In defense of deference: International human rights as standards of review

Many wellfunctioning democracies defer to regional human rights courts about how they treat their inhabitants. In turn, these international judges sometimes defer to the domestic courts' assessment. What are we to make of this reciprocal deference? The relationship seems fraught with problems: Unaccountable and distant legal experts sometimes— but not always— set aside democratic decisions about complex societal challenges, for the sake of vague legal human rights. Indeed, it is the judges themselves who interpret and develop these rights, in practice taking on legislative tasks. Somewhat surprisingly, the deference of such international courts (ICs) to states fuels further concerns about ‘juristocracy’: the practices of deference are said to be so opaque that strong states do as they will, while the weak human rights courts adjudicate as they must. Does the mutual deference of democratic states and human rights courts collide with both the Scylla of permitting states to violate human rights and the Charybdis of juristocracy— or can the practices of deference avoid both? Considerations of comparative epistemic expertise of domestic bodies in democracies and international judges may help resolve some of these concerns, or so I argue in the case of the European Court of Human Rights (ECtHR) which reviews states' compliance with the European Convention on Human Rights (ECHR). The Court is criticized when it rules against states— e.g., concerning the UK's blanket ban on prisoners' right to vote (Hirst V. The United Kingdom (No. 2), 2005). And the Court is criticized when it defers to states, such as when it accepted France's and Belgium's bans on veils justified as necessary to uphold conditions for ‘living together’ (Belcacemi and Oussar V Belgium, 2017; S.A.S. V France, 2014). Whence ECtHR's authority over

fairly well-functioning democracies? And is its deference to states consistent with its mandate to protect human rights?
At least three of its practices express deference. The Court sometimes defers to states by granting them a margin of appreciation with regard to some Convention rights, and it sometimes heeds a European consensus among the states. Furthermore, the judges apply the Convention's limited focus on civil and political rights, arguably to the detriment of social, economic, and cultural rights -which the domestic authorities must also respect.
What are we to think of such practices? After all, states create and design ICs knowing that they themselves risk in due course to be subject to the ICs' jurisdiction. No wonder that they prefer a very subsidiary role for any such review body (Besson, 2016). Do the practices simply illustrate the design result of states seeking to protect their sovereignty while agreeing to be reviewed by international judges-who in turn is eager to please its creators lest they abandon the IC?
The following account offers a partial defense of these practices that carve out some protected pockets of state sovereignty concerning human rights from the scrutiny of a regional human rights court. The defense is based on comparative epistemic expertise.
The legitimacy of the ECtHR crucially depends on whether it fulfills its task to help states protect individuals better against various kinds of domination in the form of human rights violations. The ECtHR is often-but not always-in a better position than the domestic authorities to discern the relevant facts and legal norms appropriate to adjudicate the conflict. The set of human rights of the ECHR and the complex deference practice of the ECtHR can to some extent be justified as reflecting three considerations: (a) when the ECtHR judges and domestic judges, respectively, are more likely to have the specific and limited epistemic expertise needed to curb domination by domestic majorities (b) the need for mechanisms to reduce the risk of domination by judges; and (c) appropriate respect for the intrinsic and epistemic values of democratic decision making.
These three concerns affect the somewhat limited list of human rights and the deference practices of the ECtHR, based on arguments about comparative epistemic advantage. These arguments support some limits on the domain of the review the Court should perform, which matches the present practices on several points. This account is not a mere apologia for the existing practice: this 'rational reconstruction' of when deference may be justified provides standards for critique. The present practice of the ECtHR must be specified much more closely. It should also improve on the rules to nominate and elect judges and members of the Registry of the Court, and must improve the doctrine of the margin of appreciation. Finally, the roles of sightings of an 'emerging European consensus' by the Court merit much closer critical attention.
Section 1 provides a sketch of the ECtHR and of the three aspects of concern. The ECtHR is familiar to several readers; the account here seeks to bring out some of the issues that are salient for these arguments concerning comparative epistemic competence. Section 2 presents the main argument for judicial review by the ECtHR with some practices of deference. It starts with a sketch of a democratic theory, to provide a justification of such judicial review partly to protect democratic values and procedures-with a doctrine of a 'margin of appreciation.' This account focusses on the epistemic advantages of democratic decision-making, but also appreciates other intrinsic reasons to value democratic self-governance. Section 3 uses this account to identify the nature of independence and expertise required by the ECtHR, and of the domestic democratic authorities, for review by such proclaimed authorities to be not only possible, but helpful and justifiable. A broad range of scope conditions and requirements seems important, including judges with a range of backgrounds. Section 4 applies these arguments to identify some areas for improvement concerning the present processes to select experts to the ECtHR, and regarding the current doctrines of a margin of appreciation and European consensus.

| ON THE EUROPEAN COURT OF HUMAN RIGHTS
The main role of the ECtHR is to review domestic legislation and policies against the standards of the ECHR. The express role of the ECtHR is to assist states "ensure the observance of the engagements undertaken by the High Contracting Parties" (Art 19 ECHR). The ECtHR is thus not authorized to promote and protect all human rights by all means. Rather, the task of the ECtHR is 'subsidiary' or supportive and supplementary vis-a-vis the states, to supplement and strengthen the protection offered by domestic bodies including the legislature and the judiciary. The state remains the primary responsible actor to respect human rights.
The legal rights listed in the Convention and developed by the ECtHR are thus standards the member states of the Council of Europe use to express concern for how the other state authorities treat their inhabitants. I submit that the ECHR norms should therefore be understood and assessed as specifically constructed for such policy evaluations, by certain expert judges. This 'political' account of the ECHR rights (Maliks & Karlsson, 2017) does of course not deny that individuals have other interests and legal or moral human rights that may be more important for them, and states may have other obligations toward them than such rights that trigger concern (Follesdal, 2017b).
In order to apply the ECHR, the ECtHR must often engage in interpretation of the Convention rights-possibly trespassing onto the legislative function, at the risk of juristocracy and expert overreach. This interpretive task is even more of a concern since the ECHR leaves several rights very vague-such as the right to family life. In favor of the ECtHR's interpretive activity, the states have arguably authorized the Court to engage in such interpretations, since the Preamble of the ECHR makes clear that the states commit to the Convention and its Court to help with the "maintenance and further realization of human rights and fundamental freedoms" The Court has often insisted that this requires the Court to interpret the Convention in ways that make it effective, and relevant for new societal risks (Stafford V. United Kingdom, 2002).
We should also keep in mind that the review is limited in several ways. The ECHR does not aim to 'harmonize' domestic legislation e.g., as does the Court of Justice of the EU. Instead it allows some heterogeneity of policies and legislation. The ECHR underdetermines such institutions and policies. Its task is rather to 'prune' these in accordance with the ECHR, and review the 'balancing' a state performs when it claims the need to restrict some Convention rights. The ECtHR furthermore only determines incompatibility with the ECHR. It does not invalidate or replace the law/policy but returns this to the domestic authorities for a democratic process of reform. Indeed, it is not for the Court to supervise whether and how states execute the judgments: this is the task of the Committee of Ministers (ECHR Art 46). It usually does not specify remedies for the violation except when it identifies structural problems in States Parties that cause several repeated violations.
A striking part of their practice is that the judges sometimes explicitly defer in various ways to their domestic counterparts. Consider how three features limit the Court's intrusion into democratic states' sphere of sovereign decisions. They are long standing parts of the ECtHR's jurisprudence, as evidenced by the early cases. The ECtHR sometimes grants states some discretion in the form of a 'margin of appreciation.' The justification appears to combine deference for democratic decision making and epistemic reasons: by reason of their direct and continuous contact with the vital forces of their countries, the State authorities are, in principle, in a better position than the international judge to give an opinion, not only on the "exact content of the requirements of morals" in their country, but also on the necessity of a restriction intended to meet them. (A, B andC V Ireland, 2010, para 232, cf Handyside V United Kingdom, 1976, para 48) The Court claims that a margin of appreciation is appropriate for at least three main issue areas.
• 'Balancing' the rights against other urgent issues such as emergencies, public safety, the economic well-being of the country etc-as permitted for several rights to private life, religion, expression etc (Art. 8, 9, 10). • 'Balancing' or 'trade-offs' among different private human rights in the Convention-such as between freedom of expression (Art. 10) and privacy (Art. 8). • How to apply the norms to the specific circumstances of a state, which may depend on shared values and traditions or perceived threats.
Note that the Court hardly grants any margin of appreciation when certain rights are at risk under certain emergencies, such as rights against torture or slavery. And the Court usually only grants such a margin if the domestic judiciary has performed a 'proportionality test. ' Secondly, this margin is narrower-and the review more stringent-if the judges observe an (emerging) 'European consensus' among the member states concerning particular rights which runs contrary to the accused state's policy. The Court thus seems to guide its own assessment of compliance with the standards by the policies of the other states.
The third constraint on the review by the ECtHR is not self-imposed, but a feature of the ECHR. It focusses on civil and political rights that constrain state action, rather than social and economic rights which are sometimes said to require more actions by states. While the doctrines of a margin of appreciation and of a European consensus arguably serves to guide and constrain the ECtHR, this bias in set of human rights in the Convention may exacerbate harms individuals may suffer by being subject to the judges' discretion-increasing the concerns that the ECtHR is an example of unjustified juristocracy.
In short, judicial review of certain human rights skews both the democratic processes of decision-making and the policy outcomes. Critics may worry that the bias in the substantive rights the Convention protects has regrettable consequences when it comes to the impact of judicial review on domestic, democratic processes and outcomes.
The following seeks to lay some such concerns to rest.

OF APPRECIATION' IN STRENGTHENING DELIBERATIVE DEMOCRATIC DECISION-MAKING
To justify judicial review of human rights approximately of the sort the ECtHR provides, first recall some reasons we have to value democratic decision-making. Arguably, the ECtHR may contribute to make such democratic decisions even more reliable and inclusive.

| Benefits of democratic procedures
I venture that mechanisms of democratic politics can serve at least four valuable epistemic roles in discovering and deciding policy outcomes on the basis of assessments of their effectiveness, feasibility, and justice (Follesdal, 2017a). As Landemore argues, democratic decision-procedures, characterized as inclusive deliberation followed by majority rule, are generally more able than oligarchies to tap the distributed collective intelligence of a given people. (Landemore, 2017, p. 288) Note that such an 'epistemic turn' in deliberative democracy may well acknowledge further reasons to value democratic decision-making, in particular as procedures with intrinsic value that express individuals' autonomy and political equality.
Democratic deliberation contributes to discover, specify, and assess policies and pieces of legislation on the basis of their feasibility and expected effects (Anderson, 2006;Elster, 1998, p. 7, Przeworski, 1998. Democratic deliberation may also help individuals discover and even modify their ultimate values, through the exchange of opinions and arguments. Inclusive deliberation allows many affected parties to voice their concerns about alternative policies, and this in turn may trigger and enhance other citizens' sense of justice or fairness. (Goodin, 2004;Mill, 1861Mill, [1972, p. 325). Democratic deliberation may also foster normative assessment of policies, and may lead individuals to modify or prune their self-interested plans out of a sense of justice and consideration for the interests of others, because of the process that expresses collective autonomous choice.
But such positive possibilities are by no means guaranteed: the deliberations may not always affect preference shifts in a more just or fair direction. The majority may still be insufficiently respectful or aware of the interests of others,-and the deliberations may bolster negative views about various minorities as a result of group think, xenophobia or the like (Przeworski, 1998).
Note that these benefits of democratic rule are not due to the institutions' ability to accurately "mirror" (Cohen, 1997, p. 79) or be otherwise similar to some 'ideal deliberative procedure' such as an "ideal speech situation" of outstanding philosophy seminars (Estlund, 2009). The epistemic benefits are instead due to the constrained competition among parties and the role of the opposition to government. These actors have incentives to scrutinize, criticize and offer arguably better alternatives to the policies of accountable officials. The central contribution of democratic rule is thus not to remove competition and 'politicking' in favor of consensus, but rather to foster better, genuine constrained competition on the basis of such deliberation (Shapiro, 2003, p. 7). Now, consider how judicial human rights review can strengthen these mechanisms, and hence both the epistemic contributions and intrinsic value of democratic rule.

| A defense of international judicial human rights review
The ECtHR can arguably sometimes help ensure that domestic decisions are well informed and duly carried out, by checking their compliance with the ECHR. We will return to conditions for when it can do so below. Note that to ascertain compliance with the ECHR requires local and counterfactual knowledge. Both domestic authorities and any regional review body must have information about avoidable abuse or neglect that is likely to occur due to the laws and policies of their government. They must be familiar with the culture and circumstances, as well as the risks individuals face due to complex interplay between majority culture and institutions-and about feasible alternative policies that may avoid such violations. The need for such local knowledge is one reason why the chamber of the ECtHR that hears a case always includes the judge with respect to that particular country.
Human rights courts can thus give assurance to citizens and other state governments that a government is committed to human rights,-and that majority rule among them thus is not overly risky. The ECtHR may thus help states show that they are normatively legitimate (Buchanan & Keohane, 2006, p. 408). The ECtHR may serve similar legitimating roles for the EU member states. It may help protect citizens against other EU member state governments, who now share decision-making authority over them, often with majoritarian mechanisms. It is then especially important that citizens can trust that all member state authorities exercise such powers responsibly. No political party should enjoy domestic political power that may lead them to favor EU policies that violate human rights. Such concerns are arguably even more salient insofar as EU authorities undermine the democratic bases of legitimation in the member states (Follesdal & Hix, 2006).

| A defense of the doctrine of a margin of appreciation
So if the ECtHR may sometimes help secure well-functioning democratic processes, when might this occur-and when not? I submit that the margin of appreciation is an example of when the ECtHR cannot be expected to add much value to the domestic process.
When the ECtHR grants a state a margin, it essentially defers to the domestic court's adjudication of the ECHR-of the very same member state accused of a violation. This is appropriate in those circumstances where the state organs should retain the final authority to determine compliance when the ECtHR cannot or is unlikely to provide extra protection. That is: a margin of appreciation should apply insofar and for those objectives, and under those conditions, where the domestic courts and other authorities are at least as well suited as the ECtHR to determine whether there is a breach of the Convention. For instance, there should be a very low risk that the domestic court will skew its judgment unduly in favor of the state in its dispute with its citizens.
Consider now the actual practice-or at least the claimed practice, where the Court claims that it grants a margin of appreciation for some issues of balancing the rights against other urgent issues (Art. 8, 9, 10); or for 'trade-offs' among different private human rights; or for some issues regarding how to apply the norms to the specific circumstances of a state.
What arguments may be offered for this doctrine? On these main issue areas the ECtHR seems correct to hold that domestic authorities are often better placed than itself to judge, due to local knowledge. The Court often claims that domestic authorities are in principle better placed than an international court to evaluate such local needs and conditions (Lindheim & Others V. Norway, 2012, p. 165). So the Court seems to give the domestic judiciaries the benefit of any doubt.
However, it should not abdicate its responsibility completely, but instead seek to assess the risk of human rights abuses in a more nuanced way. The ability of local authorities to strike the balance right is not enough. We should ask under which circumstances are local authorities likely to make decisions in ways that respect human rights appropriately? When, in short, will domestically enacted domestic laws and policies be sufficiently responsive to the best interests of all citizens-thus expressing the intrinsic value of their autonomy? And when will the domestic authorities have mechanisms of self-correction in this regard? Recalling the reasons to value democratic decision-making, I submit that this is more likely under conditions of wellfunctioning democratic rule under the rule of law. Such democratic polities are likely to be more responsive to human rights and self-correcting than alternative modes of governance. Under functioning democratic mechanisms and the rule of law the population deliberates about alternative policies and legislative proposals in light of their implications for all affected parties, so as to promote broadly shared interests whilst avoiding harm to anyone; and an independent judiciary protects the human rights of the inhabitants.
Under such circumstances, the ECtHR is unlikely to reliably provide a better assessment of violations of the Convention than domestic judiciaries-when the appropriate sort of deliberation has occurred in good faith. Such deliberation among citizens on a footing of equality seems central for the intrinsic value of democratic rule as expressing individuals' autonomy as political equals.
Note that this argument does not extend to granting a margin of appreciation for the rights concerning political participation itself, including freedom of the press, freedom of political debate and other rights required for well-functioning democratic decision-making (eg Animal Defenders International Vs United Kingdom, 2013, para 100-104). These rights would appear to be necessary both for the epistemic benefits and for the processes to express the intrinsic value of individuals' autonomy. This appears to match the ECtHR's practice, being very strict in granting a margin of appreciation for such rights (Brems, 1996, pp. 266, 268).
A second area where majoritarian democratic mechanisms are not particularly reliable in securing the vital interests and equal respect for all, concerns certain interests of vulnerable minorities-such as the non-derogable rights to life (Art 2), against torture (Art 3), slavery or forced labor (Art 4)-or indeed freedom of religion (Art 9). Again, this pattern appears to be in accordance with the current practice of the margin of appreciation doctrine: The margin will tend to be narrower where the right at stake is crucial to the individual's effective enjoyment of intimate or key rights … Where a particularly important facet of an individual's existence or identity is at stake. (S. And Marper V UK, 2008, para 102;cf Follesdal, 2018;Benvenisti, 1999, p. 847) A third area where the ECtHR is not reliably better at applying the Convention is arguably when it comes to the 'balancing' among rights in the ECHR. Note that such 'balancing' does not entail less stringent human rights protection, but rather how the democratically accountable domestic government gives some rights a certain weight compared to other rights. Such 'balancing' may require detailed knowledge about the domestic culture and history-consider the dilemmas between respecting freedom of expression and the right to privacy. Though while domestic authorities are likely to know more about the domestic setting than do international judges, the former do not have any obvious epistemic strength about which alternative policies may 'balance' the legitimate interests and values sufficiently well. This requires comparative perspectives which domestic authorities may be too myopic to discern.
Finally, I submit that national courts should enjoy such a margin only when the ECtHR is satisfied that the national court has duly considered several conditions, in the form of a proportionality test-in good faith (Rasmussen V Denmark, 1988). One important reason for this requirement is that even democratic deliberative majoritarian decision-making is not always well-functioning. On the other hand, international judges do not enjoy a vantage point that makes them epistemically superior to the domestic authorities. Still, the Court can reduce the risks of human rights violations by performing such a procedural check. Thus the Court typically-but not always-requires the state to have performed a proportionality test in order to grant a margin of appreciation, or at least to be lenient about such a margin: from Hirst v. the United Kingdom (no. 2) it could be deduced that the margin of appreciation would be narrower when Parliament had not analyzed and carefully weighed the competing interests or assessed the proportionality of blanket rules. (Lindheim and Others V. Norway, 2012, p. 85) Such a test confirms that state authorities have considered whether there are less human rights-invasive alternatives, and have not ignored the impact on some groups-and thus arguably has carried out the sort of deliberative process that gives us both epistemic and intrinsic reason to value democratic rule. At the same time, the Court assures the population that this is in fact the case. I submit that insofar as such proportionality testing has not occurred,-be it in well-functioning democracies or elsewhere-the presumption in favor of domestic democratic decision-making no longer stands. The process is not likely to be epistemically reliable in identifying a human rights-respecting outcome. And the process has failed to include all affected citizens as political equals, to the detriment of its intrinsic value.
Indeed, by indicating to the state that the Court may grant it a margin of appreciation only if the state has carried out a proportionality test, the Court arguably nudges the state toward such more inclusive, deliberative-and democratic-processes, valuable both for epistemic and intrinsic reasons.
On this account, the ECtHR made a mistake in its decision Schalk and Kopf v Austria, in granting the state a margin even in the absence of a proportionality test. The dissenting judges were correct: [T]he existence or non-existence of common ground between the laws of the Contracting States' [...] is irrelevant as such considerations are only a subordinate basis for the application of the concept of the margin of appreciation. Indeed, it is only in the event that the national authorities offer grounds for justification that the Court can be satisfied, taking into account the presence or the absence of a common approach, that they are better placed than it is to deal effectively with the matter. (Schalk and Kopf V. Austria, 2010, dissenting opinion of Judges Rozakis, Spielmann and Jebens; cf. Besson, 2016, p. 18;Spano, 2014) Note finally that the Court's doctrine of a margin of appreciation also includes a role for a 'European consensus' on the matter. If the Court does not see such a consensus, it may grant the state a wider margin of appreciation and perform a less strict proportionality test (Chapman V. The United Kingdom, 2001, para 93 and 104;Dickson V UK, 2007, para 78)-though the Court's actual practice seems inconsistent (Gerards, 2004, p. 154). We return to consider the significance of such a consensus below.

| A justification for the focus on civil and political rights
Against this backdrop, consider the criticism that the ECHR focusses unduly on civil and political rights. Such a priority might stand accused of being based on flawed premises about the relative importance of these sets of rights compared to social, economic, and cultural rights. Some authors claim this as evidence that judicial review rests on contested 'liberal' assumptions. Indeed some argue that there was such a philosophical bias among the western victors of the Second World War, and that this influenced the list of rights in the ECHR (White & Ovey, 2006). Implications would seem dire for the argument of this article: such a bias means that the scope of democratic decision-making is drastically restricted. Any such entrenchment shifts conflicts about rights and policy choices away from the domestic political arenas. (Bellamy, 1999, p. 166;Bellamy, 2007). It will lead to an unfortunate focus on rights and adversarial mechanisms of conflict resolution, at the expense of 'ordinary' democratic contestation, civility and "fair and reciprocal compromise, in which all give and take" (Bellamy, 1999, p. 208). The bias in favor of civil and political rights underscores how the democratic process suffers. In contrast, well-functioning democratic institutions should allow unbiased perpetual contestation about interests, rights and policies, within institutions that foster civic virtues (Bellamy, 1999, p. 135;Bellamy, 2007).
In response, two main observations are in order. Firstly, from the beginning the Convention has protected some social and economic rights concerning property and education. And even though the ECHR does not explicitly protect many social rights, in fact the Court's case law recognizes a range of individuals' social and economic rights to various social benefits-food, water, health, social security, and an adequate standard of living (Eb V France, 2008, Konstantin Markin V. Russia, 2012, Nachova & Others V Bulgaria, 2005, Opuz V. Turkey, 2009, cf Thornton, 2014. The Court has ruled against discrimination, also regarding social and economic rights (D.H. And Others V Czech Republic, 2007, cf. Palmer, 2009. It has argued that that there are socio-economic rights entailed by the right to life, the prohibition on inhuman and degrading treatment, access to justice and the right to private and family life.
Secondly, recall that the question of which rights should be protected by the ECtHR is one of institutional design, not simply a matter of which rights are somehow normatively more important. Indeed, there is no claim that economic and social rights are less necessary for a decent life. Any focus on a narrow set of political and civil human rights of the ECHR is arguably due to the circumstances when the Court enjoys comparative epistemic advantages over domestic bodies. One reason for the limited scope of rights may be that international judicial review of all social and economic rights and trade-offs among them and civil and political rights would provide little benefit to individuals, and some further risks -once domestic democratic arrangements and the rule of law protections are in place and protected by the ECtHR. The regional human right court may often be worse placed than a domestic independent judiciary to hold politicians to account about difficult trade offs among costly, complex social and economic institutions. So if domestic democracy and the rule of law works well, international review of social and economic rights may add little value, but run the risk of juristocracy. Review by the ECtHR might therefore best remain focused largely on civil and political rights to foster and maintain well-functioning democratic processes, and leave many details concerning social and economic rights to such domestic actors. However, we may still urge that the ECtHR may provide a valuable service by checking that the domestic authorities have performed a good faith proportionality test also for a broad range of social, cultural, and economic rights. Such a check seems to provide some more protection for individuals, without running the risk of juristocracy.

| Dynamic interpretation and a European consensus
One area where the ECtHR challenges the standard domestic democratic processes is when it engages in law-making. Its doctrine of relying on a 'European consensus' might dampen the fear of such juristocracy-if the doctrine is improved.
The Court is often criticized for interpreting rights beyond the intentions of the original treaty-making states. To some extent this is unavoidable. The text of the ECHR is often vaguecontaining phrases such as the 'respect for private and family life' (Art 8). The rights must also be interpreted to protect individuals in new circumstances. Examples include whether human trafficking is prohibited as a form of slavery (Rantsev V. Cyprus and Russia, 2010), or how to understand the rights to family life of children born out of wedlock when societal norms surrounding marriage have changed (Marckx V Belgium, 1979). The ECtHR thus interprets the ECHR 'dynamically', akin to taking on a legislative function.
In response to such criticisms about the ECtHR's law-making, note that states may have good reasons to sometimes delegate such law-making to independent courts, because it may be impossible for states to predict all possible situations that will arise, or the states may think that further specification of the law is better done by such a court in the particular context (Ginsburg, 2005, p. 644).
But how reduce the risk of juristocracy? The deference to a 'European consensus,' first in Tyrer vs UK arguably limits the judges' discretion (Lixinski, 2017, pp. 66-67). The Court follows the domestic developments rather than taking the lead. The Court sums it up thus: Since the Convention is first and foremost a system for the protection of human rights, the Court must, however, have regard to the changing conditions in Contracting States and respond, for example, to any emerging consensus as to the standards to be achieved. One of the relevant factors in determining the scope of the margin of appreciation left to the authorities may be the existence or non-existence of common ground between the laws of the Contracting States. (Glor V. Switzerland, 2009, para. 75) Such a policy restricts the judges' discretion in its 'dynamic interpretation' and in granting states a margin of appreciation. Such a check reduces the risk that the judges exercise unconstrained discretion.
Note that there is no claim that the fact of agreement among states enhances the authority of such interpretations, by democratic arguments or otherwise. 1 In addition, the epistemic premises this practice relies on merit scrutiny, both what such a consensus is about, and the judges ability to discern it (Dzehtsiarou, 2015;Follesdal, 2019;Kapotas & Tzevelekos, 2019). Many agree that that the consensus doctrine is currently too vague and unpredictable. Indeed, there is a risk that the Court finds a consensus or lack of such in support of conclusions the majority of judges want for other reasons. And at least until fairly recently, the Court did not have the requisite expertise and resources for comparative legal studies (Bates, 2019;Dzehtsiarou, 2015, pp. 82-101). So as currently developed, the European consensus doctrine does not appear to guide or constrain the Court enough to reduce the risk of juristocracy.
A further weakness of the current practice regarded as a guide or even check on the Court is that the topic of consensus varies drastically. The Court is sometimes concerned with legislation, sometimes with more abstract principles-even consensus as regards more philosophical assumptions e.g., about when life begins (Vo V. France, 2004). And the Court describes and frames the object of consensus in various ways. Regarding prisoners' right to vote, the Court will find a consensus or not depending on whether the Court asks how many member states have a blanket ban on the right of prisoners to vote, or how many member states have some restrictions on such a right. This is one difference between the majority view and the dissenting opinion in the UK's prisoner's voting case (Hirst V. The United Kingdom (No. 2) 2005). Furthermore, the states may not have intended to elaborate or confirm a rule or practice, but rather to establish a one-off policy to address a particular concern or trade-off. Even Judge Spielmann, a former President of the ECtHR, notes that the Court lacks relevant information about the legislative histories and how the policies operate in various jurisdictions (Dzehtsiarou, 2019, p. 41).
A further concern is that the threshold number of states required for a 'consensus' various drastically across cases (Dzehtsiarou, 2019). Former President of the Court Judge Wildhaber recognizes that on this point the "Court's case-law is in some respects fluid and even fuzzy" (Wildhaber & Hjartarson, 2013, p. 262 Leaving these issues aside, why should the Court heed such a consensus in the first place? There may be reasons for the Court to show that its law-making is constrained by the majority of states. And the Court's alleged sightings of an emerging consensus may warn states of how the Court may judge cases in the future (Stone Sweet & Brunell, 2013). This may prompt them to change policies-or to the contrary complain so loudly that the Court can reinterpret any such patterns it saw. The Court sometimes says quite clearly that its interpretation of the treaty may change in the future due to such changes among the Member States: The Court cannot but note that there is an emerging European consensus towards legal recognition of same-sex couples. Moreover, this tendency has developed rapidly over the past decade. Nevertheless, there is not yet a majority of States providing for legal recognition of same-sex couples. The area in question must therefore still be regarded as one of evolving rights with no established consensus, where States must also enjoy a margin of appreciation in the timing of the introduction of legislative changes. (Schalk and Kopf V. Austria, 2010, p. 105, my emphasis) It is difficult to discern epistemic reasons for this reliance on a consensus. Is there any reason to believe that it indicates the correct interpretation and balancing of the ECHR, warranting restrictions of the margin of appreciation (Helfer, 1993)? Neither a democratic nor an epistemic argument drawn from Condorcet seems convincing-though another epistemic argument seems plausible.
An argument in favor of consensus from democratic majoritarianism among states seems illfounded. Neither parliaments nor judiciaries can plausibly be said to have consented to this use of their own policies, even though the Court chooses to regard their practice as part of a consensus (Dzehtsiarou, 2015). As Pellet notes in a different setting: … States' behaviour converging towards such a practice can be said to be "voluntary". But State wills are aimed here at doing something, not at elaborating a rule of law. If a national court makes a decision on State immunity, its judgment will be part of the general practice. That, however, was not the aim of the court's decision: its only concern was to decide on the precise dispute it had to solve. (Pellet, 1989, p. 36) Another flawed argument sometimes offered is based on Cordorcet's "jury theorem." (Condorcet, 1785;Dothan, 2015;Posner & Sunstein, 2006). When several independent experts vote which of two competing claims are correct, as long as each expert is slightly more likely to be correct than not, the larger a majority for one claim the more likely it is to be correct. But there are two important disanalogies to this case. Firstly, one must assume independence among the individuals' decisions for the theorem to hold. But European states look to each other's jurisprudence in law-making (Dothan, 2015). A further difference concerns what counts as 'correct' answers. Condorcet's theorem holds in the first instance for choices among two alternatives, only one of which is correct. The argument also holds for choice of the unique correct answer among several alternatives (List & Goodin, 2001). However, the objective of the ECtHR is not, and arguably should not be, to harmonize the legislation of the various domestic jurisdictions to make it uniform (Carozza, 1998(Carozza, , p. 1228Merrills, 1988). But since human rights requirements underdetermine the domestic legal arrangements, the task of the ECtHR is less Procrustean: it should instead only constrain the varieties of domestic legislation.
So an actual emerging trend in legislation among other states does not seem to indicate the normatively preferred interpretation or balancing, nor should it automatically count in favor of reducing the Margin granted one state.
I submit that there may be one epistemic role of such developments in member states: it may aid the ECtHR in discovery. Agreement among European states may be an indication of new grounds for unacceptable discrimination, e.g., with reference to Art 14. Thus the Court has a history of appealing to consensus in favor of children's rights (Tyrer V United Kingdom, 1978); 'illegitimate' families (Marckx V Belgium, 1979); homosexual behavior (Dudgeon Vs. United Kingdom, 2000, p. 60), in favor of prisoners (Hirst V. The United Kingdom (No. 2) 2005 § 81), and in favor of Roma children's access to education (Orsus & Others V. Croatia, 2010).
When several states make such new legislation, policies or court judgments, this trend may give evidence that there is a group hitherto overlooked, which the Court should protect against discrimination. The Court may still be allowing states discretion in how to remove the discriminatory practices, with due respect for varieties in institutions and history. However, note that this defense of the Court's reliance on a European consensus is modest.
Firstly, such a consensus cannot be a necessary condition. That would render the Court too vulnerable to biased majorities (Benvenisti, 1999, p. 851;Fenwick, 2016, pp. 249-50). Furthermore, this epistemic contribution does not depend on the number of states: it may suffice that some states prohibit such discrimination for what appears to be very strong reasons, for the Court to discover that this is how it should dynamically interpret the Convention, or balance Convention rights. Indeed, the Court can learn such insights from other jurisdictions as well. Its judgments have referred to laws of Canada, South Africa and many other states (Hirst V. The United Kingdom (No. 2) 2005, pp. 35-39;cf. Dzehtsiarou, 2018, p. 107). On this line of argument, the appearance of consensus should prompt the Court not to blindly agree to the consensus, but rather lead the Court to look for the possible arguments presented in domestic arenas for such developments. Indeed, this role of consensus is limited: "International consensus merely served to reinforce the status of the applicants as a minority. This, in turn, serves as the justification for the Court to narrow what would have otherwise been a wide margin of appreciation." (Kagiaros, 2019, p. 306).
To conclude: what are we to make of the criticisms against judicial human rights review made by Bellamy and others (Bellamy, 1999, p. 166;Bellamy, 2007), that it shifts conflicts about rights and policy choices away from the domestic political arenas for 'ordinary' democratic contestation and "fair and reciprocal compromise, in which all give and take?" The account sketched above provides a complex account of the relationship between the Court and the domestic authorities, whereby it may serve to strengthen the sort of state autonomy we and citizens have reason to value. The Court can perform a subsidiary supportive function, helping to improve the domestic epistemic democratic processes. So I submit that the criticism does not seem to hold as a general description of the ECtHR, and in particular not when we consider the margin of appreciation doctrine.
The role of the Court is to review domestic legislation, but then to return reform tasks to the domestic authorities for renewed democratic process. This reconsideration, can arguably contribute to give minorities more voice in such deliberations, so that the compromises are more fair. And the nudging toward a better proportionality test may render states more likely to engage in more careful, epistemically more thorough public proportionality testing,-which is in part what gives democratic processes value.
However, one worry remains, namely that that judges of the ECtHR are engaged in lawmaking without sufficient democratic input or checks remain, appeals to a 'European consensus' notwithstanding. And importantly, so far this argument mainly points out how the ECtHR can contribute. For such contributions to be likely, we must look closer at some conditions concerning the independence of the Court and the impartiality and expertise of the international judges.

REQUIRED: HOW TO SELECT, GUIDE, AND GUARD THE GUARDIAN LAWMAKERS
The above argument for judicial human rights review and development of international law by experts has many lacunae. In particular, a broad range of scope conditions and requirements seems important with regard to the experts. For arguments of relative epistemic competence to hold in practice, we must consider more carefully how the judges are selected, how they are guided and supported by the ECtHR staff, and how to guard against the possible abuse of their discretion.
The review and development of the Convention rights require the ECtHR to have access to local and counterfactual knowledge about the likely impact of policies in various states for vulnerable citizens, and which alternatives may be feasible for the various governments to adopt. This is one reason why the bench always includes a judge 'in respect of' the accused state in particular.
The judges must be highly competent in legal method. And they must be-and be perceived as-sufficiently independent of the states that appoint them. Otherwise they cannot be trusted to perform an objective review of compliance with the ECHR. The difficult combination of these tasks already informs the nomination, vetting and selection processes to some extent. The Parliamentary Assembly of the Council of Europe elects the judges, one judge from each list offered by a state with three qualified candidates. The Parliamentary Assembly requires that states describe the nomination process, according to the guidelines issued by the Committee of Ministers which must include a 'fair transparent and consistent national selection procedure' (Committee of Ministers, 2012). The competition for the post must be announced in specialized literature. A Committee of the Parliamentary Assembly can dismiss the slate for various listed reasons, such as lack of members of both genders on the list (Lemmens, 2015). The Committee of Ministers has set up an Advisory Panel of Experts to examine and ensure that the candidates meet the requirements for office (Committee of Ministers, 2010). The term of office also promotes independence from the appointing bodies. Previously the judges served a six-year term, renewable once. This was changed to one non-renewable nine-year term, to reduce the risk that hopes of reappointment would affect the judges' voting and 'to reinforce [the judges'] independence and impartiality' (Council of Europe, 2005).
The epistemic qualifications of expert judges and other members of the ECtHR go beyond these requirements. It is particularly important that they as a body can consider a broad range of viewpoints and salient legal aspects of the situation and the Convention. This is of course relevant to decide a particular case, but even more important because the Court develops the Convention with impact on inhabitants in Europe as a whole, far beyond the particular parties to the case. Consider two concerns about lack of representative perspectives on the Court.

| Underprivileged perspectives-Gender equity and other
One possible bias within the ECtHR that has received attention concerns gender. There are currently almost a two-thirds majority of male judges. 2 The need to secure the broadest range of relevant facts and legal norms suggests that the Court needs to have a more varied population. This imbalance may reduce the epistemic competence of the Court. One might fear that this has an unfortunate impact on the ECtHR's interpretations and judgments, and on the general reputation and perceived legitimacy of the ECtHR. For instance, the gender imbalance may lead some to question the representativity of the ECtHR and the impartiality and fairness of the process and of the ECtHR itself (Follesdal, 2021;Grossman, 2016, p. 82-95). What are we to make of such concerns? There is hitherto insufficient research concerning whether female and male judges tend to vote differently in the ECtHR (Grossman, 2016, p. 89), and on any panel effects: Whether having both genders represented on the bench present affects the deliberations and the decisions (Terris & Romano, 2007, pp. 186-87;Wald, 2005, pp. 979-993, 989). There are several topics that merit research: how to understand and identify 'women friendly' judgments; whether these are judgments which correct existing gender discrimination, or favor the situation of groups which historically tend to consist of women. Further questions concern how 'representative' and sensitive to the group members' interests the elected members of the court actually are (Follesdal, 2021;Phillips, 1995;Young, 1997, p. 44).
The Parliamentary Assembly of the Council of Europe has taken laudable steps. It requires that the state's list of three candidates for each post as judge must include at least one man and one woman (PACE, 2004). The Assembly has often sent lists of candidates back to the state to urge compliance with this requirement. Nevertheless, Malta and Belgium have claimed lack of qualified women nominees. The de facto deference of the Council of Europe bodies toward such claims merits further scrutiny.
The discussions surrounding the gender bias of the composition of the ECtHR gives rise to a more general concern: there would seem to be parallel arguments to ensure broad representativity of the affected population of Europe among the judges, based on the epistemic need for perceptive attitudes toward perspectives and arguments from historically and presently discriminated groups.

| Professional background
Another lack of relevant perspectives that merits concern in connection with the ECtHR is the professional backgrounds of the judges. The Court should not be biased in favor of, nor against, the state interests when they conflict with human rights. However, the judges of the ECtHR appear to carry with them the professional norms of their previous places of employment, to some extent. Former employees of central state administration and diplomats tend to be more 'state friendly' in general than former lawyers with human rights organizations, as evidenced by their voting patterns (Bruinsma, 2006;Voeten, 2008, p. 428). This should not come as any surprise. But there seems to be a shift in the professional backgrounds of the judges of the ECtHR, so that the judges are generally younger and more specialized in human rights-and states tend not to select human rights activists (Madsen, 2015, pp. 259-278). This may in due course challenge the need to have expertise and perspectives on the bench with regard to the interests both of individuals and of the state. Indeed, the states may strategically select more state-friendly judges, rather than human rights-activists, to render the ECtHR more predisposed to the raison d'état' (Voeten, 2007, pp. 669-701, 672).

| The Registry
A final topic regarding the epistemic quality of the ECtHR and ways to guide its discretion concerns the Registry of the Court. It provides crucial services to the judges to enhance their epistemic quality regarding judgments and development of the Convention. Two important 'departments' within the Registry merit particular mention. The Office of Jurisconsult assists the Court to ensure quality and consistency of the case law (Rule 18B, Rules of Court). The other department is the Research and Library division, which helps with comparative research to determine the existence of a 'European consensus' or a lack thereof (Londras and Dzehtsiarou, 2018, p. 22).
The loyalty of the Registry to the objectives of the ECtHR need not be questioned, but several recent developments of the ECtHR require the Registry to 'balance' several important objectives-a difficult task even in good times. With one-term judges, the Registry becomes a more important repository of institutional memory, including precedents, since more of the judges are relatively new to the job. With increased pressure to reduce the case load, the Registry's preparations for the judgments become increasingly important. Several reports indicate that the Registry works to fulfill a certain target number of cases, with the effect that more complex cases may be left aside (Woolf, 2005, p. 55). The Registry also provides much needed assistance to judges in drafting the judgments of a case, suggesting both reasoning and outcomes. This epistemic framing helps ensure consistency and high judicial quality, but may reduce the ECtHR judges' attention to alternatives and to their real spheres of discretion.
More studies seem required to understand and assess patterns of recruitment to the Registry, the epistemic strengths and biases of its staff, and the choices that help frame the decisions of the judges of the ECtHR, for better and worse. In particular, increased focus on consistency and a more specific and public doctrine of a margin of appreciation may reduce the discretion of judges and hence diminish fears of juristocracy-and increase the possibility to check the developments.

| CHALLENGES AND PATHS OF IMPROVEMENT
This justification of the ECtHR' general practice of judicial review appears to also support the margin of appreciation doctrine. It arguably reflects the scope conditions of the epistemic comparative advantage of the Court over domestic bodies. However, several aspects of the practice of judicial review should be changed.

| Selecting more diverse expert judges to the ECtHR
To ensure the requisite epistemic tasks facing the judges of the ECtHR, their diversity of perspectives merits close attention. They must be able and motivated to jointly elicit and consider all relevant facts and legal arguments that pertain to the case at hand. Such a broad range of views is even more important since the Court actively develops the ECHR. To be defensible, such lawmaking must pay attention to the likely impacts on the populations of the Member States at large-including often overlooked or even oppressed minorities.
There is already a body within the Council of Europe instructed to consider some such diversity: A standing Committee of the Parliamentary Assembly. It shall consider the candidates "not only as individuals but also with an eye to a harmonious composition of the Court, taking into account, for example, their professional backgrounds and a gender balance" (PACE, 2016). It seems appropriate to uphold present gender requirements (Follesdal, 2021). The attention to professional backgrounds is even more important to better handle attempts to stack the Court with state friendly judges. Compare, for instance, the instructions concerning election of judges to the International Criminal Court which necessitates competence in both international law and in criminal law and procedure (Art 36). The epistemic arguments above also lend support to also have other ranges of perspectives included, such as judges who are familiar with several religious traditions, in particular Islam. Judges should also come from a range of ethnic backgrounds and be familiar with a variety of sexual orientations as well as other minorities at added risk from being ignored or treated worse by the majority populations.
With regard to the Registry, more research is needed to understand and assess patterns of recruitment to the Registry, and how its choices about priority of cases and relevant case law frames the decisions of the judges of the ECtHR. Its important roles may make diversity among its staff similarly important.

| The doctrine of a margin of appreciation
This Doctrine has received much praise and much criticism, some of both are well deserved. It expresses some respect for sovereign democratic self-government but still granting no margin of appreciation concerning the rights to life, against torture or slavery, or to rights necessary for democratic decision-making. Yet the 'Doctrine' is so vague and multifarious that even to refer to it in the singular, and to call it a 'doctrine' seems unduly charitable. More fundamentally, its current vagueness grants both the ECtHR and powerful states too much discretion, and put human rights at risk, contrary to the purpose of the ECHR.
The Doctrine creates legal uncertainty, because states are unable to predict and hence cannot avoid violations of the ECHR (Arai-Takahashi, 2013;Brauch, 2005, p. 125;Lester, 2009;Macklem, 2006). Indeed, even the judges of the Court disagree about the Doctrine (Z V. Finland, 1997, Judge De Meyer partly dissenting, and The Sunday Times V United Kingdom, 1979, Observer and Guardian V United Kingdom, 1991Wingrove V United Kingdom, 1996, dissenting judges). To some extent the uncertainty is due to the legal norms, rather than the margin of appreciation doctrine itself. But disagreements among judges about specifics of the doctrine are legion. And the result may be that powerful states do as they want, while the Court must grant them a margin as it must.
The main advice would be to make the rules and conditions of the doctrine more precise. Such specification must be guided by an understanding of why a margin of appreciation should be accepted at all, for instance along the lines indicated above.
Among the points that should be clarified, consider two. Firstly, the respect for democratic self determination and epistemic comparative advantage support a requirement that a good faith domestic proportionality test is a necessary condition for the Court to grant a margin at all: without such domestic attention to alternatives and their human rights implications, deference seems unwarranted. Secondly, for a range of issues including necessary limitations of rights, there is no reason to believe that the Court generally has good information about alternative policies that may secure the objectives at least as well, without curtailing the rights as much as their preferred option. Such assessments would seem to require comparisons among states, and expertise concerning societal norms and culture far beyond anything judges are trained for. But such information is also not readily available at the domestic level. Thus it seems that-with regard to limitations on certain human rights-the Court should carry out a procedural check on whether domestic authorities have carried out a proportionality test, assisted by the Registry's Research and Library division.

| European consensus
Finally, consider the role of the ECtHR's sightings of an '(emerging) European consensus.' The upshot of the discussion above is that three of the offered arguments appear plausible. A European consensus may guide and constrain the Court's discretion in interpretation thus reducing the risk of juristocracy, and the Court's claims about an emerging consensus may warn states of developments to come, to avert future cases. And there is an epistemic contribution: the developments in domestic jurisdictions may help the ECtHR discover new forms of violations of the Convention. The last argument does not require any specificity concerning the topic of convergence or the number of states converging. But the first argument has important implications. To function as a real constraint on judges, the objects of consensus must be more clear: is it a democratically enacted law or policy or an accepted practice or a judicial interpretation? And who shall determine what counts as an emerging consensus? The research unit at the Registry now provides comparative analysis, but seldom considers all jurisdictions, and not in all cases where the Court claims to see an emerging consensus (Dzehtsiarou, 2015).

| CONCLUSIONS
There are good reasons why even the authorities in a well-functioning democracy should defer to judicial review by distant judges of regional human rights courts. At the same time, such courts should sometimes defer to state bodies for some such assessments. The arguments presented here have sought to first show that there are epistemic reasons to welcome judicial review of the kind provided by the ECtHR, even in generally well-functioning democracies. Indeed, such review can help protect the political rights necessary for domestic democratic procedures to have both intrinsic and epistemic value. The ECtHR's independence from the domestic authorities allows it to provide some assurance to other parties-such as citizens-whether the state is indeed in compliance with its human rights obligations, including those required for intrinsically and epistemically valuable democratic procedures. Secondly, the argument defends some domains of state sovereignty from review by a regional human rights court, on the basis of considerations of comparative epistemic competence, respect for the intrinsic values of democratic decisionmaking, and the risks of juristocracy.
The account relies on several peculiar features of the ECtHR within a multilevel system of institutions marked by complex interdependence and the fact that the ECHR human rights norms are standards that underdetermine policies. The arguments for why states should enjoy some margin of appreciation for certain issues are based on considerations about when ECtHR judges and domestic judges, respectively, are likely to have the specific and limited comparative epistemic expertise needed to curb domination by domestic majorities. The margin of appreciation, properly circumscribed, also reduces the risk of domination by judges, while respecting intrinsic and epistemic values of democratic decision-making.
This account may also justify the ECHR's bias toward civil and political rights: once they are in place, the complex issues of social and political rights may better be left with domestic democratic bodies-though the Court might be well placed to perform a proportionality test also regarding such rights.
The account matches the present practice on several points-including a robust practice of the Court to grant states a margin of appreciation under certain conditions. However, this account is not a mere apologia for the existing practice. To the contrary, the present practices of the ECtHR must be specified much more closely, guided by these arguments. These range from the Council of Europe's rules to nominate and elect judges and members of the Registry of the Court, to how to specify the margin of appreciation doctrine and the roles of sightings of an 'emerging European consensus.' The current practices are too vague to guide and guard these guardians of the European Convention on Human Rights. The current practices need not be abolished, but should be improved to further reduce understandable fears of juristocracy.
The paper builds on earlier writings that have benefited from various fora for constructive criticism. The article received helpful comments at a GOODPOL seminar at the Norwegian Centre for Advanced Study 2021. I am also grateful to colleagues at PluriCourts, and especially for thorough and constructive comments from Antoinette Scherz and Martin Vestergren, as well as from an anonymous reviewer.