics, a very importantbook for geography – and for the social sciencesmore broadly. Derek Gregory (1994: 258), forone, argued that ‘Postmodern Geographies [is] abrilliant book. The work of a master-craftsman,its intellectual sparkle is the product of a rare andgenerous critical intelligence. […] the result is acarefully polished text, with each word weighedand set in place to bring out the deeper tonali-ties of the others’. Taking a brief look at thevolumes that in the last ten/fifteen yearshave tried to reconstruct the recent history ofthe discipline in the English speaking world,Postmodern Geographies (and Soja’s work morebroadly) figures prominently in discussions ofthe postmodern turn. For instance, Cloke,Philo and Sadler’s (1991) Approaching HumanGeography highlights Postmodern Geographies as‘the book’ that, together with Harvey’s work,opened the ground for a reflection on post-modern geography: their textbook also includes,indeed, a ‘window’ dedicated to PostmodernGeographies, presented as a key text in the recenthistory of the discipline. Peet’s seminal ModernGeographical Thought similarly points to thepublication of Postmodern Geographies as a funda-mental step in the affirmation of a postmoderngeography and, despite some scepticism, notesthat it was the first book that engaged with a‘materialist interpretation of spatiality’ conceivedin terms of a new ‘spatialized ontology’ (Peet,1998: 222–224). Postmodern Geographies is alsogranted attention in Johnston’s by now-classicdisciplinary reconstruction Geography andGeographers, where a number of Soja’s quota-tions open the subchapter devoted toPostmodernism (1997: 271, 275). The list ispotentially endless but the important point to bemade is that some of the most respected voices

in the discipline have, over the years, marked outthe publication of the book as a milestone ingeographical reflection and a key event in theputative postmodern turn of the discipline.

Reasserting space

Postmodern Geographies consisted of a fiercechallenge to historicism, to the subordinationof space in critical social theory, togetherwith an accusation of what Soja saw as theparalysis of modern geography – a paralysisdue not only to this subordination, but alsoto the marginalization of the discipline inmainstream critical social theory. Soja’s bookrepresented, at the same time, an attempt ofreaffirming the importance of the return ofphilosophy in geography and of geography inthe social sciences, in those years facing thepromising consequences of the spatial andcultural turns. It is probably here that wecan find the book’s most intriguing links/challenges to today’s disciplinary reflection.

In the volume, Soja reconstructed the historyof the discipline, outlining the affirmation of ahistoricist vision of critical thought (and theeventual banning of theory in geography, to putit in crude terms) as a consequence of the factthat the last two decades of the nineteenth cen-tury and the first two of the following wereessentially dominated in Europe by ‘modernmovements’ which defined ‘separate and com-petitive realms of critical social theorisation, onecentred in the Marxist tradition, the other onmore naturalist and positivist social science’ (Soja,1989: 29). Soja urged readers to look to the finde siecle – when critical theory turned resolutelyhistoricist – to argue that, after the ParisCommune, historicity and spatiality wouldnever again be considered together as explanatoryforces. After the defeat of the Parisian experi-ment, according to Soja, the spatial dimensionof critical social analysis began to fade away togive … ‘space’ to the grand revolutionarySubject of History. Critical geo graphical



readings had therefore been mar ginalized andsilenced in social theory, with the hegemony ofhistorical materialism (on the critical side) andscientific positivism (on the other) becomingvirtually overwhelming. It is in this dialecticalrelationship that the reasons for the ‘virtual anni-hilation of space by time in critical socialthought’ (Soja, 1989: 31) can be found; and it isalso here that the exile of geography from the-ory formation in Western thought can beexplained.The triumph of geometric spatialities(through the a-critical adoption of cartographicparameters), on the one hand, and of meredescriptive techniques, on the other, left moderngeography an orphan of theory and deprived itof critical perspectives.

Seen from a ‘European’ perspective, Soja’sdiscussion of twentieth-century geography,although insightful and useful, did not suffi-ciently take into account the evolution ofcontinental geography and, in particular, thetensions between the Erdkunde project – as acritical vision of the Earth – and the emer-gent state geography that influenced most ofthe developments in the discipline for almosta century up until the humanistic and cul-tural turns. The ‘spatial perspective’ did notabandon geography with the turn of the cen-tury. What did occur was that a very specificcartographic vision of space emerged andcolonized the discipline and social theorymore broadly – for a distinct set of politicalreasons linked to the affirmation and consol-idation of the bourgeois nation state. Such avision would remain with us for most of thetwentieth century, presenting itself as theonly possible conception of space. This con-ception of space was metabolized in thesocial sciences as an essential dimension of‘reality’, as a constitutive element of theworld; and it is this very fact that occludedfor many decades, especially in geography,any reflection on the deep nature of the car-tographic reasoning on space. PostmodernGeographies was genuinely concerned aboutthis lack of reflection although (in my view at

least) Soja’s reconstruction somewhat over-looked the political roots of this problem: indeciding not to engage directly with some ofgeography’s canonical texts – such as theErdkunde project, Ratzel’s opus, or with Vidalde la Blache’s definitive ‘statalisation’ of thediscipline – Soja did not fully confront thehistorical and political reasons for the dis-missal of critical spatial theory in continentalgeography. It is curious, indeed, that geogra-phers have made so little of this question inthe discussions that followed the publicationof the book, with attention mostly focusedon a whole other set of issues.

Let us go back to the text, however. With anextraordinary historical leap – unfortunatelytypical of many Anglophone reconstructions ofthe discipline – Postmodern Geographies movedfrom the crisis of critical spatial theory at theturn of the century to the new geographies ofthe 1960s and 70s. And this is where the mostfascinating and innovative part of the recon-struction begins. After having set aside thequantitative wave as a ‘mathemathized versionof geographical description’ (Soja, 1989: 51), thediscussion focused on the rise of Marxist geog-raphy in the 1970s and the ‘spatializing’language adopted by an important part ofWestern Marxists. Soja engaged with some ofthe leading Marxist thinkers of the day, fromHenri Lefebvre, John Berger and Ernst Mandelto Anthony Giddens, David Harvey andFrederic Jameson, while also interrogatingHeidegger’s spatialities and the phenomenolog-ical geographies inspired by Husserl and Sartre.

Beyond Soja’s long-standing dialogue withLefebvre’s work, also of particular importancewas PG’s engagement with Foucauldian the-ories of power – a set of ideas and conceptsby now taken for granted in critical humangeography but revolutionary in those years.Writing in 1992, Chris Philo (together withFelix Driver, 1985, one of the pioneers inexploring the implications of Foucault’swork for geography) would write: ‘Whatis surprising is the absence to date of any


EDWARD SOJAÿÿ139

sustained theoretical engagement withFoucault on the part of theoretically mindedgeographers. There is certainly one exceptionin this respect, though, and this is to be foundin the opening chapters of Soja’s importanttext Postmodern Geographies, where he exam-ines Foucault’s “ambivalent spatiality” as astraw in the wind of “reasserting space incritical social theory”. The result is athoughtful and intriguing introduction toFoucault’s geography’ (Philo, 1992: 138).

It is indeed on French social theory thatSoja draws most heavily for the new map-pings of Western Marxism and its passage topostmodernism, a passage that Soja presentsalso as a necessary reconstruction of a criticalhuman geography. Here lies, to my mind,probably the most important contribution ofthis book to the discipline of geography.Postmodern Geographies was, in fact, about thefoundation of a new ontology – of a new spa-tial ontology. A new ontology that, accordingto Soja, would lie at the centre of a new,‘postmodern’ geography; of a new way ofunderstanding space and society.

Postmodern geographies

Soja associated postmodernism with theprocess of ideology formation and culturalreproduction. Postmodernism was, primarily,a ‘periodisation’ that made ‘theoretical andpractical sense of [the] contemporary restruc-turing of capitalist spatiality’ (Soja, 1989: 159);it was a historical transition as well as a shift incritical theory. Accordingly, contemporarygeography needed to be deconstructed andsubsequently reformulated in order to be ableto cope with what Soja defined as the threedifferent paths of spatialization that hadmarked society – and critical reflection onthe social – in the preceding decades.Postmodern Geographies spoke of a transitionin the configuration of everyday materiali-ties, and a correspondent new set of critical

understandings; it noted the consequences ofa crisis of representation that necessarilycalled for entirely new ways of describingour ‘postmodern’ reality. ‘Postmodernism’was, in other words, the cultural and ideolog-ical reaction to an epochal change that Sojadefined also in terms of post-historicism –the reassertion of space in Western thought –and post-Fordism – the flexible, disorganizedregime of capitalist accumulation trans-forming the world and its political – andcultural – economies. Understanding anddescribing these new spatializations was keyto the emergence of a new postmoderngeography.

A fundamental concept in the constitutionof this new geographical language was that ofsocio-spatial dialectics, based within a newunderstanding of ‘spatiality’ itself: ‘the gener-ative source for a materialist interpretation ofspatiality is the recognition that spatiality issocially produced and, like society itself,exists in both substantial forms (concrete spa-tialities) and as a set of relations betweenindividuals and groups, an “embodiment”and medium of social life itself ’ (Soja, 1989:120). This understanding drew broadly onLefebvrian theorizations and, inspired by theFrench philosopher, Soja fundamentally rejectedthe reduction of spatiality to the materialitiesof the world – as Gregory (1994: 273–274)would note, ‘apprehended as a “collection ofthings” (“the illusion of opaqueness”)’ – or itsreduction ‘to psychological constructs, revealedas merely projections of the mind (“theillusion of transparency”)’ (Gregory, 1994:273–274). Spatiality, Soja suggested, shouldrather be theorized as socially produced space.Spatiality was part of a ‘second nature’ (Soja,1989: 129), at the same time source, mediatorand output of social action; a spatio-temporalstructuring process, argued Soja, defines ‘con-crete spatiality – actual human geography –[as] a competitive arena for struggles oversocial reproduction’ while the temporality ofsocial life ‘is rooted in spatial contingency in



much the same way as the spatiality of sociallife is rooted in temporal/historical contin-gency’ (Soja, 1989: 130).

A truly postmodern geography would thusbe a geography able to engage with the con-cepts of spatiality and socio-spatial dialecticin order to make sense of a new ontologybased on an equally novel understanding ofthe relationship between space, time andbeing and, at the same time, of a radical urbanand regional restructuring associated withthe newly produced post-Fordist capitalistspatialities. It is in this sense that Soja’s call for‘critical regional studies’ was to be read, anapproach able to provide the theoretical andmethodological language for the analysis ofthe spatial consequences of post-historicism,post-Fordism and postmodern culture.

Los Angeles constituted, for Soja, the ideallaboratory for the exploration of the post-modern city and for the deployment of thesenew critical geographical tools. ‘Nothing inthe book prepares us for the tour de forcewhich is Chapter Nine! Entitled “Taking LosAngeles apart: towards a postmodern geogra-phy”, this is a brilliant (dis)integration of thecity and the region. […] This essay alone isworth the price of the book!!!’ was MichaelDear’s (1991: 651) enthusiastic assessment.‘Taking Los Angeles apart’ inaugurated, indeed,an entirely new geographical language, a newway of doing geography. Soja’s flight overL.A. was truly breathtaking, fuelled by pow-erful insights and a rare descriptive power.L.A. was, in Soja’s (1989: 221) words, ‘theplace where it all seems to “come together’’:I do not mean that the experience of LosAngeles will be duplicated elsewhere’, hewould argue, ‘but just the reverse may indeedbe true, that the particular experiences ofurban development and change occurringelsewhere in the world are being duplicatedin Los Angeles’.

Yet what certainly became one of themost quoted bits of the book also attractedcriticism. Derek Gregory’s dissection of Soja’s

postmodern geographies in his influentialGeographical Imaginations (1994, see Pickles,Chapter 20 this volume) (which dedicates anentire chapter to the book) lamented that inSoja’s ‘spiralling tour around the city’ (Gregory,1994: 299) his postmodern geography ‘as awhole seems indifferent to the importance ofethnography and in particular to the experi-ments in “polyphony” which have so vigorouslyanimated postmodern ethnography’ (Gregory,1994: 295). This was in part a consequence,according to Gregory, of the visual metaphorsthrough which Soja conceived spatiality; ‘hisrepresentation of the landscapes of LA is con-ducted from a series of almost Archimedianpoints from which he contemplates a series ofabstract geometries’ (Gregory, 1994: 299).Gregory’s critique was motivated by the factthat, despite Soja’s explicit recognition that anytotalizing description of the postmodern city(here, L.A.) would have been impossible (would,indeed, represent a contradiction in termswithin the framework of a ‘postmodern geogra-phy’) and thus his empirical chapters could onlyconsist of a fascinating succession of fragmentaryglimpses, the excursion presented in ChapterNine was, nonetheless, conducted from aseries of external (albeit mobile) viewpoints.This criticism, in many ways not surprising if weconsider the experimental nature of the chapter,should in no way diminish the path-breakingimpact of those ‘flights’; the fact that Soja’s ‘spi-ralling tours’ of Los Angeles created an entirelynew and provocative way of writing geography.

The most serious accusations against PGwere levelled, however, at its silence withrespect to feminist theorizations of space andsociety. ‘The most startling absence from PG isany sustained engagement with feminism. Thisis perfectly consistent with the modernist castof Soja’s prosject as a whole, I suppose, and anumber of writers have drawn attention to themasculinity of the modern gaze and its system-atic devaluation of the experience of women’,Gregory (1994: 309) underlined, noting alsothat ‘the contemporary conjunction between


EDWARD SOJAÿÿ141

postmodernism and feminism makes such anomission from a postmodern geography trulyastonishing’. Criticism was hence forthcomingfrom many feminist geographers (see, for exam-ple, Deutsche, 1991; McDowell, 1992) but itwas Doreen Massey’s influential review article‘Flexible Sexism’ that set the tone of the debate.Launching a forceful attack on Soja’s and DavidHarvey’s ‘postmodern’ works, it accused bothauthors of essentially ignoring the feminist con-tribution to spatial and social theory: ‘it seems tome that the absence from, indeed the denial by,both these books of feminism and the contri-butions it has recently made, raises issues whichare important for all of us, and which rangefrom our style as academics to the way in whichsome of the central concepts of the debate areformulated’ (Deutsche, 1991: 31–32) (see alsoWoodward and Jones, Chapter 15 this volume).It is important to note that Soja would, by andlarge, accept this criticism and, in the openingpages of his subsequent book, Thirdspace, wouldadmit to what could have been perceived as agender-bias in his previous book, albeit reject-ing the idea that PG ‘can be read as amasculinist posing tout court’ (Soja, 1996: 13).Indeed, Thirdspace would extensively engagewith feminist and postcolonial theory, with par-ticular attention, in geography, to the work ofGillian Rose (Soja, 1996: 121).

If the controversy over feminism andpostmodernism marked an important passagein the reception of the book, another aspecthighlighted by many commentators was the‘thoroughly modernist’ spirit of Soja’s project(Gregory, 1994: 273). Warf (1991: 101), sug-gested, for example, that ‘Soja’s explanationamounts to an impressive historical-material-ism interpretation of the late twentiethcentury capitalism that hinges upon a decid-edly modernist, not a postmodernist view’,while for Dear (1991: 653), Soja’s Marxismeffectively left the ‘ontological project’ intact,overvaluing the contribution of Marxist main-stream, and representing most non-Marxists asfundamentally anti-postmodern.

What do we make of this criticism today?My view, after almost two decades, is thatPostmodern Geographies was in fact intentionallyconceived as a Marxist and modernist criticalrethinking of geography and of the hegemonyof historical materialism in the social sciences.Only with the last chapter on L.A. did Sojaexplicitly move beyond that tradition, experi-menting with what a postmodern geographyof a postmodern city could be. However wemay interpret the success of that experiment, Ido believe that this was the ‘substance’ of theproject behind PG. We should also recognizethat the postmodern overall had enteredAnglophone geography largely by way of anintricate but decidedly Marxist stream, and thiscan explain why the tension between adeclared postmodern relativism and structurallyoriented ‘Marxist’ descriptions of society andthe economy marked (postmodern) geographyfor so long. The emphasis placed by Soja on thetransformation of global modes of capitalistproduction is also understandable from thisperspective.

A similar point can also be made regardinga more general reflection on the potentialpolitical implications of the sort of postmod-ern praxis described by Soja. The 1990switnessed a rich debate on radical ‘postmod-ern’ politics and, looking back, it is importantto note the place of Soja’s subsequent twobooks vis-à-vis these discussions. Despite thefact that, again from my point of view,Postmodern Geographies was a decidedly ‘politi-cal’ text – it would be quite difficult to argueotherwise considering its driving ideologicalpush – it is important to note that inThirdspace first and Postmetropolis later, Sojawould explicitly engage with many of theearly critiques, such as that of having par-tially overlooked the politics of racism andsexism, or of not having fully explored, in hispostmodern politics of resistance, the poten-tial of many grassroots social movementsemerging in those times as new politicalsubjects.



Conclusion

Today, Soja’s influence on contemporaryunderstandings of the urban is still strong andreaches a genuinely international audience wellbeyond the discipline of geography. Indeed,one of the hallmarks of Postmodern Geographieswas its truly global reach. To my knowledge,the book was translated into Portuguese,Japanese, Chinese and Serbo-Croat, and partsof it into Italian and German; a special editionhas also been published in India. Copies ofPostmodern Geographies can be found not onlyin academic bookshops around the world, butalso in the book collections of modern art andarchitectural museums and galleries fromWashington to Berlin to Shanghai.

What can we say of the legacy of PostmodernGeographies at a time in which the echoes of thepostmodern wave in geography (and beyond)seem to have faded away, as evidenced by DavidLey (2003) in a recent appraisal? The first thingthat I would like to say is that although it mightbe true, as Dear (1991: 654) would have it, thatPostmodern Geographies was ‘a brilliant and fruit-ful account of one person’s encounter withpostmodernism’, Postmodern Geographies was –and is – also a book that marks a much longertrajectory, traversing the history of humangeography and of the social sciences of the lasttwo decades. For this reason, I believe thattoday we are in the position to be ableto express a serene and partially detached (timehelps) judgement on this extraordinary ini-tiative. The road opened by PostmodernGeographies, with its powerful insights andbrilliant theoretical engagements, but alsowith all its tensions and provocations, hasgiven a vital contribution to making geog-raphy the vibrant discipline that it is today,if anything for the reactions that it pro-voked and the debate that it stimulated,

both in the academic and intellectualrealms.

We can probably also accept today, with sim-ilar peace of mind, that Postmodern Geographieswas perhaps not quite yet as postmodern as thetitle intimated, but it is out of the question thatthe volume represented a path-breaking inter-vention at a moment in which there was a needfor just such an intervention – also for thosewho did not like the book and dismissed it. Andhere I do not refer only to Soja’s early engage-ment with Foucault, or to his reflection onLefebvre’s spatialities or to the concept of socio-spatial dialectics that will influence a generationof urban geographers. Nor do I only think of histimely recognition of the utility of Giddens’reflection for geographers and of Jameson’swork for all scholars concerned with thecontemporary city. Postmodern Geographies’s con-tribution has not only been important for therevolution that it brought to the reflection onthe relationship between Western Marxism,postmodernism and geography, but also for itsfantastic flight over L.A. that launched a newway of writing geography. What I believe hasbeen the major contribution of this book, espe-cially in the light of what happened in thedecades that followed, is that it showed howgeography might (and ought) to dialogue withphilosophy and social theory. What is more, as ageographer, Soja demonstrated that in order tograsp the spirit of our times, to interpret ournew modes of communication, to read – andlive – the cities of today, a geographical way ofunderstanding space and spatial theory wasnot only useful but crucial for the future ofsocial analysis. ‘Soja has irrevocably shifted theperspective of both geographers and non-geographers on the project of Geography’ –announced Dear in the by-now far away 1991;contemporary human geography has shownthat this was a fitting prediction.

Secondary sources and referencesCloke, P., Philo, C. and Sadler, D. (1991) Approaching Human Geography. New York: Guilford.Dear, M. (1991) ‘Review of Postmodern Geographies’, Annals of the Association of

American Geographers 80: 649–654.


EDWARD SOJAÿÿ143

Deutsche, R. (1991) ‘Boys Town’, Environment and Planning D: Society and Space 9: 5–30. Driver, F. (1985) ‘Power, space, and the body: a critical assessment of Foucault’s

“Discipline and Punish” ’, Environment and Planning D: Society and Space 3: 425–446.Eflin, J. (1990) ‘Review of Postmodern Geographies’, Geographical Review 80: 448–450.Farinelli, F. (1992) I segni del mondo. La Nuova Italia Scientifica: Bari.Friedland, R. (1992) Space, place, and modernity: the geographical moment,

Contemporary Sociology 21: 11–15.Gregory, D. (1994) Geographical Imaginations. Oxford: Blackwell.Harvey, D. (1989) The Condition of Postmodernity. Oxford: Blackwell.Johnston, R. (1997) Geography and Geographers. London: Arnold.Ley, D. (2003) ‘Forgetting Postmodernism? Recuperating a social history of local knowl-

edge’, Progress in Human Geography 27: 537–560.McDowell, L. (1992) ‘Multiple voices: Speaking from inside and outside “the Project” ’,

Antipode 24: 56–72.MacLaughlin, J. (1994) ‘Review of postmodern geographies’, History of European Ideas

18: 803–805.Marden, P. (1992) ‘The deconstructionist tendencies of postmodern geographies: a

compelling logic’, Progress in Human Geography 16: 41–57.Massey, D. (1991) ‘Flexible Sexism’, Environment and Planning D: Society and Space

9: 31–57.Peet, R. (1998) Modern Geographical Thought. Oxford: Blackwell.Philo, C. (1992) ‘Foucault’s Geography’, Environment and Planning D: Society and

Space 10: 137–161. Resch, R. (1992) ‘Review of Postmodern Geographies’, Theory and Society 21: 145–154.Rose, G. (1991) ‘Review of Postmodern Geographies’, Journal of Historical Geography

17: 118–121.Smith, D.M. (1991) ‘Review of Postmodern Geographies’, Urban Geography 12: 93–95.Soja, E. (1989) Postmodern Geographies. London: Verso.Soja, E. (1980) ‘The socio-spatial dialectic’, Annals of the Association of American

Geographers 70: 255–272.Soja, E. (1996) Thirdspace. Oxford: Blackwell.Soja, E. (2000) Postmetropolis. Oxford: Blackwell.Warf, B. (1991) ‘Review of Postmodern Geographies and The Condition of Postmodernity’,

Journal of Regional Science 31: 100–102.Warf, B. (1993) ‘Postmodernism and the localities debate: ontological questions and

epistemological implications’, Tijdschrift voor Econ. en Soc. Geografie 84: 163–168.


THE CAPITALIST IMPERATIVE (1989):MICHAEL STORPER AND RICHARD WALKER

Neil Coe

Most existing treatments of urbanization,regional development, and industry locationare based upon neoclassical economics andshare its assumptions and shortcomings.The three fundamental building blocks ofneoclassical theory are: (1) the central eco-nomic activity is exchange; (2) the goal ofeconomic exchange is efficient resource allocation in the service of subjective prefer-ences; (3) the natural state of the system is to come to rest at a stable equilib-rium … Our view of capitalist reality is …quite otherwise. The economy is fundamen-tally a disequilibrium system, driven to growand to change by its own internal rules ofsurplus generation, by investment to expandcapital, by fierce competition, and by techno-logical change to extract more surplus(value) from human labor. (Storper andWalker, 1989: 38)

Introduction

The 1989 publication of Michael Storperand Richard Walker’s book The CapitalistImperative was undoubtedly a key milestonein the development of a Marxist-influencedpolitical economy approach to human geog-raphy. Since the mid-1970s, and prompted byDavid Harvey’s Social Justice and the City(1973), many human geographers, and eco-nomic geographers in particular, had beenstriving to shrug off the legacy of neoclassicallocation theory and its sometimes simplisticattempts to quantitatively model the spatial

structures of the economy (e.g. Isard, 1956).Geographers increasingly began to challengethe presumed neutrality of spatial science,highlighting, for example, the completeabsence of capitalist class relations in suchanalyses. Instead, and inspired by the insightsoffered by Marxian political economy, theysought to demonstrate how the processes ofcapitalist accumulation and their associatedsocial relations were inherently geographical.In such analyses, the main foci of analysiswere no longer discrete spatial coordinatessuch as firm locations, but rather the deepstructures of capitalist social relations –such as the ongoing struggle between capitaland labour – that were in turn seen toactively shape spatial patterns of economicdevelopment.

A wide range of influential books andarticles were published by geographers in thelate 1970s and 1980s as part of this drive todevelop detailed historical geographies of thecapitalist mode of production and in so doingforge a ‘new’ industrial geography to supplantthe previously dominant neoclassical para-digm. Scott (2000) identified three strands ofresearch within this emerging body of work.One theme focused on the nature of urbanspace under capitalism, and more specifically,the links between land rent, housing provi-sion and urban planning (e.g. Harvey, 1985).Another was concerned with problems ofdeindustrialization, poverty and job lossoccurring as economic restructuring bit hardin many developed countries (e.g. Bluestone

17


and Harrison, 1982) and was closely linked todetailed investigations of regional industrialrestructuring and changing geographies oflabour (e.g. Massey, 1984). The third streamof work sought to theorize uneven geo-graphic development within capitalism at arange of spatial scales (e.g. Harvey, 1982;Smith 1984; see Castree, Chapter 8 this vol-ume; Phillips, Chapter 9 this volume). As weshall see, Storper and Walker’s The CapitalistImperative – which offers ‘a broad depictionof the complex interconnections betweentechnology, industrialization and territorialchange in advanced capitalist societies’ (Smith,1992: 81) – can be positioned as an ambitiousattempt to bridge the latter two strands ofpolitical-economic research.

The book and its argument

The argument of the book can be summa-rized, albeit somewhat crudely, as follows.The geographies of capitalist developmentare extremely dynamic, with the fortunes ofparticular places, regions and countries wax-ing and waning over time as they interactwith different waves of industrialization.Contra the assumptions of neoclassical eco-nomics and location theory, rather thanexhibiting any tendency to spatial uniformity,capitalist growth is inherently expansionaryand unstable, and produces differentiationbetween territories specializing in particularindustries. While industries exhibit an innatetendency to agglomerate in particular locali-ties, the tapestry of interconnected clustersthat results is far from static, and the produc-tion process, rather than the market, is centralto understanding this dynamism.

The most important shapers of industrialchange are technologically intensive industrieswith innovative new products and technolo-gies which have the ability to re-locate andactively create new regional clusters duringthe early stages of their development. The key

mode of economic organization within thisfluid system is not the market, the individualfirm or the workplace; rather it is the territo-rial production complex, a cost-reducingconfluence of firms, workers, knowledge,resources and infrastructure (see below). Suchcomplexes facilitate the flexibility – in termsof divisions of labour, local labour markets andtechnology – inherent to modern productionsystems. Understanding the fluctuating yetinterconnected fortunes of these territorialcomplexes, both new and old, in the core andin new growth peripheries, is critical to reveal-ing the geographical nature of capitalistgrowth and its wider impacts upon societyand class relations.

According to its authors, the central aim ofthe book was to grapple with two theoreticalchallenges (Storper and Walker, 1989: 1).First, it sought to develop an analyticalframework for understanding the geographyof economic development, and more specifi-cally, the geography of industrialization. Asthe following quote suggests, the objectivehere was to move well beyond the limitationsof the neoclassical approach: ‘We propose tothoroughly rewrite location theory from thestandpoint of political economy, and, in sodoing, to leave location theory behind infavour of a theory of geographical and terri-torial industrialization’ (Storper and Walker,1989: 3). The book sought to do so throughextensively critiquing existing theories aswell as developing the authors’ own theoriza-tion. In addition to neoclassical models ofindustrial location, several other perspectiveswere put to the sword, including regionalscience, industry/sector studies, regulationtheory, behaviouralism and systems theory.While in Gertler’s (1991: 363) words ‘thesetheories are lined up and shot down like somany empty bottles in a shooting gallery’,the authors were also ‘smart enough to bor-row ideas when they make a useful contributionto their own framework’. In forging a newanalytical framework out of this extensive

146ÿÿNEIL COE


critical review and synthesis, a specific aimwas to bring together hitherto largely sepa-rate theories of industrial location andregional development.

It is worth briefly reviewing the most impor-tant analytical concepts proposed and developedby Storper and Walker in their account:

• Geographical industrialization describes theinherently geographical nature of thegrowth and decline of capitalist industriesand the constantly shifting map of eco-nomic activity that results. There are fourkey stages: (1) new industries develop inparticular places; (2) similar and relatedactivities cluster around this new ‘hearth’;(3) industries disperse some of their pro-duction capacity away from these clusters,and (4) new industries, based on innova-tive products and technologies, shift awayand take up new locations.

• Territorial development captures these processesfrom the perspective of places. It speaks tothe ways in which territories – be theycities, regions or countries – developthrough successive waves of industrializa-tion. Over time, new places are connectedinto the capitalist system, while othersbased on declining industries will becomeless significant. Territorial development isheavily shaped by the ‘dominant ensemble’of industries at a given point in time, suchas automobile and consumer durables pro-duction in the middle decades of thetwentieth century.

• Windows of locational opportunity are momentsof locational freedom during the earlystages of development of technologicallyinnovative industries (i.e. before they haveinvested heavily in factories, etc.) inwhich leading firms can choose betweena range of viable locations. This idea isunderpinned by a recognition that indus-tries do not simply respond to thelocation of inputs such as labour, resourceinputs and suppliers, but that they also cre-ate and attract such factor inputs.

• Territorial production complexes such ascities, or city-regions, are seen as the mainmode of organizing capitalist productionrather than firms, markets or workplaces.They represent localized arenas in whichcosts for participating firms and industriesare lowered through the development ofan effective division of labour betweenand within firms, the sharing of infra-structure, the pooling of resources, thedevelopment of labour skills, and the cre-ation of an environment conducive forinnovation. Over time, however, locatingin such a complex may start to constrainan industry and leave it vulnerable toapproaches developed elsewhere.

Second, it strove to contribute to social the-ory more broadly by demonstrating howpolitical economic processes are profoundlyshaped by geography. In that sense it can beread as part of a wider project to reveal themutually constitutive interactions between thesocial and spatial structures of society (e.g. Gregoryand Urry, 1985). As Storper and Walker argue,‘Space must again become an active variable inthe social system, because human action takesplace in specific locales and social relations fromboth within and across territorial boundaries… That is human life unfolds in a thoroughlygeographical way’ (Storper and Walker, 1989: 3).The idea, then, was to fashion a form of spatial-temporal analysis that would make an impactacross the social sciences, and not just be read bythose interested in cities, industries and eco-nomic change.

Placing the book and its authorsin context

In attempting these twins tasks, The CapitalistImperative was clearly influenced by two inter-related theoretical developments in thesocial sciences more generally during the1980s, namely those concerning critical

MICHAEL STORPER AND RICHARD WALKERÿÿ147


realism and the structure-agency debate.Philosophically, the book is underpinned bycritical realism, a theory developed in thegeographical context by Sayer (1984) andwhich gained increasing currency during the1980s as scholars sought to break down thecertainties and economic logics of structuralMarxist frameworks. By contrast, realismsought to emphasize the contingent factorsthat determine how the causal relationshipsinherent to capitalism actually play out (dif-ferentially) on the ground. Such an approachis implicit in Storper and Walker’s book: ‘thenarrative is replete with “mights”, “maybes”,and “possiblys”, though one is never allowedto lose sight of the necessary conditions andthe underlying structures through whichbasic causal properties are defined. The over-all effect is of a sensitivity to the determinantsof the concrete rich enough to satisfy themost hardened of critics of so-called Marxistreductionism’ (Cox, 1991: 442). Similarly, it isalso possible to discern echoes of Gidden’s(1984) ‘structuration’ theories, which soughtto map out a middle ground between thoseseeking to emphasize either the importanceof structural forces or human agency (Gertler,1991). These underlying philosophical posi-tions are made clear when the authors statehow they ‘try to walk the fine lines betweenvarious intellectual pitfalls. We recognize boththe force of social structure and the initiativesof human agency; both necessity and contin-gency in the causation of particular events;both the widespread impact of deep structureand the relative autonomy of various parts ofsociety…’ (Storper and Walker, 1989: 5).

Returning to the different lines of politicaleconomic enquiry identified earlier, TheCapitalist Imperative operates at a level ofabstraction that falls between that of Harvey’s(1982) The Limits to Capital or Smith’s (1984)Uneven Development as compared to Massey’s(1984) Spatial Divisions of Labour (see Phelps,Chapter 10 this volume), all three of which arekey texts in geography’s political economy

canon. The former two volumes present over-arching theoretical models of the spatialconstitution of capitalism, and are rich in ana-lytical insight, but thin on empirics. Massey’stext, in comparison, presents an empiricallystrong account of regional restructuring in theUK in the 1960s and 1970s in developing hernotion of spatial divisions of labour (interest-ingly, writing in 2001, one of the authorsreflected on the importance of Massey’s bookalong with Bluestone and Harrison’s, 1982,treatise on American deindustrialization as keyintellectual ‘jumping-off points’ for the workwhich led to The Capitalist Imperative – seeWalker, 2001). Storper and Walker offer anaccount that is largely theoretical in nature,but one which describes the shifting positionof different kinds of places within the capital-ist system, and provides selective, rather thancomprehensive, empirical evidence to supportits arguments. As they argue in the last fewlines of their book, their approach to the‘inconstant geography’ of capitalism is one that‘can get hold on historical change at a middlelevel of abstraction and concreteness, betweenthe full sweep of centuries and the endlessswirl of everyday events’ (Storper and Walker,1989: 227).

A final way to position and think aboutthis book is as a product of a particular timeand place. Storper and Walker’s book was partof a wider corpus of path-breaking work bya ‘Californian School’ of economic geogra-phers – which also included Amy Glasmeier,Ann Markusen, AnnaLee Saxenian, EricaSchoenberger and Allen Scott, among others –who were using the experiences of Californiaand its growth industries to inform widerdebates about the resurgence of regionaleconomies in a post-Fordist era. This contextgives the book a certain ‘flavour’ and perspec-tive, as recognized by the authors themselvesin the Preface when they state: ‘Our perspec-tive has been skewed in important ways bythe view from California, on the edge of thebooming northern Pacific Rim; hence the

148ÿÿNEIL COE


overriding emphasis given here to economicgrowth and geographic expansion in modernindustrialization … others are better positionedto depict the devastation of unemployment in theFirst World or underdevelopment and imperial-ism in the Third’ (Storper and Walker, 1989: ix).

The authors of the book, both AssociateProfessors at the time of writing (Storper atthe University of California, Los Angeles –UCLA – and Walker at the University ofCalifornia, Berkeley), have gone on to full pro-fessorships and have firmly establishedthemselves as leading figures in economicgeography (although their influence stretcheswell beyond the subdiscipline). Walker hasremained at University of California, Berkeley,and since the early 1990s has turned his atten-tion to analysing the historical economicdevelopment of California. In his work andwriting, he has retained a strong belief in thepower of the political economy perspective(e.g. see Walker, 1998). Key subsequent publica-tions include his co-authored book withAndrew Sayer developing the notion of thesocial division of labour (Sayer and Walker,1992), and latterly, a book on Californianagribusiness (Walker, 2004). During the 1990s,Storper moved away from purely economicexplanations of regional dynamism and pio-neered the development of more socio-culturalexplanations based on knowledge, learning andthe emergence of place-specific economicconventions and practices (e.g. Storper, 1997).Interestingly, Storper himself has been criticalof what he sees as some of the excesses ofthe ‘cultural turn’ in geography and the socialsciences and has more recently been workingat the interface of geography and economics(e.g. Storper, 2001; Storper and Venables, 2004:see also Reimer, 2004).

Initial reception and critiques

This is a book that all human geographersshould read … Michael Storper and

Richard Walker achieve better than anybefore them the goal of demonstrating howcapitalism is (and can only be) spatiallyconstituted. In doing so, they advance theintellectual project and cause of our disci-pline like few others books in geographybefore this one. (Gertler, 1991: 361)

This is a work that must be read by thosewith interests in industrial geography andregional development, but one that alsohas important implications for a muchwider field of scholarship. (Wood, 1993:81, emphasis in original)

Surveying the book reviews published in thefew years after publication, and as the abovetwo quotes suggest, The Capitalist Imperativewas widely (and rightly) lauded as an impres-sive work not only by economic geographers,but also by historical and political geographers,and sociologists and economic historians.Commentators admired the clarity and read-ability of the book, and the quality and extentof the critique of existing perspectives. Interms of the own conceptual framework, therewas praise, for example, for its coherence andelegance, its ability to operate at a mid-level ofabstraction, the way in which it successfullydrew together the industrial location andregional development approaches, its revealingof the social dimensions of labour, technology,investment and the like, and the manner inwhich it successfully incorporated space anduneven development into the conceptualapparatus.

That being said, a book of this theoreticalbreadth and ambition was not receivedwithout some misgivings. We can group theseconcerns into six sets. Firstly, the critique ofexisting approaches was seen by some to beoverdone, attacking what one reviewerdescribed as a ‘cartoon version of neoclassicaltheory’ (Solot, 1990: 359), and one whichmany economists of the time would them-selves have seen as simplistic and outmoded.Secondly, for some readers, the authors were



more effective in synthesizing and surveyingexisting approaches than in advancing theirown approach. Smith (1992), for example,suggested that the regional mosaic theydepicted did not reveal the structural ‘geome-tries’ of development and underdevelopmentat different scales that characterize the worldeconomy. This relative silence on the ‘scalingup’ of their analysis is problematic in the con-temporary era of neoliberal globalization. Athird and particular area of concern was thatStorper and Walker offered no systematicattempt to sustain their theoretical argumentsthrough empirical evidence. Several reviewersnoted that the empirical evidence was highlyselective both sectorally and geographically,drawing in particular on developments in cer-tain US manufacturing sectors. The empiricalevidence was also rather piecemeal: the bookcontains a lot of small examples ratherthan large sets of quantitative data or longsustained histories of particular placesor sectors. As such, it was not able to offerexplicit information as to why particularregions went through certain changes at par-ticular times. Fourthly, particular elements ofStorper and Walker’s conceptual frameworkalso came under the microscope. The notionthat industries create regions was seen bysome as too unidirectional and lacking appre-ciation of the two-way relationships (i.e.regions also create industries) seen in Massey’swork on spatial divisions of labour. Otherssaw the analysis as too technologically deter-ministic – although the authors describedtheir approach as ‘strong technological deter-mination’ rather than determinism (Storperand Walker, 1989: 124) – and as placing toomuch emphasis on a version of the productlife cycle, with a tendency to focus only oncertain kinds of post-Fordist, flexible produc-tion. Fifthly, it was contended there were alsosome important ‘missing links’ discernible inthe book. In particular, there is little room forthe nation state (in its different scalar forma-tions) and civil society in the analysis.

Relatedly, the US-centrism inherent to theevidence base (see above) is reflected in thelack of discussion of different national varia-tions in capitalist processes shaped byinstitutional, regulatory and political forces.Finally, the book does not explicitly draw outthe policy implications that stem from its analy-sis of contemporary capitalist growth: onereviewer lamented the apparent ‘steadfast resis-tance’ to engaging in policy debates (Gertler,1991: 364).

Looking back: a lasting legacy?

Assessing the lasting impact of any intellectualcontribution is necessarily a highly qualitativeand subjective exercise. However, quantitativeindices such as citation indexes can provide astarting point for such as assessment. Accordingto the ISI Web of Science, as of September 2006the book had been cited 520 times in articlespublished in listed journals. From 1992 until2002, the book received at least 30 cites peryear, and still currently receives around 20 peryear, and that is just citations in leading jour-nals, ignoring citations in books and myriadother forms of publications. By this simplemeasure alone, then, the ongoing significanceof this book starts to become clear.

Notwithstanding some of the critiquesraised above, the book has arguably had a par-ticularly strong impact of three inter-linkedstrands of economic geography research:

• Post-Fordist production systems: Storper andWalker’s volume was a key contribution tothe aforementioned ‘Californian School’ ofresearch which theorized the agglomerativetendencies inherent in the organizationaland technological dynamics of a range ofpost-Fordist craft, high technology andbusiness service industries, and in turn pre-cipitated a lively debate within geographyabout the validity and wider applicability ofthese ideas.

150ÿÿNEIL COE


• New regionalism/geographies of innovation:relatedly, through its emphasis on labour,technology and investment as socialprocesses, and innovations as ‘placed’ inparticular localities, The Capitalist Imperativealso helped sow the seeds for the policy-oriented ‘new regionalism’ agendas thatemerged from the mid-1990s. Thesedebates have sought to recast regionaldevelopment in an era of globalization interms of differing combinations of localizedproduction cultures and conventions, insti-tutional and labour market formations, anddynamic learning and innovation processes.

• Labour geographies: the book depicts howlabour forces are ‘made’ in territorial pro-duction complexes through a variety ofsocial processes, and shows that labour ismuch more than simply a passive factor ofproduction. Instead, local labour marketsare constructed, segmented and alteredthrough the location and relocation of pro-duction. These insights were an importantprecursor to the now small but flourishingsubdisciplinary area of labour geographies(e.g. Herod, 2001).

Overall, perhaps the most important contri-bution of the book is the forging of a coherentand integrated theory of capitalist unevendevelopment that is ‘grounded’ enough to dealwith the volatility, variability and contingenciesof real world economies.

Conclusion

The economic geography of today is verydifferent from that of 1989. Influenced by thecultural turn, the latest ‘new’ economic geog-raphy is a theoretically and methodologicallyplural and diverse field, ranging from quanti-tative approaches through to post-structuralaccounts grounded in textual and discourseanalysis. For those familiar with recent devel-opments in the field, The Capitalist Imperativemay seem to be lacking in several important

respects. It may seem rather structuralist, withthe agency of places and actors rather cir-cumscribed by broader capitalist dynamics. Itmay seem overly economistic, with morecultural processes and formations under-weighted. It may seem overly productionist,with little or no mention of retail capital andprocesses of consumption, for example. Itmay appear rather technologically determin-istic. And the institutional landscape onwhich events are played out might appearrather ‘thin’, with the absence of differentstates-formations particularly notable. Thereare elements of truth to all these critiques,and indeed the authors have themselves clari-fied and in some cases modified theirpositions on these issues in subsequent writ-ings. On the other hand, these critiques are insome ways unfair, as any book is a product ofits particular time (and place). The CapitalistImperative undoubtedly drove several debatesin economic geography forward productively,and it is in those terms that it should bejudged.

As a student reader, a useful test of yourown might be to focus on some of the keyeconomic-geographical developments oftoday – the rise of India and China as eco-nomic superpowers, for example – and thinkas you read about the extent to which thebook offers a framework for understandingthose processes. You will find that the bookcan reveal a lot about some aspects – thedevelopment of new manufacturing zonesand the technologies and labour market con-ditions that underpin them, for example – butless about others, such as the importanceof Western retailers and consumption patternsin driving demand for goods, and the role ofthe Indian and Chinese states in shaping thepace and nature and development withintheir countries. What reading The CapitalistImperative will make clear, however, is that acoherent geographical-political-economicperspective must be a central part of ourattempts to understand such phenomena.



Secondary sources and referencesBluestone, B. and Harrison, B. (1982) The Deindustrialization of America. New York: Basic

Books. Cox, K. (1991) ‘The Capitalist Imperative’ (book review), Political Geography Quarterly

10 (4): 441–444.Gertler, M.S. (1991) ‘The Capitalist Imperative (book review)’, Economic Geography 67 (4):

361–364.Giddens, A. (1984) The Constitution of Society. Cambridge: Polity Press.Gregory, D. and Urry, J. (1985) (eds) Social Relations and Spatial Strucures. London:

Macmillan.Harvey, D. (1973) Social Justice and the City. London: Arnold.Harvey, D. (1982) The Limits to Capital. Oxford: Blackwell.Harvey, D. (1985) The Urbanization of Capital. Oxford: Blackwell. Herod, A. (2001) Labor Geographies. New York: Guilford. Isard, W. (1956) Location and Space Economy. New York: Wiley. Massey, D. (1984) Spatial Divisions of Labour: Social Structures and the Geography of

Production. London: Macmillan. Reimer, S. (2004) ‘Michael Storper’, in P. Hubbard, R. Kitchin and G. Valentine (eds) Key

Thinkers on Space and Place. London: Sage, pp. 282–287. Sayer, A. (1984) Method in Social Science. London: Hutchinson. Sayer, A. and Walker, R. (1992) The New Social Economy: Reworking the Division of

Labour. Oxford: Blackwell. Scott, A.J. (2000) ‘Economic geography: The great half-century’, Cambridge Journal of

Economics 24: 483–504.Smith, N. (1984) Uneven Development: Nature, Capital and the Production of Space.

Oxford: Blackwell.Smith, N. (1992) ‘The Capitalist Imperative (book review)’, Antipode 24 (1): 81–83.Solot, M. (1990) ‘The Capitalist Imperative (book review)’, Journal of Historical Geography

16 (3): 358–359.Storper, M. and Walker, R. (1989) The Capitalist Imperative: Territority, Technology and

Industrial Growth. Oxford: Blackwell.Storper, M. (1997) The Regional World: Territorial Development in a Global Economy. New

York: Guilford Press. Storper, M. (2001) ‘The poverty of radical theory today: From the false promises of Marxism

to the mirage of the cultural turn’, International Journal of Urban and Regional Research25: 155–179.

Storper, M. and Venables, A.J. (2004) ‘Buzz: Face-to-face contact and the urban economy’,Journal of Economic Geography 4: 351–370.

Walker, R. (1998) ‘Foreword’, in A. Herod, (ed.), Organizing the Landscape: GeographicalPerspectives on Labor Unionism. University of Minnesota Press, Minneapolis, pp. xi–xvii.

Walker, R. (2001) ‘Bennett Harrison: A life worth living’, Antipode 33 (1) 34–38.Walker, R. (2004) The Conquest of Bread: 150 years of California Agribusiness. New York:

The New Press. Wood, A. (1993) ‘The Capitalist Imperative (review essay)’, Antipode 25 (1): 69–82.

152ÿÿNEIL COE


THE GEOGRAPHICAL TRADITION (1992):DAVID LIVINGSTONE

Nick Spedding

Stories are always told by people, about people,for people. Geography’s story is no exception.(Livingstone, 1992: 4)

Introduction

First published in 1992, David Livingstone’sThe Geographical Tradition is widely regardedby English-speaking geographers as the mostimportant history of the discipline producedin recent years. The ubiquity of TheGeographical Tradition on reading lists forcourses on the history and philosophy ofgeography, and its rapid appearance in thejournal Progress in Human Geography’s serieson ‘classics revisited’, testify to this (Mayhewet al., 2004). If the subject matter of the bookwas mostly familiar – The GeographicalTradition examines many of the same peopleand ideas as previous histories of geography –the way in which Livingstone told his storywas not. Because it is a remarkably open dis-cipline – a core concern with the spaces andplaces of the earth’s surface potentiallyincludes far more than it excludes – it is notsurprising that geography’s past contains ahuge variety of phenomena studied with ahuge variety of tools for thinking and doing.This substantive, philosophical and method-ological diversity was recognized by historiansof geography long before The GeographicalTradition was published. The book made a bigimpact because it did far more than just listthis past variety; Livingstone set out the detail

of how and why such variety arose. In particular,The Geographical Tradition destroyed thenotion – implicit, if not explicit, in manyprevious accounts – of smooth progresstowards an ideal, objective truth. ForLivingstone, academic geography must be asubjective, plural, contested enterprisebecause it is always influenced by the partic-ular times and places in which geographerswork:

Geography has meant different things todifferent people in different settings … theheart of my argument is that geographychanges as society changes, and that thebest way to understand the tradition towhich geographers belong is to get a han-dle on the different social and intellectualenvironments within which geography hasbeen practised. (Livingstone, 1992: 347)

The inspiration for Livingstone’s contextualapproach to geography’s history came fromstudies in the history and philosophy of sci-ence, especially those that discussed thesociology of scientific knowledge. From thesecame The Geographical Tradition’s insistencethat geographical knowledge was, and is,always partial, in the sense that it is neithervalue-free nor complete. In turn, this overtscepticism towards the possibility of absolutetruth aligned Livingstone’s history withaspects of postmodernism. Postmodern think-ing was a powerful influence on muchgeographical research in the 1990s, and con-tinues to be so (see Minca, Chapter 16 this

18


volume; Woodward and Jones, Chapter 15this volume). This, combined with the qual-ity of Livingstone’s scholarship, does muchto explain why The Geographical Tradition’sreputation has diminished little over time.

The author

Based in the School of Geography at Queen’sUniversity in Belfast, Northern Ireland,David Livingstone has worked his way upthrough the academic ranks, and is nowProfessor of Geography and IntellectualHistory. In the Royal Geographical Society’sBiographies of Chartered Geographers,Livingstone summarizes his research interestsas ‘history and theory of geography, cartogra-phy and scientific cultures’. He is recognizedin academic circles as both a geographer anda historian of science. It is important toappreciate this twin identity to understandthe scope and substance of Livingstone’swork, and the high regard in which he is heldby his peers. His consistent focus on the linksbetween geography and the wider sciences,arts and humanities means that he is far lessconcerned to work within orthodox (butoften arbitrary) disciplinary boundaries thanother historians of geography have been.Before The Geographical Tradition was written,Livingstone’s publications in the 1980saddressed such issues as the shortcomings oftraditional scientific method, the use of myth,metaphors and models in science, the absenceof an essential identity for academic geogra-phy and the interaction of magic, theologyand science. He is an expert on NathanielSouthgate Shaler, the nineteenth-centuryHarvard University polymath, and the influ-ence of the neo-Lamarckian version ofevolutionary thought on the environmentaland social sciences. The common factor thatunites these themes, all of which recur in TheGeographical Tradition, is suspicion of the core

qualities – progress, certainty, objectivity –traditionally claimed for modern science.

The text

Livingstone’s (1992) history of geography startswith a study of geography’s historiography.The Geographical Tradition’s first chapter is asustained critique of the ways in which geog-raphers typically have tried to write the historyof their discipline. The title of the chapter –Should the history of geography be X-rated? – wasborrowed from a 1974 paper in the journalScience by the US historian of science S.G.Brush, entitled ‘Should the history of sciencebe rated X?’ The orthodox view of scientificmethod, associated with positivist empiricism,emphasizes particular characteristics – mea-surement, rigour, logic, objectivity, etc. – thatguarantee reliable, truthful knowledge. Studentsare usually expected to absorb and adhere tothis ideal as part of their professional training.Brush’s paper highlighted new work by histo-rians of science that undermined the orthodoxyby documenting just how subjective (i.e.dependent on the thoughts and actions of thescientists themselves) real-world science was;hence Brush’s mischievous suggestion that ‘oldschool’ scientists should introduce censorshipto prevent impressionable students fromgetting hold of such seditious literature.Livingstone’s point here was that there was noneed to label existing histories of geographywith an X-certificate because geography’s his-torians had yet to pick up on the new ways ofwriting about science. The ‘sanitized’ historiesof the discipline available, bereft of ‘social con-text, metaphysical assumptions, professionalaspirations or ideological alliances’ (Livingstone,1992: 2), were far more likely to send studentsto sleep than to the barricades. By writing TheGeographical Tradition, Livingstone intended tointroduce geographers to revisionist histori-ographies of science and to illustrate the

154ÿÿNICK SPEDDING


possibilities of these with case studies selectedfrom geography’s past. Henceforth, uncriticalfaith in philosophical and methodologicalrigour guiding the impartial hero in his questto discover indisputable truth was out; empha-sis on the various human, or, to put it bluntly,‘non-scientific’ factors was in. Historians ofgeography now had to make room for people’sluck, misfortunes, whims, mistakes, friendships,rivalries, prejudices and vested interests.

Livingstone (1992) proceeds with a detailedcritique of the ‘standard’ textbook histories.His most basic objection is that these areboring – tedious lists of who said what,when. But he argues for a greater fault – textsthat present a distorted history because oftheir presentism and internalism. Presentist his-tory is history written to match present-daystandards, without taking account of howthings were different in the past. Internalisthistory is history written to match a narrow,often arbitrary, view of how things shouldbe, ignoring alternative perspectives or fac-tors that sit outside the preferred view.Livingstone’s (1992: 4) concern is that theseprejudices combine to produce Whiggish his-tories, ‘written backwards’. These accountsstart with a particular view of the correct wayto do geography and trace a selective historythat justifies this. Triumphant, inevitableprogress towards this present state of choice ismade as a succession of big names tread thepath of righteousness. Those who do not fitare dismissed as troublesome heretics, or areignored.

The outstanding example of Whiggish his-tory of geography is Richard Hartshorne’s1939 treatise on The Nature of Geography(see Livingstone, 1992: 8–9 and 304–316).Hartshorne tried to justify his philosophicalpreference for the discipline’s identity – it wasto be defined exclusively as the study of arealdifferentiation – with blatantly Whiggish tac-tics. His choice of subtitle – A Critical Survey ofCurrent Thought in the Light of the Past – was

a dead giveaway, as was the chapter on‘Deviations from the course of historical devel-opment’! It was not surprising that Schaeffer’s(1953) oft-cited attack on Hartshorne targetedboth his philosophy and his history. Schaeffer’spaper argued forcefully against ‘Exceptionalismin geography’ – Hartshorne’s belief that thediscipline had a special, integrative, mission thatset it apart from the other sciences – in favourof a ‘scientific’ law-seeking approach. Schaeferprotested that Hartshorne’s view of what geog-raphy should be was based on a selective,incorrect reading of past authorities, in partic-ular Kant and Hettner, and produced acounter-argument that was just as Whiggish,supporting his own partial view with anequally selective roll-call of heroic works,including those of von Thünen, Christaller andLösch. As William Bunge wrote to Hartshornein 1959, ‘history, as conducted in geographicmethodological discussions in general, canprove anything and therefore proves nothing’(Livingstone, 1992: 315).

If historical analysis cannot be used tosettle philosophical debates it can help usunderstand why people thought and didcertain things. This involves a shift frominwards-looking textual accounts of science –an impersonal focus on its logic, methods,data, hypotheses and laws – to outwards-looking, contextual accounts that stress subjectivefactors. Most scientists reject the contentionthat human inputs make a difference to theknowledge they produce, claiming that thelogic and procedures of ‘scientific method’neutralize subjectivity and so ensure that thefindings of science mirror natural reality;thus, scientific knowledge has a peculiar, priv-ileged status. The opposite point of view isthat scientific knowledge is always subjectiveprecisely because science is done by people; itis a social construction. This perspective is oftentraced back to Thomas Kuhn’s work onscientific revolutions and paradigm shifts, butit was developed in detail by sociologists of

DAVID LIVINGSTONEÿÿ155


science in the 1970s and 1980s (Livingstone,1992: 12–23). The most extreme versions ofthis school of thought maintain that it isfutile to think that we know anything of anon-human reality; that science is just as sub-jective as, say, religion or literature; and thatany one claim to knowledge is just as good asany other. Livingstone (1992) does not acceptthis out-and-out relativism; if nothing else, thepractical achievements of science surely indi-cate that we must know something abouthow the world really works. However, heinsists that historians of geography shouldtake social factors seriously as part of a his-tory that treats text and context astwo sides of the same coin. For example, tounderstand why Anglo-American geographyadopted regional studies with such enthusi-asm between the two World Wars, we mustlook beyond straightforward academic argu-ments. Powerful figures such as Hartshorneand Carl Sauer (see Livingstone, 1992:290–303) advocated the turn to particular, syn-thetic studies of areal differentiation as astrategic move that would distance the disci-pline from its disreputable environmentaldeterminist past, restore its intellectual andmoral purity, and provide a new, distinct iden-tity with which to counter the rise of socialsciences such as economics and sociology.

To summarize Livingstone’s (1992) corearguments in this first chapter, we might con-clude that, firstly, there is no correct way todo geography – it is only part-determined bythe state of reality or by philosophical argu-ment. Hence, any history of the disciplinethat assumes otherwise is wrong, Secondly,contextual, social factors also control the typeof geography that is done. These operate ata range of scales, from the individual tosociety-wide, and change over time andacross space. Thirdly, and following from this,‘… there can only be a situated geography.For geography has meant different things todifferent people in different places and thusthe ‘nature’ of geography is always negotiated.

The task of geography’s historians … is thusto ascertain how and why particular practicesand procedures come to be accounted geo-graphically legitimate and hence normativeat different moments in time and in differentspatial settings’ (Livingstone, 1992: 28).

To end the chapter, Livingstone (1992: 30)suggests that we should think of geography asa tradition that evolves over time – an obvi-ous reference to Darwin’s theory of speciesevolution. He describes this as a ‘rhetoricalflourish’, but it is an apt analogy that capturesthe kind of history he wants to write. Firstly,it negates Whiggish histories, as the trajectoryof evolution is contingent, not pre-determined.Secondly, it demands consideration of bothinternal and external factors, as transmutationof a species or tradition is the product of thatentity’s (whether organism or idea) interac-tion with its environment (whether natural orsocial).

The next eight chapters present case stud-ies intended to illustrate the kind of historythat Livingstone advocates. Much of thecore subject matter consists of episodes andpersonalities familiar from the standardtextbook histories; for example, the Age ofReconnaissance, Prince Henry the Navigatorand Christopher Columbus; the Enlightenment,and James Hutton’s discovery of ‘deep time’in his Theory of the Earth; the scientificvoyages of James Cook and Alexander vonHumboldt; the formal academic theories ofgeography’s first professors, such as HalfordMackinder, Friedrich Ratzel and WilliamMorris Davis; the struggle for the discipline’sscientific identity in the wake of the Schaeffer-Hartshorne clash … all of which I studied asan undergraduate between 1989 and 1992,just before The Geographical Tradition waspublished. The book’s most original sectionsare contributed by Livingstone’s own research:on magical geography, neo-Lamarckianism,and – in what is perhaps the book’s out-standing chapter – on the applications ofenvi ronmental determinism that established



geography as the practical science of race andempire. The book’s final chapter providesa useful summary of what Livingstone consid-ers to be the chief ‘conversations’, or discourses,that have occurred as the geographical traditionhas evolved.


The Geographical Tradition was enthusiasticallyreceived: most academics agreed that, despitepersonal objections to certain of Livingstone’sviews or omissions, it was by far the best historyof geography available. Reviewers highlightedLivingstone’s prodigious knowledge of primaryand secondary sources, the novelty and powerof his contextual stance, the pace and skill of hisstorytelling, and the elegance of his writing.Denis Cosgrove (1993) described the book as ‘ascholarly tour de force’; Audrey Kobayashi(1995) claimed that it was ‘absolutely the bestchoice for those often-resented obligatorycourses on the nature of geography’, as the ‘ravecomments’ of her students showed; AlanWerrity and Laura Reid (1995) went so far asto suggest that it ‘sen[t] one running down thecorridor doing cartwheels’! The science journalNature published the only openly hostile review,by historian William McNeill (1993). Underthe banner ‘Obsessive deconstruction’ McNeillattacked what he saw as Livingstone’s excessiverelativism, arguing that ‘the word “truth” isabsent from his vocabulary’, so that TheGeographical Tradition was ‘intellectual [i.e. igno-rant of realities] history with a vengeance’.

Yet Livingstone (1992) was correct todescribe the standard histories of geographyas boring lists of facts (I recall sitting in oneof my first lectures as an undergraduate, dili-gently writing down that Columbus’s threeships were called Niña, Pinta and Santa Maria,just as I had once done at primary school!).The Geographical Tradition set new standardsof authorship for disciplinary history that wascolourful, controversial and relevant. This was

a direct consequence of the fresh perspectivethat Livingstone advanced. His authoritativeknowledge of the wider history of sciencecontributed significant new insights thatinwards-looking histories of geography, con-strained by their concern to keep withinartificial disciplinary limits, had missed. Forexample, standard accounts of the voyages ofdiscovery focused on the new lands and seasadded to western Europe’s maps; Livingstonesaw far more than just new data, arguing thatthe empirical process of exploration promotednew standards for rational knowledge –prioritizing first-hand experience overestablished authority – that revolutionizedWestern science itself (Livingstone, 1992:32–35 and 59–62). However, it was his motifof the ‘contested enterprise’ that captured thegeographical imagination. The notion of thegeographical tradition as something that wassubjective, context-dependent and, as a result,flexible, offered a stimulating alternative toself-justificatory histories, often presented insimplistic paradigm terms. Livingstone’s (1992)history demonstrated that claims to authori-tative knowledge often had more to do withshifting power relationships than with ratio-nality or truth, so that the identity ofgeography was not fixed, but was up forgrabs. This was a message that a large numberof geographers starting to explore the possi-bilities of post-modernism very much wantedto hear. Livingstone (1992) was hesitant tomake links to postmodernism in the TheGeographical Tradition, with no reference to itin the first chapter. However, he did specifi-cally align the postmodern challenge toauthority with revisionist historiography inhis 1990 essay on ‘Geography, tradition andthe Scientific Revolution’, concluding that:

… it is now clear that classical founda-tionalism [the belief, which underwritespositivist scientific method, that onlyfirst-hand experience of something pro-duces valid knowledge] is in bad shape,



and had better be given up for dead. Inthis context geographers will have toacknowledge that warranted knowledgeis relative to a body of beliefs, not a bodyof certitudes. Inevitably, this will leadto pluralism in the geographical acad-emy… We now need to realise thatnon-foundationalist discourses – in thepolitical, the affective, the moral, theartistic, the cultural, the religious, anddoubtless a host of other spheres – are aslegitimate now as they were in the fif-teenth and sixteenth centuries… This hasfar-reaching consequences. [To addressthese] is the task that confronts the presentgeneration. (Livingstone, 1992: 370)

Livingstone uses this passage a secondtime to close his history (Livingstone, 1992:345–346). With such a call to arms that soobviously tapped into the Zeitgeist of humangeography (was Livingstone’s history just asWhiggish as Hartshorne’s?) it is not surpris-ing that The Geographical Tradition became an‘instant classic’.

Despite this, The Geographical Tradition wascontroversial. Critics soon identified omis-sions, with perhaps the most vociferouscomplaint being that Livingstone’s historyignored women. Kobayashi (1995: 194) sawthis as ‘a flaw so grievous as to be almostfatal’.The feminist objection was matched bypost-colonial critics who pointed out thatthe book focused on European and NorthAmerican traditions at the expense of others.It is not difficult to understand such criti-cisms – feminist and post-colonial thinkingwere an important part of human geogra-phy’s post-modern turn, which advocated anopenness towards multiple (as opposed toprivileged) viewpoints as one of its key fea-tures. Ironically, Mona Domosh (1991a,1991b) supported her attack on the sexismof David Stoddart’s On Geography (see below)with reference to Livingstone’s previouswork advocating alternative histories of geog-raphy! There is a second irony too – those

who saw The Geographical Tradition as out-of-step with the intellectual cutting edgeinvoked exactly the kind of presentist cri-tique that Livingstone cautioned against. Butwhat was important in 1992 seems lessimportant now. Livingstone (1992: 30–31)admitted from the start that his case studieswere selective so it makes little sense to attackThe Geographical Tradition for its omissions. IfLivingstone and Stoddart wrote historiesdominated by privileged, white, westernmales this was because their subject matter ofchoice – formal academic geography beforethe Second World War – was dominated byprivileged, white, western males. It is alsonow recognized that much of the initial workinspired by feminist and post-colonial theo-ries was over-vigorous in its identification ofhegemonic conspiracies. The spectre of anall-powerful masculinity was often counteredby unsubtle feminism that, despite its claim torespect pluralities, was just as essentializing.Subsequent work has tended to be morenuanced in its discussion of identity – see, forexample, McEwan (1998) – in part because itdisplays greater sensitivity to the particularitiesof place. This need for careful historical-geographic study of the history of geographyis evident in The Geographical Tradition, andhas been developed explicitly in Livingstone’ssubsequent work.

It is perhaps more pertinent to ask if TheGeographical Tradition matches Livingstone’sown objectives. Several of his case studiesunequivocally demonstrate the need for con-textual histories. For instance, we now seeenvironmental determinism as both blatantlyracist and scientifically absurd – but, becauseit set geography up as the science of empire,it endured as a core component of thegeographical tradition for over 50 years(Livingstone, 1992: Chapter 7). This exampleclearly shows that the success of science can-not be read off as a simple function of theextent to which it matches up to real-worldtruth. Elsewhere, however, Livingstone’s



history is less convincing. His account ofdevelopments after c. 1970 is disappointing.Temporal proximity cannot satisfactorilyexcuse what is little more than a list of bignames and big ideas – just the type of historyLivingstone wanted to discard. For what issupposed to be a contextual history, it is sur-prising that he makes almost no attempt toexamine the relationship between postmod-ern thought and the specific economic,political and cultural phenomena indicativeof postmodernity outside the ivory towers(see Chapter 15). Driver (2004: 231), wasright to point out that The GeographicalTradition is ‘a very book-ish book’, in that ittends to prioritize the details of particularideas. This is evident in Livingstone’s discus-sion of evolution. If positivism was importantto twentieth-century geography because ofits style, not its substance (Livingstone, 1992:321–322; see also Taylor, 1976), should it notalso be the spirit, not the detail, whichaccounts for the influence of evolutionarydoctrines on nineteenth-century geography?Yet Livingstone concentrates on the differ-ences between Lamarckian and Darwinianversions of evolution. This narrow focuson ideas is at odds with studies by sociologistsof science, introduced in The GeographicalTradition’s first chapter, which shows that it isimportant to follow what scientists do if weare to understand how and why knowledgearises (see Demeritt, 1996, for a review of this work).

It is true that The Geographical Traditiondemonstrated a new way to write the historyof geography – but it was not the first text todo so. Papers such as Taylor’s (1976) interpre-tation of the quantification debate as a socialpower struggle, in which the revolutionariesused mathematics as a weapon against theestablished regime, or Peet’s (1985) exposé ofenvironmental determinism as the ideologi-cal tool of imperial capitalism, had previouslyemphasized the importance of subjectivefactors. The first single-author book by a

geographer to take contextual history seri-ously was David Stoddart’s On Geography andits History, published in 1986. Stoddart’s(1986) first chapter presents a manifesto forcontextual histories of the discipline thatanticipates much of The Geographical Tradition’sfirst chapter. Livingstone’s enthusiastic reviewappeared in Progress in Human Geography in1987. So, if the two books are similar, why isit that The Geographical Tradition, not OnGeography, became established as the break-through text? Both are impressive works ofscholarship, although written in very differ-ent styles. But Livingstone’s history differedfrom Stoddart’s in several key aspects: hemade more use of work from the history andsociology of science; he was more flexible inhis attitude towards what geography was, orshould, be; he was far more sceptical aboutthe power of empirical scientific method; andhe was more sensitive to matters of discourseand representation. This did not make TheGeographical Tradition a better book, but it didmake for a better match with the wants andexpectations of those attracted by the possi-bilities of post-modernism. As Livingstone’srecent research shows (Livingstone, 2005),the context in which a book is read can bejust as important as what it says. Written by ahuman geographer with strong affinities forthe humanities (Stoddart is a physical geog-rapher), and published six years after Stoddart’sbook in 1992, by which time the ‘culturalturn’ was in full swing, the impact of TheGeographical Tradition must itself be under-stood in both textual and contextual terms. Itwas very much a case of the right message inthe right place at the right time.

Conclusion

Because of The Geographical Tradition’s timelyappearance amidst a sea of persuasive, com-plementary ideas (see Driver, 1995), it isimportant not to claim too much for a single



book. But it is clear that The GeographicalTradition inspired geographers to rethink thefoundations of their discipline. Livingstone’s(1992) notions of the ‘geographical tradition’and the ‘contested enterprise’ provided pow-erful metaphors that soon became part of thediscipline’s everyday language, uniting geog-raphers with diverse outlooks – in debate, ifnot agreement (Mayhew et al., 2004: 229).Undergraduate courses on the history andphilosophy of geography had to be re-written.The history of geography, often thought ofas the poor relation of ‘proper’, contempo-rary research, received a major boost. Byplacing them firmly in the orbit of postmod-ernism The Geographical Tradition helped tomake historical studies of knowledge bothfashionable and relevant (but see Barnett,1995, for a very different view of history’srelevance). If the first of the new batch ofhistorical case studies took a predictable direc-tion under the influence of feminism and/orpost-colonialism – Barnett (1995: 418) com-mented that ‘it would be quite disastrous forthe theoretically inclined human geographerif their discipline did not have a dubiousimperialist past’ – more recent work hasopened up different times and places. Muchof this new work foregrounds the actions ofparticular personalities in particular localities,and so corrects the tendency of past casestudies, including some of those in TheGeographical Tradition, to isolate the intellec-tual. It is now common for geography’shistorians to promote the principle – pleas-ingly geographical, but firmly non-rational –that the who and the where of scientific

enquiry make a difference to the content ofthe knowledge produced. Livingstone’s con-tribution is important here (Livingstone,2003, 2005), but perhaps the outstandingexample – because, unusually, it examinesnear-contemporary history – is Trevor Barnes’swork which sets out a ‘post-prefixed’ inter-pretation of economic geography (e.g. Barnes,1996, 2001; see Kelly, Chapter 23 this volume).

It is ironic that, by stimulating cross-disciplinary revival of historical and philo-sophical debate, The Geographical Traditionundermined its core premise – that it makessense to study a single body of beliefs andactions that constitute the geographical tradi-tion. Livingstone (1992: 420) dislikes what hecalled ‘the postmodern pluralization impera-tive’ but his argument that the tradition existssimply because it is possible to tell its story,however varied that story might be, looks farfrom convincing 15 years on. Much of whatgeography had, or still has, in common comesfrom the contingent co-location of its practi-tioners in Departments of Geography. Persistentattempts to start conversations across assorteddivides (e.g. Harrison et al., 2004) perhapsindicate a collective desire for disciplinaryidentity founded on necessity, but often dolittle more than emphasize the multitude oftraditions in which geographers now partici-pate. If we follow through Livingstone’sanalogy of tradition-as-species, this pluralismshould not surprise us (or him), for as speciesevolve differences accumulate, so that, atsome point and in some place, new speciesdistinct from, and often incapable of fertileinteraction with, their ancestors appear.


Secondary sources and referencesBarnes, T.J. (1996) Logics of Dislocation. New York: Guilford Press.Barnes, T.J. (2001) ‘Lives lived, and lives told: biographies of geography’s quantitative

revolution’, Environment and Planning D: Society and Space 19 (4): 409–429.Barnett, C. (1995) ‘Awakening the dead: who needs the history of geography?’

Transactions of the Institute of British Geographers NS 20 (4): 417–419.


Cosgrove, D. (1993) ‘Book review of The Geographical Tradition’, Progress in HumanGeography 17 (4): 583–585.

Demeritt, D. (1996) ‘Social theory and the reconstruction of science and geography’,Transactions of the Institute of British Geographers NS 21 (3): 484–503.

Domosh, M. (1991a) ‘Toward a feminist historiography of geography’, Transactions of theInstitute of British Geographers NS 16 (1): 95–104.

Domosh, M. (1991b) ‘Beyond the frontiers of geographical knowledge’, Transactions of theInstitute of British Geographers NS 16 (4): 488–490.

Driver, F. (1995) ‘Geographical traditions: rethinking the history of geography’, Transactionsof the Institute of British Geographers NS 20 (4): 403–404.

Driver, F. (2004) ‘Commentary two (Classics in Human Geography Revisited – theGeographical Tradition)’, Progress in Human Geography 28 (2): 229–233.

Harrison, S., Massey, D., Richards, K.S., Magilligan, F.J., Thrift, N. and Bender, B. (2004)‘Thinking across the divide; perspectives on the conversations between physical andhuman geography’, Area 36: 435–442.

Kobayashi, A. (1995) ‘Book review of The Geographical Tradition’, Annals of theAssociation of American Geographers 85 (1): 191–194.

Livingstone, D.N. (1990) ‘Geography, tradition and the Scientific Revolution: an interpreta-tive essay’, Transactions of the Institute of British Geographers 15 (3): 359–373.

Livingstone, D.N. (1992) The Geographical Tradition: Episodes in the History of aContested Enterprise. Oxford: Blackwell.

Livingstone, D.N. (1995) ‘Geographical traditions’, Transactions of the Institute of BritishGeographers NS 20 (4): 420–422.

Livingstone, D.N. (2003) Putting Science in its Place. Chicago: University of ChicagoPress.

Livingstone, D.N. (2005) ‘Science, text and space: thoughts on the geography of reading’,Transactions of the Institute of British Geographers NS 23 (4): 420–422.

McEwan, C. (1998) ‘Cutting power lines within the palace? Countering paternity and euro-centrism in the “geographical tradition” ’, Transactions of the Institute of BritishGeographers NS 23 (3): 371–384.

McNeill, W. (1993) ‘Obsessive deconstruction. Book review of The Geographical Tradition’,Nature 362: 218 (18 March 1993).

Mayhew, R., Driver, J.F. and Livingstone, D.N. (2004) ‘Classics in human geography revis-ited: Livingstone, D.N., 1992, The geographical tradition: episodes in the history of acontested enterprise’, Progress in Human Geography 28 (2): 227–234.

Peet, R. (1985) ‘The social origins of environmental determinism’, Annals of theAssociation of American Geographers 75 (3): 309–333.

Stoddart, D.R. (1986) On Geography and its History. Oxford: Blackwell.Taylor, P.J. (1976) ‘An interpretation of the quantification debate in British geography’,

Transactions of the Institute of British Geographers NS 1 (2): 129–142.Werrity, A. and Reid, L. (1995) ‘Debating the geographical tradition’, Scottish Geographical

Magazine 111 (3): 196–197.



FEMINISM AND GEOGRAPHY (1993):GILLIAN ROSE

Robyn Longhurst

In exploring the masculinism of geographyat some length, this is not a book about thegeography of gender but about the gender ofgeography. (Rose, 1993: 4–5)

Introduction

Gillian Rose wrote Feminism in Geography:the limits of geographical knowledge early in hercareer when she was Lecturer in Geographyat Queen Mary and Westfield College,University of London. Soon after, Rose wasappointed Lecturer in Geography at theUniversity of Edinburgh, moving later to aProfessorship of Geography at the OpenUniversity. Prior to the publication ofFeminism and Geography, feminist geographershad focused attention mainly on empiricalstudies of women. The 1984 landmarkGeography and Gender by the Women andGeography Study Group (see Hanson,Chapter 11 this volume) contained severalchapters that cover a range of feminist theo-retical and methodological approaches, butfocused mainly on the effect of urban spatialstructures on women, women’s employment,women’s access to facilities, and women anddevelopment. Geography and Gender thus laidan important foundation for feminist geog -raphy but its primary aim was to examine‘geographies of gender’. Rose’s primaryaim in Feminism and Geography was to exam-ine the ‘gendering of geography’. At thebeginning of the 1990s geography was

severely in need of such an in-depth andsustained feminist examination.


When Feminism and Geography arrived in themail (it was not available in New Zealandbookstores so I had ordered it) I was writinga feminist-informed doctoral thesis on preg-nant bodies in public spaces. I was trying tothink through how bodies (which are com-monly associated with Woman, irrationality,and the private sphere) had been excluded ingeography but I wasn’t quite sure whichbodies had been excluded, in which geogra-phies, and what purpose this served. I hadbeen looking forward to the arrival of Rose’sbook hoping it might cast some light on theseissues. I was not disappointed. I devouredthe 200 page text over a weekend (it was anintense read because, as Karen Morin, 1995,notes ‘a lot is packed into this small volume’).The following Monday I recall contactingmy doctoral supervisor to report that Rose’sbook had prompted a breakthrough in myresearch.

Needless to say Feminism and Geographydidn’t just have an impact on me. At the timeof its publication, the listserv ‘Feminismin Geography’ ([email protected]) had not long been established and thebook quickly became the subject of muchdiscussion most of which centred uponunpacking the ideas contained within the

19


164ÿÿROBYN LONGHURST

text. There is a reasonably heavy reliancethroughout the book on academic prosewhich makes the work challenging, espe-cially for new or emerging scholars. In areview of Feminism and Geography, BriavelHolcomb (1995: 264) says the book is writtenin ‘the arcane language of critical social the-ory’. The listserv discussion was invaluable inthat it provided an arena for many of us toshare our ideas and opinions on the many andcomplex arguments contained in the book.

However, not only was Feminism andGeography discussed informally on listservessuch as ‘Feminism in Geography’ but it wasalso reviewed extensively in geography jour-nals such as Annals of the Association ofAmerican Geographers (Falconer Al-Hindi,1996), The Geographical Journal (Burgess,1994), Transactions of the Institute of BritishGeographers (Kofman, 1994) and GeographicalReview (Holcomb, 1995). The book’s influ-ence also extended beyond geography. It wasreviewed in feminist studies journals such asFeminist Review (Bondi, 1995) and Signs(Hayden) and in journals in other disciplinessuch as The Journal of American History(Norwood, 1994) and Postmodern Culture(Morin, 1995).

Rose (1993: 1) begins Feminism andGeography by explaining ‘The academic disci-pline of geography has historically beendominated by men, perhaps more so than anyother science’. Men have been over-representedin professional geography organizations, ingeography departments and in academicpublishing. Men’s interests, therefore, havestructured what counts as legitimate geo-graphical knowledge. Rose makes the pointthat women’s exclusion in geography is notjust about the themes of research, nor evenabout the concepts feminists employ to orga-nize those themes, but that there issomething in the very claim to knowing ingeography which tends to exclude women asproducers of geographical knowledge.

Rose (1993: 15) says she ‘desperatelywanted to be able to join in, to be part ofdebates among knowledgeable men’ and was‘seduced’ by the academy but occupied anambiguous position. Rose was empoweredby her whiteness but at the same time mar-ginalized as a woman. When I first enrolled asa young, white, working-class, female univer-sity student in 1980 I too felt enormouslyattracted to the academy and in particular tothe discipline of geography but also excludedfrom it. I felt excluded from discussions andactivities in lectures, tutorials, and on field-trips. I also felt excluded when I chose tostudy topics that were deemed to be inap-propriate for geographical study, such as in1985 when I wanted to write a Master’s the-sis on sexual violence against women. Manyyears later as a Lecturer I revelled in reading,teaching and researching geography but stillfelt like an impostor in the discipline, not juston account of the topics I chose to study (Iwas once told by a geography journal editorthat pregnant bodies were an inappropriatesubject for geography) but at a much deeperlevel. Rose’s text spoke to me. I too felt posi-tioned uneasily in the discipline. Feminismand Geography provided me with a frame-work within which to understand further mysense of being both an insider and an outsiderin geography.

Feminism and Geography challenges geog -raphers to think critically about the waygeographers produce particular kinds ofknowledge. Rose (1993) argues there is aspecific notion of knowing, and of knowl-edge, as masculine, exhaustive, rational, andassociated with the mind rather than the body(as though the two can be separated) whichmarginalizes women in the production ofgeographical knowledge.

In order to mount this argument about theintersection of power, embodied subjectivityand knowledge, Rose (1993) relies heavily onthe intellectual cross-fertilization of feminist


GILLIAN ROSEÿÿ165

poststructuralist, psychoanalytical and geo-graphical theory. Rose was highly influencedby Doreen Massey’s (1984: 129) pioneeringwork on the Spatial Divisions of Labour (seePhelps, Chapter 10 this volume) which shesays encouraged ‘studies of the intricategeography of gender and class accessedthrough local studies’. But Rose also drawson the feminist theorizing of Rosi Braidotti,Marilyn Frye, Diana Fuss, Moira Gatens,Donna Haraway, Luce Irigaray, Michéle LeDoeuff, Elspeth Probyn, and Iris MarionYoung. She also draws on the work of queertheorists such as Teresa de Lauretis and EveSedgwick. Contributions from black andpostcolonial feminists such P. Hill Collins,bell hooks and Gatari C. Spivak are also usedto further the argument that geography is a‘masculinist’ and white discipline.

Rose adopts the term ‘masculinist’ fromMichéle Le Doeuff (1991: 42, cited in Rose,1993: 4) who defines it as ‘work which, whileclaiming to be exhaustive, forgets aboutwomen’s existence and concerns itself onlywith the position of men’. Rose describesgeography as ‘masculinist’ because it has pro-duced grand theories that claim to speak foreveryone but in actual fact speak only forheterosexual, white, bourgeois men. The sub-stance of Rose’s argument is that there are atleast two different kinds of masculinity atwork in the discipline. She refers to the firstas ‘social-scientific masculinity’. This kind ofmasculinity ‘represses all reference to itsOther in order to claim total knowledge’(1993: 10–11). Rose critiques the ‘social-scientific masculinity’ in the time-geographyof Swedish geographer Törsten Hägerstrand(see Lentrop, Chapter 1 this volume). Sheexplains that while time-geography was use-ful, in that some of the earliest feministgeographers adopted it in order to ‘recoverthe everyday and the ordinary’ (Rose, 1993:22) that are often seen to typify women’slives, it was not able to account for the kind

of embodied subjectivity produced by theroutine work of mothering and domesticitythat occurs mainly in the home. Hence,‘women and their feelings somehow got lost’(Rose, 1993: 27) in time-geography.

Rose refers to the second kind of masculin-ity in geography as ‘aesthetic masculinity’. Thiskind of masculinity ‘establishes its powerthrough claiming a heightened sensitivity tohuman experience … [it] admits the exis-tence of its other in order to establish aprofundity of which it alone has the power tospeak’ (Rose, 1993: 11). ‘Aesthetic masculin-ity’ can be found in humanistic geographywhich establishes place as key concept, partlyas a response against positivism which becameso popular in the discipline in the 1960s.‘Man’ was to be put back into the centre ofthings (Rose, 1993: 43). Humanistic geogra-phers, however, tend to feminize place whichis conceptualized as the idealized Woman.They seek to know place exhaustively, whileat the same time asserting its mysteriousunknowability.

Rose’s critique of masculinism, however,does not stop with time-geography andhumanism. She also critiques a strand ofradical feminism by examining the dualismbetween Nature (feminine) and Culture(masculine). Geographers often display con-tradictory impulses towards Nature (feminine).They fear Nature and want to dominate itbut also desire it, revelling in its magnificenceand beauty. Rather than problematize thisdualism, many radical feminists simply invertit. Rose also turns her attention to culturalgeography, in particular to the concept oflandscape. Cultural geographers, she argues,fail to address their own pleasure in look-ing at landscape (see Gilbert, Chapter 12 thisvolume).

Rose understands feminist geography tobe both complicit with and resistant to mas-culinism. In some ways feminist geographershave challenged the masculinism inherent in



the discipline (for example, by consideringproduction and reproduction) but in otherways they have been complicit with mas-culinism (for example, by invoking a conceptionof Woman that loses sight of the diversityamongst women). This critical mobility thatpositions feminists both inside and outside ofgeography, as both complicit and resistant toits masculinism, offers strategic possibilities for‘imagining something beyond the discipline’shegemonic imagination’ (Rose, 1993: 117).

Impacts and effects

Feminism and Geography was hugely impor-tant in propelling ideas in certain directionsin the discipline of geography. First, itencouraged geographers, including feministgeographers, to reflect critically on their/ourown role as producers of knowledge. Rose(1993) acknowledges that feminism itself iscaught in existing masculinist discourses andthat there is no space of purity from whichWoman or women can speak. She acknowl-edges the challenge that feminism faces inneeding to build an identity for Woman,while at the same time recognizing the diver-sity amongst women in relation to race, class,and sexuality and calls for a ‘strategic’ or ‘crit-ical’ mobility (Rose, 1993: 13) as a form offeminist resistance. This idea has contributedrichly to arguments about researcher reflex-ivity and postionality that began to emergein the discipline in the early 1990s (e.g.see McDowell, 1992 and a ‘Special Issue ofWomen in the Field’ in Professional Geographer,1994).

A second contribution Rose made to bothgeography and feminist studies was to furtherthe critique of dualistic thinking. Other femi-nists, including feminist geographers such asLiz Bondi (1992) and Dina Vaiou (1992) had,prior to the publication of Rose’s (1993) book,begun to illustrate how geographical dis-course focused on the distinctions between

culture/nature, public/private, production/reproduction, western/oriental, work/home,and state/ family. They also made the argu-ment that these dualisms were gendered butRose was the first to offer a book-length,sustained engagement with this notion.

The third contribution Feminism andGeography made was to prompt work on‘embodied geographies’. Rose (1993: 7)claims ‘Masculinist rationality is a form ofknowledge which assumes a knower whobelieves he can separate himself from hisbody, emotions, values, past experiences andso on’. This allows for him to consider histhoughts (his mind) to be autonomous, tran-scendent and objective; mess and matter-freeso to speak. She points out that ‘the assump-tion of an objectivity untainted by anyparticular social position’, or any particularbody, allows masculinist rationality to ‘claimitself as universal’ (Rose, 1993: 7). These ideashelped give rise to a new area of study –‘embodied geographies’ (e.g. Duncan, 1996and Teather, 1999). Lise Nelson and JoniSeager (2005: 2) claim ‘The body is thetouchstone of feminist theory’ but up untilthe publication of Feminism and Geographyfeminist geographers had been more influ-enced by liberal and socialist, rather thanradical, versions of feminism in which ‘thebody’ was not centre-stage.

A fourth impact the book had was to openup the discipline of geography to the possi-bility of using feminist psychoanalysis tounderstand better the relationship betweenthe social, spatial and psychic. In turn, Roseopened up the discipline of feminist psycho-analysis to the possibility of focusing on issuesof spatiality. Rose claims that feminist psy cho-analytic commentaries offer an eloquentcritique of geography’s white, heterosexual,masculine gaze, a gaze torn between pleasureand its repression (Rose, 1993: 103). In 1993only a handful of geographers were engagingwith psychoanalysis, perhaps most notablySteve Pile, who in 1996 published The Body


GILLIAN ROSEÿÿ167

and the City. It was not until the latter half ofthe 1990s that feminist geographers and othersreally began to engage with psychoanalyticapproaches to bodies in an attempt to under-stand more about the psychological acquisitionof gendered, sexed and racialized identities andrelationships with others (e.g. Nast, 2000).

A fifth contribution made by Feminism andGeography is Rose’s reimagining of a geogra-phy that is not based on masculinism or adualist system that dominates the Self andmarginalizes the Other. Rose (1993) askshow might it be possible to challenge the dis-cipline so that hegemony of Man/Womangives rise to geographies of difference? Toanswer this question she puts forward thenotion of ‘A Politics of Paradoxical Space’that acknowledges the power of hegemonicdiscourse but also insists on the possibility ofresistance (Rose, 1993: 155). Feminism needsto occupy both the centre and the margin –to be mobile, multiple, and contradictory – soas to ‘threaten the polarities which structurethe dominant geographical imagination’ (Rose,1993: 155). This concept, which entails aradical rethinking of place, space and gender,has opened up possibilities for new ways ofthinking about people–place relationships(see Bondi and Davidson, 2005, 20–25 onhow feminist geographers have used the con-cept of ‘paradoxical space’).

Finally, Rose’s (1993) text changed theway that geographers and others in disci-plines such as landscape architecture, urbanplanning, feminist studies and cultural studiesengage with the visual. In Chapter Five –‘Looking at Landscape: The Uneasy Pleasuresof Power’ – Rose opens up for question thepolitics of looking, and the conflation of see-ing and knowing. This strikes to the heartof the discipline. This is an area that Roseherself has developed in theoretical andempirical research over the past decade(e.g. see Rose, 2003). She has also writtenabout methodologies for interpreting visualmaterials (Rose, 2001). Looking, seeing, and

knowing (the package deal offered by mostgeography fieldtrips) have long been main-stays of spatial disciplines such as geography.

Despite these contributions the book hasnot been without it critics. Holcomb (1995:264) suggests that Rose’s ‘sweeping condem-nation of most male-produced geographysuggests the very tendency to essentialize ofwhich she accuses men’. Holcomb explainsthat she herself became a geographer becauseshe ‘loved fieldwork’ and so was dismayed toread that this ‘initiation ritual [of] toughheroism establishes fieldwork as a particularkind of masculine endeavour’ (Rose, 1993,cited in Holcomb, 1995: 264–265).

Eleonore Kofman (1994: 496) notes that‘the repeated and insistent attacks on the tex-tual strategies of male cultural geographersleft [her] reeling’. Kofman (1994: 497) claimsshe is ‘not at all convinced that oscillatingbetween difference and unity … will under-mine masculinism’. Kofman also critiques thebook on the grounds that it lacks historicaldepth in the recounting of the masculinenarrative of the discipline and that ‘there area number of epistemological issues whichhave influenced the positions adopted in thebook but which have not been made explicit’(Kofman, 1994: 497).

Vera Norwood (1994: 834) reiterates someof Kofman’s points. Norwood argues there isinsufficient historical analysis of key men anddocuments in Feminism and Geography to‘convince even a sympathetic reader’ ofRose’s case. She says Rose (1993) criticizesmen in the discipline of creating overlyabstracted and disembodied figures of ageneric Woman but she ‘falls in the samepractice in her descriptions of male col-leagues’ (ibid.). Norwood (1994: 833) argues‘Feminism and Geography is a very ambitiousbook that often falls short of its project’.Some reviewers also criticized the ‘produc-tion’ of the book. Kofman (1994) notesthe book does not contain an alphabeti-cally ordered bibliography, which means the



reader is endlessly searching the notes forreferences. Morin (1995) points out that‘Occasional misspellings don’t help matters,and the book’s three illustrations are merelyadequately reproduced’.

Conclusion

Like any book, this one undoubtedly has itspros and cons, its supporters and detractors. Ithas been criticized on some fronts but over-all I think it is fair to say the book hasreceived overwhelmingly positive reviews.Jacquelin Burgess (1994: 226), for example,describes the book as ‘subtle and sophisti-cated’. Liz Bondi (1995: 133) says ‘Rose’scritique of geographical knowledge is pow-erful and far-reaching’. Since its publicationin 1993 Feminism and Geography has undoubt-edly proven itself to be a key text. It has beenread widely and debated vigorously. Rosesucceeded in producing a book that is

indispensable for any feminist geographer orother scholar attempting to come to gripswith feminist and/or geographical thought.Linda McDowell, on the back cover ofFeminism and Geography, comments: ‘[Rose’s]book will become essential reading foreveryone interested in philosophical andmethodological issues in geography’. McDowellwas correct; the book has become essentialreading. Of course there are still some whowould disagree and consider Rose’s (1993)insights to be misinformed, unimportantand/or peripheral to the history of geography.But maybe this is the real testimony to thebook’s success; in some ways it hasn’t slotted ineasily to the geographical canon because it stillposes a radical challenge to much of that canon.Fourteen years after publication the book itselfstill occupies something of paradoxical space.Whilst in some ways it has become a key textin the discipline, in other ways it remains ‘ille-gitimate’ and unlikely to be read by those whowould benefit most from its insights.

Secondary sources and referencesBondi, L. (1992) ‘Gender and dichotomy’, Progress in Human Geography 16 (1): 98–104.Bondi, L. (1995) ‘Review of Feminism and Geography by G. Rose’, Feminist Review 51:

133–135.Bondi, L. and Davidson, J. (2005) ‘Situating Gender’, in L. Nelson and J. Seager (eds) A

Companion to Feminist Geography. Malden, MA: Blackwell, pp.15–31.Burgess, J. (1994) ‘Review of Feminism and Geography by G. Rose’, The Geographical

Journal 160 (2): 225–226.Duncan, N. (1996) (ed.) BodySpace. London: Routledge.Falconer Al-Hindi, K. (1996) ‘Review of Feminism and Geography by G. Rose’, Annals of

the Association of American Geographers 86 (3): 610–611.Hayden, D. (1997) ‘Review of Space, Place and Gender by D.B. Massey and Feminism

and Geography by G. Rose’, Signs 22 (2): 456–458.Holcomb, B. (1995) ‘Review of Feminism and Geography by G. Rose and Gender,

Planning and the Policy Process by J. Little’, Geographical Review 85 (2): 262–265.Kofman, E. (1994) ‘Review of Feminism and Geography by G. Rose’, Transactions of the

Institute of British Geographers 19 (4): 496–497.Le Doeuff, M. (1991) Hipparchia’s Choice: an Essay Concerning Women, Philosophy,

etc. Oxford: Blackwell.McDowell, L. (1992) ‘Doing gender: feminism, feminists and research methods in human

geography’, Transactions of the Institute of British Geographers 17: 399–416.


GILLIAN ROSEÿÿ169

Massey, D. (1984) Spatial Divisions of Labour. Basingstoke: Macmillan.Morin, K. (1995) ‘Review essay: the gender of geography’. Review of Feminism and

Geography by G. Rose, Postmodern Culture 5: 2.Nast, H.J. (2000) ‘Mapping the “unconscious”: racism and the Oedipal family’, Annals of

the Association of American Geographers 90 (2): 215–255.Nelson, L. and Seager, J. (2005) ‘Introduction’, in L. Nelson and J. Seager (eds) A

Companion to Feminist Geography. Malden, MA: Blackwell, pp.1–11.Norwood, V. (1994) ‘Review of Feminism and Geography by G. Rose’, The Journal of

American History 18 (2): 833–834.Pile, S. (1996) The Body and the City: Psychoanalysis, Space and Subjectivity. London:

Routledge.Professional Geographer (1994) Special Issue on Women in the Field 46 (1).Rose, G. (1993) Feminism and Geography. Cambridge: Polity Press.Rose, G. (2001) Visual Methodologies: An Introduction to the Interpretation of Visual

Materials. London: Sage.Rose, G. (2003) ‘Just how, exactly, is geography visual?’ Antipode 35: 212–221.Teather, E. (1999) (ed.) Embodied Geographies: Spaces, Bodies and Rites of Passage.

London: Routledge. Vaiou, D. (1992) ‘Gender divisions in urban space: Beyond the rigidity of dualist classifi-

cations’, Antipode 24 (2): 247–262.Women and Geography Study Group of the IBG (1984) Geography and Gender: An

Introduction to Feminist Geography. London: Hutchinson.


GEOGRAPHICAL IMAGINATIONS (1994):DEREK GREGORY

John Pickles

All I seek to do is make a series of incisions intothe conventional historiography of geographyand show that its strategic episodes can be madeto speak to many other histories. (Gregory,1994: 14)

Introduction

From his first book, Ideology, Science andHuman Geography (1979), Gregory has beenone of the most erudite voices arguing for theimportance of critical theory in geography,shaping its form and interpreting its wider pos-sibilities for geographers and non-geographersalike. Through Regional Transformation andIndustrial Revolution (1982) and Social Relationsand Spatial Structures (Gregory and Urry, 1985)to Geographical Imaginations (1994), and subse-quently through The Colonial Present (2004),and, with Allan Pred, Violent Geographies(2006), Gregory has charted a complex andbroad geographical project of critical theory.This is a project of re-working geographicaltheory to sustain a conversation about histor-ical materialism and human agency, and in away that resists the artificiality of disciplinaryboundaries and institutions. For Gregory, disci-plines are enabling institutions, but he insiststhey must not be binding limits. Instead, heproposes critical human geography as part of atrans-disciplinary (even post-disciplinary)project integrating geographical, social,and cultural theory to understand the pro-duction of everyday bodies, places, andspaces.

Founded on a theory of history and geogra-phy that rejects disciplinary hagiography andgrand theory, Geographical Imaginations seeks tomodel a politics of sense-making withoutuniversals or absolutes, one that questionsgrand modernist narratives in favor of moregrounded and contextual cultural theory.Geographical Imaginations was published in 1994in the wake of Ed Soja’s Postmodern Geographies(1989, see Minca, Chapter 16 this volume),David Harvey’s The Condition of Postmodernity(1989, see Woodward and Jones, Chapter 15this volume), the English translation of HenriLefebvre’s The Production of Space (1991), andSusan Buck-Morss’s rendering of WalterBenjamin’s Arcades Project in Dialectics of Seeing(1989), all texts which problematize the mod-ernist project and with which GeographicalImaginations engages in turn, exploring theconnections between human geography andcritical social theory, and teasing out from themthe political value of theoretical work and theinherently geopolitical role of theory.

Writing against ‘the deadening proclama-tions about the “nature” or “spirit andpurpose” of geography, Gregory (1994: 78)turns instead to political economy, social the-ory, and cultural studies, which he defines‘as a series of overlapping, contending andcontradictory discourses that seek, in variousways and for various purposes, to reflectexplicitly and more or less systematically onthe constitution of social life, to make socialpractices intelligible and to intervene in theirconduct and consequences.’ As David Harvey(1995: 161) has pointed out: ‘It is plainly his

20


intention to take what he considers the verybest of geographical work and treat it on apar with some of the very best writings fromphilosophy, sociology, anthropology, and thelike and to illustrate the major contributionsthat geographers have made, and are making,and can make to social and literary theory.’Both Harvey (1995) and Katz (1995) havequestioned the wisdom in Gregory’s work ofeliding the geographical imaginations atwork in physical geography and the naturalsciences, and this is partly the case inGeographical Imaginations, but in terms of itsengagements with the humanities and socialsciences Geographical Imaginations remainsamong the most well developed and articu-lated texts of critical geographical theory.What began as an attempt to explain geo-graphical ideas and history for geographersended up as a text about geographical ideasand imaginations well beyond the limits ofdisciplinary boundaries.

Background: the book and its author

As with any text, it is useful to readGeographical Imaginations as an elaboration ofthe author’s earlier works, especially Ideology,Science and Human Geography, in which heinterrogated three central theoretical tradi-tions in geography: positivism, Marxism, andhumanism. In this earlier book he developeda sustained critical theory of the history andnature of positivism (from Auguste Comte tothe logical positivist and logical empiricistsand their goal of unified science to the emer-gence, promise, and dominance of spatialscience in Geography), humanist approachesto Geography (including phenomenology,hermeneutics, and interpretative approachesmore generally), and critical emancipatorygeographies (such as Marxism).

Three aspects of this earlier work remainespecially important in Geographical Imaginations.First, Gregory draws on Frankfurt Schoolcritical theory and particularly Jürgen

Habermas’s Theory and Practice (1988) and relatedwritings to develop an epistemological critiqueof essentialism and functionalism in spatial sci-ence, and to demonstrate the always interestedand political commitments of theory. Second,in Regional Transformation he takes up the chal-lenge of Marxist historian E.P. Thompson towrite historical geography and politicaleconomy from the perspective of the every-day struggles of people in particular places,not as products of structures taking shapebehind their backs but as active agents shap-ing their worlds and making their ownhistories, albeit not under conditions of theirown choosing:

Thompson regards historical eventuationas an existential struggle whose recoveryrequires a recognition of ‘the crucialambivalence of our human presence in ourown history, part-subjects, part-objects,the voluntary agents of our own involun-tary determinations.’ For, if history is‘unmastered human practice,’ and ifits subjects are ‘ever-baffled and ever-resurgent human agents’ whose effectivity‘will not be set free from ulterior determi-nate pressures nor escape determinatelimits,’ there is nevertheless a space for theinsistent return of conscious, knowledgeableagency. (Gregory, 1982: 9–10)

Third, in Social Relations and SpatialStructures (edited with John Urry, 1985) andsubsequently in Human Geography: Society,Space and Social Science (edited with RonMartin and Graham Smith, 1994), he focusedmore directly on the role of ‘human agencyin human geography’ in ways that highlightthe connections between space and society,structure and agency, and economy and cul-ture. His engagements with historicalgeography, political economy, and culturalpolitics illustrate this commitment to con-ceptual border crossing that sustains thepolitical power of theoretical work – a powerthat ‘lies in some part in its power to inter-rupt, displace, and call into question the

172ÿÿJOHN PICKLES


taken-for-granted of the world and ourplace in it’ (Gregory, 1995: 177).

Geographical Imaginations ranges across abroad intellectual landscape. Throughout,Gregory reads texts as opening possibilitiesfor dialogue among different traditions. As hewrites:

It has to be possible – and it is an impor-tant part of my project, in these essaysand elsewhere – to rise above the cynicaldisparagement of theoretical work, tointerrogate its other, creative, imaginativeand productively political values. (Gregory,1994: 49)

The result is a rich and heady wine; acultural politics of spatial practices that rangesacross the mapping impulse, spatial analysis,contemporary urbanisms, social action, andthe production of space.

Structure and arguments

Geographical Imaginations is organized in threeparts, each with an introductory essay andtwo chapters. Part One, ‘Strange Lessons inDeep Space’, was written last and can be seenas a charting of the ground for his subsequentwriting on Edward Said and the ‘MiddleEast.’ It focuses on the ways in which spaceand representation have been thought andacted upon. He begins with the visual regimethat ‘rendered’ the world as a representation –i.e. as a picture, an exhibition, and as a mereobject-form. Gregory shows how this visualregime worked to shape the colonial gazeand with it the universalist claims of Europeanscience. This was the ‘god-trick’ that pro-duced nature and its local inhabitants asmerely resources for settlement and exploita-tion. It was also the founding moment forthe creation of a view of knowledge thatassumed a separation between subject–object,and it gave rise to a deep epistemologicalbreak in geographical thinking; a carto-graphic anxiety that parallels what Richard

Bernstein (1983) had characterized as theCartesian Anxiety. Part One of GeographicalImaginations clarifies the hold this visualregime has on contemporary social and spa-tial thought and begins the process ofreworking these naturalized notions of spaceand representation, ‘remapping spaces ofpower-knowledge’ (Gregory, 1994: 33) todevelop a more thorough-going conjuncturalcultural analysis of place, space, and landscape.He maps out the ways in which modern spa-tial science has been ‘socialized’ by itsengagement with political economy, socialtheory, and cultural studies, and correspond-ingly how questions of place, space, andlandscape have been taken up in other human-ities and social sciences, particularly throughphilosophy, feminism, post-structuralism, andpost-colonialism.

Part Two, ‘Capital Cities’, turns moredirectly to the connections between spaceand representation, politics and poetics, andthe role of the city in shaping contemporarytheory and politics. He reads ‘capital cities’through the variously modernist and post-modernist urban mappings of David Harvey,Walter Benjamin, Allan Pred, and Ed Soja,drawing from each the ways in which theyforeground ‘capital cities’ as generative sites ofcultural capital, but also as privileged sites ofcultural innovation and theory production(Gregory, 1994: 213). This is a cosmopoli-tanism with which he is increasingly uneasyand one that leads him closer towards Said. AsGregory writes on the first page of the book,this in itself is an act of migration, a re-centering, and a displacement, first throughhis own move from Cambridge to Vancouver(where he more directly recognized the chal-lenges and possibilities of post-colonial,multi-cultural and gendered politics and the-ory) and second through the ways in whichhe seeks to open conversations among a broadrange of contextual and anti-essentialist socialand cultural theories. While commentatorshave expressed surprise that post-colonial,multi-cultural and feminist movements had

DEREK GREGORYÿÿ173


not apparently reached Cambridge (Barnett,1995; Harvey, 1995; Swyngedouw, 1995),Gregory (1995) prefers to describe hismove to Canada in 1989 as an importantsymbolic and rhetorical border-crossing froma Cambridge and Britain of discipline andmetropolitanism to a Vancouver and Canadain which intense struggles over social andpolitical identity were being fought in andbeyond the university in the postcolony (seeSparke, 2006).

Part Three, ‘Between Two Continents’ inves-tigates further the ‘uncomfortable’ spacebetween historical materialism and postmod-ernism. Gregory reads Harvey’s The Conditionof Postmodernity and those who, in respondingto its arguments, have struggled with the waysin which it addresses the tension between the-ory’s desire for unity and a fear of ‘insufficiency,absence, fragmentation.’ He draws on WalterBenjamin’s critical interventions in Marxismand cultural theory to problematize theseuneasy commitments to either grand theory orpostmodernism. In particular, he draws onSusan Buck-Morss’s (1989) rendering of WalterBenjamin’s massive Arcades Project (Dialectics ofSeeing) to re-frame his own spatial materialism.In Benjamin’s thought, the movement of his-tory is constructed out of fragments of the past,a view of the city as a kind of shattered urban-ism comprising myriad fragments of localizedand everyday practices. The result is a criticalsocial theory lodged against what Deutsche(1995: 172) astutely called ‘the certainties of thesingular spatial consciousness which erases thetraces of its erasures in a foundational vision ofsocial totality.’

Conversations

At times, Geographical Imaginations is not aneasy text and responses to it have varied. Forsome it is erudite but esoteric, richly theoret-ical but lacking in concrete or practicalengagement (Barnett, 1995; Swyngedouw,

1995; Harvey, 1995). For many readersGeographical Imaginations appeared as a breathof fresh air, an exciting and full-blownencounter with social and cultural theories ofall kinds, animated by geographical imaginar-ies that few at that time had been able toconjure (Deutsche, 1995). Explaining the dif-ferences in these responses is not easy andprobably has more to do with the tenor ofother debates in the mid-1990s than with thesuccess or failings of Geographical Imaginations.At the time, Geography had experienced adecade of theoretical shifts from politicaleconomy (Marxism and post-Marxism), toidentity politics (feminism, race studies, andgender and sexuality studies) and a post-structural cultural studies committed to apolitics of contingency, context, and theconcrete.

One of the central goals of the book is todemonstrate the importance and value ofclose, careful reading. Indeed, at one level,Geographical Imaginations might be inter-preted as a response to the readings ofLefebvre that had entered Geography in the1980s and an attempt to model alternativereadings that weren’t (or couldn’t ever be)closed around a single interpretation. Thisquestion of closure has become even moreimportant in his subsequent books, first in hisimpatience with ‘Geography with a CapitalG’ and second with what in The ColonialPresent he refers to as the sense that the worldexists in order to provide examples of ourtheorizations of it. This was a call for an ethicof reading in which authors and their writ-ings are not read in terms of a logic of friendsand enemies, of battle and victory, but one inwhich authors and texts are read for theinsights they provide and the work they do,placed in conversation with each other.Gregory is driven by a fear of closure, by thatfalse certainty that leads others to ‘know’ thatanswers are clear and pathways are known.Above all, he strives to keep the questionsopen, to highlight the political commitments

174ÿÿJOHN PICKLES


of all claims to absolute grounds, fearing thatotherwise we close off possible politics-to-come. The violence of the fixed category andthe ‘disciplined’ subject/mind is his target,and this allows him to roam widely amongspatial analysts, ethnographers, Marxists, andothers without fear or regret that distinct dis-ciplinary-markers were disrupted or newones not set. For some readers the result hasbeen a deep anxiety about the project of post-structuralism and deconstruction. Swyngedouw(1995: 388) sees this commitment to philoso-phies of indeterminancy, uncertainty (or betterperhaps over-determination) as a laudable goal,but one that leaves political decision andGregory himself ‘up for grabs.’ What,Swyngedouw asks, are the practical politicalcommitments that follow from such rich anderudite textual and theoretical analyses?

Gregory begins Geographical Imaginationswith a suggestion that the book can be read asa reactivation of three dialogues, first withpolitical economy, second with social theory,and third with cultural studies. What does this‘working-through’ these dialogues signify andhow are we to read the path to cultural studies(Gregory, 1994: 6)? Clive Barnett has suggestedthat this is fundamentally a book focused on‘theory’ and ‘space,’ and in many ways this is thecase. But Gregory reads his own efforts differ-ently as an attempt to articulate more directlya critical, non-metropolitan, spatialized theoryof culture and a geographically nuanced theoryof history.

This is the broader and more interestingconversation with which Gregory asks us toengage. Far from ignoring the political lessonsof the Marxist writer Antonio Gramsci, as Katz(1995: 167) suggests, Geographical Imaginationscan be read as taking up Gramsci’s claims aboutconjunctural analysis in ways that have enor-mous importance for how we understandgeography and the political possibilities ofcritical theory. In this Gramscian sense, a con-juncture describes ‘the complex historicallyspecific terrain of a crisis which affects – but in

uneven ways – a specific national-social forma-tion as a whole’ (Hall, 1988: 127). But this‘terrain’ of conjunctural analysis is not merelycontextual. The spatialized and territorializedunderstanding of context as located and placedmust also be understood in terms of the alwaysrelational nature of context and place. Thereader of Geographical Imaginations is thus askedto consider what critical human geographylooks like when it takes conjunctural, contex-tual, and relational analysis seriously and whathistorical materialism and cultural studies looklike when they embody thoroughly spatializedpractices.

Conclusion

It is, I think, central to understanding this mas-sive book and what others have at times seenas an unwieldy text whose politics is ‘up forgrabs,’ that we appreciate the ways in whichthe various finite ontologies and fracturedepistemologies Gregory weaves into his analy-sis enable a certain kind of cultural studies toemerge, one that connects ‘the history of thebody with the history of space.’ The modernistsensibility, the god-trick of grand theory, andthe utopian gesture are here displaced in favorof a geography ‘that recognizes the corporealityof vision and … requires a scrupulous attentionto the junctures and fissures between manydifferent histories: a multileveled dialoguebetween past and present conducted as a history (or an historical geography) of the present’ (Gregory, 1994: 416). And it is hereGregory’s reactivation of dialogues works itsway into cultural studies. As Larry Grossberghas written:

Cultural studies is a project not only toconstruct a political history of the present,but to do so in a particular way, a radicallycontextualist way, in order to avoid repro-ducing the very sorts of universalisms (andessentialisms) that all too often characterize



the dominant practices of knowledgeproduction and that have contributed(perhaps unintentionally) to making thevery relations of domination, inequalityand suffering that cultural studies desiresto change. Cultural studies seeks to embracecomplexity and contingency, and to avoidthe many faces and forms of reduction-ism. (Grossberg, 2006: 2)

Geographical Imaginations builds a geogra-pher’s bridge to this complex and contingentconjunctural history of the present and themultiple spatialities that shape it, and it doesso in order to clarify the always political pos-sibilities of such a spatialized history of thepresent, a cultural politics that becomes evenclearer in The Colonial Present and ViolentSpaces. While his critics ask Gregory to moredirectly address crucial issues of race or

gender, he is already more interested in howwe can understand what Stuart Hall (1995:53–54) – in suggesting that he himself has‘never worked on race and ethnicity as a kindof subcategory’ – called ‘the whole social for-mation which is racialized.’ That is, the pathto cultural studies is about the production ofcertain kinds of conjunctural truth. These arenever fixed, but they intervene in the waysin which social life is produced (Hall, 1997:157). This is ‘a description of a social forma-tion as fractured and conflictual, alongmultiple axes, planes and scales, constantly insearch of temporary balances or structuralstabilities through a variety of practices andprocesses of struggle and negotiation’(Grossberg, 2006: 4-5): a vital conversationindeed for geographers and non-geographersalike!

176ÿÿJOHN PICKLES

Secondary sources and referencesBarnett, C. (1995) ‘Why theory?’, Economic Geography 71: 427–435.Buck-Morss, S. (1989) The Dialectics of Seeing: Walter Benjamin and the Arcades

Project. Cambridge: The MIT Press.Deutsche, R. (1995) ‘Surprising Geography’, Annals of the Association of American

Geographers, 85 (1): 168–175.Gregory, D. (1979) Ideology, Science and Human Geography. London: Palgrave

Macmillan.Gregory, D. (1982) Regional Transformation and Industrial Revolution. London:

Macmillan.Gregory, D. (1994) Geographical Imaginations. Oxford: Blackwell.Gregory, D. (1995) ‘A Geographical Unconscious: Spaces for Dialogue and Difference’,

Annals of the Association of American Geographers 85 (1): 175–186.Gregory, D. (2004) The Colonial Present: Afghanistan, Palestine, Iraq. Oxford: Blackwell. Gregory, D. and Pred, A. (eds) (2006) Violent Geographies. London: Routledge.Gregory, D. and Urry, J. (1985) Social Relations and Spatial Structures. London:

Palgrave Macmillan.Gregory, D., Martin, R., and Smith, G. (eds) (1994) Human Geography: Society, Space,

and Social Science. London: Macmillan.Grossberg, L. (2006) ‘Does cultural studies have futures? Should it? (Or what’s the mat-

ter with New York?)’ Cultural Studies 20 (1): 1–32.Habermas, J. (1988) Theory and Practice. Boston: Beacon Press.Hall, S. (1988) The Hard Road to Renewal: Thatcherism and the Crisis of the Left.

London: Verso.Hall, S. (1995) ‘Negotiating Caribbean identities’, New Left Review 208: 3–14.


Hall, S. (interviewed by David Scott) (1997) ‘Politics, contingency, strategy’, Small Axe 1:141–159.

Harvey, D. (1989) The Condition of Postmodernity: An Enquiry into the Origins of CulturalChange. Oxford: Blackwell.

Harvey, D. (1995) ‘Geographical Knowledge in the Eye of Power: Reflections on DerekGregory’s Geographical Imaginations’, Annals of the Association of AmericanGeographers 85 (1): 160–164.

Jones, J.P. (1995) ‘Derek Gregory’s Geographical Imaginations: Dialogic Invitation’,Annals of the Association of American Geographers 85 (1): 159–160.

Katz, C. (1995) ‘Major/Minor: Theory, nature, and politics’, Annals of the Association ofAmerican Geographers 85 (1): 164–168.

Lefebvre, H. (1972/1991) The Production of Space. Oxford: Blackwell.Soja, E. (1989) Postmodern Geographies. New York: Verso.Sparke, M. (2006) In the Space of Theory. Minneapolis: Minnesota University Press.Swyngedouw, E. (1995) ‘Geographical Imaginations: Book review’, Transactions of the

Institute of British Geographers NS 20: 387–400.



GEOGRAPHIES OF EXCLUSION (1995): DAVID SIBLEY

Phil Hubbard

In order to understand the problem of exclusionin modern society, we need a cultural reading ofspace, what we might term an ‘anthropology ofspace’, which emphasizes the rituals of spatialorganization. We need to see the sacred whichis embodied in spatial boundaries. (Sibley,1995: 72)

Introduction

The cover of Geographies of Exclusion(subtitled Society and Difference in the West) isa striking black and white photo entitled‘Woolloomooloo Girl’ by the renownedfashion photographer, Henry Talbot. Yet it isclear this is not from a fashion shoot. Taken inan inner city suburb of Sydney in the 1950s,it is of young girl, her hair matted and lank,her clothes simple and plain, clambering overa backyard fence (her backyard?). Whethershe is escaping or breaking in is uncertain,but either way we suspect her action is atransgression; she is daring to defy adultauthority and crossing the boundary thatseparates her world from a world that isdeemed off-limits. The physical boundary –the fence – is at one and the same time asocial boundary, between the ensconcedworld of childhood and a public space that isdefined as adult. Given the photo is of aninner city area, we might also infer that thegirl is crossing a class boundary, moving fromher working-class territory into the social

space occupied by the (middle-class) photog-rapher. Our reactions to this picture may bevery different, therefore, depending on ourown positionality. Maybe we regard the girlas a threat – a ‘feral’ street urchin who showslittle respect for authority? Or perhaps weadmire her for her seeming lack of fear, hersense of adventure, her spirit?

Initially, it is hard to see why this imagemight have been deemed particularlyappropriate for what is an incredibly wide-ranging exploration of the construction ofthe socio-spatial boundaries that divideus along lines of gender, colour, class,sexuality, age and disability. However, questions of childhood development andsocialization are central to Sibley’s theorizedexploration of what drives us to constructboundaries between those who we feelaffinity with and those Others that weregard as different or discrepant. Indeed,psychoanalytical ideas about how a childpositions itself in relation to other humans –as well as non-human aspects of the ‘objectworld’ (Sibley, 1995: 5) – are at the heart ofSibley’s analysis of how difference is createdand maintained spatially. Drawing on thework of those psychologists and psychoana-lysts who have reworked and extended‘objects relations theory’ to consider howwe project feelings of disgust onto particularpeople and objects while we introject others(which become part of our sense of Self),Sibley’s book encourages us all to reflect on

21


180ÿÿPHIL HUBBARD

the processes by which we become (orbecame) fully formed individuals, and tothink about the exclusions and repressionsthat are inherent to those processes.

However, the image on the cover is also sug-gestive of another theme within Geographies ofExclusion; namely, the power of the media toperpetuate stereotypes in which purity andpollution are mapped onto specific groupsand identities. It is perhaps significant that atthe time that Sibley was writing Geographiesof Exclusion, there was something of a ‘moralpanic’ in the UK about the ways in whichteenagers’ behaviour often challenged andundermined adult hegemony, at the sametime that it endangered younger children.Indeed, another highly charged image in thebook is of a pair of teenage boys on a child’sclimbing frame, glaring at a younger boy. AsSibley relates, these teenagers appear ‘out ofplace’ in a space designed for children, yetneither do they appear old enough to accessthe spaces and sites of adulthood. In the con-text of a children’s playground, they are atransgressive and polluting presence: froman adult or parent’s perspective, they mightbe regarded as ‘folk devils’ who threatenyounger, innocent ‘angels’. Sibley relatessome of the consequences of this, noting thatadolescents were increasingly finding theirpresence in public spaces regarded as prob-lematic by developers, planners and towncentre managers who suggested they threat-ened the integrity of ‘family’ consumerspaces (see Vanderbeck and Johnson, 2000).For Sibley, this demonstrates that attempts toorder a continuous social category (age) intosharp and neat categories (child/adult) alwayscreates ambiguities and indistinction – andthat this has wholly negative implications forthose who are represented as existing in theseliminal zones.

Consequently, the image of the girl ina Sydney back-alley captures some of the keyempirical and theoretical concerns that informSibley’s book, an inspired intervention in

geographical debates about socio-spatialsegregation. What is perhaps particularlyremarkable about the book, however, is thatwithin just 190 brief pages, it manages toconstruct an impassioned argument for geo-graphers to adopt more reflexive, engagedand inclusive modes of enquiry, as well asoffering a highly original analysis of the waythat questions of subjectivity and power con-struct boundaries at different spatial scales. Bydrawing parallels between the exclusion ofparticular knowledges and the exclusion ofparticular subjects, Sibley was able to producea book that speaks to a range of debates,marking it out as a text that was of signifi-cance well beyond the sub-discipline of socialand cultural geography.

Key themes: diversity, differenceand defilement

Never the most prolific of geographers,David Sibley’s previous publications hadnonetheless demonstrated his commitmentto a radical and engaged geography. Outsidersin Urban Society (Sibley, 1981) was a largelyempirical exploration of the lives and spacesof travelling gypsy communities in the UK,based on years of close contact with suchgroups. A key theme in this work was theway that sedentary society (in the form oflocal authority councillors and planners)sought to limit this mobile population’s resi-dence to ‘official’ sites, typically located inmarginal sites on the edge of cities, awayfrom those wealthy areas of residence wheresuburban dwellers might complain about thepresence of these ‘dirty’ and ‘dangerous’Others. From the perspective of the seden-tarist, removing these polluting Others thus‘purified’ urban space; conversely, the associ-ation of gypsy-travellers with devalued andoften derelict sites served to cement theimagined association between their mobilelifestyles and decay and dirt.


DAVID SIBLEYÿÿ181

Hardly a conventional geographic text,Outsiders in Urban Society drew on ongoingMarxist debates about deprivation and classstruggle, but sought to extend and amplifythese arguments by exploring the differentsenses of order inherent in both ‘mainstream’and ‘outsider’ society. However, given itsfocus on gypsy-travellers in the UK, thewider theoretical implications of Outsiders inUrban Society remained rather muted, and itwas not until a 1988 ‘review’ piece in thejournal Society and Space that Sibley offered amore explicit account of spatial purification.Herein, Sibley described exclusion andconstraint as the outcome of an interplaybetween processes of individuation (onthe one hand) and ideologies promotingproperty-ownership and capital accumulation(on the other). Underpinned by ideasderived from structuration theory, Sibley thusaffirmed the importance of transcendingstructure/agency models in human geogra-phy (a widespread concern at the time)through an approach attuned to the reciproc-ity of individual, society and the environment.Here, he drew on the work of sociologistAnthony Giddens, as well as geographersAllan Pred and Derek Gregory. What was par-ticularly distinctive about Sibley’s account,however, was his deployment of the ideas ofanthropologist Mary Douglas and education-alist Basil Bernstein. The former had writtenextensively of purification rituals in tribalsocieties, highlighting the persistence sinceancient times of attempts to impose a sym-bolic order in which the boundaries betweencleanliness and dirt were maintained throughboth routine and ceremonial practices: Thelatter had written of the different forms ofcontrol evident in the educational sphere,noting the differences between closed andopen curricula, where the latter exhibit littleconcern about boundaries between disci-plines or forms of knowledge. From this,Sibley took the idea that we can talk ofstrongly- and weakly-classified spaces, with

the contemporary urban West principallycharacterized by strong spatial boundaries,homogeneous spaces and a distinct lack ofsocial mixity. Purification rituals thus entailattempts to maintain this socio-spatial order,removing those who ‘pollute’ otherwise purespaces.

By Geographies of Exclusion, Sibley (1995:76) admitted that structuration theory wasnow regarded as somewhat passé, thoughcontinued to insist that it offered a useful wayof thinking about the distribution of power.However, he acknowledged a significant cri-tique of structuration theory, namely thatit split society into structure (context) andagency (individual intentionality) withoutexploring the idea that each individual con-stitutes society differently. As such, Sibleyshifted to a theoretical perspective in whichstructure/agency issues are reframed as ques-tions of the relation between Self/Society.Here, Sibley found the insights of particularpackages of psychoanalytical thought useful,not least the distinctive object relations per-spective developed by Melanie Klein. Notingthat, from the moment of birth, a childexperiences ‘discomforts of being’ (cold, light,noise, etc.), Klein surmises that the childbecomes reliant on a mother-figure. However,as the child develops, and recognizes its sepa-ration from the maternal (in what Lacantermed the ‘mirror stage’), Klein argues thatit seeks to redress its sense of loss by bringingsome people and things into the Self whilerejecting and distancing itself from ‘badobjects’. As Sibley notes, the fact that mostobjects (and people) are not polarized alonga continuum of good–bad means that sub-jects exhibit a great variety of reactions todifference, with some possessing a greaterboundary consciousness than others. Hence,while some people are prone to embrace dif-ference, others reject it (and most exhibit amix of the two tendencies). Sibley (1995: 7) hence begins from the positionthat anxieties associated with maternal separa-


182ÿÿPHIL HUBBARD

tion set in motion a series of processes bywhich individuals seek to constructa boundary between an ‘inner (pure) Self andthe outer defiled Self ’. However, at thispoint he acknowledges one of the major crit-icisms levelled at psychoanalytical theory inthe social sciences – i.e. its tendency toessentialize and generalize about processesof difference-making – by arguing foraccounts more alert to the cultural specificityof boundary-making. Here, he notes that it is‘socialized and acculturated’ adults who teachchildren about the dangers of dirt and pollu-tion, and it is their obsessions and anxietiesthat transmit to the child. Accordingly, henotes that many of the key anxieties aboutdirt, soil, faeces and bodily residues thatassume so much significance in many instancesof boundary-making are not innate, but aredistinctly modern and Western (or at leastwhite Northern European) concerns. Hissubsequent discussion of Julia Kristeva’snotion of the abject (defined as that whichwe seek to distance ourselves from, butremain haunted by) is hence illustrated withexamples that are specific to particular geo-graphic place and time (even if some enjoystrong historical resonance).

Sibley’s deployment of object relationstheory to explain social-spatial issues was byno means the first invocation of psychoana-lytic theory by a geographer. However, it wascertainly one of the most sustained attemptsto explore how ‘inner’ torments and anxietiesmanifest themselves in attempts to order the‘outside world’. Moreover, it was also one ofthe most distinctive, combining Sibley’s read-ing of psychoanalytic theory (Klein, Kristeva,Winnicott, and Erikson in particular) witha range of diverse scholarships drawn fromanthropology, history, educational studies,feminism and sociology. In fact, few of themajor reference points in Geographies ofExclusion are from within geography’s estab-lished canon. This makes it a potentiallychallenging read, particularly for those not

acquainted with psychoanalytical theories(which were, at least at the time, far from thecentre of the geographical curriculum). YetSibley’s writing style – never verbose andoften refreshingly straightforward – providesconsiderable clarity, and the book itselfunfolds through a series of linked chapterswhich outline how feelings about differencebecome mapped onto groups regarded asdeviant and dirty, and in turn, explain howthis produces exclusionary urges whose man-ifestations are clear at a variety of spatialscales.

Throughout the book, Sibley throws insuggestive examples of the ‘imperfect’ peoplewho, through history, have been depicted astroublesome Others who need to be located‘elsewhere’. By highlighting instances ofxenophobia and discrimination againstgroups such as prostitutes, gypsies, those withimpairments and sexual and racial minorities,he is able to demonstrate that there is aremarkable recalcitrance in the languagesand attitudes expressed towards these Othergroups in the urban West. Nonetheless, heinsists that the boundaries of society arealways shifting, and is at pains to illustratehow some groups have drifted into (or con-versely, out of) categorizations of purity. Theexamples that litter the text are consequentlymany and various, and Sibley draws on arange of popular texts and representations,including films (Taxi Driver, Simba), TV adverts(for Persil washing powder, Volkswagen cars)and even the images on a packet of biscuits.These illustrations allow Sibley to vividlysupport his underlying argument whilstdemonstrating the very banal and taken-for-granted nature of many of the representationswhich figure our reactions to difference.

The first hundred pages or so of Geographiesof Exclusion hence develop an astute andrichly illustrated geographical interpretationof social exclusion. It is at this point thatSibley flips his argument to turn his gazeonto the discipline of geography. Here,


DAVID SIBLEYÿÿ183

his chief contention is that geography (likeother disciplines) is based on a hierarchy ofknowledge which privileges particular(white, male, middle class) views of the world,but disables and marginalizes other view-points. This was not an uncommon argumentat the time, given geographers’ growingenthusiasm for post-structural theories whichare open to multiplicity and difference. YetSibley chooses to underline his critique ofgeography as exclusive by highlighting theway that certain bodies of knowledge havebecome mere footnotes in geographical his-tories because their authors peddled a‘localized’ knowledge which was viewed aspolluting the scientifically elegant andordered accounts that were regarded as cen-tral in establishing geography as a spatialscience. Sibley thus alights on the work ofW.E.B. Dubois, whose perspectives on thesocial geographies of the black inner citywere dismissed because of his own racial sta-tus, and the proto-feminist geographiesproduced by Florence Kelly and Jane Addamsat the Hull House in the University ofChicago at the turn of the nineteenth cen-tury. Focusing on these rather obscure figuresallows Sibley to suggest that geography isneither attuned to the range of exclusionsthat fracture the social world nor the exclu-sions that permeate the discipline – implyingthat geography itself needs to transcend dis-ciplinary boundaries in order to producemore inclusive and useful accounts of socialdifference.

A critical assessment

While Outsiders in Urban Society had offered afocused case study of social marginalizationthrough spatial process, Geographies of Exclusionoffered a more wide-ranging and explicitlytheoretical overview. Whether it actuallyoffers a cohesive geographical theory of exclu-sion is, however, a moot point – one which

Jonathan Smith (1996: 630) raises in a criticalreview in which he suggests most readerswill ask ‘what any of this has to do withgeography?’ For instance, while Sibley dwellson the importance of stereotyping, and usesmultiple media representations to illustratethe pervasive nature of stereotypes ofOtherness, it is unclear whether he attributesparticular importance to the media as amaker of territorial divisions or whether heis using these examples simply to illustratethe nature of more widespread social anxi-eties. Further, while Chapters One to Sixunfold in a laudable logical progression, mov-ing from a consideration of anxieties aroundsubjectivity through to the ways in whichthese anxieties are projected onto specificpeople and places, it remains unclear if Sibleyis offering a holistic interpretative frameworkthat might be used by others in their explo-rations of Otherness, or whether he is simplydetermined to make a general argumentabout the connections between Self, societyand space. Indeed, in the Introduction, Sibley(1995: ix) modestly claims his intent was ‘notto provide a comprehensive account ofexclusionary process’, merely to illustratesome of the ‘more opaque’ instances ofexclusion that persist in the West.

Further, Sibley’s use of Kleinian psychoan-alytic theory, while original, may be deemedinappropriate and ill-fitted to considering thewide range of exclusions extant in the urbanWest (and beyond). Reviewing the book,Cresswell (1997) alleged that Sibley locatedthe construction of difference in a ‘natural’transcultural and transhistorical moment,namely, the separation from the maternal. ForCresswell, this led Sibley to assume there is a‘natural’ desire to construct boundaries, withthe subsequent form of those boundariesbeing the product of social and culturalforces. Cresswell is clearly troubled by this,and is prompted to ask how the individualanxieties Sibley talks of might be ‘scaled up’to the level of the social and cultural. The


184ÿÿPHIL HUBBARD

idea that group behaviour may be no moreor less than the culmination of individualbehaviours is indeed problematic (and it isnotable that Cresswell’s own analyses ofexclusion and transgression bequeath themedia with much more importance in theprocess of constructing social identities – see,for example, Cresswell, 1996).

Yet even if some commentators regardedSibley’s attempt to integrate psychoanalysisinto human geography as not wholly suc-cessful, the subsequent citation and uptakeof Sibley’s ideas suggest that Geographies ofExclusion was an important and influentialintervention in geographical debates. Onefactor here was the sheer timeliness of itspublication. Geographies of Exclusion appearedat a time when talk of social exclusion was tothe fore of political and academic agenda,with 15 years of Conservative rule in the UKseen as exacerbating the gap between thehaves and the have-nots. Given cities wereseen to be fracturing along multiple axes ofdifference (race, class, age, sexuality, etc.), theclass-based heuristics of Marxism appearedincreasingly unable to grasp the complexitiesof the processes that were condemning someto life on ‘no-go’ estates while others enjoyedunprecedented affluence. As such, the ideathat there was a growing group not justmarginalized but actively excluded from par-ticipation in modern (consumer) society wasencouraging many geographers to explore thecultural, social and political basis of inequality,and not just its economic ‘underpinnings’.

In this sense, Sibley’s thoroughly encultur-ated interpretation of socio-spatial exclusionappeared at a time when a putative ‘culturalturn’ was encouraging geographers to explorethe role of representation and language in themaking of social categories (see Hubbardet al., 2002). Such approaches effectively re-energized social geography’s long-standinginterest in spatial segregation by consideringthe cultural politics of marginalization, withnotions of ‘social sorting’ and class differenti-ation being supplemented by a new-found

interest in questions of socio-cultural exclu-sion, marginalization and its resistance. Relatedto this was the increasing attention devotedto all number of Others in human geographyin the 1990s, with several studies of thoseliving with bodily impairments (e.g. Wilton,1998), the homeless (Takahashi, 1997), sexualminorities (Hubbard, 1999) and ethnicminority groups (Holloway, 2004) drawingexplicitly on the theoretical ideas outlined bySibley in Geographies of Exclusion. Interestingly,Sibley makes repeated reference to represen-tations of humans as animalistic in Geographiesof Exclusion, presaging a major interest in ahuman geography of animals (as opposed tozoogeography or biogeography); in later col-laborative work he was to focus explicitly onthe place of the animal in civilized society(Griffiths et al., 2000), making a significantcontribution to the understanding of animalsas Others (Wolch, 2002).

Geographies of Exclusion appears highly pre-scient in other senses, not least its deployment ofpsychoanalytical theory. At the time, humangeography’s engagement with psychological the-ory was largely limited to ‘scientific’ cognitive-behavioural models of the mind whichafforded little importance to the realm of theunconscious, with all that implies about therepression of particular hopes, desires and fears.In his acknowledgements, Sibley credits thegeographer Chris Philo for encouraging himto persist with his more ‘bizarre’ ideas, notingthat these were nonetheless becoming quiteconventional in geography. Published aroundthe same time as Steve Pile’s (1996) The Bodyand the City, and Pile and Thrift’s (1995)Mapping the Subject, Geographies of Exclusion didindeed appear at a moment when humangeography was ready to engage with psycho-analytical theories in a serious and sustainedmanner, with the long-held view that psycho-analytic theory offered little insight into theproduction of space dissipating in the wake ofcareful re-interpretation. Sibley’s engagementwith object relations theory nonetheless dis-tinguishes his approach from some of the


DAVID SIBLEYÿÿ185

deployments of Freud and (especially) Lacanwhich have been adopted by geographers infields such as cinematic geography, feministgeographies and the examination of ‘emotionalgeographies’ (Aitken and Craine, 2002; Bondiand Davidson, 2004). Significantly, perhaps,Sibley says little about the precarious achieve-ment of gender and sexual identities in andthrough space – an area in which psychoana-lytical geographies have proved particularlyinsightful (Philo and Parr, 2003). Indeed, whencompared with many other geographical takesinspired by psychoanalysis, Sibley presents adecidedly anaemic and disembodied accountwhere there is little attention given to the rela-tion of looks, bodies and surfaces that serve toposition us as subjects-in-the-world.

But perhaps the most cutting critique of thebook was offered by Smith (1996) when heidentified Sibley’s book as a somewhat polemic‘countercultural’ geography. Suggesting Sibleyopposed the ‘usual bogies’ of depersonalizedmiddle-class suburbanites whose fear, preju-dice and obsession quashes the counterculture,Smith (1996: 631) made the case for a geogra-phy which attempts the ‘hard work’ ofexploring the antimonies of mixing, toleranceand trust (on the one hand) and excellence,judgement and criticalness (on the other). Theaccusation here is that Sibley developed anargument designed to suggest that exclusionresults from prejudice and irrational intoler-ance; to the contrary, Smith suggestedexclusion can often be justified in quite ratio-nal ways and that Sibley had not explored thereal-life circumstances in which decisions toexclude some individuals or groups are made.Working through one of Sibley’s examples (i.e.the aforementioned exclusion of teenagersfrom children’s playgrounds) Smith arguedthat the exclusion of older children might beentirely appropriate: even if they do not intendto ‘play badly’, he suggests they should beexcluded because they have the potential todo great harm to younger children because oftheir relative strength. He continues by askingwhether other forms of difference (such as

criminalized drug-use) might be valid.Whether criminals be tolerated and includedor shunned and excluded is clearly a problem-atic debate, but Smith seems to side with theview that, again, exclusion is often justified.Initially, we might read Smith’s criticisms as areactionary response to Sibley’s call for mixityand diversity, yet there is a real underlyingconcern here that nuanced empirical work isneeded to explore the social norms and tiesthat bind specific communities (i.e. their geo-graphies of inclusion) before we attempt tomake generalizations about the causes andeffects of geographies of exclusion.

However, as a general demonstration of thepotential of psychoanalytical theory in humangeography, Geographies of Exclusion has yet tobe bettered. It also continues to have value asa clarion call for geographers to pursue moreparticipatory approaches – even if commenta-tors like Smith were uncertain if it set out asuitably objective framework for exploringsocio-spatial relations. In the book, Sibleynotes repeatedly that geographers in the 1990swere becoming increasingly attuned to notionsof difference, due in part to the de-centringimpulses of postmodern theorization (seeespecially Ed Soja’s Postmodern Geographies asdiscussed by Minca, Chapter 16 this volume).Yet he also noted the gaping chasm betweentheir willingness to recognize diversity and theextent to which they were prepared to movebeyond their centred, academic viewpoint:

The social sciences, and human geogra-phy in particular, might now be betterequipped to challenge xenophobia, racismand other exclusionary tendencies becauseof a greater intellectual awareness ofdifference … However … there is still adistance between authors (mainly male)and their subjects. (Sibley, 1995: 184)

Railing against what he saw as an emerg-ing trend for geographers to engage withtexts rather than people, Sibley impliedgeographers ‘talked the talk’, but seldom‘walked the walk’. Sibley accordingly


186ÿÿPHIL HUBBARD

implored geographers to go out into theworld, to engage with people and to producemore inclusive knowledges. This was hardlythe most original of arguments (as heacknowledged), and revisited many of thesentiments enunciated in the wake of the‘relevance’ debate of the 1970s. And, unfortu-nately, it could be argued that Geographies ofExclusion (unlike Outsiders in Urban Society)was not the best illustration of a text writtenwith the close cooperation and participation ofthe groups Sibley talks of (its sheer scope andrange militates against that). Yet read sympathet-ically, we can view Sibley’s book as a continuingprovocation to those geographers who keep dif-ference at an arm’s length lest their own neatlycompartmentalized and institutionalized world-view is disturbed. Facing our own fears, Sibleyconcludes, may sometimes be an appropriateway of getting to grips with difference.

Conclusion

Though at times idiosyncratic, Geoographies ofExclusion nonetheless bears repeated reading

because of its sheer lucidity. While it does notspell out a particular method for exploringgeographical exclusion, it clearly illustratesthe value of psychoanalytical packages ofthought in explications of social difference.Given this – and the subtext that thediscipline of geography needs to be anti-exclusionary – it is clear that the book con-tinues to have relevance within the disciplineas a whole, as well as having had a more sub-stantive and obvious imprint on the socialand cultural geography that followed it. Inmany ways, it is disappointing that Sibley didnot follow-up Geographies of Exclusion with amore focused study that demonstrated howthe conceptual building blocks of his theorymight be integrated (much of his subsequentwork in fact focuses on the problematicboundary between town and country, givenall that implies about distinctions betweenpurity and danger – see Sibley, 1997, 2003).Yet for all this, Geographies of Exclusionremains a highly suggestive text whose syn-thetic approach shows the value of a geographythat is prepared to transcend establisheddisciplinary boundaries.

Secondary sources and referencesAitken, S.C. and Craine, J. (2002) ‘The pornography of despair: lust, desire and the

music of Matt Johnson’, ACME, An International E-Journal for Critical Geographers1 (1): 91–116.

Bondi, L. and Davidson, J. (eds) (2004) Emotional Geographies. Chichester: Ashgate.Cresswell, T.M. (1996) In Place/Out of Place. Minneapolis: University of Minnesota

Press.Cresswell, T.M. (1997) ‘Geographies of Exclusion: Society and Difference in the West, by

David Sibley’ (book review), Annals of the Association of American Geographers 87(3): 566–567.

Griffiths, H., Poulter, I. and Sibley, D. (2000) ‘Feral cats in the city’, in C. Philo andC. Wilbert (eds) Animal Spaces, Beastly Places: New Geographies of Human–AnimalRelations. London: Routledge.

Holloway, S. (2004) ‘Rural roots, rural routes: discourses of rural self and travelling otherin debates about the future of Appleby New Fair 1945–1969’, Journal of Rural Studies20: 143–156.

Hubbard, P. (1999) Sex and the City: Geographies of Prostitution in the Urban West.Chichester: Ashgate.


DAVID SIBLEYÿÿ187

Hubbard, P., Kitchin, R., Bartely, B. and Fuller, D. (2002) Thinking Geographically.London: Continuum.

Philo, C. and Parr, H. (2003) ‘Introducing psychoanalytic geographies’, Social andCultural Geography 4 (3): 283–293.

Pile, S. (1996) The Body and the City. London: Routledge.Pile, S. and Thrift, N. (eds) (1995) Mapping the Subject. London: Routledge.Sibley, D. (1981) Outsiders in an Urban Society. Oxford: Basil Blackwell.Sibley, D. (1988) ‘Survey 13: Purification of space’, Environment and Planning D –

Society and Space 6 (4): 409–421.Sibley, D. (1995) Geographies of Exclusion: Society and Difference in the West. London:

Routledge.Sibley, D. (1997) ‘Endangering the sacred: nomads, youth cultures and the English coun-

tryside’, in P. Cloke and J. Little (eds) Contested Countryside Cultures. London,Routledge, pp. 218–231.

Sibley, D. (2001) ‘The binary city’, Urban Studies 38: 239–250.Sibley, D. (2003) ‘Psychogeographies of rural space and practices of exclusion’, in

P. Cloke (ed.) Country Visions. Harlow: Pearson.Smith, J. (1996) ‘Geographies of Exclusion: Society and Difference in the West, by David

Sibley’ (book review), Geographical Review 86 (4): 629–631.Takahashi, L. (1997) ‘The socio-spatial stigmatisation of homelessness and HIV/AIDS:

towards an explanation of the NIMBY syndrome’, Social Science and Medicine 45:903–914.

Vanderbeck, R.M. and Johnson, J.H. (2000) ‘That’s the only place where you can hangout: Urban young people and the space of the mall’, Urban Geography 21 (1): 5–25.

Wilton, R. (1998) ‘The constitution of difference: space and psyche in landscapes ofexclusion’, Geoforum 29: 173–185.

Wolch, J. (2002) ‘Anima urbis’, Progress in Human Geography 26 (6): 721–742.


CRITICAL GEOPOLITICS (1996): GEARÓID Ó TUATHAIL

Jo Sharp

Geography is about power. Although oftenassumed to be innocent, the geography of theworld is not a product of nature but a productof histories of struggle between competingauthorities over the power to organize, occupy,and administer space. (Ó Tuathail, 1996a: 1)

Introduction

In Critical Geopolitics Gearóid Ó Tuathail(1996a) presents a thorough challenge toconventional geopolitics, which he formu-lates as the theories and practices of statecraft.This monograph has been hugely influentialin political geography, pushing forward acritical vision of the subdiscipline influencedby some of the forms of poststructuralismthat have driven Geography’s ‘cultural turn’.The effect of the ‘cultural turn’ on politicalgeography has generated a move towardnon-traditional political geographical knowl-edges and a concern with the everyday as avalid space of political analysis rather thanfocusing solely on the formal arena of statepolitics and international relations. CriticalGeopolitics has facilitated a challenging ofaccepted boundaries within the discipline ofpolitical geography ‘a geopolitical perspectiveon the field of geopolitics’ (Ashley, 1987: 407)as one commentator put it – to examinethose relationships that were previously takenfor granted.

Critical Geopolitics first examines the tradi-tion of geopolitics through the work of its key

proponents at the turn of the twentiethcentury (e.g. Mackinder, Ratzel and Haushofer).In the central section Ó Tuathail moves tothe arguments of critical commentators suchas Bowman, Wittfogel, Lacoste, Ashley andDalby, before moving on to the final sectionwhere he exemplifies his own vision ofcritical geopolitics. The book finishes witha discussion of the changing nature ofgeopolitics, and especially the influence oftechnology and media, arguing for the neces-sity for critical analysis of these new trends:‘The challenge for critical geopolitics today isto document and deconstruct the institu-tional, technical, and material forms of thesenew congealments of geo-power, to prob-lematise how global space is incessantlyreimagined and rewritten by centers ofpower and authority in the late twentiethcentury’ (Ó Tuathail, 1996a: 249).

From conventional to criticalgeopolitics

For Ó Tuathail, conventional geopolitics isunderstood as an approach to the practiceand analysis of statecraft and internationalrelations which considers spatial relations toplay a significant role in the constitution ofinternational politics. British geographer andstrategist Halford Mackinder popularized theterm geopolitics in his famous address to theRoyal Geographical Society, ‘The GeographicPivot of History’, (1904) which promoted

22


the study of geography as an ‘aid to statecraft’.Mackinder accordingly proposed that geopol-itics offered one such way in which geographerscould inform the practices of internationalrelations. Geopolitics studied the ways inwhich geographical factors shaped thecharacter of international politics. These geo-graphical factors included the layout ofcontinental masses, their physical size, and thedistribution of physical and human resourceswithin them. As a result of these factors, cer-tain spaces are seen as either easier or harderto control, distance is seen as influencing pol-itics (because proximity was regarded aspotentially leading to susceptibility to politi-cal influence), and certain physical featurespromote security or lead to vulnerability.

Mackinder’s best-known geopolitical argu-ment is presented in his ‘Heartland Thesis’which insisted upon the importance of theterritory in the centre of Asia (the AsianHeartland) in the history of great states.Mackinder believed that whichever state hadcontrol over this territory held a more or lessimpenetrable position and this would providea powerful position from which to dominatethe world. For Mackinder, unless checked bypower in the ‘outer rim’ of territory proxi-mate to the Heartland, the occupying powercould quite easily come to control firstEurope and then the world.

Ó Tuathail links the rise of geopoliticswith the end of European exploration. Heillustrates this with the case of Mackinder,who argued that at this point the world hadbecome ‘known, occupied and closed’ andthat it was ‘no longer possible to treat variousstruggles for space in isolation from oneanother, for all are part of a single worldwidesystem of closed space’ (Ó Tuathail, 1996a:27). Ó Tuathail thus claims that geopoliticswas central to modernist ways of viewinginternational space. The emergence of thisnew science of international politics,Ó Tuathail (1996a: 29) argued, produced ‘aCartesian perspectivalist organization of spacewith a detached viewing subject surveying a

worldwide stage whose intelligibility is, for thefirst time, becoming visible and transparent’.Thus:

Geopolitics emerged during the last fin desiecle as part of an imperial, Eurocentricplanetary consciousness. This was a mas-culinist, ex-cathedra vision of a dangerousworld viewed from the commandingheights of governmental and academicinstitutions. Geopolitics (and the wider dis-cipline) were thus elements in what MartinJay (1988) has called the ‘scopic regimes ofmodernity,’ rational, ordering and control-ling. (Heffernan, 2000: 348)

This is an important argument as it offersa critique of suggestions that space was sub-ordinated or invisible in intellectual thoughtin the twentieth century (an argument mostforcefully made by Soja, 1989; see Minca,Chapter 16 this volume). Ó Tuathail argues(1996a: 24) that the enduring influence ofgeopolitics shows that spatial thinking wasindeed influential in twentieth-century polit-ical thought, shaping international politicsfrom imperial conquest to the Cold War.Heffernan (2000: 349), however, suggests thatÓ Tuathail may have overmade this point,and Ó Tuathail himself has argued the natureof space in geopolitical reasoning is highlyreductionist. Conventional geopolitics hencereduces spaces and places to concepts or ide-ology. The complexity of global space issimplified to units which singularly displayevidence of certain political and cultural ele-ments assumed to characterize that place. Forinstance, geopolitics perhaps reached itsheight in terms of influence during the ‘ColdWar’ that followed World War II. In thatperiod, American geopoliticians, influencedby Mackinder’s Heartland Thesis, worriedabout the power of the Soviet Union. Theyaccordingly explained the possibilities ofSoviet expansion not as a complex politicalprocess of adaptation and conflict, but as aresult of proximity. The Domino Theoryassumed that Soviets, Communists and

190ÿÿJO SHARP


socialists everywhere ‘were, and are, unquali-fiedly evil, that they were fiendishly clever,and that any small victory by them wouldautomatically lead to many more’ (Glassner,1993: 239). This led to preemptive militaryaction. US Admiral Arthur Redford, speakingin 1953, for example, argued that anAmerican nuclear strike on Vietnam wasessential in order to halt a Viet Minh victory,because that would set off a chain reaction ofcountries falling to the Communists, ‘like arow of falling dominoes’ (in Glassner, 1993:239). The argument went that the UnitedStates had to fight and win in Vietnam, for ifSouth Vietnam went communist then auto-matically, like falling dominoes, Cambodia,Laos, Thailand, Burma, South East Asia, andultimately, other parts of the world would dolikewise. This process would not stop until itreached the last standing domino, the US.

Having outlined conventional geopolitics,the remainder of the book turns to ÓTuathail’s conceptualization of critical geopol-itics. Ó Tuathail insists that geopolitics ‘doesnot simply “happen” but is instead practicedby agents at discrete sites of knowledge pro-duction, from where it is disseminated andenforced’ (Agnew, 2000: 98). Here, Ó Tuathailapplies the theories of French philosophersDerrida and Foucault to the ‘problematic ofgeopolitics, the politics of the production ofglobal political space by dominant intellectu-als, institutions, and practitioners of statecraftin practices that constitute “global politics”’(Ó Tuathail, 1996a: 185). He exposes thepower inherent in any enunciation of spatial-ity through his transformation of the noun‘geography’ to a hyphenated verb geo-graphy(literally, from the Greek, earth-writing (geo-graphien)). The hyphen is an important literarydevice as it ‘ruptures the givenness of geopol-itics and opens up the seal of the bonding ofthe ‘geo’ and ‘politics’ to critical thought’ (ÓTuathail, 1996a: 67). In his rewriting ofgeopolitics as critical geopolitics, space ispower: no description of political events in theworld is merely a reflection of some prior or

exterior condition but instead is a ‘willto power’ – a use of particular geographicalrepresentations to construct possible interpreta-tions and limit meaning. There is always apolitics to describing the world because there isalways a choice of whose descriptions are used.

While traditional geopolitics regards geog-raphy as a set of facts and relationships ‘outthere’ in the world awaiting description, crit-ical geopolitics understands geographicalorders to be created by key individuals andinstitutions and then imposed upon theworld as frameworks of understanding. Criticalgeopolitical approaches seek to examine howit is that international politics are imaginedspatially or geographically and in so doing touncover the politics involved in writing thegeography of global space. For geopoliticians,there is great power available to those whosemaps and explanations of world politics areaccepted as accurate because of the influencethese have on the way the world is under-stood. In turn, these understandings haveprofound effects on subsequent politicalpractice.

Ó Tuathail’s deconstructionist approach isnot about defining geopolitics but ratherunderstanding how this particular form ofknowledge has been used to particular ends.As he puts it:

How has the term ‘geopolitics’ beencharged with particular meanings andstrategic uses within differing networks ofpower/knowledge? How has it been putto use in differing times and places? Theterm ‘geopolitics’ poses a question to useevery time it is knowingly evoked andused. (Ó Tuathail, 1996a: 66)

Critical impact and reception

It is difficult to disentangle Critical Geopoliticsfrom a number of wider shifts in the natureof political geography in the early 1990s,but there is no doubt that, as an exemplar of

GEARÓID Ó TUATHAILÿÿ191


192ÿÿJO SHARP

critical geopolitical enquiry, it has had a verysignificant influence on the subdiscipline.Critical geopolitics has included studies ofthe language of statecraft (for example,Dalby’s, 1990b, work on the characterizationof the Soviet Union by Reagan’s advisors,The Committee on Present Danger) and for-eign policy (e.g. Sidaway’s, 1998, work onrepresentation of the Gulf ) to demonstratehow international politics are imagined spa-tially and to uncover the politics involved inthis process. However, it is not only the geo-graphical imaginations of the powerful thatare studied in critical geopolitics: less formalarenas of politics have also been examined.The ‘cultural turn’ has fully impacted and thedefinition of ‘the political’ bringing questionsof representation and the politics of identityvery much into the heart of study. This hasled to the development of a ‘popular geopol-itics’ which has looked at the way in whichglobal and national political identities havebeen formed through representations ofthreats in the media, such as Cold War mag-azines (Sharp, 1996), or films (Sharp, 1998;Power and Crampton, 2005). Despite (orbecause) of this influence, there have been avariety of critical engagements with criticalgeopolitics and specifically Ó Tuathail’s work.

Probably the most significant and wide-spread critique focuses on the meaning of‘critical’ in Ó Tuathail’s formulation. AsAgnew (2000: 98) has argued, ‘Criticalgeopolitics is not just an alternative theory tothat of geopolitics because it refuses the driveto certainty and objectivity that such a the-ory requires’. As Ó Tuathail himself notes,critical geopolitics is not an absence ofgeopolitics but a variation of it. Ó Tuathailtoo creates geographies when he writesabout geopolitics, and his book, like the textsof those he seeks to critique, contains silencesand a will to power in his own cartography ofinternational relations. For example, his appearsto be a view from nowhere; there is no senseof an embodied critic ‘Ó Tuathail’ – only

a relentless unveiling and revealing of allgeopolitical texts that he encounters (Sharp,2000a). Ó Tuathail claims his own work isnot a totalizing survey of ‘geopolitics’ but ‘aset of contextual explorations of the prob-lematic imperfectly marked by the term“geopolitics’’’ (Ó Tuathail 1996a: 18). However,his critique does seem to share a number ofelements with its object. Critical Geopoliticsitself is characterized by a Cartesian perspec-tivalism: ‘The object of site here is not somuch the landscape or the globe but thearray of pre-existing geopolitical texts viewedand read by the detached theoretical eye/I’(Smith, 2000: 368). Thus, critics are suggest-ing that, just as the geopoliticians that comeunder Ó Tuathail’s scrutiny present them-selves as all-knowing observers of the world,and predictors of its political future, so too doeshe stand apart, detached and all-seeing of theirworks. Ó Tuathail’s account of the geopoliticalwritings of previous and current intellectualsof statecraft is as much a visualization ofthe ‘world-as-exhibition’ (Mitchell, 1988) asMackinder’s famous presentation to the RoyalGeographic Society was, providing his ownheroic narration, not of world domination orprediction, but of the cunning theorist unmask-ing the powerful statesmen and their advisors.

For Agnew (2000: 98), this denial of theneed for ‘ontological commitments’ is a keylimitation of Ó Tuathail’s Critical Geopolitics.Agnew (2000: 98) wonders what kind of pol-itics might be possible from this perspectiveand suggests that ‘deconstructing the termsand strategies of geopolitics tells us how butnot why geopolitical knowledge is con-tructed where it is and by whom’. Paasi(2000: 284) has similarly suggested that thereis a need to move from text and metaphorthat had dominated radical political geogra-phy to more immediate forms of analysis andintervention. Others, notably feminist critics,have agreed suggesting that the textual focusof Ó Tuathail’s critique constrains the natureof politics, instead insisting on a form of


geopolitics that is attentive to the embodi-ment of the people through whose lives it isarticulated (Dowler and Sharp, 2001). Thrift(2000) argues for the need for criticalgeopolitics to think of bodies as sites of per-formance in their own right rather than assimple surfaces for discursive inscription.Discourses do not simply write themselvesdirectly onto the surface of bodies as if thesebodies offered blank surfaces of equal topog-raphy. Instead these concepts and ways ofbeing are taken up and used by people whomake meaning of them in the different globalcontexts in which they operate. Various the-orists have tried to work Ó Tuathail’s ideasthrough notions of embodiment to producea more materially grounded critical geopoli-tics (e.g., MacDonald, 2006), including ÓTuathail himself (Ó Tuathail, 1996b), and interms of the bodies of those caught up in therepresentational practices described in CriticalGeopolitics (e.g. Hyndman, 2003). Sparke’s(1998) ethnographic account of the produc-tion of Timothy McVeigh, the 1995 ‘OklahomaBomber’, as a political subject, points to theimportant and complex interrelationshipsbetween the production of geopoliticalimages and their actual impact on people’sdaily lives.

Feminist critics have noted a more generalfailure for Critical Geopolitics to acknowledgefeminist heritage, so that the widening of ‘thepolitical’ is seen to come from poststructuralistsensibilities rather than feminist argumentsabout the personal being the political. Indeed,elsewhere it has been suggested that ÓTuathail’s account reproduces geopolitics as anessentially masculinist practice. Ó Tuathail’sintellectual history of geopolitical practitionersand critical geopoliticians is certainly a historyof Big Men (in order): Mackinder, Ratzel,Mahan, Kjellen, Hausehoffer, Spykman,Wittfogel, Bowman, Lacoste, Ashley, and Dalby.A few women are allowed into the footnotes,but the central narrative is one of the exploitsand thoughts of men (see Sharp, 2000a: 363).

The story of geopolitics presented in CriticalGeopolitics is resolutely male, not just when dis-cussing the masculinist history of geopoliticalstrategies of elite practitioners, but also the inter-ventions of ‘critical geopoliticians’. Sparke(2000) is surprised that Ó Tuathail’s theories ofvision does not draw upon the pioneering workof Haraway (1988) about the ‘God trick’ of theall-seeing eye/I of modern science. Moreover,while women were not part of the hallowedintellectual societies which discussed the prac-tised geopolitics, their bodies most certainlywere caught up in the resulting political geogra-phies: flows of migrants, representations of otherplaces in geographical imaginations, construc-tions of ‘women and children’ to be protectedby the state, new social movements and ecolog-ical resistances (see Enloe, 1989, 1993).As Enloe(1989: 1) suggests at the beginning of herattempt to make feminist sense of internationalpolitics – Bananas, Beaches and Bases – ‘if weemploy only the conventional, ungenderedcompass to chart international politics, we arelikely to end up mapping a landscape peopledonly by men, most elite men’.

It is also worth noting that Ó Tuathailmaintains a clear distinction between highand low culture, apparently agreeing with the‘intellectuals of statecraft’ that he quotes thatthe High Politics of Statecraft and interna-tional relations are beyond the ken ofordinary people, and beyond the effects ofthe politics of everyday life (Sharp, 2000a).Popular culture is given short shrift, andwhen it is mentioned, is described as ‘largelypropagandistic’ and offering only ‘crude andconspiratorial reasoning’ (Ó Tuathail, 1996a:114 and 121). A variety of work whichexamines the significance of ‘popular geopol-itics’ has emerged as an alternative (e.g. Sharp,2000b). The point here is not just that criti-cal geopolitics needs to study a wider range oftexts per se, but is an intellectual critique of ÓTuathail’s focus on elitist texts.

In many ways, therefore, Critical Geopoliticsassumes the intellectuals of statecraft are



somehow beyond or outside of hegemonicnational culture, or that their pronouncementsare somehow unaffected by the circulation ofideas and beliefs therein. But, in order to maketheir arguments believable to their audiences(in many cases, ‘ordinary’ people, and not justother intellectuals of statecraft), they must referto concepts and values that have consonance forthe population at large, if their support is to beassured. As Ó Tuathail argued in his 1992 paperwith John Agnew, ‘geopolitics is not a discreteand relatively contained activity confined onlyto a small group of “wise men” who speak inthe language of classical geopolitics’ (Ó Tuathailand Agnew, 1992: 194). Simply to describe aforeign policy is to engage in geopolitics and sonormalize particular world views. If this is thecase, then surely the media – both high and lowculture alike – is intimately bound up withgeo-graphing the world, as are a range of activ-ities normally described as occurring outwiththe sphere of international politics. By forcingapart the ‘geo-politics’ of the everyday and the‘geopolitics’ of statecraft, Ó Tuathail too readilyaccepts a ‘neo-realist’ view of state actors as theprimary agents in world politics, rather thanaccepting the fluid and contested nature ofhegemonic values and norms (Sharp, 2000a).

A critical assessment

Appearing at a crucial moment in geogra-phy’s ‘cultural turn’ (see Hubbard et al., 2002),Critical Geopolitics prised open a space for dis-cussions in political geography addressing therole of representation and discourse. In thatsense, it remains an important read, mappingout an agenda for subsequent studies. Yet itneeds to be read with an eye for its blindspotsand occlusions. For example, although ÓTuathail suggests in his introduction that‘[i]dealized maps from the centre clash withthe lived geopolitics of the margin’ (ÓTuathail, 1996a: 2), Critical Geopolitics does

not contain much examination of resistanceto dominant geo-graphings. In the majorityof the book, resistance and alternative geo-graphings lie in the textual interventions ofcritical geopolitics. Marxist critics have beenquick to point out alternative forms of action(Smith, 2000), while others have argued foran ‘anti-geopolitics’ to open up space forforms of resistance beyond the textual.Routledge defines ‘anti-geopolitics’ as:

An ethical, political and cultural forcewithin civil society – i.e. those institutionsand organizations that are neither part ofthe processes of material production in theeconomy, nor part of state-funded or state-controlled organizations (e.g. religiousinstitutions, the media, voluntary organiza-tions, educational institutions and tradeunions) – that challenges the notion thatthe interest of the state’s political class areidentical to the community’s interests. Anti-geopolitics represents an assertion ofpermanent independence from the statewhoever is in power. (Routledge, 1998: 245)

Ó Tuathail’s division of life into political-effectual and non-political-ineffectual spheresalso silences a whole range of people andgroups from the operations of internationalpolitics. This division of international anddomestic politics reinforces the public-domesticspheres division characteristic of patriarchalcapitalist society, where women are effec-tively contained within the mundane spaceof home and kept from public space, which isthe space of politics, change and new possi-bilities. Ó Tuathail insists that it is importantto maintain a certain specificity to the term‘geopolitics’ to keep it as a ‘careful genealog-ical approach to the problematic of thewriting of global space by intellectuals ofstatecraft’ (Ó Tuathail, 1996a: 143). However,in line with his own arguments that criticalgeopolitics should be seen as a project whichexamines to what end particular geopoliticaldiscourses are used, rather than trying to

194ÿÿJO SHARP


define the term, the effects of this move needto be examined. This argument has beentaken up by Matt Sparke (2005) in his exam-ination of the geographing of nationalidentity and international relations. Sparkehighlights the necessity for untiring critique:

the work of describing the graphing ofthe geo is never done … it is a reminderof a responsibility to examine othergraphings, other geographies, that evenavowedly antiessentialist work may havewritten out of the geo. (Sparke, 2005: xxxi)

Critical Geopolitics is a challenging bookwhich covers difficult concepts and draws onthe complex writings of theorists such asFoucault and Derrida. It brought together anew way of thinking about space and powerand has had a significant influence on politi-cal geography and the discipline ofgeography itself. That there have been somany critiques of the book, and so manyattempts to develop critical geopolitics innew directions, is testament to the ambitionof his arguments rather than any limitationsof his vision.


Secondary sources and referencesAgnew, J. (2000) ‘Global political geography beyond geopolitics: review essay’,

International Studies Association 2 (1): 91 –99.Ashley, R. (1987) ‘The geopolitics of geopolitical space: towards a critical social theory of

international politics’, Alternatives XIV: 403–434.Dalby, S. (1990a) ‘American security discourse: the persistence of geopolitics’, Political

Geography Quarterly 9 (2): 171–188.Dalby, S. (1990b) Creating the Second Cold War: The Discourse of Politics. New York: Guilford.Dalby, S. (1991) ‘Critical geopolitics: Difference, discourse and dissent’, Environment and

Planning D: Society and Space 9 (3): 261–283.Dalby, S. (1994) ‘Gender and critical geopolitics: reading security discourse in the new

world order’, Environment and Planning D Society and Space 12, 525–542.Dowler, L. and Sharp, J. (2001) ‘A feminist geopolitics?’ Space and Polity 5 (3): 165–176.Enloe, C. (1989) Bananas, Beaches and Bases: Making Feminist Sense of International

Relations. Berkeley: University of California Press.Enloe, C. (1993) The Morning After: Sexual Politics at the End of the Cold War. Berkeley:

University of California Press.Glassner, I. (1993) Political Geography. Chichester: Wiley.Heffernan, M. (2000), ‘Balancing visions: comments on Gearóid Ó Tuathail’s critical geopol-

itics’, Political Geography 19: 347–352.Haraway, D. (1988) ‘Situated knowledges: the science question in feminism and the privi-

lege of partial perspective’, Feminist Studies 14: 575–599.Hubbard, P., Kitchin, R., Bartley, B. and Fuller, D. (2002) Thinking Geographically. London:

Continuum.Hyndman, J. (2003) ‘Beyond either/or: a feminist analysis of September 11th’, ACME

2 (1).MacDonald, F. (2006) ‘Geopolitics and the “vision thing”: regarding Britain and America’s

first nuclear missile’, Transactions, Institute of British Geographers 31 (1): 53–71.Mitchell, T. (1988) Colonising Egypt. Berkeley: University of California Press.Ó Tuathail, G. (1996a) Critical Geopolitics. Minneapolis: Minnesota University Press.


Ó Tuathail, G. (1996b) ‘An anti-geopolitical eye: Maggie O’Kane in Bosnia, 1992–1993’,Gender, Place and Culture 3 (2): 171–185.

Ó Tuathail, G. and Agnew, J. (1992) ‘Geopolitics and discourse: Practical geopoliticalreasoning in American foreign policy’, Political Geography 11 (2), 190–204.

Ó Tuathail, G. and Dalby, S. (1998) ‘Introduction: rethinking geopolitics: towards a criticalgeopolitics’, in G. Ó Tuathail and S. Dalby (eds) Rethinking Geopolitics. London:Routledge, pp. 16–38.

Paasi, A. (2000) ‘Review of Rethinking Geopolitics’, Environment and Planning D: Societyand Space 18 (2): 282–284.

Power, M. and Crampton, A. (2005) ‘Reel Geopolitics: Cinemato-graphing Political Space’,Geopolitics 10: 193–203.

Routledge, P. (1998) ‘Anti-geopolitics: introduction’, in G. Ó Tuathail, S. Dalby andP. Routledge (eds) The Geopolitics Reader. London: Routledge, pp. 245–255.

Said, E. (1978) Orientalism. New York: Vintage.Sharp, J. (1996) ‘Hegemony, popular culture and geopolitics: the Reader’s Digest and the

construction of danger’, Political Geography 15 (6/7): 557–570.Sharp, J. (1998) ‘Reel geographies of the new world order: staging post-Cold War geopol-

itics in American movies’, in S. Dalby and G. Toal (eds) Rethinking Geopolitics. London:Routledge, pp.152–169.

Sharp, J. (2000a) ‘Remasculinising geo(-)politics? comments on Gearóid Ó Tuathail’sCritical Geopolitics ’, in Political Geography 19 (3): 361–364.

Sharp, J. (2000b) Condensing the Cold War: Reader’s Digest and American Identity,1922–994. Minneapolis: University of Minnesota Press.

Sidaway, J. (1998) ‘What’s in a gulf?: from the “arc of crisis” to the Gulf War’, in S. Dalbyand G. Toal (eds) Rethinking Geopolitics. London: Routledge, pp. 224–239.

Smith, N. (2000) ‘Is a critical geopolitics possible? Foucault, class and the vision thing’,Political Geography 19: 365–371.

Soja, E. (1989) Postmodern Geographies: The Reassertion of Space in Critical SocialTheory. New York: Verso.

Sparke, M. (1998) ‘Outside inside patriotism: the Oklahoma bombing and the displacementof heartland geopolitics’, in Ó G. Tuathail and S. Dalby (eds) Rethinking Geopolitics.London: Routledge, pp.198–239.

Sparke, M. (2000) ‘Graphing the geo in geo-political: Critical Geopolitics and the re-visioningof responsibility’, Political Geography 19 (3): 373–380.

Sparke, M. (2005) In the Space of Theory. Minneapolis: Minnesota University Press.Thrift, N. (2000) ‘It’s the little things’, in K. Dodds and D. Atkinson (eds) Geopolitical

Traditions: a Century of Geopolitical Thought. London: Routledge, pp. 380–387.

196ÿÿJO SHARP


LOGICS OF DISLOCATION (1996):TREVOR J. BARNES

Philip Kelly

[K]nowledge is acquired in many differentways; there is no single epistemology thatreveals the ‘truth’. To see how knowledge isacquired, we must examine the local context;that is, we must see how knowledge isobtained, used, and verified in a particular placeand time. This view is relativist; there are noabsolutes because one’s knowledge is always‘local’ in origin. (Barnes, 1996: 95)

Introduction

If there is one core idea that the reader ofLogics of Dislocation is intended to take away, itis that there is no core idea that should guidethe sub-discipline of economic geography(or indeed any project of knowledge produc-tion). Barnes’ book is an argument againstfoundations, essences, fundamentals and uni-versals. Instead, he invites students andresearchers to think carefully about wherethey, as individuals, are coming from, howthey are constructing their arguments, andthe specificities of the contexts they aredescribing and explaining. Logics is, there-fore, a call for modesty in our theoreticalprojects and an attempt to encourage anongoing conversation between a plurality ofperspectives.

Reading Logics of Dislocation

Logics is a challenge. Barnes writes exceptionallywell, but his prose demands close attention as heengages in tight philosophical arguments only

occasionally leavened by empirical examples.The range of literatures and disciplinestouched upon by Logics is also quite dizzying:aside from geography, Barnes also engagessubstantially with economics, sociology,regional science, philosophy, and mathematics.To his credit, however, Barnes is never cleverlyobscure just for the sake of it, and always lucidenough to carry the careful reader along withhis argument.

While the reader might be tempted to diveright in with the first chapter, it would be amistake to skip the Preface. Barnes uses thosethree short pages to explain some importantcaveats and contexts for the book that fol-lows. In particular, he lays bare his owncomplicated relationship with the approachesthat he will review and critique – a relation-ship we will examine more closely below.

In the first chapter, Barnes sets out someimportant concepts that underpin much ofthe book. In particular, he explains the criticaldistinction between Enlightenment and post-Enlightenment thought. Enlightenmentthinking, emerging as modern science, rosefrom the mysticism of religious belief inEurope, carried with it four key characteristics:

1 The belief that progress could be achievedthrough the application of science andrationality – namely the idea that theworld could be improved without divineintervention.

2 The belief that human beings areautonomous, self-conscious decision-making units with fixed characteristicsand identities.

23


3 The belief that an order existed in the(natural and social) world, discoverablethrough rational inquiry free from poli-tics or subjective biases.

4 The belief that universal truths are possi-ble, unchanging in time and space,creating a direct correspondence betweenideas and reality.

In short, the Enlightenment project wasfounded upon a powerful set of ‘-isms’ – uni-versalism, foundationalism, essentialism, andrationalism. Together these provided theontological basis for modern science, alongwith all of the power, creativity and destruc-tion that accompanied it. Barnes’ focus, ofcourse, is on one small corner of ‘modernscience’ – economic geography. The zenith ofscientific approaches to economic geographywas in the 1950s and 1960s, when a group ofyoung and mathematically inclined scholarsredefined the field, emphasizing the utility offormal models of economic location andbehaviour, the expression of such models inthe formal language of mathematics, and theuse of quantitative data to test and refinethem. Variously labelled location theory, spa-tial science, or the quantitative revolution,this was a set of approaches that influencednot just economic geography, but the disci-pline as a whole (see Billinge et al., 1984; alsoJohnston, Chapter 4 this volume).

The 1970s and 1980s saw a strong reactionagainst these approaches (e.g. Gregory, 1978)and, in economic geography, new debatesemerged based upon Marxism, critical real-ism, locality studies, and flexible production(see Phelps, Chapter 10 this volume). Whileeach of these promised an alternative to thescientific pretensions of the previous genera-tion, Barnes argues that they all failed to liveup to the ideals of a post-Enlightenment phi-losophy – at best, they ‘tottered’ between anEnlightenment search for certainty and apost-Enlightenment rejection of such a pos-sibility. The chapter then turns to highlight

three approaches that do seem to offer thatpromise – post-developmentalism, feministapproaches to labour markets, and post-colonialism. In the rest of the book, however,Barnes chooses not to elaborate on thesethree sets of approaches. Instead, he devoteshimself to laying the philosophical ground-work that opens up economic geography totheir promise of a post-Enlightenment approachto knowledge production.

In the second part of the book, Barnes setsabout dissecting some of the key tenets ofEnlightenment thought. In Chapter Two, hedemonstrates that conceptions of ‘value’ inboth Marxist and neoclassical economics arerooted in ways of thinking that continuallyseek essences rather than trying to under-stand the distinctive way economic value iscreated in particular places. Barnes arguessuch processes need to be seen as contextualrather than universal, and are not reducible toa ‘final cause’. In Chapter Three, Barnesexamines the way human actors are repre-sented in economic theory – usually asrational economic actors (‘homo economicus’)rather than complex multifaceted individualscoloured by social, cultural and politicalprocesses. Barnes’ larger argument here is thathomo economicus is flawed not just because heis ‘unrealistic’, but because he represents amisguided attempt to find universal ratherthan contextual truths. In both chapters,Barnes carefully takes apart the philosophicalassumptions that have guided (often uncon-sciously) economic geography. He does so,however, not by asserting the superiority ofan alternative worldview, but by carefullyexamining the internal logical consistency ofexisting approaches. While his argumentsmay be abstract, the type of knowledge pro-duction Barnes advocates as an alternative isvery much grounded in everyday experience:contextualized, ethnographic and attentive tothe richness and variety of economic life.

The third part of the book then moves awayfrom examining what various orthodoxies have

198ÿÿPHILIP KELLY


to say about the world they study, and shiftstowards thinking about who is saying it andhow they are convincing others of their argu-ments. In Chapter Four, Barnes examines therise of scientific rationality in human geogra-phy and looks at the personalities and placesinvolved. He argues that proponents of spatialscience seldom applied the scientific methodin its purest form, and insists the rise of loca-tion theory in economic geography was basedupon the sociology of power relations withinthe academy, rather than the inherent superi-ority of a new set of ideas. In Chapter Five,Barnes examines the concepts and modelsused by economic geographers, paying par-ticular attention to the metaphors that helpto construct theory. Using the philosopherRichard Rorty as his guide, Barnes arguesthat metaphors can be useful tools for theo-retical debate, but they can constrain ourthinking in important ways. Thus, themetaphors widely used in economic geogra-phy, such as the gravity model – whichassumes that consumers, for example, will be‘drawn’ to the most efficient and economicalsource of a product – should not be seen asrevealing fundamental processes, but as aids toour thinking that prove useful from time totime and in particular contexts.

Having examined the models, metaphorsand men of economic geography (the pro-tagonists he discusses are indeed all men,although of course not all economic geogra-phers were; see Phelps, Chapter 10 thisvolume), Barnes turns in Chapter Six to itsmathematics – the language of quantification.Here he shows how mathematics, in the formof inferential statistics, is not a medium ofultimate objectivity, neutrality and truth, butis in fact a very partial way of representingthe world. Again, Barnes has a philosopher-guide, in the form of Jacques Derrida, whosemethod of linguistic deconstruction he appliesto question the ‘certainties’ that reside inmathematical descriptions of the world.Here, the ascendance of a particular mode of

analysis in economic geography is shown torest on decidedly uncertain foundations, andto be largely a reflection of the (local) institu-tional power wielded by its proponents.

In Part Four of Logics of Dislocation, Barnesturns to three individuals to find examples ofthe type of thinking that he advocates undera post-Enlightenment framework. They aresomewhat unusual choices – all white maleswhose work dates from the mid-twentiethcentury and none of them contemporaryeconomic geographers – but for various rea-sons, each holds an important place inBarnes’ affections. The first, Piero Sraffa, wasan Italian-born Cambridge economist. WhileSraffa offered a revised, but still formallymodelled, version of several key tenets ofneoclassical and Marxist economics in rela-tion to commodity production, he had thevision to leave his models open to the geo-graphical distinctiveness of particular timesand places. In resisting the urge to seek uni-versals and foundations, Sraffa was thus a veryunusual economist.

Barnes’ second exemplar is Harold Innis, apolitical economist who sought to explainthe historical and geographical developmentof the Canadian economy. Barnes sees Innis’‘staples thesis’, in which the development ofCanada was closely tied to the geographicalcharacteristics of primary resource extraction,as a case of contextual theory or ‘local mod-elling’. Innis also appeals to Barnes as apolitical economist who thought seriouslyabout knowledge, reflexivity, language, andcommunication. In his own empiricallygrounded work on the forestry sector ofBritish Columbia, Barnes has accordinglypursued Innis’ interest in the spatiality of aresource-based economy (see, for example,Barnes and Hayter, 1997).

The third exemplar, Fred Lukermann, wasone of Barnes’ teachers during his graduatestudies at Minnesota, and was something of asubversive at the height of human geogra-phy’s quantitative revolution in the late 1950s

TREVOR J. BARNESÿÿ199


and early 1960s. Lukermann was interested inindustrial location but rejected approachesthat focused exclusively on either individuallocational decisions (in this case by flour andcement millers) or generalized models ofrational location patterns. Instead, Lukermannfocused on the specific history and geogra-phy of these sectors in particular places. Morebroadly, Lukermann regarded integrative nar-rative description as a methodology that iscontextually sensitive, resistant to the searchfor root causes, and open to wider cultural,social and political contexts in explainingeconomic geographies.

Locating Logics of Dislocation

We carry around our geographies andhistories. (Barnes, 2006: 42)

A central argument of Logics, and the reasonit finishes with detailed cases studies of threespecific thinkers, is that theories are notabsolute truths but are always rooted in thepersonal and societal contexts of their authorsand adherents. It is appropriate, therefore, thatwe look a little closer at where Trevor Barneshimself is coming from. Obviously, ideas andintellectual positions are constructed in com-plicated ways that are only ever known to thethinker in question. Barnes in particular isnot an easy scholar to ‘place’ – he doesn’t slotneatly into any single paradigmatic pigeon-hole. Nevertheless, it is possible to discernin Logics the traces of a personal intellectualtrajectory.

Barnes’ early academic training was atUniversity College London in the mid-1970s. Quantitative spatial science and locationtheory were the mainstream orthodoxies ofthe day. With a joint degree in geography andeconomics, Barnes was exposed more thanmost to the scientific certainties and mathe-matical language that were then the currenciesof both disciplines. For Barnes and the

generation that preceded him, the develop-ment of statistical and other technical skills,the impulse to formally model spatial phe-nomena, and the reduction of human actionto a set of rational decision-making processeswas at the core of geographical study.Industrial location, urban systems, or patternsof international trade, were seen as productsof ordered and logical systems that could berevealed with appropriate methodologies.

The first significant point in Barnes’ intel-lectual trajectory, then, is this early training inthe rationalism of economics and an appreci-ation of geography as a spatial science. It is, ashe notes himself in the Preface to Logics, his‘temptation’ and his ‘habit’. Barnes (1996: vi)confesses to being ‘deeply smitten by thelogic and orderliness of spatial science’, but atthe same time feeling an urge to underminethese certainties. Indeed, readers of Logics willquickly notice Barnes’ mode of argumenta-tion in the book is very much a product offormal logical thinking. In that sense thebook is very aptly titled – although it seeks todislocate conventional thinking in variousways, its method for doing so is very muchrooted in processes of logical argumentation:Barnes turns upon and dismantles the scien-tific approach with its own tools. Barnes’fascination with ‘scientific’ geography is alsoevident in his writings since the publicationof Logics, which have provided ethnographicand philosophical insights into the social net-works and intellectual underpinnings ofgeography’s ‘quantitative revolution’ (e.g.Barnes, 2001, 2003). Thus, Barnes is notstraightforwardly a critic of Enlightenmentthought – his fascination with it representsboth an exorcism of a ghost in his own past,and a continued excitement with its technicalsophistication and mathematical precision.

To understand Barnes’ impulse to decon-struct the scientific economic geography inwhich he was trained, we need to recognizethat while his formal education was focused onquantitative spatial science, his extra-curricular

200ÿÿPHILIP KELLY


reading reflected the alternative, radical,streams of thought that were emerging in the1970s through the work of David Harveyand others (see Castree, Chapter 8 this vol-ume). Barnes has noted his debt to Harvey inparticular: ‘[m]y academic life from the timeI entered university in 1975 had pivotedaround Harvey’s works, first, Social Justice andthe City (Harvey, 1973) that I bought duringmy first term, and then his various essaysappearing in Antipode, which as a journal wasthought so seditious that it was kept underlock and key at our library. In fact, it wasSocial Justice that persuaded me to attendgraduate school’ (Barnes, 2004: 408)

At the University of Minnesota, Barnes(2002) recalls stuffing envelopes for theRadical Geography Newsletter, and his doc-toral research brought quantitative and radicaltraditions together in the form of analyticalMarxism. This represents a form of radicaleconomics inspired by the categories ofMarxist thought, but using sophisticatedmathematical tools to substantiate its argu-ments. Perhaps because of the paradigm-straddling intellectual acrobatics this requires,it has not been a widely adopted approach ineconomic geography, and Barnes’ book withhis doctoral supervisor, The Capitalist SpaceEconomy, represents one the few examples(Sheppard and Barnes, 1990).

The broader point to note here is thatBarnes’ training spanned a period of signifi-cant change in human geography. Whatstarted out as a radical fringe, gradually, overthe course of the 1980s in particular, becamea central focus of economic geography.Alongside Harvey, the work of RichardWalker, Michael Storper, Neil Smith, RichardPeet, Doreen Massey and others (see Phelps,Chapter 10 this volume; Phillips, Chapter 9this volume; Coe, Chapter 17 this volume),all made significant impressions. But onceagain, Barnes’ inclination was to cast a sub-versively sceptical eye upon this ascendant setof approaches. In the Marxist framework that

they apply, Barnes sees a set of logics andcertainties just as rigid as those of the spatialscientists they criticized. Thus a core argu-ment of Logics is that both Marxist andneoclassical economic approaches suffer fromthe same problems of essentialism, founda-tionalism and universalism.

Without entirely burning his bridges withspatial science or Marxism, then, Barnes hascarved out a role for himself as the restlessanalyst of knowledge production in eco-nomic geography. It is a role that sits mostcomfortably with a third set of philosophicalinfluences: post-structuralism. In simple terms,post-structuralism represents an approach toknowledge that rejects any notion of founda-tions or fundamentals. It argues that there areno absolutes, no ‘bedrock notions’, no ‘God’seye’ views of the world. Instead there are par-tial and contestable knowledges, necessarilysituated in the personal experiences of indi-vidual knowledge-producers, closely linkedto the language of theory construction, andthe power relations that surround it (seeDoel, 2004).

Logics of Dislocation, then, is Barnes’answer to the puzzle of how one writes apost-structural economic geography thatdefines the state of the art in the field whileat the same time denying the notion thatthere is a definable field or that anyone hasthe right to try to define it. It is also an exer-cise in theoretical ecumenicalism – Barnes’purpose is not to debunk ‘old’ approachesthat draw upon neoclassical economics orMarxism, but is instead to strip them of theirveneer of certainty and universality in orderthat they can engage in conversations witheach other. Indeed the ‘conversation’ is animportant metaphor for Barnes.

If Barnes’ scholarly training bridged theeras of dominance for spatial science and rad-ical political economy in geography, hissubsequent home at the University of BritishColumbia has been a hotbed for the impor-tation of post-structural ideas into human



geography. The Department he joined in1983 included several prominent critics ofboth spatial science and Marxism, includingDavid Ley and James Duncan; feministapproaches were later represented with thearrival of Geraldine Pratt in 1986 (who hadpreviously been a graduate student in thedepartment). A year later, Derek Gregory,whose trenchant critiques of spatial scienceand applications of social theory to geogra-phy had long influenced Barnes’ thinking,also arrived. Perhaps more than any othersingle department, UBC geographers havecollectively brought the post-structural writ-ings of Foucault, Derrida, Haraway, Bourdieu,Said and others into the mainstream ofhuman geography (see Gregory, 1994; Pickles,Chapter 20 this volume). Barnes contributedto, but was also doubtless influenced by, thismilieu – and out of it emerged Logics.

Logics of Dislocation in context

Having explored Barnes’ own flight pathtowards Logics, it is also important to note theintellectual state of economic geography at the time of its publication in 1996. Inretrospect, 1995–97 was a fine vintage ineconomic geography, with several importantand lasting contributions. Three contribu-tions marking the vanguard of economicgeography appeared almost simultaneouslywith Logics. Susan Hanson and GeraldinePratt’s (1995) influential study of gender andlabour markets in Worcester, Massachusetts,had been published the previous year and wascited by Barnes as an example of the‘post-prefixed’ economic geography he wasadvocating. Hanson and Pratt explored therole of space and gender in the constitutionof a local labour market without reducing thephenomenon to over-arching frameworksbased on class, rational economic decision-making, or gender. They also combined bothquantitative and qualitative research methods.

A similarly contextual approach was alsobeing advocated by Jamie Peck in Workplace(1996). Peck sought to construct an argu-ment for seeing labour markets as locallyconstituted through a variety of institutionaland regulatory forms. The labour market wasthus presented not as a homogenous process ofclass relations under capitalism, nor as theoutcome of neoclassical universals based onmarket mechanisms, but instead as the prod-uct of local contingencies. In returning tolocality and contingency, Peck’s argumentwas very much in the same spirit as Logics ofDislocation. A third significant contemporaryof Logics of Dislocation was the first edition ofJ.K. Gibson-Graham’s The End of Capitalism(1996). Gibson-Graham sought to rethinkthe ways in which class struggles intersectwith other social processes, and to displacethe foundational centrality of capitalism inunderstanding political-economic structures.In their attention to the power of languageand representation, and in their non-essentialist focus on multiple processes (classplaced alongside gender, sexuality, regional-ism, etc.), they too were on the samewavelength as Barnes.

Shortly after Logics of Dislocation was pub-lished, a collection of research essays byeconomic geographers titled Geographies ofEconomies appeared and staked out the expandedterritories being explored in the field (Lee andWills, 1997). While Barnes had provided thephilosophical arguments for a non-essentialistand pluralist economic geography, this col-lection showed how economic geographerswere operationalizing such an agenda.

The vanguard of economic geography was,therefore, quite closely aligned with Barnes.It is, however, also instructive to consider thestate of debate in human geography and thesocial sciences more broadly. The time atwhich Barnes was writing, in the early 1990s,was the apogee of controversies over post-modern approaches to knowledge. In thesame year as Logics of Dislocation was

202ÿÿPHILIP KELLY


published, for example, a heated exchangeoccurred in the pages of the Annals of theAssociation of American Geographers concerninga post-modern perspective on the meaning ofpoverty proposed by Lakshman Yapa (1996).Yapa argued that Indian poverty could beseen as socially constructed through dis-courses of development and scarcity. NandaShrestha (1997: 710) angrily replied that suchan argument, focusing on the power of lan-guage, amounted to ‘aimless intellectualpontification’. Looking back on the debatenow, it appears rather overblown, with over-statements and misunderstandings fuellingpassions on both sides. A decade later, the ideaof social constructivism is well-establishedand human geography has moved on fromthe days of ‘pomo’ and ‘anti-pomo’ rhetoric.The 1990s, however, were the height of the‘pomo’ wars. In some respects, Logics reflectsthis context with its earnest insistence on anti-essentialism and its post-modern rejection offoundational knowledge.

Initial reception and critiques

Logics of Dislocation represented the culminationof almost a decade of work by Barnes. Many ofthe arguments in the book had appeared in aseries of papers published since 1987 and insome cases these papers had themselves stimu-lated responses and debate (for example, seeBarnes 1993, 1994; Bassett, 1994, 1995). Thepublication of Logics, however, broughttogether Barnes’ perspectives on knowledgeproduction in economic geography and laidout the implications for future practice in thefield. Reviewers of the book were enthusiasticabout the clarity and force of its arguments, butseveral found themselves reluctant to go all theway with Barnes into the realm of relativistphilosophy (for the most thorough considera-tion of the book, see Bassett, 1996).

Some reviewers raised issues concerningthe ways in which the book’s arguments were

made, without questioning the argumentsthemselves. For example, it was noted thatLogics tends to be retrospective rather thanprospective. Barnes’ focus on spatial scienceas a target throughout much of the bookseemed misplaced given how little influencethat paradigm held by the 1990s. His choiceof three retired or deceased white malescholars as ‘exemplars’ also raised eyebrows,given that the first chapter of Logics hadpointed towards feminism, post-colonialismand post-structuralism as promising ways for-ward for Economic Geography. These were,however, criticisms that Barnes had antici-pated (see pages v–vii), noting that his choiceof targets reflected his own past and expertise.Furthermore, Barnes argued that while theEnlightenment thinking that he was dissect-ing was most explicitly evident in spatialscience approaches, the rationalism thatunderpinned them had lingered in manysubsequent frameworks. Some reviewers,however, also accused Barnes of providingslightly caricatured portraits of the theoristshe was representing. While Mirowski (1997)suggests Piero Sraffa’s work was not quite assensitively contextual as Barnes portrays,Bassett (1996) conversely argues that AndrewSayer’s critical realism is more subtle thanBarnes allows.

A second and more substantive set ofcriticisms focused directly on the socialconstructivist approaches to knowledge pro-moted by Barnes. This critique took fourdistinct directions. First, while Logics advo-cates a plurality of perspectives, it remainssomewhat silent on how we go about choos-ing between these different understandings ofthe world. As Bassett points out, Barnes does,on occasion, call for more ‘suitable’ or ‘appro-priate’ metaphors, and ‘a more satisfactoryeconomic geography’ (Bassett, 1996: 228), allof which implies that there is some a prioribasis for making such judgements. But if allknowledge is relative rather than absolute,then how do we determine what is more



suitable, appropriate and satisfactory? Thisalso connects with a second criticism, con-cerning the politics that are implied byBarnes’ arguments. Like much of the ‘post-modern’ philosophy that was being debatedat the time, critics suggested that Barnes’approach provided no ‘project’ – no funda-mental beliefs that could serve as a rallying cryfor action. Thus while Logics advocates femi-nist, post-colonial and post-developmentalperspectives as the most promising waysforward, and all are avowedly political pro-jects, the relationship between theory andpolitical action is not developed in the book.Barnes provides no roadmap for bringingtogether the ‘militant particularisms’ of localstruggles that would be based on localknowledge. It is important to remember thatthe Marxism that Barnes criticizes for itsessentialism provided a strong dose of politi-cization for geographers. A third critiqueconcerns Barnes’ rejection of grand narra-tives of knowledge and his insistence that weshould use whatever ‘works’ in a specific con-text. And yet, Barnes is quite damning ofspatial science and Marxism, leaving littleroom for ‘conversations’ with them and see-ing little value in applying their techniques. Itseemed to some, then, that Barnes was him-self constructing a grand and linear narrativeof paradigm replacement, rather than dia-logue. A fourth and related critique askswhether, in practice, elements of grand narra-tive and contextual theory might co-exist,rather than being incompatible as Barnessuggests. Bassett (1996: 456) points out thatAndrew Sayer, the critical realist targeted inLogics, has argued that ‘local knowledges maybe appropriate for objects which are indeedlocal, but grand theories and narratives areneeded for objects which are large or widelyreplicated’. This statement conflates scales ofprocess with levels of abstraction, but thepoint remains that we still need some of thevocabulary of grand theories when we con-struct local knowledges.

Logics’ legacies: A prelude to neweconomic geography

Even if Logics of Dislocation is just read nar-rowly as a critical intellectual history ofeconomic geography, it is still a fine achieve-ment. There are few, if any, studies of specificsubfields within geography that tackle theirsubject matter with such a combination ofphilosophical rigour and ethnographic insight.But Logics of Dislocation amounts to muchmore than a disciplinary history – it is also amanifesto for future directions in the field. Itis, however, a very delicate manifesto, whichstudiously avoids being programmatic or pre-scriptive. Instead, Barnes’ contribution was tolay the philosophical foundations for the‘new economic geography’ that was alreadyemerging. We can identify several specificways in which Barnes’ arguments provided aprelude to much of the economic geographythat has been conducted since the 1990s.

First, Logics of Dislocation explains to eco-nomic geographers that they need not seeklaws and principles, or read economic land-scapes as manifestations of universal processes,but should instead look for processes thatare specific in time and space. Barnes arguesthat explanations of economic geographiesthemselves have a geography – different expla-nations fit in different places. This recognitionof contingency is visible to varying degreesin a range of geographical research over thelast decade, from the tracing of commoditieschains and networks (Leslie and Reimer,1999), to examining the role of geographi-cally specific institutions (Peck, 1996), tothinking about embeddedness (Hess, 2004).

Secondly, Barnes’ arguments for non-essentialism endorsed a range of approachesthat rethought what is meant by ‘the eco-nomic’ and attempt to incorporate a widerrange of social processes into studies of eco-nomic life. Economic geographers have beenparticularly active in seeking to incorporatecultural practices and the multiple subjectivities

204ÿÿPHILIP KELLY


(gendered, racialized, regionalized, etc.)occupied by economic actors. Thinkingabout economic actors in non-essentialistways has thus opened up a wide range ofpossibilities, including the fields of labourgeography, ethnic economies and feministlabour market studies.

Third, the methodological corollary ofcontextualism and non-essentialism is oftenethnographic analysis, based on close obser-vation of the ways in which economicprocesses are actually practised. These processesmight include the tacit knowledge andnetworks that lead to regions of intensetechnological innovation, or the workplaceprocesses and performances that lead to cer-tain patterns of gendered segmentation in thelabour force (see Saxenian, 1996; McDowell,1997). To capture these processes economicgeographers have increasingly used qualita-tive methods in their research.

Fourth, Barnes’ advocacy for reflexivity inthe research process and in the constructionof explanations has also been widely taken onboard. This implies not just thinking aboutwhy we choose to ask certain questions asindividual researchers, but also to think aboutthe politics of our research and where it

stands in relation to social power relations. It isnow an expected component of methodolog-ical rigour to reflect upon the ‘positionality’of the research in the research process (e.g.,see Tickell et al., 2007).

Conclusion

It would be an overstatement to suggest thatLogics brought about all these new approachesin economic geography – most were alreadyapparent when the book was published. Whatcan reasonably be claimed, however, is that thebook laid a philosophical foundation for a newapproach to knowledge production in eco-nomic geography, and much of the research inthe field that has appeared since has been con-sistent with this vision. Indeed, through a seriesof edited collections published since Logics,Barnes has continued to chart (still ever sodelicately) the new directions in the field(Sheppard and Barnes, 2000; Barnes andGertler, 1999). For students seeking to under-standing some of the philosophy behind thecontemporary practice of economic geogra-phy, and where the field has come from, Logicsof Dislocation remains their best guide.


Secondary sources and referencesBarnes, T. (1993) ‘Whatever happened to the Philosophy of Science?’ Environment and

Planning A 25 (3): 301–304.Barnes, T. (1994) ‘5 ways to leave your critic – a sociological experiment in replying’,

Environment and Planning A 26 (11): 1653–1658. Barnes, T. (1996) Logics of Dislocation: Models, Metaphors and Meanings of Economic

Space. New York: Guilford Press.Barnes, T.J. (2001) ‘Lives lived, and lives told: biographies of geography’s quantitative

revolution’, Environment and Planning D – Society and Space 19: 409–429.Barnes, T.J. (2002) ‘Critical notes on economic geography from an aging radical. Or

radical notes on economic geography from a critical age’, ACME: An InternationalE-Journal for Critical Geographies 1: 8–14.

Barnes, T.J. (2003) ‘The place of locational analysis: a selective and interpretive history’,Progress in Human Geography 27: 69–95.

Barnes, T.J. (2004) ‘The background of our lives: David Harvey’s The Limits to Capital’,Antipode 36: 407–413.


Barnes, T.J. (2006) ‘Between deduction and dialectics: David Harvey on knowledge’, inN. Castree and D. Gregory (eds) David Harvey: A Critical Reader. Oxford: Blackwell,pp. 26–46.

Barnes, T.J. and Hayter, R. (eds) (1997) Troubles in the Rainforest: British Columbia’sForest Economy in Transition. Victoria: Western Geographical Press.

Barnes, T. and Gertler, M. (1999) The New Industrial Geography: Regions, Regulationsand Institutions. London: Routledge.

Bassett, K. (1994) ‘Whatever happened to the Philosophy of Science – some commentson Barnes’, Environment and Planning A 26 (3): 337–342.

Bassett, K. (1995) ‘On reflexivity – further comments on Barnes and the sociology of sci-ence’, Environment and Planning A 27 (10): 1527–1533.

Bassett, K. (1996) ‘Reconstruction or Dislocation: Barnes or Sayer on the PoliticalEconomy of Space?’ Geoforum 27 (4): 453–460.

Billinge, M., Gregory, D. and Martin, R. (1984) Recollections of a Revolution: Geography asSpatial Science. New York: St Martin’s Press.

Doel, M.A. (2004) ‘Poststructuralist geographies: the essential selection’, in P. Cloke,P. Crang and M. Goodwin (eds) Envisioning Human Geographies. London: Arnold,pp.146–171.

Gibson-Graham, J.K. (1996) The End of Capitalism (as we knew it): A Feminist Critique ofPolitical Economy. Cambridge, MA: Blackwell.

Gregory, D. (1978) Ideology, Science and Human Geography. New York: St Martin’s Press.Gregory, D. (1994) Geographical Imaginations. Oxford: Blackwell.Hanson, S. and Pratt, G. (1995) Gender, Work and Space. London and New York:

Routledge.Hess, M. (2004) ‘“Spatial” relationships? Towards a reconceptualization of embeddedness’,

Progress in Human Geography 28 (2): 165–186.Lee, R. and Wills, J. (eds) (1997) Geographies of Economies. New York: Arnold.Leslie, D. and Reimer, S. (1999) ‘Spatializing commodity chains’, Progress in Human

Geography 23 (3): 401–420. McDowell, L. (1997) Capital Culture: Gender at Work in the City. Oxford: Blackwell.Mirowski, P. (1997) ‘Logics of dislocation: Models, metaphors, and meanings of economic

space’, Canadian Geographer 41 (3): 335–336.Peck, J. (1996) Work-Place: The Social Regulation of Labor Markets. New York: Guilford Press.Saxenian, A. (1996) Regional Advantage: Culture and Competition in Silicon Valley and

Route 128. Cambridge, MA: Harvard University Press.Sheppard, E. and Barnes, T.J. (1990) The Capitalist Space Economy: Geographical

Analysis After Ricardo, Marx and Sraffa. London: Unwin Hyman.Sheppard, E. and Barnes, T. (2000) A Companion to Economic Geography. Oxford:

Blackwell.Shrestha, N. (1997) ‘On “What causes poverty? A postmodern view”: A postmodern view

or denial of historical integrity? The poverty of Yapa’s view of poverty’, Annals of theAssociation of American Geographers 87 (4): 709–716.

Tickell, A., Sheppard, E., Peck, J. and Barnes, T. (eds) (2007) Politics and Practice inEconomic Geography. London: Sage.

Yapa, L. (1996) ‘What causes poverty? A postmodern view’, Annals of the Association ofAmerican Geographers 86 (4): 707–728.

206ÿÿPHILIP KELLY


HYBRID GEOGRAPHIES (2002):SARAH WHATMORE

Sarah Dyer

… the hybrid invites new ways of travelling.(Whatmore, 2002: 6)

Introduction

Hybrid Geographies is a challenging book. Itstheoretical and empirical scope is wide rang-ing and impressive. It draws on and developsa number of different areas of literature tolocate them firmly within the discipline ofgeography. In this book, Whatmore challengesus to fundamentally rethink the ways inwhich we understand nature and the naturalworld. In exploring our conceptualizations ofnature we are required to think through thepolitics of nature, to traverse history and theglobe, and to question our assumptions aboutthe relationship between humans and thenatural world. This is also a book that chal-lenges many other assumptions. It occupiesmany contradictory, or seemingly contradic-tory, positions at once. It is poetic and yetfactual, imaginative but not imaginary. Thebook takes as its subject the ‘fleshiness’ of thematerial world and yet it is an intricately the-oretical text. It is, as Whatmore (2002: 6)herself says, philosophical but not philosophy.Moreover, it is neither a conventional mono-graph with a beginning, middle and end,nor a book of essays. It is, again in her ownwords, ‘an effort to germinate connectionsand openings that complicate’ (Whatmore,2002: 6). A book which has the express aim

of complication can seem a little unnerving –especially for the undergraduate reader.Consequently, it is a book that can be asconfounding as it is pleasurable.

In this chapter I aim to provide readers anentry into the text and an understanding ofwhy this text has become a key text in humangeography. I begin by giving an overview ofthe book, describing the contemporary rele-vance of the topics addressed in HybridGeographies and geographers’ interest in these.I then turn to the main arguments Whatmoremakes and the evidence she draws on to makethem. I highlight concerns raised by someabout the truth status of the claims made inthe book. In the next section I describe theacclaim with which the book was received atthe time of publishing. I summarize three keyquestions discussants of the book raised in aconference session about Hybrid Geographies.These provide a thought-provoking platformfrom which readers can perform their owncritical reading. I then end with a short con-sideration of the place Hybrid Geographies haswithin a more unified human-and-physicalgeography.

Overview

The book begins by invoking a set of debateswe are used to seeing on the front pages ofour newspapers and on the evening news.In any given week, the media covers many

24


stories which concern our changing relation-ship with the natural world. These storiesinclude concerns over farming and the foodwe eat, such as the ‘crisis’ in the UK in 2001over the spread of foot and mouth disease orthe ongoing debates about genetically modi-fied crops (GMOs). Watching the eveningnews we are confronted with stories aboutthe risks wild animals pose to humans. Thespread of bird flu from 2003 onwards,the contraction of the disease by people, andthe subsequent debate about the likelihoodof human-to-human infection, is typical. Wealso see the converse concerns, about thethreats humans pose to the natural world, bethey marked by extinction rates of endan-gered animals or rising sea levels. Similarly,stories abound of the new and challengingways in which science and technologyare enabling us to alter our own (human)nature, from extending fertility to enhanc-ing of our capabilities through drugs. Asthe popular press recognize, these storiesand the issues raised are of great concern topeople. We are challenged to think aboutwhat we can and can’t control, how weshould behave, and whose responsibilityit all is.

Whatmore’s premise is that geographybrings important tools to the understandingof these issues. In order to introduce herreaders to hybrid geographies, Whatmore(2002: 4–6) describes the main bodies ofwork with which she engages. These arevariously labelled ones of ‘life politics’ or‘bio-sociality’ as they involve a reconfiguringof socio-political life, the material world,and the relationships between the two.Geographers have been interested in theseissues and these literatures long before HybridGeographies. However, Whatmore’s project isa creative synthesis of existing literatures andto locate them firmly within geography.Elsewhere, Whatmore has argued that socialscience often ends up following the agendasand lead of the life sciences:

… not enough energy is being invested inthe much harder and less eye-catchingwork of contextualising the biotechnologiesthat are making the headlines in the varie-gated histories and geographies throughwhich life and knowledge has been co-fabricated. (Whatmore, 2004: 1362)

Geography has the ability to transcend the‘science wars’ that pit social against naturalsciences, she argues, and to enable social sci-entists to contribute fully to debates about‘life politics’.

In order to demonstrate the value of geog-raphy, Whatmore works through a number ofexamples in Hybrid Geographies. The book isdivided into three sections: bewilderingspaces; governing spaces; and living spaces.Each section has its own short introductionand a couple of chapters giving what can bethought of as case studies. Each case can beread as self-contained, as illustrating thetheme of the section it is in, or as furtheringthe arguments of the book as a whole. Shesets out her agenda and theoretical underpin-ning in the first and last chapter of the book.I will not summarize the case studies here, asI could not do justice to Whatmore’s rich andthorough descriptions. I will turn instead to awider consideration of the argument thebook makes.

Arguments and evidence

Hybrid Geographies is, at its core, a book aboutrelations; it is about the relationship betweenthe social and the natural; it is about the rela-tionships of long-dead leopards that foughtin the Roman amphitheatres and of ele-phants in British zoos; it is about therelationships that construct each of thesecreatures and, through juxtaposing the storiesof these animals in this book, Whatmoreposes questions about the relationships betweenthese creatures. It may seem odd to talk aboutanimals having relationships, but that is

208ÿÿSARAH DYER


Whatmore’s starting point. She sets out inthis book to challenge conventional ways ofthinking about nature and to challenge us tothink about the relationships between societyand nature. It is only by challenging suchconventional dualist thinking we have thechance of honestly representing and acting inthe world.

The nature-society dichotomy is a long-standing and powerful one. For a long time(in the Western world) these two realms havebeen thought to be ontologically separate,mutually exclusive. It is associated with otherdichotomies under the labels Modernity orthe Enlightenment. Something is social ornatural, active or passive, agent or acted upon.In this schema nature is separate fromhumanity and humans have the monopolyon knowledge, agency and morality. Humansare subjects in the world. They are able toknow and to manipulate nature, which is pas-sive and object. Being separate does not meanlocated in different places, although that toois seen to be true (rural versus urban), butseparate ontologically, in the most fundamen-tal way. Whatmore is by no means alone inher attempt to describe an alternative modeof thought. However, the fundamentalpremise of Hybrid Geographies is that seeing theworld as being divided up into either the nat-ural or the social (and all the other associateddichotomies) is neither useful nor honest. Thefundamental aim of the book is to enlarge thenotion of social and to de-couple the subject–object dichotomy (Whatmore, 2002: 4): towrite of what Whatmore has called elsewhere,a ‘more-than-human geography’ (Whatmore,2003: 139).

It is important to understand that this isnot just a book about the relations of natureand society but a book that asks us to under-stand the world as constituted by relations.Rather than being ‘pure’ or ‘discrete’ andbelonging to the natural rather than thesocial entities, things such as elephants or soybeans are constituted through relations. They

are hybrids. Our world is a world constitutedthrough hybridity. And the same is true forpeople. Although we might assume humansobviously belonging to the realm of thesocial, as Whatmore shows when she dis-cusses the colonization of Australia, peoplehave always managed to construct a divisionbetween social and natural, which excludessome people. Having posited the world asrelational, the author provides a tool for mak-ing sense of it: hybrid geographies.

In order to understand the relationshipsthat constitute things, the author insists, wemust follow the journeys these things take.These might be literal journeys, such as thetransportation of leopards to Roman amphithe-atres, or the movement of captive-bredcrocodiles into the wild. They might alsoinvolve translation into information, such asthe drawing up of a statute or the monitor-ing of captive animals’ breeding stock, andthe movement of this data. Consequently, inthis book we travel back and forward in his-tory and traverse the globe. Whatmoreconjures up ‘wormholes’, holes between thepast and the present, the distant and the local,the strange and the familiar. These are notimaginary journeys (although they do seek tobe imaginative ones), neither is this booksimply a report of empirical work. Theauthor seeks to present us with ‘closely tex-tured journeys’ (Whatmore, 2002: 4). Theyencompass both the very personal (for exam-ple, invoking the moment she became amember of staff at UCL or getting lost on acar journey in Australia) and the more com-mon academic representations of the world.The journeys the people and animals in thebook take (literally and figuratively) consti-tute them.

Whatmore is magpie-like in her collectingof evidence, which she deploys to constructthese geographies using both words andimages. Using plans of the journeys theseanimals take (for example, Figure 2.4 onpage 25), description of the bureaucratic

SARAH WHATMOREÿÿ209


discourses, and ‘thick’ descriptions of thespaces these animals inhabit, she paints com-pelling pictures of their geographies ofhybridity. She illustrates how what we thinkof as ‘nature’ does not pre-exist the spaces andrelationships which it occupies.

Whatmore’s use of evidence in her argu-ment has attracted some attention. As Duncan(2004: 161) has noted, the empirical material inthe book is ‘marshalled as illustrations’ in amethodological, theoretical, and ethical project.While this is not in itself necessarily problem-atic, there is no conventional methods sectionwhich lays out the methods of selecting,collecting, and analyzing empirical data. Thismakes it hard for the more conventionallyminded social science reader to judge the mer-its of evidence and arguments with which theyare presented. Braun (2005: 835) sees HybridGeographies as ‘insistently empirical … (that it)seeks to explain nothing more than what isimmediately at hand’. While he is sympatheticwith Whatmore’s approach, he suggests herown knowledge claims, which are presented asdetached and objective, do not reflect her pro-fessed claim that knowledge is relational andprecarious. Demeritt (2005: 821) agrees: ‘Theepistemic modesty about partial and situatedknowledge is somewhat belied by some quitestrong claims about how that world actually is.’Conventionally, a methodology offers readersan account and defence of the mechanics ofthe research project(s). Such an account recog-nizes the precarious and inter-subjective natureof social science truth claims.

The methodological commitments demon-strated in Hybrid Geographies are fundamentalto many geographers’ vision of the ways inwhich the discipline ought to develop.Whatmore has since argued that we needto be risky and imaginative in the methodswe use:

(O)ne of the greatest challenges of ‘more-than-human’ styles of working as I see it isthe onus they place on experimentation …(There) is the urgent need to supplement

the familiar repertoire of humanist meth-ods that rely on generating talk and textwith experimental practices that amplifyother sensory, bodily and affectiveregisters and extend the company andmodality of what constitutes a researchsubject. (Whatmore, 2004: 1362)

Certainly, Hybrid Geographies meets thischallenge although, as noted above, it may be ina way that some geographers find a littleopaque. In a paper on Australia, Instone (2004)considers the book’s methodological lessons.She argues (Instone, 2004: 134) that Whatmorehas challenged geographers to focus ‘on thepractices through which nature is manifested insocial action’. Such a relational analysis requires,she argues, a ‘multilayed, multivalent, embodiedand situated approach’ (Instone, 2004: 131),with ‘connectivity at its heart’ (Instone, 2004:136). Such an approach will include textual,visual, oral and/or material evidence. Thesemethods force geographers into ‘a process offiguring what matters’ (Instone, 2004: 138).

The multilayering and ‘risky-ness’ of themethods espoused in Hybrid Geographies aremirrored by the prose. In the preface toHybrid Geographies Whatmore (2002: x) saysshe has tried to ‘hold onto some sense of thisenergetic fabrication in the writing’. Thisemphasis on opening up arguments and con-nections rather than closing them down orsettling them, can lead to rather dense text.However, it is an important strand in herargument. Whatmore (2005: 844) says she setsout to perform her philosophical commit-ments rather than merely state them. Shewants to open up new ways of thinkingabout nature and to make these imaginativelyas well as rationally plausible.

This leads to a style that some find delight-ful but others find frustrating. It allows thereader, and sometimes requires them, to beplayful. For example, there is no exclusivelylinear argument in the book. This book pre-sents the reader with a robust structure, albeita malleable one. The three sections structure

210ÿÿSARAH DYER


although there is not necessarily a lateral pro-gression of the arguments through the book.At its core, though, as Demeritt (2005: 882)notes, this is still at its heart an academic textwhich follows the appropriate scholarly con-ventions such as citation. For, in a sense, thebeginning chapters counter conventionalviews of nature as apart from the social, whilethe later chapters provide an alternative con-struction. Although the final chapter is not aconclusion to the book (it does not rehearsethe arguments or sum up the findings), itreturns to the aim set out at the beginning ofthe book to enlarge our notion of the social.

Impacts and significance

Hybrid Geographies received much interest andacclaim at the time of publishing. It wasdescribed variously as ‘splendid book, brim-ming over with interesting ideas’ and with‘compelling’ arguments (Duncan, 2004: 162)and marking ‘an important and impressivemilestone’ (O’Brien and Wilkes, 2004: 149). Itwas welcomed as ‘(r)einvigorating geographyand making it relevant to interdisciplinarywork’ (O’Brien and Wilkes, 2004). Geographersof a certain bent had been working on thequestions and literatures of science and tech-nology studies for a while. They embracedHybrid Geographies as a book that worksthrough the implications of geographic thoughtfor these concerns and promises the contribu-tion geography can make to this academic area.In the years since its publication the book hasbeen much cited, becoming what can betermed, in ANT parlance, an obligatory passagepoint, it is a necessary point of reference formany geographers of nature and informs theirthinking on the relational co-constitution ofnature and society.

In 2003, Noel Castree organized an author-meets-critic discussion at the annual confer-ence of the UK’s Royal Geographic Society(RGS). Those involved in this exchange pub-lished their thoughts in Antipode (Demeritt,

2005; Philo, 2005; Braun, 2005; Whatmore,2005). These papers provide thoughtful andsympathetic critique of the book. Braun(2005: 835) describes it as a book that‘believes in the world’. Demeritt (2005: 818)calls it fascinating. Philo (2005: 824) charac-terizes it as an ‘excellent book, whichseriously moves on debates’. All of thecontributions recognize that the book ischallenging and provide entries into, andquestions to enrich, our reading of the HybridGeographies. Having already described theconcerns expressed in part in this exchangeabout the status of the truth claims Whatmoremakes, I am going to concentrate here onthree further questions. The first is Demeritt’sconcern about how far we ought to extendWhatmore’s analysis of hybridity. The secondconcerns Braun’s calls for a clarification ofthe relationship between knowledge andethics/politics. Finally, I describe Philo’s anx-iety about the place of animals in the text.

The first question posed in the Antipodeexchange concerns the scope of Whatmore’sanalysis. I have said that at its core, HybridGeographies is a book about relations. Oneturn of phrase Whatmore (2002) uses in thisbook is that of ‘becoming…’, for example,she talks of becoming a leopard and becom-ing a soybean. In doing so, Whatmore seeksto highlight the networks and journeys thatco-construct leopards and soybeans. Whilewe might be convinced that this is a usefulway of thinking about the capacities of ani-mals, we ought to ask is it a useful way ofthinking about their fleshiness. Demeritt(2005: 829) asks whether we want to acceptan elephant’s skin and trunk are partial andprovisional achievements. He accepts such acritical realist challenge makes assumptionabout what partial and provisional mean.However, this is an important question to askin our reading of Hybrid Geographies. WhileWhatmore makes a compelling case for thesocio-natural hybridity of some aspectsof nature, how far does the scope of herargument extend?



The second of the Antipode questions con-cerns the place of animals in this text.Although this is a book about animals, thereis a sense in which they are absent, or at leastveiled, in the text. As Philo puts it:

… might it not be that the animals – indetail, up close, face to face, as it were –still remain somewhat shadowy pres-ences? They are animating the storiesbeing told, but … they stay in the mar-gins. (Philo, 2005: 829)

Although she discusses corporeal ‘fleshiness’,her examples are drawn from bureaucraticand all too human discourses. As anothercommentator has said:

Although at times … we may occasionallybe tempted to ask where the elephants are,in some ways this is the point here – thatthere are many aspects that define an ani-mal, what it can do, or will be made to do,which are constituted in wide circulationsof materials moving in and beyond itsfleshy body, though admittedly these canbe somewhat abstracted forms here.(Wilbert, 2004: 91)

Duncan too notes how the examplesWhatmore works through effectively illustratehow social networks interpenetrate the natural(Duncan, 2004: 162). He wonders what hybridgeographies of animals that are less enrolled inhuman projects would look like and posit thatthey would be less human-centred grantinggreater agency to animals.

The final consideration raised by theAntipode exchange that I will mention here isthe construction of ethics. The project ofethics is central to Hybrid Geographies. In anearlier paper, Whatmore lays out the basis ofthe relational ethics which makes up the‘conclusion’ of Hybrid Geographies (Whatmore,1997). Despite the centrality of ethics to theHybrid Geographies project, there is a sense inwhich it remains difficult to get a handle onthe resultant ethics. Braun (2005: 838) has

argued that ‘the question of knowledge andits relation to politics is not always clear inHybrid Geographies’, noting in particular theabsence of a mention of capitalism in heranalysis. Wilbert (2004: 92) has describedrelational ethics as rather vague. There is asense in which it is unclear what differencehybrid geographies make ethically. What canwe now understand, now do, that we couldn’tbefore? For example, conventional humanistethical theories take human agency andsubjecthood to denote a capacity for respon-sibility. Whether newly-enlarged hybridgeographies of agency and subjectivity bringwith them such moral responsibility is some-thing on which Whatmore remains silent.

The exchange in Antipode raises thesequestions as part of a thorough engagementwith Whatmore’s book and against a back-ground of praise for its ambitions andexecution. They provide useful questionsto enrich our own readings of HybridGeographies. The extent to which theseconcerns are judged important or workedthrough rests not with Whatmore herself butwith the extent to which the text proves use-ful to geographers, other academics, activistsand beyond. As Whatmore (2005: 842) herselfnotes, ‘books have lives of their own – prolif-erating connections as they travel that exceedany authorial intentions’.

Conclusion

Hybrid Geographies has been received as animportant and challenging book. It hasalready become a classic among a communityof geographers dealing with the relationshipbetween nature and society, already widelycited. To some working in this area, though,its analysis is too apolitical to be adopted. Forothers, its dense prose and flouting of impor-tant social science convention (such as anaccount of methods) make it inaccessible.

212ÿÿSARAH DYER


However, despite such reservations HybridGeographies has become known as an impor-tant and agenda-setting book.

This is a book that firmly positions geog-raphy as relevant to important issues facingthe world and the academic treatment ofthese issues. In a sense, this will be its acid test.The challenge is twofold. There is the needfor geographers to communicate with thosebeyond geography. There is also the challengeof creating methods and language for geog-raphers to communicate with those in thediscipline working in areas other than theirown. Elsewhere, Whatmore (2003) has dis-cussed the ability geography ought to have tomake sense of biological processes, such asdiseases and micro-organisms, as much aspolitical ones, such as international tradeagreements. She gives the case of banana

production and its threat by a virulent leafdisease called Black Sigatoka:

This obscure airborne fungus put thebanana in the headlines in European newspapers, heralding generic engineering asthe banana’s only salvation. Geographyshould be a discipline that fosters skills todeal as effectively with this pathologicalprocess as with the organisational rela-tions of banana production, but seems lessinclined to do so. (Whatmore, 2003: 139)

If Whatmore’s book Hybrid Geographies canprompt such research agendas, and if it playsa role in the conversation between humanand physical geography, it will be a real suc-cess. The book will be creating its ownhybrid networks. It is certainly up to thechallenge.


Secondary sources and further readingBraun, B. (2005) ‘Writing geographies of hope’, Antipode 37 (4): 834–841.Duncan, J. (2004) ‘Hybrid Geographies: Natures, Cultures, Spaces’ (Book review),

Social and Cultural Geography 5 (1): 160–162.Demeritt, D. (2005) ‘Hybrid geographies, relational ontologies and situated knowl-

edges’, Antipode 37 (4) 819–823.Instone, L. (2004) ‘Situating Nature: on doing cultural geographies of Australian nature’,

Australian Geographer 35 (2): 131–140.O’Brien, W. and Wilkes, H. (2004) ‘Hybrid Geographies: Natures, Cultures, Spaces’

(Book review), The Professional Geographer 56 (1): 149–151.Philo, C. (2005) ‘Spacing lives and lively spaces: partial remarks on Sarah Whatmore’s

Hybrid Geographies’, Antipode 37 (4): 824–833.Whatmore, S. (1997) ‘Dissecting the autonomous self’, Environment and Planning D:

Society and Space 15 (1): 37–53.Whatmore, S. (2002) Hybrid Geographies: Natures, Cultures, Spaces. London: Sage.Whatmore, S. (2003) ‘From banana wars to black sigatoka. Another case for a more-

than-human geography’, Geoforum 34 (2): 139.Whatmore, S. (2004) ‘Humanism’s excess: some thoughts of the “post-human/ist”

agenda’, Environment and Planning A 36 (8): 1360–1363.Whatmore, S. (2005) ‘Hybrid Geographies: Author’s responses and reflections’,

Antipode 37 (4): 842–845.Wilbert, C. (2004) ‘Hybrid Geographies: Natures, Cultures, Spaces’ (Book review), Area

36 (1): 91–92.


CITIES (2002): ASH AMIN AND NIGEL THRIFT

Alan Latham

Cities have to be seen less a series oflocations on which categorical attributes arepiled, and more as forces and intensities thatmove around and from which, because of theirconstant ingestions, mergers and symbioses,the new constantly proceeds… Life in the citycontains magical powers; it is full to brimmingwith an abundance of life… (Amin and Thrift,2002: 91)

Introduction

Urban geography and urban studies is a dis-cipline populated by Big Things. Cities for astart. They are by definition big. Motorwaysand mass transportation systems, urban rede-velopment projects, and suburban shoppingmalls are pretty big too. Then there areskyscrapers, mega-projects, new-towns, edge-cities, again all large, obvious, written acrossthe landscape and close to the heart ofurban geography’s sense of itself. And that isto say nothing of deindustrialization, subur -ban-sprawl, or inner-city dereliction andabandonment. Nor is it to mention the tens ofthousands of miles of pipes, cables, and fibreoptics that allow a contemporary city to func-tion and which urban geography, led byinnovate scholars like Stephen Graham andSimon Marvin (2001), has recently – if a little belatedly – discovered as being central tocontemporary patterns of urbanization. Urbangeography is also a discipline populated withBig Theories. Much of the most interesting

work that formed the so-called ‘quantitativerevolution’ in human geography was urban.Melvin Webber (1964) was talking about the‘non-place urban realm’ decades before themuch trendier French anthropologist MarcAugé (1995) rediscovered the term in the1990s, while Brian Berry with his centralplace hierarchies and rank size models wasthe world’s most cited human geographerthrough much of the 1960s and 70s (seeBerry, 1973). Indeed, Berry was toppled fromthis throne by another urban geographerwith a Big Idea, David Harvey. A committedand intellectually charismatic Marxist whohad begun his career as a quantitative geog-rapher, Harvey (1973, 1982, 1989) arguedthat the ‘urbanization of capital’ was at theheart of the evolution of contemporary cap-italism. According to him, if we want tounderstand how capitalism works we have tounderstand how cities work (see Castree,Chapter 8 this volume; Woodward & Jones,Chapter 15 this volume). More recently theUCLA-based urban geographer Ed Soja(1989, 1996, 2000) has been undertaking ahugely ambitious rethinking of the ontolog-ical foundations of human geographythrough the contemporary urban landscape(see Minca, Chapter 16 this volume). And, ifthat were not enough Big Ideas for one sub-discipline, Michael Dear (2000, 2002) hasbeen trying to do something similar to Soja ina series of books that seek to trace the outlineof a postmodern urbanism that does notmerely mark out a new pattern of urbanization,

25


216ÿÿALAN LATHAM

but also the rudiments of a new postmodernsocial science.

Given its attraction to the Big, it is per-haps of little surprise that urban geographyhas not been particularly good at, or indeedoften even interested in, making sense ofmany of the smaller elements that make upa city. The day-to-day routines of shoppingand household provisioning, the life of pub-lic places like parks, sidewalks and shoppingmalls, the networks of friendship and enthu-siasm, and so forth that give urban life somuch of its texture have not often had aprominent place within the sub-discipline’stheoretical or empirical imagination. It hasby and large been urban sociologists,anthropologists and social historians, whohave furnished us with these kinds of moreintimate, more street level, narratives of thecity. It is not just that urban geography hasnot been much good at making sense ofthese diverse everyday practices. Thereseems to be something in the quality ofthese practices that is resistant to the BigTheorizing dominant throughout urbangeography. A key question for urban geogra-phy is whether this matters? Does urbangeography need to be able to speak to theseeveryday practices? Is the texture of thisintimate city, the ‘sense of the marvelloussuffusing everyday life’ (Aragon, 1971: 23)that it speaks to, necessarily important towhat urban geography is about? Could itsimply be within the ‘natural’ division of thesocial sciences that urban geography shouldonly try to speak of the big? That urbangeography’s natural terrain is the broadoverview, the synthesizing, somewhat dis-tanced, gaze? Or, does the inability to writeabout the everyday ecologies of cities repre-sent something more significant? Does itsymbolize a failure of the urban-geographi-cal imagination that limits its usefulness, itsapplicability?

Ash Amin and Nigel Thrift’s (2002) Cities:Re-imagining the Urban, is a bold and ambitious

attempt to address – if not always directlyanswer – the above questions. In the space ofjust over 150 pages they seek to outline a setof novel ways of thinking about urban life,ways of thinking that transcend, or at thevery least go beyond, the orthodoxies pre-vailing throughout both urban geographyand urban studies. This might sound likeanother attempt at yet more Big Theory, theproduction of yet more Big Ideas. But Aminand Thrift’s argument throughout Cities ismore subtle than that. Urban geography andurban studies is, they argue, underpinned bya series of largely taken-for-granted, andapparently obvious, assumptions about justwhat elements make a city a city. Amin andThrift want to place into question theseunderlying assumptions, and at the same timesuggest an alternative framework throughwhich to go about understanding the urban.They want to try to create a more open,more vital, and more populated style of urbananalysis. A style of analysis that recognizes theongoing incompleteness, the fuzzy-ness, thestrange, often unpredictable, elements thatmake up a city. In Amin and Thrift’s ownwords:

The city has no completeness, no centre,no fixed parts. Instead, it is an amalgamof often disjointed processes and socialheterogeneity, a place of near and far con-nections, a concatenation of rhythms;always edging in new directions. This isthe aspect of cities which needs to beexplained. (Amin and Thrift, 2002: 8)

What Amin and Thrift want to do is notsimply demonstrate the importance of the so-called everyday. They want to try to steer usaway from thinking in simplistic terms of bigor small, near and far, global or local, andinstead encourage us to explore how throughall sorts of relationships cities emerge throughcomplex patterns of connection. In short, theywant to invite their readers to try and explorea different kind of urban ontology.


ASH AMIN AND NIGEL THRIFTÿÿ217

Key arguments: Toward a newontology of the urban

How then should we go about developingthis new kind of urban analysis? What toolsdo we need? What styles of thinking? ForAmin and Thrift there are two key elementsthat need to be addressed in answer to thesequestions. The first question concerns themetaphorical repertoire through which urbangeography and urban studies frames itsunderstanding of cities. This needs to beenlarged and invigorated. Put simply, weneed a new, and more inventive, set ofmetaphors in addition to those we are usedto working with. Without new metaphors –and the shift in style of thinking that newmetaphors bring with them – urban geogra-phy and urban studies will be unable to doproper justice to the heterogeneity and com-plexity that Amin and Thrift are pointing to.Secondly, they explore the contention thaturban geography and urban studies needs toreach beyond the relatively narrow range ofintellectual traditions upon which it haddepended for much of its history. Drawingon a heterogeneous and often neglected col-lection of thinkers ranging from seventeenthand eighteenth-century European philoso-phers such as David Hume, John Lock andBaruch Spinoza, through to early twentieth-century writers such as the psychologistWilliam James, the sociologist Gabriel Tarde,and the philosophers Henri Bergson andAlfred Whitehead, and later to the work ofMichel Serres, Gilles Deleuze, and finallythrough to contemporary writers like theFrench philosopher Bruno Latour. Amin andThrift (2002: 27) argue that we need todevelop ways of considering cities that placestress on what they call ‘an ontology ofencounter or togetherness based on the prin-ciples of connection, extension and continuousnovelty’ .

This hardly sounds like we are clarifyingmatters any. But it is worth sticking a little

longer with the intellectual inspirations ani-mating Amin and Thrift’s account becausethe whole sense of the argument that ener-gizes Cities from the start of the book to thevery finish depends on the distinctiveness ofthese inspirations. For Amin and Thrift, theanalysis of cities that emerges through anengagement with the intellectual traditionsmentioned above has a very distinctive style:

All philosophies of becoming have a num-ber of characteristics in common. One is anemphasis on instruments, on tools as a vitalelement of knowing, not simply as a passivemeans of representing the known. The sec-ond is their consideration of other modes ofsubjectivity than consciousness. The third isthat ‘feelings’, howsoever defined, areregarded as crucial to apprehension. Thefourth is that time is not a ‘uniquely serialadvance’ … but rather exists as a series ofdifferent forms knotted together. Fifth,becoming is discontinuous, ‘there is abecoming of continuity, but no continuityof becoming’ … And finally, and mostimportantly, this means that ‘prehensions’(ideas about the world) can constantly bebuilt. More and more can be put into theworld. (Amin and Thrift, 2002: 27–28; thequotes are from Whitehead, 1978)

So, Amin and Thrift want to think aboutways of writing about cities that activelyaccount for the role of all sorts of non-human actants in the life of cities. They wantto explore ways of writing about the citiesthat actively engage with how they areinhabited by all kinds of bodies, non-humanand human. They want to develop ways ofwriting about cities that recognize how citiesare made (and emerge) through multiple, andoften conflicting, orderings of time and space.And they want to try and find ways of writ-ing about urban life that respect the complex,heterogeneous, becomings that define cities –ways of writing that pay attention to themany unlikely, surprising, and unexpected


218ÿÿALAN LATHAM

patterns of association that are constantly ani-mating cities in all sorts of ways.

Key themes and content: networked,machinic, and powerful cities

To develop a sense – feeling might be a moreappropriate word – of what an analysis of theurban organized through such an analyticalstyle might look like, Amin and Thrift offerthe reader five different accounts of city lifein the substantive chapters that follow thebook’s first chapter. At the risk of some sub-stantial simplification, the thematic concernsof these five chapters can be summarizedunder three main headings, developing ideasof cities as networked, as machinic and as‘powerful’ (Amin and Thrift, 2002: 105).

In relation to the first of these, Amin andThrift work through debates about the distan-ciation of social and economic relations,alighting on ideas that geographers need tothink about cities as sites in distanciated net-works. Of course, the idea of cities as networksis not new. The urban sociologist ManuelCastells (1989, 1996) has placed the idea ofnetwork at the centre of his analysis of infor-mational cities for almost two decades. And wellbefore that, researchers interested in urbancommunity such as Barry Wellman (1979) andClaude Fischer (1982) had made extensive useof the term. Amin and Thrift, however, substan-tially extend, and reformulate, the networkmetaphor. For them the metaphor of the net-work is productive, not just because it speaks ofpatterns of connection, but because of how itsuggests novel ways of considering how con-nections are built. In Amin and Thrift’s thinkingnetworks include not just humans and socialinstitutions. They also include all sorts of non-human entities that allow a network to comeinto being. They include the computers, thecommunication systems, the software, thebureaucratic procedures, the meeting places; inshort all the various assemblages of materials

that allow a network to achieve the stability itneeds to be called a network. Thus, the net-works they are talking about should moreproperly be called, following the work of writ-ers like Bruno Latour (1993) and MicheleCallon (1998), actor-networks.

Thought about in these terms, the metaphorof network is useful for Amin and Thrift forat least four reasons. Firstly, it is usefulbecause it places into question both who andwhat is acting. It allows us to think about howsocial action is distributed across a range ofentities rather than originating from anapparently autonomous individual. One ofthe most striking things about contemporarycities is the number of non-human actantsthat are monitoring, analyzing, and actinginto the world. So, for example, Amin andThrift (2002: 125) stress ‘the modern cityexists as a haze of software instructions’.These instructions form a kind of digitalunconscious without which the modern citysimply would not be able to function. A sec-ond strength of the network metaphor is thatit helps to think about how boundary andscale are defined in a world characterized inall sorts of ways by distanciated relationships.Networks often generate topologies of connection that bear little relationship toEuclidian space. So, to highlight but oneexample, within a transnational corporationwhat counts as ‘local’ is that which is inte-grated into the organization – part of itscommunication structure and bureaucraticprotocols – rather simply that which is sim-ply nearby. Thirdly, the network metaphor isalso useful because it highlights the knowl-edge, effort, skill and work that goes intomaking the world have the spatial-temporalfeel – the rhythms – that it does. Amin andThrift, thus, spend a great deal of time focus-ing on the amount of resources that go intothe maintenance of the ordinary life of a city,and the precariousness of this achievement.They describe cities populated not just withhigh powered professionals, and global


corporations, but ones that are also cruciallydependent on all sorts of technicians, and ser-vice workers to keep them going. Fourthly,and finally, the metaphor of network is usefulas it encourages us to think of cities as notprimarily defined by proximity per se (as it isin so much of contemporary urban studies)but as ‘sites’ (Amin and Thrift, 2002: 63) forthe staging of certain kinds of proximity.

Like the idea of the city as network, themetaphor of cities as machines also featureslarge in Cities. The notion of cities asmachines clearly has an established heritagewithin urban studies and urban geography(Mumford, 1934; Molotch, 1976). But againAmin and Thrift (2002: 78) have somethingdistinctive in mind when they write of theneed ‘to understand the city as a machine’:

In using [machine] we do not mean toimply that the city can be understoodthrough mechanical metaphors – withinputs, mechanisms and outputs – butrather as a ‘mechanosphere’, a set of con-stantly evolving systems or networks,machinic assemblages which intermixcategories like the biological, technical,social, economic, and so on, with theboundaries of meaning and practicebetween the categories always shifting.(Amin and Thrift, 2002: 78)

As with the metaphor of the network,Amin and Thrift’s use of the metaphor of themachine – and the associated pronoun,machinic – is productive because it points tohow cities are built through the organizing ofall sorts of materials and tools. The notion ofcities as machines also stresses how all thesematerials can not only ‘act’, but also develop togenerate all sorts of unlikely and unexpectednew assemblages that exceed the origins plans,or patterns, of use intended for them.

So, the metaphor of the city as machine –at least in the sense Amin and Thrift use it –removes humans from being the necessarycentre of any account of cities. Instead, they

become but one element, or set of elements,albeit very important, along with all theother elements that make up the ecology ofcities. This ethological approach introduces awhole new population of entities into urbangeography. If we focus on cities as ecologiesin the fullest sense then suddenly we are con-fronted by a world populated not just withpurposeful actors, but also all kinds of differ-ent entities, ranging from viruses andbacteria, to buildings, rivers and deserts, tobrown rats, cockroaches, and urban foxes – andthis is to say nothing of the intricate techno-logical assemblages already mentioned in thediscussion of networks above. This ethologicalapproach also prompts us to consider the needto understand ourselves as biological entities,as yet another species in the mechanosphere ofcities, moved by all sorts of passions.

In adding passions to the metaphor ofmachine, Amin and Thrift are drawing directinspiration from the French philosopher GillesDeleuze’s rereading of the seventeenth-century Dutch philosopher Baruch Spinoza.As Amin and Thrift (2002: 84) put it, Spinoza’s‘thought can best be conceived as a kind ofphysics of bodies in which the human body isnot a self-contained whole but built out ofbodies with our own’. To think about passions,then, is also to think about the ways in whichcities are made up of all sorts of communitiesorganized through particular assemblages ofpassion. These can range from the intimate andemotionally intense such as with the connec-tion of families, of lovers, of friends, to thecommunities of enthusiasm that includepractices as diverse as ‘Civil War re-enactmentsocieties, the showing of cats and dogs, ortrainspotting’ (Amin and Thrift, 2002: 83)and fishing, to the assemblages of mutual pur-pose and indifference that are the crowdedspaces of big city mass public transportationsystems. And finally, to think about passions,and think about cities as machinic, is also tohighlight how cities emerge out of the knot-ting together of a whole range of complex



mechanisms for generating and regulatingpassions. Cities function through the constantand iterative ‘engineering of certainty’ (Aminand Thrift, 2002: 93) through all kinds of orga-nizing mechanisms, from postal systems, andtransportation networks, to the sports leaguesand popular media that help lend the ebb andflow of a city’s economy of passions a certainstructure.

To think about the ‘engineering of cer-tainty’ leads directly to the third and finaloverarching theme of Cities; cities as sites ofpower. In stressing the importance of net-work and flow, and parallel to that the placeof ‘passions’, Amin and Thrift might seem tobe outlining an overly cheery, voluntaristic,account of cities. To read their argument inthese terms is to misinterpret what Amin andThrift are trying to do. Throughout the pagesof Cities they stress that cities are striated withpower in all kinds of ways. However, forAmin and Thrift if the ‘ontology of becom-ing’ that they are outlining is to be treatedseriously, then there is a need also to recon-sider how the power that circulates in citiesworks. So, Amin and Thrift develop a distinc-tive approach to power:

Rather than focusing on the conventionalinterest in urban studies on the domina-tion and oppression by certain kinds ofactors and institutions, our concern iswith power as a mobile, circulating forcewhich through the constant re-citation ofpractices, produces self similar outcomesmoment by moment. (Amin and Thrift,2002: 105)

This way of thinking through power is per-haps best understood as ‘diagrammatic’, in thesense that French historian Michel Foucaulttalked of forms of modern power and controlbeing organized through certain ‘diagrams’ forconsidering how society should be assembledand governed. These ‘diagrams of power’ oper-ate as ‘abstract machines’ (Amin and Thrift,2002: 106) that work to process the world into

a certain kind of consistency. Modern cities,then, might be seen as being organizedthrough at least four diagrams: bureaucracy,production, sensuality, and imagination. Citiesare permeated with a will to count, collect, cat-egorize, and order all sorts of information andmaterial. Equally, they are driven by a will toproduce, whether that be economic growth,clean streets, or educated citizens. They arealways organized around tasks of produce,either more of something, or in some cases less.Rarely are they organized around keepingthings exactly as they are. Similarly, cities gen-erate and organize all sorts of sensualities, andjust as they generate them, so do they almostsimultaneously throw up all sorts of techniquesand technologies to manage, regulate, and con-trol them. And, lastly, cities are remarkable fieldsfor imaginative endeavour. The shaping andregulating of this imaginative space is central tothe exercise of any modern organization.

Reception

Cities is not an easy book to digest. It isdesigned to challenge and question thereader’s view of what cities are and how theyshould be understood. This wilful awkward-ness has shaped much of Cities’ criticalreception to date. On the one hand, in theshort time since its publication the book hasbeen widely read and highly cited bothwithin urban geography, urban studies, andindeed elsewhere in the social sciences. Thisis hardly surprising, given the prominentplace both Ash Amin and Nigel Thrift hadalready established for themselves withinurban geography and urban studies by thetime of Cities’ publication: between themAmin and Thrift had published dozens ofarticles and book chapters on a wide range ofurban issues, and both held professorships atprominent British geography departments(Amin at Durham University, Thrift at theUniversity of Bristol). As such, Cities has been

220ÿÿALAN LATHAM


read by many as simultaneously a kind ofsummary of Amin and Thrift’s key arguments,and an overview (following Thrift, 1996) ofhow one might go about doing a so-called‘non-representational’ style of urban geography.

On the other hand, it is not at all clear howsustained urban geography and urban studies’engagement has been with the argumentsmade in Cities. While most of the reviews ofthe book in academic journals were broadlypositive, almost all raised doubts abouthow a researcher would actually go aboutdoing Amin-and-Thrift-ian inspired urbanresearch. In part these doubts were because, asone sympathetic reviewer wrote (Savage,2003: 807), its ‘perspective is so distinctivethat it does not engage with much “actuallyexisting” urban scholarship’. It follows that toget more than just a passing sense of the kindof accounts that Amin and Thrift are askingurbanists to try to write requires the reader toinvest an enormous amount of intellectualenergy going back to writers like BrunoLatour, Gilles Deleuze, and Alfred Whiteheadthrough which the argument of Cities isdeveloped. And while, as Amin (2007) showsin a recent article surveying urban geography,there is a growing body of literature that fol-lows the lines of thinking outlined in Cities,this writing has hardly come to define themainstream of the sub-discipline.

More tellingly, many reviewers questionedthe applicability of Amin and Thrift’s newstyle of urban theory. These questions wereboth conceptual and methodological. Somewondered what kinds of methods and tech-niques researchers would need to use tooperationalize Amin and Thrift’s theoreticalvision. Would they simply use those theywere already using? Or were a whole bunchof new methods required? Or some mix ofthe old and the new? These are questionsthat, as Ian Gordon (2003: 519) wrote in areview in Progress in Human Geography, ‘arenever really confronted’. Gordon also won-dered why it was that – in his reading – ‘all

notion of inequalities of interest and powerdisappear’ (Gordon, 2003: 520) in Amin andThrift’s account. For all the interesting andnovel elements of urban life that Cities throwup, it seems to miss what many urbanresearchers would consider the ‘brute factsabout urban life’ (Gordon, 2003: 520). That isto say, it somehow loses a sense of all the waysin which cities are organized around ques-tions of life and death; how they are definedby profound and often deep-seated povertyand exclusion, how they are often sickeninglyunequal and unfair places. As Lynn Staeheli(2004: 86), put it in her review in UrbanGeography ‘[i]n this time of the revanchivistcity, the “soft” politics implied in [Cities]seem somehow not enough’.

Working with Amin and Thrift

Populated with urban anglers, foxes, street-lights, airports, computer software, networksof friendship and enthusiasm, computerscreens and keyboards, hands and fingers,roller-coasters, bacteria and viruses (microbi-ological and electronic), lawns, automobiles,databases and accounting systems, films andnovels and guidebooks, regulations of all dif-ferent kinds, telephone networks, styles oftalk, league tables, all sorts of non-human andhuman passions, Cities: Re-imagining the Urbanpresents a striking, distinctive, and intellectu-ally challenging attempt to re-think urbangeography and urban studies. It also offers asuccinct introduction to the intellectual pro-jects of two of the most important urbangeographers writing today.

Nonetheless, Cities is more than just asummary of arguments that Amin and Thrifthad outlined in previous publications. AsAmin and Thrift write at the start of thebook:

We see [Cities] as a kind of stagingpost towards a different practice of



222ÿÿALAN LATHAM

urban theory based on the transhumanrather than the human, the distanciatedrather than the proximate, the dis-placed rather than the placed, and theintransitive rather than the reflexive.(Amin and Thrift, 2002: 5)

Cities, then, is much more about outlininga project in the making than a summary of awell-established intellectual position. Thismakes for exciting reading. ThroughoutCities one often has little sense of whereAmin and Thrift will go next in their argu-ment. And equally, one is also constantlysurprised by the material that they draw onto elaborate their argument. Do they reallythink that a history of lawns would tell us agreat deal about contemporary patterns ofurbanization? Do they really think that weshould carefully consider the sociality ofurban anglers? That we need to be aware ofthe conversational rhythms of telephoneconversations?

Well, the obvious answer is, ‘Of course theydo!’ The real question the reader needs to askafter reading through Cities is not, ‘Do Aminand Thrift’s examples hold up?’ It is, ‘DoesCities outline an approach to thinking aboutcities that actually works?’ And, along withthat question, the reader must also ask, ‘DoAmin and Thrift offer us enough of a sense ofwhat this new style of urban analysis, thisstudy of cities as sites of constant becoming,involves?’ Which leads to a third, and equallyimportant, question, ‘Do Amin and Thriftmanage to map out a way of thinking aboutcities that actually offers urban geography and

urban studies more analytical leverage thanthe theoretical approaches that it already has?’Well, the answer to each of those questions,unfortunately, is that it is too early to tell.Amin and Thrift’s book is too short, and, theenormous productivity of each of the co-authors aside, not nearly enough material hasbeen published elsewhere on cities in thestyle that Cities is arguing for to really makea careful judgement on any of these threequestions.

Conclusion

Ultimately, Cities – as all new ways of think-ing are – is dependent on a kind of wager. Awager that in working to show up all sorts ofnovel elements of urban life, that in trying tothink about cities in a style outside of theestablished grooves of conventional urbangeography and urban studies, it will invigo-rate and expand our understanding of howcities can be thought about in ways thatexisting approaches simply cannot. In a sense,then, one can only discover if the ideas ani-mating Cities work by trying to work withthem and see what one comes up with. So,whether Cities will cure urban geographyand urban studies of its love of Big Thingsand Big Ideas is perhaps doubtful. But, readwith the sense of intellectual adventure withwhich Amin and Thrift approached its writ-ing, Cities is one of those surprisingly rarebooks within human geography, a book thatbears repeated study.

Secondary sources and referencesAmin, A. (2007) ‘Re-thinking the urban social’, City 11 (1): 100–114.Amin, A. and Thrift, N. (2002) Cities: Re-imagining the Urban. Oxford: Polity.Aragon, L. (1971) Paris Peasant. London: Picador.Augé, M. (1995) Non-Places: Introduction to the Anthropology of Supermodernity. London:

Verso.Benjamin, W. (1973) Charles Baudelaire. London: Verso.Benjamin, W. (1978) One Way Street and Other Writings. London: Verso.



Berry, B. (1973) The Human Consequences of Urbanisation: Divergent Paths in the UrbanExperience of the Twentieth Century. London: Macmillan.

Callon, M. (1998) Laws of Markets. Oxford: Blackwell.Castells, M. (1989) The Informational City. Oxford: Blackwell.Castells, M. (1996) The Network Society. Oxford: Blackwell.Caygill, H. (1998) Walter Benjamin: The Colour of Experience. London: Routledge.Dear, M. (2000) The Postmodern Urban Condition. Oxford: Blackwell. Dear, M. (ed.) (2002) From Chicago to LA: Making Sense of Urban Theory. Beverley Hills:

Sage.Fischer, C. (1982) To Dwell Among Friends. Berkeley: University of California Press.Gordon, I. (2003) ‘Review of Amin and Thrift Cities: Re-imagining the Urban’, Progress in

Human Geography 27 (4): 519–520.Graham, S. and Marvin, S. (2001) Splintering Urbanism. London: Routledge.Harvey, D. (1973) Social Justice and the City. Oxford: Blackwell.Harvey, D. (1982) The Limits to Capital. Oxford: Blackwell.Harvey, D. (1989) The Urban Experience. Oxford: Blackwell.Latour, B. (1993) We Have Never Been Modern. Hassocks: Harvester Wheatsheaf.Lefebvre, H. (1991) The Production of Space. Oxford: Blackwell.Lefebvre, H. (2004) Rhythmanalysis: Space, Time and Everyday Life. London: Continuum.Molotch, H. (1976) ‘The city as a growth machine’, The American Journal of Sociology

82: 309–318.Mumford, L. (1934) Technics and Civilisation. New York: Harcourt and Brace.Savage, M. (2003) ‘Review of Amin and Thrift Cities: Re-imagining the Urban’, Sociology

37 (4): 806–808.Soja, E. (1989) Postmodern Geographies. London: Verso.Soja, E. (1996) Third Space. Oxford: Blackwell.Soja, E. (2000) Postmetropolis. Oxford: Blackwell.Staeheli, L. (2004) ‘Review of Amin and Thrift Cities: Re-imagining the Urban’, Urban

Geography 25 (1): 84–86.Thrift, N. (1996) Spatial Formations. London: Sage.Thrift, N. (2005) Knowing Capitalism. London: Sage.Webber, M. (1964) Explorations into Urban Structure. Philadelphia: University of

Pennsylvannia.Wellman, B. (1979) ‘The community question: the intimate networks of East Yorkers’,

American Journal of Sociology 84, March: 1201–1231.Whitehead, A.N. (1978) Process and Reality. New York: Free Press.


FOR SPACE (2005): DOREEN MASSEY

Ben Anderson

For the future to be open, space must be opentoo. (Massey, 2005: 12)

Introduction

On page 108 of For Space, an impassionedbook that discloses the theoretical and politicalchallenges of thinking space, a map of part of

the South-East of England is inscribed with avery simple if perhaps initially puzzling phrase:‘ceci n’est pas l’espace’

The phrase recalls Rene Magritte’s famousinscription below a painting of a pipe: ‘cecin’est pas une pipe’ (this is not a pipe). Initially,like Margritte’s phrase, it may seem odd –counterintuitive perhaps – since we are beingrather bluntly informed that a map of roads

26

Figure 26.1 ‘Ceci n’est pas I’espace’ (Figure 11.1 in Massey, 2005)


and motorways, railway lines, topography,fields and villages is not space. Odd becausemaps have become central to how we thinkabout and imagine space. Yet maps, perhapsthose we are most familiar with, function byrepresenting space as an ordered surface inrelation to which the observer is positionedoutside and above. Massey’s point is a simpleone that is now echoed in a critical literatureon cartography – that hegemonic types ofmapping represent space as a ‘completedhorizontality’ – in which the dynamism ofchange is exorcised in favour of a totality ofconnections. Mapping is one of a numberof ways in which the disruptiveness ofspace is tamed. Offering an alternative non-euclidean imagination of space, that disruptsthis and other problematic accounts of space,is therefore the pressing task that animates ForSpace: a book that Massey (2005: 13) summa-rizes as comprising ‘an essay on the challengeof space, the multiple ruses through whichthat challenge has been so persistently evaded,and the political implications of practising itdifferently’.

The basis to an alternative approach tospace can be articulated in a set of threeintertwined propositions:

• Space is the product of interrelations; thuswe must recognize space ‘as constitutedthrough interactions, from the immensityof the global to the intimately tiny’(Massey, 2005: 9).

• Space is the sphere of the possibility ofthe existence of multiplicity; that is space‘as the sphere in which distinct trajecto-ries coexist; as the sphere therefore ofcoexisting heterogeneity’ (Massey, 2005: 9).

• Space is always under construction; ‘it isalways in the process of being made. It isnever finished; never closed’ (Massey,2005: 9).

For Space is an argument for the recogni-tion of these three characteristics of space

and for a lively, heterogeneous, progressivepolitics that thereafter responds to them. Thethree propositions therefore aim to enable usto ponder the challenges and delights of spa-tiality and subsequently open up the politicalto the challenge of space – perhaps disrupt-ing how political questions are formulated,perhaps intervening in current argumentsand perhaps contributing to alternative imag-inations that enable different spaces to be.

The double aim of For Space – to simulta-neously open up our thinking of the spatialand the political – resonates with Massey’swork over the past two decades. Fromresearch on industrial restructuring and thesocial division of labour (see Massey, 1984;Phelps, Chapter 10 this volume), through totheoretical work on the emergence and dis-ruption of power-geometries (see Massey,1994), Massey has been a consistent advocateof the political necessity of teasing out themutual imbrications of the spatial and thepolitical. If For Space therefore chimes withseveral of Massey’s abiding concerns then italso resonates with the emergence of a rangeof poststructuralist geographies that associatespace with dynamism and thus qualities ofopenness, heterogeneity and liveliness (see,for example, Amin and Thrift, 2002; Doel,1999; Murdoch, 2006; Whatmore, 2002). Theother context she writes in is, however, thepersistence of a set of problematic associa-tions around space that we have inheritedfrom a set of philosophical lineages and thatare constantly articulated in contemporarypolitics. The first section of this essay reviews,therefore, Massey’s critical engagement withother imaginations of space. Section twomoves on to draw out the alternative con-ception of space that For Space outlines byreturning to explicate the three propositionsintroduced briefly above. Section three thenthinks through more precisely how Massey’salternative conceptualization of space offersand promises Human Geography a type of

226ÿÿBEN ANDERSON


‘relational politics’. In the conclusion I raise aseries of questions about the relationalapproach to space that For Space exemplifiesand argue that what is distinctive about thebook is that it offers a specific ethos ofengagement which trusts that ‘there arealways connections yet to be made, juxtapo-sitions yet to flower into interaction (or not,for not all potential connections have to beestablished), relations which may or may notbe accomplished’ (Massey, 2005: 11).

‘Unpromising associations’

The title of Doreen Massey’s book, For Space,provokes a simple question. Why For Space?The title declares that space matters. That itinflects how we engage, understand andapproach the world. So conceptualizing spaceshould, therefore, be a pressing concern forus – it should cause us problems, make usthink, and interest us. Yet the title is not aboutspace or thinking space or questioning space. Bydeclaring she is for space Massey affirms thepossibilities and potentialities enabled byspace(s). I will interrogate these possibilitiesin sections three and four but before we candisclose them we need to interrogate the‘unpromising associations’ that, for Massey,serve to conceptualize or assume space to besimply the negative opposite of time. Despitethe reassertion of space in social theorywhich has made space part of the lexicon ofthe social sciences and humanities over thepast two decades or so, deeply ingrainedhabits of thought continue to tie space to a setof dehabilitating assumptions. These areassumptions that are fundamentally embeddedin the framing of a range of contemporaryproblems. Central to the history of modernity,for example, has been a translation of spatialheterogeneity into temporal sequence. Differentplaces are interpreted as occupying differentstages in a single temporal sequence in the

various stories of unilinear progress thatdefine the West against the rest (such as mod-ernization or development). Talk of the‘inevitability’ of neo-liberal ‘globalization’,to give another example, assumes both afree unbounded space and that globaliza-tion takes only one form. In both cases –and we can think of others such as the ideathat space can be annihilated by time – thecontemporaneous heterogeneity of theworld is all too easily forgotten and thusdifference erased.

In aiming to discern how such a taming ofthe spatial is also present in a range ofphilosophers and political theorists, Massey’sconcern is not, it should be noted, simplywith how time has been prioritized overspace – a claim that has been central to thereassertion of space but is itself tied to prob-lematic assertions that we live in uniquely‘spatial times’ (e.g. Soja, 1989). Instead sheinterrogates how space has been attached to aset of ‘unpromising associations’ in the workof a set of theorists and theories broadlyunderstood as either structuralist or post-structuralist (including Althusser, Bergson,Laclau and Derrida). She describes her rela-tion with these theorists and schools ofthought in strikingly affective terms. In rela-tion to their treatment of space she is:

Puzzled by a lack of explicit attention theygive, irritated by their assumptions, con-fused by a kind of double usage (wherespace is the great ‘out there’ and the termof choice for characteristics of representa-tion, or of ideological closure), and,finally, pleased sometime to find the looseends (their own internal dislocations)which make possible the unravelling ofthose assumptions and double usages andwhich, in turn, provokes a reimaginationof space which might be not just more tomy liking, but also more in tune with thespirit of their own enquiries. (Massey,2005: 18)

DOREEN MASSEYÿÿ227


Despite her puzzlement and irritation, thelast line in this quote stresses that Massey’sengagement with this range of thinkers is areparative rather than dismissive one. Ratherthan condemn them, and in that act of dis-missal separate her own approach from theirs,Massey’s critique aims to disclose a range ofnew potential openings. Each of the theorists,and schools of thought, offer something toMassey’s project. From Bergson she under-stands questions of the dynamism of life – ofliveliness. Structuralism offers an understand-ing of how the identity of entities is madeout of relations; whilst deconstruction heraldsa constant enlivening interruption to space.Yet in her engagement with each she arguesthat space takes on a set of two ‘unpromisingassociations’ that either implicitly or explic-itly tame space and refuse the challenge ofunderstanding its singularity as the realm of‘radical contemporaneity’. First, a conceptu-alization of space as static that equates spacewith a stabilization of life. Space is assumedto conquer the inherent dynamism of timeby imposing an order upon the life of thereal – ‘spatial immobility triumphs temporalbecoming’ (Massey, 2005: 30). Second, a con-ceptualization of space as closed and thusawaiting the enlivening effects of temporalityfor change or anything new to take place.Instead then of thinking space as the verycondition of and for radical contemporaneity,that is the sphere of co-existing multiplicity,space is tied to the chain stasis/closure.

Alternatives

It is because of the promise of space, that iswhat it could offer us or may give us, thatMassey critiques the unpromising associa-tions that, firstly, casts space as separate fromtime and then, secondly, devalues space bymaking it the negative opposite of time. Inother words, her engagement with theorists,

and schools of thought, is animated by abelief that imagining an alternative under-standing of space is a pressing intellectual taskbecause it is simultaneously a means ofresponding to spatial politics. This task istherefore not only to critique taken-for-granted uses of space but to offer alternativeconceptualizations that could help the diffi-cult work of building alternatives to various‘power-geometries’ – including neoliberalglobalizations. Massey’s positive alternativeconceptualization of space can be placed inthe context of a range of diverse engage-ments that think space and place in terms ofrelationality (i.e. where relations, types ofconnection or association between entities,precede identity). Such a move resonateswith a set of trajectories in human geographythat no longer conceptualize space as a ‘con-tainer’ in which other entities or processeshappen. Instead, any space or place, from theintimate space of a body to the space of theglobe, are precarious achievements made upof relations between multiple entities. Spaceshave to, in other words, be made and remadebecause relations are processual. A namedspace, such as London or Newcastle, does nothave a permanent essence.

Relational thought takes a number ofquite different forms in Human Geography.Harvey (1996), in advocating a type ofdialectical materialism influenced by a longlineage of process thinking, argues that spaceis made by (biological, physical, social, cul-tural) processes and that these processes arethemselves constituted by relations betweenvery different kinds of entities. Thrift (1996),advocating a ‘modest’ style of theory that heterms non-representational, conceptualizesspace as a site of becoming that has to beconstantly performed in and through numer-ous everyday practices. There is much thatHarvey and Thrift disagree on, but whatenables them both to be cast, like Massey, asrelational thinkers is that discrete spaces and

228ÿÿBEN ANDERSON


places are permanencies that are only everprovisionally stabilized because of the multi-tude of entities in relation that they areconstituted from.

For Space is perhaps the most detailed state-ment of an approach cast in terms of relations,and relationality, so it is important to pause andunpack in more detail the three propositionsthat make up the core features of Massey’salternative approach. First, in concert with theclaims of relational thought, Massey (2005:107) argues that space is constituted through itsrelations. Outside of these relations a space hasno existence. There is no difference herebetween spaces we would, ordinarily, considerto be ‘big’ or ‘small’.All are products of relationsbetween all manner of heterogeneous bits andpieces (that are simultaneously natural, social,political, economic and cultural). Space is thusa sphere ‘of dynamic simultaneity, constantlydisconnected by new arrivals, constantly wait-ing to be determined (and therefore alwaysundetermined) by the construction of newrelations. It is always being made and is alwaystherefore, in a sense, unfinished (except that“finishing” is not on the agenda)’. This meansthat, secondarily, space is the sphere of multi-plicity because it is made out of numerousheterogeneous entities. Space is the gatheringtogether of multiple openended, intercon-nected, trajectories to produce what Massey(2005: 111) terms that ‘sometimes happen-stance, sometimes not – arrangement-in-relation-to each-other’. This multiplicity meansthat space is the condition for the unexpected.Third, and consequently, space is an ongoingachievement that is never finished or closed.Stabilities and permanencies, a place thatappears unchanging, for example, are provisionalachievements that have to be constantly madeand remade (even if this process of making andremaking is hidden or taken-for-granted).

An example that Massey uses that exempli-fies how these three propositions functiontogether to disclose space differently is an

example of a train journey from London toMilton Keynes. In a journey you are not sim-ply travelling through space or in space (thatis from one named place – London – toanother – Milton Keynes). This would makespace into a simple container within whichother things only happen. Instead youminutely alter it – if only a little bit by virtueof your presence in one place and your absencefrom the other place – and thus contribute toits being made. Yet as space is altered – by youractive material practices – the places are them-selves constantly moving on and changing asthey are constituted out of processes thatexceed you:

At either end of your journey, then, a townor city (a place) which itself consists of abundle of trajectories. And likewise with theplaces in between. You are, on that train,travelling not across space-as-a-surface (thiswould be the landscape – and anyway whatto humans may be a surface is not so to therain and may not be so either to a millionmicro-bugs which eave their way throughit – this ‘surface’ is a specific relationalproduction), you are travelling across trajec-tories. That tree which blows now in thewind out there beyond the train windowwas once an acorn on another tree, will oneday hence be gone. That field of yellowoil-seed flower, product of fertiliser andEuropean subsidy, is a moment – significantbut passing – in a chain of industrialisedagricultural production. (Massey, 2005: 119)

Human geography and arelational politics

From this evocative image of spaces emerg-ing, and passing away, during a train journeywe get a sense of the delight, or perhaps evenwonder or joy (see Bennett, 2001), thatMassey fosters as she carefully composes heralternative conceptualization of space and



place as relational and thus fundamentallyopen. Another example she returns to is theplace of Keswick – a town in the LakeDistrict, UK – a town that is bound to theromance of the timelessness of the hills, apre-given collective identity (based on a typeof farming) and now modern practices oftourism. Using the case of a visit to Keswickby her and her sister, Massey argues that whatis special about this place, and all others, is its‘throwntogetherness’ – the way that verydiverse elements that cross categories such asthe natural or social come together to fostera particular ‘here and now’. This is whatmakes places specific – this gathering ofdiverse entities into relation:

This is the event of place. It is not justthat old industries will die, that new onesmay take their place. Not just that Hillfarmers round here may one day aban-don their long struggle, nor that thatlovely old greengrocers is now all turnedinto a boutique selling tourist bric-a-brac. Nor, evidently, that my sister and Iand a hundred other tourists soon mustleave. It is also that the hills are rising, thelandscape is being eroded and deposited;the climate is shifting; the very rocksthemselves continue to move on. Theelements of this ‘place’ will be, at differ-ent times and speeds, again dispersed.(Massey, 2005: 140/141)

In the example of Keswick as a particularplace, and of the train journey as a type ofmovement, we see how the three propositionsfoster a shift in how we think about andencounter space – a shift announced in aproposition that Marcus Doel (2000) makes:echoing Massey and drawing on a range ofpoststructuralist thought he argues that ‘itwould be better to approach space as a verbrather than a noun. To space – that’s all.Spacing is an action, an event, a way of being’.

For Space can, therefore, be read as attemptto think space as a verb – a move that ties

space to a set of problematics that have beenseen as the provenance of time. How to thinkthrough the emergence of new spaces andplaces? How to live with difference withinspaces and places? How to engage with theinterconnections that tie together what wemay consider to be ‘separate’ spaces andplaces? Space becomes, therefore, the veryground of the political because to think spa-tially is to engage with the existence ofmultiple processes of coexistence. That is, itopens up a type of relational politics based onthe ‘the negotiation of relations, configura-tions’ (Massey, 2005: 147).

What is at stake is how politics makes adifference from within ‘the constant and con-flictual process of the constitution of thesocial, both human and nonhuman’ (Massey,2005: 147). How would a relational politicsdisclose and intervene in the constellation oftrajectories that produce particular places orspaces? Massey offers three practices that fol-low from opening up the political to thespatial – that is to ‘the challenge of our con-stitutive interrelatedness’ (Massey, 2005: 195).First, a politics of receptivity that is open tothe ‘throwntogetherness’ of place – the waythat a place is ‘elusive’ because it is made outof multiple trajectories. Thus a politics ofplace would not be simply a politics of ‘com-munity’ but would involve processes of‘negotiation’ that would confront the fact ofdifference via ‘the range of means throughwhich accommodation, anyway always provi-sional, may be reached or not’ (Massey, 2005:154). The key, though, is that there are noportable rules because of the uniqueness ofplace: ‘the negotiation will always be aninvention; there will be need for judgement,learning, improvisation’ (Massey 2005: 162).Second, and following on, there can be rulesof space and place that cosily determine apolitical position, i.e. no spatial principlesfrom which a position is simply deduced.Take, for example, arguments about the‘openness’ of particular spaces. These are

230ÿÿBEN ANDERSON


frequently fraught with contradiction. Sothose on the right of the political spectrummay argue for the free movement of capitalbut against the free movement of labour,whilst those on the left may argue for the freemovement of people but against unbridledfree trade. As Massey (2005: 166) stresses‘abstract spatial form, as simply a topographicspatial category, in this instance openness/closure, cannot be mobilised as a universaltopography distinguishing left and right’. Thekey instead is to think through the relationsthrough which the spaces, and thus differenttypes of openness and closure, are con-structed without privileging a-priori eitheropenness, movement and flight, or closure,stasis and immobility. Openness is not thesame in the case of the free movement ofcapital as it is in the free movement of peo-ple. Third, if a relational politics requires bothnegotiations due to ‘throwntogetherness’ anda politics of the terms of openness and clo-sure, it also requires a politics of connectivitythat takes account of wider spatialities of rela-tions. The fact of connectivity raises a host ofdifficult questions about responsibilities thatit is the task of a spatial politics to open up:

It questions any politics which assumesthat ‘locals’ take all decisions pertaining toa particular area, since the effects of deci-sions would likewise exceed the geographyof that area; it questions the predomi-nance of territorially based democracy ina relational world; it challenges an all-too-easy politics which sets ‘good’ localownership automatically against ‘bad’external control (Amin, 2004). It raises theissues of what might be called the respon-sibilities of the local: what, for instance,might be the politics and responsibilitiestowards the wider planet of a world citysuch as London? (Massey, 2005: 181)

To finish with a set of open questionsis therefore appropriate because what ispromised by a relational politics is an expansion

of the problems that animate ‘the political’.This is an expansion that is energized not bythe laying out of a set of invariant principlesbut by the gradual emergence of a distinctivestyle or ethos of engagement with the world:an ethos that strives to be attentive to theconsequences of our varied interrelatedness.It therefore resonates with other currentattempts to foster geographical imaginationsthat engage the world differently in andthrough relational imaginations of space.Whatmore (2002), animated by a range ofnon-representational theories, argues for anethos of generosity that would enable us tounderstand the complex entanglements thatfold humans and non-humans into specific‘hybrid’ geographies. Gibson-Graham (2006),carefully sketching a post-capitalist politics,offer a hopeful stance that would disclose therelations that foster spaces of hope in order todisrupt the mastery of neoliberal capitalism.By resonating with these and other shifts ingeographic thought and practice, Massey(2005) offers a means of thinking through pol-itics of interrelations that is sensitive toheterogeneity of space and thus the genuineopenness of the future, i.e. the very conditionof the political.

Such an ethos of engagement with theworld emerges from a positive understandingof space based simply on ‘a commitment tothat radical contemporaneity which is thecondition of, and the condition for, spatiality’(Massey, 2005: 15). It therefore achieves twoeffects. On the one hand the relational alter-native disrupts many of the taken-for-granted understandings about the relationbetween space and time that have a hold overthe popular and political imagination and arealso still played out by theorists that geogra-phers are otherwise happy to encounter.Massey discloses an evasion of space and issensitive to the ideological and hegemonicwork that an association between space andthe closed, immobile and fixed does. On theother hand, a relational approach to space



fosters the emergence of a new set of ques-tions that force us to wonder again aboutthe task of spatial thought. Massey constantlydiscloses how thinking space fosters a com-mitment to radical contemporaneity. Thesetwo effects combine to open up the politicalto the challenge of space and thus disclose ahost of new political questions and problemsand therefore, perhaps, the faint outline of ageography based on practices of relationality,a recognition of implication and a modestyof judgement.

Conclusion

For Space exemplifies what a relationalapproach to theorizing space and place bothoffers and promises the ethos and politics ofcontemporary human geography. There are,therefore, a set of questions about relations andrelationality that are emerging in human geog-raphy that may become central to how ForSpace is critiqued, evaluated and incorporatedinto the geographical imagination.

• On the one hand how do we understandthe term ‘relation’ given that there aremany forms of ‘elation’ (such as encounter,belonging, etc.). On the other, how do we

understand relations of non-connection –what we could term ‘non-relations’?

• How to understand the durability of par-ticular places or spaces? How do certainconstellations of relations repeat andendure? Alternatively, how to disclosethose space times that flicker out of exis-tence or those space times that nevercame to be?

• How to understand differences in spacesbased on size, i.e. how to theorize scalefrom a relational and thus non-Euclideanperspective?

• How to engage in differences in degreeand in kind within and between the enti-ties that make and are made by relationalspaces, i.e. how are the capacities to act ofa human different to the capacities to actof a non-human?

• How to engage with radical alterity fromwithin a system of relational thought. Thatis how to engage with relations that remainunknowable, undecided or indeterminate?

• How to engage with other types of spacesthat Human Geography is only beginningto encounter – such as spaces constitutedthrough the circulation of images or spacesanimated by the distribution of affect – orthe multiple topological forms that rela-tional space can take (network spaces,Euclidean spaces, fluid spaces, etc.)?

232ÿÿBEN ANDERSON

Secondary sources and referencesAmin, A. and Thrift, N. (2002) Cities: Re-imagining the Urban. London: Polity Press.Bennett, J. (2001) The Enchantment of Modern Life: Attachments, Crossings, and Ethics.

Princeton and Oxford: Princeton University Press.Crang, M. and Thrift, N. (2000) Thinking Space. London: Routledge. Doel, M. (1999) Poststructuralist Geographies: The Diabolical Art of Spatial Science.

Edinburgh: Edinburgh University Press.Doel, M. (2000) ‘Un-Glunking geography. Spatial science after Dr Seuss and Gilles

Deieuze’, in M. Crang and N. Thrift (eds) Thinking Space. London and New York:Routledge, pp. 117–135.

Gibson-Graham, J.-K. (2006) A Postcapitalist Politics. Minnesota: University of MinnesotaPress.

Harvey, D. (1996) Justice, Nature and the Geography of Difference. Oxford: Blackwell. Massey, D. (1984) Spatial Divisions of Labour. London: Macmillan.


Massey, D. (1994) Space, Place and Gender. Cambridge: Polity Press. Massey, D. (2005) For Space. London: Sage.Murdoch, J. (2006) Post-Structuralist Geography. London: Sage.Soja, E. (1989) Postmodern Geographics: The Reassertion of Space in Critical Social

Theory. London: Verso.Thrift, N. (1996) Spatial Formations. London: Sage.Whatmore, S. (2002) Hybrid Geographies. London: Sage.



Index

actor-network theory 79, 133agency, human 148, 173–174agency, of nature 78–79Amin, A. 39, 41, 85, 89, 215–224, 226,

231, 233animals 72, 184–186, 208–212Antipode (journal) 68, 126, 132, 201,

211–212architecture 99, 104, 142

Barnes, T. xv, xx, 17, 23, 25, 31, 32, 34,41, 51, 84, 88, 89, 129, 134, 160,197–206

Behavioural geography 83, 117, 146, 184Berkeley School 101Berkeley, University of 24Berry, B. 11, 18–19, 21, 215body see embodimentBourdieu, P. 126, 202Bretton Woods 128Bunge, W. 9–16, 18, 19, 23, 26, 32, 155Buttimer, A. 43, 57

Californian School 148 see alsoLA School

Cambridge, University of 18, 34,173–274

Canada 14, 102, 117, 174, 197, 199capitalism 61–70, 75, 76, 79, 83, 101,

105, 106, 109–114, 127, 128, 129,130, 133, 141, 145–152, 202, 212,215, 231

cartography 10, 154, 192, 226central place theory 10–11, 19, 28, 215Chicago, University of 4, 17, 183children 95, 180, 182, 185, 193Christaller, W. 18–19, 117, 155 Chorley, R. 11, 18, 25–27, 30clusters, economic 147, 149cognitive 44, 56, 184 consumption 64, 66, 67, 128, 151corporations 117, 118, 218Corbridge, S. 78, 80, 109–116Cosgrove, D. 91–108, 130, 134, 157, 161Cox, K. 9, 10, 11, 15, 33–42, 68, 148, 152critical realism 135, 148, 198, 203cultural geography 99, 101, 102, 106, 129,

130, 131, 165, 167, 180, 185, 186cultural turn 84, 130, 137, 138, 149,

151, 189, 192, 194

culture 1, 2, 43, 44, 46, 48, 57, 58, 95,100, 101, 102, 104, 106, 126, 130,132, 151, 165, 166, 172, 175, 182,184, 190, 193, 194, 204, 228

CURS initiative 81

deconstruction 129, 157, 177, 191,199, 228

Deleuze, G. 217, 219, 221demography 1, 2, 40, 110, 112Dicken, P. 117–124diffusion 1–7, 13, 18, 19, 114disability 181discourse 96, 120, 127, 130, 132, 133,

151, 157, 158, 159, 166, 167, 171,193, 194, 203, 210, 212

economic geography 1, 23, 117, 130,150, 151, 160, 197–205

Economy 34, 37, 38, 39, 40, 54, 61, 63,66, 67, 69, 72, 83–89, 96, 110,111, 112, 114, 119, 127, 128,146–151, 194, 197–205

embodiment 47, 55, 56, 73, 139, 164,166, 193, 212, 219

Enlightenment 127, 156, 197, 203, 209ethics xiii, 57, 125, 126, 129, 194, 210,

211, 212, 213ethnography 58, 140, 175, 194, 198,

200, 204, 205externalities 34–36

feminism 49, 57, 62, 91–97, 101, 105,129, 130, 132, 140, 141, 158, 160,165–170, 173, 174, 183, 185, 193,198, 202, 203

gender 40, 44, 64, 74, 78, 83, 84, 91–97,102, 105, 126, 130, 132, 133, 141,163–168, 173, 174, 176, 179, 185,202, 203, 205

gentrification 37, 71, 76, 77, 78, 130geographic information systems 6, 13,

22, 23, 30geographical imagination 95, 125, 157,

167, 171–176, 232geopolitics 77, 189–195Germany 12, 21Giddens, A. 136, 142, 181globalization 40, 41, 76, 77, 112,

117–122, 150, 151governance 37, 114

Gregory, D. 31, 32, 87, 89, 136, 137,139, 140, 141, 147, 152, 171–178,181, 198, 202, 206

Hägerstrand, T.1–8, 18, 19, 24, 166 Haggett, P. 11, 15, 17–24, 25, 26, 29,

30, 32, 53, 58, 62, 135 Haraway, D. 165, 193, 202Harvey, D. 21, 22, 24, 25–32, 33, 39,

41, 57, 58, 61–70, 71, 78, 80, 86,111, 115, 125–134, 135, 136,137, 138, 141, 143, 146, 148, 152,171, 172, 173, 174, 177, 201, 215,223, 228, 232

Harvard University 5, 154health 35, 46, 93hegemony 126, 128, 167, 180historical materialism, 109, 131–133,

136–137, 141, 171, 174–175 seealso Marxism, structuralism

homeless 34, 184humanistic geography 43, 53, 54,

55, 57, 92, 103, 105, 129,138, 165

hybridity 207–213

iconography 99, 102 identity 43, 45, 46, 47, 57, 78, 94, 102,

106, 129, 166, 174, 192, 195, 230ideology 63, 97, 99, 105, 109, 110,

111, 139, 171, 190Innis, H. 199, 201Iragaray, L. 165

Kant, I. 2, 72, 155Klein, M. 81–183knowledge production xiii, xv, xvi,

xix, 22, 25, 26, 27, 29, 54, 78, 92,149, 153, 155, 156, 158, 159, 164,165, 166, 171, 172, 176, 183, 186,189, 191, 197, 198, 201, 203, 205,209, 210

LA School 149Lacan, J. 183, 185Lake District, UK 230Le Doeuff, M. 165Los Angeles 141–142, 148 landscape 63, 74, 100–106, 130, 160,

173, 229landscape art 72, 100, 102, 104

Hubbard-Index:Hubbard-Index.qxd 4/17/2008 5:51 PM Page 235

language barriers xivscientific 28, 54, 200gendered 73, 164national 192representation 182, 200

Lefebvre, H. 57–58, 135, 138, 142, 174Livingstone, D. 153–162locality studies 83–84, 198, 202locational analysis 11, 17–23, 117 London 11, 85, 228, 231London, University of 163 (Queen

Mary and Westfield College), 200(University College)

Lukermann, H. 199–200Lund University 1–2, 5, 18

machines, city as 221–222Mackinder, H. 189–190,Magritte, R. 225 Manchester, University of 137Marble, D. 11, 18Marxism 30, 39, 57, 61–70, 78–80,

100–102, 109, 112, 126–127, 129,131, 136–142, 174, 184, 198,201–203

masculinity 92, 96–97, 105, 131, 132,140–141, 158, 163–167, 193

Massey, D. 47, 48, 49, 50, 51, 57,83–90, 121, 123, 131, 132, 134,138, 141, 146, 148, 150, 152, 165,169, 202, 225–235

materiality 104, 136–137, 141, 209–212,218–222

migration 2–4, 40, 62, 93, mobility 48, 50, 56, 74, 228modelling 1–6, 11, 13, 17–23, 26–28,

146, 198–200modernity 101, 125, 127–128, 130,

209, 227money 65–69, 74morality 77, 112, 158, 212moral panic 180Morrill, R. 5

nationalism 77nation-state 77, 79, 106, 119, 120neo-classical economics 39–40, 83, 117,

145–146, 149, 198–199, 201, 204New York 33 non-place 215 see also placelessnessnon-representational theory 55, 131,

221, 228, 231

objectivity, in geographical analysisxiv, 4, 10, 27, 62, 153–154, 166,185, 192, 230

Ó’Tuathail, G. 189–196otherness 183Oxford, University of 83Oxfordshire 99

paradigm 10, 61–62, 78, 155, 157,203, 205

Paris 62, 126, 130, 137

patriarchy 94–95, 131 performativity 248phenomenology 44–49, 172physical geography xvi, 10, 47, 83, 159,

172, 213place xiii, 10, 43–50, 53–58, 86–87, 122,

158, 165, 167, 180, 186, 228–230placelessness 43–50 see also non-placeplanning 92political economy 71, 127, 129, 131,

145–149, 171–175, 201positionality xv, 127, 129, 179, 205positivism 14, 29–30, 73, 83, 127,

137–138, 154, 172postcolonial theory 101, 105–106, 114,

141, 158, 160, 165, 173–174,203–204

post-Fordism 121, 128–129, 139–140,148, 150

postmodern epoch 49–50postmodernity 125–133, 135–142,

157, 159, 174, 185, 216poststructuralism 46, 49, 50, 57, 92, 96,

131–133, 151, 165, 173, 189, 193,201–203, 226, 228

power of labour 65of landscape 101–103of the local state 35–41, 220of the nation-state 79, 113, 189–191of place 50relations 90–91, 100, 132, 159,

199, 201Pred, A. 1, 4, 8, 173, 181Probyn, E. 165probability theory 3, 28production

of commodities 65–67, 109, 112,118–122, 128–130, 146–151,198–199, 201, 229

of knowledge 27, 54, 96, 164, 198,201, 206

of nature 71–79of space 57, 77, 171–173, 184, 191

psychoanalysis 166, 184–185

quantification 11–12, 14, 21, 26, 149, 199quantitative revolution 9, 11, 23, 25,

53, 86, 117, 198–200, 215queer theory 129, 165

race 60, 102, 166, 176, 184reflexivity 166, 199, 205regional geography 1, 10, 85Relph, E. 43–52, 53, 59Representation 100–105, 127, 129, 131,

141, 173, 182–185, 191–192, 202resistance 143, 194Royal Geographical Society 117, 154,

189, 192, 211 Rose, G. 49, 51, 57, 59, 103, 105, 108,

136, 141, 143, 163–170

Said, E. 173, 202scale 20, 36–38, 49, 71–79, 118, 120,

180, 182, 232

sexuality 96–97, 166, 179, 184Sibley, D. 179–188Social Area Studies 22social construction 96, 131, 155social movements 141, 193socio-spatial dialectic 136Soja, E. 58, 59, 135–144, 171, 173, 177,

185, 190, 196, 215, 223, 227, 233space

absolute 65, 76–77, 79paradoxical 167–168production of 57, 77, 171–173,

184, 191relative 65, 76–77, 79

space-time 14, 31 see also time-spacespatial science xvii, 17, 34, 40, 54, 62,

147, 202–204spatial divisions of labour 77, 86,

120, 226Sraffa, P. 199, 203statistics 13, 15, 20, 22, 28, 199Storper, M. 88, 90, 145–152, 201structuration theory 87, 148, 181subjectivity 78, 155, 164–165, 181,

183, 212, 217 Sweden 1, 18, 26, 117

Taylor, P. 21, 159technology 146, 149, 150–151,

208, 211territory 190textbooks xii-xiv, 25, 29–30, 120, 166Third World 75, 110, 112, 119, 131Thrift, N. 39, 42, 55, 56, 59, 69, 70,

86, 87, 88, 89, 118, 121, 122,130, 135, 161, 184, 187, 193,215–224, 226, 228

time 4, 7, 12–13, 56, 113, 128, 136,156, 165, 198, 204, 217, 227–228

time-space 127, 131–132, 232 see alsospace-time

Tobler, W. 11, 18transnationalism 40transnational corporations (TNCs)

118, 218travel time 12Tuan, Y-F. 43, 51, 53–60, 129, 134

United Kingdom 94–95, 180, 184, United States 128, 191 urban ecology 219urban geography 33, 38, 41, 215–222urbanism 13, 215uneven development 71–78, 84, 86–87,

109–111, 120, 149utility theory 34–35

Vancouver 173

Walker, R. 61, 68, 145–152, 201Washington, University of 15Whatmore, S. 207–214, 226, 231, 233Wilson, A. 12workplace 146, 205

Young, I.M. 165

236ÿÿINDEX

Hubbard-Index:Hubbard-Index.qxd 4/17/2008 5:51 PM Page 236

key texts in human geography -...

Documents