Jump to ContentJump to Main Navigation
Governing with JudgesConstitutional Politics in Europe$

Alec Stone Sweet

Print publication date: 2000

Print ISBN-13: 9780198297710

Published to Oxford Scholarship Online: April 2004

DOI: 10.1093/0198297718.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 14 November 2019

The Politics of Judging

The Politics of Judging

(p.127) 5 The Politics of Judging
Governing with Judges

Alec Stone Sweet

Oxford University Press

Abstract and Keywords

The theme of the need to reconceptualize traditional models of legitimation such as separation of powers doctrines is brought into focus. The question of the democratic legitimacy of review is most commonly addressed by grafting Kelsenian constitutional theory onto the classical distinctions between the judicial and the legislative functions. However, this model is found to be theoretically incoherent and empirically inaccurate. An alternative perspective is advanced, arguing that the legitimacy of constitutional review is a product of the participatory nature of constitutional adjudication, driven by strategic interaction that is necessarily pursued through normative argument.

Keywords:   Negative legislator, Kelsenian model, legitimacy, New constitutionalism, separation of powers

Previous chapters have demonstrated the great extent to which constitutional politics have undermined orthodox understandings of how legislative and judicial powers are exercised in Western Europe. Traditional separation of powers notions were employed as benchmarks for evaluating subsequent developments, and certain logics of delegation were used to help explain the move to constitutional review. But I did not explicitly deal with what Cappelletti (1989) has rightly called the ‘mighty problem’ posed by review: the question of its democratic legitimacy. In this chapter, I examine how this question is most commonly addressed in Europe, by grafting Kelsenian constitutional theory on to the classical (but not therefore Kelsenian) distinctions between ‘law’ (the judicial function) and ‘politics’ (the legislative function). I then offer a very different perspective on the question, arguing that the legitimacy of constitutional review is a product of the participatory nature of constitutional adjudication.

Judging, Law‐Making, and Separation of Powers Doctrines

In 1973, the German Federal Constitutional Court ruled on an individual complaint, brought by the newspaper, Die Welt, attacking a decision of the Federal Supreme Court. The Supreme Court, the nation's highest civil law jurisdiction, had upheld an award for damages in favour of Madame Soraya, the wife of the former Shah of Iran. Die Welt had published what turned out to be a completely fabricated interview with Soraya, in which it revealed embarrassing, intimate details of her personal life, leading her to sue. The Supreme Court, proceeding forward from a line of case law dating from the 1950s, found that articles 1 and 2 of the German constitution created a damage remedy, within the framework of the Civil Code, in situations wherein one individual has violated another individual's ‘general right of personality’ (the functional equivalent of ‘privacy’ in US law). The (p.128) Civil Code, however, explicitly prohibited damage awards in such instances, and the Bundestag had twice rejected (1959, 1967) proposals to grant such a remedy. The newspaper argued that judicial authorities had violated constitutional separation of powers doctrines: the courts had not only ignored clear statutory commands, but had amended the Civil Code on their own.

In its decision (GFCC 1973b), the Court ruled that ordinary judges have a duty to ignore relevant statutory provisions that would lead, if applied, to fundamentally unjust results. Instead, when necessary to achieve justice, judges are to reconstruct and even add provisions to the codes, in light of the structure of constitutional rights. The decision not only comforted the position of the Supreme Court. It effectively ‘constitutionalize[d] natural law by requiring the [ordinary] judge to reject fundamentally unjust laws’ (Currie 1993: 212), to reconstruct creatively, outmoded ones, and to fill gaps caused by legislative inaction. When ‘legislation has not kept up with the rapid pace of social development . . . and a changed society's substantive notions of justice’, the Court declared: ‘One cannot blame the judge, if compelled to decide every case submitted to him and convinced that he cannot rely upon [the] future intervention of the legislature, he does not adhere to the literal meaning of the existing written law in a case where adherence would largely sacrifice justice’.

As Quint (1989: 281) has it, the Soraya decision embodies ‘the natural implication of the influence of constitutional rights on private law in German constitutional doctrine’, post‐Lüth (see Chapter 4). But it is also symptomatic of a more general phenomenon: European judges are breaking free of the traditional boundaries that have long circumscribed their work. I have largely focused on the activities of constitutional judges, but consider the following examples drawn from the world of the ordinary courts

  1. In Italy, beginning in the 1970s, ‘progressive judges’ began to mould, by creatively interpreting, the various codes, in order to ‘adapt’ them to constitutional principles and values (documented by Bognetti 1982). Judges, and especially those sitting on lower courts, now typically develop rights provisions, including: equality before the law; rights to work, free expression, and health; due process standards; and the open‐ended norm of ‘personal freedom’. They do so for their own purposes, without first referring questions to the constitutional court. As important, their reading of these rights are now often in advance of the ICC's case law. Italian statutes ‘have undergone . . . a process of internal transformation’ (Bognetti 1982: 440), (p.129) subverting legislative intent, but also the monopoly of the constitutional court on constitutional interpretation.

  2. In Spain, general labour contract law, ostensibly governed by statute, is also being transformed, as judges on the labour courts gradually incorporate constitutional principles, such as freedom of expression and privacy rights, into their understanding of the labour code. As in Italy, ordinary judges have gone beyond, and at times have even contradicted, more restrictive interpretations of rights issued by the Spanish Constitutional Tribunal (del Rey Guanter 1995). Labour judges often prefer to reconstruct statutory provisions in the light of their own versions of constitutional rights, rather than attack statutes before the SCT as unconstitutional. In bypassing the Tribunal, they enhance their own autonomy, and avoid potentially undesirable decisions (which, once pronounced, could be enforced against them by the amparo mechanism).

  3. In France, the Council of State (France's highest administrative jurisdiction), once hostile to the notion that the Constitutional Council's case law could constitute a source of law binding upon it, today actively engages in creative constitutional decision‐making of its own. In a recent case involving the right to property (Errera 1998), an apparent conflict between the constitutional court's jurisprudence and the Council of State's own doctrines provoked the administrative judges to find creative ways of blending their own case law with the Constitutional Council's, thus reconfiguring rules—now part (judge‐made) administrative law, part statute, and part (judge‐made) constitutional law—governing the taking of property by local authorities.

Ordinary judges today regularly use the techniques of constitutional law adjudication to manage the problems that confront them in their work‐place. Judges may even be using such techniques to enhance their own policy‐making authority, vis‐á‐vis legislatures and constitutional courts. The brute empirical reality is that parliaments have lost their implied monopoly on law‐making, constitutional courts today share their authority to interpret the constitution, and ordinary judges—certainly not slaves to the codes1—participate in constitutional politics. Stated differently, it has become increasingly obvious that traditional separation of powers doctrines, however deeply embedded in consciousness we might suppose them to be, are increasingly less relevant to the realities of European governance.

(p.130) The Challenge of the New Constitutionalism

Separation of powers doctrines underpin theories of the state, reflecting, but also conditioning, how political legitimacy is to be understood and assessed. Such doctrines are deeply normative constructs. Read descriptively, as rule systems, they purport to model the relationship between structures and functions in any complex system of governance. Read prescriptively, as sets of idealized expectations how about structure and function ought to be related, they provide standards for evaluating the appropriateness of official conduct.

Traditional, Continental separation of powers doctrines are in deep crisis. Conceived as a set of prescriptions, they appear increasingly obsolete and incoherent. Conceived as descriptions, they obscure more than they clarify what is actually going on in the world. Although other very important factors are at work,2 systems of constitutional justice have operated—inherently, in my view—to undermine the utility and relevance of the classic doctrines. Structural differentiation, but overlap in functions, are the norms. We can easily list the institutional characteristics that distinguish a European legislature from a European court; but today judges legislate, parliaments adjudicate, and the boundaries separating law and politics—the legislative and judicial functions—are little more than academic constructions.

If all this is so, some readers may be asking if and why they should bother with such formalities. This is a serious question, to which I have two quite different responses. First, many social scientists probably do not have any immediate reason to concern themselves with the fact that constitutional politics have undermined separation of powers doctrines, or the question of what, if anything, ought to be done about it. After all, such doctrines, excessively formal simplifications of how political systems are organized, operate at levels of abstraction that are often too high to be of practical use in research. For most purposes, it is probably enough to take seriously the fact that judges (constitutional and ordinary) play an increasingly active, sometimes crucial, role in a wide range of policy domains.

My second response takes the form of an alert. How such controversies are deliberated and settled will likely shape the deep changes in European law and politics already underway. Separation of powers doctrines help those who govern (e.g. government ministers, members of parliament, judges, and administrators) make sense of their own role in the political (p.131) system, and of how they are expected to interact with other actors. They thus constitute not only models of the polity, but models of action. To the extent that these models actually condition how governmental actors take decisions, they deserve our attention. Today, judges and legislators not only participate in constitutional politics, but their work is deeply impacted by the evolution of constitutional rules, and these behaviours are not modelled well by traditional separation of powers doctrines. Consider the following description of the judicial function in Europe:

the focus of all law‐making authority within the state is the sovereign legislature; law is a closed system of logically arranged and internally coherent rules; all legal disputes must be resolved by reference to such rules; courts of law, independent of the legislature, are the proper agencies for interpreting law; courts should interpret law literally and in strict accordance with the legislator's will; their function, therefore, is to administer the law as written. (Kommers 1997: 124; see also Merryman 1985)

We know that constitutional courts often dictate statutory provisions that legislators ratify; that constitutional politics create gaping holes in the legislature's law, once presumed to be closed; and that ordinary judges increasingly behave as if they recognized a higher duty to interpret statute as if it were in conformity with the constitution, thus participating directly in the legislative function. Nevertheless, how the relationship between law and politics is most commonly conceptualized has lagged far behind what is actually happening.

The emergence and consolidation of a new model of the relationship between legislative and judicial power, one that incorporates the normative logic of the new constitutionalism in Europe, would effectively complete the transformation of European governance that I have been describing. Not only would such a model gradually replace the existing one by, among other things, serving to discredit the traditional dogmas as outdated, and those who espouse them as not sufficiently versed in the niceties of constitutional law. A new model would accelerate and reinforce the movement already taking place, operating to legitimize it. At stake, then, is how the nature of European democracy is to be redefined, albeit in the form of a quasi‐official state ideology, given that existing separation of powers doctrines are no longer tenable.

The process of developing this new model has already begun. It is most visible in the work of legal scholars, especially in the activity of constitutional lawyers, but judges are also intimately involved. It bears mention in advance that although aspects of these debates echo those found elsewhere, (p.132) they remain distinctly European. The institutional logic of parliamentary democracy, the historically rooted suspicion of judicial power, and a tradition of highly formalist legal scholarship give them a specific character. Contemporary North American ideas about judicial review and its legitimacy are available, but American and Canadian debates have had no sustained or serious influence on how Continental legal scholars understand the work of their own constitutional courts.3

Three general approaches or orientations to the question of the legitimacy of constitutional review are current. The dominant approach (e.g. Brewer‐Carias 1989; Favoreu 1988, 1990, 1994; Von Brünneck 1990, 1992; García de Enterría 1991) is explicitly Kelsenian, in that its proponents invoke his constitutional theory (see Chapter 2). It entails reconceptualizing traditional separation of powers principles in light of Kelsen's ideas about constitutional legality, taking into account the formal requirements of contemporary systems of constitutional justice. The constitution, which expressly provides for the exercise of review, is the ultimate source of legitimacy. Although tinkering with the classic doctrines has been unavoidable, scholars have also tried to show that they have retained as much of the traditional separation of powers edifice as possible.

The second approach (e.g. Ponthoreau 1991; Troper 1995; Gascón Abellan 1994; Guastini 1991), a distinctly minority one, is commonly labelled ‘realist’. For realists, the constitutional law, and therefore the polity, is constructed through processes of interpretation, which breathes life and agency into the law, expanding its domain and import. Realists argue that the constitutional law is developed within interpretative communities, comprised not only of constitutional judges, but also of legislators, administrators, ordinary judges, scholars, and the interested public. The monopoly of the constitutional judge, presumed by the Kelsenians, is therefore rejected. The perspective is broadly congruent with my own, to the extent that realists see the building of the constitutional law as a participatory process, involving a wide range of actors, public and private.

The third approach is critical, but even less representative of typical academic discourse. The expansion of judicial authority and the attendant ‘juridification’ of political life are viewed as inherently oppressive phenomenon (e.g. Habermas 1986, 1992; Teubner 1987), and stifling of what is needed—a liberating, ‘deliberative’ democracy. In this view, the law—and particularly academic discourse about the law—tends to construct the outside (p.133) environment in a particular way, by favouring the reproduction of the law's own internal logic, insulating the law from the world ‘external’ to it, and by recasting social problems as legal ones (see King 1993; Beck 1994). Of course, the more any project to legitimize constitutional review succeeds, the worse off we are. Although there are connections between this approach and my own,4 I will not deal with it further except insofar as I treat the work of legal scholars, which I do find insular and expansionist, as an important stage in the construction of the law.

We now turn to the Kelsenian orthodoxy. I focus only on those elements that are related to conceptualizing sources of legitimacy; I simplify (or ignore altogether) many otherwise important arguments.

The Kelsenian Model

Kelsen and his latter‐day followers have been concerned, above all else, with the problem of legal validity, in particular, how public acts come to be invested with normativity, that is, with their formal authority as binding law. Put as a question: how do we determine which public acts constitute authoritative rules, binding on their subjects, backed up by the enforcement capacities of the state? Simplifying, Kelsenians argue that any given act is only valid, or normative, if it is enabled by, and does not conflict with, a specific—and formally superior—legal rule. Furthermore, all legal rules, in order to confer validity to lower order rules, must be enforced by some authority (a judge or ‘jurisdiction’). A closed, self‐referential, hierarchically ordered, system of norms is thereby established: every legal rule is lawful by virtue of another, higher order, legal rule. Thus, a ministerial decree or a police action taken in pursuance of a statute must respect the terms of that statute or be invalid, as controlled by judicial authorities; and the statute itself must conform to constitutional dictates or be invalid, as controlled by constitutional judges. The legality of the norm, which reduces ultimately to constitutional legality, and the legitimacy of the legal system, are virtually one and the same thing.

In today's Europe, the common source of all legality, and therefore of political legitimacy, is the constitution—the summit of the hierarchy. Modern European constitutions: (p.134)

  1. (1) establish state institutions and confer on them the power to make and enforce other legal rules;

  2. (2) lay down procedural rules that govern the production of other legal norms, like statutes, and the constraints (like rights) on legislating that must be respected;

  3. (3) fix the rules through which constitutional revision can take place, firmly distinguishing these rules from those governing the production of infraconstitutional law, like statutes;

  4. (4) authorize constitutional courts to defend the normative superiority of constitutional law vis‐à‐vis all other legal rules.

To this point, Kelsen's modelling of the constitution as a system of higher order—but, nevertheless, a positive body of—norms adequately describes how Continental legal systems are, in fact, organized. The constitutional law, as adopted by the sovereign people, comprises both a positive source of law, and the common source of legitimacy for all other legal norms. Further, Kelsenian constitutional theory and principal–agent (P–A) models of delegation are broadly congruent with one another.5 A simple P–A model of constitutional politics links, as in a chain, authoritative acts of delegation from one set of official actors to another (see Chapters 1–3). The sovereign people (first order principals) decide—albeit through their representatives—to be governed by a constitution. The constitution is the normative instrument through which the people have delegated power to governmental bodies, like legislatures. And the statute is the normative instrument through which governments and legislatures (agents of the electorate, but second order principals vis‐à‐vis ordinary judges and administrators) delegate certain specific responsibilities and powers to the courts and the administration. Different sets of principals exercise different types of controls depending on the normative act at hand and the precise terms of the delegation. But the ultimate source of authority, and legitimacy, is the constitution, which is assumed to express the will of the sovereign people.

As a formal construction, the legitimacy of European constitutional review can hardly be questioned, flowing as it does from specific, authoritative acts of delegation from one ultimate source of legitimacy: the sovereign people. However, when we look in detail at how constitutional politics actually operate, other aspects of the model quickly run into difficulty. At its core are two views of the world that are increasingly at odds with reality. First, although the point does not follow from Hans Kelsen's (p.135) own writings,6 Kelsenians today reproduce the separation of law and politics that is basic to the state theory that reigned prior to the advent of the new constitutionalism. The boundaries separating law‐making from judicial functions, although breached by the introduction of constitutional review, are otherwise supposed to have retained their integrity. Second, Kelsenians insist that the erosion of legislative sovereignty has been no more than necessary to guarantee the normative superiority of the constitution through constitutional review.

An attempt to show that the European state, and especially the juridical order, can accommodate constitutional review with minimal subversion of legislative supremacy animates the argument. Kelsenians argue that the constitutional court is the sole judge of the constitutionality of statute, whereas the ordinary courts are judges of how statutes are to be applied (Renoux 1994). Further, although the constitutional court necessarily possesses the power to invalidate unconstitutional legislation, its law‐making function is purely ‘negative’, while the government and parliament act as ‘positive’ legislators (see Chapter 2). Undeniably, the monopoly of the constitutional court on constitutional interpretation of statutes has broken down, and in some places—like Germany—was never established (see Quint 1989). Just as clearly, constitutional courts act as positive legislators: not only do they regularly say no to legislators, they also draft and amend legislation, commanding legislators to behave in certain ways, and not in others.

We know why this modern‐day Kelsenian model, as a description of what is going on, is in trouble. Theoretically, the project is incoherent, for reasons that Kelsen himself would be quick to identify. Kelsen knew that establishing review would fatally subvert the separation of political and judicial functions (see note 6), by politicizing the judge. He warned against codifying rights in the constitution precisely because he knew that codification would politicize constitutional judges even further, making of them not only legislators, but supreme legislators. Although legal scholars today appear to be parroting Kelsen's constitutional theory, they also claim that constitutional courts inherently function to protect constitutional rights, and that this function is basic to the legitimacy of review. Further (see Chapter 4), the orthodox position of modern Kelsenians, and many judges, (p.136) is that rights possess ‘supraconstitutional’ status (their contents can not be altered by constitutional revision), which is akin to a natural law, not a positivist, position. In a sentence, orthodox constitutional scholarship in Europe only selectively embraces elements of Kelsen's constitutional theory, while implicitly rejecting the fundamental positivism of the rest of his thinking. In consequence, Kelsenians are led to defend what cannot be defended, namely, that constitutional judges are only negative legislators.

Empirically, the move to a judicially enforceable constitutional law, replete with human rights (and positive rights) possessed of direct effect, did more than just undermine legislative sovereignty in a formal sense. The new constitutionalism created opportunities for a range of non‐judicial actors to pursue their interests through the constitutional law, and made judges responsible for developing and exercising new powers of decision‐making. The more judges have effectively resolved the constitutional disputes referred to them, the more opportunities for more litigation, and more constitutional rule‐making, expand. Although those who negotiated the constitution (the leaders of political parties) did not intend as much, the construction of a self‐sustaining, expansive constitutional politics—which gradually but inevitably weakens the hold of statute on judging, and gradually but inevitably diffuses the techniques of constitutional adjudication, as a general mode of governance, to judicial and legislative actors—has been the general outcome. In P–A terms, because constitutional and ordinary judges operate in relatively permissive environments, guided by rules that they themselves curate, it is no surprise that their activities routinely escape mechanisms of control available to their principals. The hierarchy of rules and controls assumed by the hybrid‐Kelsenian—and the P–A—model is in fact little more than an assumption.

Although this model remains dominant, and in France reigns as quasi‐official dogma, some legal scholars, and even a few judges, have begun to raise questions. There is growing awareness, for example, that the distinction between the positive and the negative legislator can not be sustained; at the same time, it is constantly asserted that this distinction remains basic to how the legitimacy of constitutional review is normally conceived (e.g., Auguiar de Luque 1987; Díez Moreno 1986; Pizzorusso 1990; Rubio Llorente 1988a, 1991). Thus, Gascón Abellan (1994: 75) complains that ‘the distance that separates the constitutional court [of today] from the Kelsenian court is abysmal’. And Rubio Llorente (1988b: 170–85), a law professor and former vice‐president of the Spanish Constitutional Tribunal, has noted the ‘awkwardness’ and the ‘insufficiency of the theoretical model inherited from Kelsen’, and has called attention to the fact (p.137) that the so‐called ‘negative legislator’ too often ‘assume[s] the role of legislator in toto’.

Why not simply acknowledge that constitutional courts sometimes act as positive legislators, and that ordinary judges sometimes act as constitutional judges? To do so, Kelsenians fear, would effectively destroy classical separation of powers schemes—‘the very foundation of all constitutional architecture’ (Rubio Llorente 1988b: 174–5); judges would then be left exposed to charges of usurpation of the political function. In my own view, the fact that European constitutions explicitly establish their own normative superiority7 provides all of the formal legitimacy needed (although formal legitimacy is only part of the equation). Further, in practice if not always in scholarship, the dictates of the new constitutional law tend to overwhelm rules derived from traditional Continental separation of powers schemes, rather than vice versa.

Four ‘Functions’ Of Constitutional Review

In Europe, doctrinal commentaries on the constitution and on case law comprise the vast bulk of research in public law; surprisingly little is written on constitutional courts and their relationships with other actors, such as parliament and ordinary judges. Nevertheless, there exists an important strain of scholarship, again Kelsenian in inspiration, that seeks to legitimize constitutional review with reference to the (always virtuous) impact of review on parliamentary systems of governance (e.g. Brünneck 1990, 1992; Favoreu 1986,1988, 1990, 1994). It is claimed that constitutional courts perform four basic ‘functions’: (1) they operate as a ‘counterweight’ to majority rule; (2) they ‘pacify’ politics; (3) they legitimize public policy; and (4) they protect human rights.

These functions are essentially regulatory. First, as Favoreu puts it (1994: 560; see also Brünneck 1990: 40) constitutional courts act as ‘either a counterweight’ against a parliamentary majority that is ‘too powerful’ [e.g. in France and Spain], or as a ‘substitute’ legislator where a parliamentary majority ‘does not exist’ (e.g. Italy). Second, constitutional review tends to ‘pacify’ politics: ‘quarrels’ that before would have been fought out in partisan terms, unrelentingly, are ‘appeased’, and settled more reasonably—with reference to the constitutional law (Favoreu 1986: 62; 1988: 19, 39). As Brünneck (1990: 41–2) writes, ‘in society, characterized by extreme politicization and by the total supremacy of majority principles’, constitutional (p.138) review not only settles conflict about the legality of any parliamentary act, more or less permanently, it also facilitates the building of a consensus in favour of other democratic principles, including ‘the safeguarding of rights’ and the ‘principle of legality’. Third, although they deny that constitutional courts ever ‘block’, ‘veto’, ‘censor’, or ‘prevent’ decisions taken by parliament, judges act to ‘guide’, ‘direct’, ‘authenticate’, and ‘correct’ decisions (Favoreu 1988; Brünneck 1990; Vedel 1988), ‘putting reforms on the right normative track . . . the constitutional one’ (Favoreu 1994: 578). Thus, far from obstructing the will of the majority, constitutional judges legitimize it. Last, in the absence of constitutional review, the argument goes, human rights would enjoy no protection.

I have evaluated these formulations elsewhere (Stone 1992a: chs. 4, 6, 7; Stone 1994a), and will not repeat these criticisms here. It is enough to note (as Kelsen himself would have) that constitutional judges can fulfil these functions effectively only to the extent that they in fact behave as (very powerful) ‘positive’ legislators. If, for example, constitutional judges did not annul legislation as unconstitutional and, at the same time, tell (or at least signal to) legislators how they should go about their legislative business, then constitutional review could function neither to ‘correct’ statutes nor to put law‐makers ‘on the right track’. To take another example, the safeguarding of rights may very well provoke, rather than appease, partisan controversies; clearly this occurred in the German abortion cases (Chapter 4), and in the French media pluralism decisions (Chapter 3). Modern‐day Kelsenians seemingly do not notice, let alone explore, these points. Instead, Brünneck (1990: 46–7), after elaborating on these functions, simply reasserts the distinction between the negative and positive legislator, and declares that ‘nearly everyone’ insists that the distinction remains robust. For Favoreu (1994: 578; see also Vedel 1992), even if it is admitted that the constitutional court may at times appear to block the majority's will, or to usurp the legislator's discretionary prerogatives, constitutional rule‐making can always be voided by constitutional revision

What grounds the legitimacy of the constitutional judge is that [an annulment] is to be analysed not as barring but as redirecting the legislative act: further, whatever obstacle has been erected by the constitutional judge is not definitive and can be removed by the adoption of a constitutional revision.8

(p.139) We are thus thrown back into the world of principals, agents, and direct mechanisms of control (see Chapter 3) which, we know, are far less effective than Kelsenians suppose.

Constitutional Politics and Legitimacy

Given that constitutional review subverts the schema that has traditionally served to legitimize judicial power on the Continent, how do we account for the fact that European constitutional courts have succeeded not only in enhancing their capacity to shape legislative and judicial processes, but have at the same time seen their own institutional legitimacy, vis‐à‐vis other governmental institutions,9 increase? I approach this question not as a legal theorist or political philosopher might, but as a social scientist.10 I wish to explain, as simply and clearly as possible, why specific actors behave as they do, and how these behaviours produce specific, relatively predictable, outcomes.

I argue that the political legitimacy of constitutional review has been constructed over time, in interactions among three sets of actors: (1) litigants, those who activate constitutional review processes; (2) judges, those who adjudicate constitutional disputes; and (3) academic lawyers, those who work to build the constitutional law through their scholarly activity. I reject the law/politics distinction on empirical grounds. The decision‐making behaviours that constitutional review engenders are always both ‘judicial’ and ‘political’; and institutions that participate in constitutional review processes necessarily engage in each other's activities. Equally important, as a matter of causality, the politicization of constitutional review processes, by litigants, comprises the essential first step, the trigger mechanism, that enables judicialization; and judicialization processes serve to legitimize constitutional review, by establishing and continuously reinforcing the centrality of constitutional case law within legislative and judicial processes. Underlying this argument, indeed underlying the account of constitutional politics in this book, is a theory of action. This theory of action integrates strategic behaviour and normative reasoning.

In Chapter 1, we examined the relationship between rule structures and self‐interest. I argued that, in any complex set of social behaviours, norms, (p.140) and interests are likely to become intertwined in ways that are often difficult or impossible to unravel. We know that people often obey rules routinely, without giving much thought to them; but we also know that people also weigh the potential consequences of choosing behaviour that does not conform to normative expectations. More complicated yet, sometimes what we observe as the enactment of rules may actually be the surface manifestation of the desire, on the part of individuals, to avoid punishment or to access desired benefits.

In the discussion that follows, I assume that all three sets of actors behave strategically all of the time: they seek to maximize their well‐being given the constraints they face. In constitutional politics, however, actors can pursue their interests only through normative argument, and effective normative arguments can only be fashioned by reasoning through rule structures. In such a social system, interests are constantly reconstituted as legal discourse, a fact which has heavy consequences for how the system develops. At any given point in time, perfect information about the precise nature of the most important set of constraints, the constitutional law, rarely if ever exists; rather constitutional adjudication and its associated practices lead to the clarification of these constraints. As we will see, each of the three sets of actors use normative discourse for their own purposes; but each does so with different priorities in mind, and seeks different kinds of outcomes.


Individuals activate constitutional review processes—they politicize constitutional justice—in order to alter policy outcomes in their favour. They know that the constitutional court can overrule parliaments and courts, and that the court can also control legislative and judicial behaviour in other ways. Individuals litigate if the potential benefits of obtaining an annulment or revision of a law (or public policy, or private practice) outweigh the costs of the legal challenge. Litigants will make the best possible legal arguments on their own behalf, but such arguments will express, rather than disguise, motivations. Constitutional challenges constitute an instrumental, relatively porous form of normative discourse, one provoked by, and open to, social facts (e.g. the details of a particular dispute, the relative power of disputants, the distribution of resources in society, and so on). It is through litigation that the constitutional law secures and maintains its political relevance and vitality.

Constitutional litigation helps to legitimize constitutional review in at least two ways. First, in activating constitutional review litigants delegate, (p.141) to constitutional judges, policy issues that could have been dealt with in other, non‐constitutional, forums. Acts of delegation are (at least implicit) acts of legitimation. Consider the politics of abstract review. Governing majorities often complain, after an annulment, that constitutional judges have thwarted democratic principles and subverted legislative sovereignty. Yet, when these same parties are in opposition they do not hesitate to make use of abstract review, and they characterize their victories as strengthening democratic principles embodied in the constitution. Parliamentary oppositions could choose to defer to the wishes of the majority, after losing parliamentary votes, as the price to be paid for losing elections; instead, seeing the opportunity to alter the majority's legislative programme, they initiate abstract review.11 In this way, legislators—across the political spectrum—have participated, willingly and continuously, in processes that reduce the scope of their own authority and discretion, and to enhance that of the constitutional court. More generally, we find more, not less, constitutional litigation over time, across Europe, for the simple reason that individuals find it useful to pursue their interests through constitutional review. That legislative and judicial processes are increasingly conducted in the shadow of the constitutional court is important empirical evidence of the enhanced social legitimacy of review itself.

Second, in activating constitutional review, litigants help to build the constitutional law. Litigation not only transfers questions of social, economic, and political import to judges; it provokes constitutional rule‐making.


In resolving disputes, constitutional judges seek both to preserve the normative superiority of the constitutional law and to ensure that the constitution becomes, or continues to be, the essential reference point for the settlement of like cases that may arise in the future. Judges know that their activities will be evaluated by two groups: (1) present and future litigants, and (2) doctrinal authorities. The judges are cross‐pressured. Litigants hope to win the case and to see their own political values enshrined as authoritative constitutional law. Failing that, they hope not to lose too much, or see values that they oppose enshrined as case law. Legal scholars expect constitutional courts to exercise their rule‐making powers in ways that enhance the overall coherence of the constitutional architecture, (p.142) which they have helped to build. At a minimum, this means producing decisions that can be read as being consistent with antecedent case law and established doctrinal understandings.

To succeed at their task, the judges deploy several techniques. First, they avoid declaring either side a clear loser, preferring, wherever possible, to issue partial victories, splitting the difference between the disputants. Of course, for those who challenge a statute's constitutionality, even a partial victory equates as a loss for those who supported the law (since, in the absence of constitutional challenge, the law could not be challenged). None the less, in abstract review processes, for example, it is striking how few total annulments have been issued by constitutional courts. Instead, the judges remove those provisions considered to be contaminated by unconstitutionality, allowing the rest of the law to be applied; or they tell the legislature how it must correct the law if it wishes to pursue a given legislative reform. This point is seemingly never noticed, and certainly not commented upon publicly, probably because it would focus attention on non‐normative, extraconstitutional elements of constitutional adjudication.

Other, explicitly legalistic, aspects of constitutional decision‐making help judges counter suspicion that constitutional review produces clear winners and losers, over time. In developing balancing and proportionality tests (Chapter 4), for example, judges exploit the social logic of long‐term reciprocity among potential litigants (society), and construct the law as a flexible instrument of dispute resolution. These tests make it clear: that each litigant's legal interest is a legitimate one; that the court none the less must take a decision, by weighing each side's interest against the other; and that future cases pitting the same two legal interests against one another may well be decided differently, depending on the facts. Balancing and proportionality tests hold sway precisely where: (1) the law is most indeterminate; and (2) most in danger of being constructed in a partisan way.12 Balancing tests constitute normative tools for managing this indeterminacy over time; they also enable the judges to maintain their own room to manoeuvre in cases likely to come before them in the future.

Second, constitutional judges labour to portray their decision‐making process as inherently ‘judicial’, and therefore meaningfully distinct from ‘political’ (i.e. legislative) processes. They have borrowed procedures from ordinary courts for example, in order to make more ‘concrete’ what would otherwise be essentially abstract processes of decision‐making. Thus, (p.143) although no written rules require such practices, the French Constitutional Council solicits from the government and parliamentary majority briefs that respond, point‐by‐point, to legal arguments contained in referrals from the parliamentary opposition (see Stone 1992a: ch. 5). The other courts, too, have actively sought to enhance the adversarial quality of constitutional review procedures, to increase the perception of fairness, to enhance the ‘judicial’ qualities of the proceedings, but also to encourage those actors being controlled to engage in the kind of rule‐governed reasoning that the court does.

In any event, the judges are cognizant of what litigants wants, and what doctrinal authorities expect. The administrative staffs of the courts under consideration compile dossiers that contain, in addition to litigants' briefs, discussions of the pertinent case law and relevant scholarly commentaries. In abstract review cases, all of the parliamentary debates, committee reports, and other documents that issue from the legislative process are included. In concrete review cases, the court receives the essential documents related to the litigation that generated the referral.13

Although the socio‐political stakes of constitutional litigation are always apparent, the judges will nevertheless push private interests and social facts further into the background, repackaging them as inherently normative arguments. Thus, the French government had a long list of reasons why every important French bank needed to be nationalized in 1981–2, and these reasons were iterated and reiterated for months leading up to the constitutional adjudication stage of the legislative process. The French Council recognized these arguments, which had nothing to do with the constitution, only in so far as they could be said to constitute a ‘public interest’, which the constitution had authorized the French government and parliament to pursue. But because the 1789 declaration of the rights of man contains a provision guaranteeing citizens the right to property and fair compensation in the event of expropriation, the Council proceeded to what were, in effect, balancing and proportionality tests, and then found the Socialist's bill lacking (Council 1982a). My point is not only that constitutional judges are fully aware of the partisan, or instrumental, nature of the litigation that comes before them, but that they also react by requiring such interests to be expressed in the rule‐governed discourse that constitutional decision‐making generates.

Finally, and most obviously, constitutional courts portray their decision‐making as if it were a pure exercise in logic. The logic is ultimately (p.144) expressed as a syllogism: the constitution, as higher law, requires x; the act which has been attacked as unconstitutional falls either within or outside category x; to the extent that said act falls outside of category x, it is void as unconstitutional. Of course, in order to arrive at this point, judges have reasoned through the case and through the pertinent rule structures. Often this reasoning is complex, sometimes tortuous. Nevertheless, in that final, crucial instance in which the law is determined, the decision perfects its own internal normativity. And it is this normativity that gets passed on to doctrinal authorities and future litigants, as part of the law.

The Decision

Constitutional courts are, by law, required to justify their decisions in written form. Yet how they do so has more to do with their own quest for legitimacy, given the environment in which they find themselves. Judges seek to create the conditions under which constitutional review will not only survive, but thrive; that is, they seek to reconstruct their environment in ways that are conducive to the continuing vitality of constitutional review. To thrive, constitutional review must draw other actors into the rule‐governed discourse of adjudication; and the court must also convince its audiences that how it exercises its powers is meaningfully constrained by rules internal to normative reasoning (and therefore are not arbitrary powers). Michel Troper (1995: 294–5), a French legal theorist, has put these points together elegantly

The internal constraints on the constitutional judge's power are essentially those linked . . . to the need to justify the judge's decisions. This need does not relate exclusively, or even principally, to the requirement to provide justification . . . The judge in fact intends to give his decisions the [highest] level of effectiveness possible. In other words, he wants to hand down a decision that will produce effects, not just in concreto, thanks to the annulment of the law, but also in abstracto. . . . The norm which will have thus been established can then be imposed on other constitutional authorities, such as parliament, because it is obvious that any future law which contradicts this decision will in turned be deemed constitutional, as well as other courts, because they likewise could not, without difficulty, avoid basing their own decisions on this same norm, except by establishing the objective existence of another more general principle which would justify the exception. The justification is thus an attempt to enclose the other bodies within a logical conclusion from which they cannot escape. However, this logical conclusion also constrains the constitutional judge himself, because once he has confirmed the objective existence of a norm, he cannot therefore ignore it.

(p.145) What is peculiar to the political power exercised by constitutional judges is that, in deciding, they not only constrain the activities of governments, parliaments, administrators, ordinary judges, and the citizenry. They constrain themselves.

Opinion Styles and Dissents

Two models of opinion‐writing styles coexist in Europe. The first, represented by France and Italy, is the more traditional. The French and Italian constitutional courts follow conventions established by the high administrative and civil courts. Decisions are relative short and declaratory of the law; they invoke the precedential authority of prior case law through the use of linguistic formulas that are pointedly repeated. The second model, developed first in Germany but quickly adopted in Spain, more resembles American practice. Constitutional decisions are longer, more wide‐ranging, even literary. Each important point of law raised by each litigant may be argued through to its conclusion, in the light of existing case law and alternative (but ultimately rejected) lines of argument. The German and Spanish courts commonly cite the work of legal scholars and even other courts, like the US Supreme Court. Although a decision written in the style given by the first model could never be confused for one written in the style of the second, French and Italian constitutional rulings have, over time, become much longer, more openly argumentative, and less terse and syllogistic. I interpret this change as a predictable response to the increased politicization of constitutional justice. As constitutional judges know, the politicization of their offices by litigants can only be effectively countered with more and better normative arguments.

In Germany and Spain, votes are published and dissenting opinions allowed; in France and Italy, dissents are prohibited. Those who favour the practice argue that dissents enhance the court's legitimacy by showing ‘that the arguments of the losing side were taken seriously by the court’ (Landfried ed. 1989: 231). Opponents invoke the legitimizing power of public unanimity. A small handful of studies on voting patterns in the German and Spanish courts exists, which show that groups of judges do tend to vote together, and that judges appointed by the same parties tend to belong to the same groups (e.g. del Castillo Vera 1987). These tendencies, which are quite weak, are often overwhelmed by disagreements about the law and constitutional doctrine (Lamprecht 1992).

In Italy, but not France, there exists a vigorous debate over whether to allow dissenting opinions (e.g. Panza and Reposo 1981; Roselli 1994). In (p.146) any case, the style of opinion‐writing in Germany and Spain more easily accommodates dissents. If France or Italy did move to permit the publication of minority opinions, it is likely that a more literary, discursive model of opinion‐writing (such as that found in the US, Germany, and Spain) would gradually emerge.

The Precedential Authority of Case Law

It is commonplace in comparative law to note, first, the formal absence of the doctrine of stare decisis in Continental legal systems and then to assert that this absence constitutes a crucial difference between legal systems. However, the best systematic research has shown, in Summers and Taruffo's summation (1991: 487), ‘that there are no great differences in the use [of precedent] between the so‐called common law and civil law systems’. Courts, and especially higher courts, constantly invoke the authority of prior decisions. In Spain, scholars now openly debate whether or not the principle of stare decisis has in fact already emerged (Díez Picazo 1994b). In Germany, lawyers can be fined for failing to invoke relevant precedents in their pleadings. The notion that judicial decisions could not create binding precedent inhered in traditional separation of powers doctrines, since to admit otherwise would have required recognizing that judges participated in the legislative function. Today, with legislative sovereignty in full retreat, judges more openly exploit the legitimizing resources that asserting precedent provides. In contemporary Europe, virtually all of the constitutional law that matters is case law.

Decisions rendered by constitutional courts are recognized as formal sources of binding law in all four countries. In France, Germany, and Spain, how courts arrive at their decisions—the judges' reasoning—is also binding; in Italy, the doctrine of the ‘living law’ gives the Supreme Court (Cassazione) some latitude to interpret constitutional case law in light of its own, judicially constructed, doctrine and cannons of statutory interpretation (Chapter 4). Finally, in Germany and Spain, failure on the part of any public official to abide by the terms of constitutional case law constitutes grounds for individual complaints. In France, some judicial authorities continue to resist the authority of the Council's reasoning (Chapter 4).

Doctrinal Activity

Legal scholars work to construct the law as a coherent body of rules, by elucidating the content and meaning of specific norms, and by defining the (p.147) relationship of any given norm to the greater normative system. Constitutional case law grounds the scholar's work, conditioning while not entirely determining it. One function performed by the legal scholar is to explain the constitutional court's jurisprudence, and to integrate the court's rule‐making into the law; still the constitutional law, as curated by the legal scholar, remains analytically distinct from case law. As a form of normative discourse, doctrinal activity is—relative to litigating—maximally opaque: it self‐consciously ignores the world external to the law itself. Put differently, doctrinal activity reconstructs the law as a radically autonomous discursive structure, cut off from the greater socio‐political environment to which the law, and the judges of the law, would otherwise belong.

In Europe, the social power of public law scholars has depended critically on their capacity to insulate the law from the social world, and especially from ‘politics’: the world of political parties, ideologies, interests, and ‘non‐legal’ values.14 This way of doing things—the maintenance of the law/politics distinction as an article of disciplinary faith—has reproduced itself over many generations. That Continental legal scholarship is highly formalist, relatively immune to critical perspectives on the law, largely disinterested in questions of legal interpretation, but none the less committed to enhancing the prestige and legitimacy of doctrinal and judicial power are tendencies that have been widely commented upon (e.g. Merryman 1985; Stone 1992a: chs. 1, 4). For our purposes, what is important to emphasize is that legal scholars, in pursuing their own corporate interests, operate to legitimize the court and its case law.

They do so in several ways. Constitutional scholarship, first, refrains from being too critical of any decision. Confronted with an aspect of a decision that appears inconsistent with prior case law, or with established understandings of the constitution, for example, doctrinal authorities will typically downplay the ‘mistake’, narrowing its relevance to the specific case at hand, and reasserting the full scope of the prior version of the law in all other relevant cases. Scholars have thus invited the constitutional court to correct itself in the future, and shown the court how to do so without having to admit that an ‘error’ had ever been made.

Second, scholars extract from the case law those purely normative elements that can be incorporated into the rule system they are building, all but ignoring other elements. They do so almost instinctively, so normal has the reflex become. Thus, most standard texts on constitutional law in (p.148) Germany, Italy, and Spain, make little or no mention of who litigates and why, what kinds of legal arguments were made and rejected, or even how the constitutional court reasoned through rules. (In fact, most students of constitutional law in German, Italian, or Spanish universities do not read any case law for their courses, but rather a treatise—a ‘synthesis’ of the law written by a constitutional scholar.) In France, scholars have even produced a Code Constitutionnel (Renoux and de Villiers 1994), that combines exegesis of the constitution, provision by provision, with discussion of how relevant decisions have clarified the meaning of the constitutional text. In constructing a ‘pure’ system of constitutional law, scholars enhance the court's authority, to the extent that constitutional rule‐making is portrayed as the by‐product of purely normative reasoning.

Bernard Schlink, one of a very small number of constitutionalists who have written reflectively on their discipline, is quite critical of this result, which he characterizes as a kind of ‘constitutional court positivism’, wherein constitutional scholars seek to ‘canonize the constitutional court's case law’, by ‘harmoniz[ing] these decisions into a coherent doctrinal corpus’ (1993: 730–1). If the legal scholar appears to treat the constitutional court as if it were ‘the mouth’ of the constitution, the scholar appears to be ‘the mouth’ of the constitutional court. What is clear is that the relationship between constitutional adjudication and doctrinal activity is pervasively symbiotic. Scholars need an authoritative, ‘judicial’ interpreter of the constitutional law, to structure, but also to give salience and urgency, to their own activities; and constitutional courts rely heavily on legal scholars to disseminate and explain their decisions to politicians, judges, the interested public, and, often enough, even to the constitutional judges themselves.15

Although my argument is that the scholarly impulse towards systematizing the constitutional law helps to legitimize that law, I do not mean to imply that scholars are blind to the politics of constitutional review. On the contrary, legal scholars actively participate in these politics other than through their doctrinal production. For obvious reasons, the more authoritative a legal scholar is within the academy, the more likely that scholar will be solicited by potential litigants, such as governments and political parties, for advice and for drafting referrals. Further, in all four of the countries under consideration here, eminent public law scholars have not only been named to, but have dominated, constitutional courts. I argue, instead, that the drive towards systematizing the law constitutes a deeply ingrained (p.149) response to the fact that constitutional law only develops in reaction to politicization. The constitutional law is, inherently and by definition, political law, but legal scholars rightly insist on that law's normative qualities. In doing so, they help to constitute and perpetuate their own authority.

Summary: Constructing the Constitutional Law

One important measure of the social legitimacy of constitutional review is the extent to which review has provoked normative discourse. In Europe, constitutional courts have drawn an ever‐widening range of actors, public and private, into participating in, and perpetuating, that discourse. Although I have argued that each set of actors participates in constitutional politics with different purposes in mind, the core activities of each tends to push for more, not less, constitutional review, and for more, not less, rule‐governed discourse. The result is that constitutional review process function as permanently constituted forums for the construction of the constitutional law.

Figure 5.1 summarizes and depicts the argument. This process of constructing the law, represented as the line moving from left to right, involves three sets of actors: litigants, judges, and legal scholars. As we move from left to right, the nature of the normative discourse changes. Litigation activity, the far left‐hand pole, requires that self‐interests—private or partisan—be expressed as legal interests; the discourse is overtly instrumental. Doctrinal activity, the far right‐hand pole, produces a relatively ‘pure’

                   The Politics of Judging

Fig. 5.1. Constructing the constitutional law. Note. The figure depicts three stages of the process through which the constitutional law is constructed. The line, a continuum, is constituted by the relative presence of policy considerations within three kinds of normative discourses. At point A, litigants express (in the form of normative arguments) specific, identifiable socio‐political interests, which they hope will be furthered by subsequent constitutional decision‐making. At point B, judges co‐ordinate socio‐political interests and normative structures, by taking policy decisions, but also engaging in constitutional rule‐making. At point C, legal scholars represent the law as an autonomous, apolitical, rule system, eliminating as far as possible the policy‐relevance of constitutional innovation.

(p.150) normative discourse, as divorced as possible from socio‐political interests. Judges co‐ordinate abstract rule structures and concrete disputes and, in doing so, build the constitutional law. Litigants keep the law open, and legal scholars work to close it. The court is advantaged by both activities. To use an analogy, litigation pours in, turning the mill; constitutional judges operate the mill, separating the wheat from the chaff; and legal scholars produce neat loaves that are easily stacked.

Last, it bears emphasis that I consider each of these stages to be equally, and profoundly, political processes. Governance is, in my view, how rules are adapted to the experiences and exigencies of those who live under them. If political scientists have reason to care about how rules are produced and stabilized within social systems like political systems, than they have reason to take seriously the sources and consequences of normative deliberations.


Chapters 2–5 show that legislating, litigating, and constitutional adjudication are—increasingly—mutually constitutive processes. Under certain conditions, each establishes the context for the others. The empirical findings lead us to deny the utility of traditional separation of powers schemes. Such schemes do more to obscure than to illuminate how laws are made, how judges take decisions, and how the constitution evolves.

Separation of powers ideologies, of course, are less suited to the accurate description of how the world of government actually works than they are to putting in order that (potentially chaotic) world. Moreover, they ground arguments designed to secure the legitimacy of public authority, including the judicial. Nevertheless, I have argued in this chapter that traditional, Continental separation of powers notions provide an inherently weak basis for discussions about the political legitimacy of constitutional review. I conclude by sketching three different ways to conceptualize the legitimacy of the constitutional judge, each of which admits (at least partly) the policy‐making consequences of constitutional review.

The first is drawn from the logic of delegation and hierarchy of laws notions. Kelsen argued that the exercise of constitutional review could effectively operate not to obstruct but to legitimize the work of the legislature, but only if constitutional judges were not given jurisdiction over rights provisions. There existed no logical means, he demonstrated, for judges to defend rights without at the same time supplanting the legislator. (p.151) Kelsen's prognosis was correct: if Europeans wish to have judicially enforceable rights then they must accept that constitutional judges will fully participate in the legislative function. It follows that the legitimacy of constitutional rule‐making is ultimately tied to the legitimacy of rights provisions. The calculus: is the polity better off without constitutional rights?; and should legislators alone decide how constitutional rights are to be enjoyed and protected in law? The answer to both questions, in most of Europe today, is a clear and resolute No.16

Second, if we accept that constitutional judges behave as adjunct legislators, that this is a core component of their job description, provided for by the constitution, then legitimacy issues are recast in important ways. Most debates about the legitimacy of constitutional review are debates about whether or not constitutional courts behave as ‘judicial bodies’ are expected to behave, that is, as adjudicators (applying pre‐existing law to resolve disputes) not law‐makers; and these expectations are derived from outmoded separation of powers schemes. But constitutional courts were neither meant, nor originally intended, to be ‘judicial bodies’ traditionally conceived (which is not to say that the ordinary courts make less law than do constitutional courts).17 The founders recognized the mixed politico‐legal nature of these new jurisdictions, just as they recognized the mixed nature of constitutional law. Constitutional courts were instead expected to participate in the legislative function. We get closer to reality if we go beyond the question of whether constitutional judges legislate when they protect rights, and ask instead: do constitutional judges, in fact, protect rights better than governments and parliaments do, or would do in the absence of constitutional review? To the extent that we can answer this question in the affirmative, the legitimacy of constitutional review is that much more secure.

This book, and the latter part of this chapter suggests a third argument, one that will be revisited in the concluding chapter. The legitimacy of constitutional review, like the development of the constitution, is constructed, (p.152) over time, by practice, by constitutional politics. The development of the constitution, as organized by constitutional rule‐making, tends to draw other actors into specific kinds of legal discourses, thus reinforcing the centrality—and legitimacy—of constitutional justice within the polity.


(1) I do not mean to imply that parliaments really did exercise continuous control (parliaments did not), or that ordinary judges did not make the codes their own (judges did). But, today, judges do so overtly.

(2) The growth in the state, social welfare, the complexity of government given modern communications, etc.

(3) The extent of transatlantic cross‐fertilization is insignificant. American constitutional law scholars are especially insular in their approach to judicial review, know little or nothing about constitutional review elsewhere, and therefore rarely address European concerns.

(4) I have tried to show that constitutional review has generated an expansive (rather than simply narrow) and relatively participatory (rather than exclusively elite‐dominated) deliberative mode of governance, a mode of governance that would not have emerged in the absence of constitutional review.

(5) This connection is never made.

(6) For Kelsen, law‐making is political in that law constitutes the formalization of politics, or the will of those authorized to make the law. Constitutional courts, because they have been delegated the authority to interpret and therefore to make the constitutional law, are recognized by Kelsen to perform a political function. I thank Michel Troper for reminding me of this point. For further discussion, see Ch. 2.

(7) That is, by (1) distinguishing between legislative procedures and procedures for revising the constitution, and (2) conferring review powers on a constitutional court.

(8) The argument recalls one made by Kelsen (1942: 187): the ultimate source of unconstitutionality is that the majority decided to adopt the desired reform as a statute, rather than as a constitutional revision.

(9) Opinion polls regularly show higher public support from citizens than do other governmental institutions.

(10) I do not deny the contribution legal theory or political philosophy can make, and indeed make use of it.

(11) There are no systematic differences, across countries or among political parties within countries, in how or why political elites make use of abstract review.

(12) For example, when two rights claim conflict, or when a right conflicts with a constitutionally derived public interest being pursued by government

(13) For reports on the internal procedures, staffing, budgets, and so on, of European constitutional courts, see Table Ronde (1990, 1994).

(14) In the public law, relative to the private law, this task has proved to be especially difficult (see Stone 1992a: chs. 2, 4).

(15) Judges rely on and use doctrinal materials; they also consult, and have even been known to discuss important decisions with doctrinal authorities.

(16) Recall that everywhere but France, the constitution explicitly delegates the responsibility and the powers to constitutional judges to protect rights.

(17) According to Golay (1958: 183; see also Kommers 1994: 12), the founders of the German Basic Law ‘recognized . . . that in many issues on which it is called to adjudicate, [the GFCC] acts as a kind of third legislative chamber, whose decisions have the same effects as legislation . . . ’. In France, neither legal scholars nor politicians understood the Constitutional Council to be a ‘judicial’ body until the late 1970s (Stone 1992a: ch. 4). And in Italy, opponents of constitutional review regularly characterized the proposed ICC as a ‘superparliament’ or ‘third parliamentary chamber’ (Pizzorusso, Vigoriti, and Certoma 1984: 334).