10

Deliberative Democracy and Constitutional Review

Christopher F. Zurn

 

1       The Counter-Majoritarian Difficulty with Judicial Review

For myself it would be most irksome to be ruled by a bevy of Platonic Guardians, even if I knew how to choose them, which I assuredly do not. If they were in charge, I should miss the stimulus of living in a society where I have, at least theoretically, some part in the direction of public affairs. Of course I know how illusory would be the belief that my vote determined anything; but nevertheless when I go to the polls I have a satisfaction in the sense that we are all engaged in a common venture.[1]

Learned Hand

 

I believe that many reservations about the judicial role of reviewing and potentially overturning statutes enacted by a democratically-elected legislature, reservations memorably expressed by Judge Learned Hand, arise from an apparently deep tension in our professed political ideals: namely, the tension between democracy and constitutionalism. For it would seem that if an institution such as the United States Supreme Court—a governmental body of nine individuals who are only remotely responsible to the electorate—has the power, in the name of the constitution, to overturn the considered will of the people as it has been formulated and executed by the legislature and the executive—governmental bodies more directly responsible to the people through periodic elections—then the power to shape their own destiny is not ultimately in the hands of the people, but resides in a ‘bevy’ of paternalistic guardians.

Alexander Bickel has formulated this problem as the ‘counter-majoritarian difficulty’ with judicial review.[2] Since representative forms of democracy must involve the legislative enactment and executive enforcement of the will of the people, and since the will of the people is expressed in the majoritarian decisions of their elected representatives, any governmental agency that overrules the outcomes of legislative practices appears not only undemocratic, but fundamentally anti-democratic. Formulated in this traditional way, the problem yields a rather traditional answer: judicial review is a justifiable mechanism for securing the minority rights enshrined in the Constitution against the will of the majority and the vicissitudes of the legislative process. As Jesse Choper puts it: ‘the overriding virtue of and justification for vesting the Court with this awesome power is to guard against governmental infringement of individual liberties secured by the constitution.’[3] Furthermore, according to Choper, the Supreme Court is the proper institutional body for this counter-majoritarian power precisely because it ‘is insulated from political responsibility and unbeholden to self-absorbed and excited majoritarianism. The Court’s aloofness from the political system and the Justices’ lack of dependence for maintenance in office on the popularity of a particular ruling promise an objectivity that elected representatives are not—and should not be—as capable of achieving.’[4]

This tension between judicial review and democracy underlies several recent controversies in the philosophy of law and broader public debates: concerning, for instance, the proper level of judicial ‘activism’ with respect to other branches of government, the proper methods of interpreting statutes and constitutional provisions, and the acceptability of particular Supreme Court decisions made by the Warren, Burger, and Rehnquist courts concerning, for example, school segregation, the right to privacy, and campaign finance regulation.[5] The most insistent contemporary polemics against the Supreme Court have charged it with excessive judicial activism and with the paternalistic imposition of ‘new values’ against the people’s will. It is worth noting that this charge does not have an intrinsic ideological bent. Although the contemporary objection to judicial activism in the name of democracy is closely associated with the Reagan administration—and especially with Robert Bork, one of its nominees to the Court[6]—very similar charges were raised in the beginning of this century by Progressives dismayed at the Court’s Lochner-era decisions overturning labor laws in the name of individual property rights.[7] Interestingly, in light of the 2000 presidential campaign and the U.S. Supreme Court’s effective decision of the electoral result,[8] heated charges of ‘judicial usurpation’ have come, once again, from the left of the ideological spectrum. Jurisprudential theory, meanwhile, has been preoccupied with methodological debates concerning ‘interpretivism’ and ‘noninterpretivism.’[9] The issue here is whether judges should restrict themselves to a ‘strict construction’ of the constitution in terms of the written text or the original intent of the framers, or whether they should go beyond such argumentative resources and adjudicate hard cases on the basis of values and norms that cannot be fairly discovered within the ‘four corners’ of the relevant constitutional provision, the constitution as a whole, and perhaps also its history.[10]

In order to adequately address these questions about the role of the judiciary in the separation of powers and the proper methods of constitutional adjudication, I believe we need to first address more fundamental issues in political theory concerning how best to conceive of constitutionalism and democracy, and the relationship between them. For instance, the traditional justification of judicial review as a counter-majoritarian institution assumes a particular vision of what democracy is and what it should aspire to be. Many of those who either attack or defend the American form of institutionalizing constitutional review apparently share the presumption that, at its core, democracy denotes a certain type of political process: majoritarian self-legislation as expressed through electorally accountable representative bodies. Constitutionalism is then understood as a pre-structuring of this procedure with certain legitimacy-ensuring side constraints that guarantee the non-interference of the state in areas of private life delineated by individual rights.[11]

Recent work in democratic theory has seriously questioned this particular conception of democracy, and recommended the adoption of a ‘deliberative’ conception of democracy. What precisely deliberative democracy means and entails, however, remains hotly contested. The focal problems ranged under this banner run the gamut from how to expand participation in the political marketplace of ideas, to how to encourage the civic virtue necessary for collective development of and reflection on a shared social ethos, to how to publicly ground a just liberal order in the face of competing comprehensive worldviews, to how to design political procedures so that their outcomes can be understood by all as the result of agreement on the basis of the best available reasons.[12] With this shift in the underlying conception of democracy, the objection to judicial review merits another look, especially since jurisprudential theory has not been immune to these developments.

In this paper, I look at four recent theories of constitutional review with special emphasis on the underlying conceptions of constitutionalism and democracy, and their relationship, that they employ.[13] Although I begin with the supposition that a deliberative conception of constitutional democracy is conceptually and normatively superior to more traditional conceptions, the arguments presented here should provide additional support for the deliberative conception by showing how it can productively approach political and jurisprudential puzzles raised by earlier conceptions.[14] John Hart Ely, Michael J. Perry, Ronald Dworkin, and Jürgen Habermas have all offered accounts of judicial review involving richer, more deliberative models of constitutional democracy.[15] I argue that the significant differences between their respective accounts are best understood as arising from different positions taken on two cross-cutting distinctions inherited from Locke and Rousseau. On the one hand, there is a difference concerning the preferred mode of collective decision making processes in a democracy: aggregative versus deliberative. On the other hand, there is a difference concerning how the legitimacy of collective decisions should be understood: substantively versus procedurally.

Ely’s theory responds to the potential for judicial paternalism by combining Locke’s aggregative conception of political processes as the search for the will of all, with Rousseau’s procedural account of legitimacy (Section II); Perry attempts to combine a substantivist account of legitimacy with a deliberative model of political processes (Section III); and, Dworkin combines a (different) substantivist account of legitimacy with a hybrid theory of political processes (Section IV). I argue that, although each has significant contributions and insights, all run into problems because of their underlying theories of constitutional democracy. These arguments lead to some recommended adequacy criteria for a theory of constitutional review (Section V). With these in mind, I argue that the best understanding of constitutional review is yielded by Habermas’s theory of constitutional democracy, for it satisfactorily combines a sufficiently differentiated conception of deliberative processes of democratic opinion and will formation, with a resolutely proceduralist conception of democratic legitimacy in terms of the collective process of citizens’ attempt to institutionalize the general will through law.[16] However, I also argue that there are lingering threats of judicial paternalism in Habermas’s account, particularly with respect to his apparently sanguine approach to extant forms of institutionalizing the function of constitutional review in an unaccountable judiciary (Section VI). I conclude with some recommendations concerning alternative approaches to institutionalizing this function, and some open questions concerning how to properly interpret contemporary constitutions (Section VII).

Before turning to the arguments for judicial review, let me briefly explicate the two cross-cutting distinctions that structure the paper. The first distinction—between aggregation and deliberation—concerns how democratic decision making processes are conceived of: either in terms of the Lockean notion of an aggregation of individual preferences concerning individual interests or the Rousseauian notion of a deliberation about that which is in the equal interest of all. According to the one, the common good of the citizenry can be determined by finding the largest sum of sufficiently identical individual interests. According to the other, the common good can only be determined by collectively testing hypothetical proposals to find those based upon reasons all citizens could reasonable accept.[17] Of course, this distinction between an aggregative and a deliberative model of democratic processes largely follows Rousseau’s distinction between how to determine the will of all versus the general will.[18] The second distinction—between substantivism and proceduralism—concerns how we should think of the legitimacy of democratic decisions: as arising from their permissibility within some antecedently given moral limits, or, as arising simply from the fact that they are the outcome of certain decisions mechanisms that enjoy the presumption of rationality.[19] Whereas Locke argues that governmental decisions are only legitimate if they are not in conflict with the substantive moral constraints of a natural law that is binding even in the state of nature, Rousseau argues that the decisions of a sovereign legislative assembly are legitimate simply because the deliberations have been procedurally structured in a way that all members can understand themselves as subject only to those laws they have given to themselves. In short, this distinction is between substantive and procedural conceptions of democratic legitimacy.[20] Another way to put this distinction is to say that a procedural account of legitimacy sees the outcomes of a decision process as justified simply because the specified conditions of the procedure have been met; a substantive account of legitimacy sees the outcome of a decision process as justified only if that outcome accords with some determinate ideals that are logically independent of the decision procedures employed.[21]

Of course, neither distinction is all or nothing; rather instances should often be characterized as falling along a continuum between the two poles. For this reason, I prefer to think of the distinctions as analytic distinctions between ideal types, keeping in mind that actual examples of democratic theories, and their specific claims, may contain admixtures of both contraries. I do claim, however, that the two distinctions can be analytically useful in seeing the contours of the debates, in particular, by helping to highlight the commitments and entitlements of the various positions. Beyond the inherent imprecision of the distinctions, we also need to keep in mind that the complex theoretical requirements of any political and legal theory—including those of deliberative democracy and constitutional review—will further undermine a futile search for theoretical positions that are entirely contained by a rigid, dichotomous understanding of the distinctions. Thus, for example, a decision process that looks to rationally aggregate already-given individual preferences will require at least some communication between participants, if only for individuals’ private information-eliciting and strategic purposes. And in some cases, rational deliberators may consider the simple aggregative weight of majority preferences as probative, even if not dispositive, to the inherent inferential strength of reasons justifying a proposal.[22]

Likewise, even the most severe adherent to a procedural account of legitimacy must admit that the recommended procedures are recommended because they model or incorporate at least some substantive value, good, norm, or ideal. After all, this substantive component forms the reason for adopting the decision procedure in the first place.[23] Conversely, even the most ambitious attempts to specify a full and complete panoply of substantive principles and values as legitimacy requirements for political decisions will recognize an inexpugnable role for merely procedurally legitimate decision processes in unforeseen or indeterminate cases. Admitting the possibility that one’s preferred substantive theory of legitimacy can’t be used to decide all issues, one must admit that some decisions are legitimate simply as the result of recommended procedures.[24] The non-pure character of these distinctions is especially evident in acceptable theories of democracy. No sensible theory will claim that the legitimacy of any and every state decision or action hangs entirely or exclusively on a matter of either substance or procedure. Substantivists will usually claim that, even though many democratic decisions are justifiable simply because they result from a recommended procedure correctly followed, some determinate substantive content—defined independently of any procedures actually followed—sets constraints on the range of acceptable outcomes of any democratic processes. And if a procedural account of legitimacy is to be more than an arbitrary and unjustifiable stipulation of pointless rules, it must explain the legitimacy-conferring power of its recommended procedures in terms of some principles or ideals the procedures are purported to serve: increasing rationality, ensuring equality, allowing for autonomy, and so on.

Nevertheless, as I hope to show, these two crosscutting distinctions become crucially important when we come to theories of constitutional review and the judicial institutionalization of such review. For, by slighting deliberative forms of decision-making processes, Ely’s theory is led to put forward an impoverished theory of the requisite duties of constitutional review. And by combining a substantivist account of legitimacy with theories that locate deliberative processes about the meaning of constitutional norms exclusively in a politically unaccountable judiciary, Perry’s and Dworkin’s theories are both led to theories of judicial review that pose threats to the basic ideal of popular sovereignty: that citizens should always be able to understand themselves as subject only to those laws that they are themselves the joint authors of. Even Habermas’ general theory of constitutional review, which is convincing on both dimensions of legitimacy and process, does not fully comprehend the institutional ramifications of that general theory and needs to be modified.

2        Ely: Procedural Referees of the Political Marketplace

John Hart Ely has put forward one of the most influential theories of the proper role of judicial review in a constitutional democracy. He begins with the traditional objection to judicial review as the overturning of majority will by a body that is electorally unaccountable. ‘A body that is not elected or otherwise politically responsible in any significant way is telling the people’s elected representatives that they cannot govern as they’d like.’[25] This gives rise to an objection to any actual judicial decisions that overturn enacted laws on the basis of values and ideals that cannot be reasonably discovered within the ‘four corners’ of the Constitution. Such an imposition of values external to constitutional provisions would then seem to be a kind of judicial paternalism.

The most immediate response to the specter of judicial paternalism is to insist that judges overturn statutes only on the basis of a strict ‘clause-bound interpretivism,’[26] and, if the text cannot support a decision, they should simply adopt a passive stance. The attraction of strict interpretivism, combined with a plea for judicial passivism, is that it seems consistent with both the common understanding of adjudication as merely the application of positively enacted laws and the democratic ideal that the legislature is the proper forum for the articulation and justification of the fundamental values that get transformed into legal norms. However, as Ely argues, there are a number of crucial constitutional provisions (such as the equal protection clause of the Fourteenth Amendment) that are open-textured and need to be filled in.[27] Furthermore, the very content of such provisions invites interpretation that reaches beyond their manifest textual content. If so, then the strict clause-bound interpretivist must admit that reliance on the manifest content of the relevant provision would force judges to adopt a noninterpretivist method. ‘The constitutional document itself, the interpretivist’s Bible, contains several provisions whose invitation to look beyond their four corners—whose invitation, if you will, to become at least to that extent a noninterpretivist—cannot be construed away.’[28]

According to Ely, a dilemma now arises. Although strict interpretivism fails by its own standards, none of the proposed noninterpretivist strategies for filling in constitutional provisions are able to escape the charge of a paternalistic imposition of values by an electorally unaccountable body. All of the candidates for discovering extra-textual fundamental values that might guide adjudication result, in the end, in judges applying substantive criteria to the outcomes of legislative processes, processes that are themselves supposed to be the well-spring of the substantive values embedded in legal norms. Whether these fundamental values are found in the judges’ own values, in natural law, in neutral principles, in moral philosophy, in tradition, in current socially-shared values, or in predictions about the future progress of the constitutional project, all of these substantive approaches violate the democratic ideal of legislative self-government: they in effect involve the substitution of extra-legislatively determined values for legislative value decisions.[29]

Rather than advert to the Supreme Court’s role as a protector of substantively guaranteed minority rights, however, Ely proposes a purely proceduralist theory of constitutional adjudication. He accepts that the open-textured nature of central constitutional provisions requires review processes to fill in those provisions. And he accepts that the legitimacy of legally-enforced values can only be secured through the legislative process of representative self-government. Judicial review should therefore secure precisely those procedural conditions necessary to ensure that the legislative process, which gives rise to substantive decisions, is fair and open to all actors in the political marketplace. Courts would then act as referees over the process of the democratic genesis of law, and, in seeking to concretize constitutional provisions, they should adopt a ‘participation-oriented, representation-reinforcing approach to judicial review.’[30] According to Ely, this means that the Supreme Court should especially aim to correct two types of distortions in the political process. First, they should ensure that the legislative process is open to all viewpoints on something close to an equal basis. Thus, especially high scrutiny should be given to legislation that enables electoral winners to block the channels of change by denying access to positions and power to those who are currently not in power. Second, the Supreme Court should be particularly attentive to legislative processes that systematically disadvantage society’s traditional unequals by providing goods only to citizens in the mainstream. ‘Insofar as political officials had chosen to provide or protect X for some people (generally people like themselves), they had better make sure that everyone was being similarly accommodated or be prepared to explain pretty convincingly why not.’[31] Rather than restoring these imbalances on the grounds that the good in question is tied to some fundamental value that all citizens should have, the court should rather ensure that minorities traditionally discriminated against were equally represented in the political process. The court should adjudicate on the basis of the participational goals of broadened access to political processes and equal access to the bounty of representative government.

Ely’s justification for having an unelected body as the referee of legislative processes can now be seen as arising from two commitments: to a Rousseauian conception of purely procedural democratic legitimacy, and to a Lockean conception of the democratic process as a marketplace of competing interests aiming to enact the aggregative will of all. Since the legitimacy of positive law is based not upon the substantive content of its directives but upon the procedural conditions of its genesis, it becomes particularly important to ensure that those conditions are fairly structured and as open as possible to all citizens. Ely’s proceduralist commitment is supported by rejecting a reading of the U.S. Constitution as a statement of fundamental values or moral commitments, whether static or evolving. Rather, according to Ely, a proper reading of the Constitution and the underlying premises of the American system of representative government

 

will reveal ... that in fact the selection and accommodation of substantive values is left almost entirely to the political process and instead the document is overwhelmingly concerned, on the one hand, with procedural fairness in the resolution of individual disputes (process writ small), and on the other, ... with ensuring broad participation in the processes and distributions of government.[32]

 

Because legitimacy hangs on fair political procedures, some institutional oversight is needed. But since, with Locke, Ely conceives of the political process as a marketplace of competing, self-interested parties, fairness can only be ensured on the supposition of an impartial, disinterested third party empowered to adjudicate disputes. Thus the oversight of the procedural conditions of the political process cannot be entrusted to one of the sides to the dispute—namely, the legislature. Rather, an independent, unelected judiciary is institutionally well-situated to play the required referee role in a dispute between citizens and their representatives. Ely’s Lockean conception of political democracy as a kind of negotiation amongst strategically acting individuals and groups, simply trying to maximize their pre-political interests, thus plays a central role in his theory of judicial review.

 

The approach to constitutional adjudication recommended here is akin to what might be called an ‘antitrust’ as opposed to a ‘regulatory’ orientation to economic affairs—rather than dictate substantive results it intervenes only when the ‘market,’ in our case the political market, is systematically malfunctioning. (A referee analogy is also not far off: the referee is to intervene only when one team is gaining unfair advantage, not because the ‘wrong’ team has scored.)[33]

 

Judicial review of legislation is thus justified, not because of a belief in the special competence of judges to be able to discern, and paternalistically enforce, the moral truth, but precisely because they are unelected, and so institutionally situated as disinterested parties in procedural disputes between the electors and the elected.[34]

Difficulties with Ely

Given his Rousseauian commitment to procedural legitimacy and his Lockean commitment to politics as the aggregation of private interests, it is not surprising that Ely’s theory has been attacked both by those who reject a Rousseauian account of legitimacy as insufficient for explaining the moral content of politically enacted rights, and by those who reject the Lockean conception of representative democracy as insufficient for explaining the deliberative, intersubjective character of political decision making.

First there are those who criticize Ely’s theory for its insufficient attention to other individual rights besides those that can be plausibly defended in terms of their direct relevance to the political process.[35] If the role of constitutional review is confined solely to refereeing the political process, then it seems that the Supreme Court will no longer have much claim as a defender of non-political individual civil and social rights. Ely’s political proceduralism seems to leave no room for a claim that the legitimacy of any democratically enacted statute is called into question if it infringes on certain inalienable moral rights of individuals; rights that should be guaranteed by a counter-majoritarian judiciary employing substantivist criteria as checks on the rightness of any given outcome. Ely himself considers this objection, and rejects it on the grounds that individual liberties are sufficiently secured by the underlying American theory of government:

 

I went through a period of worrying that the orientation here recommended might mean less protection for civil liberties. ... Reflection has convinced me that just the opposite is true, that freedoms are more secure to the extent that they find foundation in the theory that supports our entire government, rather than gaining protection because the judge deciding the case thinks they’re important.[36]

 

But Ely owes us something more here about what that theory of government is if his response is to remain more than mere hand waving. For one might think, like Rousseau, that true self-government is impossible unless all citizens alienate all of their rights before having some of them bestowed back upon them by a benevolent sovereign power. In fact, one of the few indications of Ely’s underlying theory of democracy—proffered in a footnote—should give pause to those who are looking for a strong defense of individual liberties. ‘I have suggested that the appeal of democracy can be best understood in terms of its connections with the philosophical tradition of utilitarianism. ... Since nothing in the ensuing analysis depends on this claim, it is omitted here.’[37] I think, to the contrary, that a fair amount does in fact so depend. An Achilles’ heel of both republicanism and utilitarianism is their difficulty in giving sufficiently deontological justifications for individual liberties. As Ely is unwilling to go into any detail concerning his account of democratic legitimacy, a quick dismissal of a hyperbolically constructed thought experiment at the end of his book will do little to assuage traditional liberal worries here.[38]

Another way to see how important Ely’s underlying assumptions about democracy are to his theory of judicial review is to consider his conception of representative processes themselves. Recall that he models collective decision making as a kind of political marketplace, whereby individual members register their de facto preferences concerning the likely impact a decision will have on their private interests. On this account, voting—whether by citizens in elections or representatives in legislating—is an expression simply of an individual’s belief concerning the best way to secure his or her own good in the light of his or her contingently given preferences.[39] In contrast, a deliberative conception of democracy insists, with Rousseau, that political processes should be oriented towards shaping collective arrangements that will be in accord with the general will, not merely the will of all. Here voting is understood as a way of individuals’ expressing their current convictions on which proposed governmental action will be the best way to secure that which is in the equal interest of all, or at least that which can be reasonably expected to realize a generalizable individual interest. On this account, the process of collectively deciding upon how we are going to live our lives together under government requires debate and the giving of reasons—reasons which all could potentially accept for themselves. In this process of deliberation, citizens themselves may in fact alter their pre-political preferences to bring them into line with the requirements for living with others. In this sense, voting is not a mere registration of preferences, but is a specific mechanism adopted by mutually deliberating actors in order to reach some decision under time, knowledge, and coordination constraints. Voting is a way of temporarily calling a halt to deliberations under pressing needs for action.[40] Even if this description seems overly idealistic for a great number of routine political decisions, it still seems that the ideal of democracy includes the notion that citizens can only understand the laws as products of their own free will if those decisions have been reached on the basis of thoughtful deliberation and opinion formation. If this notion of a ‘republic of reasons’[41] is a crucial part of the democratic ideal, then the duties of Ely’s judicial procedural referees will extend further than merely ensuring against ‘antitrust violations’ of the political marketplace.

Ely’s procedural justification of judicial review is attractive precisely because it does not rely upon the superior insight of judges into matters of moral principle or truth.[42] It is thus not subject to the skepticism concerning judicial moral competence which, combined with an insistence on the democratic principle of popular sovereignty, led to worries about judicial paternalism in the first place. However, without some fuller account of democratic legitimacy, Ely’s reliance on antitrust-style procedural legitimacy leads to liberal concerns about the security of non-political, individual civil and social rights. In addition, Ely’s purely Lockean account of democratic processes in terms of pre-political preference aggregation ignores the intersubjective deliberation about ends and responsiveness to public reasons that are ideally a part of democratic self-rule. Below I will argue that Habermas’s account of judicial review fruitfully carries forward Ely’s insights by developing an enriched notion of procedural legitimacy and offering a more differentiated account of political processes. Before this, I look at two theories of judicial review grounded in (different) substantive conceptions of legitimacy in order to point out the perils of such theories under conditions of pluralism.

3       Perry: Keepers of the Substantive Flame of American Exceptionalism

Like Ely, Michael J. Perry starts his theory of judicial review with a recognition of the indeterminacy of constitutional provisions. But from there, Perry develops a defense of judicial review that is committed to the Rousseauian notion of deliberative forms of decision making and the Lockean notion of substantive moral constraints on the legitimacy of outcomes. Perry starts from the notion that political discourse is an attempt to come to a collective, ethical self-understanding about our moral and religious aspirations. In this process of becoming clear about who we are as Americans, the U.S. Constitution takes on the dual roles of a founding corner-stone of our identity, and of the guiding beacon that can lead us to a realization of deep moral truths. However, since the provisions of this identity-constitutive document are indeterminate with respect to their application to specific situations and with respect to the precise contours of their moral content, they need to be specified more completely.

Who should carry out such specifications? Perry’s answer is unequivocal: a politically unaccountable judiciary. The members of the legislature are ill suited to carrying out the subtle discussions needed to discern the objective hierarchy of values in a truly dialogic manner, as their capacities for judgment are impaired by the ever-pressing task of getting reelected. ‘A [legislative] regime in which incumbency is (inevitably?) a fundamental value seems often ill suited, in a politically heterogeneous society like the United States, to a truly deliberative, dialogic specification of the indeterminate constitutional norms.’[43] Since ‘specifications of indeterminate constitutional directives are a species of political-moral judgment [and] ... a dialogic capacity is an important element of the capacity for good judgment,’[44] we need a coterie of guardians of the moral truths of our society who can engage in subtle dialogical interchanges with other judges. It is of paramount importance that they have been entrusted with these duties on the basis of their special capacities for good judgment, and that they be able to discuss, amongst themselves (through their written opinions) the reasons for their decisions.[45]

Thus, for example, Perry justifies judicial review in human rights cases on the basis of those objective values discovered in the history of the American process of moral self-development. Since legislators are beholden to conventional convictions and dogma through the electoral process, they ‘are not well suited to deal with such issues in a way that is faithful to the notion of moral evolution or, therefore, to our religious understanding of ourselves. Those institutions, when they finally confront such issues at all tend simply to rely on established moral conventions and to refuse to see in such issues occasions for moral reevaluation and possible moral growth.’[46] In contrast, noninterpretive judicial review, particularly on such important issues as human rights, enables us ‘as a people, to keep faith with ... our religious understanding of ourselves as a people committed to struggle incessantly to see beyond, and then to live beyond, the imperfections of whatever happens at the moment to be the established moral convictions.’[47] If there can be right answers to some moral questions, especially those concerning fundamental human rights—and there can be if we accept Perry’s arguments for an ineliminably religious natural law theory of morality[48]—then ‘the politically insulated federal judiciary is more likely, when the human rights issue is a deeply controversial one, to move us in the direction of a right answer ... than is the political process left to its own devices, which tends to resolve such issues by reflexive, mechanical reference to established moral conventions.’[49] The basic purpose and justification for the institution of constitutional review is then to serve as a beacon and indicator of the exceptional moral truths that were discovered at the start of our collective religious-political learning process; judicial review is further justified through the superior capacities for moral discernment and dialogue found in a politically-insulated judiciary.

Difficulties with Perry

Although a number of criticisms may be raised against this justification for judicial review, I want to focus on two that arise from Perry’s institutionally- and ethically-constricted account of the public use of practical reason. The first concerns his theory of political processes relevant to judicial review and the second his substantivist account of political legitimacy in Aristotelian terms. As should be clear, Perry’s theory of judicially instituted constitutional review, in insisting that there are certain moral truths that any legitimate governmental directive has to respect, has been driven to recommending precisely the kind of judicial paternalism that Hand and Ely were both worried about. From his account of the legitimacy conditions of a constitution and his argument about the superior dialogical and moral capacities of the judiciary, Perry argues that noninterpretive review—review, that is, that depends on the discernment and specification of extra-constitutional moral content—should be entrusted to an unelected body of guardians who will enforce our own best moral interests even over our own objections as expressed through the legislature. However, the point of endorsing either an extremely restrained approach to adjudication, as Judge Hand recommends, or a proceduralist account of the role of the judiciary as a referee in the political marketplace as Ely does, is precisely to capture the ideal of popular sovereignty and its rejection of forms of political paternalism. If citizens are to be able to understand themselves as both free and equal under law, they must be able to understand the state’s laws as laws they have given to themselves—not as laws that have been imposed upon them by a wise council of tutors in moral truth.[50]

The first objection to Perry’s specific argument for noninterpretive judicial review is that it is driven in part by a false dichotomy concerning the location of political judgment: it must either be found in the legislature or the judiciary. But this overlooks those broader, uncentered public spheres and associational fora in which citizens discuss and debate, and form the opinions that then are (or ought to be) fed into the formally organized channels of political organizations. The false dichotomy ignores, in other words, precisely those sites of political dialogue and judgment that theories of deliberative democracy have focused on. Even if we accept a substantive account of political legitimacy (though I think we should not), there are at least two potential further locales of that form of political judgment Perry believes is required for constitutional specification: the executive branch and the broader non-governmental public sphere. With regard to the executive, it is at least more accountable to the electorate than the judiciary, and so concerns about paternalistic review might favor some forms of executive review, though Perry doesn’t consider such alternatives to the status quo. He also ignores the broader public as a potential source of contributions to constitutional dialogue and judgment.

I should note that Perry does explicitly recognize the paternalistic objection to the institutionalization of constitutional review in a politically unaccountable judiciary, and he specifically recommends two institutional reforms to the United States’ arrangements that would make judicial review ‘more responsive to ‘We the people’ now living, who, after all, unlike our dead political ancestors, are supposed to be politically sovereign’: term rather than life appointments for federal judges,[51] and adoption of a mechanism akin to the Canadian ‘notwithstanding clause’ of Section 33 of the Canadian Charter of Rights and Freedoms.[52] While the first would make the judiciary, especially the Supreme Court, more accountable to the electorate through more frequent judicial appointments and their attendant confirmation hearings, the second would facilitate increased political and moral dialogue between the courts and other branches of government and would increase the power of non-judicial branches concerning issues of constitutional fundamentals. ‘Were it adopted, the Canadian innovation would present the people—or the people’s political representatives in the Congress and the White House—with more rather than fewer opportunities to exercise their constitutional and moral responsibility.’[53] Perry argues that such an arrangement becomes compelling precisely when the crucial premise of superior judicial competence looses its cogency; that is when ‘we are skeptical both about the capacity of ordinary politics to specify constitutional indeterminacy and about the capacity of many of our judges and justices to do so.’[54]

Yet, again, the comparative evaluation of capacities for moral discernment that Perry makes does not grapple with the possibility of citizens themselves exercising these powers in non-governmental public fora, precisely the kind of fora one might look to when one centers an account of democracy on a Rousseauian account of procedures of public deliberation aiming to ensure true popular sovereignty. His rejection of the possibility of extra-governmental sources of moral judgment forces a false dichotomy in identifying the proper location of paternalistic guidance: the choice for Perry seems only between locating it in the legislature or the judiciary. For all of his insistence on the importance of dialogue and discussion on fundamental constitutional issues, this debate is to be institutionally restricted to those who can be expected to have the requisite ethical capacities of judgment: judges, law professors, and lawyers. Noting that the specific capacity for contextually-sensitive moral and political judgment is possessed, for Perry, only by those who have a virtuous character,[55] one wonders what could be the empirical support for the claim he needs to vindicate that the judiciary houses such persons to a greater degree than the public at large? Once we expand the purview of those we might want to compare for their discernment capacities, the institutional argument that legislative debate is subject to distorting pressures is alone insufficient to justify confidence in the judiciary’s empirical claim to superior moral competence

This is related to the second problem with Perry’s account: his account of dialogic, practical reasoning is ethically constricted, specifically to ethical-political explorations concerning the proper contours of the good life for us, given who we are in the light of our particular, constitutive moral and religious traditions. Like other neo-Aristotelian theories of practical reason, Perry’s has tendencies towards a partialistic and potentially exclusionary perfectionism that does not sit well with the manifest ethical pluralism of contemporary societies. Since Hobbes, political theory has struggled to come to terms with the wars of religion and the increasing ethical pluralism of heterogeneous populations in modern nation-states and internationally. The prospect for justifying basic political institutions and decisions in terms of a substantive ethos specific to one or more systems of revealed religion is quite limited in contemporary pluralistic contexts. Whether one simply starts with the fact of a plurality of incompatible and warring comprehensive doctrines as Hobbes does,[56] or tries to explain the origins of pluralism as the outcome of the burdens of judgment facing reasonable and rational persons attempting to come to an understanding about how to live together in a free and open society as Rawls does,[57] hopes for agreement on some notion of a religiously-based American exceptionalism seem quixotic at best. In fact, Perry’s protestations that both the American political community at large and the judiciary are and should be pluralistic are reminiscent of Rousseau’s call for (limited) religious tolerance at the end of The Social Contract. As long as we all accept the ‘civil profession of faith’—after all, we shall be banished if we do not accept it, or killed if we renege on an earlier acceptance—then we should be ‘tolerant’ of the different particular dogmas that interpret and specify that commonly accepted faith.[58] Perry’s specific brand of constitutional originalism indicates that this communal faith need not be—and is not for Americans—a faith in one revealed religion and with a specific doctrinaire interpretation. Rather, it is a civil profession of faith in the original constitutional framers’ insights into moral truth as refracted through the lens of various Protestant doctrines. Nevertheless, insofar as all citizens do not share this perfectionist vision of ethico-religious truth, any unaccountable body that decides substantive issues in accordance with that vision of truth will be even more disconnected from the practices of inclusive self-government precisely for relying upon it in justifying their decisions. Perry’s ideal of the practical reasoning involved in constitutional review thus exacerbates, rather than solves, the democratic deficit of judicial review.[59]

In sum, the account of practical reason underlying Perry’s justification of judicial review locates its exercise solely in those officials populating formal governmental bodies and presupposes a dedifferentiated picture of moral reason as a perfectionistic process of ethico-religious self-clarification. Although he is sensitive to the normative claims of popular sovereignty and deliberative politics, his theory of judicial review exacerbates its paternalist taint by combining a substantive account of political legitimacy with an account of political processes focused exclusively on formal governmental bodies and the elites that populate them.

4       Dworkin: Guardians of the Moral Law in the Forum of Principle

Like Ely and Perry, Dworkin develops his theory of adjudication in response to the indeterminacy of constitutional provisions and the inadequacy of strict interpretivist responses to the problem. According to Dworkin, without some theoretical guidance, the crucial judicial decision concerning what level of generality to adopt in reading abstract, open-textured provisions that clearly contain moral content is left without an anchor. Of course judges must attend to the specific language of the provision, but, in addition, they must decide particular cases consistently with precedent and must uniformly apply a principle invoked in one case to other cases involving similar issues. Even so, these three counterweights to arbitrary specification—that is, of text, precedent, and consistency—are jointly insufficient.[60] In addition, a judge constrained by what Dworkin calls ‘integrity’ must also be guided by moral principles.[61] In a sense, the ‘moral reading’ of the United States Constitution that Dworkin favors follows from the very constraint of the text itself: the abstract clauses of, for instance, the Bill of Rights, ‘must be understood in the way that their language most naturally suggests: they refer to abstract moral principles and incorporate these by reference, as limits on government’s power.’[62] Part of a consistent moral reading of the Constitution, according to Dworkin, involves a distinction between principles protecting individual rights, understood as deontic trumps, and governmental policies intended to further the realization of particular goods or values. In contrast to Perry’s theory, the constitution does not simply enshrine a particular constellation of religious values that must be weighed against each other and transitively ordered in each case by judges. Rather, individual rights have lexical priority in political arguments: they express principled considerations that cannot be simply weighed on the same level as various competing values, goods, policy goals, and the like.[63]

Even if we accept that the Constitution must be read morally, at least in part, and that this involves understanding rights deontologically, this does not yet guide the judge in deciding how to formulate these abstract moral principles and rights. At this point, Dworkin puts forward his preferred conception of democracy. This conception combines a hybrid theory of democratic processes drawing on both Locke and Rousseau, with a Lockean substantivist account of democratic legitimacy.

He begins by distinguishing two kinds of collective action: statistical and communal. ‘Collective action is statistical when what the group does is only a matter of some function, rough or specific, of what the individual members of the group do on their own, that is, with no sense of doing something as a group.’[64] ‘Collective action is communal, however, when it cannot be reduced just to some statistical function of individual action, when it presupposes a special, distinct, collective agency. It is a matter of individuals acting together in a way that merges their separate actions into a further, unified, act that is together theirs.[65] These action types clearly correspond to the distinction between the will of all and the general will: Lockean democracy through aggregation aims at collective actions that satisfy the pre-political preferences of individuals taken as individuals, while Rousseauian democracy through deliberation aims at collective actions that satisfy the requirements of the people acting together as citizens.[66]

Dworkin then argues that, if the communal conception of collective self-determination is the right characterization of democracy, then any adequate democratic regime must meet certain conditions. These conditions will then furnish substantive checks on the legitimacy of the outcomes of collective decisions. In particular, in order to treat each member of the collectivity as an equal moral member, each member must be afforded ‘a part in any collective decision, a stake in it, and independence from it.’[67] Having a part in collective decisions means that each citizen must have an opportunity to influence those decisions in a way in that does not systematically discriminate against him or her on the basis of morally arbitrary qualities. This condition forms the justification for political procedures concerning voting and representation, and for the expressive and associational liberties required to actualize them. Having a stake in collective decisions means that the ‘community must express some bona fide conception of equal concern for the interests of all members, which means that political decisions that affect the distribution of wealth, benefits, and burdens must be consistent with equal concern for all.’[68] This condition prohibits the community from disregarding, in their decisions, the differential impact that a proposed policy might have for the needs and interests of all of its members. It does not require that distributions be strictly egalitarian; rather it insists that the interests of all be fairly considered in setting up distributive arrangements.[69] Finally, the condition of independence sets limits upon the scope of collective powers over individuals’ lives, commonly through individual liberties against state infringement on how citizens choose to realize their individual conception of the good life.

Dworkin’s next move is to argue that democracy should be properly understood as a form of communal, not statistical, collective action. His basic idea is that all of the arguments for a purely statistical notion of democratic action presuppose the communal conception of collective action, and so presuppose that the conditions of moral membership in a collective venture have been satisfied.[70] Thus, the core of democratic self-government cannot be thought of as simply rule by a majority, since this is a statistical notion of collective action. Rather, majority rule must be structured so that it meets the principled conditions of communal collective action, and this is precisely the function of constitutional structures that set limits upon how members of a political community may be treated. Dworkin’s theory thus

 

denies that it is a defining goal of democracy that collective decisions always or normally be those that a majority or plurality of citizens would favor if fully informed and rational. It takes the defining aim of democracy to be a different one: that collective decisions be made by political institutions whose structure, composition, and practices treat all members of the community, as individuals, with equal concern and respect.[71]

 

In this way, Dworkin attempts to redefine the concept of democracy by splitting the notion of democratic political processes into two kinds along the lines of Locke and Rousseau—namely, statistical and communal—and insisting upon a substantive account of constitutional legitimacy along Lockean, natural law lines.[72] His specific moral reading of the constitution and his conception of democracy draws from the Rousseauian notion of collective action certain substantive conditions that can be applied to the outcomes of any political procedure in order to test for their legitimacy.

 

Democracy means government subject to conditions—we might call these the ‘democratic’ conditions—of equal status for all citizens. When majoritarian institutions provide and respect the democratic conditions, then the verdicts of these institutions should be accepted by everyone for that reason. But when they do not, or when their provision or respect is defective, there can be no objection, in the name of democracy, to procedures that protect and respect them better.[73]

 

For Dworkin, then, democracy properly understood is achieved whenever the substantive outcomes of a political process—whatever those processes happen to be and however they are institutionalized—are legitimate in light of the ideal of equal status for citizens.

Given this singular redefinition of democracy, the question remains concerning its relation to the judicial review of legislation. Dworkin’s answer is predictable: there should be a division of labor between those governmental bodies concerned with issues that are appropriate to statistical collective action and those concerned with ensuring the legitimacy conditions of communal collective action. Since the legitimacy conditions concern individual rights and fundamental moral principles, they should be handled by an independent judiciary that has the requisite competences, and lacks the distorting pressures of power blocs and private interests. Like Perry, Dworkin believes that legislatures cannot fill this role, since their debates are rarely of high quality with respect to fundamental moral principles, their decisions are often substantially influenced by power blocs, and they usually aim at compromises that undermine the deontic quality of principles. The institutional solution is to entrust guardianship of the ‘democratic’ conditions to an insulated ‘forum of principle’ within which judges can draw on their special legal competence for integrating all of the relevant considerations needed for legitimate decisions.[74] According to the persuasive redefinition of democracy as whatever institutional arrangements best fulfill the legitimacy conditions of natural-law-like fundamental axioms concerning the equal moral status of individuals, judicial review does not compromise democracy; to the contrary, it enhances democracy. ‘Individual citizens can in fact exercise the moral responsibilities of citizenship better when final decisions involving constitutional values are removed from ordinary politics and assigned to courts, whose decisions are meant to turn on principle, not on the weight of numbers or the balance of political influence.’[75]

To this defense of judicial review based upon a skeptical portrayal of ‘ordinary politics’ and a claim to special judicial competence, Dworkin adds an interesting empirical argument. His claim here is that, when an issue becomes an adjudicated constitutional issue, an issue of fundamental political morality, rather than simply an interest-based claim to be decided through bargaining and compromise, the quality of the public debate on that issue increases. ‘When a constitutional issue has been decided by the Supreme Court, and is important enough so that it can be expected to be elaborated, expanded, contracted, or even reversed, by future decisions, a sustained national debate begins, in newspapers and other media, in law schools and classrooms, in public meetings and around dinner tables.’[76] This debate will be of much better quality than what could be produced through the legislative process on its own, and will have the participatory benefits of involving a larger percentage of the citizens in public deliberations sensitive to the complexity of the considerations involved.[77]

Difficulties with Dworkin

As an empirical counterfactual, the claim that the quality of public debate is improved by judicial review is difficult to evaluate. Jeremy Waldron suggests that the quality of debate over a controversial issue like abortion has in fact been of equally high caliber in countries like Britain and New Zealand where it cannot even become a constitutional issue. He then suggests why this might be so: ‘It is sometimes liberating to be able to discuss issues like abortion directly, on the principles that ought to be engaged, rather than having to scramble around constructing those principles out of the scraps of some sacred text, in a tendentious exercise of constitutional calligraphy.’[78] But even if Dworkin’s empirical claim were correct, its not clear that the argument in fact supports judicial review. Rather, it would certainly lend weight to some institutionalized form of constitutional review, but this could take place in the legislature, in the executive, in an independent constitutional court, and so on. In other words, it leaves open the possibility that constitutional review could be carried out by a governmental body that was more accountable to citizens than the United States Supreme Court is.

In order to justify placing the function of constitutional review in a politically unaccountable judiciary, Dworkin needs the additional claim of a special judicial competence that outstrips that of ordinary citizens when it comes to constitutional issues. His books such as Taking Rights Seriously and Law’s Empire can be read as sustained attempts to vindicate this claim. The portrayal of the tasks of adjudication as, literally, Herculean certainly lends credibility to the notion that only judges can carry out the complex tasks of ensuring sensitivity to the often technical language of statutes, the consistency of principle application across disparate cases, the coherence of various historical precedents, and so on. But all this is, I think, somewhat misleading, since the issue here concerns—according to Dworkin’s argument from the improvement of public debate—only whether an independent judiciary has a special competency in basic moral-political reasoning, and not a special competency in specifically legal consistency and integrity. Is it really true that only judges have the requisite competence to detect and interpret the basic moral principles that underlie the conditions we set on our collective political arrangements, and that this competence should be grounds for allowing them to not only set the basic terms and limits of subsequent debate, but also to decide the issue for a significant period of time?

In a sense, the problem here is that Dworkin has tried to split the difference between Locke and Rousseau on the character of democratic processes. On the one hand, he has assigned the deliberative argument over constitutional essentials—the debate over the proper legal structure of the general will—to the judiciary (under the tutelage of moral philosophers). On the other, he has assigned the self-interested bargaining—the struggle of individual actors and social powers to secure their own interests—to the legislature.[79] This division of labor is only warranted, however, on a rather pessimistic characterization of citizens’ capacities to engage in arguments over principles. The veracity of this pessimism, and Dworkin’s faith in judicial competence, are indeed empirical matters. But it is worth asking whether judges are any less inclined to attempt to hide their own biases, ideological preferences, and interests behind a screen of principle and legalese than ordinary citizens in ordinary political dialogue.[80]

Besides these empirical questions concerning the character of democratic processes, there is a deeper, normative problem that Dworkin’s defense leads to: once again, the specter of judicial paternalism. In opting for a substantivist account of legitimacy, Dworkin has severed the internal connection between legal legitimacy and the procedural conditions of a law’s genesis that Rousseau, and Kant following him, argued for.[81] If fundamental moral principles are to be detected and specified by an independent judiciary, with guidance from preferred moral philosophers, then the principle of popular sovereignty is severely compromised. For popular sovereignty entails that citizens are only free in a form of political association if they can somehow understand themselves as the authors of the laws that structure their interactions, as both sovereign citizens and legal subjects at the same time. Dworkin seems to be saying, in effect, that the people are allowed to be sovereign with respect to policy decisions, but when it comes to principles and rights, they must simply submit to the paternalistic imposition of the ‘conditions of democracy’ by an unaccountable Hercules. Under this division of labor, the moral competence of citizens does not and cannot extend to collective decisions concerning the conditions under which they are going to regulate their lives together. Even if Dworkin prefers to retain the label of ‘democracy’ only for those regimes that substantively guarantee the natural rights of equal respect and concern for citizens, something important has been lost when popular sovereignty does not extend to decisions concerning the very conditions under which we collectively act as a political community.[82]

Another way to put this same point is to focus on what showing equal respect to each individual citizen as a moral member of the community means. For if there are in fact disagreements amongst citizens not only about values and policies, but also about how to determine and specify the fundamental moral principles that are to structure their political interactions, then it seems perverse to shut them out of the debate over those constitutional essentials. Of course Dworkin does not wholly deny citizen input into constitutional debates: citizens justifiably have the political process of amendment open to them, and there is indirect representative control over court appointments. Nevertheless, on the much more frequent—and usually more contested[83]—questions of how to specify the constitutional provisions and amendments, treating citizens as autonomous moral agents seems to require an institutional openness to the full spectrum of information, reasons, and arguments that they might think relevant to that task of specification. Effectively shutting them out of those debates requires the presumption that judges know better than citizens do themselves how to live their lives as free and autonomous citizens in a form of political association under law.

5       Criteria for an Adequate Theory of Judicial Review in Deliberative Democracy

I have outlined three theories of judicial review with special attention the answers each gives to two questions: How are the laws passed in a constitutional democracy legitimate? and, What is the proper characterization of democratic political processes? Ely combines a proceduralist account of legitimacy with a market-modeled account of political processes. Perry combines an ethically substantivist account of legitimacy with a fallen image of legislative processes inspired by a neo-Aristotelian account of deliberative practical wisdom. Dworkin combines a natural-law style substantivism with a two-track account of political processes: majoritarian processes to determine will of all policies and judicial processes to determine general will principles. I have tried to show how each fails, in important ways, to account for salient features of democratic legitimacy and democratic process. From this, I suggest that the following six criteria must be met by an adequate account of constitutional review, within the theoretical framework of deliberative democracy.

Since substantivist accounts of democratic legitimacy appear to lead to worries about judicial paternalism—especially under conditions of value pluralism—it appears that we need, first, a thoroughly procedural account of how democratic politics can warrant the legitimacy of decisions; one without recourse to contestable substantivist checks such as natural law or the objective hierarchy of values. Second, we would still like to be able to account for the deontic character of rights ascribed to persons, without going beyond a procedural account of legitimacy. Thus approaches that either reduce rights to values that must be weighed against each other and transitively ordered in each new conflicting situation, or that rely on deeply contested metaphysical claims about the natural grounds of human rights will both be unacceptable. Third, we would like this proceduralist account to be able to defend the individual civil liberties that make private autonomy possible, instead of being limited only to those political rights of participation that make public autonomy possible.

With respect to democratic processes, it seems that we need, fourth, an account of the diversity of forms of practical reason. Since each of the accounts of judicial review runs into problems when it ignores the variety of types of political interaction and reason-giving, we would like an account of democratic processes to be able to comprehend and distinguish between Lockean aggregation, Rousseauian consensus, ethical-political self-clarification, and bargaining. Fifth, democratic theory needs to pay more attention to the distinctions between, and the types of interaction amongst, different public fora. At the very least, we need a good account of the difference between formally organized and governmentally institutionalized public arenas of debate (paradigmatically legislative bodies), and, informal, non-institutionalized and heterogeneous arenas of debate (what is currently being called ‘civil society’). Finally, we need clearer considerations of the different areas and relative levels of competence on the part of judges, legislators, and non-official public actors, in order to be able to determine whether or not constitutional review should be institutionalized in an independent judiciary. I turn now to Jürgen Habermas’s theory of democratic constitutional review, since it promises to fulfill many of these criteria.

1       Habermas: Guardians of the Conditions of Procedural Legitimacy

Habermas’s theory of judicial review—a theory, I will argue, that fulfills many of the above-mentioned criteria—begins not from jurisprudential considerations, but from a combined normative and sociological theory of constitutional democracy oriented towards procedurally-structured participatory deliberation. I begin by explicating the Rousseauian conception of procedural legitimacy that underlies Habermas’s defense of constitutional democracy and justifies the function of constitutional review; then I turn to his differentiated account of democratic processes that prioritizes deliberations aimed at the Rousseauian general will but that does not deny the import or place of Lockean aggregation, ethical-political self-clarification, and bargaining; and finally I examine issues concerning the institutionalization of constitutional review in an independent judiciary and the adjudicative scope of its mandate in order to assess the extent to which there is lingering paternalism even in his account.

Democratic Legitimacy

According to Habermas, the fundamental normative idea of democracy can be modeled in a principle of democratic legitimacy: ‘only those statutes may claim legitimacy that can meet with the assent (Zustimmung) of all citizens in a discursive process of legislation that in turn has been legally constituted.’[84] This principle of democratic legitimacy results from the ‘interpenetration’ of the specific form of law as a medium for action coordination[85] and the more general requirements for the justification of norms of action.[86] The crux of this democratic principle for the purpose of this paper is that it points to a solely procedural test for the legitimacy of laws: statutory legitimacy hangs solely on whether a law has been enacted in the correct way, not on whether it fulfills some antecedently specified substantive normative criteria for goodness or rightness.

If we should next ask why democratic procedures alone grant legitimacy, Habermas’s answer is that they warrant the expectation of rational outcomes.

 

The democratic procedure for the production of law evidently forms the only postmetaphysical source of legitimacy [for legal rules]. But what provides this procedure with its legitimating force? ... Democratic procedure makes it possible for issues and contributions, information and reasons to float freely; it secures a discursive character for political will-formation; and it thereby grounds the fallibilist assumption that results issuing from proper procedure are more or less reasonable.[87]

 

So a Rousseauian conception of popular sovereignty—as reasonable and open debate, deliberation, and decision concerning issues of public interest, rather than the Lockean notion of democracy as mere electoral accountability of representatives—is at the heart of Habermas’s procedural account of democratic legitimacy. In this way, the basic ideal of democracy—collective self-determination—can be secured: if the legally-structured political community ‘constitutes itself on the basis of a discursively achieved agreement,’ then the resulting regime and its positive laws are legitimate according to ‘the idea of self-determination: [namely, that] citizens should always be able to understand themselves also as the authors of the law to which they are subject as addressees.’[88] The crucial difference, however, is that while Rousseau’s conception of legitimacy rather unrealistically required a full assembly of all citizens while relying only on the small size and extreme homogeneity of the population to generate reasonable laws, Habermas argues that a much more exacting specification of procedural requirements is required under conditions of value pluralism and in large, complex, modern nation-states in order to underwrite the expectation of generating reasonable laws.

In order to specify exactly what types of procedures could carry this weight of legitimation, Habermas explicates the pragmatic presuppositions of legally-constituted democracy, drawing on both the requirements of the form of law and the normative ideal of self-government. This leads to a defense of constitutionally structured democracy whereby a system of five incompletely specified categories of rights indicate what types of rights individuals would need to legally grant each other if they wish to legitimately regulate their interactions through the medium of law. First, as individuals mediating their horizontal relationships through law who recognize each other as free and equal, individuals would need to grant each other 1) rights to the greatest amount of equal subjective liberties, 2) equal membership rights in the legal community, and 3) equal rights to the legal protection and actionability of their rights. These three categories of rights can, according to Habermas, be procedurally justified in terms of the meaning of legality and of the pragmatic presuppositions of raising and defending normative validity claims. The basic idea is that without such rights, any individual’s assent to constitutional or statutory enactments could not be assumed to rest on each individual’s reasoned acceptance, but might be a result of distortion and exclusion through direct or indirect forms of force, coercion, fraud, and so on. The category of 4) equal rights to participation in processes of political opinion- and will-formation then follows from the requirement that members of the legal community also be authors of the laws they are subject to. With this fourth category of rights, members first form themselves into a political community that must interpret and elaborate the more specific, ‘saturated’ rights that will fill in the abstract categories of rights. Finally, the equal status of these four categories of rights can only be made more than a merely formal guarantee if citizens also ensure that all have an equal opportunity to utilize such rights through 5) ‘basic rights to the provision of living conditions that are socially, technologically, and ecologically safeguarded, insofar as the circumstances make this necessary.’[89] Habermas claims that this last category of rights is only instrumentally and contextually justified: they are those rights necessary for ensuring to citizens equal opportunities for utilizing their other civil, membership, legal, and political rights, yet their sufficient provision is only required under contingent social conditions where the mere formal assurances of equal civil and political rights cannot alone secure an equal opportunity for their use among all citizens.[90]

It is important to see that this ambitious combination of political philosophy, legal analysis, and communicative pragmatics promises to provide a robust defense of the types of individual civil liberties apparently missing from Ely’s contrasting procedural account of democratic legitimacy.[91] Recall Ely’s inadequate responses to the objection that his defense of judicial review would allow the infringement of individual rights by duly-followed democratic procedures in those cases where the right infringed is not clearly a requirement of proper political processes. On the one hand, he expressed confidence in the underlying libertarian content of traditional American political practice, and other the other, he suggested that the justification for individual freedoms and democracy should both be based in the same theory while merely hinting at the utilitarian roots of that theory. The first response suffers not only from historical amnesia, but more importantly simply doesn’t address the objection to his account of legitimacy. The second response should simply worry the objector more. In contrast, by basing a defense of rights to both individual liberty and political participation in the same proceduralist theory of constitutionally-structured democracy, Habermas can offer a straightforward response. A clear violation of an individual right—such as the torturing of an innocent person—does violate the procedural conditions of democratic legitimacy once we see that such violations make it impossible for individual assent to laws to be understood as the result of reason, rather than coercion or exclusion from the opinion-forming and decision-making processes. If democratic legitimacy requires that individuals must be able to give their reasoned consent to those laws they are subject to, and that reasoned consent is impossible to secure without 1) maximal equal subjective liberty rights, 2) equal membership rights, and 3) equal rights to the legal protection and actionability of those rights, then popular sovereignty under law presupposes constitutionally guaranteed individual liberties. Of course, this theory of procedural legitimacy cannot show that no morally unacceptable rights-infringing law could ever be passed under our current best understanding of required procedures. But, it seems equally correct to say that no substantivist theory of legitimacy could meet this extreme argumentative burden either: it is always possible that our best understanding of ex-ante moral constraints is not sufficient to protect against unforeseen possibilities of injustice.

Yet Habermas’s argument avoids the recourse to substantivist defenses of rights in terms of either natural law or religious truth that Dworkin and Perry make respectively. This should ensure that Habermas’s defense of constitutionalism—and by extension, constitutional review—does not fall prey to the problems of paternalism and partialism that substantivist accounts of legitimacy apparently lead to under conditions of value pluralism. In addition, Habermas claims to be able to comprehend the deontological character of rights because of their unconditional justification: rights to both private autonomy and political participation have the status of individuals’ legal claims that may not be abrogated by considerations of the collective good and preferred policy initiatives, nor may be treated as merely one among other competing goods to be weighed and transitively ordered on a case-by-case basis. Thus, Habermas’s account of the legitimacy requirements of constitutional democracy fulfills the three conditions I outlined earlier: it can defend more than simply rights to political participation, it is proceduralist rather than substantivist, and, it understands rights deontologically rather than teleologically.

But given this procedural concept of legitimacy that stresses the importance of citizens’ reasoned deliberations about and decisions upon substantive issues confronting the polity, it is somewhat unclear what the role of constitutional review might be. For, if the substantive normative content of laws gains its legitimacy only by being enacted in accordance with the procedural requirements of popular sovereignty, why shouldn’t any and all outcomes of proper legislative procedures have the force of law?[92] Like Ely, Habermas answers this in terms of the need for the maintenance of exactly that system of rights that secures statutory legitimacy through procedural—that is to say, constitutional—correctness. Constitutional review is understood and justified as a surety for just those democratic procedures adherence to which confers legitimacy on positive laws. Since state actions are legitimate only on the condition that they have resulted from fair and open procedures, there must be some way of reviewing the correctness of those procedures, which includes ensuring individuals’ procedurally required civil, membership, legal, political, and social rights.

According to Habermas, the constitutional review of ordinary statutes and governmental policies should be thought of, in ideal-typical terms, as a form of application discourse, not as a type of justification discourse.[93] In a discourse aiming at the justification of a general norm of action, all those potentially affected by the proposed norm must come to an agreement as participants in a rational discourse if the acceptance of the norm is to be valid. Parliamentary law making can thus be understood as a type of justification discourse, where proposed statutes are debated and considered by representatives of all those potentially affected, before the law is decided upon through a vote intended to secure finality under time and knowledge constraints. In contrast, application discourses do not aim at the justification of general norms of action tailored to standardized situations. Rather, they aim to apply already justified norms to the concrete features of a specific action situation. Because several valid norms may be prima facie relevant to the given situation, an application discourse aims to clarify the relevant features of the situation in order to make possible a determination of which of the potentially applicable norms is appropriate. Ordinary judicial proceedings can thus be understood a type of application discourse, where a sufficiently exhaustive characterization of the facts of the case relevant from the point of view of potentially applicable legal norms should make possible an impartial judgment about the unique applicability of the appropriate, and hence decisive, legal norm. In the legal context, of course, such an application discourse will often turn on a proper determination of the hierarchical relations between the potentially applicable legal norms, where the decisive features of the situation actually concern the present state of the relevant law. Returning now to the constitutional review of statutes and policies, Habermas conceives of it as a type of application discourse, seeking an impartial application of already justified higher level constitutional norms to those legal norms justified through ordinary legislative procedures. In determining whether higher order constitutional norms are applicable to ordinary legal norms, constitutional review ensures that the procedural conditions of democratic legitimacy—basic rights to private and public autonomy—have been fulfilled.

Of course, at this point, no specific recommendations concerning how such review should be institutionalized follow from this account; all that is established is that some form of constitutional review is needed. Note also that for Habermas, unlike those who argue for judicial review as a distinctly anti-democratic counterweight required by constitutionalism’s principle of individual rights, constitutionalism and democracy are not competitive or antithetical principles whose conflict is to be resolved by an unaccountable judiciary.[94] On the contrary, legally constituted individual rights and legally constituted rights to political participation presuppose one another. Popular sovereignty is only legitimate if it respects the legal status of subjects as independent, so that their agreement can be supposed to rest on their autonomous consent and not on coercion. What, how, and with respect to what properties equal rights to private autonomy are to be equally enjoyed by all and legally enforced through the mechanisms of state coercion can, however, only be legitimately determined through citizens’ use of their rights to political participation. Habermas has recently summarized this complex argument for the interdependence of private autonomy and public autonomy, and thus for the interdependence of constitutionalism and democracy:

 

There is no law without the private autonomy of legal persons in general. Consequently, without basic rights that secure the private autonomy of citizens, there also is no medium for legally institutionalizing the conditions under which these citizens, as citizens of a state, can make use of their public autonomy. Thus private and public autonomy mutually presuppose each other in such a way that neither human rights nor popular sovereignty can claim primacy over its counterpart.[95]

 

The very processes of deliberative politics themselves that aim to transform conflicting opinions, desires, and interests into democratically sanctioned legal programs require that rationality-enhancing procedures and autonomy-ensuring conditions have been met if participants are to understand the results of such procedures—just because they are the results of such procedures—as legitimately binding upon them. Thus for Habermas, as for Ely, constitutional review secures the legitimacy of political outcomes by insuring their procedural conditions.

In contrast to Ely, however, Habermas believes that much more activity is implied by this guarantor role, and so for whatever institutional organ or organs are to play the role. Those reviewing the constitutionality of statues cannot simply be ‘anti-trust’ style referees in the political marketplace. As is already clear by the extensive system of rights Habermas takes as requisite for constitutional democracies, a reviewing body will have to scrutinize legislative processes and outcomes not only for violations of rights to political participation, but also for those of individual civil liberties, membership rights, rights to legal protection, and those social and ecological rights necessary for ensuring the equal opportunity of all citizens to actualize their legally ensured private and public autonomy. Hence, Habermas recommends a quite activist constitutional review precisely with respect to the procedural requirements for legitimate democracy:

 

If one understands the constitution as an interpretation and elaboration of a system of rights in which private and public autonomy are internally related (and must be simultaneously enhanced), then a rather bold constitutional adjudication is even required in cases that concern the implementation of democratic procedure and the deliberative form of political opinion- and will-formation.[96]

 

In addition, because Habermas has a richer and more differentiated account of democratic politics than Ely and looks beyond the bare electoral relationship between representative bodies and the citizenry for the core circuit of democratic accountability, this also leads to an expansion of the purview of constitutional review over Ely’s theory.

Democratic Processes

Rather than the Lockean picture of politics as entirely a marketplace of competing individuals and groups trying to push their pre-political, private and corporate interests through the legislature, Habermas develops a picture of politics as also including (at least sometimes) the Rousseauian, deliberative search for the general will. If this more expanded conception of democratic politics is warranted, it is no longer adequate to attend only to those processes of bargaining, compromise, and aggregation that exhaust the pluralist model of democracy. Rather, democratic theory must be able to account for the diversity of forms of practical reasoning that can play different roles in political activity: not only the interest aggregation and fair bargaining that Ely attends to, but also those consensus-oriented debates over what is in the equal interest of each citizen that Dworkin points to in terms of principled morality, prudential reasoning about proper means to pre-given ends, as well as the kind of ethical-political self-clarification and reflection on constitutive histories that Perry focuses on. Often some or all of these different types of practical reasons are bundled together in the justification discourses that constitutional conventions and legislatures engage in during enactment.

 

Legal norms … can be justified not only with moral but also with pragmatic and ethical-political reasons; if necessary, they must represent the outcome of a fair compromise as well. … Valid legal norms indeed harmonize with moral norms [as just], but they are ‘legitimate’ in the sense that they additionally express an authentic self-understanding of the legal community, the fair consideration of the values and interests distributed in it, and the purposive-rational choice of strategies and means in the pursuit of policies.[97]

 

Once we attend to these different forms of practical reason, however, it becomes clear that the substantive normative content entrenched in constitutional and statutory law is quite complex, often comprising a syndrome of justice, ethical-political, pragmatic, and fairly bargained claims. If constitutional review is thought of as the application of already-justified constitutional norms to legislatively justified statutory norms, then, like any form of application discourse, it will often involve unpacking a syndrome of different kinds of practical reasons embedded in laws. Hence, the scope of activity for purely procedural constitutional review is again increased from what Ely recommends, moving beyond the latter’s focus on the legislative constriction of participating parties and the governmental use of suspect classifications, to include the quality and propriety of justifying reasons as well.[98]

 

The legitimating reasons available from the constitution are given to the constitutional court in advance from the perspective of the application of law—and not from the perspective of a legislation that elaborates and develops the system of rights in the pursuit of policies. The court reopens the package of reasons that legitimated legislative decisions so that it might mobilize them for a coherent ruling on the individual case in agreement with existing principles of law; it may not, however, use these reasons in an implicitly legislative manner that directly elaborates and develops the system of rights.[99]

 

Finally, the institutional scope of constitutional review is much greater on Habermas’s model than on Ely’s. Habermas claims that in order to ensure the procedural correctness of law-making activities, it is not enough to look only to formally organized legislative and quasi-legislative governmental bodies. These bodies are part of what Habermas calls the ‘strong public sphere’, where decisive will-formation occurs. Besides this strong public sphere, there is a ‘weak public sphere’ characterizable in terms of non-governmental civil society that contributes information, diverse perspectives, opinions, and reasons to the collective processes of political debate.[100] Ideally, on Habermas’s model, the legitimate circulation of power would operate by the formation of ‘communicative power’ in the weak public spheres that identify and thematize problems, conflicts, and deficits in the everyday life of citizens, the taking up of this public opinion into legislative contexts and its transformation into laws that can then direct the administrative power of the state to achieve the action coordination indicated. Ultimately, it is only the robust deliberative character of opinion formation in the ‘weak’ public sphere that warrants the expectation of rational outcomes from representative, parliamentary procedures, and grants legitimacy to politically adopted programs. Therefore, if constitutional review is to be oriented towards protecting and promoting participatory opinion- and will-formation, it will need to be much more than an impartial referee between voters and their representative bodies: it will have to ensure that the ‘sluice-gates’ through which public opinion gets channeled into the legally structured strong public sphere remain unobstructed.[101] And finally, if we agree with Habermas’s social-theoretic claim that obstructions to and distortions in these channels occur not only through governmental power but also from economic and social powers, then those entrusted with the power of constitutional review will have a great deal of work on their docket. According to Habermas the function of constitutional review can be summarized as simply guaranteeing the procedural fairness and openness of democratic processes. Yet, concretely, the tasks involved are manifold: keeping open the channels of political change, guaranteeing that individuals’ civil, membership, legal, political, and social rights are respected, scrutinizing the constitutional quality and propriety of the reasons justifying governmental action, and ensuring that the channels of influence from independent, civil society public spheres to the strong public sphere remain unobstructed and undistorted by administrative, economic, and social powers.

Institutionalization and Lingering Paternalism in Habermas

The question now is what institutional arrangements would best carry out all of these various tasks of constitutional review, while being sensitive to the principle of popular sovereignty? Even if we accept that democracy, properly understood, requires a robust form of constitutional review, it is not clear that an electorally unaccountable body structured as a judicial panel is the best mechanism to carry out the manifold tasks of a procedural guardian of democracy. Habermas briefly considers alternative ways of institutionalizing constitutional review: for instance, in a special committee of the legislature or in the executive administration.[102] After categorically rejecting the latter as a subversion of the executive’s proper constitutional role of being directed by legislatively enacted positive law, he notes that constitutional review ‘belongs without question among the functions of the legislature. Hence it is not entirely off track to reserve this function, even at a second level of appeal, to a legislative self-review that could be developed into a quasi-judicial procedure.’[103] It is worth noting that this passage only claims that legislatures must take account of the constitutionality of proposed legal norms; since they may not simply pass off this task to others, it ‘belongs without question among the functions of the legislature.’ The passage does not recommend locating final and sole powers of constitutional review in the legislature since, as Habermas goes on to note, an ordinary legislative body—as opposed to a constitutional assembly—does not have the same disposition over the content of constitutional rights as it does with respect to ordinary statutory content.

However, he apparently drops further consideration of alternative designs in favor of an extended discussion of judicially institutionalized constitutional review and of various interpretive and methodological problems raised in the German and American contexts. At this point, I think, one might justly wonder whether Habermas has forsworn the utopian-critical potential of his broader project in favor of a type of ameliorist ‘justificatory liberalism’ that merely intends to show why the way we do things around here is pretty much just fine as it is. Granted, his further discussion of the self-understanding of the adjudicative methodology of the German High Court is intended to move it from a value-balancing to a proceduralist jurisprudence, and this recommendation is motivated largely by worries about the judicial paternalism that can result from a method focused on reinforcing what the judiciary takes to be the ethical identity of its society. And his extended discussion of American jurisprudential debates about adjudicative methodology is likewise focused on worries about types of constitutional interpretation that lead to overreaching on the part of the Supreme Court. Yet these arguments about how a constitutional court should adjudicate already presuppose that the institutionalization question is settled. Perhaps the richness of the German and American jurisprudential debates simply distracts Habermas from a more wide-ranging consideration of issues concerning the separation of powers and institutional design.

Nevertheless, his theory must face the same problem raised by other theories: why isn’t the common institutional arrangement of judicial review paternalistic? I believe that Habermas presents, rather obliquely, two kinds of considerations here. His first response is that a procedural understanding of the system of rights will not in fact lead to judicial paternalism, since judges reviewing the constitutionality of statutes need not have recourse to any substantive political or moral ideals justifiable apart from those already contained in constitutional provisions and legislatively-enacted statutes. Although the rights specified in the constitution are to be understood as having substantive, deontic content, they are designed to be exactly (and no more than) those rights procedurally required for realizing the principle of popular sovereignty in a legal form, and so, exactly those rights individuals would have to grant each other if they intend to regulate their interactions as free and equal consociates under law. Since the system of rights is procedurally justified in the first place, whatever governmental organ is charged with interpreting that system does not need to rely upon metaphysically secured theories of natural rights or objective value hierarchies. The basic idea is that the process of constitutional review does not itself require the justification of the normative content of the system of rights, but only requires the rational application of normative content already embodied in constitutional provisions; provisions that are already justified in terms of the legal and normative requirements of an association of free and equal citizens engaged in the process of ruling themselves. In this sense, judges are in the same position with respect to the constitution as they are with respect to ordinary statutory application: they must unpack the normative reasons bundled together in the constitutional provision or statute in order to determine which of the relevant, potentially competing norms is applicable in a particular context. Above all, a constitutional court must avoid taking itself as trying to secure, through its jurisprudence, a substantive hierarchy of values or catalog of natural rights that ought to be, but are not currently, contained in the constitution: ‘By assuming that it should strive to realize substantive values pregiven in constitutional law, the constitutional court is transformed into an authoritarian agency.’[104] In short, because the constitution itself is to be understood as largely procedural in character and because a constitutional court ought merely to apply that content, we need not worry about democratically unaccountable judges interpreting and imposing substantive normative content above and beyond what is already instantiated in law.

Of course, this response to the objection to a judicial instantiation of review puts the cart before the horse: it recommends an interpretive method to the judiciary, while presupposing a judicial institutionalization as given. The argument cannot itself establish the proper separation of powers here; it presupposes that issue as settled. It may well be that a constitutional court ought to understand its work product as merely an application of the already-justified normative content embedded in constitutional and statutory norms, but this self-understanding can only ward off the danger of particular paternalist decisions for a body that is—according to the objection—institutionally placed in such a way that it is constantly in danger of encroaching on the principle of popular sovereignty.

Habermas needs here some further argument to show that, compared to other branches of government, and to other possible forms of institutionalization, a court sitting at the apex of the judiciary, whose members are at most quite indirectly responsive to citizens’ public use of reason, will have some higher degree of competence at performing constitutional review. This constitutes his second response to the objection to judicial review from popular sovereignty, namely, that the institutionalization of constitutional review in a judicial body is recommended by an understanding of the separation of governmental powers along the lines of specialized discursive functions. Here Habermas relies on a form of the claim to judicial competence, though one unlike Perry and Dworkin’s claim that judges are better moral reasoners than elected officials. While the legislature specializes in the function of justifying legal norms, the judiciary specializes in the rational application of prima facie justified legal norms to particular situations. The judiciary’s competence is not based upon judges’ special character traits or particular capacity for moral reasoning, but upon the judiciary’s institutional competence in dealing with the specialized form of legal discourses of application—as opposed to pragmatic, moral, and ethical justification discourses—which are operationalized in an independent court system and a juridical form of argumentation through decisions backed by opinions.

 

Legal [juridical] discourse can lay claim to a comparatively high presumption of rationality, because application discourses are specialized for questions of norm application, and can thus be institutionalized within the surveyable framework of the classical distribution of roles between the involved parties and an impartial third party. For the same reasons, however, they cannot substitute for political discourses that, geared for the justification of norms and policies, demand the inclusion of all those affected.[105]

 

Because the review of legislatively enacted statutes for their constitutionality is still a matter of the rational application of already justified legal norms—those embodied in both constitutional provisions and statutory enactments—the judiciary has the requisite competence for this function that other governmental bodies are lacking. While legislative bodies are specialized in, and designed for, the justification discourses involved in making laws and establishing policy goals, and administrative bodies are specialized in, and designed for, the pragmatic discourses involved in the selection of efficient means to legislatively-given policy goals, judicial bodies are specialized in, and designed for, the application discourses involved in determining the uniquely appropriate valid law applicable to concrete situations. Given the intricacies and difficulties of application discourses, only judicial bodies have the requisite competence to ensure a rational procedure of applying laws, especially since ‘the complex steps of a constructive interpretation’—as the central part of an impartial judicial application discourse—’certainly cannot be regulated through procedural norms.’[106]

How strong an argument for judicial review is this? Much of the answer turns on the strength of the analogy between ordinary jurisprudence and constitutional review; that is to say, on Habermas’s claim that both are instances of application discourse, and so justifiably judicial tasks due to the heightened rationality of juridical discourses specialized for resolving disputes impartially between involved parties. Although it is obscured in the United States context, where legislatively enacted statutes are reviewed by the Supreme Court only on the occasion of a test case with all of its concrete details, ordinary jurisprudence is clearly not directly analogous to constitutional review. To begin with, in the former case, the question before the court is how to sufficiently describe the legally relevant facts of the situation so that exactly one of several prima facie relevant legal norms can be shown to be uniquely appropriate, and so dispositive of the case. In constitutional review, however, the question is whether lower-level legal norms, such as those embodied in statutes or administrative policies, can be made consistent with higher-level constitutional norms. In the former case, a semantically universal norm is being applied to particulars; in the latter a semantically universal norm is being applied to another such norm.

One might use this difference to object to Habermas’s argument by claiming that, since constitutional review involves a different form of practical reasoning that ordinary application discourses, it must be a form of justification discourse, and so properly carried out by a legislative body.[107] If, however, this objection proves anything, it proves too much. For surely any form of judicial discourse that hinges on ascertaining the proper hierarchy of valid legal norms—including the relationship between various positive rules and legal principles—also would not be in the strict form of an application discourse. Although a case may end by applying one legal norm to a fact situation, the decisive work is often done in properly ascertaining the relevant priority relationships between various legal norms, and this usually involves subsuming some semantically universal norms under others. If the special competence of judicial bodies does not extend to this form of reasoning about the hierarchical relationships between norms in a system of norms—if the logic of such reasoning is that of a justification discourse—then it is unclear why judicial bodies should ever have a legitimate say on such matters, irrespective of whether legal norm conflicts arise amongst ordinary legal norms or between these and constitutional norms. Perhaps we might reserve the term ‘application’ for those cases where a semantically universal norm is applied to particular facts, but then Habermas’s argument should just be restated to say that judicial bodies have a specialized competence for ‘adjudicative’ discourses, that is, those involving both norm application and norm prioritization.

Nevertheless, defending judicial review as merely an extension of the judiciary’s ordinary capacities to the adjudication of norm conflicts between constitutional and statutory norms is insufficient to dispel worries about judicial paternalism. This is because crucial constitutional provisions are deliberately open textured and the meaning of their specific content—usually debated in terms of their applicability—is often the subject of reasonable and deep disagreement and the prevalent understanding of that meaning often changes over time. This is precisely the problem of constitutional indeterminacy that forms the starting points for the competing theories of constitutional interpretation put forward by Ely, Perry, and Dworkin. And this is the heart of the disanalogy between ordinary adjudication and constitutional review.[108] In both ordinary statutory adjudication and constitutional review, the judiciary may well need to further specify the content of the relevant norms through a constructive interpretation of them. However, in the former case, the judiciary is on the same level as the legislature since, in a sense, the court is rendering to the legislature a rebuttable interpretation of the meaning of the original statutes, an interpretation that can be easily rejected by the legislature through a new enactment. In the case of constitutional review, in contrast, the court is only on the same level as a constitutional assembly, not the ordinary legislature. Here it is interpreting, through further specification, the relevant open-textured provisions of the constitution, yet it could only be rendering such a rebuttable specification to the people as a whole, understood, however, as a constitutional assembly. For, on the procedural account of democratic legitimacy and the deliberative account of democratic processes, the task of realizing the system of rights in concrete terms for an historically specific community is decidedly a matter for the people reasoning together as a whole, and not for any appointed set of wise tutors.

Another way of seeing the danger here is that in applying constitutional tests to statutes and policies, a constitutional court may engage in forms of constitutional specification that rely on reasons available legitimately only to democratic processes of self-government, and thereby surrender the court’s claim to legitimacy based upon its narrow specialization in legal discourse. Commenting upon Robert Alexy’s call for the judicial use of the full gamut of ‘general practical discourse’ in ordinary judicial proceedings,[109] Habermas clearly recognizes this danger—an even more pressing danger in the case of constitutional review:

 

I am still not quite clear about the role of what Alexy calls ‘general practical discourse.’ Here, different types of argument—prudential, ethical, moral, [and] legal arguments—are supposed to come in one package. I have the suspicion that this conception is not sufficiently sensitive for the desired separation of powers. Once a judge is allowed to move in the unrestrained space of reasons that such a ‘general practical discourse’ offers, a ‘red line’ that marks the division of powers between courts and legislation becomes blurred. In view of the application of a particular statute, the legal discourse of a judge should be confined to the set of reasons that legislators either in fact put forward or at least could have mobilized for the parliamentary justification of that norm. The judge, and the judiciary in general, would otherwise gain or appropriate a problematic independence from those bodies and procedures that provide the only guarantee for democratic legitimacy.[110]

 

Constitutional review, especially where it involves specifying the system of rights observance of which grants legitimacy to positive law, may very well involve just this kind of a use of reasons that the judiciary does not have legitimate disposition over. In the end, if—to recall Rousseau’s test of popular sovereignty—citizens are to understand themselves as both under positive law and free, they must be able to simultaneously understand themselves as authors of the system of rights they are subject to.

Of course, the combination of judicial institutionalization and constitutional indeterminacy that proves threatening to citizens’ political autonomy may indeed be mitigated by the adoption by supreme courts of a form of constitutional interpretation that is oriented towards the reinforcement of deliberative democracy. And this is indeed what occupies Habermas’s chapter called ‘Judiciary and Legislature: On the Role and Legitimacy of Constitutional Adjudication.’[111] Nevertheless, the adoption of a legitimate form of constitutional jurisprudence cannot itself justify placing the power of constitutional review in a judicial body.

In summary, neither of Habermas’s two obliquely presented arguments for the institutionalization of constitutional review in an independent, politically unaccountable judiciary is compelling. First, his claim that a resolutely procedural form of legal interpretation would guard against judicial paternalism simply begs the question of how to institutionalize this power in the first place. His second argument hangs on the contention that the rationality of application discourses can be best ensured by an institutionally separated judicial power specialized in, and limited to the ambit of, such application discourses. Insofar as any controversies arise over how to specify indeterminate norms or prioritize competing norms, juridical resolutions can be safely offered, as rebuttable presumptions, to properly democratic actors specialized in justification discourses. Here, however, the analogy between ordinary and constitutional controversies breaks down, since in the former case the proper democratic actors are the weak and strong public spheres of civil society and the legislature, while in the latter, a constitutional assembly.

It seems to me that the most cogent argument for entrusting a judicial panel with the power of constitutional review remains Ely’s argument from judicial independence. In a controversy between political actors about the constitutive rules of political cooperation—the rules which, in a procedural conception of democracy carry significant weight for legitimation—none of the interested parties can be counted on for an impartial resolution of the controversy. In this way, a judiciary that is independent of normal political accountability is in a unique institutional position to guarantee that the procedural conditions of democratic processes are correctly fulfilled. However, the limitations of Ely’s theory of judicial review result from the rather thin conception of democracy that he employs, namely that democratic self-government means merely a process of aggregating pre-political preferences through elections in order to achieve overall social utility. Thus, we need to recognize, with Habermas, that an adequate guarantor of the legitimacy of the democratic process will need to take on tasks beyond policing elections and ensuring against monopolistic distributions disfavoring insular minorities: it will need to secure the full gamut of individuals’ rights to private and public autonomy, scrutinize the quality of reasons justifying governmental actions, and ensure that the circuit of communicative power from the weak public spheres through the strong public sphere and into administrative power remains undistorted by economic and social powers. Furthermore, the fact that constitutional review involves complicated adjudicative considerations concerning the application, specification, and prioritization of constitutional and statutory legal norms entails that rational outcomes of such review can not be warranted through a determinate set of procedural rules. Judicial bodies do, however, specialize in dealing with such adjudicative complexities, and so we should expect heightened rationality from the outcomes of juridical discourses employed for constitutional review. As I have noted, it is surprising that Habermas displays somewhat of a lack of imagination concerning the institutional design of constitutional review, preferring to take currently prevalent structures for granted while focusing on the proper self-understanding of constitutional courts and the adjudicative methods that they should adopt. Yet I believe that his basic theory of constitutional review, especially since it properly conceives of the relationship between constitutionalism and deliberative democracy as mutually presuppositional rather than antithetical, gives us a solid basis for finding better forms of institutionalization. The arguments I have advanced so far, if successful, point to the following criteria for such a body: independent of ordinary political processes, capable of rationally cognizing complex adjudicative arguments through juridical forms of reasoning, sensitive to the manifold tasks involved in securing the procedures of deliberative democracy in order to warrant the expectation of legitimate outcomes from ordinary political processes, capable of significant interventions in ordinary political processes when they malfunction, yet not capable of introducing substantive normative content into the system of rights without serious opportunities for citizens’ participation and influence, nor capable of decisions with virtually indefeasible finality.

2       Forms of Institutionalization and Open Questions

In a more exploratory vein, it is worth considering various ways of organizing a political power that might be able to fulfill these criteria better than, for instance, the United States Supreme Court. Surely Habermas is right to reject locating the job of constitutional review exclusively in the executive branch, as this would transform the administration’s role in the legitimate circulation of democratic power from that of being programmed by statutory enactments into that of the specifying the normative content of basic legal norms. However, in the United States, before Justice Marshall’s declaration of the power of judicial review in Marbury v. Madison, many envisioned that all three branches of government would equally participate in the job of constitutional interpretation and application.[112] Yet with this scheme, citizens lose one of the crucial components that enables law to act as an effective medium for society-wide action coordination: namely, legal certainty, here with respect to the basic law of the nation-state itself. Another unacceptable scheme that Habermas does not consider would be the popular election of constitutional judges. With this approach the institutional importance of an impartial judiciary would be lost—an impartiality that Ely pointed to as essential in procedural conflicts between citizens and their representatives—turning the constitutional court into another site of partisan politics rather than an impartial guardian of the procedural requirements of deliberative democracy. Such a scheme misunderstands the meaning of democracy by focusing on the election of candidates as the privileged locus of democracy, thereby losing sight of Rousseau’s concept of voting as only a necessary mechanism for the conclusion, under time and knowledge constraints, of open debate and discussion oriented towards a rational agreement. Furthermore, as elected constitutional judges orient themselves towards reelection, there is a potentially serious loss in the increased rationality of adjudication that is carried by written opinions, concurrences, and dissents focused on the legal and constitutional issues at stake, rather than on polemics designed as campaign tracts. Another form of institutional design—limited rather than life terms for appointed constitutional judges—seems neither to significantly heighten nor minimize the dangers of judicial paternalism. Although this arrangement reduces the long-term deleterious effects of a single judge’s willfulness in imposing substantive ideals against the legislature, it also makes it possible for a sort of revolving door of judicial paternalists.

More promising institutional designs, it seems to me, begin with the establishment of a constitutional court that does not also have responsibilities of being the apex of the federal judiciary. Like many European constitutional tribunals, such a court would be specialized in the review of the procedural conditions of legislative enactments and administrative programs.[113] This would at least have the effect of focusing public consciousness on its decisions as specifically constitutional concerns and so would function to deepen public appreciation and understanding of constitutional democracy as the ongoing elaboration of an interlocking system of civil, legal, and political rights.[114] One problem with this kind of arrangement is that appellate cases often raise constitutional issues, and so it seems that a court supreme over the various appellate branches would of necessity deal in constitutional issues, thus collapsing the functional distinction.[115] Here it may be helpful to distinguish two different ways in which ‘constitutional issues may be raised.’ On the one hand, a case may implicate a constitutional provision or principle in such a way that it decides the case; on the other hand, a case may crystallize a conflict of laws between constitutional and normal legal norms. If this distinction is tenable, then it seems possible to arrange the judicial separation of powers along its lines so that ordinary courts up through the appellate system and including the supreme appellate court would deal with constitutional issues of the first variety, while constitutional conflicts of law would be referred to an independent constitutional court.

Insisting on a form of constitutional self-review of statutes within legislatures themselves would significantly reduce the collisions between a constitutional court and the legislature, collisions that are often popularly perceived as resulting in ‘judicial usurpation.’ As Habermas himself notes:

 

It is worth considering whether the legislature could not also scrutinize its decisions, exercising a quasi-judicial review of its own … in a parliamentary committee (also) staffed by legal experts. This method of internalizing self-reflection on its own decisions would have the advantage of inducing legislators to keep the normative content of constitutional principles in mind from the very start of their deliberations. This normative content is lost, for example, when, in the crush of parliamentary business, moral and ethical questions are redefined as negotiable questions open to compromise. In this regard, the institutional differentiation of a self-referential procedure for reviewing norms, whose operation would remain in the hands of parliament, might perhaps contribute to a more rational legislative process.[116]

 

Even if such a legislative arrangement could not have the final word on all constitutional questions—especially when the legislature itself distorts democratic processes in the manifold ways Ely points to—it would still reinforce the internal normative connection between constitutionalism and popular sovereignty during the process of democratic will-formation itself. And this reinforcement, in addition to the discipline it would impose on legislators, would significantly increase the deliberative character of legislative politics.

Finally, arrangements akin to the Canadian ‘not-withstanding’ clause that Perry recommends would, by reducing the finality of judicial review, furnish a real potential for balancing between the twin dangers of an unaccountable judicial interpretation of abstract, controversial constitutional principles and the potential injustices of a federal legislature that closes itself off from the input of weak public spheres or that denigrates private autonomy through overreaching pursuit of goals and interests shared by a majority of citizens. Such a proposal is made especially attractive by the way in which it focuses both governmental and non-governmental deliberation on the procedural conditions that make legitimate self-government possible. By forcing reconsideration of not only the specific issue at play, but also of the concrete meaning and implications of the principles they have agreed to live by, citizens would have an opportunity to collectively reflect upon and debate about the meaning of the system of rights that they are trying to bring to legal life. Of course, by allowing a constitutional court to have a first chance at defining the relevant legal issues, the advantages of juridical discourse specialized for matters of adjudication are brought to bear without the democratic deficits associated with strong and final forms of judicial review.

Instances of judicial review are indeed never completely final given the basic possibility for amendment by citizens and legislatures built in as a basic structure of modern constitutions. But the relative finality of judicial pronouncements is inversely proportional to the ease with which amendments can be voted on and adopted. While any amendment procedures must be more strenuous than those for statutory enactment given that the constitution itself is to specify the procedural conditions guaranteeing the legitimacy of ordinary legislation, if constitutional amendment is close to impossible then the democratic elaboration of the system of rights becomes a chimera and the possibility of judicial usurpation becomes all too real. Thus the proper institutionalization of constitutional review crucially depends on balancing between these two extremes in the light of other structural and political features of a specific constitutional state. The United States could certainly benefit, it seems to me, from an easing of the amendment requirements. This is the apparent lesson learned from the U.S. experience by later constitution writers in other countries attuned to the demands of democracy who almost universally made the amending process easier. Such procedures relieve the pressure on a constitutional court to single-handedly develop constitutional principles and the system of rights in accordance with new insights into the demands of justice. This might increase the legitimacy of constitutional court decisions as the burdens of progressive constitutional refinement would not be shouldered by inventive interpretations of written constitutional language. For example, the Supreme Court has recently developed a doctrine of gender as a suspect classification for the purposes of equal protection jurisprudence in the wake of the defeat of the Equal Rights Amendment despite its wide support amongst citizens and its clear consonance with the principles underlying the U.S. system of rights. However necessary from the perspective of a principled articulation of the system of rights, the court’s doctrine is not based in any clear constitutional text, text that would have been provided by the ERA. Easing amendment requirements would also, of course, allow citizens and accountable political actors to more readily modify and/or reverse the edicts of a constitutional court.

Although these suggestions for procedurally required legislative self-review, an independent constitutional court, and finality-reducing mechanisms such as the not-withstanding clause and easier amendment procedures are offered as exploratory proposals for alternative institutional designs, I believe they are more consonant with the legitimacy requirements of deliberative democracy than mere acceptance of, and attempts to tame, current forms of judicial review. They are offered in the hope that the realization of popular sovereignty through democratic constitutionalism is an ongoing historical project, one that must ultimately be open to the participation of all citizens. As Habermas nicely explicates the notion of an ongoing constitutional project,

 

The constitutional state does not represent a finished structure but a delicate and sensitive—above all fallible and revisable—enterprise, whose purpose is to realize the system of rights anew in changing circumstances, that is, to interpret the system of rights better, to institutionalize it more appropriately, and to draw out its contents more radically. This is the perspective of citizens who are actively engaged in realizing the system of rights.[117]

Open Questions Concerning Constitutional Interpretation

I have argued that the best conception of the purpose and functions of constitutional review is that put forward by deliberative democrats who side with Rousseau on two crucial distinctions: that the legitimacy of collective decisions is best understood procedurally and not substantively, and that collective decision making processes are best modeled as aiming at the general will and not merely the aggregative will of all. In particular, I have tried to show how the various justifications of constitutional review put forward by Ely, Perry, Dworkin, and Habermas can be clarified and critically evaluated in terms of their respective positions on these two distinctions. The guiding premise of the paper is that jurisprudential debates often move too quickly to questions concerning the proper methods that a specific supreme court should adopt in interpreting a nation-state’s constitution, even though much of each theory’s characteristic work is really being carried by its underlying conception of the relationship between constitutionalism and democracy, and its resulting position on the proper institutionalization of constitutional review. I have tried to eschew such jurisprudential issues, not because I think them unimportant, but because we need first to have a good grasp on the purpose and functions of constitutional review in a democratic context before exploring what the best interpretive methods might be.

If however, we accept the main lines of Habermas’s conception of constitutional review—even with reservations about his sanguine take on current institutional designs of judicial review—at least three interesting problems arise that can be briefly indicated but left for another day. First, if a supreme court were aiming at reinforcing deliberative democracy, and concerned about the potential costs to popular sovereignty of a judiciary that rules on constitutional content, it apparently ought to adopt the minimalist form of adjudicative interpretation recently recommended by Cass Sunstein.[118] This would involve avoiding both the articulation of broad rules that might lead to unintended errors in other cases in the future, and the expression of deep justifications for decisions which might establish new legal principles as precedents. Sunstein’s minimalist strategy of judicial review would, in essence, by leaving as much undecided as possible in the resolution of the case at hand, seek to leave as much open to participatory political deliberation as possible. By focusing tightly upon the issues raised by particular cases—whether reviewing lower court decisions in individual actions or abstractly reviewing the constitutionality of statutes and programs—a minimalist court could avoid adding any substantive normative content to the statutes, policies, and constitutional provisions involved. Habermas does not, however, endorse the minimalist form of constitutional interpretation recommended by Sunstein. As we have seen, not only does Habermas endorse Ely’s argument for ‘a rather bold constitutional adjudication … in cases that concern the implementation of democratic procedure,’[119] he significantly expands the purview of such review through a broader conception of the required procedural conditions and a more capacious account of democratic processes. Furthermore, in his work on legal interpretation, Habermas attempts to give an intersubjectivist twist to Dworkin’s ambitious program for constructive interpretation, a jurisprudential program heavily committed to deep justifications of legal decisions in terms of fundamental political principles. However, tailored as it is to a substantivist conception of political legitimacy—one that, as I argued above, would be quite willing to redefine any political procedures as democratic as long as they generated the correct outcomes—Dworkin’s brand of depth maximalism seems particularly prone resulting in a paternalistic judiciary. It is not self-evident that merely giving a dialogical reading to Dworkin’s Hercules, and stressing the pragmatics of juridical discourse, will be sufficient to dispel such worries, especially when such an interpretive method is used in constitutional review.

Second, even if problems about the institutional location of constitutional review were resolved, we still might have a worry about constitutional decisions concerning the fifth category of social and ecological rights, decisions that appear more prone than others to problems of judicial paternalism. Recall that these rights are designed to be those necessary for ensuring that citizens have equal opportunities to utilize their other rights to equal liberty, membership, legal protection, and political participation. Yet the substantive content of these rights appears to contain teleological components concerning highly contested and potentially partialistic conceptions of the good to a much higher degree than the other four categories of rights. If this is the case, then the fact that a constitutional court must guard the system of rights that makes private and public autonomy possible would seem to entail that the constitutional court would need to rely on controversial judgments about the good life to properly specify such rights, judgments that the deontic character of the liberal scheme of rights is designed to avoid in the first place. In order to guarantee the procedural legitimacy of political processes, then, the constitutional court would apparently need to introduce teleological considerations not contained in the syndrome of legal reasons bundled together in statutory justification discourses and so, would risk adding substantive normative content to their interpretations in a way that is potentially paternalistic with respect to pubic, participatory self-determination.

Finally, similar sorts of questions about the normative content that a court relies on in deciding cases of constitutional review arise with respect to the various legal paradigms tacitly presupposed and relied upon in adjudication. Once it becomes clear that individual judicial decisions often hinge in part on a specific paradigmatic social model—an implicit set of assumptions about and pre-understandings of one’s society—and that such legal paradigms contain normative ideals, it is no longer legitimate to simply allow judges to decide upon the content of the reigning paradigm. ‘The dispute over the correct paradigmatic understanding of the legal system … is essentially a political dispute. In a constitutional democracy, this dispute concerns all participants, and it must not be conducted only as an esoteric discourse among experts apart from the political arena.’[120] Yet, it is unclear exactly how to get such an inclusive discourse going on such an abstract, theoretical topic as the background understandings of legal concepts and how to keep such a debate, should it occur, at an acceptable level of quality. Here again, there are important issues of interpretive jurisprudence and comparative legal competence that might be fruitfully addressed by a theory of constitutional interpretation oriented by the desire to foster deliberative democracy.

References

Ackerman, Bruce. We the People: Foundations. Cambridge, MA: Harvard University Press, 1991.

________. We the People: Transformations. Cambridge, MA: Harvard University Press, 1998.

Alexy, Robert. ‘The Special Case Thesis.’ Ratio Juris 12, no. 4 (1999): 374-384.

Aquinas, Saint Thomas. Summa Theologiae. Vol. 43. 60 vols. New York: McGraw-Hill, 1964.

Aristotle. The Complete Works of Aristotle. Translated by Jonathan Barnes. Vol. 2. 2 vols. Bollingen Series. Princeton, NJ: Princeton University Press, 1984.

Barber, Benjamin J. Strong Democracy: Participatory Politics for a New Age. Berkeley: University of California Press, 1984.

Bickel, Alexander M. The Least Dangerous Branch: The Supreme Court at the Bar of Politics. Second ed. New Haven, CT: Yale University Press, 1986.

Bohman, James. ‘Survey Article: The Coming of Age of Deliberative Democracy.’ The Journal of Political Philosophy 6, no. 4 (1998): 400-425.

Bohman, James, and William Rehg, eds. Deliberative Democracy: Essays on Reason and Politics. Cambridge, MA: The MIT Press, 1998.

Bork, Robert H. ‘Neutral Principles and Some First Amendment Problems.’ Indiana Law Journal 47 (1971).

________. ‘Judicial Review and Democracy.’ In Encyclopedia of the American Constitution, ed. Leonard W. Levy, Kenneth L. Karst and Dennis J. Mahoney, 3, 1061-1064. New York: Macmillan, 1986.

________. The Tempting of America: The Political Seduction of the Law. New York: The Free Press, 1990.

Brewer-Carías, Allan R. Judicial Review in Comparative Law. New York: Cambridge University Press, 1989.

Cheeves, John. ‘Committee Oks Ban on Nude Club Dancing.’ Lexington Herald-Leader, February 11 2000, B 1.

Choper, Jesse H. Judicial Review and the National Political Process. Chicago, IL: University of Chicago Press, 1980.

Cohen, Joshua. ‘An Epistemic Conception of Democracy.’ Ethics 97 (1986): 26-38.

________. ‘Democracy and Liberty.’ In Deliberative Democracy, ed. Jon Elster, 185-231. New York: Cambridge University Press, 1998.

Dryzek, John S. Deliberative Democracy and Beyond: Liberals, Critics, Contestations. New York: Oxford University Press, 2000.

Dworkin, Ronald. ‘The Model of Rules.’ University of Chicago Law Review 35 (1967).

________. Taking Rights Seriously. Cambridge, MA: Harvard University Press, 1978.

________. ‘What Is Equality: Part I, Equality of Welfare.’ Philosophy & Public Affairs 10, no. 3 (1981): 185-246.

________. ‘What Is Equality: Part II, Equality of Resources.’ Philosophy & Public Affairs 10, no. 4 (1981): 283-345.

________. A Matter of Principle. Cambridge, MA: Harvard University Press, 1985.

________. Law's Empire. Cambridge, MA: Harvard University Press, 1986.

________. Freedom's Law: The Moral Reading of the American Constitution. Cambridge, MA: Harvard University Press, 1996.

Elster, Jon. ‘The Market and the Forum: Three Varieties of Political Theory.’ In Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg, 3-33. Cambridge, MA: The MIT Press, 1997.

Ely, John Hart. Democracy and Distrust: A Theory of Judicial Review. Cambridge, MA: Harvard University Press, 1980.

________. On Constitutional Ground. Princeton: Princeton University Press, 1996.

Estlund, David M. ‘Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority.’ In Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg, 173-204. Cambridge, MA: The MIT Press, 1997.

Favoreu, Louis. ‘Constitutional Review in Europe.’ In Constitutionalism and Rights: The Influence of the United States Constitution Abroad, ed. Louis Henkin and Albert J. Rosenthal, 38-62. New York: Columbia University Press, 1990.

Freeman, Samuel. ‘Constitutional Democracy and the Legitimacy of Judicial Review.’ Law and Philosophy 9 (1990-1991): 327-370.

Gunther, Gerald. ‘Judicial Review.’ In Encyclopedia of the American Constitution, ed. Leonard W. Levy, Kenneth L. Karst and Dennis J. Mahoney, 3, 1054-1061. New York: Macmillan, 1986.

Günther, Klaus. The Sense of Appropriateness: Application Discourses in Morality and Law. Translated by John Farell. Albany: State University of New York Press, 1993.

Habermas, Jürgen. Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy. Translated by William Rehg. Cambridge, MA: The MIT Press, 1996.

________. The Inclusion of the Other: Studies in Political Theory, ed. Ciaran Cronin and Pablo De Greiff. Cambridge, MA: The MIT Press, 1998.

________. ‘A Short Reply.’ Ratio Juris 12, no. 4 (1999): 445-453.

Hamilton, Alexander, James Madison, and John Jay. The Federalist Papers. New York: Bantam Books, 1982.

Hand, Learned. The Bill of Rights. Cambridge, MA: Harvard University Press, 1958.

Hart, H. L. A. The Concept of Law. Second ed. Oxford: Clarendon Press, 1994.

Hobbes, Thomas. Leviathan Cambridge Texts in the History of Political Thought, ed. Richard Tuck. New York: Cambridge University Press, 1991.

Locke, John. Two Treatises of Government, ed. Peter Laslett. New York: Cambridge University Press, 1988.

Mansbridge, Jane. Beyond Adversary Democracy. Chicago: University of Chicago Press, 1983.

Michelman, Frank I. Brennan and Democracy. Princeton, NJ: Princeton University Press, 1999.

________. ‘Morality, Identity and 'Constitutional Patriotism'.’ Denver University Law Review 76, no. 4 (1999): 1009-1028.

Nino, Carlos Santiago. The Constitution of Deliberative Democracy. New Haven: Yale University Press, 1996.

Perry, Michael J. The Constitution, the Courts, and Human Rights: An Inquiry into the Legitimacy of Constitutional Policymaking by the Judiciary. New Haven, CT: Yale University Press, 1982.

________. The Constitution in the Courts: Law or Politics? New York: Oxford University Press, 1994.

________. The Idea of Human Rights: Four Inquiries. New York: Oxford University Press, 1998.

Peterson, Victor. ‘A Discourse Theory of Moral Judgment.’ Ph.D. Dissertation, Northwestern University, 1998.

Post, Robert. Constitutional Domains: Democracy, Community, Management. Cambridge, MA: Harvard University Press, 1995.

Rawls, John. Political Liberalism. Paperback ed. New York: Columbia University Press, 1996.

________. A Theory of Justice. Revised ed. Cambridge, MA: Harvard University Press, 1999.

Rousseau, Jean-Jacques. Discourse on Political Economy and the Social Contract. Translated by Christopher Betts. New York: Oxford University Press, 1994.

Sunstein, Cass R. The Partial Constitution. Cambridge, MA: Harvard University Press, 1993.

________. Legal Reasoning and Political Conflict. New York: Oxford University Press, 1996.

________. One Case at a Time: Judicial Minimalism on the Supreme Court. Cambridge, MA: Harvard University Press, 1999.

Waldron, Jeremy. ‘Rights and Majorities: Rousseau Revisited.’ In Liberal Rights: Collected Papers 1981-1991, 392-421. New York: Cambridge University Press, 1993.

________. ‘Freeman's Defense of Judicial Review.’ Law and Philosophy 13 (1994): 27-41.

________. ‘Judicial Review and the Conditions of Democracy.’ The Journal of Political Philosophy 6, no. 4 (1998): 335-355.



[1]Learned Hand, The Bill of Rights (Cambridge, MA: Harvard University Press, 1958), 73-4.

[2]Alexander M. Bickel, The Least Dangerous Branch: The Supreme Court at the Bar of Politics, Second ed. (New Haven, CT: Yale University Press, 1986), 16.

[3]Jesse H. Choper, Judicial Review and the National Political Process (Chicago, IL: University of Chicago Press, 1980), 64. This defense of constitutionalism as a libertarian counterweight to majoritarianism reaches back, in the American context, to the defense of the newly proposed U.S. Constitution in Alexander Hamilton, James Madison, and John Jay, The Federalist Papers (New York: Bantam Books, 1982). Also to be found there is Hamilton’s insistence upon the independence of the judiciary from the legislative and executive branches: see especially Federalist 78. Of course the original justification for giving the judiciary the supreme power to interpret and enforce the U.S. Constitution was put forward by Chief Justice John Marshall in Marbury v. Madison, 5 U.S. 137 (1803): as the only legitimate interpreters of the law, the judiciary is the only governmental power in a position to decide what the law is. But, if the Constitution is the supreme law of the land, and if a legislatively enacted statute is in conflict with the Constitution, then a supreme judiciary must make the law internally consistent by striking down the statute. This argument from the requirements of legal consistency should be distinguished from Choper’s argument from the importance of minority rights. A third argument for judicial review is also (at least) implicit in Federalist 78: the judiciary plays a crucial role in the checks and balances established by the institutional separation of governmental powers, and judicial review is one of the judiciary’s more powerful weapons.

[4]Choper, Judicial Review, 68. Here Choper (and, as I will explain below, Ronald Dworkin) follows Bickel’s contention that only the judiciary has the relevant capabilities to be a forum of principle. As Bickel puts the point: ‘Courts have certain capacities for dealing with matters of principle that legislatures and executives do not possess. Judges have, or should have, the leisure, the training, and the insulation to follow the ways of the scholar in pursuing the ends of government,’ The Least Dangerous Branch, 25-6.

[5]The cases generating the most controversy concerning these issues are: Brown v. Board of Education, 347 U.S. 483 (1954) (segregation); Griswold v. Connecticut, 381 U.S. 479 (1965), Roe v. Wade, 410 U.S. 113 (1973), Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833 (1992), and Bowers v. Hardwick, 478 U.S. 186 (1986) (privacy); and, Buckley v. Valeo, 424 U.S. 1 (1976), and Nixon et Al. v. Shrink Missouri Government (2000) (campaign finance).

[6]For Bork’s views on the proper scope and method of judicial review, see Robert H. Bork, ‘Neutral Principles and Some First Amendment Problems,’ Indiana Law Journal 47 (1971). For a short summary of his conception of the tension between judicial review and democracy, see his ‘Judicial Review and Democracy,’ in Encyclopedia of the American Constitution, ed. Leonard W. Levy, Kenneth L. Karst, and Dennis J. Mahoney (New York: Macmillan, 1986). Barely reworked versions of these two articles can be found amongst much political polemic in Robert H. Bork, The Tempting of America: The Political Seduction of the Law (New York: The Free Press, 1990).

[7]For an interesting discussion of the Lochner era and the charge of judicial activism, see Cass R. Sunstein, The Partial Constitution (Cambridge, MA: Harvard University Press, 1993), especially chapter 2. One large difficulty for polemicists on the right is finding a way to endorse the court’s unanimous nullification of settled legislative will in Brown v. Board of Education while rejecting the privacy decisions which equally overturned settled legislative decisions concerning access to contraceptives and abortion. Leftist polemicists face a complementary problem of distinction: namely, how to endorse these Supreme Court decisions while retaining any role for majoritarian self-government.

[8]Bush v. Gore, 531 U. S. (2000).

[9]The terms were introduced in John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (Cambridge, MA: Harvard University Press, 1980), and appear to have been adopted as standard (though contested) in the literature since.

[10]Ely points out that judicial ‘activism’ and ‘self-restraint’ are categories that cut across interpretivism and noninterpretivism, virtually at right angles,’ Democracy and Distrust, 1. In principle, a strict interpretivist court may be quite active in rejecting the statutes of an assertive legislature, and a non-interpretivist court may adopt a passive role toward statues expanding the scope of constitutional provisions. In other words, the possible combinations of adjudicative methodologies and comparative judicial role will depend on the contingent history of legislative actions and past judicial decisions. From a quite different perspective Michael J. Perry, The Constitution in the Courts: Law or Politics? (New York: Oxford University Press, 1994), has argued that an interpretivist approach to adjudication that attends closely to original intent entails nether judicial passivism nor activism (which he calls ‘minimalism’ and ‘nonminimalism’). Contrary to other originalist theorists like Bork—who argued in 1971 (‘Neutral Principles’) that an originalist or noninterpretivist approach to adjudication led to a commendable form of judicial passivism—Perry in fact argues for an activist (i.e., nonminimalist) originalism.

[11]Starting from Rawlsian principles of justice, Samuel Freeman understands constitutionalism as a kind of pre-commitment to legitimacy-ensuring side constraints on majoritarian decision making. See his Samuel Freeman, ‘Constitutional Democracy and the Legitimacy of Judicial Review,’ Law and Philosophy 9 (1990-1991).

[12]For a good overview, see James Bohman, ‘Survey Article: The Coming of Age of Deliberative Democracy,’ The Journal of Political Philosophy 6, no. 4 (1998). Some of the representative work in deliberative democracy can be found in the following: Benjamin J. Barber, Strong Democracy: Participatory Politics for a New Age (Berkeley: University of California Press, 1984), Joshua Cohen, ‘An Epistemic Conception of Democracy,’ Ethics 97 (1986), Jürgen Habermas, Between Facts and Norms: Contributions to a Discourse Theory of Law and Democracy, trans. William Rehg (Cambridge, MA: The MIT Press, 1996), Jane Mansbridge, Beyond Adversary Democracy (Chicago: University of Chicago Press, 1983), Carlos Santiago Nino, The Constitution of Deliberative Democracy (New Haven: Yale University Press, 1996), and the collection of essays James Bohman and William Rehg, eds., Deliberative Democracy: Essays on Reason and Politics (Cambridge, MA: The MIT Press, 1998).

[13]This paper assumes the review of legislative statutes by an independent, unelected judicial body for their constitutionality as the paradigmatic instance of judicial review to be scrutinized from the perspective of deliberative democracy. Closely allied to this function is the judicial review of the constitutionality of executive and administrative actions, rules, and regulations. I will not be concerned with the many other functions a body such as the United States Supreme Court often carries out—even when they involve constitutional concerns—such as: serving as the apex of a federal appellate judicial system by deciding between conflicting lower court holdings, ensuring the internal consistency of adjudicative law, reviewing state court decisions for consistency and constitutionality, reviewing criminal trial procedures, settling disputes between federal and state governments, settling jurisdictional disputes between the legislative, executive and judicial branches, reviewing administrative actions for their consistency with statutory and common law, and so on. These latter functions are either tied directly to the basic functions of an independent judiciary within a tripartite separation of powers, or are derivative from them depending on particular nation-state’s federal organization. As such, these functions must be justified in terms of a general political theoretic doctrine of the constitutional separation of powers, a doctrine not restricted to constitutional democracies alone.

[14]Deliberative conceptions of democracy have often been attacked as empirically inadequate to the variety of extant political practices ranged under the label of democracy. Although this charge is well beyond the scope of this paper, numerous theorists have argued that such conceptions are not, at least, wildly idealistic. See especially the books cited above by Barber, Habermas, Mansbridge, and Nino. Of course, even if actual democratic practices do not often live up to the standards of theories of deliberative democracy, this does not undercut the normative claims made by those theories.

[15]Other important jurisprudential theorists who emphasize richer, more deliberative conceptions of constitutional democracy in their accounts of judicial review include: Bruce Ackerman, We the People: Foundations (Cambridge, MA: Harvard University Press, 1991). and We the People: Transformations (Cambridge, MA: Harvard University Press, 1998), Frank I. Michelman, Brennan and Democracy (Princeton, NJ: Princeton University Press, 1999), Robert Post, Constitutional Domains: Democracy, Community, Management (Cambridge, MA: Harvard University Press, 1995), and Cass R. Sunstein, One Case at a Time: Judicial Minimalism on the Supreme Court (Cambridge, MA: Harvard University Press, 1999). I intend to deal with these theorists in a future paper focusing on which adjudicative methods and forms of legal interpretation are appropriate in the light of a conception of constitutional review informed by deliberative democracy.

[16]It may help to think of the two distinctions as forming a four-cell matrix, and the theories considered as metaphorically occupying different parts of it. Thus if we place Ely’s theory in the first quadrant, combining an aggregative account of decision processes with a procedural account of legitimacy, then Perry’s belongs in the quadrant diagonally across, combining a deliberative account of decision processes with a substantive account of legitimacy. While Dworkin also employs a substantive account of legitimacy (though one different than Perry’s), he puts forward an account of decision making processes that recommends both aggregative and deliberative modes. His theory thus occupies the two quadrants of the matrix on the side of substantive legitimacy. Habermas’s theory, by contrast, occupies the other half comprising a mixed account of decision processes and a proceduralist theory of democratic legitimacy.

[17]An early and influential example of this distinction in the deliberative democracy literature can be found in Jon Elster, ‘The Market and the Forum: Three Varieties of Political Theory,’ in Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg (Cambridge, MA: The MIT Press, 1997) (originally published 1986). Elster distinguishes between ‘the economic theory of democracy’ [which] … is a market theory of politics, in the sense that the act of voting is a private act similar to that of buying and selling,’ (25) and a forum theory of politics in which ‘rather than aggregating or filtering preferences, the political system should be set up with a view to changing them by public debate and confrontation,’ (11). As will become clear in the paper as I consider different theories of constitutional review that have elements of deliberative democracy, questions concerning what exactly the various forms of deliberation are, which kinds are preferred, who should do the deliberating, about what, in what fora, and so on, are all contested issues between the various jurisprudential theories.

[18]Jeremy Waldron, ‘Rights and Majorities: Rousseau Revisited,’ in Liberal Rights: Collected Papers 1981-1991 (New York: Cambridge University Press, 1993), characterizes this distinction as one between Bentham’s and Rousseau’s notions of democratic decision making: see especially 394-400. So, for instance, Waldron explains the different conceptions of voting on each account of democratic processes: ‘Bentham’s voter is taken to be expressing a preference of his own; his vote represents a possible individual satisfaction. Rousseau’s voter is not supposed to express his personal preference; rather he affirms his personal belief about the best way to promote the general good. The Benthamite political system sums votes as utilitarianism sums satisfactions, while the Rousseauian political system counts votes to determine the preponderance of opinion,’ 398-399. Rousseau’s work is filled with many different ideals of democratic decision making, and these different ideals are often conflated. For instance, Rousseau also seems to think that citizens’ assemblies should focus on articulating their underlying solidarity which is ultimately based upon a certain kind of consensus of feeling arising out of the similarity of their mores, education, socialization, and collective history. In this paper, I make no claim to accurately represent all of Rousseau’s (nor Locke’s) actual positions. Rather, I will focus on the notion of reasoned deliberation rather than authentic collective self-reflection as the ‘Rousseauian’ model of deliberative democratic processes.

[19]I use the term ‘legitimacy’ (and its cognates) here in its normative, not its descriptive sense. In general, a legitimate political decision will, either directly or indirectly, lead to state actions that are normatively permissible, are defensible on the basis of good reasons, and/or give citizens good, prima facie obligatory moral reasons for obeying whatever actions are commanded by the decision. Thus, I am not here directly concerned with factual matters, about, for instance, the extent of social obedience to the state, the degree to which a state is perceived by its members or others to have a monopoly on the coercive use of force within its territory, or the extent of motivations for conformity versus disruption, and so on.

[20]Establishing that my contestable interpretations of the conception and importance of these two distinctions to Locke and Rousseau are correct goes beyond the ambit of this paper. Interested readers might refer to the following passages, among others, from the two: Locke on substantive legitimacy constraints on legislation via the natural law: §§ 134-142, Locke on aggregative, majoritarian democratic processes: §§ 95 -99, Rousseau on pure procedural legitimacy through political autonomy: Book I, Chapters 5-8, Book II, Chapters 1-2, 4, and 6 Book III, Chapter 1 and, Rousseau on deliberative democratic processes: Book I, Chapters 7-8, Book II Chapter 3, Book III Chapters 1-5, 12-15 and 18, Book IV Chapters 1-3. See The Second Treatise of Government in John Locke, Two Treatises of Government, ed. Peter Laslett (New York: Cambridge University Press, 1988) and The Social Contract in Jean-Jacques Rousseau, Discourse on Political Economy and the Social Contract, trans. Christopher Betts (New York: Oxford University Press, 1994).

[21]This way of putting the distinction roughly corresponds to the distinction made in David M. Estlund, ‘Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority,’ in Deliberative Democracy: Essays on Reason and Politics, ed. James Bohman and William Rehg (Cambridge, MA: The MIT Press, 1997) between a procedural theory of legitimacy and a ‘correctness’ theory of legitimacy which claims that a political decision ‘is legitimate if correct, and otherwise not,’ 174.

[22]As will become clear by the time I have considered Perry’s, Dworkin’s, and Habermas’s respective accounts of democratic decision-making processes, the two-part distinction between aggregation and deliberation is insufficiently differentiated to both accurately characterize the extant diversity of public reasoning and to theoretically articulate the various kinds of reason-responsiveness different institutional actors ought to play. In particular, I will suggest that a theory of constitutional review needs to account for at least four kinds of public reasoning processes: preference aggregation, deliberative consensus, ethical-political self-clarification, and fair bargaining.

[23]Consider John Rawls’ example of a system of fair gambling as an exemplar of pure procedural justice, where ‘there is no independent criterion for the right result: instead there is a correct or fair procedure such that the outcome is likewise correct or fair, whatever it is, provided that the procedure has been properly followed,’ John Rawls, A Theory of Justice, Revised ed. (Cambridge, MA: Harvard University Press, 1999), 75. Here we seem to have an example of unalloyed procedural legitimacy, except that the legitimacy of the outcome results not only from following the procedure, but also from the supposition that the procedure will satisfy or operationalize a substantive ideal: namely, fairness or correctness or justice.

[24]Consider one of the most comprehensive and ambitious theories of substantive legitimacy: Aquinas’s natural law theory. It clearly recognizes, on the one hand, the perfection and immutability of substantive natural law principles and, on the other, the indeterminacy and mutability of applications of those principles to human reality. Hence, it recommends certain decision procedures, adherence to which confers legitimacy on the outcomes: for example, legal ‘dispensations’ (deviations) from the letter of the law by authorized rulers, and, in general the claim that one of the three roots of legal justice is to be found in the criterion of establishment by a just authority. See specially Questions 94-97 of Saint Thomas Aquinas, Summa Theologiae, 60 vols., vol. 43 (New York: McGraw-Hill, 1964).

[25]Ely, Democracy and Distrust, 4-5.

[26]Ibid., 11.

[27]Following Ludwig Wittgenstein’s later reflections on language and rule-following, H. L. A. Hart argues that the open texture of legal rules—their characteristic incapacity to fully specify all correct applications of their provisions to particular cases—is entailed by the law’s use of general terms. From this essential, inexpungible characteristic of language and our incapacity to foresee all possible changes in social conditions, Hart argues that when judges apply legal rules to specific cases, they will inevitably have wide discretion in choosing how to interpret statutes in new situations within the penumbra of the statute’s meaning. See chapter VII ‘Formalism and Rule-Scepticism’ in H. L. A. Hart, The Concept of Law, Second ed. (Oxford: Clarendon Press, 1994). (First Edition, 1961). An early and influential attack on Hart’s doctrine of judicial discretion is found in Ronald Dworkin, ‘The Model of Rules,’ University of Chicago Law Review 35 (1967).

[28]Ely, Democracy and Distrust, 13. Ely analyzes a number of such open-textured provisions in the United States Constitution: the First Amendment’s protection of speech, the prohibition on cruel and unusual punishments in the Eighth, the Ninth Amendment’s provision that ‘the enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people,’ and, the due process, privileges and immunities, and equal protection clauses of the Fourteenth Amendment.

[29]Ibid., 43-72.

[30]Ibid., 87.

[31]Ibid., 74.

[32]Ibid., 87.

[33]Ibid., 102-3.

[34]Ely extends this same reasoning in order to justify judicial review of relations between executive administrations, the legislature, and the people. See, for example, ibid., 131-134 and 136-170.

[35]See for instance Ronald Dworkin, A Matter of Principle (Cambridge, MA: Harvard University Press, 1985), 59-69.

[36]Ely, Democracy and Distrust, 102, footnote *.

[37]Ibid., 187, endnote 14.

[38]Ibid., especially 181-3. See also Ely’s rather undeveloped response to the objection that utilitarian theories of democracy are indifferent to individual rights on 15-18 and 306-311 of his John Hart Ely, On Constitutional Ground (Princeton: Princeton University Press, 1996). Here, as in Democracy and Distrust, Ely can takes advantage of the ambiguities of a theory that operates at both the level of abstract political theory and of constitutional theory within a pre-given context of settled rights in a specific country. Thus he can simultaneously insist—without really noting the tension—that ‘Some nonpolitical rights undoubtedly should be protected’ even though such protection will not be sufficiently secured through majoritarian political processes (On Constitutional Ground, 15), and that courts should only be concerned to ‘enforce for minorities those rights that the majority has seen fit to guarantee for itself’ (ibid., 16). Apparently the ‘should’ in the first quote has merely the force of an admonition to the majority. Of course, if one is the fortunate heir of a constitutional assembly where the majority did in fact see fit to enforce an extensive schedule of individual liberal rights, then it will not seem particularly problematic to endorse a political theory that can only understand the justification of individual rights in terms of benevolent majoritarian preferences.

[39]Note that this type of preference-satisfaction voting will easily lead to interest groups and blocs in conflict with one another, since individuals recognize the effectiveness of grouping together with others who have sufficiently similar preferences and preference rankings. This may well lead to the kind of factional power-politics the writers of The Federalist Papers were keenly worried about.

[40]As Rousseau puts it in Book IV, chapter 2 of The Social Contract: ‘When a law is proposed in the assembly of the people, what they are asked is not precisely whether they accept or reject the proposal, but whether it is or is not in conformity with the general will, which is their will; everyone, by voting, gives his opinion on the question; and counting the votes makes the general will manifest. When an opinion contrary to mine prevails, therefore, it proves only that I had been mistaken, and that the general will was not what I had believed it to be. If my particular will had prevailed, I should have done otherwise than I wished; and then I should not have been free. This argument, it is true, presupposes that all the characteristics of the general will are present also in majority decisions; when they cease to be, whatever view may be adopted, liberty exists no longer,’ 138.

[41]The phrase is from Sunstein, The Partial Constitution. See especially chapter 1 for an interesting discussion of how the notion of deliberative democracy involves a commitment to a ban on governmental action based on ‘naked preferences.’

[42]Skepticism towards the presupposition of special judicial insight into moral principles is nicely captured in Nino’s phrase ‘epistemic elitism’: ‘The common view that judges are better situated than parliaments and other elected officials for solving questions dealing with rights seems to arise from an epistemic elitism. It assumes that in order to arrive at correct moral conclusions, intellectual dexterity is more important than the capacity to represent vividly and to balance impartially the interests of all those affected by a decision. It is understandable that scholars who celebrate the marvels of judicial review should identify themselves more closely with judges than politicians and, thus, are inclined to think, as Michael Walzer remarks, that what they deem to be right solutions—their own—would be more readily obtained by judges than politicians,’ The Constitution of Deliberative Democracy, 189.

[43]Perry, The Constitution in the Courts, 107.

[44]Ibid., 111.

[45]As should be clear, Perry’s account of judicial deliberation and decision is heavily indebted to Aristotle’s account of the capacity of phronesis, even to the extent that both claim a natural hierarchy amongst humans based upon their comparative capacities for insight into the proper relation between moral universals and particulars. See Aristotle’s Nicomachean Ethics in The Complete Works of Aristotle, trans. Jonathan Barnes, 2 vols., Bollingen Series, vol. 2 (Princeton, NJ: Princeton University Press, 1984), especially Book V. It is thus little surprise that both share a skepticism about democracy except as a stability-enhancing handout to the people, who are thereby led to believe that they have an actual role in determining fundamental principles and making policy. In the Politics, after conceding that democracies may exhibit epistemic gains from the collective deliberations of more individuals, Aristotle asks ‘what power should be assigned to the mass of freemen and citizens, who are not rich and have no personal merit?’ His answer clearly expresses his fear of the inferior majority: ‘There is still a danger in allowing them to share the great offices of state, for their folly will lead them into error, and their dishonesty into crime. But there is a danger also in not letting them share, for a state in which many poor men are excluded from office will necessarily be full of enemies. The only way of escape is to assign them some deliberative and judicial functions,’ but no direct role in the great offices; Politics, Book III, Chapter 11, 1281b 24-31, page 2034 of Vol. 2 in The Complete Works of Aristotle.

[46]Michael J. Perry, The Constitution, the Courts, and Human Rights: An Inquiry into the Legitimacy of Constitutional Policymaking by the Judiciary (New Haven, CT: Yale University Press, 1982), 100.

[47]Ibid., 101.

[48]These arguments are most clearly articulated in Michael J. Perry, The Idea of Human Rights: Four Inquiries (New York: Oxford University Press, 1998). In this short book, Perry argues that human rights can only be understood in religious terms, that, so understood, human rights are universally binding and context transcendent, and that human rights are grounded in ‘the very order of the world—the normative order of the world,’ 38. But note that all Perry needs to make the argument referred to here—the argument from the claim that there are right answers to fundamental value questions, to the justification of judicial review—is the claim to strong moral cognitivism, not any specific version of moral realism. In other words, his argument needs only the premise that, on fundamental questions of individual rights, there are right and wrong answers. See his arguments in The Constitution, the Courts, and Human Rights, especially pages 96-114.

[49]Perry, The Constitution, the Courts, and Human Rights, 102.

[50]It is interesting to note that Perry’s argument here shares with Rousseau the same combination of a neo-classical distrust in the original reflective capacities of the masses with a conception of political deliberation as reflection on a homogeneous, collectively practiced form of ethical life. In Rousseau this theoretical combination results in those puzzling passages where Rousseau considers that the polis will require some singular, original genius of a lawmaker in order to give the people the laws at first that they are later supposed to give to themselves. See Book II, Chapter 7, ‘The Legislator’ of The Social Contract, 76-79. Perry’s argument simply substitutes a paternalistic collective body—the judiciary—for a single father of the laws. One significant difference, of course, is that Rousseau believes that the masses will eventually develop the requisite reflective capacities through their enculturation within a free society, whereas Perry seems to have a priori reservations about the very possibility of an egalitarian distribution of moral capacities.

[51]Term appointments of members of constitutional tribunals, and appointment by legislatures rather than the executive, are common arrangements in European constitutional democracies.

[52]See his arguments on pages 196-201 of The Constitution in the Courts. The relevant clause of the Canadian Charter allows the legislative branch to pass a statute that would be in conflict with specific judicial decisions concerning the requirements of fundamental freedoms, legal rights, and equality rights (though not democratic rights and mobility rights). Such an exceptional act by the legislature in passing a law ‘notwithstanding’ judicial specifications of rights becomes inoperative after five years unless restated.

[53]Ibid., 201.

[54]Ibid., 197. I return to this institutional proposal below in my consideration of Habermas’s arguments. Although I endorse the proposal, I do so not based on considerations of respective capacities for moral judgment as Perry does.

[55]See for instance, in The Constitution in the Courts, his discussion of judgment and moral indeterminacy in Chapter 5, especially pages 72-76. The virtue requirement for phronetic capacities is made clear in Perry’s approving quotation on page 111 of A. Kronman: ‘To possess good judgment … is not merely to possess great learning or intelligence, but to be a person of a certain sort, to have a certain character, as well.’ (citing A. Kronman, ‘Living in the Law,’ 54 University of Chicago Law Review 835, 837 (1987)).

[56]See especially chapters 11 (‘Of the difference of Manners’) and 12 (‘Of Religion’) of Thomas Hobbes, Leviathan, ed. Richard Tuck, Cambridge Texts in the History of Political Thought (New York: Cambridge University Press, 1991). Of course, Hobbes has a multifaceted (and perhaps internally contradictory) psychological, epistemological and sociological explanation for the wars of religion: ‘And in these foure things, Opinion of Ghosts, Ignorance of second causes, Devotion towards what men fear, and Taking of things Casuall for Pronostiques, consisteth the Naturall seed of Religion; which by reason of the differrent Fancies, Judgements, and Passions of severall men, hath grown up into ceremonies so different, that those which are used by one man, are for the most part ridiculous to another.’ ‘So that I may attribute all the changes of Religion in the world, to one and the same cause; and that is, unpleasing Priests; and those not onely amongst Catholiques, but even in that Church that hath presumed most of Reformation,’ chapter 12, pages 79 and 86.

[57]See John Rawls, Political Liberalism, Paperback ed. (New York: Columbia University Press, 1996), especially xxiii-xxx, 36-38, and 54-58.

[58]See Rousseau, The Social Contract, Book IV, chapter 8, especially 166-167. Recall that the distinctly non-neutral dogmas of this civil profession should, for Rousseau, include: ‘the existence of the Divinity, powerful, intelligent, beneficent, prescient, and provident, the life to come, the reward of the just and the punishment of the wicked, the holiness of the laws and the social contract; such are the positive dogmas. As for those excluded, I limit them to one: intolerance; it belongs to the religions we have rejected,’ 167.

[59]Habermas makes much the same argument concerning Perry’s ethically constricted notion of practical reason and its problems for a democratic understanding of judicial review: ‘Perry sees the constitutional judge in the role of a prophetic teacher, whose interpretations of the divine word of the Founding Fathers secures the continuity of a tradition that is constitutive of the community’s life. … By assuming it should strive to realize substantive values pregiven in constitutional law, the constitutional court is transformed into an authoritarian agency,’ Between Facts and Norms, 258.

[60]The constraints of precedent might be thought of extending vertically through time, while consistency might be thought of as a horizontal requirement: see Ronald Dworkin, Freedom's Law: The Moral Reading of the American Constitution (Cambridge, MA: Harvard University Press, 1996), 83.

[61]Ronald Dworkin, Law's Empire (Cambridge, MA: Harvard University Press, 1986) is dedicated to spelling out the adjudicative constraints involved in the comprehensive notion of legal integrity. Especially noteworthy here is the suggestion that judges think of their decisions as part of a chain novel written by many different authors, where each ‘chapter’ (i.e., each decision backed by opinion) aims to make the best sense of the story (i.e., the developing legal system of a particular nation-state) as a whole. See especially chapter 7 ‘Integrity in Law’ where he claims that a judicial decision must not only ‘fit’ the ongoing practice of the law (and so be constrained by text, precedent, and consistency), but also must ‘justify’ that practice as the best that it can be (and so be constrained by the best interpretation of relevant moral-political principles instantiated in that practice.).

[62]Dworkin, Freedom’s Law, 7.

[63]Ronald Dworkin, Taking Rights Seriously (Cambridge, MA: Harvard University Press, 1978) develops the distinction between principles, policies, and legal rules (see chapters 2 and 3, especially pages 22-31 and 71-79), and defends the conception of rights as deontic requirements of principle which trump considerations of policy (see especially chapters 4 and 6).

[64]Dworkin, Freedom’s Law, 19.

[65]Ibid., 20.

[66]Dworkin himself makes this connection explicit on page 20 of Freedom’s Law: ‘Rousseau’s idea of government by general will is an example of a communal rather than a statistical conception of democracy. The statistical reading of government by the people is much more familiar in American political theory.’ I assume that this more familiar American political theory is the pluralistic theory of democracy inspired by Locke.

[67]Ibid., 24.

[68]Ibid., 25.

[69]For Dworkin’s preferred conception of distributive equality see his ‘What Is Equality: Part I, Equality of Welfare,’ Philosophy & Public Affairs 10, no. 3 (1981) and ‘What Is Equality: Part II, Equality of Resources,’ Philosophy & Public Affairs 10, no. 4 (1981). Note, however, that he believes that the specific conception of equality put forward in these two articles—one requiring a rather extensive redistribution of wealth from the rich to the poor—is not a requirement of the United States Constitution, and so should not form the basis for American judicial review: Freedom’s Law, 36. His point here about the requirement of ‘stake’ is thus intended to express a more abstract requirement of the concept of equality in any adequate concept of democracy.

[70]On pages 21-33 of Freedom’s Law, Dworkin considers, and rejects, three main types of arguments given to support a merely statistical conception of democratic processes: from the high values of collective political liberty, political equality, and community.

[71]Ibid., 17, emphasis added.

[72]I use the adjective ‘Lockean’ to modify natural law here to indicate that Dworkin’s theory does not descend from Aristotle and Aquinas, but rather insists, with Locke, that certain substantive moral tenets are binding on any and all actions, independently of any collective processes of deliberation or decision, and that these take the shape of rights held by individuals as trumps over collective actions. As he puts this point, ‘if the political principles embedded in the constitution are law … in spite of the fact that they are not the product of deliberate social or political decision, then the fact that law can be, in that sense, natural argues for the constraint on majority power that a constitution imposes,’ Dworkin, Taking Rights Seriously, viii, emphasis added.

[73]Dworkin, Freedom’s Law, 17, emphasis added.

[74]Dworkin inherits this distinction between the legislature as the forum of policy and the judiciary as the forum of principle from Alexander Bickel’s defense of judicial review in The Least Dangerous Branch. Bickel, of course, was much less concerned to reconcile the tension between constitutionalism and democracy given his libertarian conception of the former and his pluralist conception of the latter. Thus Bickel simply equates the counter-majoritarian character of judicial review with its counter-democratic character.

[75]Dworkin, Freedom’s Law, 344.

[76]Ibid., 345.

[77]Dworkin cites the debate over abortion after the Roe v. Wade decision as an example: ‘the public discussion of [abortion] in America has involved many more people, and has been more successful at identifying the complex variety of moral and ethical issues involved, than in other countries where a political compromise was engineered. In France, for example ... ,’ ibid., 345.

[78]Jeremy Waldron, ‘Judicial Review and the Conditions of Democracy,’ The Journal of Political Philosophy 6, no. 4 (1998), 339. Not only does the constitutionalization of a controversial issue sometimes constrain the terms and arguments that may be employed in broad public debates, but it may also lead to significantly degraded legislative debate on an issue. For instance, during a recent Kentucky congressional debate concerning possible enactment of a law prohibiting nude dancing, one legislator opposed to the measure actually voted for it, and was reported to have given this explanation for the vote: the legislator ‘said the bill is clearly unconstitutional, and … voted for it only so it can be struck down in the courts if it becomes law,’ John Cheeves, ‘Committee Oks Ban on Nude Club Dancing,’ Lexington Herald-Leader, February 11 2000.

[79]In The Least Dangerous Branch, Bickel argues from this distinction to a justification for judicial review as ‘a principle-defining process that stands aside from [and above] the marketplace of expediency,’ 69.

[80]The biases and distortions in judicial opinions and decisions have come under withering empirical scrutiny by both legal realists and critical legal theorists.

[81]Of course, Dworkin does not suggest that legitimacy is always disconnected from procedurally-correct enactment; he does, after all, endorse some ordinary democratic procedures with respect to some (policy) issues. In other words, his theory of democratic legitimacy is mixed in that it allows for some procedurally-secured legitimacy, but with ex ante substantive constraints on outcomes. Thus the internal link between legal legitimacy and actual procedural genesis is only severed with respect to constitutional essentials: ‘matters of principle’ in Dworkin’s language. I am not suggesting that Dworkin is entirely deaf to the notion of popular sovereignty; only with respect to those issues of moral controversy and fundamental principle best left to judicial guardians of the moral law. My thanks to an anonymous reviewer for insisting on this clarification.

[82]Freeman, in ‘Constitutional Democracy and the Legitimacy of Judicial Review,’ recognizes this loss more clearly than Dworkin. He argues in a Rawlsian fashion for a defense of constitutional review based upon a substantivist account of democratic legitimacy. According to Freeman, democracy is not a utilitarian procedure based on majoritarianism, but a form of popular sovereignty in which citizens precommit themselves to certain basic norms of individual liberty. Judicial review can then be understood as one possible mechanism by which a democratically constituted populace might seek to enforce the substantive constraints enshrined in their constitutional precommitment. I think that this proposal also succumbs to problems similar to those found in Dworkin of recommending a kind of judicial paternalism, and precisely because of its insistence on substantive criteria for legitimacy. For an interesting appraisal, see Jeremy Waldron, ‘Freeman's Defense of Judicial Review,’ Law and Philosophy 13 (1994).

[83]Cass Sunstein gives one very plausible reason why the more abstract principles enshrined in constitutional provisions are subject to much less disagreement and conflict than the more specific interpretations and applications of those principles: the former are often the result of incompletely theorized agreements, and so do not raise as many points of contention. See Cass R. Sunstein, Legal Reasoning and Political Conflict (New York: Oxford University Press, 1996), especially 35-61. In One Case at a Time, he suggests that ‘constitution-making is often possible only because of the technique of producing agreement on abstractions amid disagreements about particulars,’ 11, that is, of employing incompletely theorized agreements.

[84]Habermas, Between Facts and Norms, 110.

[85]Habermas’s account of the form of legality largely follows Kant’s analysis of the differences between morally coordinated and legally coordinated action, though Habermas gives a significantly more ‘sociological’ reading to legal relations, one informed by legal positivism. Law’s basic function is to stabilize the behavioral expectations of socially interacting agents in a way that unburdens them from the high cognitive, motivational, and organizational demands of social action coordinated through face-to-face communicative action relying only on agents’ moral competences. In this way, law allows for the development of domains of interaction (such as the economy) where actors are free, within constraints, to treat others as merely facilitators of, or impediments to, the realization of their own desires. Unlike morality, law addresses agents simply as purposive-rational actors rather than autonomous moral agents, and attends only to the external relations between social actors. As law contains publicly promulgated rules of action, it relieves actors of the cognitive burdens of figuring out what the right thing to do is in typical situations, while giving reasonable assurance that they can expect like norm-conformative behavior from others and so increases the reliability of interactions. Further, unlike morality, law abstracts from the reasons that motivate actors, demanding only that, for whatever reason, they comply. The only weakly motivating ‘force’ of good moral reasons is replaced by law through the coercive threat of sanction. Finally, by structuring the emergence of organized forms of cooperation through secondary rules that allow for the production of primary rules, define jurisdictional powers, and found corporations, associations and so on, law relieves actors of the organizational demands that purely moral action would require, while creating the possibility for large-scale action coordination and regulation across complex and far-flung social institutions. See Habermas, Between Facts and Norms, especially 104-118.

[86]Habermas’s analysis of the normative components of law in Between Facts and Norms differs from his earlier analyses of the relation between law and morality, where there was a danger of following Kant too closely by simply subordinating legal relations to moral demands. Whereas in his earlier work, Habermas had followed Kant both in the typology of legal versus moral action, and in the normative conception of the relationship between legal and moral norms, Habermas now agrees only with Kant’s analysis of the legal form, while insisting that legal norms should not be conceived of as merely derived from more fundamental moral norms. Habermas now claims that the basic legal principle—the principle of democracy (quoted in the text above)—and the basic moral principle—the principle of universalization (U), ‘A norm is valid when the foreseeable consequences and side effects of its general observance for the interests and value-orientations of each individual could be jointly accepted by all concerned without coercion,’ Jürgen Habermas, The Inclusion of the Other: Studies in Political Theory, ed. Ciaran Cronin and Pablo De Greiff (Cambridge, MA: The MIT Press, 1998), 42—are equiprimordial specifications, tailored to the different forms of legal and moral norms, of a more general principle of discursive legitimacy (D): ‘Just those action norms are valid to which all possibly affected persons could agree as participants in rational discourse,’ Between Facts and Norms, 107.

[87]Habermas, Between Facts and Norms, 448. He continues ‘The democratic process bears the entire burden of legitimation. It must simultaneously secure the private and public autonomy of legal subjects. This is because individual private rights cannot be adequately formulated, let alone politically implemented, if those affected have not first engaged in public discussions to clarify which features are relevant in treating typical cases as alike or different, and then mobilized communicative power for the consideration of their newly interpreted needs. The proceduralist understanding of law thus privileges the communicative presuppositions and procedural conditions of democratic opinion- and will-formation as the sole source of legitimation. The proceduralist view is … incompatible with the Platonistic idea that positive law can draw its legitimacy from a higher law,’ (second emphasis added), 450. Of course, the same reasons underwrite the rejection of Platonism here and the rejection of Kant’s direct derivation of legal norms from moral imperatives discussed in the previous footnote.

[88]Ibid., 449.

[89]Ibid., 123.

[90]I believe that this fifth category of ecological and social rights raises particular problems in formulating a legitimate method of constitutional interpretation, especially for contemporary legal systems that have become increasingly materialized in the wake of the development of the modern welfare state. I indicate what these problems might be at the end of the paper, but can only adequately treat them in a separate paper on adjudicative methodologies.

[91]Like Habermas, Joshua Cohen, ‘Democracy and Liberty,’ in Deliberative Democracy, ed. Jon Elster (New York: Cambridge University Press, 1998) argues that there is not a contradiction between individual liberties and the principle of democracy. In this article, Cohen uses the examples of liberties to free religious exercise, to wide freedoms of expression, and to pursue popularly denigrated moral tastes and pursuits. The key to his argument is the claim that democracy, in order to operationalize its basic aim that state power follow from collective decisions arising from citizens considered as equals, must be deliberative and not merely aggregative. But deliberativeness implies that the proper conditions of public communication and the public reason conditions for public justification are both met. Finally, according to Cohen, these conditions require not just political liberties but individual liberties as well. So Cohen employs a structurally similar strategy to Habermas’s: individual libertarian rights are justified as parts of the legitimacy conditions for deliberative democracy. The most important difference is that while Habermas sees the internal connection between private and public autonomy in terms of the legal order that is required to structure democracy, Cohen sees the internal connection between deliberative democracy and the liberty of the moderns in terms of the requirements of public reason (understood in a largely Rawlsian manner).

[92]This is precisely the thought that motivates Rousseau to apparently reject all forms of constitutionalism in the name of popular sovereignty: ‘Public decisions … cannot put the sovereign under any obligation towards itself; and in consequence, it is contrary to the nature of the body politic that the sovereign should impose on itself a law that it cannot infringe,’ The Social Contract, Book I, Chapter 7, page 57. The crucial difference here is that, whereas in Habermas’s proceduralist republicanism the constitutional structuring of democracy is intended to legally insure political legitimacy, in Rousseau’s civic republicanism the expectations of civic virtue, civic homogeneity, and small size carry an extremely heavy burden in ensuring that political deliberations are truly oriented towards the general will. In large modern, socially complex, and teleologically diverse nation-states, such heightened expectations are unwarranted. In fact, Rousseau himself was pessimistic about the possibility of realizing such expectations even in small, modern city-states: ‘All things rightly considered, I cannot see how it is henceforward possible among us for the sovereign to retain the exercise of its rights unless the state is very small,’ ibid., Book III, Chapter 15, page 129.

[93]In his distinction between justification and application discourse and his analysis of the latter, Habermas is heavily indebted to Klaus Günther, The Sense of Appropriateness: Application Discourses in Morality and Law, trans. John Farell (Albany: State University of New York Press, 1993).

[94]Recall that this antithesis between individual rights and democracy is the starting point for most theorists of judicial review who rely upon a pluralist concept of democracy, e.g., Alexander Bickel and Jesse Choper.

[95]Habermas, The Inclusion of the Other, 260-261.

[96]Habermas, Between Facts and Norms, 279-280.

[97]Ibid., 155-156.

[98]Strictly speaking, Ely is concerned with the propriety of some reasons used to justify some unequal distributions of the bounty of representative government, specifically with respect to whether such distributions are the results of mere animus, prejudice, or other unconstitutional motivation with respect to a society’s habitual unequals. He in fact recommends, within the context of the history of American constitutional jurisprudence concerning the equal protection clause of the 14th Amendment, that a series of differentiated suspect classifications can be employed to sniff out illicit reasons justifying certain legislation. However, the larger point I make in the text is that Habermas recommends a constitutional review of the quality and propriety of justifying reasons employed in legislative justification discourses across any number of issues, and recommends it not on the basis of a contingent constitutional development in the light of a specific history of racial subordination, but because such a review of justifying reasons is an inherent part of constitutional application discourses in the context of deliberative democracy.

[99]Habermas, Between Facts and Norms, 262.

[100]See ibid., especially chapter 8, pages 359-387.

[101]It is strange that the centrality of this two-track model of the public sphere—a model that occupies over one quarter of Between Facts and Norms—is completely elided in many readings of Habermas’s recent political theory. Consider this strikingly confused example from a recent Oxford monograph that cites Between Facts and Norms as its source: ‘Habermas’s emphasis on elections as the main channel of influence from the public sphere to the state would also strike many political scientists as old-fashioned,’ John S. Dryzek, Deliberative Democracy and Beyond: Liberals, Critics, Contestations (New York: Oxford University Press, 2000), 26.

[102]Habermas contemplates such options on pages 241-242 of Between Facts and Norms.

[103]Ibid., 242.

[104]Ibid., 258.

[105]Ibid., 266.

[106]Ibid., 261.

[107]Vic Peterson, in personal correspondence, has pressed the disanalogy of constitutional review and application discourses. Although he claims that constitutional review follows the logic of a justification discourse, he also believes that this does not entail that it should not be performed by a judicial body. I don’t see how a theory of constitutional review, committed to the principle of popular sovereignty, and to a procedural conception of democratic legitimacy could admit that an unaccountable institution could legitimately justify constitutional content. My thinking on these issues has been greatly spurred by discussions with him, and by his notable criticisms of Habermas’s and Günther’s theory of moral application discourses in Victor Peterson, ‘A Discourse Theory of Moral Judgment’ (Ph.D. Dissertation, Northwestern University, 1998).

[108]Frank Michelman has made this problem central to his conception of the relationship between liberal jurisprudence, as embodied by William Brennan, and political self-government. In particular, he argues that even though a political theory of self-government under modern conditions of value pluralism might achieve agreement on basic principles at higher levels of abstraction—for instance in constitutional provisions—such a strategy cannot prevent inevitable disagreement concerning the correct specification of those principles necessary to decide their applicability to real world issues. Liberal jurisprudence is seen, under the right conditions, as a procedural solution that is consistent both with the legal requirements for decisiveness and the normative requirement that all reasonable citizens feel they can give their considered respect to a legal system, even when they vehemently disagree with some of its laws. As long as the political procedures—including the procedures of judicial review—can be seen as making a good faith effort to correctly specify the abstract basic principles through mechanisms open ‘to the full blast of sundry opinions and interest-articulations in society,’ (Brennan and Democracy, 60) citizens can understand themselves as self-governing. Michelman has also argued that Habermas’s solution to this problem of constitutional specification is to be found in his concept of a shared, nation-specific political identity arising out of a shared constitutional history. According to this reading of Habermas, ‘constitutional patriotism’ provides citizens with the faith that, even though they vehemently disagree about their applicability, all are arguing about the same constitutional principles; what they disagree about is who they are as a people. See Frank I. Michelman, ‘Morality, Identity and 'Constitutional Patriotism',’ Denver University Law Review 76, no. 4 (1999), especially 1022-1028. Even if this is Habermas’s ‘solution’ to the problem (which I don’t think it is), I am wary of its implications for judicial interpretation. For if Michelman’s account is right, then it appears that Habermas is recommending that a supreme court see its constitutional decisions as hanging on historically contingent and ethically suffused visions of a nationality-specific good life; precisely the kind of ‘value jurisprudence’ that Habermas indicts the German courts as incorrectly engaged in.

[109]See Robert Alexy, ‘The Special Case Thesis,’ Ratio Juris 12, no. 4 (1999).

[110]Jürgen Habermas, ‘A Short Reply,’ Ratio Juris 12, no. 4 (1999).

[111]Habermas, Between Facts and Norms, 238.

[112]Gerald Gunther, ‘Judicial Review,’ in Encyclopedia of the American Constitution, ed. Leonard W. Levy, Kenneth L. Karst, and Dennis J. Mahoney (New York: Macmillan, 1986).

[113]‘The essential difference between the United States and the European models is in the way constitutional review is organized. In the United States, constitutional review is exercised by the entire court system; in Europe it is exercised by a unique, specialized court,’ Louis Favoreu, ‘Constitutional Review in Europe,’ in Constitutionalism and Rights: The Influence of the United States Constitution Abroad, ed. Louis Henkin and Albert J. Rosenthal (New York: Columbia University Press, 1990), 40. Conventionally in the literature on comparative constitutionalism, this difference is styled as one between the American and Austrian forms of judicial review; Allan R. Brewer-Carías, Judicial Review in Comparative Law (New York: Cambridge University Press, 1989) labels it as the difference between ‘diffuse’ and ‘concentrated’ systems and extends the analysis to many Latin American nations. Favoreu, in his clear article, also points out that this difference in institutional design often leads to a different form of constitutional justiciability: ‘In Europe, moreover, in general, the constitutionality of a law is examined in the abstract, not, as in the United States, in the context of a specific case; therefore the lawfulness of legislation is considered in general, without taking into account the precise circumstances of any particular case. This is because in Europe constitutional issues are generally raised by a public authority (the government, members of Parliament, courts) and not by individuals,’ 41. Adopting such a form of ‘abstract review,’ it seems to me would also help to disentangle the issues of a law’s constitutionality from the particular substantive outcomes of cases that often become the focus of debates under the United States regime of ‘concrete review.’ Although in practice abstract review is usually undertaken as ‘a priori review’that is, undertaken by a court before a statute’s enforcement—while concrete review usually carried out ‘a posteriori,’ there seems no theoretical inconsistency in the abstract consideration of the constitutionality of a statute after it has been in force for some time. I don’t see that anything aside from contingent problems of institutional design that weighs on the temporal difference—even as such problems may be formidable.

[114]Most United States Supreme Court decisions in fact do not involve the abstract review of statutory enactments and administrative programs for their constitutionality, but rather concern the application of laws and administrative programs to particular fact situations and the internal consistency of federal appellate court decisions.

[115]Vic Peterson has raised this problem in personal correspondence.

[116]Habermas, Between Facts and Norms, 241.

[117]Ibid., 384.

[118]See Sunstein, One Case at a Time: Judicial Minimalism on the Supreme Court.

[119]Habermas, Between Facts and Norms, 280.

[120]Ibid., 395.