Why Judicial Review?

2.1 Theoretical framing

Greater attention to the motivations for judicial review is necessary in the theoretical literature as certain reasons have dominated the discussion. Broadly speaking, it is possible to divide potential motivational grounds into two categories: epistemological and functional. Epistemological arguments emphasise the comparative advantage of the judiciary in interpretation. In divining the meaning and application of a particular right, courts are said to be more reliable in interpretive exercises than legislatures and executives. Functionalist arguments are epistemically modest although possibly more empirically demanding. It is not presumed that courts possess greater moral insight than other branches of government; rather, judicial review garners its institutional advantage through its socio-political function(s).

In the prevailing scholarship on the legitimacy of judicial review, epistemological reasons are endowed with a certain pre-eminence. In this universe of argument, we find methodological agreement between two of the most-cited bookends of the debate. Dworkin establishes the question as follows: ‘The best institutional structure is the one best calculated to produce the best answers to the essentially moral question of what the democratic conditions actually are, and to secure stable compliance with those conclusions.’¹⁰ Likewise, this epistemic baseline stands central in Waldron’s critique of the substantive legitimacy defences of judicial review: ‘Outcome-related reasons, by contrast, are reasons for designing the decision-procedure in a way that will ensure the appropriate outcome (i.e., a good, just, or right decision)’.¹¹

This typology is not watertight. Both types of reasons can be nestled together. Ronald Dworkin often adds a functional claim to his epistemic one: ‘democracy requires that the power of elected officials be checked by individual rights’ and the ‘responsibility to decide when those rights have been infringed is not one that can be sensibly be assigned to the officials whose power is supposed to be limited’.¹² Other authors offer coterminous and longer justificatory lists.¹³ Moreover, the claims can substantively overlap. To take Dworkin again, he sometimes asserts the interpretive advantage of the judiciary in more functionalist tones: judicial intervention is said to not only ensure, on balance, better answers but it also restructures public discussion about rights by foregrounding principled reasoning.¹⁴ Nonetheless, in legal and political theory epistemic reasons are often foregrounded, which suggests that we need a more critical analysis of this methodological choice. Moreover, these grounds are somewhat divorced from the functional and instrumental reasons that are commonly marshalled in practice for establishing the institution of judicial review.

Equally, we need to think carefully about which types of functional reasons should count and how. For instance, in a recent article, Fallon repeats the classical lines of a functional argument:¹⁵ Courts must possess the opportunity to invalidate legislation because the mere existence of legislation is likely to be more harmful to rights.¹⁶ Judicial review provides therefore a critical and additional veto check against such risks.¹⁷ However, as shall be seen, Fallon’s reasoning has been subject to significant critique on the grounds that it cannot account for the multitude of legislation that seeks to positively protect rights.

2.2 Methodology

A second reason for examining motivational reasons is the methodological role they play in debates over legitimacy, institutional competence, and effectiveness. For normative legitimacy assessments, it is common to weigh process against output reasons in establishing when a particular coercive institution is legitimate or not.¹⁸ In Waldron’s well-known critique of judicial review, its undemocratic features are weighed against its supposed epistemic outputs. Given his cursory approach to setting out the motivational reasons for establishing judicial review, it is possible that his balancing assessment might be different if a broader palette of reasons were included.

Similar cost-benefit or balancing approaches are taken in debates over the institutional competence of the judiciary. It is often asked whether judges should possess powers to review complex and polycentric issues, ranging from national security through to the allocation of limited budgetary resources. The concern is that courts risk distorting efficient and effective public policy. In addressing this tension, Jeff King sets up an expertise-accountability trade-off: the quality of expertise for a government’s position (institutional competence) is to be balanced against the risks to individual rights (accountability function of judicial review). Thus, when a State cites “collective” expertise (a position endorsed by a government agency/department, UN agency, or professional association) this ‘greatly skews the accountability trade-off towards deference to expertise’.¹⁹ However, King’s analysis privileges one functional reason for judicial review. As we shall see, there might be other grounds that justify judicial intervention when his trade-off favours strong judicial abstention or deference. Thus, the question is of relevance to the judiciary itself as it weighs competing factors in deciding when and how to exercise its discretionary powers.

A growing literature has also tracked the effects of rights adjudication.²⁰ A key question in designing such research is determining what types of impact we expect from courts. The principal schools of thought have focused on either material impacts (changes in policies and social realities)²¹ or symbol and constitutive impacts (changes in politics and attitudes);²² although others straddle both camps. As Scheingold put it, ‘it is necessary to examine both the symbolic and the coercive capabilities which attach to rights’.²³ To a large extent, this research is guided by an underlying normative debate on the legitimacy or usefulness of judicial review and public interest litigation.²⁴ Yet, it is interesting to observe that there are fewer empirical studies on some of the normative reasons for adjudication analysed in this article, in particular the epistemic quality of judicial reasoning and the effects on deliberative reasoning.

2.3 Practical effects: A patchwork of judicial review

The final reason is that motivational grounds appear to affect the design and trajectory of judicial review in practice. On the one hand, the judicialisation of rights has flourished across the world in the wake of the Cold War.²⁵ Numerous courts occupy an important and sometimes central place in the protection of constitutional and international rights. This transmogrification is evident in the constitutional reforms in a swathe of new democracies,²⁶ constitutional renewal and heightened judicial engagement in older democracies (including Norway),²⁷ and a spreading tapestry of international courts and complaint mechanisms.²⁸ The twinning of electoral democracy with national and international judicial review constitutes a persistent feature of contemporary constitutional reform and practice. Exclusion of the latter from this equation is typically met with strong domestic and international protest.²⁹ Further, there is an expansion of rights that are subject to review. From a limited number of narrowly framed civil rights, adjudicators are now ruling on a broader swathe of rights as well as duties. As Figure 1 demonstrates, there has been a remarkable and commensurate rise in the constitutional recognition of various civil and social rights.³⁰

Figure 1

Yet, this trend is not universal. The reach of judicial review does not extend to the four corners of the world. For a start, many Asian and Middle Eastern States cannot be found on this constitutional map: courts in these regions are granted fewer powers and are more tightly restrained, while international treaty protocols for individual complaints go unratified.³² For example, the average level of acceptance of international human rights adjudicative mechanisms is strikingly low for these two regions: 1.05 and 0.6 compared to 3.08 for the rest of the world. Yet, these States ratify substantive human rights treaties at a rate just below the global average.³³ Conforming with Ginsburg’s observation of the emergence of national judicial review, the presence of electoral democracy is largely a necessary but not sufficient condition for explaining acceptance of international human rights review.³⁴ While judicial review has emerged, sometimes surprisingly, in more authoritarian regimes, it is often highly fragile.³⁵

However, ambivalence is not constrained to regions that with a sizeable share of authoritarian and anocratic³⁶ governments. In more mature democracies, constitutional reform processes have halted at the door of enhanced judicial review.³⁷ Recently, in Norway, parliamentarians could not agree on formalising the Supreme Court’s powers of judicial review which it had claimed and exercised for 194 years.³⁸ In Europe and Latin America, different coalitions of States have sought to weaken the powers of regional human rights bodies³⁹ while the tribunal for the Southern African Development Community was stripped of its powers to consider individual complaints.⁴⁰ Moreover, this uncertain picture of State commitment may be evident in assertions of patchy compliance with judgments, including by some Western European democracies.⁴¹

Jung and Rosevear argue that there has been a slowdown in constitutional recognition of judicial review of social rights. Noting the lower rate in the period 1990-2004 compared to the period 1974-1989,⁴² they suggest that the rise of the Washington-based consensus tempered recognition. However, this statistical and causal interpretation is questionable, and using the same data, Figure 2 reveals that the general trend line remains upwards.⁴³ Nonetheless, what the graph does demonstrate is the equal persistence of the non-recognition of judicially enforceable social rights during recent constitutional reform.⁴⁴ This was notable in Norway’s recent constitutional reform: The right to education and the social rights of children were included in a reformed bill of rights, but the right to health and adequate standard of living were rejected by a parliamentary super-minority.⁴⁵

Figure 2

Does this patchwork of institutionalisation reflect normative differences over the importance or risks of judicial review? Explaining the rise of judicialisation is the subject of a growing body of empirical work. Thus, the fragmented nature of expanded judicial review suggests that normative dissensus remains a factor. It is not just a matter of time before policymakers, the legal profession and the entire public become accustomed to the idea; it is a site of deeper disagreement. This makes the march of judicial review less inevitable and more conditional on changes on ideas as much as politics and culture.

3.1 The authenticity of case-based review: Evidential particularism

A commonly-cited epistemic advantage of judicial reasoning is its factual palette: the particularity and authenticity of concrete cases. The argument runs that legislative and executive reasoning tends to be dominated by general and stylised considerations. Yet, at least in the field of individual rights, such reasoning may be less appropriate, occluding the practical and problematic effects of rules (or lack thereof) on disparate individuals. Bilchitz sets out this critique of legislatures in customary fashion:

General decision-making across a range of cases can obscure the problems that may arise in particular instances to which that general decision may apply. General decision-makers may simply overlook or fail to give sufficient weight to the problems that may be faced in particular cases.⁶⁵

The claim is alluring enough. It resonates deeply with the defence of judge-made common law in Anglo-American jurisdictions. Rules and principles develop and mature best through the inductive and analogical reasoning of courts in actual cases, avoiding the ‘perils of prophecy’ through a ‘long course of trial and error’.⁶⁶ In rights jurisprudence, the scenario is common enough. A law may be highly defensible on general grounds, but its impact falls disproportionately, whether unfairly or unwittingly, on particular individuals or groups. Video surveillance, efficient criminal trials, religious education, conditions for unemployment benefits and so on may constitute positive public aims, but their consequences are unlikely to impact individuals in a uniform manner.

Nonetheless, there are serious problems with this argument (putting aside its somewhat anti-democratic overtones). First of all, it is equally possible to encounter the reverse scenario. Laws may be motivated by very particular situations without regard for their general and systemic effect on rights. A terrorist bombing, the abuse of social benefits by one family, an alleged rape by a member of an ethnic minority, a surge in homelessness in urban business districts, may all trigger legislative solutions that possess no generalised justification or grounding in empirical reality. In these circumstances, we would want courts to lift rather than concentrate the perspective.⁶⁷ Thus, we may wish to fully reverse Bilchitz’s position and ask whether the legislature has properly engaged in ‘general decision-making across a range of cases’ and evaluated the systemic impacts of rights, which may be represented (not just actualised) by an individual in a legal case.

It might be retorted that over the last two centuries, the degree to which legislation is targeted at such specific groups has waned in mature democracies. Nonet and Selznik describe a general shift from repressive legal regimes to autonomous and responsive law.⁶⁸ Repressive law is concerned the maintenance of order and selective subordination in the interests of the rulers and elite, while autonomous law offers impartial, neutral, and equal treatment and responsive law attends to the needs and values of the disempowered and disadvantaged. But vestiges of repressive law persist and its potency remains latent.⁶⁹ Every policy narrative requires an “enemy” and, even when law and politics are framed in autonomous terms, repressive motivations may be observable if not transparent.

A good illustration is A & Others v. Secretary of State for the Home Department.⁷⁰ The September 11 bombings in New York led the British parliament to embellish its fresh and comprehensive Terrorism Act of 2000. Foreigners could be detained and deported if the relevant government minister believed they were a risk to national security and suspected their involvement in international terrorism. The provision created a “prison of three walls”: detainees could voluntarily leave for their home or third country. Yet, if they did not or could not due to fears of torture, detention would continue ad nauseum. While passing the law, the government sought to immunise it from challenge by making a derogation order from rights to liberty and security in the European Convention on Human Rights on the grounds of a public emergency.

The primary question for the House of Lords was not whether a general rule had particular effects. The law was all about particularity: its force was trained on a particular group – to which all eleven defendants as foreigners belonged. Instead, the court was confronted with three general questions: Was there a public emergency justifying the derogation? Was the legislation proportional to its aim?⁷¹ Finally, was it discriminatory against foreigners? The majority sided, somewhat reluctantly, with the government on the first question on the grounds that, in determining the existence of a public emergency, the executive possessed greater institutional competence and benefitted from a wide margin of appreciation.⁷² Yet for the rest, the answer was negative. The legislation failed the proportionality test. The use of immigration measures was unlikely to advance the stated security goals: deportees would still be free to plan attacks against the UK, as would nationals.⁷³ It was also discriminatory: no reasonable and objective criteria existed for imposing harsher treatment on non-nationals given the considerable number of British citizens involved in or suspected of international terrorism.⁷⁴

Of relevance are the considerations that were weighed. The particular circumstances of the defendants did play a role: the court took seriously the consequence that innocent foreigners could be held incommunicado ad infinitum.⁷⁵ However general considerations were equally important in the proportionality test, and ultimately decisive. They revealed an inconsistency between the legislative measures and the stated aims. Not only was the law difficult to square with the need for generalised equal treatment, the law, as one commentator put it, made ‘no sense in security terms’.⁷⁶

Secondly, even if we persist with the individualist perspective, it may be empirically limited. The particularity justification may be just that - rather particular to Anglo-American jurisdictions. Not all constitutional and international rights cases arrive in the form of individual complaints. The form of judicial review varies significantly.⁷⁷ Adjudicators in many jurisdictions are granted the power to abstractly review legislation, issue advisory opinions, and entertain complaints by legislators or collectives/organisations. These powers are particularly prevalent in civil law jurisdictions, while liberal standing rules in some common law countries permit public interest complaints. The latter two powers exist in regional quasi-judicial procedures and the final in some southern common law countries, particularly South Asia. To varying degrees, the production of factual evidence and evidence of particular violations is required, but it may not be central to the case. Moreover, some adjudicatory bodies may launch investigations and inquiries.⁷⁸

It might be objected that this criticism is unfair. Not all defenders of judicial review support abstract or collective forms of review. In their view, a lack of particularity may deprive the claim of legally manageable content⁷⁹ or the meaningful context for the interpretation and application of a right.⁸⁰ This is partly true. The individualised structure of much judicial review does carry certain benefits, although perhaps more of a functional than epistemological kind.⁸¹ But collective forms of review offer complementary benefits. As will be argued later, it can ameliorate the critique that rights are too individualised in their focus, capture the collective dimension embedded in most rights, provide broader guidance and legal certainty to the meaning of particular provisions and allow courts to rule on important questions when individual applicants are pressured to abandon or settle their claims.

Waldron goes a step further and labels the particularist virtues of Anglo-American courts a mere ‘myth’.⁸² In appellate review, the traces of ‘the original flesh-and-blood rights-holders’ have “vanished” as the argument becomes more abstract. According to him, complainants are selected by advocacy groups ‘in order to embody the abstract characteristics that the groups want to emphasize as part of a general public policy argument’.⁸³ This critique is pertinent though overstated. In the common law world, the factual record from lower courts is left largely intact (although it can be more easily contested in civil law jurisdiction). Moreover, while public interest advocates do try to identify more sympathetic claimants and narratives, their degree of control over litigation can be marginal.⁸⁴

Yet, Waldron is right in pinpointing the generalised element of judicial review. In 1976, Chayes identified this feature as a particular turn in civil law from that of retrospective and bipolar litigation dominated by individualised remedies to the more forward-looking model of public law characterised by multiple parties, a more predictive and evaluative approach to fact-finding and the presence of structural or general remedies.⁸⁵ The Chayesian paradigm shift is of course highly stylised. It ignores the long tradition of these general features in private law the fact that that most public law cases are modest in ambition or concern individualised administrative remedies.⁸⁶ However, Chayes is most likely correct that general considerations may be more prominent in cases that seek ‘vindication of constitutional or statutory rights’ rather than ‘private rights’.⁸⁷

To sum up, the particularity of judicial review may give the courts a slight epistemic advantage over legislators and executives. Though if adjudication can and should concern broader principles and policies, questions remain over a judicial comparative advantage on this terrain, as we shall see in section 2.3. Two further factors may ground such a claim.

3.2 A semi-public deliberation: Informational exposure, decisional seclusion

Adjudication is a unique institution on account of various structural features which work in opposite directions. In theory, the judiciary is fully exposed to an array of arguments and facts but is secluded from political pressure, permitting it to make non-partisan or principled decisions. An adjudicator is an extroverted perceiver and an introverted decision-maker, making them sensitive to conflicting accounts but independent in judgment.

The first characteristic is a hallmark of deliberative democracy theory. Robust exposure to different views is said to produce better decision-making. Michelmann sets up the question as to comparative epistemic advantage as follows:

(O)ne condition that you think contributes to greatly to reliability is the constant exposure of the interpreter – the moral reader – to the full blast of the sundry opinions on the question of rightness of one or another interpretation, freely and uninhibitedly produced by assorted members of society listening to what the others have to say out of their diverse life histories, current situations, and perceptions of interest and need.⁸⁸

Even though evidence from empirical research on deliberative democracy suggests that these assumptions have limits and defects,⁸⁹ we can accept for the moment that interpretive reliability improves with full argument and informational exposure. The question is whether courts possess any particular comparative advantage, structurally or in terms of the different incentives and costs different actors possess in accessing information.

The second structural characteristic of judicial deliberation is the requirement of impartiality and independence. The political insulation of the courtroom may allow dispassionate, and consequently better, reasoning. Sibley argues that conduct can only be ‘deemed reasonable by someone taking the standpoint of moral judgment’ and this often requires the intervention of a third party:⁹⁰

To be reasonable here is to see the matter – as we commonly put it – from the other persons point of view, to discover how each will be affected by the possible alternative actions; and, moreover, not ‘merely’ to see this (for any merely prudent person would do as much) but also be prepared to be disinterestedly influenced, in reaching a decision, by the estimate of these possible results.⁹¹

These features of informational exposure and decisional seclusion can be captured within a principal-agent model, as done in effect by Kis.⁹² We begin by asking why political authority is delegated first from the people to elected representatives. In Kis’s view, most mature democracies prefer a system of electoral democracy over direct democracy because ordinary citizens face challenges in obtaining relevant information. Condorcet’s jury theorem – that a majority is more likely to get the right result than a minority - does not work at scale. Referenda generate few incentives for citizens to become fully informed since the weight of their respective vote is so small: it approaches zero as the population becomes large. Bentham made precisely this exact point in 1788: ‘the greater the number of voters the less the weight and value of each vote, the less its price in the eyes of the voter, and the less of an incentive he has in assuring that it conforms to the true end and even in casting it all’.⁹³

Thus, the greater the complexity of the issue and the more information needed the ‘more serious the danger that the voters’ judgment will not be just unreliable but subject to some systematic distortion’.⁹⁴ On balance, Kis concludes that representative institutions have an epistemic advantage over citizens. The costs of accessing and processing information is low (due to economies of scale, research staff, and bureaucratic channels) and the incentives are high (better deliberation could produce more votes).⁹⁵ The result is that the main tool citizens possess in a democratic process is one of accountability, namely periodic elections: it ‘permits the citizenry to remain sovereign while paying obedience and commands issued by a representative assembly’.⁹⁶

With the rise of complexity in modernity, this account has a certain resonance. According to Stone Sweet, enhanced complexity generates a demand within dyadic relations (two entities) for more triadic forms of governance, by parliaments, executives, and judiciaries.⁹⁷ It also coheres with certain empirical research. Hibbing and Thiess-Morse note the paradoxical positions of voters in surveys and focus group studies.⁹⁸ While participants expressed a preference for significant control over decision-making, few expressed a preference for strong direct democracy.⁹⁹ They warmed initially to the idea of direct rule but quickly raise feasibility concerns, many of them related to complexity and lack of information. Such research does not rule out a role for direct democracy - it is arguably a powerful means to catalyse and tame rather than manage and direct the political process -¹⁰⁰ but it reveals the limits to its contribution and support.

Returning to Kis, this comparative advantage of legislators over citizens can nonetheless generate its own perverse effects. The asymmetry of information between the elected and the electors invites the more “predatory” politician to adopt ‘mistaken electoral beliefs’ in order to avoid electoral defeat.¹⁰¹ If rights are affected, this is serious ‘because collective self-government depends on each citizen being treated as an equal’.¹⁰² As citizens in most jurisdictions lack direct levers of control over the judiciary, the risks of the judiciary falling prey to this moral hazard is comparably less. Thus, the very feature that makes courts susceptible to charges of democratic illegitimacy may strengthen their epistemological capacity.

It is not too difficult to identify cases in which courts correct predatory information asymmetries. In the seminal prison litigation cases in the USA, it was strikingly revealed.¹⁰³ Until the 1970s, all branches of government in the State of Alabama had declined to address prison conditions; there was clearly no electoral gain in addressing the problem. While courts had acted traditionally in a highly deferential manner, it was the instigation of litigation in 1971 that eventually permitted a full and public description of prison conditions, a system ‘so thoroughly pervaded by violence, overcrowding, inedible food, and brutal methods of punishment’.¹⁰⁴ The force of the evidence was so strong that even the state’s attorney general conceded that they had no case to advance. The entire prison system was ruled unconstitutional.¹⁰⁵ A cursory reading of the European Court of Human Rights judgments on police brutality and prison conditions in Western and Eastern European States makes for equally salutary reading.

Moral hazard, though, is just one feature of informational asymmetry. Kis could have strengthened his case by an analysis of both the resource and time constraints and disincentives to deliberate that representative institutions face. The deliberative democracy school emerged precisely due to the concern that legislatures lacked the willingness or capacity to meet deliberative criteria, such as reciprocity (providing mutually acceptable reasons), publicity (a sufficiently open forum) and accountability (deliberating with and giving reasons to relevant agents).¹⁰⁶ The reasons for this informational asymmetry are practical and structural.

Practically, there is a limit to the time legislators can devote to reflection, deliberation and constituent consultation; and staff and bureaucrats may be unable to generate sufficient or quality analysis on the topic.¹⁰⁷ In most parliaments across the world, there is an enormous volume of legislation and regulation that must be accommodated. Modern parliaments operate therefore under severe time constraints and strictly regulate the time spent on debating legislation.¹⁰⁸

Structurally, legislatures are not designed for public deliberation but rather for the creation and maintenance of legislating and/or governing majorities and, depending on the political conditions, the aggregation of citizen preferences. There is no reason why public deliberation should be the dominant modus of legislative reasoning: negotiations, deals, threats etc. are all part of the political modus. Moreover, ‘If a representative is guided by factional or short-term interest, no amount of deliberation can induce an impartial stance’.¹⁰⁹ Scholars have pointed to deliberative paradox thrown up by legislative processes: if discussions occur publicly, parties may be more likely to harden their bargaining position which defeats the point of genuine deliberation.¹¹⁰

The combination of these constraints and incentives are evident in practice. Reviewing literature on European parliaments, Rasch concludes that:

In general, plenary debates on legislation in assemblies of (at least) parliamentary systems tend not to be deliberative. Arguing seldom affects information and preferences in a way that become important at the final voting stage. Outcomes almost always are known in advance…. Instead, it is much more common for legislators to use plenary debates as an arena for stating their reasons, revealing their preferences and explaining their vote – primarily to outside party members, media and the general public.¹¹¹

This finding contradicts claims by judicial review critics such as Waldron who praise legislatures as paragons of deliberation. Waldron writes, ‘In this regard, it is striking how rich the reasoning is in legislative debates on important issues of rights in countries without judicial review.’¹¹² Yet, his only example is a debate on abortion in the British parliaments from the 1960s. Also, he fails to reconcile his own normative perspective with his own empirical observations: he has bemoaned earlier the lack of debate in the New Zealand parliament, characterising it as an ‘empty chamber’.¹¹³

Overall, this quasi-deliberative claim for courts is attractive. In the context of fully fleshed-out litigation, a court may be exposed to a greater volume of relevant information and informed expertise and address these in a disinterested and thoughtful manner. However, there are four significant caveats.

First, not all cases are complex. Kis acknowledges for instance that complexity may not loom large for the general public on some issues, particularly issues of ‘personal morality’.¹¹⁴ He names capital punishment, gay marriage or affirmative action since ‘people have the opportunity for acquiring shared experience’.¹¹⁵ This fact may explain why both laws and jurisprudence on these issues arouse significant controversy, and both legislators and courts risk charges of elitism. Nonetheless, even here, Kis notes the importance of information. Judicial intervention may help deepen the understanding of the causes of inequality, which may be needed in order to assess clams for affirmative action or gays and lesbian rights, particularly if one has had no interaction with the persons who would benefit from the law.

Secondly, the informational advantage of courts is both conditional and contextual. Legislative and executive members can avail themselves of information and expertise from parliamentary committees, public consultations, in-house legal expertise and closer attention to mass media,¹¹⁶ while the significance of predatory political behaviour for rights may vary considerably. As to the adjudicatory process, it possesses its own in-built constraints: significant time and resources can be consumed by procedural machinations and cases vary considerably in terms of exposing courts to a full range of opinions and facts. Indeed, the strictures of legal method and argument may screen out morally relevant information and arguments (see further discussion in section 2.3). Moreover, while adversarial litigation generally allows weaker parties a chance to more fully put their case, alternative or more fine-grained opinions may not secure a hearing unless there is a good process for amicus curiae briefs, expert opinions and media/public exposure of on-going litigation.

Thirdly, exposure does not equate to comprehension. What is missing in Kis’ account is the differential capacity of both institutions to process complex information. Judiciaries are not experts across all policy domains and face, like any institution, their own challenges with complexity. The trend towards expert decision-making agencies is unmistakable. As will be argued in the next sub-section, the judiciary is adept at expanding their expertise within cases and over time, but one cannot overstate their capacity.

Fourthly, the assumption that courts are independent from politics is contested. Decades of empirical research on judicial behaviour demonstrate that legal reasoning is only one factor in decision-making.¹¹⁷ Indirect political pressures or personal ideology may compromise judicial independence. The extent to which judiciaries are independent will be taken up later, but for now it is important to note the dilemma.

3.3 The mode of decision-making: Justification and method

A final epistemic advantage may arise in the mode of judicial deliberation: the form and substance of reasoning. As to form, judges must State their reasons. For those obliged to provide public and written justification, certain expectations are common: a logical and defensible progression of argument; a demonstration that relevant facts and competing views have been considered; and awareness that the decision has consequences. Conversely, it might be thought that other branches of government are not so formally constrained. Legislation is passed without any formal process that demands coherent and rigorous reason giving.

This argument has merit but, ultimately, its force is modest. Empirically, not all courts are compelled to provide reasons.¹¹⁸ Judges in some civil law jurisdictions (e.g. France) provide scant reasoning: authority is presumed to flow from hierarchy and the proper operation of the legal process. Such a result is consistent with the centralised ordering of the civil law judiciary with rigorous internal systems for maintaining legal consistency across decisions.¹¹⁹ This is to be contrasted with the decentralised and coordinate common law systems. Individual judges possess enormous powers that must be justified publicly to political actors and other members of the judiciary:¹²⁰ reasons emerge as an important form of accountability.

However, one should not overstate the discrepancy between legal systems. In some instances, it a difference in style: e.g., Germanic-influenced courts prefer a highly deductive and impersonal style of form and reasoning. Moreover, there is considerable intra-jurisdictional borrowing in the contemporary design and reform of judicial systems. In particular, the rise of constitutional and international rights review has prompted courts lower in the chain to be more expansive in their statement of reasons. Otherwise they risk being perfunctorily overruled. The European Court of Human Rights has demanded that on certain topics reasons must be given or easily discernible from judicial findings (such as determination of criminal guilt).¹²¹ The Court’s justification of its position is apposite in highlighting the reason-centric focus of legal process:

[F]or the requirements of a fair trial to be satisfied, the accused, and indeed the public, must be able to understand the verdict that has been given; this is a vital safeguard against arbitrariness. As the Court has often noted, the rule of law and the avoidance of arbitrary power are principles underlying the Convention … In the judicial sphere, those principles serve to foster public confidence in an objective and transparent justice system, one of the foundations of a democratic society…¹²²

Turning to the legislative side of the equation, we should ask whether legislatures experience analogous opportunities, pressures, and incentives to state reasons. Further, do they develop reason-giving cultures over time? Structurally, there are grounds for thinking so. Legislation is debated in parliament, often in repeated phases. Justification for positions on at least particularly controversial issues will be demanded by the media, constituents, opposition politicians, and possibly more diplomatically by bureaucrats. The result is that the legislative record can contain substantive reasoning. In four case studies of constitutional questions in legislative debate, Tushnet draws back the curtain to reveal a certain level of reason-giving, such as legislative attention to constitutional rights, parliamentary committees which scrutinised legislation, and the expansion of legal advisors within the executive and legislative branches.¹²³ From this he draws the conclusion that ‘the performance of legislators and executive officials in interpreting the constitution is not, I think, dramatically different from the performance of judges’.¹²⁴

However, the argument in favour of legislatures can be taken only so far. First, as discussed earlier, there are constraints and incentives that operate to restrict the level of deliberation. These are part of the calculus of legislators. Secondly, the statements of reason tend to be thin. Tushnet argues that sometimes there is ‘flesh on the bones’ in legislative debates, but he acknowledges that most of the times the reasoning constitutes ‘skeletons’ of argument.¹²⁵ While he counter-asserts that this is not much different from the verbal discussions between justices,¹²⁶ the fact is that judges are at least required to set out those reasons. Even if they may be a rationalisation of intuitions, according to the legal realist tradition, they must be transparent. They can provide the basis for substantive discussion, occasionally in the media but at least amongst the contesting parties, policymakers, politicians and the legal profession.

These considerations might lead us to count the stating of reasons as a slight form of epistemic advantage for courts. However, we have only considered form. We must also ask a more substantive question: do the methods used by courts produce more reliable interpretations? This is a much more difficult challenge. And, it is at this point, that the twinning of moral and legal epistemological arguments for judicial review splinters, twists, and potentially falls apart.

The legalistic reflex is to defend the method of judicial reasoning by reference to doctrinal expertise. In interpreting constitutions and treaties, judges are more likely to follow an established legal method as the relevant legal system dictates: placing the correct weight on legal text, precedent and practice, principles and theories, case and background facts etc. Yet, this very modus of reasoning may count against the courts. If the metric for reliability is good moral reasoning, recourse to legal method may demonstrate precisely why we do not want courts to be given the task of rights interpretation. As Waldron puts it, judicial review does not provide ‘a way for a society to focus clearly on the real issues at stake when citizens disagree about rights’ but rather ‘distracts them with side-issues about precedent, texts, and interpretation’.¹²⁷ He notes, however, that in contemporary constitutions, the gulf between quality moral reasoning and acceptable legal reasoning may not be as great.¹²⁸

This framing presents a sharp choice: If the case for judicial review rests on quality legal reasoning, it is open to the normative charge that it may be poor moral reasoning. If the case rests on quality moral reasoning, it is open to the empirical critique that the reasons judges give must be more legalistic in nature. It seems one has to choose.

Yet, some critics would not even allow that. They would go even further and claim that the first choice is fallacious: courts are not engaged in anything that can be substantively identified as legal reasoning or method. Courts are principally guided by moral and political convictions rather than any objective or external legal criteria.¹²⁹ Stated legal reasons are but a subterfuge. Thus, one is only left with an argument from the standpoint of moral reasoning and that is likely to be weak if the official legal reasons are covering unstated moral preferences. This critique is overdriven. No one school of thought has empirically explained judicial outcomes as discussed above.¹³⁰ Gibson’s observation from 1983 seems to stand the test of time:

In a nutshell, judges’ decisions are a function of what they prefer to do, tempered by what they think they ought to do, but constrained by what they perceive is feasible to do. Thus, judicial decision making is little different from any other form of decision making. Roughly speaking, attitude theory pertains to what judges prefer to do, role theory to what they think they ought to do, and a host of group-institution theories to what is feasible to do.¹³¹

Nonetheless, the realist or attitudinalist point demonstrates, at least, that one has to be cautious in resorting to legalistic defence of interpretive method.

How can the argument be saved? There are at least three ways forward in trying to re-scramble the egg of law and morality. The first and most common way forward is to limit the scope of the claim to a subset of areas where the two clearly overlap. The legal process school, among others, argues that the articulation of rights and duties should be limited to those areas where the judiciary has institutional competence. In Lon Fuller’s classic treatment of institutional competence, adjudication is positively viewed as contributing a distinct form of decision-making to public policy: ‘a form of social ordering institutionally committed to ‘rational decision’ that is based on ‘participation through proofs and arguments’.¹³² However, judges are likely to make mistakes in adjudicating in areas where they must access information not presented by parties or that trigger complex and polycentric repercussions of a judgment.¹³³ This consequentialist argument fits neatly with some liberal arguments that judicial review thus should be limited to negative liberty rights.¹³⁴ Therefore, we might be more comfortable that courts will be more reliable in these sorts of cases.

However, the argument struggles on various points. Most of the controversial cases discussed in this piece have involved classical liberty rights. Such a demarcation of competence may be arbitrary or a mere historical reconstruction.¹³⁵ In addition, courts have developed various techniques to improve their access to information and expertise in more complex cases, from amicus curaie submissions through to experimental and reflexive doctrinal techniques and remedies.¹³⁶ Such approaches seek to maximise the respective competences and contributions of the courts and other branches of government to solving a problem.

A second and alternative way forward is to embrace the standard of moral reasoning and contend that courts are superior in this respect.¹³⁷ Dworkin does this in four moves. First, he argues that in practice lawyers and judges ‘instinctively treat the Constitution as expressing abstract moral requirements that can only be applied to concrete cases through fresh moral judgments.’¹³⁸ Secondly, while he concedes it would be ‘revolutionary’ for judges to make such a concession,¹³⁹ he argues that this moral approach largely fuses with legal method. This is because basic legal rights are framed in abstract moral terms: treating them as moral rights constitutes the only sensible means of legal interpretation.¹⁴⁰ Thirdly, he contends that moral reasoning on rights must be based on principles rather than policy: there must be “distributional consistency from once case to the next” because rights ‘do not allow for the idea of the strategy that may be better served by unequal distribution of the benefit in question’.¹⁴¹ Finally, courts as a “forum of principle” are better placed for such a task. They are constrained by ‘articulate consistency’:¹⁴² Legislatures and executives are likely to give undue weight to policy considerations due to their orientation towards the advancement of “general welfare”.¹⁴³

Dworkin’s defence of courts as forums of principle is compelling in the abstract but is practically limited. His later model of “law as integrity”, which embeds judges more firmly in their legal context (Anglo-American although potentially exportable to civil law systems), places a constraint on principle-based reasoning. In the “chain gang”, judges are both backwards and forwards-looking: seeking to both fit their chosen principle with previous interpretations and the best available and contemporary moral interpretation.¹⁴⁴ Such coherence may not always be achievable. As the Sherbet saga showed, judges are likely to possess different preferences for weighing the past and future. A further constraint is Dworkin’s acknowledgement that rights can be relative as much as absolute: they are not always full trumps: sometimes, policy considerations will be strong enough to dominate.¹⁴⁵ The necessary weighing of arguments raises the question of judicial expertise, their capacity to determine the weight or force of policy considerations, something Dworkin does not address yet is common in almost all rights cases.

Dworkin’s best case seems to be that courts will generally give more weight to moral principles but that legal precedent and policy considerations may sometimes dominate. Even if we accept this argument, it is not patently clear that it amounts to the most compelling case for judicial review. Dworkin’s approach ends up mirroring the sorts of claims by the legal process school – that judges are expert in one type of approach. In this case, it is principles rather than discrete cases.

However, if courts are called to also assess various policy considerations, we should expect them to be up for this task. This provides a third way forward. What is notable in the religious freedom and counter-terrorism judgments discussed above is the presence of significant consequential reasoning by courts - their close scrutiny of the policy-based claims of governments. However, the overall picture is different from Dworkin’s: courts emerge not so much as forums of principle, with immaculate moral reasoning, but rather as forums of principled pragmatism. It is a mode of reasoning well summarised and promoted by Carter and Burke. They argue that ‘well-reasoned legal decisions’ are those fit best together ‘the facts established at trial, the rules that bear on the case, social background facts, and widely shared values’, rather providing epistemic clarity: ‘Law does not provide a technique for generating “right answers”’.¹⁴⁶

If principled pragmatism were the standard, how would courts perform? Most parliamentarians and ministers also aspire to be principled pragmatists - marrying value or ideological commitments with complex reality. Given that rights tip the balance in favour of principled reasoning, we might expect generally courts to have some epistemic advantage. However, some of the cases discussed above suggest that courts can be superior in consequential reasoning and inferior in principled reasoning.

3.4 The limits of epistemological justifications

Are courts better placed to interpret rights? Of the various arguments traversed above, the strongest arguments for adjudicators are their political seclusion and their bias towards principled forms of reasoning. Their relative freedom from partisanship and the demands of coherence seem particularly to enhance the prospects of reliable interpretation, at least legally and possibly morally. Epistemological reliability might be enhanced further when courts are better exposed to the particularities of alleged rights violations, are able to secure a broad evidential and informational base, and are required to state fully reasons.

Nonetheless, none of these features provides a knockout blow. On balance, courts emerge with some sort of prima facie interpretive distinction, but how significant is it? Moreover, will it endure in the most important or significant rights cases? In many ways, the epistemological and somewhat legalistic justification of rights review suffers much the same problem as the notion of representative and deliberative parliamentarism. Both ideas are heavily stylised and vulnerable in practice - so-called ‘nirvana fallacies’.¹⁴⁷

In addition, an emphasis on the epistemological virtues of judicial review brings potential adverse consequences for rights practice. It encourages the conflation of judicial reasoning with the “real” meaning of rights. Such conflation might, of course, be instrumentally beneficial for compliance: it adds to the authority of judicial reason. Yet, it is also perilous. There is the danger of rigidity: the ideational or legal space for alternative moral and political conceptions of rights (whether more expansive, restrictive, or contemporary) is diminished. Further, it may encourage courts to be risk adverse. Judges may reason that restraint or refuge in doctrines such as justiciability is to be preferred to a charge of incompetence or illegitimacy. A legal and institutional environment open to the prospect of judicial infallibility may encourage both judicial dynamism, sensitivity, and innovation.

4.1 Political legitimation and trust building

The first functional role of judicial review is the most ephemeral. Judicial review provides one answer to a foundational question in political theory and practice: how to justify the coercive power of the State or majoritarian-based government.¹⁵¹ By pointing to judicial review of basic rights, an assurance is offered to individuals who must comply with law or who may suffer its neglect. Authority is legitimated accordingly through the guarantee of certain procedural and substantive protections.

This reciprocal function of judicial review produces two instrumental benefits of normative significance. First, it may constitute a condition precedent for the initial formation of a “demos” or political community. Secondly, it may help build and maintain trust within society over time, particularly between distrustful factions, groups or classes. Social trust is increasingly recognised as crucial to achieving most ends in a modern State.¹⁵²

Curiously, this function of judicial review reverses an assumption that is thought to disqualify it, namely plurality of opinion. The very existence of disagreement between individuals and groups, together with uncertainty over the consequences of majoritarian rule, may precipitate the establishment of judicial review. As Rosenfield puts it: ‘in heterogeneous societies with various competing conceptions of the good, constitutional democracy and adherence to the rule of law may well be indispensable to achieving political cohesion with minimum oppression’.¹⁵³

The mechanism by which this occurs has been described as one of “insurance”.¹⁵⁴ Empirically, we expect judicial review to emerge when individuals, groups and political parties feel compelled to protect themselves against future and unacceptable political risks. The result is a constitutional settlement that requires commitment to mechanisms in which all actors accommodate an acceptable, but not excessive, degree of risk. Such institutions transform ‘fuzzy uncertainty (where anything is possible)’ into a ‘specific assessable risk (of betrayal) that a trustor is prepared to accept’.¹⁵⁵

These risks may be apparent and visible at the time of constitution-making. For example, a losing side in an armed conflict, a small political party, or a linguistic minority, may all feel particularly vulnerable. In addition, judicial review of basic rights can provide an on-going means by which a State can legitimate its authority and sustain trust with different groups. Constitutional adoption does not imply constitutional acceptance: Violence, capital flight, migration, patronage and non-compliance with law (criminal, tax, labour, social security law conditions etc.) represent alternatives to submission to authority. Judicial review provides one means by which dissent is redirected back into political processes by helping ensure that the burdens of compliance are tolerable.

This function is not limited to the domestic level. International review represents a form of insurance. Notably, ratification of international human rights treaties and complaint protocols surges in periods of domestic constitutional reform. Morasvic argues that such actions exploit the ‘symbolic legitimacy of foreign pressure and international institutions to unleash domestic moral opprobrium’.¹⁵⁶ As a result, the ‘domestic balance shifts in favour of protection of human rights’ as a State ‘seeks to avoid undermining its reputation and legitimacy at home or abroad’.¹⁵⁷ Further, we might expect judicial review to expand at the international level in order to legitimate the growing coercive power of new global actors. For example, the adoption of the EU Fundamental Rights Charter was justified partly on the grounds that Europeans citizens should be protected from adverse decisions by the European Commission.¹⁵⁸

The political legitimation claim faces, however, four challenges. Each dampens the strength and breadth of the argument. First, there may be alternatives to constituting and sustaining trust. Judicial review is not the only means to enhance the sociological legitimacy of State authority. Other options may be available and function more efficiently or effectively: e.g., federalism, regional representation, representation quotas, reserved senate seats, affirmative action, advisory councils with veto powers (e.g. for indigenous groups), standing, or ad hoc commissions to investigate human rights abuses. Constitution-making should not be blind to these options.

However, the benefit of these alternatives is contingent. When a threat is less distinct and stable, and/or a group is geographically and politically dispersed, these approaches may be less relevant or feasible. There are likewise risks that specialist institutions will ossify over time or face political attack and de-funding. In these circumstances, judicial review may be a more practical and durable solution. It is diachronically flexible and more deeply institutionalised as a third branch of government. The judicial model may also be cheaper, in terms of financial costs as well as “agency costs” or “sovereignty costs”. The legal training of judges and their relative insulation from politics with a capital P may mean they evince sufficient fidelity to the legal text and the constitutional bargain.¹⁵⁹

Secondly, the shape of the insurance mechanism may be warped. Some scholars claim that the protection of minority elite interests is the dominant reason for the creation of judicial review.¹⁶⁰ Hirschl contends that the ‘constitionalization of rights and the establishment of judicial review’ are ‘driven in many cases by attempts to maintain the social and political status quo and to block attempts to seriously challenge it through democratic politics’.¹⁶¹ By way of example, the leading Framers of the US Constitution were anxious to institutionalise multiple forms of control, including judicial review, to check parliamentary majorities and ‘popular irrational passions’ that would threaten property rights and encourage irresponsible fiscal policy.¹⁶²

These elitist claims are arguably overstated. Normatively, it is problematic to simply associate civil and political rights with elite privilege. For instance, there may be good reasons to enshrine the right to property, the most recognised right in the world,¹⁶³ even if its application has not always been progressive. The right to property has been a condition precedent to the establishment or maintenance of mass and social democracy (locking in economic elites to the bargain) and an important protection for marginalised and indigenous groups, who sometimes face the gravest property rights violations.¹⁶⁴

Empirically, Hirschl’s assertion that elite interests dominate contemporary constitution making and treaty ratification is hard to square with all of the evidence. Constitutional settlement has often hinged on identity-based questions – e.g., linguistic, religious, or political rights for minorities – as well as socio-economic risks for particular groups.¹⁶⁵ In Latin America, and elsewhere, the historical social policy failures of democracies and political clientilism generated stronger demands for social rights as a means of re-orienting politics. While in Sweden (1974) and South Africa (1994), socio-economic rights were introduced as a quid pro quo for recognising economic liberties.¹⁶⁶

Thirdly, to what extent will the insurance mechanism materialise in practice? Ginsburg deduces that is subject to the balance of power. The extent to which dominant parties will accept judicial review is dependent on the extent to which they must evince commitment in order to secure authority.¹⁶⁷ If power is relatively diffused at the time of constitution-making, the insurance model will emerge in a strong form as all parties face future risks. If power is more concentrated, its emergence may be dependent on formal constraints: e.g., the securing of a supermajority to approve constitutional reform. In other words, ‘where constitutions are designed in conditions of political deadlock or diffused parties, we should expect strong, accessible judicial review’.¹⁶⁸

Morascvic makes an analogous though transposed argument at the international level.¹⁶⁹ He argues that the emergence of strong judicial review is dependent on the distribution of new democracies, strong democracies, and dictatorships. The former will be strong supporters, seeking ‘to employ international commitments to consolidate democracy – “locking in” the domestic political status quo against their nondemocratic opponents’.¹⁷⁰ However, in strong democracies, the benefits of such insurance weigh less with the result that the “sovereignty costs” of international review tip the balance towards an opposition.¹⁷¹ Strong democracies will be joined by more authoritarian States which seek to avoid significant challenge to the existing domestic order.

These rational choice theories provide an empirical check on the extent to which judicial review will emerge as an insurance mechanism, formally or substantively. However, there may be other causal pathways even when power is not spatially or diachronically diffuse. For instance, even if a dominant party is free from constitutional deadlock, it may face diffuseness and resistance elsewhere in society. Judicial review may be a means of establishing deeper consensus with all groups in society: a regime has to be legitimated in practice. At the international level, realist and constructive theories reveal how States may be respectively compelled or convinced of the need to establish judicial review; ranging from external foreign pressure or incentives to join trade regimes or the acculturation and ideational effects of the spread of rights and the rule of law.

Finally, there is the risk of insincere commitment, particularly if the motivation is elicited by once-off deadlock requirements or incentives or short-legitimacy gains. Authoritarian States may expand judicial power, adopt bills of rights or ratify international treaties simply to distract attention from its repressive apparatus.¹⁷² The State cloaks itself in the judicial robes of legitimacy.¹⁷³ Even in strong democracies, this risk of a “false positive” exists: international judicial review may be supported simply ‘as an export-trade’.¹⁷⁴ Many democratic governments supported the ECHR on the basis that it was ‘merely a Europeanization of their own national practices’ rather than the creation of a new domestic constraint.¹⁷⁵

However, even if judicial review helps shore up a despotic order, legitimises external relations, or permits national branding, these commitments may be nonetheless significant.¹⁷⁶ They represent a Faustian bargain for any governing regime: citizens may feel empowered to turn to courts, which may be responsive in turn.¹⁷⁷ The European Court of Human Rights is a striking example. For instance, the first cases challenging the criminalisation of sodomy were filed only two years after the convention came into force in 1953 – a claim that most States certainly did not expect at the time.¹⁷⁸ Even in authoritarian States can this Faustian bargain transpire and impact social trust. Some independent surveys indicate rises in social trust after judicial reforms in China, even if the State’s motivations were partly suspect.¹⁷⁹

4.2 Accountability for commitments

The previous sub-section argued that judicial review can provide one important means of legitimating state power and establishing social trust. We now turn to the most common functional justification for judicial review, which is accountability. It is related to the idea of insurance and social trust, as discussed above, but is different in purpose and nature. It is neither concerned with inducing participation in a polity nor fostering social cohesion and trust between particular groups within society. Rather, the objective is to ensure that commitments to rights are honoured, which is to be achieved regardless of which groups originally promoted their recognition. As Raz writes, a ‘natural way to proceed is to assume that the enforcement of fundamental rights should be entrusted to whichever political decision-procedure is, in the circumstances of the time and place, most likely to enforce them well, with the fewest adverse side effects’.¹⁸⁰

The classical lines of the accountability argument are well-rehearsed in a recent article by Fallon: courts must possess the opportunity to invalidate legislation, not because they of their interpretive prowess, but because the existence of legislation, rather than its absence, is likely to be more harmful to rights.¹⁸¹ The function of judicial review is simply to provide an additional veto check against potential harms.¹⁸² However, Fallon’s reasoning is partly flawed. The straight line that he draws between his premise (best protection of rights possible) and his method (invalidation of legislation) comes easily undone. What if legislation is seeking to positively protect rights? Such rights might include personal security, fair trial, political participation, socio-economic rights, and even many property rights. In such instances, judicial review risks being irrelevant or even counterproductive if it leads to rights-protecting legislation being struck down on the basis that it conflicts with negative rights.

Waldron – to whom Fallon is responding – warns precisely of the consequences of such lopsided rights protections if positive rights are excluded from judicial review. He states:

The text of a Bill of Rights may distort judicial reasoning not only by what it includes but also by what it omits. Suppose the members of a given society disagree about whether the Bill of Rights should have included positive (socioeconomic) as well as negative (liberty) rights. … Judges may give more weight to property rights or to freedom of contract, say, than they would if property and freedom of contract were posited alongside explicit welfare rights. And giving them greater weight may lead judges to strike down statutes that ought not to be struck down—statutes that are trying to make up the deficiency and implement by legislation those rights that failed to register in the formulations of the Bill of Rights.¹⁸³

Fallon is aware of the problem and introduces a number of caveats. He assumes that such legislation would not come into conflict with ‘fundamental rights” (assuming economic rights, such as property and the like, are not included within a constitution) and that courts should refrain adjudicating cases which raise conflicts of rights. Yet these concessions allow Tushnet, largely rightly but somewhat excessively so, to argue that Fallon’s case for judicial review becomes highly restricted (and largely indistinguishable from Waldron’s limited acceptance of judicial review). This is because there are regular conflicts between rights and the exclusion of economic liberties is pragmatically difficult.¹⁸⁴

What Tushnet does not do is articulate how Fallon’s model might have been reformulated to avoid or mitigate its internal inconsistencies.¹⁸⁵ The way to proceed is not to confuse or conflate accountability with a particular political theory, which in Fallon’s case is a Lockean stream of political liberalism that stresses negative liberty – ‘the absence of coercion by others’.¹⁸⁶ Judicial accountability as an idea resonates in broad a range of political theories and traditions. It even appears in some versions of republican liberty. Eliminating domination requires positive action since inferior status makes freedom tenuous and contingent. Judicial review could help disperse power and build such positive freedom. As Pettit puts it:¹⁸⁷

[B]eing included, having an audible voice, does not reduce to being satisfactorily represented. What is even more important, especially with the administration and judiciary, is that there is room for you and those of the relevant kind to protest to the representative bodies in question, in the event of your believing that things have not been properly done. You must be able to complain and appeal; you must be able to state a grievance and demand satisfaction.

Likewise, accountability can be justified by liberal contractarian theories. Behind a veil of ignorance, individuals are likely to choose a set of institutional arrangements that realise basic and material interests. As Føllesdal states:

Liberal Contractualism grants that democratic, majority rule among elected and accountable representatives may be one important mechanism to ensure the protection and furtherance of the best interests of citizens. But other arrangements may also be required, such as super-majoritarian features, constraints and checks on parliament and government of various kinds. There is no prima facie normative preference for unrestrained parliaments.¹⁸⁸

This overlapping consensus of perspectives suggests that it is useful to begin with a general concept of accountability. The notion has become, admittedly, ubiquitous. There is market accountability, criminal accountability, shareholder accountability etc. However, political accountability contains, at least, the compelling idea that intuitional or powerful actors must take responsibility for their commitments. It requires that they explain and justify their actions (answerability) and be subject to sanctions if their conduct or explanation for it is found wanting (enforceability).¹⁸⁹

This definition helps in broadening our understanding of the role of courts. On the one hand, it is concerned with the material enforcement of particular rights for certain individuals. The South African Constitutional Court clarifies its role in this regard in striking down the death penalty: ‘The very reason for establishing the new legal order, and for vesting the power of judicial review of all legislation in the courts, was to protect the rights of minorities and others who cannot protect their rights adequately through the democratic process.’¹⁹⁰

On the other hand, it is equally concerned with scrutinising decision-making or the lack thereof: in other words, answerability. This answerability function can have broader systemic effects: ‘courts serve as information generating function that facilitates the accountability of the various parts of the State (or even private providers) to each other, using formal rights and their judicial gloss as yardsticks’.¹⁹¹ The earlier discussion of religious freedom jurisprudence is illustrative. The State was forced to confront and answer the question as to whether seemingly neutral eligibility rules for social security and drug prohibition caused unjustifiable harm to conscientious religious minorities. Courts provide a mechanism that requires answerability, State reconsideration, and response.¹⁹² One can certainly quibble with the outcome in any of the US Supreme Court judgments, but all judgments served this function.¹⁹³

These twin features of accountability can work likewise for more positive rights or obligations. Take the seminal case of Airey v Ireland before the ECtHR.¹⁹⁴ The complexity of Irish divorce laws meant the dissolution of marriage was unattainable for those spouses unable to afford and retain a lawyer. The Court refrained from challenging the stringency and complexity of the divorce law but forced the State to be accountable for its inequitable consequences: in practice, not all citizens possessed a right to a fair trial and respect for family life. The result was the subsequent creation by the Irish government of a system of civil legal aid for family matters.¹⁹⁵

This accountability function of judicial review also helps clarify certain intriguing features of rights litigation. A large number of cases do not concern the invalidation or absence of legislation. Rather, they involve claims for implementation of existing legislation. In this modus, the distinction between judicial and administrative review becomes blurred. In jurisdictions as diverse as India, Colombia, and the ECtHR, the bulk of cases concerns rights-related statutory commitments. Why do litigants turn to judicial review, via constitutional and treaty rights, in such cases? The reason seems to be that it empowers them to leverage a more effective response from governments, which may be impervious to weak or highly individualised administrative remedies.

Two challenges stand out for the accountability claim. The first is relevance. Why should a State with a relatively strong record on rights supervision, which has been achieved with minimal judicial engagement, be tasked with the institution? Many rights-rich democracies might fear that judicial review will unleash a flood of frivolous, vexatious, or excessively idealistic claims rather than play any significant accountability role.¹⁹⁶ Take the hard case: the Nordic countries. They sit atop most global indexes on civil, social, and political rights.¹⁹⁷ Thus, why should Sweden, a State without a tradition of domestic judicial review, feel compelled to embrace the practice or ratify international mechanisms? Or why should Norway, whose tradition of judicial review has been limited to a small number of civil rights and a doctrine of deference in property rights cases, be expected to expand its remit of review?¹⁹⁸ As Hirschl argues:

Variations on a combination of well-established, ex-ante parliamentary preview and restrained ex post judicial review, deployed in the Nordic region, have proven effective in mitigating the counter-majoritarian difficulty embedded in excessive judicial review and ensuring an alternative, nonjuristocratic way of going about protecting rights.¹⁹⁹

Yet, even in these States, it is possible to mount a claim for judicial accountability. The first is the danger of atrophy and backsliding. Many of these States face new challenges in upholding rights protections due to increasing social diversification (especially through immigration) and worsening economic conditions, with the current exception in Norway. It is not clear that existing mechanisms of protection are always sufficient. The second is that the picture of Nordic success on civil and social rights must be nuanced in light of various and significant rights violations, particularly with regard to the treatment of Roma,²⁰⁰ Sami,²⁰¹ institutionalised children,²⁰² prisoners,²⁰³ and suspected communists.²⁰⁴ The Nordic countries do not possess a clean slate before the European Court of Human Rights and many judgments reveal violations that may not be systemic and grave as elsewhere but certainly worthy of condemnation.²⁰⁵ For these States, judicial review permits civil society to help “perfect” the realisation of rights and democracy - ²⁰⁶ an opportunity used increasingly by mainstream Nordic civil society organisations, social movements, and specialist legal support structures.²⁰⁷

Beyond this materialist defence, the adoption of domestic and international review may strengthen accountability efforts elsewhere in the world. By expressing a symbolic commitment to judicial review, other States are encouraged or pressured to submit to similar institutions and comply with their judgments.²⁰⁸ Many rights-confident democracies express their decision to internationally self-bind in largely these terms.²⁰⁹ Now, this symbolic defence is “other-regarding”, which may make it prone to questions over its relevance (there are multiple States) and instrumental justification (States seem to gain diffuse benefits but carry specific “sovereignty” costs). However, it is important to note that rights promotion may also be justified on instrumental grounds:²¹⁰ Promotion of political rights may reduce armed conflicts and military and humanitarian expenditures;²¹¹ civil rights may improve the conditions for foreign investment;²¹² and social rights may level the playing field for international trade.²¹³ It can also strengthen the call by these States in bilateral relations and development cooperation for rule of law reforms while also ensuring that they do not constitute a precedent for authoritarian or democratically fragile States.²¹⁴

The second principal challenge to the accountability argument is effectiveness. Are courts actually able to deliver? Will judges be responsive to rights claims and will compliance and broader impact follow their decisions? Or more particularly, will courts be at least effective as realistic alternatives to judicial review in ensuring accountability in practice?

Various critical theories claim that courts will be biased towards elite and/or the middle class. A distributive bias may be structural in origin. Advantaged groups are better able to secure access to legal representation, strategically maximise the benefits of repeat litigation, and accommodate the institutional passivity and duration of court-based procedure.²¹⁵ Or it may be attitudinal: courts are swayed by their personal ideologies that are likely to be elitist in orientation.²¹⁶ They seek to maintain the social and political status quo²¹⁷ or are swayed by the hegemonic discourse of neo-liberalism. ²¹⁸ However, the evidence is mixed and complex on distributive bias. Courts also regularly side with marginalised groups in many states while research indicates that the benefits of decisions can be spread over multiple groups.²¹⁹ Judicial fairness or responsiveness is a function of multiple factors, not least the nature of reviewable rights, the appointment processes for the judiciary, forms of access to justice, and the inflection of background public policy and political economy conditions.²²⁰ Thus the institutional design and trajectory of judicial review and its trajectory is as important as its existence.

The same applies to the general effects of courts. Realist claims that courts, without the power of purse or sword, are unable to catalyse material impacts need to be examined carefully.²²¹ A court’s coercive and authoritative capacities can be mobilised directly (through the crafting of appropriate remedies) or indirectly (through social mobilisation or bureaucratic/elite alliances) to advance the realisation of rights and control/direct state power. A growing array of comparative studies reveal the ability of courts to conditionally leverage change, from the reduction of hunger and homelessness through to the advancement of prisoner rights and rights to sexual orientation and gender identity.²²² Even at the international level institutionalist studies that foreground the role of political power and judicial legitimacy have found that enforceability is contingent on the ability of actors to leverage international judgments in domestic politics.²²³ While this empirical research should certainly dampen the enthusiasm of some scholars and advocates for the accountability function of courts, it evinces at least that such an institutional project can be much more than a forlorn hope.

4.3 Structured public deliberation

The final argument for judicial review is that courts can help structure public deliberation on rights through its role in the “constitutional order”. In explaining this functional motivation, I will first dismiss two related claims: democracy and rule of law.