Some Clarifications

John Simmons apparently contemplates the object of my critique in the following passage:

In these sentences, however, Simmons does not quite identify my topic, which differs in two ways. First, the claim that he describes is one of both necessity and sufficiency, whereas only the matter of adequacy interests me here. The question of whether a political obligation is at all possible in the absence of democracy, which some theorists deny,^{Footnote 7} is not my immediate concern. Although the necessity of democratic rule happens to be a feature of Dworkin’s case for its sufficiency, my critique has no bearing on whether it is essential. Furthermore—this is the second difference—I do not consider whether democracy solves the problem of political obligation. Since a political obligation need not be a “negative” one to obey the law—it might, instead, be a “positive” one to take part in, for example, democratic decision-making (when not legally compelled to do so)—the problem on which I focus is no more than a problem of political obligation.^{Footnote 8}

I explore the potential for democracy to solve this specific problem. I reflect on the plausibility of the assertion that those whom the law addresses are morally bound to obey it whenever it is democratic. But this problem must not be confused with a more general one. It is not equivalent to the question of whether democracy has moral worth: rather, it corresponds to the narrower question of whether democracy is morally valuable in the restricted sense that those to whom the laws of a democratic state apply are necessarily obliged to obey them. As Simmons notes, democracy might be justified in various ways: “[T]o justify an arrangement is to show either that it is permissible for us (to create or live under it), that is best (optimal) for us, or that it is mandatory for us (to support or create it).”^{Footnote 9} My question is about the third sort of justification, which Simmons calls “legitimation”.^{Footnote 10} I hesitate to use his label, though, since it presumes acceptance of his view that the moral legitimacy or the “right to rule” of a political society is the “logical correlate” of a moral obligation to obey its laws.^{Footnote 11} Not everyone agrees: Philip Pettit, for instance, connects legitimacy to a moral obligation, not to obey, but to oppose any directives that seem objectionable “in ways allowed by the system”.^{Footnote 12} Allen Buchanan disagrees with Simmons, too. He distinguishes between “political legitimacy”, which “[a]n entity has […] if and only if it is morally justified in exercising political power,”^{Footnote 13} and “political authority”, which “an entity has […] if and only if, in addition to (1) possessing political legitimacy, it (2) has the right to be obeyed by those who are within the scope of its rules; in other words, if those upon whom it attempts to impose rules have an obligation to that entity to obey it.”^{Footnote 14} Given these variations, I try to avoid needless confusion by eschewing, except when absolutely necessary, the language of legitimacy here. Instead, I focus on whether democracy has value such that those of whom a democratic state makes practical demands inevitably have a moral obligation to obey it.

A final question with which mine must not be (but is, I suspect, often) confused is whether voluntary participants (and perhaps even those who are eligible yet choose not to take part) in a democratic process are, as a consequence of their participation (or abstention), morally required to obey the laws that the process generates. This enquiry is really about consent, not democracy, as a ground of political obligation. It investigates whether, in the absence of coercion, voting (and maybe also not voting) in an election constitutes or implies consent to the laws that result.^{Footnote 15} It does not ask, as I do here, whether nothing but the democratic origin of those laws automatically furnishes those who are subject to them, including those who were not entitled to contribute to their production, with a moral duty to do as they require.

A Preliminary Objection

Although my question should now be clearer, perhaps my intended search for a positive response to it is obviously hopeless and ought immediately to be forsaken. This potential charge of futility is motivated by the apprehension that no democratic argument for a political obligation can fulfil the supposed need for “particularity”. Simmons is the leading, but certainly not the only, proponent of this alleged condition, which, he says, “requires, roughly, that […] an account […] be able to explain why the moral duty (or obligation) to obey is owed specially to one particular society (or to its subjects or governors) above all others (namely, to ‘our own’ societies), rather than offering only some moral reason for obedience that would bind one equally or more imperatively to obey or support the laws or political institutions of other societies.”^{Footnote 16} From this purported necessity, Simmons infers that no political obligation can be derived from a general duty to promote justice, for example, given its inability to discriminate between the reasonably just states—none can be expected to be wholly just—to which it demands allegiance.^{Footnote 17} Also among the so-called “natural-duty” arguments that he rejects are those that seek to ground a moral obligation to obey the law in the professed value of democratic institutions:

Simmons’s accusation of neglect is probably warranted: Thomas Christiano, for instance, fails to demonstrate that his case for democratic authority deals satisfactorily with this issue, notwithstanding his acknowledgement of it.^{Footnote 19} Rather than doing more, though, perhaps he and other natural-duty theorists should challenge the philosophical consensus that particularity is essential.

They could do so by alleging that Simmons, rather than justifying, actually suggests reasons against its inclusion among the attributes that a moral obligation to obey the law must exhibit. As evidence of his self-contradiction, they could juxtapose his attempt to vindicate the necessity of particularity by appealing to conventional morality^{Footnote 20} with his warning against uncritical acceptance of orthodox beliefs about states because of the formation of such beliefs in circumstances of domination.^{Footnote 21}

Natural-duty theorists could also contest the assumed need for particularity by denying that it is an implication of the “political” character of obedience to legal norms. Astonishingly, this quality has seldom been discussed by philosophers and Dorota Mokrosińska exaggerates only somewhat when she claims to “enter unexplored territory” in considering it.^{Footnote 22} She contends that a moral obligation to obey the law is not “political” unless “the law in question is the law of one’s own polity”.^{Footnote 23} Yet she could be said to shift without adequate justification from a “descriptively correct” understanding of politics as “a domain of asymmetrical power relations” to the normative conception of politics as just co-operation that underlies her insistence on particularity.^{Footnote 24} After all, that moral questions are raised by these relations does not mean, as Mokrosińska presumes, that an account of politics must itself answer them. Incorporation of a vision of justice into the very meaning of politics is not an inevitable consequence of her plea for a “normative counterpart” to the descriptive view.^{Footnote 25}

To be compelling, these prospective criticisms of the demand for particularity must be developed. Yet the requisite elaboration can be omitted here,^{Footnote 26} since the one entirely democratic case for political obligation of which I am aware would, if successful, exhibit the quality that Simmons and others deem crucial. Insofar as Dworkin’s argument is democratic, it is, consistent with its reputation, associative, too. As such, it promises, as do all associative arguments, to solve the supposed problem of particularity. Whether this problem is genuine matters much less to it than to theories of democratic authority that lack its communitarian character and so need not be presently determined.

Dworkin’s Isolation

Is Dworkin really the only champion of the claim that democracy is sufficient for political obligation? Before ascertaining his position, I show that, perhaps contrary to expectations, none of several prominent theorists of democratic authority holds that those whom a democracy tells to act or not to act in specified ways are inevitably obliged to obey its dictates. Each theorist calls for something (and most call for a lot) more.^{Footnote 27}

Consider, first, Peter Singer’s view. He lauds democracy as a “fair compromise” because it gives everyone an “equal say” in its decisions.^{Footnote 28} In his opinion, “there are strong reasons for playing one’s part in supporting and preserving a decision-procedure which represents a fair compromise” and “[t]o disobey under these circumstances is to reject the compromise and to attempt to use force to impose one’s views on others.”^{Footnote 29} Significantly, however, he recognises the likelihood of democratic outcomes that continually discriminate against specific persons and concedes that “an equal vote is insufficient to ensure that the system operates as a fair compromise between all parties.”^{Footnote 30} He concludes: “If ‘popular government’ just means that the government derives its powers equally from all the people, this alone is not quite sufficient to give rise to […] reasons for obedience”.^{Footnote 31}

Scott Shapiro agrees that democracy need not engender political obligation. He regards democratic government as “socially necessary, empowering and fair” and, like Singer, complains that someone who flouts its laws behaves “like a dictator: he unilaterally ‘dictates’ the terms of social interaction to others and thereby exercises inappropriate control over the lives of his fellow citizens.”^{Footnote 32} He is clear, though, that compliance is not always obligatory:

For Shapiro, the same values both warrant obedience to democratic norms and restrict their authority. Structurally identical, but substantively different, is Christiano’s theory, according to which democracy generates a moral obligation to obey the law insofar as it publicly realises equality.^{Footnote 34} It does not do so, Christiano says, when it “attempts to violate the basic democratic rights or liberty rights of its citizens” or when it “fail[s] to ensure that each citizen has a decent economic minimum of economic resources with which to live a good life” or, finally, “when permanent minorities are formed in the society that never get their way in the democratic assembly.”^{Footnote 35} In each of these circumstances, democracy violates the principle of public equality that otherwise, for Christiano, gives it authority.

David Estlund articulates a less unified view, but he, too, identifies situations in which democratic laws are no longer morally binding. He locates the authority of democracy in its propensity to make better laws than other generally acceptable forms of government.^{Footnote 36} Yet he stresses that its epistemic value is contingent on voters’ motivations: “If actual democratic procedures are to have any tendency to produce just decisions it seems likely that this is because participants will have some significant tendency to aim for justice, and not only for some narrower personal ends.”^{Footnote 37} He also indicates substantive constraints. Having insisted that the authority of democracy is compatible with a degree of injustice, he acknowledges that some laws “must be too unjust” for obedience to them to be morally required.^{Footnote 38} Although he declines to “offer a theory of where these limits lie,”^{Footnote 39} he does provide a few words of guidance:

Jeremy Waldron, in contrast, repudiates all limits of this type. He is adamant that majority-decision is the only method of law-making that is consistent with respect for individuals because no other gives each of them an equal say in resolving urgent matters of justice on which they are divided.^{Footnote 41} Democratic authority, he says, depends on everything, even democratic rights themselves, being “up for grabs”.^{Footnote 42} Its scope, for him, is boundless. Yet, notwithstanding his rejection of all checks on majority-decision, he does not suppose that this procedure is sufficient for law to have authority. Like Estlund, he admits that those who take part in the process cannot be self-interested if compliance with its results is to be obligatory. Instead, they must, he says, “vote on the basis of good faith and relatively impartial opinions about justice, rights, and the common good.”^{Footnote 43} Since these motivations cannot be guaranteed, Waldron must be added to the list of theorists for whom democracy is insufficient for political obligation. He gets closer than any of the others to declaring its adequacy, but even he denies that it is enough.

Dworkin on Democracy and Political Obligation

None but Dworkin, at least among the theorists with whose work I am familiar, really thinks that democracy is sufficient for political obligation. His unique belief follows from his view of democracy as self-government or, in Isaiah Berlin’s words, “positive liberty”.^{Footnote 44} “It seems essential to the idea of democracy,” says Dworkin, “that democracy provides self-government”.^{Footnote 45}

Yet can democracy actually be understood in this way? Consider what democracy would need to be for every individual to govern him- or herself. For a start, it could not be representative, since an individual relinquishes his or her autonomy whenever he or she permits someone else to impose duties on him or her. Indeed, all constitutional restrictions, and not just the constraint of representation, are inimical to his or her self-government. Quite simply, everything—both procedural and substantive—would need to be open to resolution. Individual self-rule cannot be realised insofar as electoral regulations, bills of rights and other constitutional provisions limit what, how or when people are able to decide.

Among the limitations that could not be present is the norm that the will of the majority ought to prevail. Whenever some people are subject to the wishes of others, each of the former lacks autonomy. “No one,” says Jack Lively, “determines a decision who has voted against it.”^{Footnote 46} Yet the incompatibility of majority-rule and individual self-rule does not mean that every person would need to have the right to veto any proposed law. The introduction of a requirement of unanimity would simply replace one procedural restriction on decision-making with another and enable a single person to thwart the wishes (and so the autonomy) of all others. Hence, individual self-government would require complete agreement, which, of course, no procedural rule could ensure, on every change to the law. Such unanimity is quite improbable. But autonomy would seem to require even more.

Suppose that a man—it could just as well be a woman, but the example is not mine—refuses to obey a law to whose introduction he (and everyone else) previously agreed. If the state were to insist on his obedience to this law, would he rule himself? Robert Paul Wolff—the author of the example—contends that he would, since “a man who is constrained by the dictates of his own will is autonomous.”^{Footnote 47} Arguably, though, Wolff is mistaken: someone to whom an option is not available due to human, including his or her own, behaviour lacks autonomy. Self-rule implies that every person must be able to change his or her mind. It thus necessitates complete agreement on existing as well as new laws.

Consequently, individual autonomy requires anarchy, not democracy. Yet this correlation should not be surprising, given that anyone subject to government, of which democracy is a species, is, by definition, not autonomous. Ross Harrison puts the point well:

But the conclusion that everyone must be heteronomous in a democracy relies on a particular view of agency. It supposes that people behave individually and not together as a single body. Crucially, Dworkin challenges this assumption. He denies that the conduct of a group must be understood as “some function, rough or specific, of what the individual members of the group do on their own, that is, with no sense of doing something as a group.”^{Footnote 49} Instead, he says, their behaviour might be collective “in a way that merges their separate actions into a further, unified, act that is together theirs.”^{Footnote 50} He thus suggests a “communal”, as opposed to a “statistical”, account of democracy in which “political decisions are taken by a distinct entity—the people as such—rather than any set of individuals one by one.”^{Footnote 51}

Yet scepticism about the relevance of autonomy to democracy is unlikely to be eased and may even be intensified by this proposal. As Dworkin himself acknowledges, the idea of communal agency seems to invoke “a baroque metaphysics which holds that communities are fundamental entities in the universe and that individual human beings are only abstractions or illusions.”^{Footnote 52} For him, though, communal action depends “not on the ontological primacy of the community, but on ordinary and familiar facts about the social practices that human beings develop.”^{Footnote 53} A community, he says, is “created by and embedded in attitudes and practices.”^{Footnote 54} It acts as a distinct entity whenever persons—he gives the example of musicians in an orchestra—“recognize a personified unit of agency in which they no longer figure as individuals, but as components”.^{Footnote 55}

On this “practice” view of community, people can regard the decisions of their government, including those with which they disagree, as their own. Before they can do so, however, they must be “genuine” members of the community.^{Footnote 56} They cannot rule themselves, says Dworkin, unless they belong to a community whose government is committed to, even if it may fail properly to implement, two principles of “dignity”, according to which “each human life has a special kind of objective value” and “each person has a special responsibility for realizing the success of his own life, a responsibility that includes exercising his judgment about what kind of life would be successful for him.”^{Footnote 57} Dworkin maintains that people can be autonomous “[o]nly so long […] as [their government] accepts the equal importance of their lives and their personal responsibility for their own lives and tries to govern them in accordance with its sincere judgment of what those dimensions of dignity require.”^{Footnote 58} A “true” community, then, satisfies the conditions of “moral membership” that equal concern and respect implies.^{Footnote 59} These requirements are of two kinds.

First, a “genuine” community must meet “the genetic or geographical or other historical conditions identified by social practice as capable of constituting a fraternal community”.^{Footnote 60} In other words, “true communities must be bare communities as well.”^{Footnote 61} Second, “moral membership” depends on certain “relational conditions”. In a “genuine” community, “each person must have an opportunity to make a difference in the collective decisions, and the force of his role—the magnitude of the difference he can make—must not be structurally fixed or limited in ways that reflect assumptions about his worth or talent or ability, or the soundness of his convictions or tastes.”^{Footnote 62} There must, then, be “universal suffrage and effective elections and representation” plus “free speech and expression of opinion, not just on formal political occasions, but in the informal life of the community as well.”^{Footnote 63} Moreover, “the political process of a genuine community must express some bona fide conception of equal concern for the interests of all members, which means that political decisions that affect the distribution of wealth, benefits, and burdens must be consistent with equal concern for all.”^{Footnote 64} Finally, “moral membership” requires “liberal tolerance of unpopular sexual and personal morality.”^{Footnote 65} A “genuine” community “must not dictate what its [members] think about matters of political or moral or ethical judgment, but must, on the contrary, provide circumstances that encourage them to arrive at beliefs on these matters through their own reflective and finally individual conviction.”^{Footnote 66} It must, that is, be “integrated” rather than “monolithic”.^{Footnote 67}

With these demanding “relational” conditions, Dworkin separates “true” from “bare” communities. The latter are instances of the former in which every member has, in summary, “a part in any collective decision, a stake in it, and independence from it.”^{Footnote 68} A community that satisfies these conditions, says Dworkin, is a democratic one whose members rule themselves as “partners”.^{Footnote 69} It is, moreover, one whose government is legitimate, which means, for Dworkin, albeit not for some others, including Pettit and Buchanan, that obedience to its laws is morally required:

Dworkin thus maintains that democracy is sufficient (and, indeed, necessary) for political obligation. Community is, too—hence his prominence as an associative theorist—as is positive liberty and, for that matter, compliance with human rights^{Footnote 71} and even law itself, when properly understood as the voice of a principled society.^{Footnote 72} Since the conditions that he specifies for each—democracy, community, positive liberty, human rights and law—are identical, all must exist to the same extent.^{Footnote 73} Their co-dependence follows from their mutual subordination to equal concern and respect, which is the political manifestation of dignity and, consequently, the “sovereign virtue of political community”.^{Footnote 74} This abstract ideal gives meaning to every element of political morality, not least democracy, which “confirms in the most dramatic way the equal concern and respect that the community together, as the custodian of coercive power, has for each of its members.”^{Footnote 75} By construing democracy as part of the unified realm of value over which dignity reigns, Dworkin understands it so broadly that, in the event that it ever exists,^{Footnote 76} the people whom it governs necessarily owe obedience to its directives.

Against Monism

In this final section, I challenge Dworkin’s expansive interpretation of democracy and thereby his distinctive belief in its sufficiency for political obligation by contesting the value-monism—the hegemony of dignity—on which his interpretation depends. Following Isaiah Berlin, I suggest that political morality should be understood, not as a coherent set of values between which any tensions are merely apparent, but as a variety of ideals that can genuinely conflict. I start by outlining Berlin’s conception of morality. I then claim that Dworkin’s dismissal of it is too hasty. I query his assumption that it lacks methodological support by indicating—I cannot do more here—an alternative to his interpretive approach that he fails to consider and that arguably upholds Berlin’s view—a view with which Dworkin’s broad conception of democracy and the political obligation that it entails do not tally.

According to Berlin, the harmony between moral, including political, values that Dworkin and other theorists assert is “demonstrably false.”^{Footnote 77} He states:

Berlin denies that practical obstacles, which presumably include limited time and imperfect rationality, cause real clashes between moral ideals. To the extent that congruence is a theoretical, though not an immediate, option, genuine conflict is absent. Instead, says Berlin, the latent incompatibility of values is a property of morality. More specifically, he thinks that the impossibility of “perfect harmony” follows from value-pluralism.^{Footnote 79}

This view of morality entails the prospect of conflicts between the values of which it consists. It is, therefore, not equivalent to, even if it is consistent with, the “reasonable pluralism” that John Rawls discusses.^{Footnote 80} Whereas value-pluralists focus on tensions between moral ideals, Rawls is concerned with disputes between moral agents. Reasonable pluralism, for Rawls, means a diversity of moral opinions. In contrast, Berlin’s pluralism is a theory of the values about which people disagree. These two types of pluralism are distinct.

Berlin’s theory can also be differentiated from relativism. That values often clash does not mean that they exist only within particular, rather than across all, cultures. Indeed, Berlin insists on the universality of some ideals.^{Footnote 81} But one can separate value-pluralism and relativism without rejecting the latter. To make this distinction, one might simply note that value-pluralism (and, for that matter, value-monism) is consistent with both universalism and relativism.

If value-pluralism is neither moral disagreement nor relativism, then what is it? Its proponents allege not only that morality consists of numerous values—this opinion is consistent with value-monism—but also that many values have inherent worth.^{Footnote 82} The distinctive claim of value-pluralists is that various ideals are intrinsically significant. While value-monists place no more than one value in this category, theorists such as Berlin hold that many values are ends-in-themselves.^{Footnote 83} Insofar as some of the ideals to which value-pluralists attribute intrinsic worth actually share the same name—according to Berlin, for instance, at least two such values are called “liberty”^{Footnote 84}—their total number is even greater than one might initially suppose. But it might be fewer than the sum of all moral ideals, since value-pluralists need not ascribe inherent worth to every value.

Crucially, no rational comparison is possible between values whose worth is intrinsic. Such appraisal requires another value in terms of which their relative significance might be assessed. But the necessary homogeneity cannot exist between values that are ends-in-themselves. Since the worth of these ideals is not conditional on the degree to which they meet an additional norm, they are incommensurable. The absence of a common standard by which to evaluate them entails that none can be better than or equal to any other.^{Footnote 85}

This incomparability accounts for the potential conflicts between moral ideals that value-pluralists describe. The authenticity of these clashes follows from the incommensurability of the values involved. No further value can resolve the incompatibility of values whose worth is intrinsic to them. Berlin thus dismisses the prospect of reconciliation between, most notably, two kinds of liberty: the positive sort, which he equates with self-government and, like Dworkin, connects to democracy, and the negative sort, which he defines as the absence of human interference and associates with civil liberties.^{Footnote 86} By ascribing intrinsic worth to these two ideals he denies that a third might prevent genuine conflict between them: “[N]egative and positive liberty are not the same thing. Both are ends in themselves. These ends may clash irreconcilably.”^{Footnote 87}

Yet value-pluralists do not suppose that all values collide. In fact, Berlin notes that “[a] good many values are perfectly compatible”.^{Footnote 88} Conflict obviously cannot exist between an instrumental value and its end. Moreover, some intrinsic values might always fit together. Notwithstanding their incommensurability, they may never actually be at odds with one another. Were they to clash, however, their inherent worth would ensure the authenticity of the conflict between them. Finally, collisions between instrumental values with a common end are merely apparent. For subordinate values really to clash, they must be means to different and contradictory ends. Their incompatibility is parasitic on (and, therefore, might be translated into) friction between the incommensurable values from which they derive.

Whenever genuine conflicts between values do occur, though, loss is a necessary consequence. A “trade-off” is impossible between two ideals whose worth is inherent: more of one cannot compensate for less of another. Instead, one of these values must be “sacrificed” if they clash.^{Footnote 89} It must be relinquished, rather than exchanged, for the other. Berlin recognises that both of these values cannot be preserved and that the unavoidable loss of one might even, but need not, be tragic:

Hence, Berlin denies the possibility of moral unity. He holds that many values, including positive and negative liberty, are ends-in-themselves. The incommensurability of these ideals explains his belief in the authenticity of clashes between them. His value-pluralism thus contradicts an account of morality to which Dworkin and many other philosophers subscribe.

According to Dworkin, however, Berlin and his fellow value-pluralists fail to justify their understanding of morality. They cannot, he says, simply assume that values might conflict: to do so “begs the question”.^{Footnote 91} He warns them against “any lazy conclusion” about the character of morality and declares that they “must do the work” on which their theory depends.^{Footnote 92} He asks how they might account for the possibility of clashes between moral ideals. Given the persistent lack of consensus about the meaning of these values, he denies that the language of moral actors, whether past or present, can be decisive.^{Footnote 93} He thereby dismisses lexicography as a method on which a value-pluralist might rely. Moreover, he rejects the possibility, which he calls “nonsense”, of a quasi-scientific analysis of moral values.^{Footnote 94} Instead, he insists on analysis of a different sort. He believes that the meaning of a moral value is a moral issue. It is a question, he says, “not of dictionary definition or empirical discovery but of substantive moral and political philosophy.”^{Footnote 95}

Yet one might think that Berlin practises interpretation of just this sort. Although his methodology is not obvious—as George Crowder remarks, “he nowhere provides any extended defence of pluralism’s truth”^{Footnote 96}—one might take him to engage in a consequentialist form of such analysis when he avers that a value-pluralist is less likely than a value-monist to bring about tyranny.^{Footnote 97} Then again, Berlin might simply regard this alleged effect as an incidental benefit of, as distinct from a justification for, his account of morality. Besides, as Steven Lukes observes, he offers no reason to associate belief in value-monism more closely with this political outcome.^{Footnote 98} He does not establish a necessary connection between the tyrannical inclinations of the value-monists to whom he refers and the type of moral theory that they endorse. In addition, he says nothing about the terrible conclusions with which value-pluralism is arguably consistent. According to Dworkin, a value-pluralist might tolerate extreme poverty due to a perceived conflict between its eradication and the liberty of the rich.^{Footnote 99} Hence, even if Berlin employs moral analysis, he does so unconvincingly.

Moreover, he does so rather differently from Dworkin, who looks not to practical consequences when making sense of moral values, but to their fit with one another. To make sense of any value, Dworkin says, one must “understand what is good about it.”^{Footnote 100} In other words, “if we want to understand what freedom or democracy or law or justice really is, we must confront the difficult question of how to identify a value’s value.”^{Footnote 101} From this, he infers “that a defense of some particular conception of a political value like equality or liberty must draw on values beyond itself: it would be flaccidly circular to appeal to liberty to defend a conception of liberty.”^{Footnote 102} Therefore, his account of negative liberty—“freedom to spend your own rightful resources or deal with your own rightful resources in whatever way seems best to you”^{Footnote 103}—is an “aspect” of his view of equality,^{Footnote 104} which, in turn, is “sensitive” to it.^{Footnote 105} They are, for him, “interconnected”.^{Footnote 106} Both are also fused with his account of the rest of the ideals, including positive liberty, of which political morality consists. He stresses the need for “interpretations of each of these values that reinforce the others—a conception of democracy, for example, that serves equality and [negative] liberty, and conceptions of each of these other values that serves democracy so understood.”^{Footnote 107} Finally, political philosophy “must aim to construct these political conceptions, moreover, as part of an even more inclusive structure of value that connects the political structure not only to morality more generally but to ethics as well.”^{Footnote 108} The principles of dignity enable the realisation of this aim, in his opinion.

But is such moral interpretation, of which value-monism seems the inevitable consequence, really necessary? Dworkin appears justified in dispensing with both lexicography and quasi-scientific analysis as means by which an account of morality might be justified, but he is wrong to think that the sole remaining option is to portray moral ideals “in the light of each other.”^{Footnote 109} He ignores “pure”—that is, non-moral—interpretation as another way in which a philosopher might defend his or her moral theory. This form of analysis involves exclusive reliance on “meta-theoretical” norms, of which, in my opinion, clarity, coherence, comprehensiveness and consistency are primary. I elaborate on this methodology elsewhere.^{Footnote 110} Here, I ask only whether it supports value-pluralism.

One might doubt that it does so, given Berlin’s renunciation of “neutral conceptual analysis” and his emphasis on the “anti-marxist” implications of his value-pluralism.^{Footnote 111} But a description of morality whose justification is non-moral might nevertheless be contentious. Moreover, Berlin occasionally alludes to meta-theoretical norms. He invokes the value of comprehensiveness, which concerns the extent to which a theory fits with conventional beliefs, when he points to the “ordinary experience” of moral dilemmas.^{Footnote 112} He also mentions the value of clarity as a reason for his understanding of “the essence of the notion of liberty”.^{Footnote 113} Despite these references to meta-theoretical criteria, however, his case for value-pluralism remains ambiguous. Whether Berlin relies on these norms alone or appeals to them as part of a moral analysis—recall his speculative comments on the tyrannical inclinations of value-monists—is not evident. Yet the fact that Berlin does not obviously base his moral theory entirely on meta-theoretical norms does not preclude such support for value-pluralism. Indeed, a philosopher whose understanding of morality is grounded solely on these criteria ought to deny that a single value, such as dignity, is the source of all the rest.

A pure theorist should, for instance, regard value-monism as insufficiently comprehensive due to its rejection of the moral dilemmas that many people seem to endure. Only value-pluralism respects this prominent feature of moral practice. “It fits with salient aspects of modern moral experience,” observes Crowder, “in particular with our sense of the multiplicity of genuine values, and of the distinctness of those values which is highlighted by those cases where we have to choose among them.”^{Footnote 114}

Of course, Dworkin would respond that a philosopher must interpret and cannot merely report these apparent dilemmas. Yet comprehensiveness is not the sole ambition of a pure theorist: he or she also strives for clarity, consistency and coherence. Value-pluralism appears to satisfy these criteria, too. It is clear insofar as it refines the ideals to which moral actors subscribe—Berlin’s distinction between positive and negative liberty is the most obvious example of such precision—and it is no less likely than value-monism to be consistent.^{Footnote 115} Because value-pluralists might combine some values—they need not hold that ideals always have the potential to collide—their theory is also coherent to the extent that they do so. Of course, their understanding of morality would be even more coherent were they simply to renounce value-pluralism and to treat all and not just some values as compatible. A pure theorist ought to resist this move, however. To conclude that moral ideals never conflict, he or she must exaggerate the methodological significance of coherence.

Much more must be said about value-pluralism as a pure theory of morality, not least about the resolution of the dilemmas that it has the potential to generate. I hope, though, that my brief remarks on it are nevertheless adequate to prompt scepticism about Dworkin’s value-monism and, therefore, his expansive conception of democracy. If, as I propose, morality is understood to consist of discrete values with the potential to conflict, then democracy cannot be thought to incorporate everything that a good-faith commitment to dignity implies. Conceived more narrowly, such government might be part of an argument for a moral obligation to obey the law. It might not even be that—I need not say here—but it surely cannot be any more. Hence, the sufficiency of democracy for political obligation remains to be established.