after the rights revolution final print - sujit choudhry · important changes in the constitutional...

LS06CH18-Choudhry ARI 28 July 2010 15:19

RE V I E W

S

IN

A D V A

NC

E

After the Rights Revolution:Bills of Rights in thePostconflict StateSujit ChoudhryFaculty of Law, University of Toronto, Toronto, Ontario M5S 2C5, Canada;email: [email protected]

Annu. Rev. Law Soc. Sci. 2010. 6:18.1–18.22

The Annual Review of Law and Social Science isonline at lawsocsci.annualreviews.org

This article’s doi:10.1146/annurev.lawsocsci.093008.131445

Copyright c⃝ 2010 by Annual Reviews.All rights reserved

1550-3585/10/1201-0001$20.00

Key Wordsconstitutional design, comparative constitutional law, comparativepolitics, civil wars

AbstractBills of rights are now central components of liberal democratic con-stitutions. But debates over the character and content of bills of rightsare no longer at the center of more recent rounds of postconflict con-stitutional politics. This review puzzles through the rise and decline,but persistence, of rights-based constitutionalism. Neither comparativeconstitutional law nor constitutional politics offers the answer. The lit-erature on civil war settlement suggests that bills of rights serve twofunctions in postconflict constitutions: a regulative role to check theabuse of public power and a constitutive role to serve as the basis of anew constitutional identity. Bills of rights cannot do the work that isexpected of them. Politicized judiciaries, constitutional underenforce-ment, and the ex post nature of judicial review undermine the abilityof the bill of rights to serve as a credible commitment against futureabuses of human rights. Moreover, the idea of a bill of rights as a sourceof shared political identity abstracted from a contingent political andhistorical context is unlikely to succeed in practice.

18.1


THE PARADOX OF BILLSOF RIGHTSAt a public lecture in January 2009, PeterGalbraith—most recently the UN Secretary-General’s Deputy Special Representative forAfghanistan and a long-time advisor to theKurdish regional government in Iraq’s north—suggested that the prospects of Iraq survivingas a single state remained dim. According toGalbraith’s argument, the very design of theIraqi constitution—its highly decentralizedfederalism, the potential for governorates(provinces) to combine into larger regionalauthorities, the extensive provincial controlover oil reserves—provided a road map toIraq’s eventual partition (Galbraith 2008). Inpassing, Galbraith shared his firsthand accountof the negotiations surrounding Iraq’s 2005Constitution. The issues that had generatedthe most controversy were the characterand scope of Iraq’s federal structure and theorganization of executive and legislative powerat the center. By contrast, the negotiatingparties agreed to the text of Iraq’s bill of rightswith a minimum of debate. Although thisis an oversimplification—the precise role ofIslam and religious personal law in the Iraqiconstitutional order, for example, intersectedin complex ways with the bill of rights,producing lengthy negotiations over how therelationship between these different parts of theconstitution would operate (Deeks & Burton2007)—it appears to be an accurate account ofone of the most recent and closely observedinstances of postconflict constitutional design(see also Bouillon et al. 2007, Dawisha 2009,O’Leary et al. 2005, O’Leary 2009). Indeed,the July 2009 report of the Iraq ConstitutionalReview Commission, which consists entirelyof draft amendments to the 2005 Constitution,continues this pattern, addressing thornyissues such as the allocation of oil revenuesand the status of Kirkuk while proposinghardly any changes to the Bill of Rights (IraqConstitutional Review Commission 2009).

Now compare South Africa. In the early1990s, as South Africa underwent its negotiated

transition from apartheid to liberal democracy,the constitutional agenda was broad. But the de-sign of South Africa’s bill of rights was an issueat the very center of the country’s constitutionalpolitics (Spitz & Chaskalson 2000). Although itwas accepted that, given South Africa’s back-drop of racial injustice, a bill of rights was un-avoidable, there was considerable disagreementover its details. These issues included, inter alia,the application of the bill of rights to private ac-tion (horizontality), a general limitations clauseand proportionality analysis, the inclusion ofsocioeconomic rights, the exact nature of theproperty rights clause, the scope of freedom ofassociation to encompass rights for organizedlabor, and the choice of a specialist constitu-tional court to wield ultimate authority to en-force the constitution. This process was learnedand highly comparative, drawing on the expe-rience of rights-based constitutionalism froma standard set of cases that serve as compara-tive models of lessons learned and dangers tobe avoided: Canada, Israel, Germany, India,and the United States. Indeed, in many ways,the South African debate shaped the academicagenda of comparative constitutional law andpolitics in the years to come.

The jarring contrast between South Africaand Iraq—constitutional transitions separatedby a little over a decade—illustrates a paradox atthe heart of postconflict constitutional design.On the one hand, bills of rights are now cen-tral components of liberal democratic consti-tutions. This reflects a fundamental shift in thedesign of democracy. As Huntington (1991) hasargued, democratization has occurred in threewaves: The first commenced in the 1820s inthe United States and ended in 1926; the sec-ond ran between 1943 just prior to the AlliedVictory in Europe and proceeded through thepostwar period with decolonization in 1964; thethird wave began in 1974 with the overthrow ofPortugal’s dictatorship, continued with the endof military dictatorships in Spain, Greece, andLatin America, reached the communist coun-tries of Eastern and Central Europe in the1990s as well as South Africa, and may or maynot be over. Huntington does not describe the

18.2 Choudhry


important changes in the constitutional pack-age associated with democratization from waveto wave. In the first wave, this package wouldhave consisted of competitive, multiparty elec-tions for a legislature, a politically accountablehead of the executive that is either directly orindirectly elected, an independent bureaucracy,independent courts, a separation of the partyand the state, etc. In the second and, decisively,in the third wave, this constitutional packagecame to include rights-based constitutionalism.A bill of rights that is entrenched and supremeover legislative and executive action, backed upby judicial review by independent courts, is nowwhat we associate with a normal state, and theonus of justification has shifted to those wish-ing to omit these arrangements from any newconstitution. This is the “Rights Revolution”(Ignatieff 2000). In the postconflict state, theadoption of constitutions that entrench civil andpolitical rights has become part of the standardscript of peacebuilding, along with the develop-ment of competitive politics and open markets(Paris 2004).

But on the other hand, Iraq demonstratesthat debates over the character and content ofbills of rights are no longer at the center ofmore recent rounds of postconflict constitu-tional politics. Bell’s (2008) overview of peaceagreements—which often provide the substan-tive framework for postconflict constitutionalsettlements—confirms this view. In exam-ples as wide-ranging as Afghanistan, NorthernIreland, Bosnia Herzegovina, Kosovo, Cyprus,Iraq, and Sri Lanka, actual and proposed peaceagreements tend to combine two strategies: thedisaggregation and the dislocation of power.Disaggregation involves the fragmentation ofpower at the center through power-sharing de-vices associated with the consociational modelchampioned by Lijphart (1977), whereas dis-location encompasses significant territorial de-volution and an acknowledgment of the multi-national character of the polity, and may evenopen up the notion of sovereignty itself, con-stitutionalizing a role for kin states of nationalminorities, as has occurred in Northern Ireland.It is these issues, not bills of rights, that are at

the heart of contemporary postconflict consti-tutional politics.

I want to puzzle through the rise anddecline (but persistence) of rights-basedconstitutionalism—hence my title, “After theRights Revolution.” Methodologically, I do sothrough an engagement with various distinctbodies of literature: comparative constitutionallaw, comparative constitutional politics, civilwar studies, political theory, and ultimately,constitutional design for divided societies. Mytentative conclusion is that constitutional ac-tors have come to understand that rights-basedconstitutionalism cannot do all the work it hasbeen expected to do.

THE CRITICAL LITERATUREWhat changed between the South African andIraqi transitions? The Rights Revolution hasbeen the central preoccupation of the closelyallied fields of comparative constitutional lawand comparative constitutional politics, so itis to them that we should first turn for in-sight. Explaining this apparent shift in the prac-tice of contemporary constitutionalism poses achallenge for both fields because they do notaddress it.

Comparative Constitutional LawThe literature on comparative constitutionallaw is large, growing, and increasingly intellec-tually ambitious. The bulk of the literature oncomparative constitutional law has been preoc-cupied with a set of discrete questions that havealso figured centrally in the constitutional pol-itics of rights-based constitutionalism:

! Should constitutions contain a justiciablebill of rights, or does a bill of rights trans-fer too much power to the judiciary andjudicialize politics? Although this debatefirst emerged in the United States (Bickel1962), it is now thoroughly comparativein character. Dworkin’s (1996) defensesof rights-based constitutionalism, for ex-ample, are both a theory of constitutional

www.annualreviews.org • Bills of Rights in Postconflict States 18.3


interpretation applicable to the AmericanBill of Rights and a general theory of lib-eral constitutionalism that supports a billof rights combined with judicial reviewas the best instrument to institutionalizethe liberal commitment to treating per-sons with equal concern and respect inany jurisdiction. In the legal literature,Waldron (1999) has offered the most ful-some response to Dworkin, challengingthe instrumental argument that rights arebetter protected by rights-based constitu-tionalism than by democratic legislaturesand that judicial review, regardless of itsconsequences, is democratically illegiti-mate because it undermines the right toequal democratic participation.

! What should the institutional arrange-ments surrounding the enforcement ofa bill of rights be? Debates surroundingthe entrenchment of rights-based consti-tutionalism often suppose that there is nomiddle ground between legislative and ju-dicial supremacy. However, the optionson the table are actually broader. For ex-ample, there may be a distinct “Com-monwealth model of constitutionalism”that combines legislative supremacy withweak judicial review, an arrangement thatenjoys advantages over both legislativeand judicial supremacy because it vestsfinal decisions with elected legislatureswhile improving their decisions by in-jecting matters of principle into legisla-tive debates (Gardbaum 2001). But doesweak-form judicial review tend to de-generate either into strong-form judicialreview because political majorities willrarely reverse a court judgment or intolegislative supremacy because they willroutinely do so (Tushnet 2008)? Do weaksystems of judicial review create an in-terinstitutional dialogue between courts,executives, and legislatures over ques-tions of rights protection, which operatesdifferently than the monologue of judi-cial supremacy, and make rights review

a joint responsibility of these institutions(Hiebert 2005, 2006a,b; Roach 2001)?

! Should ultimate responsibility for consti-tutional adjudication be diffused through-out the judicial system, with appeals toa generalist Supreme Court (as in theUnited States and other common law ju-risdictions), or should it be centralizedwithin a specialist constitutional court,which has emerged as the model ofchoice (Comella 2004, Garlicki 2007)?The claimed virtues of the centralizedmodel are legal certainty, the facilitationof constitutional transitions in contextswhere the previous judiciary was com-plicit in the abuses of the old regime,democracy (because it acknowledges thespecial political character of judicial re-view), and efficacy (in civil law systemswhere conceptions of the judicial roleinhibit the second-guessing of legisla-tures). There is now a serious debate,however, over how centralized a con-stitutional court system can remain inpractice, because, for example, ordinarycourts must engage in constitutional in-terpretation in order to decide when aconstitutional issue arises.

! Should a bill of rights contain merelythe traditional, first-generation civil andpolitical rights? Or should second-generation social and economic rights tohousing, health care, and social assistancebe included as well? What are the justi-fications for entrenching social and eco-nomic rights? Are socioeconomic rightsinstrumentally valuable because they areintegral to ensuring the equal enjoymentof civil and political rights, or are theyvaluable in their own right because ofthe important interests that they protect(Bilchitz 2007, Fredman 2008, Tushnet2008)? Does the entrenchment of socialand economic rights pose the twin dan-gers of political demobilization and re-gressive judicial interpretation, makingpolitics a more secure basis for the pro-tection of these interests (Bakan 1997)?

18.4 Choudhry


Other scholars raise questions of insti-tutional competence, arguing that socialand economic rights pose challenges forjudicial enforcement because they involvequestions of resource allocation and re-quire ongoing judicial oversight, unlikenegative rights that are allegedly cost freeand immediately satisfiable, and suggestthat an administrative law approach to so-cioeconomic rights meets those concerns(Sunstein 2001).

! Should a bill of rights be drafted to ac-commodate the activities of the regula-tory, redistributive state, and if so, how?Should a bill of rights contain constitu-tional reminders to positive state actionsuch as the Directive Principles of StatePolicy in the Indian constitution, and per-haps instruct courts to defer to constitu-tional challenges to public policies thatimplement these principles? Should a billof rights protect property rights, and ifso, how should it balance the obligationto compensate expropriations against theneed to provide constitutional space forextensive schemes of land redistribution(Alexander 2006, Allen 2000)? Should aconstitutional guarantee of equality per-mit preferential programs on the basisof personal characteristics that are oth-erwise prohibited grounds of discrimina-tion, and if so, should it impose any con-stitutional discipline on these constraints(Galanter 1984)?

! To what institutions and relationshipsshould a bill of rights apply? Should a billof rights merely apply vertically—that is,to government institutions and hence tothe relationship between citizens and thestate? Or should a bill of rights be draftedto apply horizontally to private actors andprivate relationships as well, as is the casein South Africa (Barak 1996, Gardbaum2003)? Is this textual choice really deter-minative (Tushnet 2003)?

! How should a bill of rights be inter-preted? Should courts have recourse tocomparative materials, as positive models

that set out the contours of a shared, uni-versal model of rights protection, or asinterpretive foils to further constitutionalself-understanding, including serving asnegative antimodels of comparative con-stitutional experience—through dialogi-cal interpretation or constitutional en-gagement (Choudhry 1999, 2004, 2006;Jackson 2005)? Are there general princi-ples that should govern the interpretationof a bill of rights? Here, a debate has beenjoined between Antonin Scalia (1998) andAharon Barak (2006), between a textual-ist and purposive interpretation of a billof rights.

This is a vast and rich literature. However, itdoes not answer the question at hand. The fun-damental reason is its prescriptive orientation,providing arguments for and against certainchoices in the design and interpretation of billsof rights. The principal audience outside theacademy consists of constitutional negotiatorsand courts. As Hirschl (2006) has noted, thisbody of work does not attempt to explainthe rise of rights-based constitutionalism.Indeed, scholars are very much part of thisphenomenon, not only through their research,but also as academic consultants in foreignconstitution-drafting processes. The literatureis therefore doubly ill-equipped to respond tothe eclipsing of rights-based constitutionalismin postconflict situations. First, it lacks theanalytical resources to explain this develop-ment. Second, on its own terms, it containsprescriptions that do not always respond to theinterests and constitutional claims of politicalactors, who increasingly do not debate thefiner details of bills of rights.

Comparative Constitutional PoliticsA more promising source of insight is the largebody of work on comparative constitutionalpolitics that has accompanied the Rights Rev-olution, which is self-consciously explanatoryin character. The focus has been on the ex-pansion of judicial power and, indeed, extends



far beyond rights adjudication to encompassthe broader question of the “judicialization ofmega-politics,” which often does not involvebills of rights at all (Hirschl 2008). Nonethe-less, the narrower question of the spread ofjudicial review attached to a bill of rights hasattracted sustained attention and has gener-ated several possible explanations. The mostinsightful explanations have been offered inGinsburg’s (2003) Judicial Review in NewDemocracies: Constitutional Courts in Asian Casesand Hirschl’s (2004) Juristocracy. Their prin-cipal contribution was to reframe the ques-tion away from Bickel’s (1962) inquiry intowhy courts can and would usurp power fromdemocratically elected officials and toward aninquiry into why powerful political actorswould bind themselves through the adoptionof bills of rights backed up by judicial review.Ginsburg’s “insurance model” and Hirschl’s“hegemonic preservation thesis,” although dif-fering in important respects, share the centralargument that political elites have adopted billsof rights to constitutionally entrench their nar-row policy preferences and, more generally, toprovide themselves with the legal resources tochallenge future policy decisions in the eventthat they lost power, for example, through theelectoral process. Thus, rather than fetteringpower, the goal of adopting right-based consti-tutionalism is to preserve it.

Although they do not narrowly address post-conflict constitutional design, both accounts—especially Ginsburg’s—suggest that bills ofrights should be standard components of post-conflict constitutional settlements. The rea-son is that fundamentally, postconflict consti-tutional design involves the redistribution ofpolitical power through strategies of disaggre-gation and dislocation. Parties who anticipatelosing power in the future—whether preexist-ing elites or not—should accordingly demandbills of rights because they offer an additionalset of political resources that are not tied to con-trol of executives and legislatures. Conversely,if other parties anticipate controlling politicalinstitutions, they will resist rights-based con-stitutionalism. Ginsburg, for example, explains

the precise design of constitutional courts as theoutcome of these kinds of political calculations.But it also follows that these conflicting inter-ests should be reflected in competing proposalsfor the substantive provisions of a bill of rights.The apparent lack of friction over the content ofIraq’s bill of rights therefore poses a challengeto this line of analysis.

Recent scholarship in comparative constitu-tional politics offers a potential explanation ofthe diminished controversy over bills of rightsin constitution-drafting processes and of theirpersistence in constitutional design. This is thephenomenon of constitutional convergence. Inthis account, the texts of bills of rights cluster ina narrow band because of the relatively limitednumber of different textual models available forvarious constitutional provisions. This conver-gence occurs with respect to both the rangeof rights included and their specific language.There are only so many ways to draft a provisionon the right to equality, freedom of expression,socioeconomic rights, and limitations, and mostbills of rights contain a similar set of rights. Forthis reason, there is an off-the-shelf quality torecent bills of rights. It follows that given thelimited scope for constitutional choice, thereshould be accordingly less controversy over theterms of a bill of rights.

There are two possible mechanisms for thisasserted constitutional convergence. One isrooted in international law and theories of pre-commitment (Ginsburg 2006). It has been ar-gued, in the European context, that one of theprincipal drivers for accession to the EuropeanConvention on Human Rights in the postwar pe-riod came from fragile democracies that had notyet fully consolidated and that were eager toprecommit themselves to respect human rightsand guard against the risk of a return to non-democratic rule through a mechanism of en-forcement beyond the control of any one state(Moravcsik 2000). This argument can be ex-tended to the design of contemporary constitu-tions. Thus, constitutions are framed to ensurecompliance with international legal obligations,not only by providing efficient mechanismsfor accession to those obligations, but also by

18.6 Choudhry


incorporating international legal obligationsinto constitutions themselves. The goal of con-stitutional entrenchment is to place certainpolicies beyond the reach of domestic politicalactors. Applied to bills of rights, theories of pre-commitment would link domestic schemes ofrights protection with the terms of internationalhuman rights treaties. The most direct methodof constitutional precommitment would be theincorporation and constitutional entrenchmentof international human rights treaties, as oc-curred in Bosnia. Another would be a bill ofrights that closely tracked international humanrights obligations. The relative decline in thecontroversy of rights-based constitutionalismin postconflict constitutional settlement mightsimply reflect the availability of an internationallegal script. Conversely, the ongoing debatesover questions of constitutional structure (fed-eralism, executives, legislatures) may be a func-tion of the absence of international legal normsthat govern many critical issues of constitu-tional design.

Another explanation for the decline in therelative centrality of rights-based constitution-alism to constitutional politics, but their per-sistence in constitutional design, might drawon the literature on policy diffusion (Elkins& Simmons 2005, Goodman & Jinks 2010).Numerous studies across a variety of policyarenas (e.g., educational policy, environmentalpolicy) have demonstrated that states are con-verging on public policies, even in the absenceof binding international obligations that directthem to do so. Indeed, it has been argued thatconvergence is an accelerating process, witheach decision by a state to converge on a pol-icy increasing the probability of future conver-gence by other states. The argument here is thatconvergence is not a function of precommit-ment to international law, but rather is a func-tion of processes of policy learning or accultur-ation. Policy learning arises in conditions wheredomestic constitutional drafters operate undersevere cognitive restraints, because of a lack ofexperience with liberal democracy, for example,and negotiation occurs under tight time framesand under the threat of a return to violence.

Foreign constitutional models that are readilyavailable from jurisdictions with consolidatedliberal democracies are particularly attractivein these situations. Policy acculturation occurswhen states adopt public policies through a pro-cess of socialization and assimilation into thenorms of international society on how statesare expected to behave. The normative valueof a bill of rights and “variations in nationalresources, social histories, and economic devel-opment within states” (Goodman & Jinks 2010)matter less than the fact that rights-based con-stitutionalism has become an integral part ofwhat is expected of a normal state.

A recent study, however, complicates con-siderably the simple narrative of constitutionalconvergence in the realm of bills of rights (Z.Elkins & T. Ginsburg, unpublished). First, arelatively limited set of rights have becomenearly universal, such as the rights to freedom ofexpression, freedom of religion, freedom of as-sembly, freedom of association, and freedom ofmovement, as well as the right to life, the rightagainst torture, and the right against cruel, in-human and degrading treatment. Beyond thatcore, however, there remains considerable vari-ation, with many bills of rights not extend-ing much further and others containing over45 rights. Moreover, the variation in the num-ber of rights contained in bills of rights is in-creasing, not diminishing, over time. Second,although the number of constitutions contain-ing civil and political rights has grown steeply,relatively few guarantee socioeconomic rights.Third, there remains a gap between interna-tional human rights law and domestic bills ofrights. The core of universal rights in domesticbills of rights is far smaller than the approxi-mately 30 rights found in the leading interna-tional human rights treaties [the gap is smallestbetween the International Covenant on Civiland Political Rights (ICCPR) and domesticconstitutions].

Thus, constitutional convergence cannotexplain the declining salience of debates overrights in postconflict constitutional design. In-deed, a simple comparison of the Iraqi andSouth African constitutions confirms this view.



Thus, whereas the South African bill of rightsprovides for horizontal application and has twosets of provisions on socioeconomic rights, amultipronged provision on equality, a rightto just administrative action, labor (i.e., tradeunion) rights, and education rights, the Iraqibill of rights lacks most if not all of these. Thefact that Iraq’s bill of rights was drafted af-ter South Africa’s further underscores how thescope for domestic constitutional choice andpolitical controversy over a bill of rights re-mains considerable.

Civil WarsInstead of proceeding from the bodies of schol-arship most focused on bills of rights to seewhat insights they offer on bills of rights as el-ements of postconflict constitutional design, Istart at the other end—from the research oncivil wars—to see what insights they offer onthe role of bills of rights in post–civil war (i.e.,postconflict) constitutions. As I discuss, this lit-erature does not address bills of rights explicitly,although it is highly suggestive.

Civil wars have generated an enormous bodyof research, especially in the last two decades.In no small part, this is because of their mas-sive human toll. By one count, between 1945and 1999, there were 146 civil wars with a to-tal death toll of more than 20 million, whichtook place in approximately one-third of thecountries in the world. By contrast, over thesame period there were 25 interstate wars inwhich approximately 3 million died. Not onlydid civil wars produce far more casualties, theyare in general longer, end more often in deci-sive victories, and recur in a higher proportionof cases than wars between states (Fearon &Laitin 2003). In addition, the costs of civil warcannot be measured through deaths alone. Civilwars have produced human and economic dev-astation through mass refugee flows, famines,economic collapse, and, underlying all of these,the destruction of the ability of the state to pro-tect its citizens and provide a stable and peacefulframework for economic, social, and politicallife.

This research has focused on the de-terminants of civil war onset and duration.Methodologically, it consists of the construc-tion and analysis of large data sets that combineinformation about civil wars (onset, termi-nation, casualties, and number and identityof combatants) and various characteristics ofstates (per capita income, ethnic heterogeneity,religious diversity, population, presence ofmountainous terrain, oil production, instabil-ity, level of respect for democracy and civilliberties, discrimination, new states, naturalresources, and commodity exports). Thesestudies appear to suggest that civil war isassociated with “poverty, large populations, alow level of economic development, a priorhistory of civil war, and political instability” aswell as the existence of diasporas and presenceof natural resources, among other factors(Walter 2009, p. 244). However, there remainsenormous debate over how to interpret theseresults, which in turn has generated a vastliterature and numerous schools of analysis.Roughly put, the principal theories of civilwar onset are those that emphasize grievance,greed, and fear (Collier & Hoeffler 2004, Kaly-vas 2007, Toft 2010). Grievance-based theoriesattribute civil wars to inequalities in access tonatural resources, land, income, and/or politi-cal representation, denials of civil liberties, anddiscrimination. Greed-based theories link civilwar to struggles over natural resources betweengovernments and rebel groups. Fear-basedtheories posit a link between civil war and statecollapse, which leads to a security dilemmawhereby the absence of the state leads ethnicgroups to arm for reasons of self-defense, whichis misinterpreted as an aggressive measure andfuels a spiral of armament, leading to conflict.Moreover, although grievance, greed, and fearmight provide the motivation for civil war, theyare “too common to distinguish the cases wherecivil war breaks out” and must be supplementedby an account that explains the vulnerability ofa state to insurgency (Fearon & Laitin 2003).

What is striking is that constitutions arelargely missing from the dominant strand of re-search on civil wars. Yet in a fundamental sense,

18.8 Choudhry


civil wars can be understood as constitutionalphenomena. Above all, the mission of consti-tutionalism is to channel political conflict, dis-agreements that would otherwise spill into thestreets and be settled according to violence, intoinstitutions that operate peacefully according tolaw and reach decisions that members of a polit-ical community accept as authoritative. In situ-ations of civil war, constitutions do not performthis basic function. So it would seem that to fullyunderstand civil war onset and duration, weshould understand the role that constitutionsplay. Are constitutions the causes of the break-down of politics? Are they casualties of broaderforces that engulf the constitutional system andlead to civil war? Or is the role of constitutionsa complex combination of the two? These areimportant questions that scholars of civil warhave yet to ask directly.

However, another line of research pointsthe way to bridging this gap. The focus hereis not on the factors associated with the riskof civil war onset, but rather on the explana-tion of civil war settlement. Whereas some au-thors see the onset, duration, and settlement ofcivil wars as interrelated phenomena that aredue to the same factors, others view the suc-cess and failure of civil war settlement as dis-tinct issues. As Toft (2010) explains, the lit-erature on the failure of settlement has threestrands. One posits that settlements fail whenthey do not share power between former com-batants adequately, whereas a second lays theblame at the feet of spoilers who see more gainfrom renewed civil war than from peace. But athird strand of research has focused on the dif-ficulty that states have in making credible com-mitments to rebel groups, which in turn makesnegotiated settlements to civil wars difficult toreach or difficult to implement (Walter 2009).Walter describes two kinds of restraints on theability of states to credibly commit: weak legaland political institutions and highly politicized,fixed cleavages. Both factors point to problemsin constitutional design, although Walter doesnot develop the constitutional implications ofher analysis, perhaps because her own researchplaces great emphasis on the role of third-party

enforcement to overcome commitment prob-lems (Walter 2002). Nonetheless, it is possibleto suggest what these constitutional implica-tions could be. Weak legal and political institu-tions serve as barriers to credible commitmentbecause they undermine the certainty that gov-ernments will adhere to civil war settlements.As Walter (2009, p. 251) puts it:

Governments and potential rebel groupswould have little to fear from a negotiated set-tlement if they were certain that the termswould be implemented and enforced overtime . . . . Groups worry, however, that gov-ernments will renege on their promises, ex-ploit the peace, and use an agreement to theirlong-term advantage. This fear is especiallystrong in countries where political and legalinstitutions are not strong enough to checkexecutive control.

Walter does not identify the kind of insti-tutions that would have to exist to control theexecutive for commitments to be credible. Butit is clear that independent and impartial courtswould be central to this constitutional strat-egy. Moreover, courts would potentially workalongside other institutions with complemen-tary but narrower mandates that would supple-ment the role of the courts and that the courtsin turn would support.

Walter (2009, p. 252) describes the secondfactor—“cemented cleavages”—as raising thefollowing problem:

Credible commitments are also particularlydifficult to make to minority groups in coun-tries with fixed political cleavages. This is be-cause the majority almost always has the num-bers to override promises made to a weakerrival, and the minority can do little to preventthis.

This is a rather different point that bun-dles together two different concerns. One isthe well-known fact about the character ofpolitical cleavage in ethnically divided soci-eties (Choudhry 2008). In societies that are not



ethnically divided, political and ethnic cleav-ages are cross-cutting, which promotes politi-cal moderation and blunts partisan divisions. Inaddition, this account of political cleavages isclosely tied to the case for a competitive modelof democratic politics. It assumes that politicsis characterized by shifting coalitions and ma-jorities and that political parties will competefor median voters at the center of the spec-trum and, hence, that political competition cre-ates pressures toward moderation. In addition,there is no permanent exclusion of any seg-ment of society from political power, encour-aging political losers to accept their loss in thehopes that they will win another day. In eth-nically divided societies, none of these assump-tions holds. Cleavages are mutually reinforcing,not cross-cutting, with ethnic division mappingonto political division and political competi-tion not producing moderation, but immodera-tion. The danger is of a majority dictatorship indemocratic form that will not have the same in-centives to check the abuse of public power pro-duced by political competition. Ethnic minori-ties, who are persistent losers with no prospectof wielding political power, would have the in-centive to exit politics and turn to violence.

A second concern is majoritarian tyranny onthe part of an ethnic majority. Walter’s (2009)point appears to be that even if effective legalinstitutions exist, the terms of the agreementthemselves could be undone through normal le-gal processes. This implies several claims aboutlegal form and entrenchment: that an agree-ment would take legal form, that an agreementimplemented through statute could be undoneby a simple legislative majority, and perhapseven that constitutional entrenchment is ul-timately ineffective in the face of determinedmajorities.

Walter cites these factors as lowering thepossibility of settlement to civil war becausethey undermine the ability of the state to cred-ibly commit. But alternatively, these factorscould have an effect on the character of a consti-tutional settlement to civil war. Credible com-mitment is a lens through which to assess the va-riety of constitutional options. Consider a bill of

rights. In the postconflict context, the salienceof a bill of rights would appear to be very high.The breakdown of the previous constitutionalorder, leading to a downward spiral of violenceand civil war, is almost always accompanied bygross abuses of human rights at the hands ofthe previous regime, the failure of the regime toprotect members of minorities, and the ineffec-tiveness or unwillingness of institutions to holdperpetrators accountable. Indeed, these variousforms of failure in human rights protection areoften cited as a reason to impugn the legitimacyof, and to withdraw support from, the previ-ous constitutional order. A postconflict settle-ment would by necessity include a bill of rightsas a political acknowledgment of past wrongsand, more concretely, as an effort to preventfuture abuses from ever occurring again. Thefunction of a bill of rights is to serve as a hardcheck on political power by enabling minori-ties to invoke the machinery of the courts to setbinding constraints on political decision mak-ing. Thus, a bill of rights is meant to serve asa credible commitment to minorities that theyshould set aside violence and acquiesce to andparticipate in the new constitutional legal or-der. Other related policy instruments that arenot typically understood as linked to constitu-tionalized rights protection should be under-stood as important counterparts of this post-conflict constitutional strategy. For example,the reform of the security sector—among otherthings the professionalization of the military,the separation of roles of the military (focusedon external defense) from the police (confinedto domestic law enforcement), and the estab-lishment of civilian oversight—should be un-derstood as closely aligned with this role for abill of rights (Toft 2010). I call this functionof a bill of rights the “regulative conception”(Choudhry 2008, 2009). The question raised ishow bills of rights fare in terms of their regu-lative ability to serve as instruments of crediblecommitment.

However, this is not the only conception ofa bill of rights that is suggested by the civil warliterature. Roeder & Rothchild (2005) have re-cently advocated “nation-state stewardship” as

18.10 Choudhry


a model for constitutional design to end civilwars. Their presentation of this model has twodifferent limbs, one focused on political insti-tutions and the other on political identities.The institutional dimensions of this model arebased explicitly on the U.S. Constitution, hav-ing three elements: civil liberties (protected bya bill of rights), a separation of powers that cre-ates multiple majorities, and a system of checksand balances across these different institutions.Moreover, the task of government is limited.Political identity must stem from the fact thatthe people of a state “share a common senseof nationhood” (Roeder & Rothchild 2005,p. 337). This sense of shared identity need notbe particularly deep; it “may only be a consen-sus that they should remain together in a singlestate that maintains domestic law and order anddefends them against foreign enemies” (Roeder& Rothchild 2005, p. 338).

The authors do not develop the link betweeninstitutions and political identity. On one read-ing, the sense of nationhood is a given that ex-ists independent of and prior to political insti-tutions and shapes their design. Thus, whethera state should exist or be partitioned dependson whether “the ethnic groups share a commit-ment to a common state.” In addition, the scopeof limited government is determined by therange of political disagreement—that is, sharedpolitical institutions may only make decisionsregarding issues on which there are no funda-mental disagreements among former combat-ants. But on another reading, these institutionscreate a shared sense of identity. In particular,consider Roeder & Rothchild’s (2005, p. 342)explanation of the function of a bill of rights:

The rights possessed by ethnic minoritiesmust also be identical to the rights enjoyed byother types of majorities and minorities in civilsociety . . . so that ethnicity is not privileged asa basis of political participation. In the contextof ethnically divided societies a major purposeof homogeneous individual civil rights is toencourage the proliferation of interests and,particularly, interests that cross ethnic groupsin defense of civil rights. In these conditions,

subgroups within the ethnic majority are morelikely to jump to the defense of the rights ofethnic minorities to defend the rights theyshare in common.

Roeder & Rothchild could mean no morethan the following: If rights are insecure forsome, they are insecure for all. The bond isone of self-interest. But this politics of rightsgrounded in shared self-interest could thickeninto a politics of rights grounded in commonnational identity. This is what I have termedthe constitutive conception of a bill of rights(Choudhry 2009). A bill of rights calls uponcitizens to abstract away from race, religion,ethnicity, and language, which have previouslyserved as the grounds of political identity andpolitical division, and instead to view them-selves as citizens who are equal bearers of con-stitutional rights. To serve as an instrument ofnation building, a bill of rights must transformthe political self-understanding of citizens.

THE LIMITATIONS OF BILLSOF RIGHTSSo there are two conceptions for bills of rightsin postconflict constitutional design: the regu-lative and the constitutive. In this section, I ar-gue that the diminished salience of bills of rightsin postconflict constitutional negotiations re-flects the fact that they cannot fully fulfill theseaims.

The Regulative Conceptionof Bills of RightsFirst consider the regulative conception of billsof rights. There are three factors that under-mine the ability of bills of rights to serve ascredible commitments against future abuses ofhuman rights in postconflict settings: (a) politi-cized judiciaries, (b) constitutional underen-forcement, and (c) the ex post nature of judicialreview.

Politicized judiciaries. Bills of rights dependon the independence and impartiality of the



judiciary for their enforcement. This poses aproblem in many postconflict societies becauseof the close identification of the judicial systemwith the previous constitutional regime. Oneof the most important lessons of the compar-ative study of authoritarian regimes has beenthat courts have served as central instrumentsof state control. This has been well documentedin the central cases of “wicked” legal systemsof Nazi Germany and apartheid South Africa(Dyzenhaus 2009). However, the insight ex-tends well beyond those cases to a broader spec-trum of nondemocratic regimes (Ginsburg &Moustafa 2008, Helmke 2005, Hilbank 2007,Moustafa 2007). The enforcement of securitylaws against political opponents is perhaps thecentral role of courts in these contexts. How-ever, the body of law that courts enforce againstpolitical opposition to protect existing constitu-tional regimes is much broader. These includelaws restricting participation in political institu-tions (e.g., laws governing political parties andelections) and freedom of expression (e.g., thelaws on libel). More generally, these courts areintegral components of the very framework ofpreconflict constitutional regime, which has of-ten been the source of the political grievancesgiving rise to civil war.

The allegation that a judiciary is politicizedentails two distinct claims, both of which posechallenges for bills of rights as part of postcon-flict constitutional settlements. First, individualjudges may be partisans of the previous politicalregime, closely allied at a personal and profes-sional level with executive and legislative office-holders or, more generally, part of the rulingparty. These judges acted as agents for, ratherthan as checks on, the wielders of power withinthe previous constitutional order. The questionis whether they can be trusted to adjudicate im-partially under a new constitutional scheme thatincludes a bill of rights that renders unconstitu-tional precisely the conduct that was previouslylegal. Second, distinct from the personal po-litical affiliations of judges, the judiciary as aninstitution might have an ideological commit-ment to the preexisting constitutional regime,which is reflected in its jurisprudence. In

particular, the preexisting constitutional doc-trine might be built around robust notions ofdeference to executives and legislatures. The is-sue is how these judges, and the doctrinal frame-work that they have constructed and withinwhich they operate, will respond to a new con-stitutional dispensation that includes a bill ofrights that fundamentally reshapes the charac-ter of the constitutional order.

Sri Lanka illustrates both dimensions ofpoliticization (International Crisis Group2009). The principal vehicle for executivemanipulation of the judiciary has been thepresident’s power to appoint the chief justiceof the Supreme Court, who sits atop the SriLankan judicial hierarchy, as well as other highcourt judges. High court judges are increasinglydrawn from the ranks of the attorney general’sdepartment, which itself has become morepoliticized. These judges tend to rule in favorof the executive in security cases. Moreover,the recently retired chief justice has used hisextensive powers to favor the Sri Lankan gov-ernment. He excluded Supreme Court judgescritical of the government from panels thatheard constitutional cases. In addition, throughhis power to appoint the members of the Judi-cial Service Commission, which is vested withresponsibility for promotions, discipline, andtransfers, the former chief justice interferedwith proceedings against the government withregard to human rights abuses. In addition,the Supreme Court has found unconstitutionalSri Lanka’s access to the Optional Protocolto ICCPR (allowing for individual rights ofcomplaint) and issued an implausible advisoryopinion on Sri Lanka’s compliance with theICCPR, which is allegedly motivated bythe desire to prevent Sri Lanka from losingtrade concessions tied to compliance withinternational human rights standards. Outsidethe area of rights adjudication, the SupremeCourt has struck down interim power-sharingarrangements in the Northeast and the mergerof the North and Eastern provinces (a keyTamil demand). Writing before the cessa-tion of hostilities in 2009, the InternationalCrisis Group (2009, p. 1) explicitly links the

18.12 Choudhry


politicization of the Sri Lankan judiciary withthe problem of credible commitment:

Without addressing this corrosion [i.e., of ju-dicial independence], Sri Lanka is unlikely toforce the stable political compromises thatmight now be available . . . . Courts presentlyprovide no guarantee of personal securityor redress against arbitrary state violence,let alone the possibility of transitional justice,necessary for a transition from violence. Theyare more likely to destabilize political com-promises that could help mitigate Sri Lanka’senduring social fissures.

Politicized judiciaries raise acute dilemmasof transitional justice that can be recast asproblems of credible commitment. On the onehand, the wholesale replacement of the existingjudiciary is not an option. It would contradictthe logic of a negotiated settlement wherebyall parties participate in the exercise of publicpower. Ironically, it might also subvert respectfor the rule of law because it would suggest thatthe particular occupants of judicial office mat-ter more to adjudication than impersonal legalmaterials. On the other hand, simply vestingthe existing judiciary with the responsibilityfor interpreting a new bill of rights is not anoption either, because it might undermine theenforcement of a new bill of rights. In severaltransitional democracies, the solution has beento leave the preexisting judiciary in place andto create a specialist constitutional court withsupreme and potentially exclusive authorityto interpret the bill of rights, with mem-bers appointed by the new regime. But thisinstitutional innovation may be of limited ap-plicability. The central examples of the creationof new constitutional courts—in South Africaand in Eastern and Central Europe—havebeen in democratic transitions that restored orestablished majority rule. The constitutionalcourt becomes a vehicle by which a newlyempowered political majority asserts its au-thority over the discredited courts of the priorregime, kept in place as a matter of necessity.By contrast, in many civil wars, the rebel group

represents a political minority that turned toviolence because of persistent losses in majori-tarian politics. On a new constitutional court,they would at best continue to be a minority.A supermajority decision rule on the new con-stitutional court would only partially addressthis concern because it would create a minorityveto but would not encompass a positive powerof decision on cases involving bills of rights.

Constitutional underenforcement. Even ifthese concerns regarding the politicization ofthe judiciary could be met, another barrierto credible commitment is the phenomenonof constitutional underenforcement. Debatesover the legitimacy of judicial review pre-suppose an account of the nature of rightsadjudication in which constitutional reasoningconsists of highly abstract, moral argumentrooted in principles of political morality—i.e.,Dworkin’s (1996) forum of principle. Theprovisions of bills of rights themselves, whichincorporate principles of political morality byreference, fuel this portrayal of constitutionalinterpretation. However, in recent years,constitutional scholars have observed thatthere is a wide gap between this picture andthe actual practice of judicial review. Courtsdeliberately and systematically engage in theunderenforcement of constitutional provisions(Sager 1978). There are four distinct strategiesof constitutional underenforcement: decisionalavoidance, decisional minimalism, decisionaldeference, and remedial deference.

Decisional avoidance refers to a set of tech-niques whereby courts decline to rule on con-stitutional issues that are brought before them.These techniques may consist of rules govern-ing standing (i.e., that a litigant has sufficientinterest in a case to launch it), ripeness (i.e., thata dispute is not premature), mootness (i.e., thata dispute is still alive), political questions doc-trines or doctrines of nonjusticiability (i.e., thatthe issues are political, not legal, and that theyare not amenable to judicial review), and thedoctrine of constitutional avoidance (i.e., thatcases should be resolved on nonconstitutional



grounds if possible). Bickel (1962) termed thesevarious techniques the “passive virtues.”

Decisional minimalism, by contrast, refersto the manner in which courts reason abouttheir judgments (Sunstein 1999). Maximalistjudgments are broad and deep—that is, theylay down constitutional principles with poten-tially far-reaching implications in a broad va-riety of future cases and ground those judg-ments in deeply theorized accounts of a bill ofrights. Minimalist judgments, by contrast, arenarrow and/or shallow—that is, they are limitedin scope to the case before the court, by designdo not set out principles applicable to a broadset of future cases, and are based on reasoningthat is not theoretically sophisticated. Judicialminimalism is often a product of the doctrinaltests developed by courts to interpret and im-plement broadly worded guarantees in bills ofrights (Fallon 2001, 2006).

Next, there is decisional deference. Inmost jurisdictions, rights adjudication con-sists of a two-stage process that determinesfirst whether a right is violated and secondwhether that violation is justified accordingto a proportionality analysis. In these systems,rights are generally given a broad interpreta-tion, and countervailing interests are addressedexclusively under proportionality. This con-trasts with the one-stage approach—so-calleddefinitional balancing—that is prevalent inAmerican constitutional doctrine, which con-flates the scope of a right with its strength.At a superficial level, the broader reading ofa right might suggest the eschewal of consti-tutional underenforcement, which is often ev-ident in definitional balancing. However, un-derenforcement occurs at the proportionalitystage, through a deferential approach to pro-portionality analysis that protects rights veryweakly.

Finally, there is remedial deference. Vio-lations of bills of rights can spawn a broadvariety of remedies that vary considerably intheir stringency. At one end of the spec-trum, constitutional remedies consist of dec-larations of constitutional invalidity, damages,and various forms of mandatory relief (e.g.,

injunctions) that end unconstitutional state ac-tion and directly order governments to com-ply with their constitutional obligations. At theother end of the spectrum are much more def-erential forms of relief, such as suspended dec-larations of invalidity and declaratory relief thatpreserve unconstitutional states of affairs, givegovernments time to come to terms with theirconstitutional obligations, and rely on the goodfaith of government to act constitutionally.

Although many of these practices of con-stitutional underenforcement have been identi-fied in the practice of American courts, identicalor similar techniques are a widespread featureof constitutional adjudication. They are moti-vated by legitimate concerns regarding compar-ative institutional competence and democraticlegitimacy, which raise fears of judicial overen-forcement. But separately and in combination,they also have the effect of seriously limiting theability of bills of rights to discipline all forms ofunconstitutional conduct, by deliberate design.As a consequence, even a system of depoliticizedcourts cannot provide a credible commitmentagainst future abuses of rights.

Among the principal casualties—directlyrelevant to postconflict constitutionalsettlements—are constitutional guaranteesof equality. A principal grievance leveled ata particular previous regime may be that itdistributed primary social goods in the Rawl-sian sense—liberty and opportunity, incomeand wealth, and the bases of self-respect—ona discriminatory basis. So a constitutionalguarantee of equality would be a sine quanon to a bill of rights. This kind of provisionwould encounter two kinds of discriminationclaims: that the state has engaged in directdiscrimination against members of certaingroups or that the state has engaged in indirectdiscrimination through the adoption of faciallyneutral criteria that are unequal in theireffects on different groups. The former sortof discrimination would be easy to identifyand constitutionally condemn. But the latterhas proved much more difficult to discipline,even under bills of rights that prohibit indi-rect discrimination. The reason is the real

18.14 Choudhry


possibility of judicial error. The proof of indi-rect discrimination turns on complex questionsof evidence and raises the additional questionof whether policies with discriminatory effectsare nonetheless justified. Given that indirectdiscrimination is far more pervasive than directdiscrimination, this means that a principalsource of state-created inequality will remainbeyond judicial oversight.

Ex post judicial review. Institutions to pro-mote public accountability can operate eitherex ante or ex post. Ex ante institutions operatebefore a decision is undertaken; ex post institu-tions operate afterward. Judicial review undera bill of rights is, in most cases, a process of expost review. To be sure, there are exceptions,such as preenactment legislative scrutiny that isengaged in by the French Conseil Constitutionneland other courts with such limited jurisdiction.Moreover, forms of parliamentary rights reviewcan similarly monitor legislation for compliancewith a bill of rights during the legislative pro-cess. But ex post review is the norm.

Ex post review can operate in different ways.In systems of diffuse review, it is accompa-nied by a series of rules on standing, mootness,and ripeness. Although the precise details varyacross jurisdictions, the net effect of these rulesis to limit access to ex post review to situations inwhich a litigant has a sufficiently strong interestin a constitutional claim; the claim refers to anactual dispute, not a hypothetical scenario; andthe dispute remains live. Although it is often ar-gued that systems of centralized judicial revieware different because review is abstract and isnot tied to the resolution of a particular dispute,this is a caricature. Cases in centralized systemsstill arise from claims brought by individuals orpolitical actors and through the ordinary courtsor directly to a constitutional court. In the caseof individual claims, a live legal dispute givesrise to the case, and the referring judge appliesthe ruling once it is received to resolve the caseat hand. There is no material difference in chal-lenges to executive conduct. The distinctionis whether rulings on statutes find the statuteas a whole unconstitutional (centralized and

diffused systems) or in addition can find a statuteunconstitutional in how it is applied (diffusedsystems) (Comella 2004).

Ex post judicial review is pervasive. Theprincipal reason for this is informational. Tointerpret and apply bills of rights, courts re-quire information regarding the impact of thoserights. The persons who possess that informa-tion are those whose rights have been infringed.The principal advantage of ex post over ex antereview is that it creates the incentive for thosemost affected to bring the relevant informationto the courts. Indeed, the link between the ef-ficacy of bills of rights and ex post judicial re-view is so widely accepted that a recent set ofconstitutional amendments granted the ConseilConstitutionnel powers of ex post review.

However, ex post review suffers from seriousweaknesses. The first problem is conceptual—i.e., the conception of constitutional wrongs onwhich it relies. If violations are conceptualizedas discrete wrongs that occur at a fixed point intime, a finding of unconstitutionality necessar-ily occurs after a right has been violated, anda remedy is provided after the fact. This ana-lytic structure—which analogizes between vio-lations of bills of rights and violations of rightsunder private law—has been aptly termed themodel of “transactional harm” (Levinson 2002).The model has several failings: that harms arenot discrete events in the past but are often con-tinuing, that harms are not experienced merelyby individuals but by groups, and that reme-dies can rarely compensate for the rights vi-olation. Accordingly, a tradition of Americanpublic law scholarship has developed a distinctmodel of public law adjudication. Although di-verse in its substantive focus and prescriptions,the shared commitment has been on preventingfuture unconstitutional conduct (Chayes 1976,Fiss 1988). This has also led to a redescriptionof constitutional practice.

The unstated assumption in this reconcep-tualization and redescription of public law prac-tice in consolidated constitutional systems isthat there is not ongoing, systematic noncom-pliance with bills of rights. Discrete viola-tions of bills of rights, even if serious, are not



synonymous with the breakdown of the con-stitutional order. Thus, the confined scope ofrights violations and the existence of mecha-nisms to prevent their reoccurrence offset thefact that those harms cannot be compensated.Together, they provide a credible incentive forparties whose rights are violated to adhere to theconstitutional order. Such a commitment is fre-quently lacking in situations of conflict. Abusesof human rights are not isolated but, when un-dertaken at the hands of the state, are deliberate,wanton, and systematic. Unlawful arrest, illegaldetention, internment, torture, rape, murder,and mass expulsions are widespread. These cir-cumstances magnify the inadequacy of ex postforms of judicial review, especially because theyoften occur in the face of constitutional andother legal prohibitions on such conduct. Thus,in postconflict constitutional negotiations, expost mechanisms of accountability are far lessattractive than ex ante controls that preventsuch abuses from occurring in the first place.And so constitutional negotiations focus lesson rights-based constitutionalism and more onpower sharing and vetoes in executive and leg-islative institutions.

The Constitutive Conceptionof Bills of RightsAccording to the constitutive conception, a billof rights constitutes the demos that it also con-strains. It encodes and projects a certain visionof political community—in particular, the ideaof a political community as consisting of rights-bearing citizens of equal status. To serve as aninstrument of nation building, a bill of rightsmust alter the very self-understanding of citi-zens. Can a bill of rights serve this constitutivepurpose?

We first need to unearth the genealogy ofthis idea. The idea that a bill of rights can serveas an instrument of shared national identityhas recently been championed by Habermas(1996), contributing to a long-standingdebate over the nature and character of na-tionalism. Habermas distinguishes betweenShickalsgemeinschaft (ethnic nationalism) and

Verfassungspatriotismus (constitution patrio-tism or civic nationalism). According to theethnic conception, political communities areimagined as emerging from peoples or nationsunited by a common bond that exists indepen-dently of and prior to the creation of a politicalcommunity and that is the object of loyalty,belonging, or identification. Citizenship ina political community tracks membership inthe underlying nation. In its extreme versions,the ethnic conception defines nations in termsof descent and precludes the acquisition ofmembership by outsiders. More moderateformulations define the nation in terms ofa shared language, history, religion, and/orcultural traditions, allowing for persons wholack these characteristics to embrace them andbecome members of a political community.According to constitutional patriotism, a po-litical community is based not on a prepoliticalbond but, rather, on an allegiance to sharedprinciples of political justice flowing from aliberal political morality and to a common setof political institutions through which thoseprinciples are realized. A political community isimagined as a voluntary association of citizensconsidered free and equal, who constitute apolitical community because of a shared beliefthat they should associate for political ends.Citizenship can be held by any person willingto affirm and uphold the principles of politicaljustice that lie at the foundation of the politicalcommunity.

The most recent discussions of constitu-tional patriotism have revolved around, first,the normative question of its ability to serve asthe basis for political legitimacy in societies thataspire to be liberal democracies and, second, thesociological question of whether it can providesufficient social unity for a liberal state to realizemany of its most important objectives. The am-bition of liberal constitutionalism is that a con-stitutional order must be legitimate and mustenjoy the allegiance of a sufficient number of itscitizens to work. In the liberal conception, theconditions for the legitimate exercise of publicpower are the rights and institutions of repre-sentative government that one finds in a typical

18.16 Choudhry


liberal-democratic constitution, such as a bill ofrights. The ambition of constitutional patrio-tism is that these same conditions also supplythe necessary motivational element for thoseinstitutions to work. Additionally, the connec-tion between legitimacy (normative) and stabil-ity (sociological) is not contingent. Rather, it isconceptual—it is the ambition of constitutionalpatriotism that citizens view themselves as partof the same constitutional-legal order, preciselybecause that order is legitimate.

Contemporary discussions of constitutionalpatriotism originated with the reunification ofGermany. The question was how Germansshould make sense of reunification. There weretwo options on the table. One was ethnic na-tionalism, according to which German reuni-fication brought ethnic nation and state backinto alignment, after a four-decade interrup-tion. The other was constitutional patriotism.For Habermas, the core of the German politicalidentity was the Basic Law, Germany’s postwarconstitution. Reunification was justified as therestoration of democracy and the rechstaat in aterritory that had lacked both since the rise ofHitler.

However, the notion of constitutional patri-otism was soon applied to postconflict consti-tutional contexts such as Afghanistan, Rwanda,Bosnia, and Northern Ireland, in which thecleavages between groups were ethnic (Barber1996, Muller 2007). Canovan (2000, p. 422)argues that the goal of constitutional patri-ots in these states was to move beyond “morepragmatic schemes for coping democraticallywith the tensions between potentially hostilegroups” where it was taken as a given that eth-nic identity would be the principal basis for po-litical mobilization. The goal of constitutionalpatriotism was “to lift individuals above theirascriptive identities into a shared public spherewhere all are equal citizens with a shared loyaltyto an impartial state” (Canovan 2000, p. 422).Thus, for Northern Ireland, “[r]ecognising acommon set of rights in a document that all cancommit to, at least in part, is seen as an impor-tant element in building a new political society,providing the possibility at least of common

identification with the basic document” (Mc-Crudden 2001, p. 378; Kavanagh 2004).

Can a bill of rights constitute a nationalidentity? Comparative experience fromGermany and Canada—two consolidateddemocracies with highly developed systemsof rights adjudication that serve as importantcomparative models—counsels skepticism.Consider Germany. The best answer toHabermas came from Yack (1996). Yackargued that a purely abstract constitutionalpatriotism could not explain the defense ofGerman reunification over the simple restora-tion of liberal democracy in East Germany orwhy Germany did not unify with the formercommunist dictatorship of Czechoslovakia,with which it shared a border. Constitutionalpatriotism in Germany was accordingly bestunderstood as an appeal to a certain audience,united by a shared historical memory and com-mon historical experiences that gave the rulesand institutions of liberal democracy a partic-ular salience. The more general point is thateven in nations that claim to define citizenshipin civic terms, those principles are nested in acontingent context—a constitutional narrativedrawing on a web of political memory forgedby shared experiences, challenges, failures, andtriumphs, which is often but not necessarilytied to a particular set of institutions.

Now consider Canada. Canada adopted aconstitutionally entrenched bill of rights, theCanadian Charter of Rights and Freedoms, in1982. The principal political objective behindthe adoption of the Charter was not to insureagainst the potential loss of political power bythreatened political elites, but rather to combatsubstate nationalism in the province of Que-bec through the creation of a pan-Canadianconstitutional patriotism that would competewith and eventually overwhelm Quebec nation-alism. In a federal state such as Canada, be-cause citizens share these rights irrespectiveof language or province of residence, a bill ofrights was meant to serve as a transcendent formof political identification—the spine of com-mon citizenship that unites members of a lin-guistically diverse and geographically dispersed



polity across the country as a whole. As theGovernment of Canada stated, the consti-tutional entrenchment of “individual humanrights for all Canadians . . . is a fundamentalcondition of nationhood” and is “fundamentalto the will of the nation to survive” (Govern-ment of Canada 1968, p. 18).

The Charter has not met its constitutivegoal. Outside of Quebec, the Charter has gen-erated a new pan-Canadian patriotism. How-ever, within Quebec, the Charter has notserved to bind francophone Quebeckers to theCanadian constitutional order. The differentialeffect of the Charter on Canadian political iden-tity has been evident in a series of constitutionalepisodes over the past two decades (Choudhry2009). The conflicting reactions to the MeechLake Accord within and outside Quebec illus-trate this point (Cairns 1988). Quebec had op-posed the adoption of the Charter, and the Ac-cord was a series of constitutional amendmentsthat together were an attempt to bring Quebecinto the constitutional fold. Outside of Quebec,the Meech Lake Accord was criticized because(a) the intergovernmental process that pro-duced it was inconsistent with the notion thatcitizens were the constituent actors in the con-stitutional process, and (b) there was fear that aprovision recognizing that Quebec constituteda distinct society within Canada (the DistinctSociety Clause) would undermine equal citizen-ship by authorizing Quebec to limit the Char-ter in a manner not open to other provincialgovernments. Within Quebec, the reaction wasexactly the opposite, on the theory that (a) theconstituent actors in the constitutional processwere Canada’s nations, of which Quebec wasone, and (b) the Distinct Society expressly rec-ognized this fact.

The broader lesson is that it is very difficultfor bills of rights, on their own, to serve a consti-tuting role in defining a new political identity.The German example suggests that for a bill ofrights to serve as the basis of a common politicalidentity, it must be situated in a contingent his-torical and political context—i.e., it depends on“supplements of particularity” (Markell 2000,p. 40). This history is often the topic of intense

disagreement, as it is bound up in competing ac-counts of the sources and character of the con-flict that led to violence in the first place. TheCanadian example suggests that the task is evenmore challenging where the existence of com-peting nationalisms within the same state makesthe dominant question of constitutional pol-itics the conflict between competing nationalnarratives. This is the multinational or pluri-national state, a pervasive form of postconflictpolity (Kymlicka 1995, Keating 2001). Indeed,there is the danger that rather than transcend-ing those national narratives, a bill of rights willbe drawn back into it. The difficulties may begreater still for countries emerging from violentconflict.

CONCLUSION: LESSONS FORCONSTITUTIONAL DESIGNFOR DIVIDED SOCIETIESIn his case against bills of rights backed up byjudicial review, Waldron (2006) makes a “con-ditional case,” based on a series of assumptionsabout the institutional and political features ofmodern liberal democracies. These assump-tions delineate the circumstance in which hemakes the core case against judicial review.Societies in which one or more assumptions donot hold are noncore cases. Societies emergingfrom violent conflict fail to meet one or moreof Waldron’s assumptions. They do not havedemocratic institutions in reasonably goodworking order, they do not have a set of judicialinstitutions in reasonably good order, manyindividuals and officials are not committedto rights, and disagreements about rights arenot nearly always in good faith. It is preciselyin this context that the case for rights-basedconstitutionalism is strong.

Indeed, bills of rights are expected to doa great deal of work in postconflict constitu-tions. Advocates of rights-based constitutional-ism look to bills of rights to constrain abusesof public power and to constitute a new polityfounded on constitutional patriotism. But thedeclining intensity of debates over bills of rightsin postconflict constitutions suggests that bills

18.18 Choudhry


of rights cannot do the work that is expected ofthem. Politicized judiciaries, constitutional un-derenforcement, and the ex post nature of judi-cial review undermines the ability of the bill ofrights to serve as a credible commitment againstfuture abuses of human rights. Moreover, theidea of a bill of rights as a source of shared polit-ical identity abstracted from a contingent polit-ical and historical context is unlikely to succeedin practice.

And so it should not be surprising that billsof rights have attracted minimal attention in thecopious literature on constitutional design individed societies. For the past three decades,a major question in comparative politics hasbeen that of constitutional design in ethnicallydivided societies, which overlap with many ofthe most recent examples of postconflict design.

There is no shortage of constitutional prescrip-tions for managing ethnic divisions of thesesorts, such as electoral system design, the struc-ture of and relationship between the executiveand the legislature, federalism, and legal plural-ism. Nonetheless, bills of rights have attractedno sustained analysis. Indeed, Lijphart’s (1977)Democracy in Plural Societies and Horowitz’s(1985) Ethnic Groups in Conflict—the two classictexts in the field, which have set the parame-ters of the debate for over two decades—barelymention it. By contrast, Lijphart, Horowitz,and a legion of scholars have spilled a vastamount of ink on comparatively narrow is-sues, such as the choice between the alternativevote and proportional representation. This mayhave been an oversight. Or it may have beenprescient.

DISCLOSURE STATEMENTThe author is not aware of any affiliations, memberships, funding, or financial holdings that mightbe perceived as affecting the objectivity of this review.

ACKNOWLEDGMENTSAn earlier version of this paper was presented as a keynote address at the 21st AnniversaryConference of the Center for Comparative Constitutional Studies at the University MelbourneSchool of Law in November 2009. I thank Carolyn Evans, Theunis Roux, Wojciech Sadurski,Adrienne Stone, and Alison Young for their helpful comments. Nicolas Businger, Nathan Hume,and Michael Sabet provided excellent research assistance.

LITERATURE CITED

Alexander GS. 2006. The Global Debate over Constitutional Property: Lessons from the American Takings Jurispru-dence. Chicago: Univ. Chicago Press

Allen T. 2000. The Right to Property in Commonwealth Constitutions. Cambridge, UK: Cambridge Univ. PressBakan J. 1997. Just Words: Constitutional Rights and Social Wrongs. Toronto: Univ. Toronto PressBarak A. 1996. Constitutional human rights and private law justice. Rev. Const. Stud. 3(2):218–81Barak A. 2006. The Judge in a Democracy. Princeton, NJ: Princeton Univ. PressBarber BR. 1996. Constitutional faith. In For Love of Country: Debating the Limits of Patriotism, ed. MC

Nussbaum, pp. 30–37. Boston: BeaconBell C. 2008. On the Law of Peace. New York: Oxford Univ. PressBickel AM. 1962. The Least Dangerous Branch: The Supreme Court at the Bar of Politics. Indianapolis: Bobbs-

MerrillBilchitz D. 2007. Poverty and Fundamental Rights: The Justification and Enforcement of Socio-Economic Rights.

New York: Oxford Univ. PressBouillon ME, Malone DM, Rowswell B, eds. 2007. Iraq: Preventing a New Generation of Conflict. Boulder, CO:

Lynne Rienner



Cairns A. 1988. Citizens (outsiders) and governments (insiders) in constitution-making: the case of MeechLake. Can. Public Policy. 14(Suppl.: The Meech Lake Accord):S121–45

Canovan M. 2000. Patriotism is not enough. Br. J. Polit. Sci. 30(3):413–32Chayes A. 1976. The role of a judge in public law litigation. Harvard Law Rev. 89(7):1218–316Choudhry S. 1999. Globalization in search of justification: toward a theory of comparative constitutional

interpretation. Ind. Law J. 74:819–92Choudhry S. 2004. The Lochner era and comparative constitutionalism. Int. J. Const. Law 2(1):1–55Choudhry S, ed. 2006. The Migration of Constitutional Ideas. Cambridge, UK: Cambridge Univ. PressChoudhry S, ed. 2008. Constitutional Design for Divided Societies: Integration or Accommodation. New York:

Oxford Univ. PressChoudhry S. 2009. Bills of rights as instruments of national building in multinational states: the Canadian

Charter and Quebec nationalism. In Contested Constitutionalism: Reflections on the Canadian Charter of Rightsand Freedoms, ed. JB Kelly, CP Manfredi, pp. 233–50. Vancouver: Univ. B. C. Press

Collier P, Hoeffler A. 2004. Greed and grievance in civil war. Oxford Econ. Pap. 56(4):563–95Comella VF. 2004. The European model of constitutional review of legislation: toward decentralization? Int.

J. Const. Law. 2(3):461–91Dawisha A. 2009. Iraq: A Political History from Independence to Occupation. Princeton, NJ: Princeton Univ. PressDeeks AS, Burton MD. 2007. Iraq’s constitution: a drafting history. Cornell Int. Law J. 40(1):1–87Dworkin R. 1996. Freedom’s Law: The Moral Reading of the American Constitution. New York: Oxford Univ.

PressDyzenhaus D. 2009. Hard Cases in Wicked Legal Systems: South African Law in the Perspective of Legal Philosophy.

Oxford: Clarendon. 2nd ed.Elkins Z, Simmons B. 2005. On waves, clusters, and diffusion: a conceptual framework. Ann. Am. Acad. Polit.

Soc. Sci. 598:33–51Fallon RH. 2001. Implementing the Constitution. Cambridge, MA: Harvard Univ. PressFallon RH. 2006. Judicially manageable standards and constitutional meaning. Harvard Law Rev. 119:1274–

332Fearon JD, Laitin DD. 2003. Ethnicity, insurgency, and civil war. Am. Polit. Sci. Rev. 97(1):75–89Fiss OM. 1988. Coda. Univ. Toronto Law J. 38:229–44Fredman S. 2008. Human Rights Transformed. New York: Oxford Univ. PressGalanter M. 1984. Competing Equalities: Law and the Backward Classes in India. New York: Oxford Univ. PressGalbraith PW. 2008. Unintended Consequences: How War in Iraq Strengthened America’s Enemies. New York:

Simon & SchusterGardbaum S. 2001. The new commonwealth model of constitutionalism. Am. J. Comp. Law 49:707–60Gardbaum S. 2003. The “horizontal effect” of constitutional rights. Mich. Law Rev. 102(3):387–459Garlicki L. 2007. Constitutional courts versus supreme courts. Int. J. Const. Law. 5(1):44–68Ginsburg T. 2003. Judicial Review in New Democracies: Constitutional Courts in Asian Cases. Cambridge, UK:

Cambridge Univ. PressGinsburg T. 2006. Locking in democracy: constitutions, commitment, and international law. N.Y. Univ. J.

Int. Law Polit. 38(4):707–59Ginsburg T, Moustafa T. 2008. Rule by Law: The Politics of Courts in Authoritarian Regimes. Cambridge, UK:

Cambridge Univ. PressGoodman R, Jinks D. 2010. Socializing States: Promoting Human Rights through International Law. New York:

Oxford Univ. Press. In pressGovernment of Canada. 1968. Federalism for the Future. Ottawa: The Queen’s PrinterHabermas J. 1996. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Transl.

R Rehg. Cambridge, MA: MIT PressHelmke G. 2005. Courts Under Constraint: Judges, Generals, and Presidents in Argentina. Cambridge, UK:

Cambridge Univ. PressHiebert JL. 2005. Interpreting a bill of rights: the importance of legislative rights review. Br. J. Polit. Sci.

35(2):235–55Hiebert JL. 2006a. Parliamentary bill of rights: an alternative model? Mod. Law Rev. 69(1):7–28

18.20 Choudhry


Hiebert JL. 2006b. Parliament and the Human Rights Act: can the JCHR help facilitate a culture of rights?Int. J. Const. Law 4(1):1–38

Hilbank L. 2007. Judges Beyond Politics in Democracy and Dictatorship. Cambridge, UK: Cambridge Univ. PressHirschl R. 2004. Towards Juristocracy: The Origins and Consequences of the New Constitutionalism. Cambridge,

MA: Harvard Univ. PressHirschl R. 2006. On the blurred methodological matrix of comparative constitutional law. In The Migration

of Constitutional Ideas, ed. S Choudhry, pp. 39–67. Cambridge, UK: Cambridge Univ. PressHirschl R. 2008. The judicialization of mega-politics and the rise of political courts. Annu. Rev. Polit. Sci.

11:93–118Horowitz DL. 1985. Ethnic Groups in Conflict. Berkeley: Univ. Calif. PressHuntington S. 1991. The Third Wave: Democratization in the Late Twentieth Century. Norman: Univ. Oklahoma

PressIgnatieff M. 2000. The Rights Revolution. Toronto: House of Anansi PressInternational Crisis Group. 2009. Sri Lanka’s judiciary: politicised courts, compromised rights. Asia Report

No. 172, BrusselsIraq Constitutional Review Commission. 2009. Proposed Amendments to Constitution of IraqJackson VC. 2005. Constitutional comparisons: convergence, resistance, engagement. Harvard Law Rev.

119(1):109–28Kalyvas S. 2007. Civil wars. In The Oxford Handbook of Comparative Politics, ed. C Boix, SC Stokes, pp. 416–34.

Oxford: Oxford Univ. PressKavanagh A. 2004. The role of a bill of rights in reconstructing Northern Ireland. Hum. Rights Q. 26(4):956–82Keating M. 2001. Plurinational Democracy: Stateless Nations in a Post-Sovereignty Era. New York: Oxford Univ.

PressKymlicka W. 1995. Multicultural Citizenship: A Liberal Theory of Minority Right. New York: Oxford Univ. PressLevinson DJ. 2002. Framing transactions in constitutional law. Yale Law J. 111(6):1311–90Lijphart A. 1977. Democracy in Plural Societies: A Comparative Exploration. New Haven: Yale Univ. PressMarkell P. 2000. Making affect safe for democracy?: On “constitutional patriotism.” Polit. Theory 28(1):38–63McCrudden C. 2001. Not the way forward: some comments on the Northern Ireland Human Rights Commis-

sion’s consultation document on a bill of rights for Northern Ireland. North. Irel. Legal Q. 52(4):372–84Moravcsik A. 2000. The origins of human rights regimes: democratic delegation in postwar Europe. Int. Organ.

54(2):217–52Moustafa T. 2007. The Struggle for Constitutional Power: Law, Politics, and Economic Development in Egypt.

Cambridge, UK: Cambridge Univ. PressMuller JW. 2007. Constitutional Patriotism. Princeton, NJ: Princeton Univ. PressO’Leary B. 2009. How to Get Out of Iraq with Integrity. Philadelphia: Univ. Pa. PressO’Leary B, McGarry J, Salih K, eds. 2005. The Future of Kurdistan in Iraq. Philadelphia: Univ. Pa. PressParis B. 2004. At War’s End. Cambridge, UK: Cambridge Univ. PressRoach K. 2001. The Supreme Court on Trial: Judicial Activism or Democratic Dialogue. Toronto: Irwin LawRoeder P, Rothchild D, eds. 2005. Sustainable Peace: Power and Democracy after Civil Wars. Ithaca, NY: Cornell

Univ. PressSager LG. 1978. Fair measure: the legal status of underenforced constitutional norms. Harvard Law Rev.

91:1212–64Scalia A. 1998. A Matter of Interpretation: Federal Courts and the Law. Princeton, NJ: Princeton Univ. PressSpitz R, Chaskalson M. 2000. The Politics of Transition: A Hidden History of South Africa’s Negotiated Settlement.

Oxford, UK: HartSunstein CR. 1999. One Case at a Time: Judicial Minimalism on the Supreme Court. Cambridge, MA: Harvard

Univ. PressSunstein CR. 2001. Designing Democracy: What Constitutions Do. New York: Oxford Univ. PressToft MD. 2010. Securing the Peace: The Durable Settlement of Civil Wars. Princeton, NJ: Princeton Univ. PressTushnet M. 2003. The issue of state action/horizontal effect in comparative constitutional law. Int. J. Const.

Law 1(1):79–98Tushnet M. 2008. Weak Courts, Strong Rights. Princeton, NJ: Princeton Univ. Press



Waldron J. 1999. Law and Disagreement. New York: Oxford Univ. PressWaldron J. 2006. The core of the case against judicial review. Yale Law J. 115(6):1346–406Walter BF. 2002. Committing to Peace: The Successful Settlement of Civil Wars. Princeton, NJ: Princeton Univ.

PressWalter BF. 2009. Bargaining failures and civil war. Annu. Rev. Polit. Sci. 12:243–61Yack B. 1996. The myth of the civic nation. Crit. Rev. 10(2):193–211

18.22 Choudhry

after the rights revolution final print - sujit choudhry · important changes in the constitutional...

Documents