In current legal language, the terms ‘judicial activism’ and ‘judicial restraint’ designate opposite approaches taken by judges to the text they are expected to interpret whenever the meaning of the words of which it is composed, or the intent of its authors, is not deemed sufficient for resolving the case. The more a judge feels himself free, in such circumstances, to give the text further meanings, the more he is considered ‘activist’. Conversely, the more a judge prevents himself from giving the text those meanings, the more he is deemed to be following a ‘restraint-based’ approach.
While focusing on the meaning of the text, these definitions connect the terms ‘activism’ and ‘restraint’ strictly to the task of interpretation. Larger definitions associate such terms with further activities of judges. Whether judges should strictly apply the rules of standing, whether judges should not consider a case until the applicant has exhausted other remedies, and whether judges should avoid deciding ‘political questions’ are among the questions that sometimes are deemed necessary for distinguishing judicial restraint from judicial activism. *1 These definitions, although no less correct than that focused on interpretation of the text as such, are not appropriate for application in a straightforward comparative account of the experiences of constitutional justice, requiring enquiry into judicial activities that diverge greatly in individual legal orders. By contrast, as will be further demonstrated, interpretation of the text not only corresponds to the most important criterion for designating a judge’s attitude as activist or not but is also particularly helpful in such a comparative account.
It has been noticed that “Individual words acquire real meaning only when they are viewed and interpreted within context. Myriad factors may combine to constitute that context: the other words within the sentence; the other sentences within the paragraph; the purpose of the text as a whole; the identity of the author and the expectations which we have of him; the identity of the reader; the social, cultural or political perspective from which he approaches the text, and so on. Thus it is naive to suppose that any text may have a fixed and settled meaning. Any given meaning which is ascribed to a text is, at least in large measure, a product of the external factors which influence its interpretation; the inherent meaning of the words which combine to form the text merely demarcate the parameters within which a range of specific meanings can be ascribed to that text.” *2
This argument becomes crucial with respect to constitutional interpretation. The fact that constitutional rights provisions tend to be comparatively indeterminate, including general invocations of liberty, equality, due process, freedom of speech, and the like, leaves them more open to judicial interpretation than most statutes, administrative regulations, or ordinances. Moreover, since constitutional provisions generally occupy the highest position in the hierarchy of norms within a domestic legal system, decisions of courts in the position of final arbiter of constitutional claims can be overruled only by a constitutional amendment or by their own subsequent decision. Finally, constitutional rights claims often raise issues that are highly controversial politically. *3 These features appear particularly clear in the case of the Constitution of Estonia, whose § 152 second paragraph states that “The Supreme Court shall declare invalid any law or other legislation that is in conflict with the provisions and spirit of the Constitution.”. While specifying that laws might infringe the Constitution whenever conflicting with its ‘spirit’ not less than with its ‘provisions’, the Estonian Constitution presupposes the literal rule’s insufficiency for a correct approach to constitutional interpretation. The ‘spirit’ of the Constitution is, in fact, unlikely to be encapsulated in single words, and even in the whole text of the Constitution. It can, rather, be apprehended through adaptation of the text to the diverse circumstances imposed with the passage of time. Rather than requiring a predetermined meaning, the ‘spirit’ of the Constitution admits shifts of meaning. This is precisely the kind of challenge that constitutional interpretation is expected to meet. It is also a challenge that contemporary constitutional texts are suited for, due to their relatively indeterminate language. It is that language which gives a constitution the capacity to survive those changes that may bring about reform of the ordinary legislation.
On the other hand, constitutional rights claims raise politically controversial issues to the extent that constitutions mirror pluralistic societies and at the same time posit the premises for their own free development. As Michelman has put it, “The legal form of plurality is indeterminacy — the susceptibility of the received body of normative material to a plurality of interpretative distillations, pointing toward differing resolutions of pending cases and, through them, toward differing normative futures.” *4
The fact that the literal rule and recourse to the intent of the Framers are frequently insufficient in guiding constitutional interpretation does not mean that courts may set aside those criteria whenever they wish. On the contrary, courts rely on other criteria only after having demonstrated that the language plainly emerging from the text or from the intentions of its authors is insufficient for resolving the case. This is not merely a recommendation. It also depicts a current judicial practice. Although ‘activism’ is sometimes seen as failure to apply a rule at hand in accordance with its meaning, or applying a rule that has no warrant in the existing legal materials *5 , it has been convincingly replied that “understood in these terms, an account of ‘activism’ is unlikely to be of much assistance. Few judges will knowingly fail to apply a rule in accordance with its meaning, or rely on a rule which has no legal warrant as they see it”. *6
These features appear sufficiently consolidated both in the American and in the European system of constitutional justice. If this is so, contrasting judges who apply their own moral values with judges following the plain meaning of the words in the law, as many commentators do, appears to be a ‘false dichotomy’. *7 The activism/restraint dichotomy presupposes instead that the language that judges, and constitutional courts in particular, have to contend with is often indeterminate. And the dichotomy exists in the attitude toward that language. The activist approach tends more easily than the restraint-based approach to rely on criteria, first and foremost the teleological, that are not directly grounded in the text. The above-mentioned dichotomy is therefore a matter of degree, being apprehended in quantitative rather than in qualitative terms.
Once definition is provided in such terms, it remains to be seen why judges should adopt an activist or instead a restraint-based approach.
According to Posner, three approaches may lie behind doctrines of restraint: deference, reticence, and prudence. The deferential approach consists in avoiding contrasts with the decisions of other branches of government, the reticent approach is founded on the assumption that judges should not be making policy decisions, and the prudential approach is suggested on the grounds that judges should avoid making decisions that may well impair their capacity to make other decisions. *8
The first two approaches appear directly related to the issue of the legitimacy of judicial decisions in a democratic system. The third one as well is related to that issue, albeit only indirectly, prudence being suggested in order to avoid decisions that would incur political reprisals interfering with the judiciary’s ability to make other decisions. *9 The approaches suggested by Posner for justifying restraint appear therefore as diverse expressions of the legitimacy issue.
In the American literature, the most important accounting of that issue is in the work of Alexander Bickel. “The root difficulty”, wrote Bickel, “is that judicial review is a counter-majoritarian force in our system”, since “when the Supreme Court declares unconstitutional a legislative act or the action of an elected executive, it thwarts the will of representatives of the actual people of the here and now; it exercises control, not in behalf of the prevailing majority, but against it”. *10 At the same time, however, Bickel was convinced that the court’s task consisted in rendering principled decisions. Judicial review, he stressed, “brings principle to bear on the operations of government. By ‘principle’ is meant general propositions […] organizing ideas of universal validity in the given universe of a culture and a place, ideas that are often grounded in ethical and moral presuppositions. Principle, ethics, morality — these are evocative, not definitional terms; they are attempts to locate meaning, not to enclose it.” *11
Bickel was also aware that “the Supreme Court touches and should touch many aspects of American public life”, but he was also convinced that “it would be intolerable for the Court finally to govern all that it touches, for that would turn us into a Platonic kingdom contrary to the morality of self-government”. *12 His solution to the ‘counter-majoritarian difficulty’ didn’t consist, therefore, in recommending to the court an exclusive reliance on the text, or on the intent of the Framers, since this would not correspond with the task of issuing principled decisions that he found typical of judicial review. He instead invited the court to exert, and further enhance, ‘passive virtues’, which he described as refraining from deciding cases, through a number of well-known jurisdictional techniques and like devices, whenever issues of principle are not at stake. This suggestion corresponds to the conviction that, while legislation is both ‘empirical’ and ‘evanescent’, “Principle is intended to endure, and its formulation casts large shadows into the future”. *13 Bickel here joined Marshall in considering the Constitution to be, as the latter put it in McCulloch v.
Bickel’s reconstruction of the counter-majoritarian difficulty appears almost unique in the American literature in that it represents the accumulation of a profound understanding of the specific features of constitutional interpretation, as demonstrated by his defence of the Supreme Court’s choices in the School Segregation Cases *14 , with a clear perception both of the substantive power already acquired by the Supreme Court vis-à-vis democratically elected institutions and of the dangers of the ‘Platonic kingdom’ that an unfettered constitutional jurisprudence might create.
In the decades that were to follow, the American debate has lost this contextual attention, being polarised on account of the dichotomy between partisans of the originalist approach *15 , whose fear of judicial activism leads them to forget the features specific to constitutional interpretation, and defenders of judicial activism — particularly in the Warren Court’s version — whose view is that law is an interpretative enterprise guided by a vision of the integrity of the political society to which the law belongs *16 , thus denying the very premise of the counter-majoritarian difficulty. Nor has the Supreme Court followed Bickel’s suggestion of relying on the ‘passive virtues’ to cope with that difficulty. *17
Notwithstanding its relative lack of impact on the subsequent American experience, Bickel’s reconstruction remains a useful basis for examination of the issue of the legitimacy of constitutional courts in a democratic system, which lies at the core of the activism/restraint dichotomy. Bickel was careful in giving balanced attention to the two factors that render constitutional review of legislation a delicate task — namely, the fact that the Constitution applies morally controversial concepts in many instances and the fact that the legislative text under review derives a special dignity from its source — a popularly elected parliament. *18
Posed in these terms, the legitimacy issue affects the European not less than the American model of constitutional justice. As is well known, the former is distinguished from the latter in that European constitutional courts are uniquely empowered to set aside legislation that runs counter to the relevant national constitution, while all American courts have the authority to adjudicate on constitutional issues in the course of deciding legal cases and controversies. The choice for courts specialising in constitutional issues was a result, in Europe , of both cultural and institutional elements. The high value given to the principle of legal certainty in countries adhering to the civil law tradition was likely to be ensured only by a special court in charge of constitutional review of legislation. On the other hand, in assignment of a special court to that task, specific rules could be adopted with respect to the selection and tenure of the judges concerned, thus minimising the democratic objection, inasmuch as the legislation that constitutional courts are empowered to strike down is the product of a democratic legislature. Relevant here is that European constitutional judges are frequently elected by the parliament, while ordinary judges are selected through more bureaucratic procedures. Moreover, constitutional judges’ tenure is greatly limited, while ordinary judges usually retain their judicial role until the age of retirement. *19
These structural features, which have characterised the European model since the approval of democratic constitutions after the demise of totalitarian regimes, were anticipated in the 1920s by Hans Kelsen, who for this reason is considered the father of the European model of constitutional justice. Kelsen not only envisioned its main structural features but added that, given those features, and particularly the fact that the effect of a constitutional court’s holding that a statute is unconstitutional consists in the formal expunction of that statute from the legal system, the court acts as a “negative legislature”, thus distinguished from Parliament with its positive introduction of statutes into the legal system. *20 The Kelsenian court was not a judge, or a political institution as Parliament was. Because of its specific power of reviewing the legislation, it wasn’t a judge, and it was not a political institution because the exercise of that specific power had no positive effect on the legal system.
These structural features appear very different from, if not opposite in nature to, those affecting the American model. However, when one compares the two models, even in terms of the legitimacy issue, experience needs to be taken into account. To what extent, then, does the European experience of constitutional justice correspond to the Kelsenian model?
Constitutional interpretation lies at the core of this question. Kelsen’s definition of the constitutional court as negative legislative actor presupposes that constitutions are centred on distribution of powers amongst diverse institutions, particularly on the devolution of legislative power to Parliament, and eventually on a list of rights framed in sufficiently determinate language. Constitutions of the 20th century, by contrast, are value-ridden documents, founded on principles framed in relatively indeterminate language. This indeterminacy paved the way for interpretation processes far more complex than those imagined by Kelsen. The court’s main task would lie in giving appropriate meaning to constitutional principles, rather than in merely ascertaining the compatibility of statutes with the text of the state’s constitution. Accordingly, the end of constitutional justice would consist in preserving the sense of those principles, rather than in pursuing the value of legal certainty perse.
These circumstances have affected the whole model of European constitutional justice, including the role of ordinary judges. The choice of a specialised and centralised court, as we have seen, resulted from the fear that, given the absence of a doctrine of precedent in the civil law tradition, ordinary judges would endanger the value of legal certainty. But the evolution of constitutional justice has changed these assumptions remarkably. Ordinary judges not only have abandoned that deference which had characterised their attitude toward democratically elected institutions since the French Revolution but, especially in those countries where constitutional review of legislation is made dependent on their own impulse, have become more and more involved in the constitutional interpretation process. On the other hand, the value of legal certainty has lost its crucial significance vis-à-vis the quest for preserving the sense of constitutional principles. Even on this basis, then, the European experience appears far closer to the American than at the moment of its foundation *21 , although the power to set aside unconstitutional statutes remains with constitutional courts, and structural features such as the appointment criteria and tenure of constitutional judges still reflect the Kelsenian model.
To the extent that the American and the European system of constitutional justice reveal increasing similarities in terms of their functioning, they are likely to be compared also in relation to the issue of the legitimacy of constitutional courts.
As has been noted in a general survey of constitutional justice in Western democracies, “constitutional review proves to have become the irreplaceable counterweight to the supremacy of the majority principle”. *22 However, that counterweight is not without problems, since, as we have already seen, constitutional review of legislation requires criteria of interpretation that give constitutional courts broad discretionary powers, in spite of the fact that, unlike parliaments, those courts are not democratically elected. Hence the fact is derived that the Bickelian counter-majoritarian difficulty, and the restraint/activism dilemma, affects the European system of constitutional justice to at least as great an extent as the US one.
The difference between the two systems rests instead on the fact that the issue of the legitimacy of the constitutional court has emerged, and still does so, on different sorts of occasions. The long tradition of US Supreme Court jurisprudence is frequently separated into eras corresponding to larger movements along the restraint/activism divide. The Lochner era, the period following the New Deal, the Warren Court , and — somewhat more controversially — the recent decades are depicted as representing different overall attitudes of the Supreme Court toward the legislator. And the difference among such attitudes depends essentially on whether the rulings tend to defer to the legislator or to declare void its statutes.
When we turn to the European courts’ experience, it is very difficult to find something similar. From time to time, constitutional review is reproached as impermissibly interfering in the legislative process — e.g., in Germany in the 1970s and in France in 1986 — but these tensions appear insufficient to bring about distinct periods of constitutional jurisprudence at diverse points along the restraint/activism divide.
In the European experience, that divide emerges instead in the context of establishment of positive criteria for legislation. Constitutional courts — the German, the Italian, and the Spanish, particularly — have abandoned the Kelsenian model also with respect to the definition of the court as a negative legislature. *23 The establishment by the court of positive criteria for legislation poses clearly the question of the court’s legitimacy, corresponding to the European version of that question: the more a court dictates positive prescriptions to the legislator, the more it applies an activist attitude, which might run counter to the democratic principle.
Positive decisions of constitutional courts have met scholarly criticism, to the extent that they anticipate the substantive content of future regulations. In that case, the court might further the tendency of the legislator to remove from himself the burden of decision. At the same time, the adoption by the court of overly detailed prescriptions for the legislative process might undermine the actualisation of the constitution through law, which in all democratic countries remains initially with legislative institutions, characterised not only by a direct democratic legitimacy but also by greater participation of the general public than that affecting the constitutional review process. *24
These recommendations are far from revealing some nostalgia for the Kelsenian model. Rather, they reflect the assumption that in democratic countries constitutional courts are expected not to insulate themselves from other institutions and from the general public but to ensure the openness of the democratic process. *25 This very assumption affords perhaps the best criterion for doing away with the Bickelian counter-majoritarian difficulty. An activist approach, particularly one pursued through positive decisions, should be deemed correct until it begins to impede further political debate and participation of the public in addressing the issue at hand.
A different view has been afforded recently with respect to the evolution and the perspectives of the European model. Centralisation of review of legislation, while presupposing that the laws ordinary judges must apply do not leave room for judicial lawmaking, appears in this view inconsistent with the pragmatic needs of modern societies, where legislation has ceased to be specific and categorical. Given these conditions, the centralised model has been undermined through interpretation of statutes by ordinary judges: before referring a question to the constitutional court, judges are expected to look for an interpretation of the statute that preserves its constitutional validity. Therefore, a division of labour has emerged within the centralised system between ordinary judges, who must interpret statutes in harmony with the constitution, and the constitutional court, the sole court authorised to set aside a statute. *26
Such division of labour, so the argument goes, should be reconsidered. The fundamental distinction should not be between interpreting and setting aside statutes under the constitution. The critical question should instead be that of whether the constitutional court has determined the meaning of the relevant clauses of the constitution in view of which the statute under consideration is to be examined. If the constitutional court has had such occasion, the case should be deemed relatively ‘easy’ in light of the constitutional court’s precedents. According to that hypothesis, ordinary courts would be authorised to set aside statutes. Only when ‘hard cases’ arise would those courts refer a question to the constitutional court, thus ensuring that the system, unlike the American one, remains centralised. *27
When looking beyond the difficulty of evaluating whether a case is ‘easy’ or ‘hard’, we see also that this proposal is clearly at odds with the structural features of the European model as provided for by the constitutions of those European states that have introduced centralised constitutional review of legislation, entrusting constitutional courts with the exclusive power to set aside unconstitutional statutes. The proposal amounts therefore to an infringement of explicit constitutional provisions. Such provisions could be modified only through explicit constitutional amendment. But how can the legislator distinguish ‘hard’ from ‘easy’ cases? In fact, the author does not suggest that his proposal necessitates constitutional revision. He suggests instead that the review should take place through judicial means. That judges would thus infringe on crucial constitutional provisions such as those concerning the monopoly of constitutional courts in setting aside unconstitutional statutes appears to the author a quite irrelevant matter. However, according to the general premises of European constitutional law, this concern is far from irrelevant. To the extent that they are provided for in constitutional texts, the structural features of constitutional justice are not likely to be at the discretion of judges.
Moreover, reconstruction of the relationship between ordinary judges and constitutional courts fails to consider the crucial role of constitutional interpretation. The fundamental distinction in the organisation of the centralised model of constitutional justice, we are told, consists in the division of labour between ordinary judges, who interpret the statute in harmony with the constitution, and the constitutional court, authorised to set aside unconstitutional statutes. The thesis clearly refers to the separate tasks of ordinary judges and the constitutional court with respect to legislation — that is, the object of constitutional review — but it neglects to take account of those tasks where the constitution itself is concerned: the parameter of constitutional review. So far, the thesis appears at least inaccurate. It is precisely on grounds of constitutional interpretation, as we have seen, that ordinary judges and constitutional courts are committed to a permanent dialogue that not only does not contrast with but, in fact, presupposes their distinctive roles in judicial review of legislation. To the extent that they have permitted such dialogue, these roles, corresponding to the structural features of constitutional justice as provided for via constitutional texts, need to be preserved rather than overruled.
Finally, and most importantly, the author suggests the need for what Comella terms “an important shift in the theoretical discussion regarding the problem of the legitimacy of constitutional review of legislation”, on the premise that “the literature on constitutional interpretation in some European countries has been too obsessed with the problem of how to distinguish between a genuine interpretation of a statute and the undue manipulation of its content”. *28 Such literature, Comella tells us, neglects more important issues: “Does democracy require that the majoritarian branches decide? May a system of judicial review be established? What sort of system? How should judges be appointed? Are there ways to understand the relationship between the courts and the legislature that make the arrangement more democratic than others? And then the difficult question: What are the standards that should guide the judge when she tries to ascribe concrete meaning to the broad and morally loaded clauses of the constitution? It is a pity that this debate is neglected in favor of a discussion about what should happen with a statute once it is found to be in tension with the constitution.” *29
The above questions, however, are not likely to be dealt with at the same level. For example, the question of whether a system of judicial review may be established, and of how judges should be appointed, does find an answer in the constitutional text. It is at the level of that answer that the question needs to be interpreted. Other questions, including that of the legitimacy issue, instead remain necessarily open to diverse approaches. At any rate, as the quotations offered in the preceding paragraphs suffice to demonstrate, it is simply disingenuous to assert that such questions are neglected in the literature on European constitutional justice.
Comella adds that “Kelsen has often been inaccurately used to buttress this incorrect understanding of the problem of legitimacy”, since he never intended the formula of the constitutional court as negative legislature to express the standard for measuring the legitimacy of the constitutional court itself. Kelsen insisted, according to Comella’s argument, that constitutional review should take place only with respect to rather specific clauses of the constitution, on the presumption that the final authority to interpret the more abstract clauses that protect, e.g., ‘justice’, ‘liberty’, or ‘equality’ should rest with the parliament. Thus, Comella believes it therefore to be a mistake “to use Kelsen to justify the idea that the problem of legitimacy arises when the constitutional court, instead of simply declaring a statute unconstitutional, partially readjusts it in order to save its validity, acting, as it were, as a ‘positive legislature’”. *30
Here Comella fails to consider the relationship between the framing of the constitutional text and the role of the court vis-à-vis the legislature that clearly arises from Kelsen’s writings. His conviction that the interpretation of general or abstract clauses of the constitution should rest with the parliament is in fact the clearest demonstration of his fear that, in the opposite case, the court would risk becoming a positive legislature. The more generally framed the clauses of the constitution are, the more their interpretation by the court might leave room for political appreciation that he thought alien to the model of the court as negative legislature. In the European constitutions of his time, as mentioned above, general clauses were rather rare, and this served to encourage Kelsen to suggest omitting them from the scope of the court’s actions. The European constitutions of today, by contrast, are framed in the language of principles — namely, of general clauses. Without this characterisation, our constitutions simply lose their significance. This shift has brought about the problem of the legitimacy of constitutional courts as positive legislatures. And this is also why relying on Kelsen with the aim of ascertaining the distance of the experience of European constitutional justice from the original model is far from being an ill-founded approach.