In his essay ‘Human Rights and Relativism’, Bernard Williams declares that ‘Whether it is a matter of good philosophical sense to treat a practice as a violation of human rights, and whether it is politically good sense, cannot ultimately constitute two separate questions’ (Williams 2005, p. 72). At first blush, this remark appears enigmatic and infuriating. Why cannot we distinguish between the philosophical merit of considering x a violation of a person’s human rights and the political expediency of treating x in those terms? Does not Williams’s dismissal of any such distinction implausibly deny that a gap exists between our theoretical understanding of human rights and our practical responses to human rights violations?

In this paper, we explain the underlying rationale behind Williams’s assertion and in so doing hope to undermine at least some of the kinds of criticisms noted above. We argue that Williams’s approach to thinking about human rights is plausible and worth taking seriously on its own terms and that it, alongside his broader realist political commitments, has significant implications for our understanding of human rights along a number of key dimensions. Importantly, we claim that, whilst it is possible to group Williams’s conception of human rights along with other so-called ‘political’ and ‘minimalist’ conceptions that have been fashionable lately in theoretical debates, his account is distinctive due to its grounding in a form of ‘political realism’. We then proceed to explain how Williams’s politically realistic way of thinking about human rights appears to cohere well with certain aspects of the reasoning of one of the most important international human rights courts, the European Court of Human Rights. The fit between the two indicates that Williams’s view has a firm grounding in political realities and that these realities, insofar as they manifest themselves in judicial practice, can be explained along lines that Williams has identified. This lends further plausibility to the view that a politically realistic understanding of human rights, of the kind urged by Williams, should be taken seriously since it reflects (at least a certain number of) commitments of actors engaged in practices of arguing and disagreeing about international human rights. We close by examining the suggestion that thinking in these terms is worryingly conservative.

Human Rights and the Nature of Politics

In ‘Human Rights and Relativism’, Williams suggests that we should think about human rights in decidedly non-ambitious terms. Whilst it might be true that various human rights declarations and instruments, as well as national constitutional texts, include a variety of items that may go far beyond a basic understanding of fundamental human rights, Williams insists that a proper philosophical inquiry into the nature of human rights should begin with ‘clear cases’ (Williams 2005, p. 63), where identification of such cases does not already involve acceptance of a substantive moral outlook, such as liberal universalism. This, in itself, is not a novel thesis. Following the publication of Rawls’s Law of Peoples (Rawls 1999), a distinctively ‘political’ and/or ‘minimalist’ way of thinking about human rights has emerged, and various authors have made similar claims in recent years (See, e.g., Ignatieff 2001; Cohen 2004; Letsas 2007; Gilabert 2011). These modes of thinking about human rights part ways with a ‘humanist’ or ‘naturalistic’ perspectives (Gilabert 2011) which proceed by elaborating on the idea that humans have certain pre-institutional rights because of their common humanity. In contrast, political conceptions take human rights practices to essentially involve claims made against institutional structures and political forms (paradigmatically nation-states). Moreover, political human rights conceptions are also predominantly (but not necessarily or unanimously) minimalist inasmuch as they place constraints on the normative shape of human rights claims, where the constraints stem from normative considerations such as global pluralism about fundamental standards of legitimacy and justice (Cohen 2004).

On the face of it, Williams’s conception of human rights appears to fit nicely with such understandings of human rights. Still, we claim that, even if such a taxonomy makes some sense, Williams’s construal of human rights has a number of features that make it stand out. Indeed, it is only by properly grasping Williams’s distinctively realistic understanding of politics that adequate light can be shed on the unique grounds of his human rights minimalism as well as his scepticism with regard to more normatively ambitious understandings of human rights. This suggests that political conceptions of human rights can be more and less realistic (or more and less moralistic). Indeed, we argue that Williams’s rejection of ‘political moralism’ distances his views from various other normative conceptions of human rights on theoretical offer, both political and naturalistic.

The central claim that runs through Williams’s essays on human rights is that a class of basic or fundamental human rights can be grounded in an understanding of the distinctive nature of politics and an appreciation of the norms of legitimation that are internal to it, rather than by prioritising any sectarian claims about the substantive moral requirements of politics. This, famously, is part of what Williams’s rejection of ‘political moralism’ and endorsement of a form of ‘political realism’ is all about. Yet it is important to recognise that Williams’s political realism is a complex position that can only be properly understood by reference to Williams’s work as a whole and, especially, his landmark and sweeping critique of contemporary moral philosophy, which is expressed most clearly in Ethics and the Limits of Philosophy (Williams 2006). In this sense, Williams’s political realism couples a certain (complex) kind of understanding of political matters with a certain (complex) kind of understanding of the conceptual and motivational limits of the ‘morality system’.

Williams’s political thought revolves around his understanding of what he calls the first political question. Williams claims that our understanding of politics must start with reference to Hobbes’s question of how we can secure ‘order, protection, safety, trust, and the conditions of cooperation’. This is the first political question because solving it ‘is the condition of solving, indeed posing, any others’ (Williams 2005, p. 3). As Williams sees it, for a political situation rather than mere anarchy to hold, a solution to the first political question must be offered. This distances his approach from views which, for example, do not begin from the search for a solution to the problem of disorder but, rather, from the search for a solution to the problem of the fairness of the terms of already existing social cooperation (for a classical example see Rawls 1999). Moreover, Williams insists that the existence of social cooperation must not be regarded as a fait accompli: a solution to the first political question is required ‘all the time, [and] … is affected by historical circumstances’ (Williams 2005, p. 3). This is crucial because it entails that there is no way to neatly insulate questions about how the benefits of social cooperation ought to be distributed from those considerations about the possibility of disintegration of a social and political order and/or the conditions of workable social cooperation. This constitutes a significant departure from approaches in political theory that simply legislate away the problem of the creation and sustainability of political order by building effective social cooperation into their very definition of politics, or which simply fail to address these issues altogether.

Yet, Williams departs from Hobbes because he does not think that any solution to the first question ‘is sufficient to determine the rest of the political arrangements’ (Williams 2005, p. 3). This is because he stresses that while the central point and purpose of politics is to save people from the kinds of inconveniences and dangers that accompany situations of basic disorder, purported solutions to the first political question can also inflict inconveniences and dangers of their own. If they do, it is hard to see how they can be regarded as solutions to the set of problems that politics exists to resolve.

With this in mind, Williams introduces the core normative component of his political thought: the basic legitimation demand (BLD). He claims that:

If the power of one lot of people over another is to represent a solution to the first political question, and not itself be part of the problem, something has to be said to explain … what the difference is between the solution and the problem, and that cannot simply be an account of successful domination. It has to be a mode of justifying explanation or legitimation. (Williams 2005, p. 5)

To this end, when a group claims to resolve the first political question, they must explain to those over whom they exercise power why they are not merely re-imposing the inconveniences and dangers that accompany conditions of basic disorder. In turn, this requires them to offer a justification or legitimation of their actions. As Williams puts it, when ‘A coerces B and claims that B would be wrong to fight back: resents it, forbids it, rallies others to oppose it as wrong’, A is effectively claiming that ‘his actions transcend the conditions of warfare, and this gives rise to a demand of justification of what A does’. When ‘A is the state, these claims constitute its claim of authority over B’ (Williams 2005, p. 6). Accordingly, the BLD requires that a justification or legitimation of the state’s actions to be offered to everyone that the state claims to authoritatively coerce.Footnote 1

This account of the first political question influences Williams’s political understanding of human rights because it trades on the claim that one of the few necessary truths about the nature of right is that ‘might is not right’; Williams insists that coercion requires legitimation and that the power of the stronger does not legitimate its own use (Williams 2005, p. 23). This has implications for our understanding of which kinds of political actions/orders are, in principle, capable of being classed as legitimate. Even though what is required for a state to satisfy the BLD will vary in light of a host of contingent factors dependent on the time and place in which the demand for legitimation arises,Footnote 2 Williams’s schematic account excludes ‘people using power to coerce other people against their will to secure what the first people want simply because they want it’ (Williams 2005, p. 23).

He claims this gives us a foothold from which to understand the most basic violations of human rights: ‘The situation of one lot of people terrorizing another lot of people is not a political situation; it is, rather, the situation which the existence of the political is in the first place supposed to alleviate (replace)’ (Williams 2005, p. 5). For a solution to the first political question to avoid merely replicating the problems of disorder of illegitimate coercion (with regard to the whole population under rule or a certain subset thereof), the state must refrain from engaging in certain kinds of actions. The most basic and fundamental conceptions of human rights therefore relate to our sense of what it is for political power to successfully avoid doing this; that is, to avoid merely being an imposition of ‘unmediated coercive power’. Actions like murdering, torturing, arresting and detaining one’s opponents, as well as some forms of surveillance, are, Williams claims, ‘abuses of power that almost everyone, everywhere has been in a position to recognise as such’, and sophisticated philosophical argument is not required to explain what is wrong with these kinds of actions (Williams 2005, p. 26).Footnote 3 The thinness of the normative basis of such fundamental human rights thus coheres rather well with Williams’s critique of the ‘morality system’, since a substantive ethical theory is not needed to establish their acceptability. This, however, does not imply that, as a matter of empirical reality, there will not be hard or borderline cases, where the recognition of the political—as opposed to the purely coercive—character of a regime may be a matter of degree.Footnote 4

There is a sense in which Williams appears to share with other human rights minimalists such as Cohen and Ignatieff (the list could also include John Rawls and Charles Beitz, since all of those authors are committed to the idea that some form of public reason constraint drives a wedge between first-best theories of human rights, whatever their content, and human rights proper), the view that basic or fundamental rights differ from less basic claims involving the proper limits of freedom of expression, or pornography, or sexual toleration (and so on). Accordingly, explanations of why violations of such basic or fundamental goods violate human rights do not have to appeal to any contestable liberal assumptions about the nature of political morality (Williams 2005, p. 18; cf. p. 26). Still, Williams does not set the normative bar here at a level determined by recourse to some form of a-contextual and a-historical moral argument. Rather, in line with his realist commitments, his position is fluid. At a fundamental level, the BLD and Williams’s critique of the ‘morality system’ are the only fixed points in the enquiry; they are flexible enough to be able to accommodate a host of different human rights practices depending on the overall configuration of the social and historical context, which calls for some form of interpretation. This in no way precludes (but also it in no way guarantees) that in the future liberal assumptions could play a more basic role (say, if they were the only ones deemed acceptable by the world’s population). In this sense, there is a certain affinity between Williams’s conception of human rights and Beitz’s attempt (Beitz 2009) to come to an understanding of human rights not on the basis of abstract moral argument but, rather, through an interpretive reconstruction of extant human rights practices.

In this vein, Williams claims that we should not forget that ‘[t]he charge that a practice violates fundamental human rights is ultimate, the most serious of political accusations’ (Williams 2005, p. 27). But he stresses that this ensures that it is an accusation that must be made responsibly. Thus, Williams claims that focusing on the basic or fundamental human rights is philosophically and politically prudent as it gives credence to the idea that human rights are not the mere expressions of a particular, sectarian political morality. This has the corollary implication that we should not count as rights violations ‘every practice we reject on liberal principle’ (Williams 2005, p. 72). Although various claims about, for example, the equality of treatment between the sexes, economic redistribution, or high levels of sexual toleration, resemble cases of human rights in so far as ‘their basis is not positive law but a moral claim which is invoked in arguments about what the positive law should be’, Williams insists we should think about such issues in a different vernacular. For example, he claims that even though we may think that people who are caught up in the grip of theocratic conceptions of government, or patriarchal ideas about women, are simply mistaken:

We may see the members of this society as jointly caught up in a set of beliefs which regulate their lives and which are indeed unsound, but which are shared in ways that move society further away from the paradigm of unjust coercion. In that case, although we shall have various things to say against this state of affairs, and although we may see the decline of these beliefs as representing a form of liberation, we may be less eager to insist that its way of life constitutes a violation of human rights. (Williams 2005, pp. 71–72)

It is consequently problematic to claim that theocratic regimes or patriarchal systems necessarily violate people’s basic human rights, because such legitimations may be accepted by the citizens in these states.Footnote 5 In this sense, they differ from the core or basic violations as these are blatant examples of unmediated coercion which transgress the ‘might is not, in itself, right’ axiom at the core of Williams’s account of politics. How far the less basic situations fall into the violation paradigm is a matter of interpretation and social understanding: ‘up to a certain point it may be possible for the supporters of the system to make a decent case … that the coercion is legitimate. Somewhere beyond that point may come a time at which the cause is lost, the legitimation no longer makes sense, and only the truly fanatical can bring themselves to believe it’. In this scenario, ‘there will be no great change in the argumentative character of the legitimation or the criticisms of it. The change is in the historical setting in terms of which one or the other makes sense’ (IBWD, 71).

One of the important implications of this view, an implication that we draw out in the next section, is that we need to think less dogmatically about the less fundamental or basic rights claims that now play an increasingly prevalent role in the political practice of Western liberal democracies. Williams notes that it is simply a fact that many European liberals ‘find it a quaint local obsession of Americans that they insist on defending on principle the right to offer any form of odious racist insult or provocation so long as by some argument it can be represented as a form of speech’. Whether or not such speech should be permitted is clearly a matter of political judgement—one which, Williams sardonically remarks, requires us to decide if ‘enemies of liberalism have been given only enough rope to hang themselves … [or] enough rope to hang someone else’ (Williams 2005, p. 19).

In one sense, this is a reasonably common (English) dismissal of the (American) predilection for prioritising principle over the undeniable virtues of political prudence. But the underlying philosophical point is an important one. Given that disagreement is endemic when we examine the less basic or fundamental cases, it is theoretically irresponsible to claim that a political practice or law violates people’s human rights merely if we reject its underlying legitimation. On the contrary, deciding when unsound legitimations should be conceived of as unjustified exercises of power is a matter of concrete political judgement and is something that requires us to focus on a host of factors, some of which are distinctively ‘moral’ but others of which are not.

This is where Williams’s ‘political realism’ becomes important once again. Given his realist inclinations, Williams suggests that it is a mistake to assume that political judgements and arrangements can be reduced to, or seamlessly derived from, a set of antecedent moral considerations or claims. On this basis, it is possible that moral considerations may suggest that a practice violates people’s human rights, but other reasons should lead us to refrain from making such a judgement. And this is not necessarily an unprincipled capitulation to political expediency. On the contrary, it may be an expression of something like a Weberian ethic of responsibility. Moreover, Williams’s realism is also well placed when it comes to making sense of potential tensions between the practice of arguing about human rights and ‘the political’ understood as a field of conflict that can potentially lead to the unravelling of political orders. These are precisely the cases where the first political question is posed anew and interacts with the content of human rights claims insofar as the latter presuppose a workable scheme of social cooperation.

We may sum up Williams’s distinctive understanding of human rights in the following set of propositions, which are informed by Williams’s realistic understanding of politics as well as his critique of the ‘morality system’. First, human rights currently play a central role in our judgements of which answers to the first political question are capable of satisfying the basic legitimation demand. Second, crucially, making concrete judgements to this effect is not a matter of ahistorical moral reasoning, but a function of situated political judgement; argument that works with specific contextual materials under conditions of disagreement. Third, the more we move away from core violations of human rights, the more there is room for argument and sensible accommodation of different assessments. Fourth, Williams suggests that we must view human rights violations as especially important political accusations. As a result, the tendency to inflate the domain of human rights on the basis of a sectarian moral theory should be resisted.

Looking to Practice: The European Court of Human Rights

There are myriad ways that one might wish to scrutinise Williams’s distinctively political view of human rights. In this section, we shall do so by showing that at least certain features of judicial reasoning about international human rights seem to lend support to Williams’s understanding. In so doing, we suggest that a number of well-established international human rights practices could be explained by reference to Williams’s realistic outlook. In fact, Williams’s conception of human rights and certain human rights practices could be understood to support each other. In particular, we shall focus on aspects of the practice of one of the most important international human rights courts, the European Court of Human Rights (ECtHR), which is tasked with the interpretation and effective implementation of the European Convention on Human Rights (ECHR). Whilst we recognise that this does not amount to an unqualified argument in favour of Williams’s approach, it does at the very least indicate the latter’s relevance, since it reveals that a Williamsian approach to human rights can shed light on important actually existing legal practices that address claims articulated on the basis of such rights.

Let us begin with the commonly accepted idea, within human rights practice, that how far in protecting individuals the interpretation of human rights should go in each specific context is a function of an internal differentiation between different kinds of human rights. Likewise, how far the implementation of a given human right should go, irrespective of internal differentiation between different kinds of rights, is also a live issue. From a doctrinal point of view, debates about the proper content to be attributed to human rights often take the form of a choice between ‘minimalism’ versus ‘maximalism’ in international human rights protection (Brems 2009).

Both of these kinds of distinction play out particularly clearly in the context of the Council of Europe human rights system of protection of which the ECHR is a part. To begin with, despite a certain amount of scholarly disagreement as to its proper significance, there seems to be a well-established practical distinction at the level of international human rights protection between so-called ‘first-generation’ civil and political rights and ‘second generation’ social and economic rights. Simplifying somewhat, we might say that the normative density of the latter is lesser than that of the former, where that means that the kinds of justiciability as well as the remedies provided for social and economic rights are typically weaker than those provided for civil and political rights. Thus, within the context of the Council of Europe, the European Social Charter, which protects rights such as the right to work or the right to organise and take part in collective action, does not recognise, as the ECHR does, a right to lodge individual applications for alleged human rights infringements. Instead, it only sets out a much weaker procedure of ‘collective complaints’, which can at most culminate in issuing a report outlining a State Party’s record of implementation of the examined rights (De Schutter 2010).

Likewise, human rights are commonly understood to belong to less and more stringent categories with regard to state duties, the most usual distinction being that between absolute and qualified rights. Thus, in the context of the ECHR, Articles 2, which protects the right to life, 3, which bans torture and inhuman and degrading treatment and 4, which bans slavery and forced labour, are unqualified or absolute rights. These are to be sharply distinguished from Articles 8–11 ECHR, which protect private and family life, the right to freedom of expression, the right to freedom of thought and religion and the right to freedom of assembly and association. Articles 8–11 ECHR set out qualified rights, which can typically be balanced against various legitimate state aims such as the protection of morals or public order and, conditional on a proportionality test, be suitably restricted (Tsarapatsanis 2015).

Though we do not have the space to present a full-blown defence of this claim here, we suggest that the distinctions drawn up above may be understood in a Williamsian way as functions of the varying importance of different human rights violations such that the closer we move to practices perceived as expressions of unmediated coercion, the more likely it is that fundamental questions of state legitimacy are to arise. Conversely, qualified human rights, such as the ones enshrined in Articles 8–11 ECHR, or social and economic rights of the kind we find in the European Social Charter can be understood as less stringent precisely because they less resemble the paradigm of unmediated coercion. Moreover, a wider range of sensible political answers to how they can be protected may exist, around which we ought to expect reasonable political disagreements to occur.

Apart from the distinction between civil/political and social/economic rights as well as that between absolute and qualified rights, in the context of the ECHR there are also, interestingly, a further number of ways in which we may attempt to use the proposed Williamsian reading to make sense of certain aspects of the ECtHR’s practice. A suitable starting point is to examine judicial doctrines such as the margin of appreciation and the so-called ‘consensus approach’. These are routinely used to assess claims of putative violation when it comes to qualified Convention rights (Tsarapatsanis 2015, 2018). The margin of appreciation doctrine, roughly, holds that states have leeway when it comes to balancing qualified ECHR rights against collective goods (i.e. ‘legitimate state aims’), especially when there is no ‘consensus’ among States Parties about where to draw the appropriate line (Dzehtsiarou 2015). Our claim is that, among other things, these doctrinal devices are used to mark the political acceptability of certain claims in ways that blend, along lines that Williams suggested, philosophical and political questions about human rights.

This happens in a number of different ways. First, political reasoning at the level of the ECHR involves the use of the doctrinal devices mentioned above and, generally speaking, amounts to driving a wedge between any kind of first-best moral understanding of the content of Convention rights and the right way to decide outcomes in each particular case. Following Williams, we might say that the extant forms of argument about the violations of qualified human rights at the ECHR level are not a function of ‘applying’ some antecedent form of (ideal) moral theory to cases, but, rather, of using a variety of historically available political materials to probe acceptable or sensible answers to specific political questions as well as to trace out a space of reasonable alternatives, where states have the possibility of choice. While the ECtHR’s answers will be found wanting in many cases if judged from a purely moral point of view, we understand Williams to be saying that it is this point of view that should be abandoned, and not the morally suboptimal ongoing practice of arguing about human rights. In a nutshell, the idea here is that the acceptance of moral suboptimality, at least from the point of view of ideal moral theory, is a rejection of political moralism when it comes to arguing about human rights violations where issues are not to do with the exercise of unmediated coercion and there is a wider range of sensible answers.

Moreover, Williams also warns us against overestimation of ‘theory’ when it comes to assessing human rights violations, at least if we understand ‘theory’ to refer to a well-ordered body of internally coherent propositions that can generate determinate answers to practical questions. Importantly, such qualms, which formed a crucial component of Williams’s critique of the ‘morality system’, were forcefully expressed in Ethics and the Limits of Philosophy (Williams 2006), and are not solely present in his posthumous political writings. Just as Williams claims that judgement is an inevitable component of good moral decision making, his work on human rights suggests that the exercise of distinctively political judgement on the part of judges is necessary if they are to skilfully reconcile demands flowing from substantive human rights protection (the ‘content’ of each specific right in question), with concerns regarding legitimacy, urgency and stability (and so on). The kind of judgement involved is contextual, prudential and more akin to a sense of practical propriety than to abstract, theoretical formulation. This kind of approach appears to cohere well with various aspects of the practice of the ECtHR, especially when a contextual judgement has to be made as to how best to reconcile considerations stemming from the need to protect a qualified right versus state sovereignty or democratic legitimacy, again mainly through the margin of appreciation and proportionality doctrines (Tsarapatsanis 2015, 2018).

Likewise, Williams urges us when thinking about distinctively political normativity to shift our emphasis from concepts such as ‘justice’ to ones such as ‘legitimacy’. Here again, his approach can be particularly helpful in endeavouring to make sense of a number of argumentative practices in the ECHR context. The starting point is the idea that the ECtHR is the ultimate interpreter of the ECHR. To that extent, the Court’s attributed meaning to ECHR rights is imposed on States Parties. Thus, the ECtHR exercises some form of interpretive power over Member States about the content of ECHR rights under conditions of reasonable disagreement. This raises issues of legitimacy of the ECtHR, of which judges seem acutely aware (Dzehtsiarou 2015). Under a broadly Williamsian understanding, there could thus be two different kinds of demands for legitimacy in supranational human rights contexts. First, human rights are devices whereby to assess the legitimacy of states insofar as the latter exercise coercion on a population within their territory. That much follows, as we have seen, from the basic legitimation demand. But also, second, the Court itself exercises power on states. As a result, it is asked of it by states to provide some account of its action, in order to render legitimate its own power.

The question whether, ultimately, such legitimacy concerns may justify departures from an optimal moral reading of human rights will depend, among other things, on whether the right in question is important enough to warrant such departures or not. Here as before, the distinction between absolute and qualified human rights resurfaces. Thus, whereas absolute human rights are typically immune to legitimacy concerns about the Court’s practice in the above second sense (and, in fact, can be readily understood as imposing legitimacy demands on states), qualified rights are typically balanced against such considerations, the doctrines of proportionality and margin of appreciation being key to how this occurs. Second, uncertainty about both the moral and empirical components of claims about rights violations brings to the normative fore considerations about risk and the cost of subsequent correction of potential errors (Tsarapatsanis 2018). The question here is typically whether the level of protection should be controversially raised straight away, or whether the Court may defer such a judgment. Third, concerns about legitimacy, as much as uncertainty and time pressure, can also sometimes justify examining the practices of states parties as well as non-parties to the ECHR with regard to the human rights issue at hand, typical of the consensus approach (Dzehtsiarou 2015). Fourth, the fact that judges have to take account of their strategic interaction with other judges, with whom they both disagree and have to cooperate at the supranational (these would be their peers, since judges typically decide on panels where other judges also sit) and domestic (where judges are expected at the very least to apply the settled and unambiguous case law of the ECtHR) levels, may also justify departures from their first-best moral conception about the content of ECHR rights, sometimes in the form of ‘compromises’. Fifth, more generally, ECHR rights are caught up in a dialectic that involves two ideas: that human rights are pre-political moral claims placing conditions on the legitimate exercise of coercion on individuals on the part of states and that human rights are products of political deliberation, negotiation and decision, where the democratic level of decision making about their initial specification is typically domestic.

Concomitantly, Williams also urges us to think about human rights in ways that bring the first political question, and thus considerations of stability, into sharper focus. The point here is that, since the first political question is never definitively resolved, considerations pertaining to sustainability of any historically concrete political framework within which human rights ‘make sense’ may compete with other kinds of considerations, such as those pertaining to the content of the rights themselves. We might put the argument with regard to the ECHR in the following form. Sustaining an efficient system of supranational human rights adjudication, as the ECHR aspires to be, requires states to mostly comply with the Court’s rulings. ‘Mostly’ should be understood here as a flexible and empirically open-ended term that denotes a range of positions between full compliance and general non-compliance. Whilst there are many factors that have been invoked to explain states’ compliance with systems such as the ECHR (Helfer and Slaughter 1997), arguably the perceived legitimacy of judgments is one of them (Dzehtsiarou 2015). Under these circumstances, the political argument pertaining to the sustainability of the ECHR framework appears to be straightforward: if there is a clear and present danger that states shall not comply with a judgment, especially when non-compliance as a response strategy seems to be endorsed by a considerable number of states, then there is a powerful reason to provide a morally suboptimal but sustainable, from the point of view of the stability of the ECHR system as a whole, response to an applicant.Footnote 6 We may thus put the argument in the form of a slogan by saying that the stability of the overall system of human rights protection can sometimes trump the finding of a violation in a specific case under the best ideal moral understanding of the right in question. Or, to put it another way, Williams’s approach helps us understand that the ECHR order (and, in fact, all legal orders pertaining to human rights protection) is not just a legal but also a political regime held together under conditions of conflict, which is sustained in part by ongoing efforts to avoid putting it under extreme pressure. Thus, in certain circumstances stability concerns may support moral suboptimality with regard to human rights protection. Here again, a distinctively political and realistic approach to thinking about human rights invites us to distinguish between rights that cannot be thus negotiated, because flouting them would amount to unmediated coercion, and rights that could. It is also clear from what we have said thus far that a Weberian ‘ethic of responsibility’ informs or should inform political judgement pertaining to the importance of legitimacy and stability considerations that are balanced against normative considerations about the content of human rights.

Finally, we want to flag a further claim: namely, that such a Williamsian approach to human rights can be readily distinguished from other kinds of approaches. In particular, such an approach contrasts with ‘moralism’ about human rights. We have already alluded earlier on that ‘moralism’ is a concept hard to pin down, but here we shall use it to refer to the idea that we may identify the content of human rights, as well as their application to particular cases, by sole recourse to moral (as opposed to political/prudential) reasoning and in a way that presupposes that the moral should systematically trump the political. In the ECHR context with which we have been concerned, such an approach is exemplified by authors such as George Letsas, who has claimed, following Dworkin’s ‘moral reading’ of the US federal constitution (Dworkin 1996), that the interpretation of the ECHR should be guided by a kind of comprehensive, systematic and ambitious moral theorising about the function and content of specific human rights (Letsas 2007). Whilst the moral reading of the ECHR has provided a robust critique of normatively less sophisticated doctrinal approaches, a Williamsian approach to human rights suggests that it may underplay the importance of the characteristically political questions that Court judges face, especially in hard cases. Thus, considerations to do with the abstract morality of human rights may well have a place in arguments about putative violations of the ECHR, but they compete with a host of other distinctively political factors, such as the considerations to do with legitimacy, sustainability and stability that we mentioned above.

Conclusion

By way of a conclusion, we want to say something in response to a likely objection that can be raised against this way of approaching these issues: that focusing on the kind of ‘political’ considerations we have outlined entails a worrying degree of conservativism.

It is certainly true that Williams is markedly more pessimistic about the ability of systematic theory to legislate for practice than many other moral and political philosophers are. He remarks that,

theory … can get ahead of ordinarily accepted practice. But there is no way in which theory can get all the way ahead of practice and reach the final determination of what can make sense in political thought; it cannot ever, in advance, determine very securely what direction might count as “ahead.” Very powerful political discourse can of course be proleptic and help to create the conditions it foresees. Liberal discourse itself has had considerable success in this, but it is a way that is markedly different from the ways in which liberalism typically sees itself. (Williams 2005, pp. 25–26)

As we have seen, his work on human rights elucidates one important way in which his anti-theoretical inclinations, and focus on judgement, contrast with more theoretically ambitious alternative approaches in political theory.

Yet as his discussion of Judith Shklar’s ‘Liberalism of Fear’ suggests, Williams’s constrained understanding of the proleptic potential of philosophical theory and focus on the centrality of the first political question, does not imply some kind of spectatorial political conservativism. Drawing on Shklar (1998), Williams stresses that ‘the particular arguments that carry forward liberal policies in particular situations must be not just practically but conceptually a matter of those circumstances’ (Williams 2005, p. 60). This leads to a bottom-up, anti-theoretical approach that treats different political (and legal) proposals ‘in light of what locally has been secured’ (Williams 2005, p. 61). In this respect, Williams holds that we have to take seriously the idea that politics is historically and contextually located and that any possible act of reform will begin from a specific point. We must try to build this recognition into our theorising in the way that top-down, morality-first theories often do not.

Some critics have worried that approaches like Shklar’s and Williams’s have little to tell us about what to do in stable constitutional democracies (Sangiovanni 2009). It is possible to imagine a related complaint being made against Williams’s approach to human rights. In particular, someone might object that it leaves us with little hope of thinking about human rights practice as a progressive political project. In the ECHR context, which was our main focus, this would play out as the idea that Williams’s approach would underwrite a practice by the ECtHR that could be too timid in the demands made on states parties.

Williams anticipates the objection that the liberalism of fear’s focus on ‘damage control’ has nothing to say to people who live in the politics of a ‘better-ordered society’ by retorting that the position can still play the important role of reminding people of ‘what they have got and how it might go away’. On our view, his focus on the core or basic human rights can, similarly, remind us of the precariousness of existing human rights practice and, in so doing, prompt us to think hard about how we might continue to ensure its ongoing stability and political relevance.Footnote 7 As we have seen, this is not just a theoretical, but also a practical problem that actors such as judges indeed have to face and resolve by exercising political judgement.

Moreover, Williams remarks that once the worst excesses of state power have been mitigated against, ‘then the attentions of the liberalism of fear will move to more sophisticated conceptions of freedom, and other forms of fear, other ways in which the asymmetries of power and powerlessness work to the disadvantage of the latter’ (Williams 2005, p. 60. For a similar thought see Williams 2005, pp. 7–8). This is why he argues that despite its ‘resolutely nonutopian’ character the liberalism of fear is not simply a politics of pessimism: ‘it can be, in good times, the politics of hope as well’ (Williams 2005, p. 61). In our judgement, there is no reason why this cannot also be true of his account of human rights.

Though Williams’s political realism undoubtedly leads to a largely preventive conception of politics (and law), this does not entail that those attracted to Williams’s way of approaching these issues cannot think in more ‘aspirational’ terms when the right conditions obtain. But if we are to think in this vein, Williams’s work offers the salutary reminder that we have to responsibly and seriously consider a host of issues that bear on political and legal questions, not only those of a peculiarly moral kind. It might also open conceptual space for resisting the urge to equate the practice of human rights with a morally optimal answer to a problem provided by experts (judges), thus helping to revitalise a more open democratic politics of human rights.