Abstract
The central contention of this article is that contemporary liberal theory is without an account of what legitimates coercing those who reject liberalism that is consistent with its own stipulations of the conditions of political legitimacy. After exploring the nature of the liberal principle of legitimacy, and in particular how it is intended to function as a way of protecting individuals from domination and oppression by reconciling freedom and public law, the article considers four different possible accounts of what might legitimate coercing non-liberals. While some of them have independent plausibility, the article argues that none of them meets the requirements of liberal legitimacy. The final section of the article considers the implications of this theoretical gap for liberal theory more widely. The argument is made that liberalism must accept that even liberal politics will necessitate the oppressive use of coercive power, i.e. compelling people to live according to wills other than their own, and that insofar as this is a position central to the recent burgeoning literature on political realism, liberalism ought therefore to be more realist.
The question this article addresses is ‘On what grounds is it legitimate to coerce people to comply with the laws of a liberal state if they reject the fundamental principles, values and norms of liberalism upon which it is founded?’ This is a classic problem in liberal political philosophy which presses liberalism on a series of its most fundamental commitments. 1 Yet it is one that has received very little attention in the contemporary literature, 2 largely, I suspect, because much liberal theory tends to work within ‘ideal theory’, which assumes that everyone accepts, and complies with, the ‘correct’ principles of justice. 3 The question of what legitimates coercing those who do not accept the political principles that regulate the basic structure of society simply does not arise in any interesting way. Nevertheless, the lack of attention that this question has received is surprising, given that one of the most serious political issues that governments across the world are currently struggling with is how to deal with or respond to the various non- and anti-liberal groups (many, though by no means all of them, religious fundamentalist) that have burgeoned in recent years. Regardless of the policies pursued, their effectiveness or the question of whether they might violate the rights (human or civil) of members of these groups, the general theoretical question of what legitimates coercing non-liberals is raised.
That liberal political theorists have paid little attention to the question of what legitimates coercing those who reject liberalism is indicative of a more general tendency in liberal thought to overlook the vital role and function of coercion in political life, something which has been of much greater concern for ‘realists’ or those advocating a ‘politics of compulsion’. 4 Whereas, as we shall see, the liberal vision of the political is one in which legitimacy functions as a constraint on political power such that coercion becomes consistent with individual autonomy, effectively reconciling coercion and freedom, realism is a tradition in which coercion, the imposition of one will upon others, is taken to be an inevitable feature of any political order. 5 The argument of this article will lend support to this realist contention and, in doing so, suggest that coercion, being forced to live according to a will other than one’s own, is an inevitable and ineradicable feature of political life which liberal theory needs to take seriously, or at least more seriously than it often does. Liberalism needs to be, in this sense, more realist. But this imperative to pay more theoretical attention to coercion is made more urgent by the fact that, as I shall argue, contemporary liberal theory, especially in its prevailing ‘political’ form which is the focus of this discussion, 6 does not have an answer to the question of what legitimates coercing non-liberals that is consistent with its own account of legitimacy. As such, the coercion of non-liberals will always represent a violation of their freedom and be, on liberalism’s own terms, an act of oppression or domination. The upshot of this argument is not intended to be that liberalism abandons its foundational normative commitments but that it recognizes politics as an activity which necessarily requires rule in the sense of one party coercing others to abide by its will.
Before commencing, it is worth just clarifying three aspects of my argument: first we need to be clear who it is that is the focus of this discussion, or which individuals we are to be concerned with in terms of the legitimate use of state power. It is quite common to see debates like this couched in terms of unreasonable persons, which is a reference to the category as employed in the later work of John Rawls. 7 Reasonableness is, however, a term of art and there is a plethora of different definitions within the literature; Rawls’s own account is too specific for our purposes, situated as it is within the general and broad structure of his political thinking. 8 Though this article will talk about unreasonable persons in relation to Rawls’s work, it will speak predominantly of non-liberals, by which will be meant those who reject the commitment to the public justification of the principles that regulate our shared political association in line with the liberal principle of legitimacy, which is at the heart of political conceptions of liberalism (which we shall explore in the following section). This rejection will often lead non-liberals to recommend illiberal visions of political association which are inconsistent with the liberal conditions of legitimacy, ones which e.g. seek to prohibit persons from pursuing certain ways of life or give the state a paternalistic role in the shaping and fostering of moral beliefs beyond the limits that liberals would generally tolerate. As such, comprehensive or perfectionist liberals, even though they endorse a liberal political order, will of course also count as non-liberals for the purposes of this article. 9
Second, it is important to be clear that I am not claiming that liberals fail to recognize the need to coerce non-liberals. No sensible liberal, indeed no sensible political theorist, would deny the need for the institutions and mechanisms of coercion (police, judicial system, code of law, etc.) necessary to enforce the rules of the state upon those who reject and/or contravene them. My claim is rather that liberalism lacks an account of what legitimates such coercion that is compatible with its own account of what legitimacy demands. Finally, it is also worth clarifying that, following several other authors, I take the question of political legitimacy, what makes it permissible for the state to coerce those who disobey its legal commands, to be distinct from, though related to, that of political obligation, what citizens are obligated to do and why. 10 This article will explore the issue of whether liberalism can provide a compelling account as to what legitimates the use of political coercion over non-liberals and will not be concerned with the question of whether such non-liberals have an obligation to obey the liberal state.
The article will proceed in the following way. First it will explicate the normative conditions that liberalism places on the legitimate employment of state power. This section will also establish why the structure of contemporary liberal theory raises the question of the legitimacy of coercing non-liberals in an urgent and interesting manner. The following section will then examine four different responses to this issue and argue that none of them is sufficient from a liberal perspective. All four are responses that I take to be consistent with the general structure of political liberalism, responses that political liberals could have given according to their own terms, if you like, though nowhere (to my knowledge) have such accounts been explored. Lastly I will then discuss how the inadequacy of these accounts indicates the need for liberalism to amend its understanding of the nature of politics and the role of coercion in political life.
Legitimacy, coercion and freedom
Liberalism is a politics of limits, both normative and institutional. Historically, liberalism developed as an attempt to defend the individual, first from the overwhelming and irresistible power of the modern centralised state (as we now call it) and then, from the 19th century onwards, the tyrannical possibilities inherent in majoritarian democracies. 11 The liberal concern with both of these developments lay particularly with the fate of individual liberty in conditions where the possibilities of pursuing policies of tyranny, oppression and domination were both numerous and relatively easy to implement, given the mechanisms of coercion afforded to those who held the offices of power in the modern state. The question arose, therefore, how all people can (given that liberals assume that, from a moral point of view, the prince deserves equal concern to the pauper) live together in modern political societies yet remain free.
One of liberalism’s most influential responses to this problem came through theoretical innovations that were initially introduced by Jean-Jacques Rousseau, though most fully explicated and developed in the work of Immanuel Kant. The central idea was to reconceptualize human freedom as, in Kant’s words, ‘independence from being constrained by another’s choice’, the right every man possesses ‘by virtue of his humanity’, he believed, to be independent from coercion by another’s will. In the Groundwork of a Metaphysics of Morals, Kant famously argued that man is not only a means for the arbitrary use of others’ wills, but rather ‘must in all his actions, whether directed to himself or also to other rational beings, be regarded at the same time as an end’’. 12 From this postulate followed Kant’s second formulation of the categorical imperative, the objective principle of morality: ‘So act that you use humanity, whether in your person or in the person of any other, always at the same time as an end, never merely as a means.’ 13
In relation to the political, or juridical, sphere, this understanding of freedom meant that we treat others as means if we force them to obey public laws which are not the will of their own. Human beings are only free, or autonomous, if they obey laws that they can be understood to have made for themselves, to have self-legislated. If this condition is not met, then the laws will violate individuals’ autonomy insofar as they force them to live according to the will of another. Part of the genius of Kant’s argument is that it reconciles the use of coercive political force with individual freedom by insisting that public laws are autonomously chosen bonds. It ensures that coercive political power is only used in ways that are consistent with, rather than a violation of, individuals’ freedom. This is true even if particular individuals need to be compelled to obey the law for, as Jeremy Waldron has excellently put it, ‘If the rule is one that the citizen has agreed to, surely little that is important in relation to liberty is lost if it is subsequently enforced against him’. 14 The function of legitimacy, as a concept, is then to provide normative moral limits and constraints on political power such that it is employed only in the service of principles which respect the freedom of all those subject to it, and directed away from its potentially oppressive and tyrannical uses. The question of whether the use of state power is legitimate becomes one of whether it is used in accordance with laws or principles which respect the freedom and moral equality of all citizens by treating them as ends in themselves. 15
Contemporary liberal theory has inherited both the normative concern regarding individual freedom which motivated the development of liberal thought, as well as the general theoretical strategy of conceptualizing freedom, coercion and legitimacy in Kant’s unique and specific way. Though Rawls’s later work on political liberalism was a conscious attempt to divorce his theory from what he saw (rightly or wrongly) as its earlier grounding in a Kantian comprehensive doctrine, his liberal principle of legitimacy clearly reflects these underlying commitments. He writes: ‘our exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in the light of principles and ideals acceptable to their common human reason’. 16 Waldron has gone so far as to call this account of legitimacy the ‘fundamentally liberal thesis’, that which distinguishes liberalism from all other forms of political order: ‘a social and political order is illegitimate unless it is rooted in the consent of all those who have to live under it; the consent or agreement of these people is a condition of its being morally permissible to enforce that order against them’. 17 That this commitment to public justification is grounded in moral concerns derived from Kantian theory has probably been most explicitly illustrated by Charles Larmore, such as when he writes, ‘To respect another person as an end is to insist that coercive or political principles by as justifiable to … [other] person[s] as they are to us. Equal respect involves treating in this way all persons to which such principles apply.’ 18 This general notion that the legitimacy of the political order depends upon it being acceptable to all those subject to it, the commitment to public justification, is a familiar feature of contemporary liberal theory.
It is important to note that liberal legitimacy does not require that the constitutional essentials be actually accepted by all citizens. Such a condition is unreasonable in large and diverse societies such as our own and would essentially give non-liberals a veto over the legitimacy of the political system. Rather, the relevant condition is whether the constitutional essentials can be represented as principles which are acceptable to those subject to them. As long as the political order can be justified to such persons as something they should assent to then this is sufficient for the purposes of legitimating the use of coercive force. Whether they are actually accepted is extraneous to the question of political legitimacy. This, of course, has important ramifications for the issue that interests us here, as the mere fact that non-liberals reject liberalism does not automatically deem their coercion by a liberal state illegitimate. The liberal claim will be that, though they reject constitutional principles that are publicly justified in a manner consistent with the liberal principle of legitimacy, they should, in some sense, accept them, or it is at least reasonable to expect them to do so. Whether this response is fully satisfactory will be examined in the following section.
Thus the liberal account of legitimacy is premised on the idea that the principles according to which state power is employed must be compatible with persons’ autonomy, and so it requires that those principles are acceptable to those subject to them. The normative function of legitimacy in liberal theory is to resolve the tension between freedom, understood as living according to no one’s will other than your own, and the need for public law backed up with the threat and possibility of coercive enforcement, and in doing so the individual remains autonomous while living together with others under a common political authority. This ensures that political power is not oppressive but respects the freedom of all those over whom it is employed. Or, put the other way around, forcing someone to live according to principles or laws which are not acceptable to them would necessarily be an act of oppression and domination insofar as it compels individuals to obey another’s will.
Hopefully it will be clear why, at least prima facie, the question of the legitimacy of coercing non-liberals represents a significant issue for liberal political theory. But it is important to appreciate that the significance of this question for our purposes does not come from the direction of the non-liberal asking why the liberal state is legitimate in coercing their compliance, but from the perspective of the consequences and implications it generates for liberal theory itself. This is because, unless it can be shown that coercive power can be used over non-liberals in a way that is consistent with their autonomy, then liberals have to accept that liberal political orders do, and despite their best efforts to the contrary, invariably necessitate the use of oppressive coercive force, i.e. coercion which violates the autonomy of those over whom it is used, as a mechanism for ensuring compliance with the law and to maintain order, even if this is only true for a minority of individuals. Larmore states the importance of this issue for liberal theory, and its deep connections with Kantian morality, well when he writes: … the use or threat of force cannot be deemed wrong in itself, for then political association would be impossible. What we [liberals] must regard as improper is rather to seek compliance by force alone, without requiring reasonable agreement about the rules being enforced. For consider the basic fact that persons are capable of thinking and acting on the basis of reasons. If we try to bring about conformity to a rule of conduct solely by the threat of force, we shall be treating persons merely as means, as objects of coercion, and not also as ends, engaging directly their distinctive capacity as persons … Thus, to respect another person as an end is to require that coercive or political principles be as justifiable to that person as they presumably are to us. This is certainly not the only sense we can give to the rich moral notion of respect. But it is the one which liberals must regard as relevant from a political point of view.
19
Four accounts
Reasonable acceptance
As we have seen, the liberal principle of legitimacy requires not that all persons actually accept the constitutional essentials but that they are principles which it is reasonable for all persons to accept (in fact Rawls claims that this only needs to be true of reasonable citizens, something I shall return to in the ‘irrelevance’ account). As Bohman and Richardson have recently argued, empirical interpretations of what it is reasonable to expect others to accept are made deeply indeterminate by the fact that there are no reasons that, as a matter of empirical possibility, are intelligibly accepted by one person that could not be accepted by any other. We can more than plausibly imagine cases, for example, of liberals who come to accept a form of fundamentalism that declares autonomy, rights, moral equality, etc. sheer anathema (and vice versa). Hence the empirical possibility of persons being able to accept reasons does not provide the grounds for excluding any reasons as unacceptable that this approach to establishing political legitimacy requires. 20 And so the ‘reasonable acceptance’ account more often claims that the legitimacy of coercing non-liberals derives from the fact that the liberal political order can be represented as something which they would affirm if they held the right sort of moral norms or dispositions, or were motivated by the right sort of reasons, such as reciprocity or equal respect. If successful, this would enable liberal theorists to argue that it is legitimate to coerce non-liberals on the basis that, though they do not actually endorse liberalism, they ought to do so, and hence in compelling them to obey the law they are being forced to obey their true rather than their actual will (what they should will rather than what they do will). They are, in effect, being forced, if not to be free, then to live according to political principles consistent with the moral norms that they should affirm.
A comparison between contemporary political liberalism and Kant’s political philosophy might be instructive here in showing why this account is not as plausible as its recent popularity would imply. Kant recognized that, at the empirical level, it looks very difficult to identify ends or norms that all are committed to in conditions where people endorse a variety of different and often conflicting moral and religious views. Kant negotiated this obstacle by distinguishing political unions from all others on the grounds that, while humans come together in most unions so as to pursue ends that they contingently happen to share (such as the love of chess in chess clubs or of literature in reading groups), political unions are directed towards an end that all persons ‘ought to share and which is thus an absolute and primary duty in all external relationships whatsoever among human beings’. 21 This end which unites the will of all turns out to be, for Kant, the right of men to live together under coercive law yet with the liberty to pursue their own conception of the good life free from the infringements from others. 22 In other words, all persons ought to be committed to autonomy and it is this end, one which all ought to share even if they do not, that provides the normative foundation for civic unions. Though Kant believed that the dignity of human freedom as expressed in the universal principle of right does impose an obligation upon us, it is not one that he expects, nor requires, us to act in accordance with. But because it is morally necessary, indeed an absolute and primary duty, for us to realize human freedom, we can be legitimately compelled by others to enter civil society and live according to the principles of right which maintain it.
The theoretical move that is crucial to Kant’s argument here is that we have a categorical rather than hypothetical duty to realize human freedom, the justification for which is tied up in his wider comprehensive philosophical and moral system. Kant was insistent that such duties could not be hypothetical as that would make them dependent, and ultimately contingent, upon individuals’ actually having the beliefs or ends that justify that duty. If an individual lacked such ends then he would, on the hypothetical account, lack such duties also. So our obligation to realize human freedom must, for Kant, be categorical in the sense that it does not depend on the existence of prior commitment to particular ends or norms. And it is the categorical nature of our political duties which justifies the use of coercion even over those who reject them.
Interestingly, much contemporary liberal theory has proceeded on the empirical assumption that people share a certain motivational disposition to treat others in ways consistent with publically justifiable principles and that this provides an appropriate basis upon which to justify a liberal political order. It is a familiar theme of the realist critique of liberalism that this assumption is fundamentally misguided and that people act on a plurality of different motives, not all of them necessarily consistent with respecting others via public justification, and that political theory ‘must start from and be concerned in the first instance not with how people ought ideally (or ought “rationally”) to act, what they ought to desire, or value, the kind of people they ought to be, etc., but, rather, with … what really does move human beings to act in given circumstances’. 23 Yet even when liberals themselves question the empirical validity of this assumption, little thought has been given to the implications of this theoretical strategy for those who reject or lack this disposition. Rawls’s distinction between reasonable and unreasonable persons, for example, is an explicit acknowledgement that not everyone shares the moral commitment to reciprocity which underpins the concern for public justification. Yet he does not pursue the ramifications of this distinction.
Charles Larmore is more frank about the fact that the concern for treating others with equal respect, which is the central norm in his theory which generates the demands of public justification, is not universally shared. He has rightly acknowledged that, while the demands generated by seeing others as deserving of equal respect might be transparent and obvious to us (liberals), it has nevertheless met with sincere rejection even from fully rational others. 24 Unlike Rawls, however, Larmore does spare a brief thought for those who lack this disposition and the implications for them of the reliance upon equal respect in liberal political theory. He writes: ‘Nothing has been said about how we ought to converse with those who refuse to show us respect. This limitation is important since with those who reject the norm of equal respect or rank their view of the good life above it, we will usually be unable to converge on any political (coercive) principles that are as justifiable to them as to ourselves.’ Yet his response, especially when seen in the light of his concerns regarding ensuring that liberal political associations are not simply characterized by the rule of force but as the rule of principles that people recognize as expressions of their own will, seems a little blithe: ‘What should be said about the less-than-ideal case is not entirely obvious. The best tack is probably to hold that political principles should be justifiable to these people as well, though with the justification premised on the (counterfactual) supposition that they do prize the norms of rational dialogue and equal respect.’ 25
This begs more questions than it answers. An awful lot of work is required both to explain the grounds on which such counterfactual justifications can be based and to justify why they trump, for the purposes of thinking about political legitimacy, the norms and beliefs non-liberals actually hold. Why should we think that an assumed set of counterfactual beliefs can in reality alter the nature of the relationship between the state and its citizens? And, of course, the possibilities afforded by counterfactuals cuts both ways and opens up the prospect for non-liberals to make exactly the same claims in relation to liberals (e.g. a theocratic political order is exactly what others would endorse if they were believers of the true faith). In the face of such counterclaims, it is not clear how we can privilege the liberal position. So without being able to say why these counterfactual beliefs represent, in some sense, non-liberals’ true or real will, it is left obscure how such an argument is supposed to lend legitimacy to the use of coercive power over them.
Kant, of course, has no need to appeal to counterfactuals in order to explain the legitimacy of coercing non-liberals. His answer to this general dilemma is not to imagine that non-liberals are actually committed to the norms of liberalism, but to see them as bearers of categorical duties which they can be compelled to execute (whether they accept them or not). This allows us not only to privilege the liberal claim but also to do so in a way that is consistent with the liberal principle of legitimacy. For if you treat non-liberals as if they were committed to liberalism, then you are necessarily imposing an alien will upon them; but if you can claim that they actually have a duty to be committed to liberalism, and importantly that it is irrational for them not to affirm it, then (assuming the rationality of the will) coercing obedience to liberal laws is merely compelling people to obey their true will. Larmore’s mistake, and one echoed by other liberals also, is to fail to move from hypothetical imperatives, with their reliance upon persons’ actual ends and beliefs to generate duties, to categorical ones. And as such, he is ultimately unable to reconcile non-liberals’ freedom with coercion as liberal legitimacy demands.
But then there is very good reason why contemporary liberals do not follow Kant’s strategy on this point: it is only plausible within the context of Kant’s wider comprehensive moral doctrine and this represents but one of a plurality of controversial comprehensive doctrines that citizens affirm in modern liberal democratic societies, over which liberals believe it is reasonable to disagree. The move to political liberalism was motivated by the very concern that it is illegitimate to force people to live according to constitutional essentials that are justified with reference to moral or religious beliefs or doctrines that citizens can reasonably disagree with. Reliance upon Kant’s moral philosophy to give legitimacy to coercing non-liberals would therefore sit in some considerable tension with the account of moral pluralism which forms the backdrop against which liberal political takes place. Indeed, the problem that contemporary liberal theory addresses, especially in its dominant ‘political’ variant, is how people can agree upon the fundamental principles which regulate their shared political community in conditions of deep, ineradicable yet nevertheless reasonable disagreement about the good. The truth of Kantianism, indeed of any comprehensive moral doctrine, will be a self-defeating strategy to employ when moral pluralism is supposed to present the problem that liberal political theorizing is intended to address.
Furthermore, while normatively constrained accounts of what reasons persons can accept allows us to exclude particular reasons on the grounds that they fail to conform with the substantive moral commitments employed, such as reciprocity or mutuality, as Bohman and Richardson say, ‘When normative constraint is imported, … all the important sorting work is done, not by the idea of reason as that all “can accept”, but rather by the relevant normative notion, such as consistency with the requirements of reasonableness’. 26 We could stipulate that by ‘reasons all can accept’ we actually mean ‘reasons consistent with the constitutive requirements of reasonableness’. But doing so would indeed ‘be a needlessly confusing terminological move’. 27 A much simpler, less confusing and more direct approach would be to require that reasons be such that they cohere with the substantive standards. But, crucially, this means that the normatively constrained account of what counts as ‘reasonably acceptable’ essentially provides a basis for assessing the permissibility of particular reasons as the grounds of political legitimacy that can be articulated completely independently of the test of what could be accepted. Insofar as this normative strategy effectively asks us to determine, more abstractly and in general, whether a given reason is consistent with the constitutive commitments of reasonableness, because these will be same for each person, the idea of what all could accept ‘drops out’. 28 And, in doing so, the connection between individuals’ wills and political legitimacy is effectively severed.
Political liberalism’s self-imposed restraint from appealing to the truth of comprehensive moral doctrines leaves it without the theoretical resources to justify why non-liberals ought to be committed to liberalism despite their actual rejection of it. It limits liberal theory to dealing with people’s beliefs essentially to as it finds them and therefore unable to appeal to the sort of Kantian arguments regarding absolute and primary duties, categorical imperatives and so on, which have and can provide a response to this difficulty (leaving aside obvious questions about whether it is a convincing or coherent response). 29 The strategy of attempting to demonstrate that certain constitutional essentials are what non-liberals would accept if they were reasonable, committed to treating others with equal respect or engaging in equal dialogue, etc., is either to say that they are what such persons would accept if they held different beliefs, or, maybe more worryingly, that they are simply principles that are consistent with the demands of reasonableness without making any further connection between those principles and the will of the individuals who have to live under them. Either way, it is difficult to see how those constitutional essentials can be represented as being self-legislated. In this sense, while contemporary liberal theory has largely retained a Kantian theoretical structure insofar as it conceptualizes freedom, coercion, legitimacy and the relationship between them, in mostly the same manner, the attempt to retain the structure while stripping it bear of any metaphysical or moral philosophy content has left it unable to coherently make the same appeal to what people should believe, or what it is reasonable to expect them to accept, in order to justify how the coercion of non-liberals is consistent with their freedom.
Irrelevance
A different strategy to pursue would be to argue that the fact that non-liberals reject liberalism is irrelevant to the question of whether it is legitimate to coerce them, given that the liberal principle of legitimacy requires only that the use of state power be justified according to principles that are acceptable to liberals. Such a strategy of deeming certain people’s rejection of liberalism as irrelevant to the question of legitimacy could be construed as Rawls’s position on this issue. For Rawls, legitimacy only depends upon the endorsement of the reasonable. If the constitutional essentials are endorsed by the reasonable then the state can legitimately exercise its coercive power over unreasonable persons without their consent. 30 As such, non-liberal persons’ rejection of liberalism is simply irrelevant to the question of whether it is legitimate to compel them to obey the laws of the liberal state or not.
Rawls never went on to justify this position (assuming it was his), though Jonathan Quong has fleshed this position out in a compelling way. In setting out the theoretical parameters of his article ‘The Rights of Unreasonable Persons’, Quong states: I do not address the critique that political liberalism is illegitimate or incoherent because it involves the coercion of unreasonable citizens on grounds that they cannot accept. Political liberalism does not address itself to unreasonable citizens because it is a theory about the freedom and equality of citizens. Since unreasonable people by definition reject this premise, their (unreasonable) views are simply of no normative interest in the process of political justification. As we shall see, this does not mean that they are not entitled to the general rights and benefits of citizenship, only that they are not part of the constituency of public justification that determines what those rights and benefits will be.
31
This second claim is correct; in the same way that when deciding how to fairly cut a cake between several people I can discard the views of the person who wants the entire cake to himself, it seems right that the views of the unreasonable can be excluded as of no bearing on the question of which constitutional essentials are compatible with respecting the freedom and equality of others. However, it is unclear how that perfectly acceptable claim regarding the basis for excluding unreasonable persons from consideration in the justificatory process of the constitutional essentials can possibly explain or validate the legitimacy of coercing those same individuals to obey the principles which that process gives rise to. What is unclear is the relationship between the question ‘which constituency of people do the constitutional essentials of a well-ordered society need to be justified to (the justificatory constituency)?’ and ‘which constituency of people can legitimately be coerced to comply with the constitutional essentials of a well-ordered society (the legitimacy constituency)?’ Quong’s response implies that he does not think that these two constituencies need to map perfectly upon one another, such that it is possible not to be a member of the justificatory constituency but be included in the legitimacy constituency.
What is striking about this separation of the two constituencies is that it represents a significant deviation from the standard liberal understanding of legitimacy I explored earlier in this article, and would seem to signify a drastic change in the function of legitimacy in liberal thought. We saw that the liberal understanding of legitimacy responds to the problem of how individuals can be free yet live together in political societies regulated by common public principles backed up by the threat of coercive force. Legitimacy, as a concept, reconciled freedom and coercion by ensuring that political power be used in a way consistent with each and every person’s autonomy. The theoretical consequence of this was to imagine a form of politics in which the oppressive use of coercion (i.e. compelling others to live according to a will other than their own), and its resulting loss of individual freedom, is absent. That legitimacy responded to the particular concern about reconciling freedom and coercion via the concept of autonomy meant that the two constituencies I described, that to whom the constitutional essentials must be justified to and that over whom coercive power can be legitimately employed, necessarily became identical. And, in doing so, liberals ensured that the constituency of people whom the state can legitimately compel to obey its laws is the same group of people who can live according to those laws yet remain free (even when coerced to obey them).
Making the legitimacy constituency smaller than the justificatory constituency, and essentially no longer taking the two groups to be synonymous with one another, significantly disrupts this picture. It allows for the possibility that there are people within the political community who can legitimately be coerced to obey the laws of a liberal state despite the fact that they reject the principles according to which the political coercive power is employed. While there might be independent plausibility to this model (realists certainly think so), it is, I contend, a substantial concession for liberal theory to make insofar as it accepts that some people are not to be treated as moral equals, that there are some who will be treated as means rather than ends in themselves. But while I want to leave open the question of whether separating the two constituencies in the way this irrelevance model does is a desirable move or not (though I do want to stress the importance of it for liberal thought more widely), crucial for our purposes is the fact that it does not actually provide us with any grounds for thinking that the coercion of non-liberals is in any way consistent with their freedom, and hence not an act of oppression or domination. Indeed, though we may rightly think that the views of non-liberals are irrelevant to the question of which liberal constitutional principles should regulate the basic structure of society, this draws further attention to the fact that coercing such people is necessarily forcing them to live according to principles that they do not accept. And so this account can only work by distinguishing between the constituency of persons to whom the constitutional essentials must be justified and those over whom coercive power can be employed in order to ensure compliance with those essentials, a distinction which the liberal principle of legitimacy purposefully does not make in order to ensure that coercion does not become oppressive by compelling people to obey laws that they reject.
Forfeiture
A third possible account could be based on the notion of forfeiture, that non-liberals’ rejection of liberalism entails them forgoing the requirement that the fundamental political principles be justified to them as the liberal understanding of legitimacy demands. This account could be stated in either a strong or a weak form depending on what it is we think non-liberals have forfeited. The strong version of the argument would be that non-liberals forfeit their moral status as free and equal persons and thus the demands of reciprocity do not arise with such individuals (in the same way that they do not apply to other living creatures that lack the appropriate moral standing). The weaker form of this account would be that non-liberals retain their moral status as free and equal persons though they forfeit their political status as citizens to whom the demands of reciprocity apply. Importantly, according to this version it is their political not moral status which non-liberals forfeit. Nevertheless, both accounts deny that non-liberals are people to whom the terms of the political association need to be justified in order for it to be legitimate to coerce them.
The strong version of the forfeiture account is of dubious liberal credence. Most liberals take freedom and equality to be essential moral qualities of persons qua their status as human beings which, as such, are inalienable in that they can be neither forfeited nor confiscated (though they can of course be ignored or violated by others). The weaker version of the argument is potentially more plausible insofar as it takes non-liberals’ rejection of liberalism to signify a forfeiture of their political status as citizens (or maybe full citizens) of a particular polity, which is a contingent rather than a necessary characteristic of persons and can therefore be forfeited in a way consistent with the liberal view of the voluntary nature of political membership. An interesting consequence of this account would be that liberals would remain obliged to treat non-liberals in ways which reflect the latter’s moral equality, for example, by not violating their basic human rights, while not being obliged to treat them as political equals by ensuring that the constitutional essentials are justified to them as liberal legitimacy demands. It would therefore be permissible to coerce non-liberals to comply with the laws of the liberal order, though it would be wrong to violate the rights due to them as human beings.
The weaker version of the forfeiture account is, I think, intuitively plausible. Importantly, it emphasizes the difference between moral and political obligations and in doing so explains why those who reject liberalism can be exempt from the demands of reciprocity though it might still be impermissible to treat them in particular ways (detention without charge, extraordinary rendition, or the use of coercive interrogation techniques being pertinent examples). But there are two fundamental problems with this account. The first is that the liberal principle of legitimacy holds in relation to all persons who are subject to the same political authority. This means, and here we meet a familiar criticism of consent-based theories of legitimacy, that unless persons actually leave the state then they are considered to be a member of that political community and, as such, subject to its power and authority. Their rejection of liberalism does not in and of itself mean that non-liberals are no longer members of the liberal state. So unless there are to be individuals who live amongst us and with us in the same society yet who are not subject to the laws of the land, the question of legitimacy must arise over all those whom reside within the borders of the state. However, even if it were the case that people could renounce their membership of a particular polity while remaining within its jurisdictional borders, the second related issue with the forfeiture account is that the liberal conditions of legitimacy are generated by the moral rather than the political status of persons; that is to say, the demand that the political fundamentals of a society be acceptable to those subject to them is generated by the moral concern that a person’s freedom is violated if they are forced to obey laws they do not accept. Though it is the case that the requirements of legitimacy only arise in relation to those with whom we share a liberal democratic polity, nevertheless the requirements or conditions of legitimacy are given by a pre-political moral commitment. Whom we owe a justification is determined by their political status as a member of the same political authority subject to the same political coercive power; that we owe them a public justification is determined by their moral status. As such, from the perspective of the liberal understanding of legitimacy, the weaker version of the forfeiture account still leaves unexplained how coercing non-liberals is consistent with their standing as free and equal persons. That non-liberals may have forfeited their right to have the political fundamentals justified to them still leaves us with the need to explain how our coercion of them is legitimate. A separate and distinct argument is still required.
Threat to stability
Rawls fully accepted that the presence of doctrines that reject liberalism and the democratic freedoms is a ‘permanent fact of life’. He went on to say that ‘This give us the practical task of containing them – like war and disease – so that they do not overturn political justice.’ 33 Rawls did not explain what policies of containment he was referring to here, though Quong has plausibly suggested that they should be understood as ‘any policy whose primary intention is to undermine or restrict the spread of ideas that reject the fundamental tenets of liberal democracy’. 34 He argues that ‘A justification for containment can be grounded … on the fundamental moral importance of normative stability in a liberal, deliberative democracy’, by which he means that it is vital citizens are able to recognize the constitutional essentials of a liberal political order as worthy of their endorsement for moral reasons internal to their specific conception of the good. And ‘It is therefore essential that doctrines which deny the freedom and equality of persons not become so prevalent that they threaten to undermine the essentials of a liberal democratic regime.’ 35 On this account, the coercion of non-liberal persons is legitimated by the fact that it prevents them from promulgating their non-liberal beliefs and destabilizing liberal regimes.
The first problem with this ‘threat to stability’ account is that it does not necessarily follow from non-liberals rejecting the liberal state that they will be engaged in the project of trying to undermine it. Many of those who reject liberalism, Marxists, anarchists, fascists, etc. are law-abiding citizens, albeit perhaps grudgingly, who choose not to promulgate their views. That is to say, though they might disagree with liberalism, nevertheless many non-liberals seek to get by, and maybe even flourish, in the liberal society in which they find themselves. They do not seek to undermine the liberal state and thus do not represent the sort of threat to stability that this account requires. The vast majority of non-liberals would, I think, fall into this category of reluctant or grudging obedience to the laws of the liberal states they reject rather than actively seeking to weaken or destabilize it.
But what about those individuals who do seek to undermine the liberal state through the transmission of their anti-liberal views, such as jihad poets or members of radical anti-democratic political parties, and hence can plausibly be viewed as threats to stability? Is it legitimate to coerce these non-liberals? The central difficulty with this account is that it lacks the necessary explanation of why the use of coercive force is consistent with their freedom and equality when they reject the fundamental moral and political principles of the liberal state. Quite clearly, such people will not be moved by the fact that their activities may undermine the liberal political order when that is their very intention. And perhaps neither should the fact that they reject liberalism affect our consideration of whether it is permissible to coerce these persons in the best interests of state stability. We might even wish to say that it is legitimate to do so. But the point is that this account still falls short of what liberal legitimacy demands insofar as it cannot connect the will of non-liberals with the principles according to which political power is employed.
Politics, realism and coercion
Unless one wants to hold to the normative position (akin to that of philosophical anarchism) that it is illegitimate for liberal states to coerce their non-liberal citizens, then the need for an account which can explain the legitimacy of compelling such people to obey the law is clear. The current absence of a plausible account represents, I think, an important gap in liberal theory, one made even more significant by the fact that several of the most pressing political problems facing liberal democratic societies seem to demand the coercion of those that reject liberalism. 36 While I rejected the reasonable acceptability account on the grounds that it was inconsistent with the background conditions of reasonable moral pluralism which characterize modern liberal democratic states, I do not want to deny that any of the other accounts previously explored could provide plausible independent accounts of the legitimacy of coercing non-liberals. That it is legitimate to coerce those who threaten to undermine liberal states or destabilize their fundamental political, social, or economic institutions seems intuitively plausible, for example. But the point I have been seeking to make is that such accounts are inconsistent with liberalism’s own account of what legitimacy demands because they seemingly violate non-liberals’ freedom in a way that runs counter to the concern for respecting all persons’ moral status as free and equal beings. Insofar as liberals force non-liberals to obey wills other than their own, they are therefore engaging in the very acts of domination or oppression that liberalism was intended to safeguard individuals from.
Accepting that coercing non-liberals necessarily represents a violation of their freedom is to concede the implausibility of a central pillar of the liberal vision of the political. As we have seen, liberal political theory attempts to reconcile individuals’ freedom with public law by claiming that the fundamental principles of the political association are the subject of citizens’ agreement and thus freely accepted bonds. Insofar as the state employs coercive force it does so in ways consistent with individuals’ wills. Liberalism is an attempt to imagine a political order in which the control that must be employed in order to ensure peace, stability and obedience to the law is not a form of oppression or domination because the principles guiding the political order are the subject of a consensus amongst the citizenry at large. The argument of this article, however, throws serious doubt on the plausibility of this image of politics, as non-liberals will always be coerced against their will when they are compelled to obey the laws of the liberal state. Liberalism’s self-understanding as a non-oppressive form of politics in which freedom and law are reconciled overlooks the fact that this is only true in relation to those who endorse the fundamental principles of the liberal order; for those that do not, compelling them to obey the law is to force them to live according to principles which they do not accept.
That politics requires compelling people to obey laws they reject seems like a banal truism of political life. We recognize in our everyday understanding of politics that the art of governing regularly requires political leaders to overrule the will of their citizens and often coerce them to do things which they do not want to do. Systematically compelling people to do what they otherwise would not seems to be a central characteristic of political rule. 37 In large part, as realist political theory recognizes, this is because politics as a human activity takes place in conditions where we need to live according to common principles of cooperation though disagree about what those principles should be, what Waldron has called the ‘circumstances of politics’. 38 Crucially, realists believe that disagreement (political, as well as moral and religious) is a permanent and inevitable feature of political life, something which politics must manage but can never fully overcome. While realism need not disavow the normative objectives of liberalism which underpin its account of legitimacy, the protection of the individual from the tyranny and oppression of state power, it cannot endorse its dominant strategy for achieving that goal. 39 Because of realism’s conflictual account of politics, it rejects as misguided the liberal requirement of a consensus on political fundamentals in the face of deep disagreement about matters such as the proper purposes to which state power should be put, the basis and grounds of political legitimacy and the just distribution of the benefits and burdens of cooperation. 40 That politics takes place against the backdrop of political disagreement confirms our experience that politics has a special relationship with the act of ruling (even in liberal democratic societies) which necessarily requires coercing some people against their will to act in accordance with constitutional essentials and laws that they reject. And insofar as this is right, ruling over those who reject the terms of the rule will unavoidably, on liberalism’s own terms, violate their freedom.
That politics inevitably requires ruling over those who reject the terms of the political association is not something that liberal theorists have paid much attention to, largely because the ideal-theory model in which they more often than not work assumes consensus on the fundamental principles of politics, in which case the question of coercing non-liberals does not meaningfully arise. Any actual disagreement with the political principles is often rejected as unreasonable or irrational (which I have suggested is incompatible with liberalism’s own accounts of the nature and origin of deep moral pluralism). Yet assuming that non-liberals are an inevitable presence in liberal societies, coercion and the violation of persons’ freedom (even if only of a minority) is an unavoidable feature of liberal regimes, as it no doubt is in all forms of political association. And so the liberal image of the political needs to be modified both in light of our lived experience of political disagreement and the need for political rule in response to such conflict, but also because theoretically, and on its own terms, liberalism is without an account which can explain how freedom and law are reconciled in relation to non-liberals. Liberalism needs to come to terms with the place and function of what it would see as oppressive coercion within its own theory. And to take coercive power more seriously means not pretending it is something else, such as freedom, or disguising it as a means to harmony.
Yet accepting the political necessity of compelling persons to obey wills other than their own is not something liberalism can all too easily accept. Liberalism does not understand itself as just another partisan political position which employs coercion and oppression in order to maintain its control. Rather liberalism sees itself as an emancipatory creed in the sense that it is the antithesis of oppression, a form of politics in which domination and its corollary loss of individual freedom is absent or overcome for all people. And so accepting that liberal regimes do violate the freedom of non-liberals by forcing them to live according to principles they do not accept, is to recognize that liberalism is indeed partisan, that it does take very firm and controversial positions on several of the most fundamental of political questions, and that it does enforce the will of one group of people over another. A significant alteration of liberalism’s self-image would therefore be called for if it were to accept that it is a partisan rather than consensus-based theory of politics. It is, in effect, a crucial concession for liberal political theory to make.
Recognizing the necessity and inevitability of coercion in political society, even a liberal one, and of politics as a non-consensual activity, is something much more amenable to realist political theory. Stears writes, for example, that … the mechanisms that are employed in order to respond to that [political] disagreement are themselves inherently coercive. They are mechanisms through which either elites or more widely dispersed social groups seek to enforce their will, or realise their desires, at the same time as to maintain stability and order of a certain sort. Politics cannot be a field of reasonable consensus; it is an inevitable realm of dispute shaped by the pursuit of control.
41