The Dworkin–Williams debate: Liberty, conceptual integrity, and tragic conflict in politics

Bernard Williams articulated his later political philosophy notably in response to Ronald Dworkin, who, striving for coherence or integrity among our political concepts, sought to immunize the concepts of liberty and equality against conflict. Williams, doubtful that we either could or should eliminate the conflict, resisted the pursuit of conceptual integrity. Here, I reconstruct this Dworkin–Williams debate with an eye to drawing out ideas of ongoing philosophical and political importance. The debate not only exemplifies Williams’s political realism and its connection to his critique of the morality system. It also illustrates the virtues and hazards of contemporary efforts to ameliorate or engineer our concepts; it indicates what political philosophy might look to in appraising political concepts; it adverts to the different needs these concepts have to meet if they are to sustain a politics of pluralism, deal with polarization, and secure the consent of those who end up on the losing side of political decisions; and it presents us with two starkly contrasting conceptions of politics itself, of the place of political values within it, and of our prospects


INTRODUCTION
In the fall of 1998, a year after the death of Isaiah Berlin, the New York Institute for the Humanities convened a two-day conference in Manhattan to examine his intellectual legacy.The conference attracted an unexpectedly large audience and was covered by The New York Times. 1 It was also the scene of a coruscating debate between Ronald Dworkin and Bernard Williams, itself the culmination of a dialogue going back to seminars they held together at Oxford in the late eighties (as part of a series informally known as 'Star Wars'). 2 As Damian Cueni (manuscript-a) points out in an important article I build on here, once their compressed remarks in New York are connected to the scattered writings in which they elaborated their points and engaged with each other's work, a surprisingly focused debate emerges. 3My immediate aim is to reconstruct this debate and show that it is rich and protracted enough to merit being labelled 'the Dworkin-Williams debate'.
What is ostensibly at issue in this debate is how we ought to deal with the conflict that Berlin perceived between the values of liberty and equality.Dworkin, striving for integrity among our concepts, seeks to eliminate this conflict, while Williams, doubtful that we either could or should eliminate the conflict, resists the pursuit of conceptual integrity.
Yet upon closer inspection, the debate turns out to be about far more than Berlin's familiar pluralist thesis that values conflict.Dworkin and Williams personify not just two contrasting approaches to the conflict between the concepts of liberty and equality, but two rival views of what our political concepts should be answerable to, what the constraints on those who would reinterpret our concepts for coherence should be, and how we should conceive of politics itself.Once these broader ideas are drawn out, the confrontation with Dworkin emerges as formative 1 Such was the demand for transcripts afterwards that the organizers were persuaded, against their original plans, to publish the presented papers together with the subsequent discussions.Aileen Kelly, Steven Lukes, Mark Lilla, Robert Silvers, Ronald Dworkin, Bernard Williams, Thomas Nagel, Charles Taylor, Avishai Margalit, Richard Wollheim, and Michael Walzer all presented papers, while Frances Kamm, Richard Bernstein, Fritz Stern, Neil Rudenstine and others contributed to the discussions.See Rothstein (1998) as well as Lilla, Dworkin, and Silvers (2001) for an account of the conference and its reception. 2See Guest (2013: 17, 247n20). 3Cueni reconstructs and explores the implications of the Dworkin-Williams debate from a jurisprudential perspective in a series of texts (manuscript-a, manuscript-b, manuscript-c, manuscript-d) that inform the present article along with our many conversations on these issues.When we first set out to reconstruct this debate, however, we had to start from scratch, as Williams's challenging essays engaging with Dworkin had been virtually ignored by commentators.This is now beginning to change.Hall (2017), Ulaş (2020), and Kyritsis (2021) touch on the Dworkin-Williams debate in passing, while Murata (2022aMurata ( , 2022b) ) draws on Williams's writings on Dworkin in some depth, albeit in a somewhat different connection.I offered a first sketch of the debate's dialectics in Queloz (2022bQueloz ( : 1252-53)-53).For congenial applications of some of the lessons it holds to legal reasoning, see now also van Domselaar (2022) and Murata (manuscript).
19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License for Williams's later political philosophy: some of his distinctive ideas take their most distinct form in his responses to Dworkin, making the debate not merely a secondary application of Williams's fragmentary and elusive political philosophy, but a key to it, which elucidates his 'political realism' and its connection to his critique of 'the morality system'.
My ultimate aim in reconstructing this debate from Williams's far less well understood perspective is not just to fill a gap in intellectual history, however, but to draw out and develop ideas of ongoing philosophical and political importance.The Dworkin-Williams debate is instructive for contemporary efforts to ameliorate or engineer our concepts.It exemplifies the virtues and hazards of such efforts, warning us that alleviating tensions between our concepts comes at a cost.The debate also indicates what political philosophy should look to in appraising political concepts; it illustrates how to devise the conceptual tools we need to make sense of political disagreement, and in particular to deal with polarization and secure the consent of those who end up on the losing side of political decisions.And the debate presents us with two starkly contrasting conceptions of politics itself, of the place of political values within it, and of our prospects of reducing conflicts between political values through philosophy.
I proceed as follows: after briefly describing Dworkin's position and his reasons for pursuing conceptual integrity ( §2), I show that Williams fundamentally articulates two worries about Dworkin's project: a pluralism-based worry about the attainability of conceptual integrity ( §3); and a concern-based worry about the advisability of striving for conceptual integrity ( §4).I then bring out why Williams thought that the concerns animating our use of the concept of liberty could act as a guide to the concept of liberty we now need ( §5), before reconstructing the two positive reasons he gave for preferring his conception of liberty over Dworkin's: that those on the losing side of a rightful political decision still need a concept capable of acting as the political sharp end of the concern for primitive freedom ( §6); and that those on the winning side need a conception that equips them to acknowledge the costs in liberty incurred by those on the losing side if they are to secure losers' consent, cultivate healthy relations of fellow citizenship, and foster respect across the aisle ( §7).Finally, after exploring how Dworkin might respond to narrow the divide between him and Williams, I show that the debate ultimately turns on how to realistically conceive of politics and its relation to high-minded political values ( §8).

DWORKIN'S PURSUIT OF CONCEPTUAL INTEGRITY
The starting point of the Dworkin-Williams debate is the observation, stressed notably by Isaiah Berlin, that the values of liberty and equality conflict.As Berlin describes the tension, 'total liberty for the wolves is death to the lambs, total liberty of the powerful, the gifted, is not compatible with the rights to a decent existence of the weak and the less gifted ' (2013d: 12-13).Conversely, the thorough enforcement of total equality carries severe costs in liberty, which has to be curtailed to redress or prevent various forms of inequality arising from disparities in wealth, resources, opportunities, and talents.Circumstances concurring, modest gains in both liberty and equality may be achievable without trade-off; but when the realization of either value is pursued more insistently, there comes a point at which one has to be paid for with the other.'It is an uncomfortable situation', Williams concludes already in his influential early essay on the idea of equality, 'but the discomfort is just that of genuine political thought ' (1973b: 249).Dworkin believes we can do better.He sees reason to hope that we might eliminate the tension between the concepts of liberty and equality.His leverage point is the idea that whether QUELOZ liberty and equality conflict 'depends on how we conceive these abstract values ' (2001a: 83). 4  This seemingly trite observation marks a crucial step: it transposes what might otherwise have been a metaphysical debate about the nature of liberty and equality into a debate about what has recently come to be known as conceptual ethics: the philosophical reflection on which concepts or conceptions we should use. 5In the sense at issue here, conceptual ethics is the ethics of concept possession: the question of which concepts we should use is the question of which conceptions or conceptualizations should be operative in our actual thinking processes.
As long as liberty is conceived as 'freedom from the interference of others in doing whatever it is that you might wish to do' (2001a: 84), Dworkin admits, liberty undoubtedly conflicts with equality.But why should we think that we are committed to this way of conceptualizing liberty?The question cannot simply be answered by an inquiry into what liberty really is, he observes-we 'can't conduct a DNA analysis of liberty ' (2001a: 86).The question has to be which conception of liberty we have most reason to accept.And perhaps the most attractive conception of liberty will turn out not to conflict with the most attractive conception of equality after all.
For Dworkin, we have several reasons to prefer conceptions that do not conflict.What makes the situation as Berlin describes it unacceptable is that it systematically confronts the state with dramatic or tragic choices-choices where, as Dworkin puts it, 'the government must not merely disappoint but must wrong some citizens no matter what it does' (Dworkin et al., 2001: 122). 6 Moreover, Dworkin also regards the fact that two conceptions do not conflict as being in itself already a reason to prefer them over conceptions that conflict: 'integrity among our concepts is itself a value', he maintains, 'so that we have that standing reason for seeking out, for preferring, conceptions of our values that do not conflict' (Dworkin et al., 2001: 127).Other things being equal, we have a standing reason to pursue what he calls conceptual 'integrity'.Conceptual integrity amounts to a form of practical coherence between value concepts that we might call congruence, where two concepts are congruent if and to the extent that the realization or instantiation of one concept does not come at the expense of the realization or instantiation of the other concept.
Dworkin therefore proposes to iron out the conceptual tension between liberty and equality by composing a definition of the concept of liberty that renders it systematically congruent with the concept of equality.If Berlin reached the conclusion that liberty and equality irremediably conflict, Dworkin believes, it was because Berlin equated liberty with freedom from interference in doing what one wants. 7In fact, however, liberty should be interpreted as a political rather than personal value: it is 'that part of your freedom that government would do wrong to constrain' (2011: 4). 8As a political value, liberty should not be understood as freedom from interference in doing what one wants, but in terms of rights distributed according to a political principle of equalityin other words, liberty should be rightful freedom. 9That does not presuppose a right to freedom; Dworkin argues instead 'for rights to liberty that rest on different bases' (2011: 4), such as rights to ethical independence, to free speech, and to due process of law. 10 4 In Dworkin's terminology, they are interpretive concepts; see Dworkin (1986: 45-86;2001b). 5The phrase 'conceptual ethics' hails from Burgess andPlunkett (2013a, 2013b). 6See also Dworkin (2000: 120-83;2001a: 80). 7See Dworkin (2011: 367). 8As Dworkin also puts it: someone's liberty is 'the area of his freedom that a political community cannot take away without injuring him in a special way: compromising his dignity by denying him equal concern or an essential feature of responsibility for his own life' (2011: 366). 9See Dworkin (2000: 120-83;2001a, 2001b).
10 For Dworkin's elaboration of what these rights amount to, see Dworkin (2011: 368-74).If we conceive of the political value of liberty in terms of equally distributed rights to liberty, this 'rules out genuine conflict with the conception of equality . . .because the two conceptions are thoroughly integrated' (Dworkin, 2011: 4).The realization of equality may of course still entrain a loss in freedom.But not every loss in freedom will be a loss in liberty.A loss in freedom will only count as a loss in liberty where there is a claim to liberty, and there can only be a claim to liberty, on Dworkin's account, where that claim can be grounded in a right.It follows that as long as rights are equally distributed, liberty must itself be equally distributed, and liberty and equality can no longer conflict.As a result, '[t]he alleged conflict between liberty and equality disappears' (Dworkin, 2011: 4).This yields an extremely neat and tidy account that immunizes the concepts of liberty and equality against conflict.It achieves this by defining the concept of liberty in terms that effectively guarantee its congruence with the concept of equality.

A PLURALISM-BASED WORRY ABOUT THE ATTAINABILITY OF CONCEPTUAL INTEGRITY
Williams, however, has strong reservations about the pursuit of conceptual 'integrity'-especially in the case of the political value concepts that Dworkin is keenest to reconcile.While Dworkin encourages us to aim for conceptual integrity and just see whether we can or cannot construct concepts that achieve it, 11 Williams does not share Dworkin's hope that there might be an adequate but entirely tensionless way of thinking about liberty and equality waiting to be discovered in this way.
Williams's scepticism towards the attempt to achieve conceptual integrity is grounded, at the most immediate and still somewhat question-begging level, in his acceptance of Berlin's value pluralism: the thesis that there is a plurality of irreducibly distinct and incommensurable values that are bound to end up pulling in competing directions when pursued in concert, not merely because time is short or the world recalcitrant, but because the values themselves inherently conflict. 12As Berlin puts it in his best-known essay: 'We are faced with choices between ends equally ultimate, and claims equally absolute, the realisation of some of which must inevitably involve the sacrifice of others', which is why 'the possibility of conflict-and of tragedy-can never wholly be eliminated from human life, either personal or social ' (2002b: 213-214).
Some pluralists, including notably Rawls, have focused on pluralism at the level of society, where the values of some members of society clash with the values of other members of society. 13ut what Berlin and Williams emphasize is that even if society were far less pluralistic, so that members of society shared roughly the same range of values, there would still be pluralism 'within the breast': even the concepts of one and the same person conflict in ways that are not resolvable without loss.14A value conflict is not 'most typically enacted by a body of single-minded 11 See Dworkin et al. (2001: 127). 12Here I rely particularly on Williams's elaboration of pluralism.On Berlin's pluralism and his influence on Williams in this respect, see Lyons (2021: 215-61), and see also Gray (2013) and Lyons (2020) for synoptic accounts of Berlin's thought.On the development of pluralist ideas between 1940 and 1980 and its connection to anti-totalitarianism, see Müller (2012).For an exploration of pluralism's implications for politics, see also Galston (2002Galston ( , 2005)).
egalitarians confronting a body of equally single-minded libertarians', Williams notes, 'but is rather a conflict which one person, equipped with a more generous range of human values, could find enacted in himself ' (1981a: 73).Of course, social pluralism and pluralism within the breast are not unrelated, since the latter is in many instances an expression of the former: the conflict enacted within one person may itself reflect the fact that the person's conceptual repertoire is the accumulated historical deposit of different social influences-be it different groups within a society, different stages in the history of that society, or even different societies.
The cardinal claim of pluralism, however, is that these values, at whichever level they are expressed, are such that all the things they pick out as being of value cannot 'ultimately be united into a harmonious whole without loss' (Williams, 2013: xxxv).The ideal situation in which, as P. F. Strawson put it, 'every god is given his due and conflict is avoided by careful arrangement and proper subordination of part to part' (2008: 30) is a fantasy, and an incoherent one at that.For the pluralist thesis is not just the weak claim 'that in an imperfect world not all the things we recognise as good are in practice compatible', but the much stronger claim, which is a claim about our concepts before it is a claim about the world in which they are deployed, 'that we have no coherent conception of a world without loss, that goods conflict by their very nature' (Williams, 2013: xxxv).Our value concepts are incongruent, and while the incongruence may be accidental in some cases, it is non-accidental in others, and hence ineliminable as long as we hold on to anything like these values.
But must we hold on to anything like these inevitably conflicting values?This is the question raised by the radically revisionist conceptual ethics of someone like Nietzsche, and, in a less radical spirit, it is also Dworkin's response to this pluralism-based worry.Dworkin does not deny that the conceptions held up by pluralists in fact conflict; what he questions is the authority of those conceptions.Pluralism, Dworkin complains, 'is too often cited as a kind of excuse for not confronting the most fundamental substantive issues', in particular 'the hard work of actually trying to identify the right conceptions of the values in question' (Dworkin et al., 2001: 124-5): the argument necessary to defend pluralism . . .must show, in the case of each of the values it takes to be in some kind of conceptual conflict with one another, why the understanding of that value that produces the conflict is the most appropriate one.(2001a: 90; see also 2006: 116) Values are something that we construct, Dworkin reminds us, and finding the most attractive conceptions is not a matter of 'excavating the shared meanings of words' or of making 'anything like a scientific discovery about the true nature of reality' (Dworkin et al., 2001: 126).
As Dworkin's claim that we should identify the most attractive conception of the values we have already brings out, however, his approach is much less radically revisionist than that of someone like Nietzsche: Dworkin does not go so far as to call into question the very use of anything like liberal and egalitarian concepts. 15Instead, he deploys the concept-conception distinction as a distinction between different levels of abstraction within the concepts already in use: a concept, for Dworkin, is abstractly characterizable in terms of what different interpretations of a value tend to share, while 'the controversy latent in this abstraction is identified and taken up' at a 48-50; 2013d: 14), and to warn against the danger of allowing this utopian ideal to determine one's moral and political practice (2002b: 212-17); but logically, these political conclusions lie downstream of a structural claim about human value concepts.
more fine-grained level by distinguishing competing conceptions of the value.This yields what Dworkin envisions as a 'treelike structure', the trunk being the concept-what people 'by and large agree about'-and the branches being the conceptions-the 'more concrete refinements' of that concept with respect to which people differ (1986: 70-71).Dworkin offers courtesy as an example: people might agree, at an abstract level, that courtesy is a matter of respect, while at the same time disagreeing over what exactly that form of respect requires.
For Dworkin, then, the question is not whether the concepts we inherited conflict, but whether the most attractive conceptions that we could be using do.He does not question our continued use of the concepts of liberty and equality.But he is less impressed by the constraints that existing ways of thinking exert on us than by the degree of freedom they leave us to revise our conceptions of our values.Even if pluralists are rights about the conceptual apparatus we inherited, we remain free to try and construct values that do not conflict, and it would be premature to accept the persistent possibility of conflict before every effort had been made to forestall it.As he insists: 'We shouldn't buy failure in advance: we should aim at integrity in an optimistic spirit' (Dworkin et al., 2001: 127).Yet Williams's pessimism with regard to our prospects of arriving at such conceptual integrity stems, at a deeper level, from the expectation that any viable set of values will include at least some values that necessarily conflict.Like Berlin, he thinks that there are some value concepts that human societies are bound to cultivate in some form-not because these are given to us by God or implanted in us by nature, nor because there is some antecedent structure in reality that these concepts are bound to reflect, but because the conjunction of certain facts about human beings and their environment systematically gives rise to the same kinds of practical problems calling for the same kinds of solutions, which notably include the development of the same kinds of concepts.
While Hobbes and Hume may be the best-known exponents of the idea that certain concepts are rendered necessary by contingent facts about us and the world we live in, 16 the same idea is still very much alive in the twentieth century. 17It underpins Berlin's conviction that values vary only within the bounds of a 'human horizon', i.e. that there is a 'minimum of moral values accepted by all men without which human societies would disintegrate ' (2015: 206); 'these values', Berlin asserts, 'are objective-that is to say, their nature, the pursuit of them, is part of what it is to be a human being, and this is an objective given' (2013c: 12). 18In this respect, Berlin self-consciously echoes his close friend Herbert Hart, who, in The Concept of Law, suggests that we can make certain generalizations about what kinds of practical needs will arise in the kinds of environments that humans inhabit, and 'as long as these hold good, there are certain rules of conduct which any social organization must contain if it is to be viable ' (2012: 192-93).In a similar vein, Stuart Hampshire remarks that if 'the underlying structure of moral distinctions has no supernatural source, it must be recognized by rational inquiry as having its origin . . . in constant human needs and interests ' (1983: 128).We can then look for bounds on the variability of moral, political, and legal concepts, imposed by practical demands faced by human beings nearly everywhere in virtue of their sharing certain very basic concerns, such as the concern to avoid violent conflict with others, to find out about the dangers and affordances of their environment, to secure the resources they need to survive, and to foster conditions enabling cooperation.Williams himself puts a version of 16 Pettit (2008), Cohon (2008), Sagar (2018), and Queloz (2020, 2023) notably foreground this aspect. 17See Hall (2020) for a discussion of this idea in the work of Berlin, Williams, and Hampshire.Other exponents of this idea include Edward Craig (1990Craig ( , 2007) ) and Miranda Fricker (2007, forthcoming), as I have argued in detail in Queloz (2021b). 18For discussions of this aspect of Berlin's work, see Müller (2019) and Riley (2019).this idea to work in Truth and Truthfulness (2002: 126) to argue that any human society needs to value the truth for its own sake in order to effectively gain and share information. 19 When the pluralist claim that some value concepts inevitably conflict is combined with the further claim that some of these concepts are ones that human societies cannot do without, the two claims form an argument for why the quest for conceptual integrity is unlikely to succeed across the board.Berlin pithily alludes to both claims in his contention that 'collisions of values are of the essence of what they are and what we are ' (2013d: 13).
Of course, the impossibility of conceptual integrity across the board does not by itself preclude its attainability for the particular pair of concepts Dworkin focuses on.But it forms an important part of the background against which Williams approaches this debate, and at least disposes one towards scepticism-as does the fact that, to pluralists steeped in history like Berlin and Williams, Dworkin's attempt to inoculate the concepts of liberty and equality against conflict is immediately recognizable as another manifestation of a familiar human urge: the deep-seated desire to eliminate tragic value conflicts, i.e. painful conflicts between what is right and what is equally right that cannot be resolved without loss, so that one is bound to do wrong no matter what one does, because there is 'no better thing to be done' (Williams, 1973a: 173). 20It is an ancient aspiration to use theorized conceptual edifices as shelters against luck, techne as a remedy to tyche. 21 Various theoretical constructions-from Stoicism through Kantianism to utilitarianism-have been erected over the ages to rationalize away value conflicts when they arose, or to prevent them from arising in the first place by devising lexical priority rules, reducing values to a common currency of comparison, or arguing that conflicts evaporate under reflection once some obligations are revealed to be merely apparent. 22A prime example is the Kantian doctrine that anything which is not a claim of morality must be a claim of 'prudence', in a specially capacious sense of the word, and is silenced when it conflicts with the claims of morality. 23Williams regards this as an all too human stratagem for reducing the risk of facing tragic choices between the claims of morality and the claims of other things one deeply cares about. 24By consigning all these competing reasons for action to the category of the merely prudential and convincing oneself of the overriding importance of morality, one seems to escape such conflicts with a clear conscience.
But this promise of protection from tragic conflict is ultimately illusory, Williams believes, because it turns on the distorting pretence that all the claims competing with morality must be claims of self-interest, and belittles even those by treating them all alike: 'How does "morality" deal with the many reasons for behaving badly that lie in the desire to be loved?As another of its "temptations", no doubt, like a craving for marmalade' (Williams, 2014a: 246). 19See Queloz (2018) for a detailed reconstruction. 20This conception of tragedy as a conflict of right and right is associated with Hegel, who writes in his Aesthetics that the 'original essence of tragedy' consists in a conflict in which 'each of the opposed sides, if taken by itself, has justification; while each can establish the true and positive content of its own aim and character only by denying and infringing the equally justified power of the other,' thereby becoming 'nevertheless involved in guilt ' (1975' ( : II, 1196)).See Robert Williams (2012: 120-42) for a discussion of this conception of tragedy, and Magee and Williams (1971: 162-165) for a discussion of its relevance to philosophy.For accounts of tragic legal choices, see Wolcher (2008) and especially van Domselaar (2017), who draws on Williams.
21 See Berlin (2013b: 196;2013c: 26-28;2014a: 25;2014b: 99-100) and Williams (1981b: 20;1995;2011: ch. 1). 22See Berlin (2002a: 291-292;2014b: 61-62, 70-72). 23For an analysis of this Kantian idea, see Bader (2015). 24As I argue in Queloz (2022aQueloz ( , 2022c, forthcoming), forthcoming).In the discussion that ensues in New York, Dworkin concedes that 'dramatic, even tragic conflicts in personal values' (Dworkin et al., 2001: 132) may be unavoidable, so that one will do wrong whatever one does.His hope, he specifies, is only that when it comes to the political values guiding state action, we can 'construct a scheme of values that strike us as right but that do not entail that unpleasant conclusion' (Dworkin et al., 2001: 132).As Dworkin sees it, pluralist scepticism towards the general attainability of conceptual integrity does not yet offer conclusive reasons to rule out its attainability in the specific case of liberty and equality; nor does it suffice to show that what holds for personal moral values guiding individual action must extend also to political values guiding state action.

A CONCERN-BASED WORRY ABOUT THE ADVISABILITY OF CONCEPTUAL INTEGRITY
Alongside this pluralism-based objection to Dworkin's project, Williams therefore mounts a second objection that leads him to be critical of the very ambition to achieve conceptual integrity in this particular case.
The main thrust of this second objection is that we discover a real need to keep open the rift between the concepts of liberty and equality that Dworkin offers to patch up with his tidy conceptual construction once we reflect on the central concerns that render the concepts of liberty and equality important for us in the first place: what is it that we fundamentally care about in these connections that leads us to have a use for anything like these concepts?Our concepts may channel our attention and shape our sense of saliency in various ways.But the merits of a proposed concept ultimately have to be judged on the basis of a prior understanding of the concerns-the needs, interests, desires, projects, aims, and aspirations-that this concept is to help us to meet.Though our concerns may be mediated and focused by our concepts, it is fundamentally the direction of those concerns that determines what we care about, what is important or relevant to us (the Latin concernere means 'be relevant to', from con-, which expresses intensive force, and cernere, which means 'sift, discern').Indeed, Williams suggests that we only really grasp why liberty and equality are two different concepts, and not just two words for the same concept, once we relate them to the concerns underlying them. 25 Of course, there are many concerns that concepts such as liberty and equality tie in with; but Williams suggests that we can to some extent cut through that complexity, because 'associated with each such value concept there is a kind of schema, a very bare outline of what our central concern is ' (2001: 92).In the case of the concept of equality, for instance, a basic human concern that might lie at its root is the concern to receive what one is due-at least, this is how Cueni (manuscript-a) helpfully glosses Williams's all too brief discussion of the concern at the root of the concept of equality. 26But our use of the concept of liberty is animated by a different and utterly basic concern, namely the universal human concern to be unobstructed in doing what one wants-in particular, unobstructed by humanly imposed coercion.He labels this the concern for 25 See Williams (2001: 92). 26In the New York debate, Williams shifts from speaking of a concern for 'equality' to speaking more broadly of a concern for 'justice', and then suggests that the concern at the root of the notion of justice is that of 'giving each person what he or she should have, or some similar idea ' (2001: 92).But his more detailed responses to Dworkin-i.e.Williams (2005aWilliams ( , 2005b))-are cast in terms of 'liberty' and 'equality'.For a detailed account of the concerns that underlie and drive the development of our notion of equality, see Sagar (forthcoming).
'primitive freedom'.'Why should human beings in general be concerned with some value of that form?', Williams asks, and answers his own question with another: 'What view would one have to take of one's desires and projects and other values if there were never even a question of its being something to be resented and resisted if others aimed to frustrate them?' (2005b: 93).His point is that anything recognizable as human agency is bound to involve some concern along the lines of the concern for primitive freedom, because pursuing any concerns at all must already involve being concerned to be unobstructed by others in doing so.A human agent who pursued concerns without caring at all about whether they were frustrated by others simply would not be intelligible to us as genuinely pursuing those concerns.
By itself, of course, that schema is 'too bare, indeed too schematic', Williams acknowledges; it 'needs, and will have indeed received, an associated social, historical, and cultural elaboration ' (2001: 93). 27For a start, this concern for primitive freedom is not yet a political concern; it is a personal concern for freedom in action, antecedent to and intelligible independently of the advent of the political.But as we shall see in §6, it is through the elaboration of a political concept of liberty that this pre-political concern for freedom finds a suitable political expression.
Hence, what we really want to know, and what Dworkin neglects in his pursuit of conceptual integrity, is to what extent a given elaboration of a concept serves, or fails to serve, the concerns we now have.For whatever exactly the concerns that basically go with the values of liberty and equality now go to, we cannot redirect those concerns 'simply nominalistically, by redefining a word', Williams insists, because 'an interest in producing a more coherent body of law is not by itself going to stop the concern going to what the concern goes to ' (2001: 94).If Dworkin's proposed conceptions fail to tie in with the concerns that give us reason to think in terms of anything like liberty and equality in the first place, we have reason not to adopt those conceptions, because they would deflect attention away from the satisfaction of our most basic concerns in these connections.It is simply no good securing conceptual integrity between two conceptions if it comes at the cost of severing the ties to the central human concerns that led us to have a use for anything like these concepts in the first place.
This concern-based objection yields an argument that promises to do more with less than the pluralism-based objection: without requiring sweeping assumptions about what values societies are bound to possess, it calls into question the very ambition to iron out certain tensions in our conceptual apparatus.It suggests that it is a fundamental mistake to look only at conflicts between concepts.We have to look also at the human concerns underlying the concepts-at what we care about in these connections.These concerns are in significant respects independent of our concepts, so that constructing more congruent, coherent, or otherwise more theoretically virtuous concepts is not by itself going to redirect the concerns underlying the concepts we inherited.And if it is by tying in with these concerns that the concepts of liberty and equality become concepts worth using in the first place, we are clearly not helped by securing congruence or 'integrity' between these concepts if they thereby lose their connection to the concerns animating their use.
Williams's second worry is thus a concern-based worry about the advisability of striving for conceptual integrity.Even if we could achieve total integrity or congruence between our concepts, we would be ill-advised to do so if it untethered our concepts from our concerns.This suggests that eliminating tensions and cultivating theoretical virtues in our concepts is not necessarily advisable.Conflicting concepts can be the better concepts in virtue of serving our concerns better.
All that Williams then has to show against Dworkin is that the concept of liberty we need by virtue of our concerns is one that conflicts with the concept of equality.

THE CONCEPT OF LIBERTY WE NEED
In effect, Williams observes, Dworkin's conception of liberty is a 'juridical conception' (2005b: 86): it is juridical in conceptualizing liberty as extending only as far as claims in liberty are granted or denied by an agreed authority on a shared interpretative basis, because the reasons that might intelligibly ground claims in liberty are exhausted by one's rights under the political system (Williams, 2005b: 86). 28As Cueni (manuscript-a) shows in a nuanced evaluation of this charge, such a juridical conception draws what appeal it possesses from its exclusive preoccupation with how a public decision-maker should regulate the relationships between citizens on the basis of a shared, principled, and consistent framework.But political debate, Williams protests, has far less in common with judicial review than Dworkin's juridical conception of liberty suggests: 'We and our political opponents-even our opponents in one polity, let alone those in others-are not just trying to read one text ' (2005b: 78).We have different concerns and hold different outlooks articulated in terms of wildly varying conceptions that yield equally varying conclusions as to how values ought to be weighed against each other. 29f, in reflecting on the political value of liberty, we truly acknowledge this, and ask, as Williams puts it, 'what we want that value to do for us-what we, now, need it to be in shaping our own institutions and practices [and] in disagreeing with those who want to shape them differently' (2005b: 75), it will emerge that what is needed in politics is not a juridical, but a more thoroughly political concept of liberty.A 'thoroughly political concept of liberty', for Williams, is one that 'acknowledges in its construction the on-going existence of political conflict ' (2005a: 126).
As these programmatic remarks make clear, Williams's guiding question is what we, now, need the concept of liberty to be.But what does it mean for a concept to be 'thoroughly political', or 'to acknowledge in its construction the on-going existence of political conflict'?And how does this amount to an argument to the effect that the concept of liberty should conflict with the concept of equality?
Williams's starting point, we saw, is the utterly basic human concern to be unobstructed by humanly imposed coercion in doing what one wants: the pre-political concern for what he calls 'primitive freedom'.To understand how a given individual's concern for freedom in action relates to the political value of liberty, one has to consider how such a personal concern for primitive freedom might relate to the perspective of some public authority that can be appealed to by the individual (Williams, 2005b: 83).As Williams indicates, the way in which freedom is conceptualized from the point of view of that authority will have to differ from the way it is conceptualized from the individual's point of view: 'the resolution of questions of how far a person's freedom should be protected or extended, how far it is good that it should be, how far he has a right that it should be, requires some degree of impartiality (a general point of view, in Hume's phrase) which is not contained in the idea of an individual's primitive freedom as such ' (2005b: 84).From such a political as opposed to personal point of view, the salient question is how the often competing concerns of different people living under a shared public authority are to be registered and dealt with.
What makes liberty a political concept is that it conceptualizes primitive freedom for this political point of view: it acts as a political lens through which the concern for primitive freedom can be focused in a political context.That, at the most basic level, is the role of the concept: it picks out that part of individual freedom that has a claim to society's attention.
To stand any chance of performing this role, however, the concept of liberty must differ substantially from how individuals might conceptualize their own primitive freedom.For one thing, the concept of liberty must be a normatively richer notion: while primitive freedom is basically a form of power that one can merely get, using the power one already has, liberty has to be something that one can lay a political claim to, which is to say a claim that makes a claim on society's attention (Williams, 2005a: 115).The mere fact that an individual has lost some freedom does not yet give that individual any claim on society's attention.Such a claim must be backed or grounded by something other than the power one already has.This is, of course, a point that Dworkin himself acknowledges, since he likewise insists that political claims to liberty must be grounded in something (namely in rights, on Dworkin's account).
Because of its political nature, moreover, the concept of liberty must also be more narrowly focused than the concept of freedom: not every loss in freedom can count as a loss in liberty, as we also saw Dworkin point out against Berlin.The reason is that, as a political value, the concept of liberty has to be able to 'co-exist with the political' (2005a: 120), as Williams puts it: the concept could not intelligibly give everyone a claim to doing whatever they happened to want-to murder whomever they wished, for instance-because that would undermine the conditions necessary to there being any political order organized by political values in the first place. 30Preventing people from murdering whomever they want is a restriction on their primitive freedom, but not one that could consistently count as a restriction on their liberty, because that restriction is necessary for there to be any kind of political order at all. 31  Someone's claim that they have incurred a loss in liberty therefore minimally needs to be socially presentable, as Williams put it, where that means that 'it can be urged consistently with accepting a legitimate political order for the general regulation of the society' (2005a: 120).Objecting already to the mere fact of being subject to a state at all would not be socially presentable in this sense, since it is a complaint that would apply to any state whatsoever, and the mere existence of a political order cannot consistently be understood as constituting, already in itself, a ground for a political complaint within that order.By contrast, objecting to the operations of Franco, or James II, would be socially presentable, because 'one could, and most objectors did, accept that these rulers should be replaced by some other rulers, and more generally they accepted a state system' (Williams, 2005a: 120).
This social presentability requirement constitutes a necessary condition on a claim of a loss in liberty being correct.But Williams and Dworkin agree that it is not yet sufficient.What further condition does such a claim have to meet in order to be correct? 30See Williams (2005b: 83-85;2009: 200). 31Resentment is the prototypical reaction to restrictions of one's liberty, on Williams's account, but he acknowledges that the feelings that go with the sense that one's liberty is being restricted do not necessarily have to be identified with resentment, because 'resentment so readily merges into other negative feelings, such as anger and dislike, not just for conceptual but also for various familiar psychological reasons' (2005b: 87).Dworkin's suggestion is that a claim in liberty needs to be rightful, i.e. grounded in rights.Under this conception, the reasons that can ground a claim in liberty are exhausted by one's rights under the political system.But insofar as one aims at congruent conceptions of liberty and equality as implying rights, one is committed to looking for conceptions that do not imply conflicting rights.Insofar as the state rightfully exercises its power in curtailing one's freedom, therefore, one can have no basis for a claim in liberty in that connection, since whatever the state thereby stops one from doing is something one has no right to do anyway.An example that Dworkin and Williams both discuss is the abolition of private schools in the name of equality of opportunity in education.Williams envisages a government that 'takes steps to make it illegal or effectively impossible, except for some fairly trivial exceptions, to run a private school ' (2001: 100).It is open to Dworkin to resist the suggestion that such a restriction would be rightful under anything like our present circumstances, Williams concedes.But Dworkin is nevertheless committed to the view that if this crackdown on private schools, or some other restriction like it, were to be rightful, then it could not come at the expense of liberty, since there could be no real tension between liberty and equality.
On Williams's view, however, this tidy picture is too tidy to make sense of the experience of life under a political order.For it is 'one datum of that experience', Williams stresses, 'that people can even recognize a restriction as rightful under some political value such as equality or justice, and nevertheless regard it as a restriction on liberty ' (2005b: 84).That is to say, even those who agree that a crackdown on private schools is rightful may still feel resentful of it; and it is paradigmatically through such feelings as the experience of resentment that people's sense of freedom is given to them, and by extension also their sense of when their liberty is being restricted (2005a: 123; 2005b: 87-88).To make political sense of this experience of resentment, i.e. to conceptualize it as reasonable even when articulated in terms of liberty rather than freedom, one needs to be able to see it as reflecting some real loss or cost in liberty.But Dworkin's proposed conception of liberty renders unintelligible the idea that one might incur a cost in liberty as a result of a rightful political decision going against one.And yet this notion of a cost in liberty incurred by those who end up on the losing side of what they acknowledge to be a rightful political decision, Williams points out, 'is at least as well entrenched in historical and contemporary experience as that of a rightful claim in liberty ' (2005b: 84).We should accordingly be suspicious of a conception of liberty that accommodates the latter but rides roughshod over the former.
However, as I propose to construe it, Williams's objection to Dworkin's tidy conception is not just that it is untrue to the less tidy experience of life under a political order.It is, more specifically, that reflection on the role of the concept of liberty 'in political argument and political conflict' (Williams, 2005b: 84) reveals needs for a concept which, while narrower in scope than the concept of primitive freedom, remains wider in scope than Dworkin's concept of liberty as rightful freedom.
In particular, the concept of liberty needs to be able to simultaneously meet the conceptual needs-the needs for certain concepts-of those on the losing side and those on the winning side of a political decision.Those on the losing side need a concept that enables them to contest even rightful political decisions by enabling them to voice reasonable complaints in liberty even when these are not backed by rights.And those on the winning side need a concept that facilitates respect across the aisle by giving them the conceptual wherewithal to make sense of the costs in liberty incurred by their political opponents.Let us consider these in turn.

CONCEPTUAL NEEDS ON THE LOSING SIDE
The first conceptual need Williams invokes, on my reading, derives from the most basic concern that underlies our use of anything like the concept of liberty to begin with: the concern for primitive freedom.If the concept of liberty is to retain its connection to that concern, it needs to be able to act as the political sharp end of the concern for primitive freedom, enabling the expression and promotion of that concern, and calls for compensation for the frustration of that concern, through political argument.
From this perspective, having a political concept of liberty will clearly be pointless unless it allows one to lay claim to more freedom than one has under the current political arrangements.But the concept can only do that if it gives people grounds for complaining about restrictions on their liberty even and especially when they are not fully identified with rightful political decisions.A concept of liberty which presupposes that one's own will is perfectly aligned with the will expressed in the rightful activity of the state, as Dworkin's proposed concept does, is pointless in this connection, since those who are fully identified with that activity are precisely those who have no cause for complaint.
To serve the concern for freedom in the political sphere, we therefore need a concept of liberty under which the reasons grounding claims in liberty are not exhausted by one's rights, but allow for reasonable complaints in liberty grounded in the residual losses in liberty one has incurred as a result of rightful state action.Only then can resentment at rightful restrictions of one's freedom intelligibly be given a political voice.The concept of liberty can meet our needs as a political concept only if it gives us intelligible grounds for contesting the prevailing understanding of rightful freedom.
To enable such contestations, the concept of liberty needs to make conceptual room for the thought that the rightful curtailment of someone's freedom can still reasonably be resented as a loss in liberty.For only then can the concept of liberty serve our concern for freedom in such situations-and such situations are sure to arise as long as there are people who are not fully identified with rightful political decisions.
Dworkin's conception of liberty as rightful freedom, by contrast, leaves no conceptual room for reasonable complaints in liberty from those who end up on the losing side of a political decision.They can of course still complain-but their complaints must appear confused or unreasonable in light of this conception, for, under this conception, the reasons that can ground their claims in liberty are exhausted by their rights under the political system.
Any construction of the concept of liberty that is to enable political contestations of what is to count as rightful freedom therefore needs to spread the idea of liberty, and hence of a cost in liberty, more widely than Dworkin's proposed concept does.While the concept of liberty cannot, consistently with the existence of the political, treat any and all complaints in primitive freedom as reasonable-as Williams puts it, 'no concept of liberty intelligible as a political value could allow anybody to murder anybody they liked' (2001: 93)-a helpful concept of liberty, and particularly one that is suitably responsive to the fact that pluralistic societies are unlikely to be of one mind about everything, will need to be far more inclusive than Dworkin's if it is to serve the concern for primitive freedom by facilitating the political contestation of rightful state action.
We will still want to rule out as unreasonable claims that are not socially presentable in Williams's technical sense; and we will also want to rule out claims that are merely a product of insincerity, ignorance, or insufficient attention to the relevant arguments.But when these fairly  (2005a: 122).For political purposes, we should regard the voicing of a responsible claim of a loss in liberty as a sufficient reason to assume that there has been a loss in liberty.This is not a metaphysical argument resting on an independent account of the nature of liberty; it is an argument in 'conceptual ethics', which advocates a certain way of thinking in politics: namely, one on which we count responsible claims of a loss in liberty as reason enough to assume that there has been a loss liberty; and it is an argument rather than the blank assertion of a liberal piety because the conceptual need engendered by the concern for freedom constitutes a reason to prefer a conception of liberty along these lines to the one advocated by Dworkin.In leaving no conceptual room for reasonable complaints, on the part of those whose desires are frustrated by rightful political decisions, that they have incurred a cost in liberty, Dworkin forecloses an important form of political argument-a form of argument that, especially in pluralistic and polarized societies in which full and general identification with the rightful activities of the state is bound to be rare, is crucial to the political expression of the concern for freedom.
Thus, a concept of liberty capable of serving the concern for freedom of those who end up on the losing side of a rightful political decision needs to allow for reasonable complaints in liberty grounded in the residual losses incurred by that decision.In other words, Dworkin and Williams agree that not every loss in primitive freedom can be reasonably resented as a loss in liberty, since the concept of liberty, just because it is a political concept, needs to be narrower in scope than the concept of primitive freedom; but Williams insists, against Dworkin, that not every reasonably resented loss in liberty has to be rightfully resented as such, because the political concept of liberty needs to be broader in scope than the concept of rightful freedom.
Of course, this will cast the net for complaints in liberty that are intelligible input to political debate fairly wide, but if the concept of liberty is to serve the concern for freedom, the net needs to be cast wide, because how seriously any of these complaints should be taken is itself a political question, not one to be settled in advance by a definition.The concept's role, insofar as it serves the concern for freedom, is to determine what can go into the funnel of political debate, not what comes out of it.As Williams puts it: A construction of liberty on these lines . . .means that, within certain limits, anyone with a grievance or who is frustrated by others' actions can appropriately complain about restrictions on his liberty.If 'appropriately' means that it is semantically, conceptually, indeed psychologically, intelligible that he should do so, that is right.If it means that it is necessarily useful, helpful, to be taken seriously as a contribution to political debate, and not a waste of everyone's time, it is not right.The point is that these latter considerations are in the broadest sense political considerations, and that is the point of the construction.(2005b: 92) This conception of liberty determines what losses in primitive freedom merit to be so much as intelligible as responsible input to political debate, but it does not by itself determine how much weight that input should be given, or what should come out of the debate.
Dworkin's conception, by contrast, collapses these two steps into one, and, in so doing, entrains a loss in freedom: in particular, a loss in the freedom to contest rightful restrictions of liberty through complaints that have a claim on society's attention.The concept of liberty can only serve QUELOZ the concern for freedom in this regard if it casts liberty as broader than rightful freedom.Instead of serving the concern for freedom that most basically animates the use of the concept of liberty, Dworkin's conception of liberty thus ends up working against the concern for freedom.
At its core, then, Williams's first positive reason to favour the conception of liberty he advocates is an argument from the concern with primitive freedom to what an adequate political concept of liberty needs to be.The concept fundamentally serves the concern with primitive freedom, focusing that concern in the political context.Dworkin's proposed conception fails to serve that concern in one crucial respect: it achieves congruence with the concept of equality at the cost of failing to serve people's concern for freedom when they end up on the losing side of a political decision and their freedom is rightfully restricted.A concept of liberty able to serve people's concern for freedom under those circumstances cannot therefore coincide with the concept of rightful freedom.

CONCEPTUAL NEEDS ON THE WINNING SIDE
The second conceptual need that a thoroughly political concept of liberty must meet, according to Williams, arises from the requirement on a shared political system to contain pluralistic and conflicting concerns.We might put this by saying that the pluralism of concerns itself engenders a conceptual need for a concept of liberty that facilitates our living together in a pluralistic society with others whose concerns radically differ from our own.One notable way in which a concept of liberty can facilitate this is by equipping those on the winning side of a political decision to make sense of, and acknowledge the costs in liberty incurred by, those on the losing side-as opposed to treating their complaints as products of error, ignorance, immaturity, confusion, delusion, or deception.The concept thereby enables what we might call respect across the aisle.Whether those on the winning side have the conceptual wherewithal to make sense of the resentment of those on the losing side as reasonable makes a great difference to the character of liberal democratic politics-in the first instance, by affecting what those on the winning side can say to those on the losing side.That, Williams insists, is also an important 'form of citizenly address, particularly in a pluralistic society ' (2001: 102).
In effect, Williams proposes to extract, from a pluralist understanding of politics, a sense of what we need the concept of liberty to be.'The idea of value pluralism', he writes, is no mere 'aestheticism of politics'; it 'tells you how to speak to the people who have to pay, not just in their interests but in their values, for things that have to be done ' (2001: 102).The legal scholar Jamal Greene observes that while the last century may have given us tools to fight political exclusion, 'in this century, we need the tools to build a politics of pluralism' (2021: xxi).The concept of liberty that Williams advocates is just such a tool: it is tailored to the conceptual needs of the politics of pluralism.As Cueni's (manuscript-a) reconstruction of this argument in Williams emphasizes, pluralistic societies are not of one mind about what political values such as liberty and equality require and how they should be weighed against other values; and just because of this, we need thoroughly political conceptions of these values.In particular, we need a conception of liberty that allows us to make sense of the complaints in liberty of those who reasonably resent what is being enacted, however rightfully, in the name of equality.
Yet all that Dworkin's tidy equation of liberty with rightful freedom encourages those on the winning side to say to those who feel they have incurred a cost in liberty although their liberty has been rightfully curtailed is that they are mistaken: they may think they incurred a cost in 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License liberty, but if only they achieved a clear-headed understanding of the concept of liberty, they would recognize that no such loss was incurred, and their resentment was therefore unreasonable. 32dopting this attitude towards our political opponents' complaints, Williams maintains against Dworkin, is 'objectionable ' (2005b: 85).In particular, '[t]elling these people that they had better wise up and revise their definition of the values involved', as Dworkin's proposed conception invites us to do, 'is not in many cases prudent, or citizenly, or respectful of their experience' (2001: 102).There are three related, but distinct objections curled up in this observation.
First, it is not prudent, because dismissing their complaints as conceptually confused only makes it harder to secure the consent of those who end up on the losing side of a political decision, and securing losers' consent is vital to the stability and health of a democracy. 33This is presumably why Williams writes that Dworkin's outlook 'does not encourage a helpful-one might say, healthy-relation to one's opponents ' (2005b: 85).What it is unhealthy for is democracy: it amplifies resentment, polarization, and ultimately the threat of violent confrontations of just the sort that the political order was meant to sublimate. 34Cueni (manuscript-a) illuminatingly compares this strand in Williams's argument to Chantal Mouffe's (2000: 13) insistence on the need for politics to prevent adversaries from turning into enemies.
Once this is understood, it also becomes clear why Williams goes on to note that brushing off one's political opponents as failing to understand the nature of liberty is not citizenly.So far from being a manifestation of civic virtue, i.e. of the character traits and dispositions associated with the successful functioning of the civil order, it is 'hostile to the relations of fellow citizenship' (Williams, 2005b: 86), undermining the civil order by threatening to alienate from the political process those who feel they have incurred a cost in liberty as a result of a decision going against them; by noting that Dworkin's view encourages a notably uncitizenly form of citizenly address, moreover, Williams is implicitly suggesting that the view falls foul of the Rousseauian tradition he sees it as standing in (2005a: 120-21; 2005b: 85), since it was Rousseau who, more than anyone in modern political theory, revived the ancient ideal of citizenly behaviour.
Third, Williams objects that to dismiss our political opponents' complaints in liberty as conceptually confused is not respectful of their experience.We should 'take seriously the idea that if, under certain conditions, people think that there is a cost in liberty, then there is'-this is a condition 'not only of taking seriously the idea of political opposition, but of taking our political opponents themselves seriously ' (2005b: 85).And '[w]hat we should take seriously', in particular, 'are their reactions, or at least their deeper reactions, rather than the extent to which we are disposed to share or morally approve of their reactions ' (2005b: 85-86).That is to say, we should take seriously the reactive attitudes they continue to have once we have subtracted the superficial reactions due merely to insufficient attention, reflection, or information.
The relation of political opposition is a relation that not only specially calls for respect, but also leaves more space for it. 35After all, as Williams repeatedly emphasizes, a political decision 'does not in itself announce that the other party was morally wrong or, indeed, wrong at all.What it immediately announces is that they have lost ' (2005d: 13).That is why regarding someone as a political opponent is subtly but crucially different from regarding them as someone one is morally at odds with.In a moral disagreement, we treat the other party primarily as someone we have to argue into seeing the moral situation aright, and hence into recognizing that they are wrong.This is what we tend to think of as respectful engagement in a moral disagreement.But in a political disagreement, Williams thinks, '[w]e should not think that what we have to do is simply to argue with those who disagree: treating them as opponents can, oddly enough, show more respect for them as political actors ' (2005d: 13).In coming to think of those with whom we disagree as political opponents, we foreground not the extent to which we disapprove of their reactions, but their reactions; not our sense that they are wrong to disagree with us, but the fact that they disagree with us.
This consideration for the fact of their disagreement and the reactions that underlie it opens up the possibility of recognizing further that they do not necessarily disagree with us merely out of error, but might have come by their political convictions much as we came by ours: through an obscure confluence of sociohistorical and biographical circumstances, passions, and interests (Williams, 2005d: 12-13).On this basis, we can then recognize that a political decision going against them may come at a real cost to them in terms of their values.This is a precondition-and it is no more than that-of acknowledging what resentment this produces as reasonable, and thus of taking their reactions seriously.
To conceive of their political opponents in this fashion, however, those on the winning side of a political decision need a concept that is, in Williams's words, less 'instructional' and 'patronizing' (2005b: 86) than Dworkin's concept.They need a concept that enables them to respect those on the losing side as people who have suffered costs in liberty.For, as Cueni (manuscript-b) observes, respect is characteristically respect under a description: we do not just show respect for people, but for people as people who possess some property, and this requires the conceptual wherewithal to make sense of them as possessing that property.Even to show respect for persons as persons already requires not just some concept of a person, but, specifically, a conception on which the fact of being a person itself already entitles one to respect under that description.A fortiori, to show respect for persons as persons who have incurred a cost in liberty when their liberty has been rightfully curtailed requires a conception of liberty that makes conceptual room for the thought that even a rightful restriction on someone's liberty can still reasonably be resented as a loss in liberty.
Those on the winning side of a rightful political decision thus need the conceptual wherewithal to acknowledge the costs in liberty suffered by those on the losing side: they need a concept of liberty that facilitates a prudent, citizenly, and respectful attitude towards their political opponents.That is part of what is required to address one's political opponents in ways that help secure losers' consent, cultivate healthy relations of fellow citizenship, and foster respect across the aisle.Hence Williams's emphatic conclusion that 'the proposed interpretation of liberty is what we need in order to live in society with others who have different interpretations of equality ' (2005a: 125-26).

THE DUALISM OF INTEREST AND PRINCIPLE
Let me summarize the argument thus far.For Dworkin, we have a standing reason to prefer concepts or conceptions that do not conflict, and this gives us a reason to use a concept of liberty under which the reasons grounding claims in liberty are exhausted by one's rights under the political system.For Williams, on the other hand, the pursuit of conceptual integrity must take a backseat when it threatens to sever the connection to the underlying concern for freedom that most basically animates our use of anything like the concept of liberty, and this is what Dworkin's proposed concept threatens to do when it denies people any reasonable claims in liberty beyond those grounded in rights.In the process of whittling away the aspects of the concept of liberty that bring it into tension with the concept of equality, moreover, Dworkin also ends up shaving off a second valuable feature of the concept of liberty: that it enables those on the losing side of a rightful political decision to make claims in liberty that those on the winning side can acknowledge as reasonable even when they are not backed by rights.I then identified two reasons to prefer a conception of liberty along the lines Williams advocates over Dworkin's tidier conception.First, to serve the concern for freedom in the political sphere, those on the losing side of a rightful political decision need a concept capable of acting as the political sharp end of the concern for primitive freedom, enabling them to express and promote that concern, or else to demand compensation for the frustration of that concern.This calls for a conception of liberty under which the reasons grounding claims in liberty are not exhausted by one's rights, but allow for reasonable complaints in liberty grounded in the residual losses in liberty one has incurred as a result of rightful state action.
Second, those on the winning side need a conception that equips them to make sense of, and acknowledge the costs in liberty incurred by, those on the losing side if they are to secure losers' consent, cultivate healthy relations of fellow citizenship, and foster respect across the aisle.This need can also only be met if the conception of liberty does not limit reasonable claims in liberty to those that can be grounded in rights.For both of these reasons, the concept of liberty we need cannot be one that is immunized against conflict in the way Dworkin proposes.By attending to the concerns to which the concepts of liberty and equality should be responsive, we thus recognize that the concepts need to carry the 'permanent possibility of conflict' (Williams, 2001: 95).That is why the pursuit of conceptual integrity in this particular case is not merely a lost cause, but an ill-advised one.
This holds wider lessons for attempts to revise or re-engineer our political concepts.Conflicting concepts may offend against philosophers' tidy-mindedness, but it is the concerns animating our use of anything like these concepts in the first place that our concepts should be answerable to, and conflicting concepts can be the better concepts in virtue of serving our concerns better.Furthermore, whatever concerns concept-users genuinely have cannot be eliminated or redirected at the drop of a definition.Even if our concepts are something we construct, we are still beholden to the concerns these concepts are to serve, and our concerns are far less amenable to revision than our definitions.
But perhaps Dworkin's ambitions for conceptual integrity can be more charitably understood in ways that narrow the divide between him and Williams; for, as we saw, and as Williams himself acknowledges (2001: 94), Dworkin does not pursue conceptual integrity solely for its own sake, or only because it would be theoretically tidier.Dworkin has an additional motive: that it would be better for the citizens if political values were such as to reduce the number of tragic choices facing a government, and hence the number of instances in which its citizens ended up being wronged by the political system-and this is itself a motive rooted in a recognizable human concern.
Other passages in Dworkin's oeuvre likewise suggest that his approach is more grounded in human concerns than the discussion hitherto allowed.In 'Do Values Conflict?A Hedgehog's Approach', for example, Dworkin also advocates conceptions of liberty and equality that are selfconsciously designed not to conflict.But he insists that, in contrast to concepts of natural kinds like gold, these conceptions are answerable to human concerns: We believe that gold is what it is quite independently of human concerns, ambitions, or needs.But that is not even remotely plausible about a political virtue like equality or liberty. . . .They are what they are because we are what we are: we believe that a government that respects liberty and equality in some way improves the lives of those whom it governs.(2001b: 255) Similarly, in Sovereign Virtue, Dworkin argues that it is in light of 'the basic assumptions' defining 'our political culture ' (2000: 128) that we must construct our interpretations of liberty and equality.He just happens to believe that an honest look at those assumptions will reveal our concern for equality to loom larger than our concern for liberty-so large, indeed, that liberty must be completely subordinated to equality: 'No theory that respects the basic assumptions which define that culture could subordinate equality to liberty . . . to any degree', Dworkin maintains.'Any genuine contest between liberty and equality is a contest liberty must lose ' (2000: 128).But it is nonetheless out of an overwhelming concern for 'equal concern' that we build society on egalitarian principles.
Crucially, however, Dworkin's view is not simply that our concern for equality is so overwhelming that it must win out whatever the loss in liberty.His view is rather that, whenever an apparent conflict between liberty and equality is decided in favour of equality, a truthful account of our concerns will show, or can at least be hoped to show, that nothing 'of value has been lost ' (2001b: 255).That is why Dworkin strives, in an experimental or tentative spirit, 'to show that no genuine conflict exists, that no right to liberty we would otherwise want to recognize would be compromised by policies our conception of equality demands ' (2000: 131).
If we can thus resituate the Dworkin-Williams debate within an inquiry into how to construct concepts in light of the concerns we need those concepts to be responsive to, this raises the question of what kind of disagreement the debate really boils down to. 36It is tempting to think that the disagreement is ultimately over how to interpret the relevant concerns at work in our political situation: Dworkin, convinced that political legitimacy depends fundamentally on the state showing equal concern for the life of each citizen, gives more weight to the need for the political system to avoid being in a position where it must inescapably wrong some of its citizens in showing concern for others; Williams, by contrast, gives more weight to the twin needs to be able to contest the prevailing understanding of rightful freedom and to engage with one's political opponents in a way that is prudent, citizenly, and respectful of their experience.
One might even be tempted to dissolve the disagreement entirely by indexing the two views to different localized political contexts: in Dworkin's case, to an American context with a history that is deeply grounded in the concept of rights, and which heavily relies on the Supreme Court for principled conflict resolution; in Williams's case, to a British political context in which it is far less clear that every political consideration should be articulated in terms of 'rights'.This comes out in Williams's review of Dworkin's A Matter of Principle (1985), which opens with an anecdote from Williams's time as chairman of a Committee on obscenity laws in the 1970s: an American lawyer had been invited to weigh in, and after he had left the room, the British legal scholar Brian Simpson said: 'I think I should explain something to the Committee.Americans believe in rights' (2014a: 256). 37Williams thus expresses something of his political culture when he complains that 36 I am grateful to Vida Yao for pressing me on this point. 37Which is not to say that Williams is generally opposed to rights-based thinking: he also writes sympathetically of the liberal predicament of 'people who need a theory of individual rights, but have lost some of the traditional reasons for asserting them ' (2014b: 318), not least as a result of Weberian disenchantment.On these wider cultural developments in the background of Williams's thought, see Krishnan and Queloz (2023).'there is something perverse in the demand to force all principled political argument into this one mould, and to ignore the wider range of conceptions that certainly have power in our political discourse ' (2014a: 260).
But there is a deeper disagreement here which survives even such relativizing interpretations.At bottom, Dworkin and Williams have different conceptions of politics itself.In particular, there are two fundamental assumptions implicit in Dworkin's approach that Williams urges us to question.
First, Williams could not agree that the disagreement boils down to how to interpret the relevant concerns at work in our political situation, because for him, the point is precisely that political opponents are 'not all trying to interpret the same text ' (2005b: 86).This mistaken assumption is one of the central characteristics of the 'political moralism' that he rejects in favour of 'political realism'.In a posthumously published essay, Williams explains that political moralism 'naturally construes conflictual political thought in society in terms of rival elaborations of a moral text: this is explicit in the work of Ronald Dworkin.But this is not the nature of opposition between political opponents ' (2005d: 12).Political debate should therefore not be modelled on judicial review; and neither should it be modelled on moral debates aiming to get others to see the moral facts aright.We need to hold on to the platitude that 'politics is neither morality nor constitutional law ' (2001: 101).
Second, Williams regards attempts to assimilate politics to either morality or constitutional law as relying, ultimately, on an overly stark contrast between interest and principle.'I suspect', he tells Dworkin in New York, 'that in these mistaken conceptions of politics there is lurking a Kantian dualism, to the effect that there is one world of interests which consists of winning and losing, and another world of principle, which is expressed in being right or wrong ' (2001: 101).While there will always be people who are disappointed in their interests by rightful state action, our more principled political values, like liberty and equality, hover above the fray.This dualism of interest and principle is what fundamentally fuels Dworkin's hope that our political values might be articulated in terms of non-conflicting rights, so that the state need not face tragic conflicts of principle entailing that it must wrong some of its citizens whatever it does.
Aiming at principled interpretations of our political values as implying non-conflicting rights may be a noble aspiration.But it becomes problematic, on Williams's view, if it relies on a dualism of interest and principle which entails that any dimension of political value that cannot be so interpreted must be demoted to the status of a mere interest, barred from rising to the level of a responsible political claim.Dworkin may present himself as setting out, with an open mind, to discover whether the pluralist thesis holds true for certain political values.But in fact, his commitment to the dualism of interest and principle ensures that he has implicitly ruled it out already from the start.This implicit dualism, correctly understood, not only fuels Dworkin's hope that the conflictual character of political values might be overcome, but actually precludes tragic conflict. 38 Accordingly, it is in contrast to such a dualism of interest and principle that Williams characterizes his own conception of politics: [M]y view is in part a reaction to the intense moralism of much American political and indeed legal theory, which is predictably matched by the concentration of American political science on the coordination of private or group interests: a division of labour which is replicated institutionally, between the 'politics' of Congress and the principled arguments of the Supreme Court (at least as the activities of the Supreme 38 I am indebted to Vida Yao on this point.Williams goes on to name Dworkin as the paradigmatic exponent of the moralism that identifies politics as embodied by the U.S. Congress with the pursuit of mere interests, and political theory as embodied by the U.S. Supreme Court with the pursuit of principled and high-minded arguments. This suggests that Williams is indeed disposed to index Dworkin's view to the American context.But this does not dissolve the disagreement, because Williams disagrees with Dworkin's view even as an account of that context.Even within a political system as reliant on the courts as the American one, principled arguments in terms of political values are not confined to the courts.Politics itself, as embodied by the legislative as opposed to the judicial branch, 'provides a dimension which can be governed by values as well as by interests ' (2001: 102).
It thus betrays an overdrawn contrast between interest and principle to conclude that if politics does not take the form of principled interpretations of our political values as implying nonconflicting rights, it must be an utterly unprincipled sphere of pure opportunism and clientelism, terminally deaf to political values and ideals.We had better not lose sight of the fact that, halfway between the politics of interest and principled legal argument, there can still be such a thing as 'the politics of principle ' (2001: 101).
In a revealing remark, Williams observes that the 'contrast between principle and interest . . .signifies the continuation of a (Kantian) morality' (2005d: 2).It reinvents, within the sphere of politics, an equivalent of the Kantian dualism between claims of morality and claims of prudence.This invites us not only to read Williams's critique of Dworkin as a detailed exemplification of his critique of political moralism, but also to read his critique of political moralism as a direct extension of his critique of 'the morality system'. 39The dualism of interest and principle, like its analogue in morality, distorts what it subordinates in its pursuit of the comforting fantasy that principled thought could be safeguarded from systematic conflict.And in doing so, it encourages both too rosy a view of our scheme of political values and too cynical a view of the actual content of politics.
What Williams advocates instead is 'a broader view of the content of politics, not confined to interest, . . .where all the considerations that bear on political action-both ideals and, for example, political survival-can come to one focus of decision ' (2005d: 12).On this view, politics is, to a non-negligible extent, a principled space; but it remains at the same time a political space, in the sense that a decision going against one signifies not that one was wrong, necessarily, but merely that one lost.Politics may involve principles as well as interests, but 'the politics of principle isn't morality or constitutional law either ' (2001: 101).
A realistic conception of politics, for Williams, is thus precisely not the threadbare realism of international relations that reduces politics to interests.This would be to turn the contrast between interests and principles into a dualism.Rather, a realistic conception of politics resituates the contrast between interest and principle within politics.This yields a notably less cynical view of politics: there is more to it than the self-serving politics of the pork barrel; there is a place in it for high-minded political values as well as for interests.But this recognition comes at the price of accepting that our principles can conflict as much as our interests, and that those in power 39 That the two are connected has also been suggested, on different grounds, by Harcourt (manuscript).
19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License may be left with dirty hands even if they only ever act rightfully. 40Sometimes, any decision the government takes must come at a real cost to some of its citizens, not just in terms of their interests, but in terms of their values, even if they are not wronged in the sense of having their rights violated.
I suggested at the beginning that the Dworkin-Williams debate formed a key to Williams's later political philosophy.We can now clearly see what otherwise remains implicit, namely that his various responses to Dworkin offer perhaps the most sustained discussion in Williams's oeuvre of what his rejection of political moralism in favour of political realism entails, and how it forms a continuation of his critique of 'the morality system' in the political sphere.The dualism of interest and principle, just like the dualism of prudence and morality, is motivated notably by a concern to reduce the uncomfortably conflictual character of our values. 41 By regaining a sense of the concerns that these various values more immediately answer to, however, we can recognize how this dualism distorts our view both of these values and of politics.These values need to carry the permanent possibility of conflict if they are to serve the concerns that animate their use.Williams's reservations thus stem from a deep conviction that, however tentative the spirit in which Dworkin strives for conceptual integrity, he must in the end be nurturing an illusory hope in entertaining the prospect that our political lives could be shielded from tragic conflict.

A C K N O W L E D G E M E N T S
I am indebted to Damian Cueni for the many probing conversations in which we refined each other's reconstructions of the Dworkin-Williams debate.I urge readers to consult his own work drawing insights from this debate.I was fortunate also in benefiting from Vida Yao's exceptionally knowledgeable comments, which improved the article as a whole and the last section in particular.In addition, I am grateful to Anna Goppel, Andreas Müller, Markus Stepanians, Matthias Rolffs, Lukas Nägeli, Marcel Twele, Samuel Tscharner, and the other participants at the practical philosophy colloquium at the University of Bern for their helpful feedback.An early version of this material was presented at the 2021 ConceptLab Workshop on Political Philosophy and Methodology at the University of Hong Kong, where I received valuable suggestions from Herman Cappelen, David Plunkett, Max Deutsch, Mona Simion, David Enoch, Matthew McKeever, and Joseph Chan.Open access funding provided by Universitat Bern.

R E F E R E N C E S
19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons LicenseQUELOZCourt are primarily interpreted at the present time).That view of the practice of politics, and the moralistic view of political theory, are made for each other.They represent a Manichaean dualism of soul and body, high-mindedness and the pork barrel, and the existence of each helps to explain how anyone could have accepted the other.(2005d:12) 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License minimal demands are met by someone's claim that they have incurred a loss in liberty, then, though we may not yet have a rightful claim, we will have what Williams calls a responsible claim 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 19331592, 2024, 1, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/phpr.13002 by Schweizerische Akademie Der, Wiley Online Library on [23/07/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License