Mattias Kumm*
The debate about judicial review is not over. In the latest round of contributions on what is one of the classical issues of Post World War II constitutionalism, Jeremy Waldron[1] and Richard Bellamy[2] restate, sharpen and refine old arguments against the authority of courts to set aside or declare null and void legislation on the grounds that it violates constitutional or human rights.
The core criticism of judicial review is
focused on two main grounds. First, at least in reasonably mature liberal
democracies there is no reason to suppose that rights are better protected by
this practice than they would be by democratic legislatures. In particular the
legalist nature of judicial rights discourse, its focus on text, history,
precedent etc., tend to unhelpfully distract from the moral issues central to
the validation of rights claims, whereas these legalistic distractions do not
burden political debate. Second, quite apart from the outcome it generates,
judicial review is democratically illegitimate. The protection of rights might
be a precondition for the legitimacy of law, but what these rights amount to in
concrete circumstances is likely to be subject to reasonable disagreement
between citizens. Under those circumstances the idea of political equality
requires that rights issues too should be decided using a process that provides
for electoral accountability. To some extent the arguments Waldron and Bellamy
make, like the debate over judicial review more generally, is unlikely to
resonate strongly in
I will argue that Waldron and Bellamy
address the right kind of concerns, but they get things exactly wrong. First,
outcomes are likely to be improved with judicial review. The essay
defends conventional wisdom against the challenge of legalist distortion, but
does so in a way that is focused specifically on contemporary European human
and constitutional rights practice. In this practice the legalist distortions
that Waldron in particular describes are mostly absent. Instead in
Second, even though the Rationalist Human Rights Paradigm does not provide much in terms of legal constraint and authoritative guidance for courts adjudicating rights claims, this does not exacerbate or confirm the legitimacy problem that sceptics claim is at the heart of the case against judicial review. The opposite is true. Under reasonably favourable circumstances of a mature liberal democracy judicial review is a necessary complement to democratically accountable decision-making. Both judicial review of legislation and electoral accountability of the legislator give institutional expression to co-original and equally basic commitments of liberal-democratic constitutionalism. Both are central pillars of constitutional legitimacy. Judicial review deserves to be defended not only on the pragmatic grounds that it leads to better outcomes, but also as a matter of principle.
At the heart of a defence of judicial review has to be an account of the point of such a practice. That account has to both fit the practice it purports to defend and articulate what is attractive about it.[4] An account can fail either because it does not meaningfully connect to an actual practice or because it does not show what is attractive about it. The rich literature on judicial review generated by US scholars[5] that generally addresses US Constitutional practice does not capture some central features of European Constitutional practice. It does not fit that practice and therefore does little to illuminate it.[6] More specifically none of that literature captures the distinct structural features central to the Rationalist Human Rights Paradigm. On the other hand those comparative or European constititutional scholars more attuned to the core features of the Rationalist Human Rights Paradigm[7] that dominates European practice have not provided well-developed persuasive accounts about why such a practice should be regarded as attractive. This essay is an attempt to provide the barebones structure of such an account. It can only present the argument in a cursory and underdeveloped way and does not claim to do justice to the rich set of questions that will be encountered or the considerable literatures that address them.
The point of judicial review, I will argue,
is to legally institutionalise a practice of Socratic contestation. Socratic
contestation refers to the practice of critically engaging authorities, in
order to assess whether the claims they make are based on good reasons. This
practice, described most vividly in the early Platonic dialogues,[8]
led to understandable frustration of many of the established authorities whose
claims Socrates scrutinised and found lacking. It led the historical Socrates
to be convicted and sentenced to death for questioning the gods of the
community and corrupting youth in democratic
The first part of the essay will highlight the core structural features of the Rationalist Human Rights Paradigm that informs much of European human and constitutional rights practice. The second part will argue that the point to institutionalise a rights-practice that has this structure is to legally establish a practice of Socratic contestation. Socratic contestation is a practice that gives institutional expression to the idea that all legitimate authority depends on being grounded in public reasons, that is, justifiable to others on grounds they might reasonably accept.[9] In practice Socratic contestation is well suited to address a wide range of ordinary pathologies of the political process. The third part first puts both the RHRP and judicial review in a historical context, before arguing that judicial review of rights is not in tension with democratic legitimacy but a necessary complement to it. I will argue that the idea of competitive electoral politics grounded in an equal right to vote and the rights-based practice of Socratic contestation are complementary basic institutional commitments of liberal democratic constitutionalism, whose legitimacy does not turn exclusively on outcome related arguments. Liberal democracy without judicial review would be incomplete and deficient. A final part will contain some tentative hypothesis about why there has been so much debate about the counter-majoritarian difficulty and judicial review and so little about the majoritarian difficulty. It is only in the Europe of the last fifty years that the liberal democratic constitutional tradition has gradually begun to emancipate itself from the authoritarian, collectivist –and often nationalist– biases that have, in the form of constitutional theories of democracy as collective self-government, continued to inform a great deal of constitutional thinking in the age of the nation state. The shadow of Hobbes continues to hover over much of contemporary constitutional theory.
Human and constitutional rights practice in
(1) It is true that not all constitutional or human rights listed in legal documents require proportionality analysis or any other discussion of limitations. The catalogues of rights contained in domestic constitutions and international human rights documents include norms that have a simple categorical, rule like structure. They may stipulate such things as: “the death penalty is abolished”; or “every citizen has the right to be heard by a judge within 24 hours after his arrest”. Most specific rules of this kind are best understood as authoritative determinations made by the constitutional legislator about how all the relevant first order considerations of morality and policy play out in the circumstances defined by the rule. Notwithstanding interpretative issues that may arise at the margins, clearly the judicial enforcement of such rules is not subject to proportionality analysis or any other meaningful engagement with moral considerations.
But at the heart of modern human and constitutional rights practice are rights provisions of a different kind. Modern constitutions establish abstract requirements such as a right to freedom of speech, freedom of association, freedom of religion etc. These rights, it seems, can’t plausibly have the same structure as the specific rights listed above. Clearly there must be limitations to such rights. There is no right to shout fire in a crowded cinema or to organise a spontaneous mass demonstration in the middle of Champs Elysées during rush hour. How should these limits be determined?
In part constitutional texts provide further insights into how those limits ought to be conceived. As a matter of textual architecture it is helpful to distinguish between three different approaches to the limits of rights.
The first textual approach is not to say
anything at all about limits. In the
The second approach is characteristic of Human Rights Treaties and Constitutions enacted in the period following WWII. Characteristic of rights codifications during this era is a bifurcated approach. The first part of a provision defines the scope of the right. The second describes the limits of the rights by defining the conditions under which an infringement of the right is justified. Article 10 of the European Convention of Human Rights, for example, states:
“Everyone has the right to
freedom of expression […]; the exercise of these freedoms […] may be subject to
such formalities, conditions, restrictions or penalties as prescribed by law
and are necessary in a democratic society, in the interest of national
security, territorial integrity or public safety”.
Similarly, Article 2 § 1 of the German Basic Law states that “every person has the right to the free development of their personality, to the extent they do not infringe on the rights of others or offend against the constitutional order or the rights of public morals”.
The first part defines the scope of the interests to be protected – here: all those interests that relate respectively to “freedom of expression” or “the free development of the personality”. The second part establishes the conditions under which infringements of these interests can be justified: “restrictions […] necessary in a democratic society in the interests of” and “when the limitations serve to protect the rights of others, the constitutional order or public morals”. The first step of constitutional analysis typically consists in determining whether an act infringes the scope of a right. If it does a prima facie violation of a right has occurred. The second step consists in determining whether that infringement can be justified under the limitations clause. Only if it can not is there a definitive violation of the right.
Even though the term proportionality is not generally used in constitutional limitation clauses immediately after WWII, over time courts have practically uniformly interpreted this kind of limitation clauses as requiring proportionality analysis. Besides the requirement of legality –any limitations suffered by the individual must be prescribed by law– the proportionality requirement lies at the heart of determining whether an infringement of the scope of a right is justified.
Finally more recent rights codifications often recognise and embrace this development and have often substituted the rights-specific limitation clauses by a general default limitations clause.[12]
Article II § 112 of the recently negotiated European Charter of Fundamental Rights, for example, states that “subject to the principle of proportionality, limitations may be made only if they are necessary and genuinely meet the objectives of general interest recognised by the Union or the need to protect the rights and freedoms of others”.
(2) The connection between rights and proportionality analysis has been thoroughly analysed by Robert Alexy.[13] According to Alexy the abstract rights characteristically listed in constitutional catalogues are principles. Principles, as Alexy understands them, require the realisation of something to the greatest extent possible, given countervailing concerns. Principles are structurally equivalent to values. Statements of value can be reformulated as statements of principle and vice-versa. We can say that privacy is a value or that privacy is a principle. Saying that something is a value does not yet say anything about the relative priority of that value over another, either abstractly or in a specific context. Statements of principle, express an ‘ideal ought’. Like statements of value they are not yet, as Alexy puts it, “related to possibilities of the factual and normative world”. The proportionality test is the means by which values are related to possibilities of the normative and factual world. Whenever there is a conflict between a principle and countervailing concerns, the proportionality test provides the criteria to determine which concerns take precedence under the circumstances. The proportionality test provides an analytical structure for assessing whether limits imposed on the realisation of a principle in a particular context are justified.
The proportionality test is not merely a convenient pragmatic tool that helps provide a doctrinal structure for the purpose of legal analysis. If rights as principles are like statements of value, the proportionality structure provides an analytical framework to assess the necessary and sufficient conditions under which a right takes precedence over competing considerations as a matter of first order political morality. Reasoning about rights means reasoning about how a particular value relates to the exigencies of the circumstances. It requires general practical reasoning.[14]
An example drawn from the European Court of Human Rights [hereinafter ECHR] illustrates how proportionality analysis operates in the adjudication of rights claims.
In Lustig-Prean and Beckett v. United Kingdom[15] the applicants complained that the investigations into their sexual orientation and their discharge from the Royal Navy on the sole ground that they are gay violated Article 8 of the European Convention of Human Rights [hereinafter ECHR]. Article 8, in so far as is relevant, reads as follows:
“Everyone has the right to respect for his private […] life. There shall be no interference by a public authority with the exercise of this rights except such as is in accordance with the law and is necessary in a democratic society […] in the interest of national security, […] for the prevention of disorder”.
Since the government had accepted that there had been interferences with the applicants’ right to respect for their private life -a violation of a prima facie right had occurred- the only question was whether the interferences were justified or whether the interference amounted to not merely a prima facie, but a definitive violation of the right. The actions of the government were in compliance with domestic statutes and applicable European Community Law and thus fulfilled the requirement of having been ‘in accordance with the law’. The question was whether the law authorising the government’s actions qualified as ‘necessary in a democratic society’. The Court has essentially interpreted that requirement as stipulating a proportionality test. The following is a reconstructed and summarised account of the court’s reasoning.
The first question the Court addressed
concerns the existence of a legitimate
aim. This prong is relatively easy to satisfy in cases where the
constitutional provision does not specifically restrict the kind of aims that
count as legitimate for justifying an interference with a specific right. In
this case the constitutional provision limits the kind of aims that count as
legitimate for the purpose of justifying an infringement of privacy. Here the
The next question is, whether disallowing gays from serving in the armed forces is a suitable means to further the legitimate policy goal. This is an empirical question. A means is suitable, if it actually furthers the declared policy goal of the government. In this case a government commissioned study had shown that there would be integration problems posed to the military system if declared gays were to serve in the army. Even though the Court remained sceptical with regard to the severity of these problems, it accepted that there would be some integration problems if gays were allowed to serve in the armed forces. Given this state of affairs there was no question that, as an empirical matter, these problems are significantly mitigated if not completely eliminated by excluding gays from the ranks of the armed forces.
A more difficult question was whether the prohibition of homosexuals serving in the armed forces is necessary. A measure is necessary only if there is no less restrictive but equally effective measure available to achieve the intended policy goal. This test incorporates but goes beyond the requirement known to US constitutional lawyers that a measure has to be narrowly tailored towards achieving the respective policy goals. The ‘necessary’ requirement incorporates the ‘narrowly tailored’ requirement, because any measure that falls short of the ‘narrowly tailored’ test also falls short of the necessity requirement. It goes beyond the ‘narrowly tailored’ requirement, because it allows the consideration of alternative means, rather than just insisting on tightening up and limiting the chosen means to address the problem. In this case the issue was whether a code of conduct backed by disciplinary measures, certainly a less intrusive measure, could be regarded as equally effective. Ultimately the Court held that even though a code of conduct backed by disciplinary measures would go quite some way to address problems of integration, the government had plausible reasons to believe that it does not go so far as to qualify as an equally effective alternative to the blanket prohibition.
Finally the court had to assess whether the measure was proportional in the narrow sense, applying the so-called ‘balancing test’. The balancing test involves applying what Alexy calls the ‘Law of Balancing’: “the greater the degree of non-satisfaction of, or detriment to, one principle, the greater must be the importance of satisfying the other”.[16]
The decisive question in the case of the
gay soldiers discharged from the British armed forces is whether on balance the
increase in the morale, fighting force and operational effectiveness achieved
by prohibiting gays from serving in the armed forces justifies the degree of
interference in the applicant’s privacy or whether it is disproportionate. On
the one hand the court invoked the seriousness of the infringement of the
soldiers’ privacy, given that sexual orientation concerns the most intimate
aspect of the individual’s private life. On the other hand the degree of
disruption to the armed forces without such policies was predicted to be
relatively minor. The Court pointed to the experiences in other European armies
that had recently opened the armed forces to gays, the successful cooperation
of the UK army with allied NATO units which included gays, the availability of
codes of conduct and disciplinary measures to prevent inappropriate conduct, as
well as the experience with the successful admission of women and racial
minorities into the armed forces causing only modest disruptions. On balance
the
The example illustrates two characteristic
features of rights reasoning. First, a rights-holder does not have very much in
virtue of his having a right. More specifically, the fact that a rights holder has
a prima facie right does not imply
that he holds a position that gives him any kind of priority over
countervailing considerations of policy. An infringement of the scope of a
right merely serves as a trigger to initiate an assessment of whether the infringement
is justified. But the fact that rights are not trumps in this sense does not
mean that they provide no effective protection. The example demonstrates that
in practice, even without such priority, rights can be formidable weapons. The
second characteristic feature of rights reasoning is the flip side of the
first. Since comparatively little is decided by acknowledging that a measure
infringes a right, the focus of rights adjudication is generally on the reasons
that justify the infringement. Furthermore, the four-prong structure of
proportionality analysis provides little more than a structure which functions
as a checklist for the individually necessary and collectively sufficient
conditions that determine whether the reasons that can be marshalled to justify
an infringement of a right are good reasons under the circumstances. Assessing
the justification for rights infringements is, at least in the many cases where
the constitution provides no specific further guidance, largely an exercise of
structured practical reasoning without many of the constraining features that
otherwise characterises legal reasoning. Rights reasoning under this model,
then, shares important structural features with rational policy assessment.[17]
The proportionality test merely provides a structure for the justification
of an act in terms of public reason.
(3) Conceiving rights in this way also
helps explain another widespread feature of contemporary human and
constitutional rights practice that can only be briefly be pointed to here. If
all you have in virtue of having a right is a position whose strength in any
particular context is determined by proportionality analysis, there are no
obvious reasons for defining narrowly the scope of interests protected as a
right. Shouldn’t all acts by public authorities effecting individuals meet the
proportionality requirement? Does the proportionality test not provide a
general purpose test for ensuring that public institutions take seriously
individuals and their interests and act only for good reasons? Not
surprisingly, one of the corollary features of a proportionality oriented human
and constitutional rights practice is its remarkable scope. Interests protected
as rights are not restricted to the classical catalogue of rights such as freedom
of speech, association, religion and privacy narrowly conceived. Instead with
the spread of proportionality analysis there is a tendency to include all kinds
of liberty interests within the domain of interests that enjoy prima facie protection as a right. The
European Court of Justice, for example, recognises a right to freely pursue a
profession as part of the common constitutional heritage of member states of
the European Union, thus enabling it to subject a considerable amount of social
and economic regulation to proportionality review. The European Court of Human
Rights has adopted an expansive understanding of privacy guaranteed under Article
8 ECHR and the German Constitutional Court regards any liberty interest
whatsoever as enjoying prima facie protection as a right. In
But what is the point of authorising courts to adjudicate just about any policy issue, once it is framed as an issue of rights within the RHRP?
(1) There is a puzzle relating to the wisdom of judicial review that shares many structural features of the puzzle of Socratic wisdom, as it becomes manifest in Plato’s early dialogues. The kind of claims that have to be made on behalf of constitutional courts to justify their role in public life, are, prima facie, as improbable as the claims of wisdom made by and on behalf of Socrates, to justify his way of life to run around and force members of the Athenian political establishment into debates about basic questions of justice and what it means to live your life well.
That puzzle is not plausibly resolved, but only deepened, by pointing to authority: True, in the case of Socrates it is the Oracle of Delphi that determines that Socrates is the wisest man.[19] Similarly, constitutional law and European Human Rights Law have authoritatively established courts with the task to serve as final arbiters of human and constitutional rights issues as a matter of positive law, presumably believing that this task is best left to them rather than anyone else. But of course the puzzle remains. How can these authorities be right? Does it make any sense? There is a puzzle here. Socrates, a craftsman by trade, denies that he has any special knowledge about justice or anything else. He is not and makes no claim to be the kind of philosopher king that Plato would later describe as the ideal statesman in the Republic.[20] In fact he insists that the only thing he does know is that he knows nothing. Similarly a constitutional or human rights court, staffed by trained lawyers, is not generally credited with having special knowledge about what justice requires and constitutional judges widely cringe at the idea that they should conceive of themselves as philosopher kings,[21] no doubt sensing their own ineptness. The only thing judges might plausibly claim to know is the law. Ironically, this is much the same as saying they know nothing, because within the rationalist human rights paradigm, the law -understood as the sum of authoritatively enacted norms guiding and constraining the task of adjudication- typically provides very little guidance for the resolution of concrete rights claims. Just as there is no reason to believe that a man of humble background and position such as Socrates is the wisest man alive, there seems to be no reason to believe that courts staffed by lawyers are the appropriate final arbiters of contentious questions of right, second-guessing the results of the judgment made by the democratically accountable politically branches using the check-list that the proportionality test provides.
But perhaps the specific wisdom of Socrates and constitutional judges lies not in what they know about theories of justice or policy, but in the questions they know to ask others who have, at least prima facie, a better claim of wisdom on their side. When Socrates is told that he is the wisest man, he goes and seeks out those who seem to have a better claim on wisdom and scrutinises their claims. It is only in the encounter with those who are held out as wise or think of themselves as wise that Socrates begins to understand why the Oracle was right to call him the wisest man alive. Socratic questioning reveals a great deal of thoughtlessness, platitudes, conventions or brute power-mongering that dresses up as wisdom, but falls together like a house of cards when pressed for justifications. His comparative wisdom lies in not thinking that he knows something, when in fact he does not, whereas others think they know something, which, on examination it turns out they don’t.
At this point it is useful to take a closer look at what the Socrates of Plato’s early dialogues is actually doing. How exactly does he engage others? First, Socrates is something of an annoying figure, insisting on involving respected establishment figures, statesmen first of all,[22] wherever he encounters them in conversations about what they claim is good or just, even when they don’t really want to or have had enough. In some dialogues the other party runs away in the end, in others the other party resigns cynically and says yes to everything Socrates says just so that the conversation comes to an end more quickly. He forces a certain type of inquiry onto others. Second, the characteristic Socratic method in Plato’s earlier dialogues is the elenchus.[23] On a general level elenchus “means examining a person with regard to a statement he has made, by putting to him questions calling for further statements, in the hope that they will determine the meaning and the truth value of his first statement”.[24] The Socratic elenchus is adversative and bears some resemblance to cross-examination. His role in the debate is not to defend a thesis of his own but only to examine the interlocutor’s. Socrates is active primarily as a questioner, examining the preconditions and consequences of the premises the other side accepts, in order to determine whether they are contradictory or plausible. Socrates does not know anything, but he wants to know what grounds others have to believe that the claims they make are true. He tests the coherence of other persons’ views. Third, Socrates does what he does in public spaces, but he does it removed from the practice of ordinary democratic politics. The type of public reasoning he engages in, he claims,[25] is impossible to sustain when the interests and passions of ordinary democratic politics intervene.
This type of Socratic engagement shares important features that are characteristic of court’s engagement with public authorities. First, courts compel public authorities into a process of reasoned engagement. Public authorities have to defend themselves, once a plaintiff goes to court claiming that his rights have been violated. In that sense, like the Socratic interlocutors, they are put on the spot and drawn into a process they might otherwise have resisted. Second, court’s engagement with public authorities shares some salient features with the Socratic elenchus.[26] At the heart of the judicial process is the examinations of reasons, both in the written part of the proceedings in which the parties of the conflict can submit all the relevant reason, to a limited extent also in the oral proceedings where they exist and, of course, in the final judgment. Furthermore in this process of reason-examination the parties are the ones that advance arguments. The court’s role consists in asking questions -particularly the questions that make up the four prongs of the proportionality test- and assessing the coherence of the answers that the parties provide it with. A court’s activity is not focused on the active construction of elaborate theories,[27] but on a considerably more pedestrian form assessing the reasons presented by others, in order to determine their plausibility. Third, this engagement takes place as a public procedure leading to a public judgment, while institutional rules relating to judicial independence ensure that it is immunised from the pressures of the ordinary political process.[28]
(2) But even if there are some important
structural similarities between the practice of Socratic contestation described
by Plato in his early dialogues and the judicial practice of engaging public
authorities when rights claims are made, what are the virtues of such a
practice? Socrates claimed that the way he lived his life -his perpetual
critical questioning - should have earned him a place of honour in
It is possible to think of the virtues of courts adjudicating human and constitutional rights in a related way.
First, the very fact that courts are granted jurisdiction to assess whether acts by public authorities are supported by plausible reasons serves as an institutionalised reminder that any coercive act in a liberal democracy has to be conceivable as a collective judgment of reason about what justice and good policy requires. It reminds everyone that the legitimate authority of a legal act depends on the possibility of providing a justification for it based on grounds that might be reasonably accepted even by the party who has to bear the greatest part of the burden. Every judicial proceeding, every judgment handed down and opinion written applying something like the RHRP is a ritualistic affirmation of this idea.
Second, it is not at all implausible that in practice the judicial process functions reasonably well to produce improved outcomes. The most persuasive way to substantiate that claim would be to analyse more closely a large set of randomly selected cases across a sufficiently wide set of jurisdictions and addressing a sufficiently wide range of issues. Such an analysis might provide a typology of pathologies of the political process that courts successfully help uncover and address. It might also uncover the limits and deficiencies of courts as they fail to live up to the task assigned to them. But none of this can be done here. Here it must suffice to provide some general observations that might go some way to establish prima facie plausibility for the claim that the availability of judicial review improves outcomes.
To begin with it might be useful to take up
another challenge by Waldron and Bellamy. Their scepticism about judicial
review producing better outcomes is not just informed by claims about the
distracting legalist nature of judicial review.
They also claim more generally, that the political process provides an
arena where sophisticated arguments can be made and deliberatively assessed. As
an example Waldron points to the abortion debate, comparing the dissatisfying
reasoning of the US Supreme Court with the rich and sophisticated parliamentary
debate in the
A much more telling example is the ECHR
case relating to gays in the military, which also originates in
There is another form of thoughtlessness
however, that judicial review is reasonably good at countering, that I will
refer to as ideological reasoning that I can only briefly describe here.
Ideological reasoning did not play a role in the case of Lustig Preen v. Beckett. But it plays a huge role in the context of
measures taken in the ‘war on terrorism’. A necessary ingredient of ideological
thinking is the idea of a powerful and vicious enemy that needs to be fought
effectively. Clearly not all claims that there is a powerful and vicious enemy
that needs to be fought effectively are ideological. Such claims might well
reflect reality, as it did when
I have identified three types of pathologies of the political process, that even mature democracies are not generally immune from and that a rights based legal practice of Socratic contestation plausibly provides a helpful antidote for. First, there is the vice of thoughtlessness based on tradition, convention or preference, that give rise to all kinds of inertia to either address established injustices or create new injustices by refusing to make available new technologies to groups which need them most. Second, there are illegitimate reasons relating to the good, which do not respect the limits of public reason and the grounds that coercive power of public authorities may be used for. Third, there is the problem of ideology. Ideological claims are claims loosely related to concerns that are legitimate. But they fail to justify the concrete measures they are invoked for, because they lack a firm and sufficiently concrete base in reality and are not meaningfully attuned to means-ends relationships.
To summarise, the legal institutionalisation of Socratic contestation helps keep alive the idea that acts by public authorities must be understandable as reasonable collective judgments about what justice and good policy require to be legitimate. This is likely to have a disciplining effect on public authorities and help foster an attitude of civilian confidence among citizens. And second, the actual practice of rights based Socratic contestation is likely to improve outcomes, because such contestation effectively addresses a number of political pathologies that even legislation in mature democracies are not immune from. Clearly both the very limited examples and the limited range of arguments that have been addressed so far do not make a comprehensive case for judicial review as Socratic contestation. But for now it must suffice to have addressed at least some powerful arguments why a certain type of judicial review, based on the RHRP, might be attractive. What remains to be explored is whether this type of judicial review raises serious issues with regard to democratic legitimacy.
There are at least two important differences between what the early Platonic Socrates was described as doing and real world judges adjudicating human and constitutional rights claims. First, the Socratic commitment to reason has something heroic about it, whereas the institutionalisation of Socratic contestation does not generally require judges to be the hero that Socrates was. Instead the impartial posture and commitment to reason-giving that characterised Socratic inquiry is secured in adjudication by means of institutional rules which guarantee relative independence from immediate political pressures. Judges find themselves in an epistemic environment, which favours, supports and immunises from serious political backlashes the kind of contestation-oriented practice, that Socrates risked dying for.[38] Second, whereas Socrates might have humiliated his interlocutors and undermined their authority, his actions did not have any immediate legal effect. The actions of courts, however, do have legal effects, often invalidating political decisions held in violation of human or constitutional rights. This raises the basic issue whether, notwithstanding a plausible claim that outcomes may be improved, legally institutionalising a practice of Socratic contestation unduly compromises constitutional democracy.
There is nothing new in understanding rights in the expansive way of the RHRP. The Declaration of Independence states that the whole point of government is to secure the rights that individuals have. And the framers of the US constitution knew that the more specific rights they enumerated in the Bill of Rights did not exhaust the rights that the constitution was established to protect.[39] In the French revolutionary tradition rights were understood in much the way the RHRP describes. The French Declaration of the Rights of Man establishes that everyone has an equal right to equality and liberty. In the enlightenment tradition that has gave rise to modern constitutionalism as defining a limited domain not subject democratic intervention. Indeed, the core task of democratic intervention in a true republic was to delimitate the respective spheres of liberty between individuals in a way that takes them seriously as equals and does so in a way that best furthers the general interest and allows for the meaningful exercise of those liberties. In this way democracy was conceived not only as rights-based, but as having as its appropriate subject matter the delimitation and specification of rights. Legislation, such as the enactment of the Code Civil, was rights specification and implementation.
Furthermore, the abstract rights, as they were articulated in the Declaration, were only specified, interpreted and implemented through the legislative process. Courts originally had no role to play whatsoever in the exercise to determine the specific content of what it means to be free and equal in specific circumstances. Courts, discredited as part of the ancien régime -the noblesse de robe- were to function as the mouthpiece of the law as enacted by the legislature and had no additional constitutional role. Rights and democracy were not conceived as in tension to one-another, but as mutually referring to one-another. Rights needed specification and implementation by democratic legislatures and the authorisation of the democratic legislatures consisted exclusively in spelling out the implications of a commitment to everyone’s right to be regarded as free and equal. Rights and democracy were co-equal and mutually dependant. Democratic actions not conceivable as rights specification and implementation -for example laws establishing one religion as the true religion- were illegitimate, as was rights specification and implementation that was not democratic. The basic rights of individuals were the exclusive subject matter of legislative intervention and, in abstract form, guided and constrained legislative intervention.[40]
The RHRP, it turns out, is little more than the constitutionalisation of this idea. There is nothing radical or new about the RHRP on the level of a conception of rights. What is new about post WWII constitutionalism is the general supervisory role of the judiciary in the process of rights-specification and implementation. In the second half of the 20th century the vast majority of countries that have gone through the experience of either national-socialist, fascist-authoritarian, communist or simply racist rule and made the transition to a reasonably inclusive liberal constitutional democracy have made a remarkable and original institutional choice. To establish a Kelsenian type constitutional court and constitutionalise rights that generally authorise those whose non-trivial interests are effected by the actions of public authorities to challenge them in court.[41] The court would then assess whether, under the circumstances, the acts of public authorities, even of elected legislatures, can reasonably be justified. Of course the primary task of delimitating the respective spheres of liberty of free and equals continuous to be left to the legislatures. Legislatures remain the authors of the laws in liberal constitutional democracies. But courts have assumed an important editorial function[42] as junior-partners and veto players in the enterprise of specifying and implementing a constitutions commitment to rights. Courts, as guardians and subsidiary enforcers of human and constitutional rights serve as an institution that provides a forum in which legislatures can be held accountable at the behest of effected individuals claiming that their legitimate interests have not been taken seriously.
But given that there is often reasonable disagreement about what rights individuals have with regard to concrete issues, should decisions relating to that disagreement not be made by a political process, in which electorally accountable political decision-makers make the relevant determinations? Was the original French institutional commitment to legislation by an elected assembly not right? Given reasonable disagreement, does the idea of political equality not demand, that everyone’s conception of how to delimitate these rights, should be given equal respect? Is the idea of political equality not undermined, when electorally unaccountable courts are empowered to override legislative decisions to make these determinations? That, as I understand it, is the core challenge posed by arguments such as those put forward forcefully by Waldron and Bellamy. In the following I will provide an argument that judicial review based on the RHRP should be regarded as basic an institutional commitment of liberal-democratic constitutionalism as electoral accountability based on an equal right to vote. There is nothing puzzling about the legitimacy of judicial review. Arguably the more interesting issue is why the practice of judicial review receives the critical attention that it does.
(1) From a historical perspective there is
a peculiar asymmetry between the critical attitude displayed towards judicial
review and the relatively untroubled embrace of representative,
electorally-mediated decision-making. Historically, the transition from direct
democracy -
At the very least it is utterly implausible to claim that through ordinary legislative procedures ‘the people themselves’ decide political questions, whereas decisions of duly appointed judges are cast as platonic guardians imposing their will on the people. Anyone who uses that language does not deserve to be taken seriously, because instead of presenting an argument they engage in a rhetorical sleight of hand. Why not say, that elected representatives have usurped the power of the people by making decisions for them? Why is the legislature the medium of ‘We the people’? And if it can be, why not say that the people themselves, through the judicial process, sometimes act to constrain a runaway legislature? What excludes the possibility of including the judiciary as a medium by which ‘We the people’ articulates itself? The rhetoric of ‘the people themselves’ sabotages clear thinking. There are no plausible reasons to identify ‘the people’ with the voice of one institution, even when that institution is a Parliament. A parliament is a parliament, not the people. You and I and the others subject to the public authorities that have jurisdiction over us, are the people. You and I, as citizens, can participate in the political process. But as individuals among millions of similarly situated individuals, practically none of us can make much difference by participating in the political process. Whether you vote or not is unlikely to ever change the government that you are under. The probability that your or my individual vote, looked at in isolation, will change anything is no higher than the probability of winning the national lottery. When we discuss political issues we may understand more deeply what we believe and who we are as citizens. Some of us may found movements and become charismatic leaders for a cause or run for office. But nothing the great majority of us will ever do is likely to bring about any meaningful change in national public policy. The most likely way that a citizen is ever going to change the outcomes of a national political process, is by going to court and claiming that his rights have been violated by public authorities. If courts are persuaded by your arguments rather the counterarguments made by public authorities, you will have effectively said ‘no, not like this!’ in a way that actually changes outcomes. In the real world of modern territorial democracy, the right to persuade a court to veto a policy is at least as empowering as the right to vote to change policy.
(2) But the puzzle deepens. The legitimacy of the political process depends on the consent of the governed. On this thinkers in the contractualist tradition as well as French and American Revolutionaries agree. Note that consent is the starting point for thinking about legitimacy, not majorities. Of course, given reasonable disagreement, actual consent is impossible to achieve in the real world. If legitimate law is to be possible at all -and given the problems that law is required to solve it had better be possible- less demanding criteria of constitutional legitimacy adapted to the conditions of real political life need to be developed to serve as real world surrogates and approximations to the consent requirement. In modern constitutional practice there are two such surrogates that need to cumulatively be fulfilled in order for law to be constitutionally legitimate. First, a political process that reflects a commitment to political equality and is based on majoritarian decision-making needs to be at the heart of political the decision-making process. This is the procedural prong of the constitutional legitimacy requirement. But this is only the first leg on which constitutional legitimacy stands. The second is outcome-oriented. The outcome must plausibly qualify as a collective judgment of reason about what the commitment to rights of citizens translates into under the concrete circumstances addressed by the legislation. Even if it is not necessary for everyone to actually agree with the results, the result must be justifiable in terms that those who disagree with it might reasonably accept. Even those left worst of and most heavily burdened by legislation must be conceivable as free and equal partners in a joint enterprise of law-giving. Those burdened by legislation must be able to see themselves not only as losers of a political battle dominated by the victorious side (ah, the spoils of victory!), they must be able to interpret the legislative act as a reasonable attempt to specify what citizens -all citizens, including those on the losing side- owe to each other as free and equals. When courts apply the proportionality test, they are in fact assessing whether or not legislation can be justified in terms of public reasons, reasons of the kind that every citizen might reasonably accept, even if actually they don’t. When such a justification succeeds a court is in fact saying something like the following to the rights-claiming litigant: “what public authorities have done, using the legally prescribed democratic procedures, is to provide a good faith collective judgment of reason about what justice and good policy requires under the circumstances; given the fact of reasonable disagreement on the issue and the corollary margin of appreciation/deference that courts appropriately accord electorally accountable political institutions under the circumstances, it remains a possibility that public authorities were wrong and you are right and that public authorities should have acted otherwise; but our institutional role as a court is not to guarantee that public authorities have found the one right answer to the questions they have addressed; our task is to police the boundaries of the reasonable and to strike down as violations of right those acts of public authorities that, when scrutinised, can not persuasively be justified in terms of public reason”. Conversely, a court that strikes down a piece of legislation on the grounds that it violates a right is in fact telling public authorities and the constituencies who supported the measure: “our job is not to govern and generally tell public authorities what justice and good policy requires; but it is our job to detect and strike down as instances of legislated injustice measures that, whether supported by majorities or not, impose burdens on some people, when no sufficiently plausible defence in terms of public reasons can be mounted for doing so”. Note how this understanding of the role of courts acknowledges that there is reasonable disagreement and that reasonable disagreement is best resolved using the political process. But it also insists that not all winners of political battles and not all disagreements, even in mature democracies, are reasonable. Often they are not. Political battles might be won by playing to thoughtless perpetuation of traditions or endorsement of prejudicial other-regarding preferences, or ideology, or straightforward interest-group politics falling below the radar screen of high-profile politics. Socratic contestation is the mechanism by which courts ascertain whether the settlement of the disagreement between the public authorities and the rights claimant is in fact reasonable. Courts are not in the business of settling reasonable disagreements. They are in the business of policing the line between disagreements that are reasonable and those that are not and ensure that the victorious party that gets to consecrate its views into legislation is not unreasonable.[43] Acts by public authorities that are unreasonable can make no plausible claim to legitimate authority in a liberal constitutional democracy. The question is not what justifies the ‘counter-majoritarian’[44] imposition of outcomes by non-elected judges. The question is what justifies the authority of a legislative decision, when it can be established with sufficient certainty that it imposes burdens on individuals for which there is no plausible justification. The judicial practice of Socratic contestation, structured conceptually by the RHRP and the proportionality test, and institutionally protected by rules relating to independence, impartiality and reason-giving, is uniquely suitable to give expression to and enforce this aspect of constitutional legitimacy. Constitutional legitimacy does not stand only on one leg.
(3) The right to contest acts of public authorities that impose burdens on the individual is as basic an institutional commitment underlying liberal-democratic constitutionalism as an equal right to vote. Just as the ideals underlying liberal democratic constitutionalism are not fully realised without the institutionalisation of genuinely competitive elections in which all citizens have an equal right to vote, they are not fully realised without a rights and public reason based, institutionalised practice of Socratic contestation. There is a symmetry here that deserves to be described in some greater length, because it helps sharpen the implications of the argument made above.
Both the constitutional justification of an equal right to vote and the legal institutionalisation of Socratic contestation do not depend exclusively on the outcomes generated. Both constitutional commitments are justified because they provide archetypal expressions[45] of basic constitutional commitments. Citizens get an equal right to vote largely because it expresses a commitment to equality. The weight of a vote is not the result of carefully calibrating different assignment of weights to outcomes. We do not ask whether it would improve outcomes if votes of citizens with university degrees, or those with children or those paying higher taxes would count for more, even though it is not implausible, that it would.[46] There are many aspects of election laws that can be tinkered with on outcome-related grounds. But any such laws much reflect a commitment to the idea that each citizens vote counts for the same to be acceptable. The same is true for the idea of Socratic contestation. It expresses the commitment that legitimate authority over any individual is limited by what can be justified in terms of public reason. If a legislative act burdens an individual in a way that is not susceptible to a justification he might reasonably accept, then it does not deserve to be enforced as law. We should not need to discuss whether or not to provide for the judicial protection of rights, even if it were not relatively obvious that outcomes are improved. What deserves a great deal of thought is how to design the procedures and institutions that institutionalise Socratic contestation. Should each individual be able to have any court address constitutional rights issues? Should there be special constitutional courts with the exclusive jurisdiction over constitutional issues? How should the judges be appointed? How long should their tenure be? What should the rules governing dissenting opinions, submission of amicus briefs, etc. be? How are the decisions by the judiciary linked to the political process? What comeback possibilities are there for the judicial branches? What are the advantages, what the drawbacks of having an additional layer of judicial review in the form of trans-national human rights protection? These are the kind of questions that need to be addressed by taking into account outcome-related considerations. But the commitment to legally institutionalise Socratic contestation reflects as basic a commitment as an equal right to vote and is, to a certain extent, immune from outcome-related critiques, much like the equal right to vote.
No doubt the successful institutionalisation
of both electoral democracy and judicial review depend on a demanding mix of
cultural, political and economic presuppositions. In
Thinking about litigation of human and constitutional rights in terms of institutionalising a form of Socratic contestation is more than an at best playful and at worst misleading analogy. It helps clarify thinking about two major questions presented by contemporary human and constitutional rights practice in Europe, that lie at the heart of the debate about judicial review. Does judicial review improve outcomes? And is it democratically legitimate? I have argued that judicial review as Socratic contestation is attractive both because it leads to better outcomes and because it reflects a deep commitment of liberal democracy.
Moreover, Socratic contestation provides an
antidote to the collectivist -and often nationalist-[47]
biases that underlies much of 20th century constitutional theorising about
democracy.[48] It is no
coincidence that in
* Professor of Law,
[1] J. WALDRON, “The Case Against Judicial Review”, Yale Law Journal, 2006, pp. 1348-1406.
[2] R. BELLAMY, Political Constitutionalism: A Republican Defense of the Constitutionality of Democracy, CUP, 2007; besides addressing the issue of judicial review, this book addresses a significantly wider range of constitutional questions.
[3] Of course this is a generalisation. In
[4] This way of framing the issue has much in common with the methodology described by R. DWORKIN, Law’s Empire, 1986; such an approach does not only provides a positive account that shows the practice in its best light, it also articulates a normative standard by which specific aspects of that practice can be criticised as falling short of what it is supposed to be.
[5] Canonical contributions include A. BICKEL, The Least Dangerous Branch, 1969; J.H. ELY, Democracy and Distrust, 1981; R. BORK and R. DWORKIN, Freedom’s Laws, HUP, 1996; C. EISGRUBER, Constitutional Self-Government, HUP, 2001; L. SAGER, Justice in Plainclothes, Yale University Press, 2004; L. KRAMER, The People Themselves, OUP, 2004; M. TUSHNET, Taking the Constitution Away from the Courts, Princeton University Press, 1999.
[6] This is a point rightly made by D. BEATTY, Ultimate Rule of Law, OUP, 2004.
[7] R. ALEXY, A Theory of Constitutional Rights, OUP, 2002; D. BEATTY, Ultimate Rule of Law, OUP, 2004.
[8] Here I follow Gregory Vlastos, who distinguished between the Socrates of the elenctic dialogues of Plato’s earlier period and the Socrates depicted in later dialogues. The elenctic dialogues are, in alphabetical order: The Apology, Charmides, Crito, Eutyphro, Gorgias, Hippias Minor, Ion, Laches, Protagoras and the first book of The Republic; See G. VLASTOS, Socrates: Ironist and Moral Philosopher, CUP, 1991, p. 46.
[9] Note how this formulation is different from the one that Scanlon uses to capture the core of the liberal contractualist conception of justice, requiring “justifiability to others on grounds they could not reasonably reject”; see T.M. SCANLON, What We Owe to Each Other, HUP, 1998. The difference between these formulations is the difference between a formulation that seeks to establish criteria for justice and one that establishes criteria for legitimacy. Even though this is in issue of some importance, it can’t be addressed here. As I will argue in greater depth below, judicial review is concerned with legitimacy, not justice.
[10] Perhaps
also for reasons relating to the structure of constitutional text in the
[11] F. SCHAUER, “Freedom of
Expression Adjudication in
[12] The Canadian Charter prescribes in Section 1 that rights
may be subject to “such reasonable limits prescribed by law as can demonstrably
be justified in a free and democratic society”. Section 36 of the South African
Constitution states that rights may be limited
by “a law of general application to the extent that the limitation is
reasonable and justifiable in an open and democratic society based on human
dignity, equality and freedom, taking into account all relevant factors,
including (a) the nature of the right; (b) the importance of the purpose of the
limitation; (c) the nature and extent of the limitation; (d) the relation between the limitation and its purpose; (e) less restrictive means
to achieve the purpose”.
[13] R. ALEXY, A Theory of Constitutional Rights, OUP, 2002. The following pages draw heavily on M. KUMM, “Constitutional rights as principles: On the structure and domain of constitutional justice”, ICON, 2004, No 2, pp. 574-596.
[14] If legal reasoning is a special case of general practical reasoning [cf R. ALEXY, A Theory of Legal Argumentation, Clarendon, 1989], reasoning about rights as principles is a special case of legal reasoning that approximates general practical reasoning without the special features that characterise legal reasoning.
[15] E.C.H.R., Lustig-Prean
and Beckett v.
[16] R. ALEXY, supra
note 8, at p. 102. Alexy illustrates the Law of Balancing using
indifference curves, a device used by economists as a means of representing a
relation of substitution between interests. Such a device is useful to
illustrate the analogy between the Law of Balancing and the law of diminishing
marginal utility.
[17]
That does
not mean that the two are identical. There are at least four differences
between substantive rights analysis and general policy assessments. First,
courts are not faced with generating and evaluating competing policy proposals,
but merely to assess whether the choices made by other institutional actors is
justified. Second, they only assess the merit of these policy decisions in so
far they affect the scope of a right. Third, specific constitutional rules
concerning limits to constitutional rights or judicial precedence establishing
rules that fix conditional relations of preference frequently exist. Fourth,
proportionality analysis leaves space for deference to be accorded to other
institutional actors. The ECHR refers to this as the ‘margin of appreciation’.
[18] For the argument that the ECJ’s human rights jurisprudence also fits the RHRP, see: M. KUMM, “Internationale Handelsgesellschaft, Nold and the New Human Rights Paradigm”, in M. MADURO and L. AZOULAI, The Past and Future of EU Law: The Classics of EU Law Revisited on the 50th Anniversary of the Treaty of Rome, Hart [forthcoming].
[19] PLATO, Apology, 21a.
[20] As Vlastos points out, only the Socrates of the middle and later
dialogues has sophisticated theories about metaphysics, epistemology, science, etc.
[21] Arguably nothing made Ronald Dworkin’s account of judging more suspect to judges then his claim that adjudication required demi-god like ‘Herculean’ intellectual labor; see ICON, 2003, No 4 [special issue on Dworkin].
[22] PLATO, Apology, 21c.
[23] For an insightful analysis, see: G. VLASTOS, “The Socratic Elenchus”, Oxford Studies in Ancient Philosophy, 1983, pp. 27-58.
[24] R. ROBINSON, Plato’s Earlier Dialectics, 2nd ed.,
[25] PLATO, Apology, 31c-32a.
[26] The claim is not that Socratic elenctic reasoning is generally like
proportionality analysis, or that cross-examination plays an important role in
constitutional litigation. Instead the claim is that courts and the early
Platonic Socrates engage in a practice, in which they challenge others to
provide reasons for their claims and then assess these reasons for their
internal consistency and coherence. In this way the two practices share salient
features. Note how in the
[27] This does not mean that there is never an occasion where theoretical sophistication is required.
[28] Interestingly
highest courts are often geographically located not in a political power centres,
but in the provinces. The ECJ is in the sleepy Duchy of Luxembourg, not in the
European political power-centre that is
[29] PLATO, Apology,
30 E.
[30] PLATO, Menon, 84.
[31] PLATO, Republic, Book I.
[32] In modern political thought it was Hannah Arendt, who, through the figures of Socrates on the one hand and Eichmann on the other, reflected on the link between justice and the practice of thinking on the one hand and evil and thoughtlessness on the other; see H. ARENDT, The Life of the Mind, Harvest, 1981, pp. 3-16.
[33] J. WALDRON, “The Case Against Judicial Review”, supra note 1.
[34] Supreme Court of the
[35] Supreme Court of the
[36] Of course I don’t know how widespread these sentiments were in the British officer corps, the ministerial bureaucracy or the parliament at the time. I am merely assuming that the British context was, at the time, not radically different from views I encountered discussing this and connected issues while doing military service in the relative progressive German Bundeswehr in the 1980s.
[37] H. ARENDT, The Origins of Totalitarianism, Harvest, 1985, pp. 461-474.
[38] L. SAGER, Justice in Plainclothes,
[39] See the 9th Amendment of the
[40] Even today
[41] Individuals can either vindicate their rights by filing a complaint with the constitutional court after exhausting other remedies, or they have to convince an ordinary court that their constitutional claim is meritorious, requiring that court to refer the issue to the constitutional court.
[42] P. PETTIT, Republicanism: A Theory of Freedom and Government, OUP, 1997. Larry Sager refers to the court as a redundant ‘quality control’ mechanism; see L. SAGER, Justice in Plainclothes, YUP, 2004. Tom Franck refers to the function of courts as ‘providing a second opinion’; see T. FRANCK, Proportionality as Global Second Opinion, unpublished [on file with the author].
[43] Of course the very fact of rights litigation suggests that there is also reasonable disagreement about the limits of reasonable disagreement. Here the original argument about reasonable disagreement about rights as the proper domain of the democratic process resurfaces on the meta-level. But whereas it is a plausible claim to suggest that disputes about justice are at the heart of what the democratic process is about, it is not as obvious that the democratic process is also good at policing the domain of the reasonable. At any rate, there is no reason not to entrust the task of delimitating the domain of the reasonable to courts, both as a matter of principle -giving expression to the link between legitimacy and reasonableness- and because it improves outcomes [see infra].
[44] For an account of the ‘counter-majoritarian difficulty’ as an
academic obsession in
[45] For the idea of legal archetype as a legal rule emblematic for a wider commitment, see: J. WALDRON, “Torture and Positive Law: Jurisprudence for the White House”, Columbia Law Review, 2005, pp. 1681-ff.
[46] Even when the right to vote is withdrawn, as it is in many states for convicted prisoners, the reasons for doing so are not outcome-oriented, but seek to punish the prisoner by expressly denying him the status of an equal member of the political community.
[47] Is it a coincidence that the only member states of the European
Union that do not have a strong form of judicial review have a monarch as a
head of state (e.g., Britain,
Holland, Denmark, Finland, Sweden) as a representative of national unity? Is it
a coincidence that, with the exception of
[48] This authoritarian and collectivist bias is something that 20th century constitutional theory shares with much of 18th century constitutional thinking. It is a remarkable feature of continental thinkers such as Rousseau, Kant or Fichte, how quickly they move from the conceptual starting point of citizens recognising each other as free and equal to apologists of a strong centralised state. The shadow of Hobbes has cast a long shadow and continues to linger over much of constitutional theory.
[49] Article 6 EUT mentions human rights, democracy and the rule of law as foundations of law. Ultimately these in turn are derived from a commitment of human dignity [see infra].
[50] Carl Schmitt’s insinuation that existential seriousness is exclusively connected to the drama of the ‘political’ as he understands it is so evidently impoverished that it is difficult not to accept it as an invitation to psychologies; see Der Begriff des Politischen, 3d ed., Duncker & Humblodt, 1963.
[51] There are two other, more specific rules derived from human dignity: first, the general prohibition of instrumentalisation of persons without their consent; second, an anti-humiliation rule. For the way that human dignity has been understood in human rights discourse, see: D. KRETZMER and E. KLEIN, The Concept of Human Dignity in Human Rights Discourse, Kluwer, 2002.