The Nature of the Authority of Constitutional Courts

3.2.1 The Mainstream Understanding of the Nature

of Legal Authority

According to the mainstream jurisprudential position about the nature of law, one of the necessary and distinctive features of a legal system is that it both possesses de facto or effective authority and either claims or is believed to have de iure or legiti­mate authority.

The kind of authority that we are considering here is of the practical, rather than theoretical, type. Practical authorities do not merely affect one’s judgment by giv­ing him or her a reason to believe in something, as theoretical authorities do, but provide instead a reason to act in a certain way (Raz 1994a, 211).

The basic idea is that judgments of legal authorities provide reasons for action of a special type, which are regarded as ‘content-independent’. As Hebert Hart has put it, a reason is ‘content-independent’ when it purports to ‘function as a reason inde­pendently of the nature or character of the actions to be done’ (Hart 1990, 101). If an authority commands me to act in a certain way, I need not to assess whether her judgment is correct or whether she is acting on the right reasons, since the directive of the authority replaces the first-order reasons that I might have to determine the action to be performed. When a policeman signals for me to stop at a crossroad, it is irrelevant whether I believe that the way is clear and I could safely cross the road at a given time. The fact that he commands me to stop provides a reason for me to do it. Content-independent reasons, therefore, are supposed to be reasons ‘simply because they have been issued and not because they direct subjects to perform actions that are independently justifiable’ (Shapiro 2002, 389).

Hence, ‘authoritative directives are unlike ordinary reasons in that they are not only reasons to act in accordance with their content, but also reasons to pre-empt other reasons for action’ (Shapiro 2002, 404).

According to Raz, there are two important features in this arbitration example. First, though the arbitrator’s decision is a reason for action, it is related to ‘other reasons which apply to the case’. It is not merely a reason to be added to these other reasons, but rather based on them, ‘to sum them up and to reflect their outcome’ (Raz 1986, 42). Because the arbitrator’s decision depends on these other reasons that apply to the disputants, Raz call these latter reasons ‘dependent reasons’. The connection between the authoritative directives and the reasons in which they are grounded is explained by the Dependence Thesis, which says that ‘all authoritative directives should be based on reasons which already independently apply to the subjects of the directives and are relevant to their action in the circumstances cov­ered by the directive’. (Raz 1986,47).

This brings us to the Pre-Emptive Thesis, which is Raz’s second and most dis­tinctive contention about authoritative pronouncements. ‘The arbitrator’s decision’, for Raz, ‘is also meant to replace the reasons on which it depends’. In agreeing to obey her decision, the disputants ‘agreed to follow her judgment of the balance of reasons rather than their own’(Raz 1986, 42). The arbitrator’s decision is a pro­tected or pre-emptive reason because it will ‘settle for them what to do’ and displace the dependant reasons on which it is based (Raz 1986, 42). To say that the law has authority means that its existence is a protected reason for performing an action, i.e.

The law enjoys effective authority, thus, ‘if its subjects or some of them regard its existence as a protected reason for conformity’ (Raz 2009b, 29). Though this feature presupposes a positivistic account of law, insofar as the content of a legal norm must be ascertained without resource to a moral argument, it also establishes an important connection between the concepts of de facto or effective authority and de jure or legitimate authority. To have effective authority the law must both ‘claim that it possesses legitimate authority’ and be ‘capable of possessing legitimate authority’ (Raz 1994a, 215). The explanation of effective authority requires that of legitimate authority, since a person has effective authority ‘only if the people over whom he has that authority regard him as a legitimate authority’ (Raz 2009b, 28).

One can notice here an important connection between law and morality, given that the effective authority of the law depends at least in part on its ability to ground the claim to authority that it raises or to convince the people subject to its authority that its normative power to enact content-independent directives is mor­ally justified.

Furthermore, Raz is convinced that the law facilitates social coordination and ‘conformity with reason’, inasmuch as it has a moral task that, put abstractly, ‘is to secure a situation whereby moral goals which, given the current social situation in the country whose law it is, would be unlikely to be achieved without it, and whose achievement by the law is not counter-productive, are realized’ (Raz 2009a, 178).

Nevertheless, this necessary connection between law and morality does not affect the Separability Thesis, which is the core commitment of legal positivism. In spite of the fact that Raz acknowledges that legal systems have a moral point, and that sometimes these systems may ‘include moral language in constitutional norms’ and allow judges to engage in moral argumentation while making validity deci­sions, the Pre-Emptive Thesis entails a strong version of the positivistic Social Sources Thesis, which argues that law ‘consists only of authoritative positivist con­siderations’, identifiable ‘without resort to further moral argument’ (Raz 1994b, 205-6).

One of the distinctive functions of the law, therefore, would be its ‘settlement function’, which is considered essential to the important coordination function assigned to the legal system. In this perspective, the law ‘provides the benefits of authoritative settlement, as well as the related but still content-independent benefits of inducing socially beneficial cooperative behavior and providing solutions to Prisoner’s Dilemmas and other problems of coordination’ (Alexander and Schauer 1997, 1371).¹

3.2.2 Off the Mainstream: Greenberg and Dworkin on the Authority of Law

Of the many comments made on Raz’s Pre-Emptive Thesis, there is at least one objection that requires our attention now, which is posed by Gerald Postema. Postema thinks that the Pre-Emptive Thesis is appealing while we consider the application of a statute, but is in trouble to square with the sort of reasoning that is required by a typical common lawyer. Common law decisions ‘establish law in the course of adjudicating particular decisions’, in such a way that the content of the ratio decidendi ‘must be extracted from the recorded opinion of the precedent’, in a reasoning process that ‘depends heavily upon the interpreter’s grasp of general ^[24] principles of common law and a shared sense of reasonableness and fairness’ that inevitably rests on ‘evaluative’ or ‘moral’ argument (Postema 1996, 95-6).^[25]

If this picture is correct, then the theory of authority under consideration might be at odds with the typical reasoning of common law. How could a defendant of Raz’s Pre-Emptive Thesis respond to this? The typical reply would be to argue that in such cases the court’s rationale is unsettled, so that a subsequent court is ‘creat­ing’, instead of ‘applying’, an authoritative legal statement. In cases where the law is unsettled the judge would have to base her decision on moral or extra-legal con­siderations.^[26] Legal authorities, in hard cases, would be under a ‘legal obligation to apply extra-legal standards’, just like a lawyer is legally bound to apply foreign rules in a case involving parties from different jurisdictions (Shapiro 2011, 272).

As a consequence, Raz believes that legal reasoning implies more than simply ‘applying’ the law. When the law is underdetermined, the task of legal officials is to reason ‘in accordance with the law’, which involves more than merely ‘establishing the law’ (Raz 1994c, 332-3). Legal reasoning, in this sense, ‘is not simply reason­ing about what legal norms already apply to the case’, but also a reasoning that has ‘valid legal norms among its major or operative premises’, and ‘combines them nonredundantly in the same argument with moral or other merit-based premises’ (Gardner 2001,215-6).

Thus, the Razian theory of authority entails a model of adjudication that is not always at odds with other theories of adjudication that adopt a very different view on the character of authoritative legal statements. Perhaps an example will illustrate this point. We can consider here one of the cases that Ronald Dworkin offered to respond to Hart’s Postscript to The Concept of Law, which concerns marked-shared liability of the producers of a drug, the consumption of which has caused someone to acquire a serious medical condition unknown at the time in which it was com­mercialized (Dworkin 2006, 143).^[27]

How would Raz and Dworkin suggest that this case should be handled by a court of justice? In the light of the theoretical gap between these two authors, one could expect significant differences between the reasoning processes of the defendants of these two different points of view. Let me verify, however, the possibility of a con­nection between those different conceptions of legality. According to Dworkin, judges ‘should try to identify general principles that underlie and justify the settled law of product liability, and then apply those principles to this case’ (Dworkin 2006, 143).^[28] The arguments that should be binding upon the judges, therefore, are those that stem from the principles of political morality that provide the best possible justification for the legal system, i.e.

In spite of Raz’s strict account of authoritative legal statements, I do not believe that a Razian judge would do an entirely different job. She would almost certainly not hold that in a case of market-shared liability the law was ‘settled’, and would probably resort to extra-legal (moral) considerations in order to determine the right principle to be adopted in the case at hand. The judge would be no longer applying, but rather interpreting the law. And once we consider Raz’s general account on interpretation it becomes clear that the interpretive attitude will not deviate very much from Dworkin’s perspective. An interpretation, for Raz, should not always be understood as a ‘retrieval’ of an original meaning. It can instead be seen as ‘an explanation of the work interpreted which explains why it is important, to the extent that it is’ (Raz 1995, 170-171).^[29] Though the process of legal interpretation is informed by the ideas of authority and continuity, there is also a room for equity considerations and for innovation in the interpretation of legal sources. Legal inter­pretation should be both backward-looking, aiming to secure fidelity to the law and continuity, and forward-looking, giving weight to other moral considerations: ‘courts whose decisions determine the fortunes of many people must base them on morally sound considerations’ (Raz 2009a, 354-5).

Though there may be some specific differences in these approaches to adjudica­tion and interpretation, it is not implausible to say that in both cases moral consid­erations determine the content of the decision and the formation of the precedent that will be referred to by other courts in the future. One could wonder, thus, why Raz and Dworkin diverge so intensely about legal authority, if they both advocate theories of adjudication that have so much in common. Why is that so?

I believe that Mark Greenberg has a persuasive answer for this question, and that such answer helps us devise an alternative account to understand both the authority of law, in general, and the authority of constitutional courts, in particular. Greenberg believes that Raz and the mainstream approaches to jurisprudence are implicitly committed to a widely accepted, yet often unarticulated, view that the law is some kind of ‘ordinary linguistic meaning or mental content’, which can be ascertained by a model that might be called the ‘command paradigm’. This conception, as Greenberg puts it, holds that ‘what is authoritatively pronounced becomes a legal norm - or, equivalently, becomes legally valid - simply because it was authorita­tively pronounced’ (Greenberg 2011,44).

What explains the popularity of the Standard Picture, for Greenberg, is the fact that most of the scholars in contemporary jurisprudence share what he designates as the Explanatory Directness Thesis, which holds that the authoritativeness of the pronouncement is (1) ‘prior in the order of explanation of the obtaining of the legal norm’ and (2) ‘independent of the pronouncement’s (specific) content and conse­quences’, in a way that (3) there are no explanatory intermediates between the authoritative pronouncement’s being made and the norm’s obtaining’ (Greenberg 2011,46).

The critics of the Standard Picture, in turn, think that it is wrong to say that the validity of a norm is established in a single moment by a single authoritative enact­ment, in the form of the Explanatory Directness Thesis. Greenberg offers two examples of general conceptions of legality which depart in a significant way from the Standard Picture of legal authority.

The first conception is Dworkin’s model of ‘Law as Integrity’. This theory is based on a normative model of community that sees the polity as a ‘community of principles’, i.e. a political community in which people share the assumption that their lives are ‘governed by common principles, not just rules hammered out in political compromise’ (Dworkin 1986, 211). The law of such community must ‘be both made and seen, so far as it is possible, to express a single, coherent scheme of justice and fairness in the right relation’ (Dworkin 1986, 219). While establishing the content of the law, legislators and adjudicators must adopt the best constructive interpretation of such principles of political morality that justify the legal system as a whole. As Greenberg summarizes, the content of the law ‘bears a less straightfor­ward relation to the content of legal texts than it does on the Standard Picture’ (Greenberg 2011, 56), since it must coincide with ‘the best constructive interpreta­tion of past legal decisions’ (Dworkin 1986, 262). The content of the law, therefore, is itself dependant on interpretation.¹

The second conception is Greenberg’s own view that the content of law ‘consists of a certain general and enduring part of the moral profile’ of a given community. A society’s moral profile, on this account, consists of ‘all of the moral obligations, powers, permissions, privileges and so on that obtain in that society’ (Greenberg ^[30] 2011, 56-7). Greenberg calls his view the Dependence View. He claims that ‘the relevant part of the moral profile is that which comes to obtain in certain character­istic ways’, and that ‘the relation between the content of the law and the content of legal utterances is, roughly speaking, that the content of the law is a certain aspect of the impact of legal utterances (and other actions) on obligations, powers, and so on’ (Greenberg 2011, 57).

Both Dworkin and Greenberg depart from the Standard Picture because neither of them is convinced that the law’s authority comes to be with a content-independent enactment. The identification of authoritative enactments, for Dworkin, is merely the ‘pre-interpretive’ stage of legal reasoning (Dworkin 1986, 65-6). When a legis­lative enactment adds something to the content of law, it does not do that by simply issuing a content-independent directive that displaces the moral reasons that one may have to act in a certain way. On the contrary, it supplements the legal practice by altering the set of principles that ‘constitutes the best total justification’ of this practice (Greenberg 2011, 59). By the same token, on Greenberg’s Dependence View the legislative enactment of a statute may also add something to the law, but even when it does so ‘the explanation will be that the enactment changes the rele­vant circumstances (described in the moral profile), thus changing what people are morally required or permitted to do’ (Greenberg 2011, 59).

There are, however, relevant differences between Dworkin’s model of Law as Integrity and Greenberg’s Moral Impact Theory of Law. Whereas Dworkin upholds that the law is constituted by the set of principles that best justifies legal practices and legal institutions, Greenberg thinks that the law is the ‘moral impact or effect’ of the actions of these institutions, i.e., ‘the moral obligations that obtain in light of those actions’ (Greenberg 2014, 1301). The content of the law, for him, is not equivalent to the principles that bind legal institutions, but is rather the set of moral obligations that result from the action of legal officials. The important question, for Greenberg, is not what morally justifies the statute, but rather ‘what is morally required as a consequence of the lawmaking actions’ (Greenberg 2014, 1303). Authoritative pronouncements, therefore, change a society’s moral profile only in an indirect way. They do not create obligations directly, but rather ‘change our moral obligations by changing the relevant circumstances’ (Greenberg 2014,1310).

For the argument developed in this paper, however, Greenberg’s criticism on the Standard Picture is more important than his own view about the nature of law and its implications for legal interpretation, insofar as my worry about Raz’s position is motivated not only by the fact that it provides an unattractive explanation for how the law comes to be or how the content of the law is ascertained, but also by its failure to account for the distinctively argumentative character of law.

Raz’s method for asserting the validity of law is in trouble to provide a general account of how the legal system works, since it focuses exclusively on how indi­vidual enactments arise and assumes that the linguistic content of a legal provision is instantly created by a single utterance of its author. According to Raz, ‘to estab­lish the content of the statute, all one need to do is to establish that the enactment took place, and what is says. To do that one needs little more than knowledge of English (including technical legal English), and of the events which took place in Parliament on a few occasions’ (Raz 1994a, 221).

This general view on legal authority, according to Dworkin, would be an ‘eccen­tric conception’ of authority, which is guilty of a ‘heroic artificiality’ and contradic­tory with common sense (Dworkin 2006, 209). For Dworkin, Raz does not take seriously both (1) the theoretical disagreements about the grounds of law - i.e., the disagreements about the law’s foundations or the meta-propositions that make a legal proposition true or false - and (2) the law’s dependency upon interpretation. The problem of Raz’s account is not merely that the content of the law is equated with the content of an authoritative utterance, but that Raz believes that the law can be ascertained in a non-argumentative way.

I fear, however, that this is not the main concern of Greenberg’s objection to the Standard Picture. Whereas Dworkin appeals to a normative ideal of ‘integrity’ that requires judges to engage in a constructive interpretation with a view to making the law the best it can be, in the light of the institutional history and the basic principles of political morality underlying the legal system, Greenberg’s conception of law is not committed to legal constructivism. The content of the law can be ascertained in a different way. To determine how a legislator alters the content of the moral profile, the Moral Impact Theory ‘makes no appeal to Dworkinian interpretation ’ (Greenberg 2014, 1301). On the contrary, it assumes that ‘working out the content of the law is not a genuinely hermeneutic enterprise’, and involves instead a ‘straightforward moral reasoning about the moral consequences of various facts and circumstances’ (Greenberg 2014, 1302).^[31]

I will assume, therefore, that Dworkin’s position is a more promising candidate for replacing the Standard Picture of legal authority and offering a powerful chal­lenge to Raz’s account of the authority of law.

3.2.3 The Argumentative Character of the Law

In the debate between Raz, Greenberg and Dworkin, I am more likely to agree with the latter. Dworkin’s view that legal authority does not antecede, but rather is estab­lished by, interpretation offers a sound explanation for legal authority because it draws a broad picture that is not at odds with the legal practice and, most impor­tantly, because it provides a plausible explanation for the distinctively argumenta­tive character of law. This feature was well captured by Neil MacCormick’s institutional approach to the practice of law (MacCormick 2005, 13). According to MacCormick, as a normative order, ‘and as a practical one’, the law is ‘in continu­ous need of adaptation to current practical problems’ (MacCormick 2005, 6). The recognition of the Rule of Law as a political ideal implies the recognition of law’s domain as the ‘locus of argumentation’ (MacCormick 2005, 13).

In this interpretation, the relative indeterminacy of law is not something to be regretted. It has to do with the ideal of the Rule of Law and with the procedural rules of argumentation that are presupposed in the institutional structure that it provides. The Rule of Law itself implies a certain degree of indeterminacy in the legal system (MacCormick 2005, 26).

If this interpretation of the political ideal of the Rule of Law is correct, legal argumentation ‘carry implications for the concept of law’ that can no longer be reconciled with most forms of legal positivism (Bertea 2008).^[32]

The idea of an argumentative character of law, however, needs to be further specified. The core point is not only that law consists in a social practice that is sensitive to reasons, or that legal rulings depend on a critical appraisal of arguments pro and against a solution to a legal problem. This feature alone is plainly consistent with a positivistic understanding of the sources of law. A Hartian positivist, for instance, will necessarily hold that the internal point of view requires from legal officials a ‘critical reflective attitude’ towards the rule of recognition. But no legal positivist denies that this commitment to the rule of recognition is consistent with a lot of disagreement about whether the requirements of the rule of recognition are satisfied in a given case, and many (if not most) legal positivists claim that their accounts of legality can explain even the kind of disagreement that Dworkin classi­fied as ‘theoretical’, in the sense that it is a disagreement about the ‘grounds’ or the foundations of law. One of the positivistic replies to Dworkin could be, for instance, that the rule of recognition remains a plausible explanation both for the identifica­tion of legal rules and for law’s capacity to guide social action even if there are some exceptional and specific cases in which legal officials disagree over the crite­ria of legality allegedly laid down in the rule of recognition, or over how such crite­ria should be interpreted.^[33]

The idea that the law is argumentative becomes more important when one con­siders the way in which the law must be an argumentative practice. For Dworkin, the basic idea is that the social practice of law is distinctively argumentative, in the sense that participants of legal discourses must take up an ‘interpretive’ attitude towards the law - recognizing that it has a point or purpose that makes it valuable as such, and constructing it in the way that makes it the best it can be, on the basis of a critical appraisal of this point. To understand the law it is not enough to identify its sources; on the contrary, one must engage in a constructive interpretation of this practice. For Dworkin, therefore, the intentionality of law - or its point - is grounded on political-philosophical values, or, in other words, on a certain conception of justice. As Ronaldo Macedo explains, Dworkin thinks that ‘legal practices only achieve the sense that they actually have in the society that we live insofar as they satisfy a requirement of legitimacy’ (Macedo Junior 2013).

I am assuming, therefore, that the distinctively argumentative character of law entails the kind of interpretive attitude that Dworkin is arguing for. If this is true, the fact that the law is the outcome of an interpretation is n ot the only consequence of the argumentative character of law. Further from being dependent on a constructive interpretation, legal propositions, for Dworkin, remain open to further interpretive activity in the so-called ‘post-interpretive’ stage of legal reasoning. Hence, legal propositions must be open to new interpretive circles.

For this reason, I think that Raz is wrong to suppose that the law is to be ‘found’ in a previously determinate set of social sources. The validity of a law cannot be merely a question of fact, but rather needs to be at least in a significant part a matter of argument. The law is hardly ever ‘settled’ and its rules are ordinarily defeasible, for they are inevitably subjected to a constructive interpretation that might lead to revisions and even to exceptions in their operative conditions (MacCormick 2005, 241).

Like the late MacCormick, I am convinced that Dworkin’s perspective of Law as Integrity is at least generally correct because it explains not only how the law becomes binding in practice, but also because it provides a better account of the argumentative character of law. On Dworkin’s theory of law, the very idea of com­munity depends on an interpretive understanding of law and legal practices. Instead of a social fact that can be ascertained by a neutral observer, the law ‘is not exhausted by a catalogue of rules and principles’. It is more specifically ‘an interpretive, self- reflective attitude that makes each citizen (who is also an interpreter) responsible for imagining what his society’s public commitments to principle are, and what these commitments require in new circumstances’ (Dworkin 1986, 413).

Nonetheless, in spite of the attractiveness of this interpretive and argumentative understanding of law, I am convinced that it is not fully capable of acknowledging the ‘essence’ of the law and providing, as Finnis has said, ‘the one feature used to characterize and to explain descriptively the whole subject-matter’ (Finnis 2011,6). Dworkin’s explanation of legal authority will probably leave out a relevant set of ‘peripheral cases’ (Finnis 2011, 11). Yet, in spite of the incapacity to establish an ‘univocal meaning’ of theoretical terms, it is interesting because it is fit to explain the law’s practical purpose and offers an account of the ‘central cases’ - or, as Aristotle has put it, the ‘focal meaning’ - of the institutions of law and legal reason­ing (Finnis 2011,9-11).

What is it, then, that Dworkin’s approach to legal authority is missing?

As I will argue in the following section, it does provide reasonable explanations about how the ‘law in general’ operates, or at least how it is supposed to operate, but it fails to explain in a fully comprehensive way the authority of constitutional courts.

3.2.4 On the Authority of Constitutional Courts

I have been arguing that the distinctively argumentative character of law poses a serious challenge to Raz and the mainstream account of legal authority. As a rule, Dworkin’s interpretivism is probably a better explanation of how legal systems operate than Raz’s Pre-Emptive model of authoritative legal enactments.

I intend to argue, in this session, that there is one exception to this partial conclu­sion. This exception is constituted by the cases where a constitutional court holds that a formally correct legal enactment is unconstitutional and refuses to enforce it in the situation at hand. In systems of strong judicial review, where courts have formal authority to strike down a particular legal statute, this power is regarded as part of the ordinary process of application of law (Waldron 2006,1354).¹¹

My point is that Dworkin is probably right about the way in which the law is built and applied in the central cases, but perhaps not in a peripheral case where a constitutional court annuls an act of parliament that was duly enacted and followed the legislative procedure established in the Constitution. In these cases, the Razian conception of authority provides a clearer explanation of how the court’s authorita­tive legal pronouncement operates in the case at hand.

In these cases, the court’s decision does not limit itself to offering a ‘constructive interpretation’ of the law, as Dworkin believes. On the contrary, its pronouncement is deconstructive in the sense that it does not merely specify a principle or add up a new norm to be considered in the future cases, but rather invalidates by a single authoritative pronouncement the majority decision reached by a legislative enact­ment. Constitutional courts in such cases provide, indeed, exclusionary reasons in Raz’s sense, pre-empting the dependent reasons that one might have to comply with the enactment of the legislature. The court’s ruling acts as an exclusionary reason ^[34] because it provides a reason to ‘refrain from acting’ on the balancing of reasons undertaken by the legislature.^[35] After the court pronounces the unconstitutionality of an act, the general normative issue is no longer ‘arguable’ or open to new argu­ments and interpretations.^[36]

This assertion may appear to some as inconsistent with the reservations that I had against the Razian picture of common law reasoning while I was discussing Postema’s objections to this account.^[37] Common law reasoning, as I argued above, is a typical form of interpretive legal reasoning. Though the practice of stare decisis implies that the case law creates a legal obligation to abide by a previous decision, common law courts ‘do not treat the formulations of law in earlier cases as exhaus­tive formulations, but as formulations which were sufficiently exhaustive in the con­text in which they were made’ (Simpson 1961, 165). Such formulations, as A W Simpson argues, are such that the case law is ‘always open to latter courts to intro­duce exceptions’ which are based, at least in part, on a moral justification for distin­guishing the case at hand (Simpson 1961,165).^[38] Instead of appealing to pre-emptive reasons, ‘arguments (in common law reasoning) to the effect that this or that is the law are commonly supported by reference to ideas which are not specifically legal’ (Simpson 1973, 87).

The similarities between common law and constitutional adjudication may lead, therefore, to an objection to the point that I am trying to argue for. How, you may ask, can I hold that in constitutional adjudication the court’s pronouncement is pre­emptive in the same way as Raz thinks that the law as a whole is? What is so special about constitutional reasoning that makes the general model that I used to explain both the legislation and the common law unfit to account for its authority?

To be coherent with what I argued earlier in this essay, I must demonstrate that there is an important difference between common law cases and cases of strong judicial review. Nonetheless, this distinction does not appear to reside in the reason­ing process of the court. In the hard cases decided by a constitutional court, Dworkin’s view that constitutional adjudication is a sort of ‘moral reading’ of the constitution is descriptively more accurate than Raz’s view that when the judges resort to moral argument or other extra-legal considerations, they are no longer using their ‘legal skills’ or ‘applying the law’.^[39] On Dworkin’s description, the rea­soning of constitutional courts is neither strictly legal nor purely moral. One of its distinctive features is that moral and political concepts are embedded in the sources of law, so that many legal concepts can only make sense if they are illuminated by moral considerations (Dworkin 2006, 51). But these moral concepts do not neces­sarily retain their original senses once they have been incorporated into legal docu­ments. As Waldron persuasively explains, ‘what we have here is a melange of reasoning - across the board - which, in its richness and texture, differs consider­ably from pure moral reasoning as well as from the pure version of black-letter legal reasoning that certain naive positivists might imagine’ (Waldron 2009, 12). This hybrid or intertwined type of reasoning stems from the interpretive attitude that one is supposed to adopt while constructing the meaning of the legal sources, and is not different from the interpretive attitude of common lawyers in hard cases. Dworkin appears to be correct, therefore, about the nature of judicial reasoning, even when we are considering the reasoning of constitutional courts. Hence, the distinction between a decision of the constitutional court and a common law form of judicial decision-making lies less on the reasoning process than on the effects of the decision.

I do not think, however, that this poses an insurmountable difficulty for the point that I am trying to raise about the pre-emptive authority of constitutional courts. To say that the reasoning process followed by a constitutional court, when it declares that a given statute is void or no longer valid, is not qualitatively different from that of a common lawyer, when she follows a precedent, does not amount to saying that the nature the authoritative settlement of these decisions is necessarily the same.

Even though some would argue that the ‘settlement function’ of constitutional courts is the same as that of the ‘rule of precedent’, I believe that this may or may not be true, depending on the case at hand.^[40]

In order to illustrate this point, let us compare two abortion cases that have been decided by different constitutional courts. In Roe v. Wade, on the one hand, the Supreme Court of the United States decided that the right to privacy, under the ‘due process clause’ of the 14th Amendment, should be extended to protect a woman’s decision to have an abortion. The court held that in regard to abortions performed during the first trimester of pregnancy, the decision must be left to the woman and to the judgment of her doctor. States lack authority to limit this right, even if it is to protect the potentiality of human life. By implication, thus, all statutes enacted to prevent women from making an abortion before the third trimester lack legal valid­ity and should be regarded as deprived of legal effects.^[41] On the other hand, in a recent case of the Brazilian Federal Supreme Court, the court did not address the issue whether the federal statute which considers abortion a criminal offence is incompatible with the constitution, but decided that women have the right to have an abortion when there is a medical diagnostic that she is carrying an anencephalic fetus, and that no legislation that considers the abortion a criminal offence shall be applicable to these cases.^[42]

By considering these cases, I am not interested in the answer to the question whether abortion is morally or legally accepted. On the contrary, I want to focus on the level of generality of the decision and on whether it leaves open a post- interpretive revision of its contents by future constitutional judges. My intention is merely to show that in the former decision the court provides to the officials and the people in general an exclusionary reason for acting in a certain way and quashes the statutes that counter the authoritative pronouncement issued in the holding of the judicial decision, whereas in the latter decision the court merely solves an ‘applica­tion problem’, with a view to determining the appropriate application of a constitu­tional norm to a situation whose features are unspecified in the literal wording of the legislative enactment (Gunther 1993, 38). What the court aimed to do in this case is merely to determine whether X has a right to abortion all things considered. The court no longer solves a ‘problem of justification’ and is not concerned with the justification of ‘only the norm itself’, but rather engages in an ‘application dis­course’ where the real issue is not whether a norm is generally valid, but if it ought to be followed ‘in a particular situation in the view of all the particular circum­stances’ of the case (Gunther 1993, 36-38). One can see, therefore, that in the latter case the court proceeds nearly in the same manner as a common lawyer. Constitutional adjudication, in such types of decision, coincides with the ‘common law conception’ of constitutionalism supported by Wil Waluchow. One of the nuclear points of a Charter of Rights, for Waluchow, is precisely to compensate for our lack of knowledge, in advance, about ‘what our rights and freedoms are’ in every dispute. By investing the judges with the power to interpret the constitution or the Bill of Rights, the legal system handles the ‘epistemic limitations in respect to the effects of the government action on moral rights’ (Waluchow 2007, 11).^[43]

I believe that these examples show that the idea of ‘constitutional adjudication’ can comprise more than the strong variant of judicial review. The two examples of constitutional decision that I offered in the previous paragraph are thus two alterna­tive classes of constitutional cases. While the former is a typical case of a strong form of judicial review, in which a court drafts an authoritative decision that annuls a legislative enactment and solves a general problem of legal validity, the latter merely interprets the law in a way that avoids a statute to come into conflict with a constitutional right. Thought the court fixes how the law shall be interpreted for the time being, the legislative text remains intact and is still open to new understandings at a future interpretive round. In other words, the text of the legislative provision still counts as a valid s ource of meaning in post-interpretive debates. Though the court adds new specific meanings to the decision of the legislature, it does not even purport to replace its general political judgment in a pre-emptive way.

Even the so called systems of ‘weak judicial review’ - such as the case of the United Kingdom, New Zealand and the Australian State of Victoria - admit the ‘interpretive mandate’ to read down the statutes that appear to violate the Charter of Rights whenever it is possible to do so.^[44] This interpretive mandate, as Mark Tushnet explains, directs the courts to ‘engage in two acts of interpretation: they must inter­pret the substantive rights protections, and then determine whether the statutory provision at issue can be interpreted in a manner consistent with their interpretation of the rights protections’ (Tushnet 2009, 26).

To be sure, even a legal system that does not allow any sort of judicial review of the legislation - and lack even a statutory authorization to ‘read down’ a legislative provision in order to make it coherent with a set of rights, as it was the case of France and the United Kindgom a few years ago - might accept the kind of ‘con­structive interpretation’ that nowadays is typical in systems of weak review. In the 1930s English decision R. v Bourne, for instance, a surgeon who performed an abor­tion on a 15 year old girl impregnated by violent rape was acquitted on the ground that the Crown was not entitled to prosecute him unless there was no reasonable doubt that he had not acted in good faith ‘with a view to protecting the life, understood as the continuing sane and healthy existence, of the girl in question’.^[45] As MacCormick reports, ‘the law’s express prohibition on performing or procuring abortion was (considered) subject to an implied exception’, in an interpretive move that remain accepted for three decades, until the Abortion Act was passed by Parliament in 1967 (MacCormick 2007, 248).

The decision of English courts in a case like R. v. Bourne does not differ in a significant way from that of a constitutional court in a case like the Brazilian deci­sion which introduced an exception to a generally accepted anti-abortion rule. They both lack the pre-emptive character that Raz is attributing to the legal system. These decisions resemble a law-making process ‘by aggregation’, which is typical not only of Roman jurists and contemporary common lawyers,^[46] but also of the medi­eval courts in Western Europe.^[47] The core point of this method of legal development is that the judge does not accept the assumption that there is always only one indi­vidual rule of law that determines the content of a legal obligation. Instead of endorsing the ‘atomistic’ assumption that legal obligations stem from a direct appli­cation of a previously constructed legal norm,^[48] the judge seeks to construct the law in a way that allows the h armonization of any given rule with the previous set of norms of the legal system to which it belongs (Hespanha 2006, 115). The excep­tions on the prohibition of abortion in these decisions are not supposed to ‘replace’ or ‘repeal’ the statute that they purport to qualify. On the contrary, they claim to cohere with it. The whole point of this form of judicial reasoning is to add or aggre­gate something to the current meaning of the previous set of laws, rather than replace the wording of the text by a new rule that is as general and comprehensive as the previous one.^[49]

We can see, therefore, that there are two ways in which the decision of a consti­tutional court can acquire its authority: first, by quashing a particular legislative provision and blocking any further deliberation about its merits, as it happens only in legal systems that adopt the strong form of judicial review; and second, by giving the appropriate interpretation of a basic right in concrete situations not envisaged when the Charter was originally issued. While the former purports to settle the law and determine its final interpretation, the latter holds that the content of the law is never final and is open to new interpretations. Whereas the former is better explained by the Razian conception of legal authority, the latter is at odds with such descrip­tion of how the legal system operates and is more compatible with Dworkin’s con­ception of the way authoritative enactments operate in legal practice.

One may wonder, at this point, what is the relevance of this distinction for the purposes of my inquiry in this essay. My answer will be as straightforward as pos­sible: the Razian view implies a conception of the authority of the courts that attri­butes to them a legal power much harder to justify than that of the Dworkinian view or any other account of legal authority that falls outside of the mainstream under­standing, which regards legal enactments as pre-emptive and content-independent. It might turn out to be possible that one can provide a moral justification for the latter even when such justification is not available to the former. But there is no reci­procity in this statement, since it is never the case that one can offer a justification to the former that would not work also as a justification to the latter.

I advocate, in the second part of this essay, two theses about the authority of constitutional courts: first, that there is no available moral justification to attribute to a constitutional court the kind of authority that Raz envisages, and second, that although the arguments that Raz deploys to justify the authority of law are incapable of providing a justification for these decisions, they might be able to justify the authority of a system of weak judicial review.

3.3