The place of unreasonable people beyond Rawls

Rawls’s ideas of reasonableness and unreasonableness

The notion of reasonableness plays a central role in Rawls’s theory. Although Rawls employed the term ‘reasonable’ in A Theory of Justice in a colloquial and unspecific way, ⁴ in his Political Liberalism ⁵ he devotes sustained attention to it, giving the term ‘reasonable’ a more technical meaning, frequently in contrast with ‘rational’. ⁶ The radical development of the notion of ‘reasonable’ consists in connecting the reasonableness of comprehensive doctrines and the question of stability through the idea of an overlapping consensus. This argument is crucial and needs to be reiterated. Recall the central question of Political Liberalism: ‘how is it possible for there to exist over time a just and stable society of free and equal citizens who still remain profoundly divided by reasonable religious, philosophical and moral doctrines?’ ⁷ This question brings to the fore both the fact of pluralism and the importance of stability for a society in which there is a plurality of incompatible comprehensive doctrines. When faced with pluralism, Rawls recognizes a serious internal problem in A Theory of Justice concerning the stability of a well-ordered society, discussed there in part III. This assumed a congruence of the good and the right and implied that justice as fairness is a comprehensive philosophical doctrine, but ‘we all know too well that Political Liberalism argues that such a doctrine may not be the bedrock of a modern liberal-democratic society; instead, this is characterized by a pluralism of incompatible yet reasonable doctrines, none of which should expect to be affirmed by all citizens’. ⁸

It is this concern, caused by the fact of pluralism, that leads Rawls to ask how it is possible that opposing reasonable and comprehensive doctrines may coexist and sustain the same political conception of justice. It is not appropriate to assume a conception of justice on the basis of any comprehensive doctrine because the exclusive enforcement of any such doctrine would lead to political oppression and instability. The stability of a liberal democracy must be reached for the ‘right reasons’: it has to be backed by a voluntary adherence to shared institutions and built on a moral consensus. Such an overlapping (moral) consensus among comprehensive doctrines may be achieved only among that class of comprehensive doctrines that are reasonable, meaning that reasonableness qualifies the group of comprehensive doctrines that can find common ground within a liberal-democratic regime.

Despite the importance of reasonableness, Rawls seems reluctant to provide a direct exposition of its meaning. Instead of defining reasonableness, Rawls prefers to associate it with the attitudes and capacities of people, referring to reasonableness as both a virtue and a kind of disposition. ⁹ There are two aspects of reasonableness: the first consists in a person’s willingness to propose and honour fair terms of cooperation. ¹⁰ In order to establish such fair terms, reasonable persons are ready to treat one another as free and equal. As equally reasonable persons, in this sense, citizens are equally entitled to receive justification for the exercise of political power.

The second aspect of reasonableness is the ‘willingness to recognize the burdens of judgement and to accept their consequences for the use of public reason in directing the legitimate exercise of political power in a constitutional regime’. ¹¹ The burdens of judgement are the source of reasonable disagreement, namely the disagreement that results from the free exercise of reason under democratic institutions. Reasonableness indicates here that reasonable persons’ ethical-political ability to share public reasons in a regime of reciprocity is based on their acknowledgement of the burdens of judgement. These two aspects of reasonableness – the willingness to propose and honour fair terms of cooperation and the willingness to recognize the burdens of judgement and the consequent fact of pluralism – substantiate what being reasonable implies. As a consequence, as will be seen, reasonable citizens accept the pluralism of reasonable comprehensive doctrines as a normal product of free reasoning and hence would never repress views different from their own. ¹²

In summary, reasonableness is primarily a political virtue of citizens who are reasonable when they take others to be politically free and equal, and equally deserving of fair terms of social cooperation. They are reasonable in so far as, having acknowledged the burdens of judgement, they support freedom of thought and conscience. Reasonableness is thus intimately connected to the basic ideas of reciprocity and fair cooperation. Understood as the ability to tolerate the holders of reasonable yet incompatible comprehensive doctrines, reasonableness is not properly an epistemological idea (though it incorporates epistemological elements); rather, ‘it is part of a political ideal of democratic citizenship that includes the idea of public reason’. ¹³

Returning to the idea of stability defended in Political Liberalism, it is possible to see how reasonableness appears to respond to the initial problem of reconciling stability with pluralism. It may be said that the institutional problem to which reasonableness is directed is exactly the problem of stability. The stability of a well-ordered society is secured by citizens’ ability to mediate between competing comprehensive claims. This ability is rooted in their reasonableness. ¹⁴ The idea of reasonableness as a political virtue responds to the problem of stability as it allows the citizens of a well-ordered society to view one another as moral partners who deserve respect and justification of the exercise of political power. What reasonableness clearly allows is for reasonable citizens to distance themselves from their comprehensive doctrines, in so far as they may suspend their beliefs and gradually adapt them in order to reach an overlapping consensus on liberal values. The underlying idea is that citizens, independently of their specific comprehensive doctrine and by means of their reasonableness, may find within their doctrine the reasons and motivations that will sustain the essentials of political justice. The overlapping consensus reached by each citizen, each in his or her own way, is stable for the ‘right reasons’ in so far as it is grounded in different, but still moral, premises. This makes the overlapping consensus in which a stable well-ordered society is grounded quite different from a ‘mere’ modus vivendi, understood by Rawls as a precarious equilibrium that depends on ‘circumstances remaining such as not to upset the fortunate convergence of interests’. ¹⁵ This implies that the question of stability cannot be addressed by ‘contenting ourselves’ with a modus vivendi; it can only be addressed by striving to realize an overlapping consensus on shared terms of cooperation and moral political values such that ‘stability is secured by sufficient motivation of the appropriate kind acquired under just institutions’. ¹⁶

There is a core idea in this. It may be said that Rawls trusts the rational agent’s capacity to split morality into two parts: on the one hand, there is the full integral morality, contained in each person’s comprehensive doctrine; on the other hand, there is an institutional morality regarding political identity, which concerns rational agents as citizens. ¹⁷ The underlying belief is that they are able to put their comprehensive doctrine truth to one side in order to behave reasonably as citizens. This is not the same as being sceptical: it means – once again – being reasonable. In this sense to be reasonable implies the ability to apply ‘the principle of toleration to philosophy itself’ and it permits citizens themselves to settle questions of religion, philosophy and morals in accordance with the views they freely affirm. ¹⁸ It may happen that citizens may regard their final ends differently from the way the political conception requires; they may reconcile the two kinds of commitment – the non-political and the political – in so far as their moral identity does not interfere with their public or institutional identity. As citizens, they morally accept common institutions, as the content of the overlapping consensus is a kind of institutional morality concerning persons as citizens. Only this institutional morality can adequately take into account reasonable pluralism by achieving stability.

Lastly, there is a further element to keep in mind: Rawls assumes that reasonable persons affirm only reasonable comprehensive doctrines. ¹⁹ This statement suggests the idea that holding reasonable beliefs is the premise to acting reasonably. Generally speaking, for Rawls it is not plausible to hold unreasonable doctrines and simultaneously behave in a reasonable way, and vice versa. I will return to this point later.

What Rawls means by ‘unreasonable’ can be equated with denying a liberal conception of justice. After detailing the characteristics of the ‘reasonable’, Rawls says that people are unreasonable when they plan to engage in cooperative schemes but are unwilling to honour, or even to propose, any general principles for specifying fair terms of cooperation. ²⁰ Being unreasonable implies not being properly motivated to support institutions, with any adherence to them equating to a ‘mere’ modus vivendi, given that the reasons for adherence are prudential and not moral reasons. Unreasonable people may not sustain the political conception of justice as a moral political conception because they reject its fundamentals. They are ready to violate them as suits their interests when circumstances allow. Rawls calls ‘unreasonable’ those people who would impose their beliefs on others because they think that their beliefs are true. ²¹ A crucial point is that we would be counted unreasonable not because our belief is considered to be false, but because we are ready to impose it on others. Given the burdens of judgement, we would be unreasonable should we deny that no truth may be agreed on and that, correspondingly, no political agreement may be grounded in truth. ²² For Rawls, no theory claiming to be true may offer a shared public basis of political justification; truth is divisive and there is no possibility of publicly justifying truth. Hence it is unreasonable to appeal to truth in the political domain as there is no shared public basis to distinguish true from false beliefs. Even justice as fairness is not itself reasonable unless it can win its support in a suitable way by addressing each citizen’s reason. This is the condition – already recalled – of stability for the ‘right reasons’.

The ‘unreasonable’ and their treatment

What about the ‘unreasonable’? Although Rawls is critical of grounding society in a ‘mere’ modus vivendi, he admits that a modus vivendi may develop over time into an overlapping consensus. Having experienced the good of living according to the principles of justice, unreasonable people may come to acquire a corresponding sense of justice that enables them to comply with institutions convincingly and intentionally. Incidentally, Rawls suggests elsewhere that in everyday life most citizens come to affirm the principles of justice without seeing any particular connection between them and their other views. ²³ This is an interesting point. Rawls seems ready to recognize that people may adhere to a political conception of justice even when they do not ground their adherence in any one reasonable comprehensive doctrine, but only have ‘spare’ arguments or motivations to sustain it. ²⁴ However, Rawls concludes this argument by saying that there is no guarantee for such an occurrence. Views that would suppress the basic rights and liberties of persons may indeed survive in society. He thinks that in a well-ordered society such views would not be strong enough to undermine substantive justice, but ‘that is the hope; there can be no guarantee’. ²⁵

Correspondingly, Rawls declares that cultivating a reasonable hope that the ‘unreasonable’ may eventually become reasonable needs time to be ‘translated’ into reality and may also remain unrealized. By distinguishing between the fact of reasonable pluralism and the fact of pluralism as such, Rawls also admits that it must not come as a surprise that in a democratic society unreasonable views will always exist. Indeed, a society may contain unreasonable and irrational and even crazy comprehensive doctrines. In such cases, says Rawls, ‘the problem is to contain them so that they do not undermine the unity and justice of society’. ²⁶ Reasonable institutions have ‘the practical task of containing them – like war and disease – so that they do not overturn political justice’. ²⁷

In summary: Rawls assumes that unreasonable doctrines are a fact of any liberal democracy; they must be kept under control in order not to undermine social unity and political stability. He is also clear about trusting liberal institutions in their capacity to educate people to cultivate in themselves a sense of justice. He believes that, in a well-ordered society, the ‘unreasonable’ are likely to become reasonable, because of the strong positive effects of living under liberal institutions. The fact that such institutions protect the liberty of the ‘unreasonable’, thus allowing them to hold on to their beliefs, is something that they will come to appreciate over time and will lead them to lend support to such institutions. His hope is that, in so doing, political stability is enhanced and the benefits brought about by stability will eventually motivate the ‘unreasonable’ to become loyal to liberal institutions. ²⁸

Building on this reconstruction of Rawls’s account of the ‘unreasonable’, I want to suggest that it is incomplete. As he hopes that the ‘unreasonable’ will eventually turn into reasonable, Rawls does not seem to be willing either to offer the unreasonable specific reasons or motivations for adhering to liberal institutions, or to listen to their possible and perhaps unreasonable motives for not adhering to them. However, under actual circumstances, Rawls’s hope may be misplaced. There may well be persons who will never end up adopting public reason and its requirements to defend their positions in the political domain. However, this does not mean or imply that they are unreasonable in Rawls’s strict sense: that is, that they are also ready to impose their beliefs on others. By the fact that they do not, and possibly will never endorse the values of freedom and equality in which social cooperation is grounded, we cannot infer that they will necessarily try to violate the terms around which cooperation is structured.

In keeping with this idea, I argue in contrast to Rawls that there may be more than one way to be unreasonable or, better, not to be reasonable. In fact, besides those unreasonable persons who will retain their unreasonableness and thus represent a threat to society, and those who are open to change – this is, the ‘reasonable-to-be’ – I suggest considering a third subset of persons, whom I call the ‘non-reasonable’. The ‘non-reasonable’ differ from those fanatics about whom Rawls is concerned and who should eventually be coerced to preserve social unity and political stability; but they are not necessarily open to becoming reasonable. Rather, although they are likely to preserve their non-reasonableness (as they do not endorse the values that form the overlapping consensus), they are ready to support liberal institutions, albeit for reasons that are not the ‘right ones’, as Rawls would put it. Non-reasonable persons may support the liberal-democratic order for their own reasons, which fall outside the domain of public reason.

Unlike Rawls, I am persuaded that nothing wrong happens if liberal institutions are sustained by people divided at the level of their fundamental commitments. Stability does not seem to be threatened by the mere fact that shared institutions are supported also by people who are sincerely unable to find any consistency or ‘continuity’ between their beliefs and the principles of justice, but who are nonetheless ready to engage in cooperation with the others. Thus depicted, non-reasonable people are those citizens who may well be able to support liberal-democratic institutions, although they are likely to do so in a specific way, based on reasons that are not what Rawls would call the ‘right reasons’, namely the reasons-that-all-can-accept. ²⁹ Therefore I suggest that the ‘non-reasonable’ might be included in the liberal-democratic citizenry although their views could not be part of a reasonable overlapping consensus. Their inclusion will rather take the form of a special but stable modus vivendi. The idea is that people may consistently hold unreasonable views (outside the domain of public reason) but be ready to cooperate with the other members of the citizenry. Such a cooperative attitude, however motivated, may provide a strong enough motive to include such non-reasonable people in the liberal-democratic citizenry and treat them as political equals.

As an example of who may be such non-reasonable people in a liberal-democratic society, consider the case of a religious group who is unreasonable in so far as its members hold an unreasonable doctrine denying equality to women. They hold this doctrine to be the truth, thus accusing all other religious groups in society of false beliefs. Although they hold such an unreasonable belief, members of this religious group willingly comply with liberal-democratic institutions, given that the protection they are afforded under such institutions turns out to be a necessary condition for their survival in society. Concrete instances of such non-reasonable groups may include Christian preachers of whatever denomination or congregation, but also communities such as Jehovah’s Witnesses. Such communities typically show some significant willingness to participate in society, as social cooperation represents a good for them and is a requisite for both their communal life and preaching activity as an essential part of their mission. Although they would probably prefer to live in the world composed entirely of Jehovah’s Witnesses, they adapt themselves to living in this world, pursuing peace and order. Their reasons for accepting the terms of cooperation under liberal-democratic institutions are unlikely to be the ‘right reasons’ Rawls has in mind: those belonging to the domain of public reason and the object of a reasonable overlapping consensus. However, they are not necessarily merely prudential reasons. Non-reasonable people of this kind may recognize liberal-democratic society as the place to accomplish their mission on earth while waiting for the real after-death eternal life. Although their reasons for cooperation may not be translated into the language of public reason, they may equally count in favour of their stable participation in a fair system of social cooperation or at least to adopt a non-belligerent conduct.

My argument thus far has been aimed at showing that there is more to Rawls’s argument about ‘not reasonable’ people and the place they may have in a liberal society. In what follows I intend to honour Rawls’s prescription to apply ‘the principle of toleration to philosophy itself’. Specifically, my claim is that the connection between unreasonable doctrines and unreasonable behaviour should be disentangled in order to sustain the idea that some people may behave ‘not unreasonably’ in the special sense specified, despite the fact that they hold unreasonable doctrines (in the Rawlsian sense) and are not prepared to renounce them. Despite their unreasonable beliefs, such non-reasonable people may behave cooperatively on the basis of motivations which, albeit non-reasonable, are ‘compatible’ with living under liberal institutions. Their participation in a well-ordered liberal-democratic society has the form of a special and stable modus vivendi.

The ‘non-reasonable’

To clarify my notion of non-reasonableness as a third alternative between the Rawlsian concepts of reasonableness and unreasonableness, I draw inspiration from two different accounts – one by Erin Kelly and Lionel McPherson, and the other by Jonathan Quong – of how to consider and treat the ‘unreasonable’ in a liberal-democratic society. ³⁰ According to Kelly and McPherson, the term ‘unreasonable’ may be interpreted either as philosophically unreasonable or as politically unreasonable. People who are philosophically unreasonable may be politically reasonable, as the politically reasonable are those people who

… are prepared to grant one another the status of free and equal persons, and to propose and abide by terms of social cooperation that they believe are fair and mutually acceptable. In public political discussion such persons would not appeal to sectarian moral and political philosophies, but only to common political values. ³¹

Kelly and McPherson do not offer a formal definition of the ‘philosophically unreasonable’ and of the ‘politically reasonable’. Rather, they refer to such examples as members of private male clubs, who are philosophically unreasonable by virtue of their views of women but may yet be politically reasonable; and a ‘Klansman’ who could make a reasonable argument, for example by citing autonomy as a political value in a way that is independent of his racist beliefs. ³² Their idea is that philosophically unreasonable people may be politically reasonable in so far as they are not attempting to pursue their unreasonable philosophical view in politics, that is, in so far as their view has ‘no practical political consequence’. On this basis, Kelly and McPherson maintain that the politically ‘reasonable’ – in spite of their being philosophically unreasonable – should be granted toleration as they conceive of it, a wide public justification: ‘the notion of toleration we are emphasizing requires that our basic political and social institutions be justifiable in terms that unreasonable as well as reasonable persons have reason to accept’. ³³ The reason for this wide public justification lies therefore in the right to potential political participation in public reasoning (which the politically reasonable are entitled to), that is, the right to be owed a justification. In assessing the Rawlsian argument, Kelly and McPherson hold that in Rawls the two ways of being reasonable – politically and philosophically – tend to conflate. For Rawls, the reasonable person is in fact the one who is ready to propose principles and standards as fair terms of cooperation and to abide by them willingly, and who recognizes and is willing to bear the consequences of the burdens of judgement. Kelly and McPherson criticize this reasoning as arbitrary in so far as it aims at excluding unreasonable persons as such from public justification.

In his reply to Kelly and McPherson, Quong denies that the ‘unreasonable’ should be granted public justification. In fact, they should not be accorded the right to be included in the constituency of public justification but should be tolerated in order to be entitled to the benefits of citizenship. To understand Quong’s position, some further explanation of his view is in order. First, he rejects the distinction between the philosophically and the politically ‘unreasonable’. According to Quong, what Kelly and McPherson claim about the philosophically but not politically ‘unreasonable’ (their being able to participate in political reasoning) is hardly tenable; ‘if you are philosophically unreasonable’ – Quong objects – ‘in the sense that you do not accept the burdens of judgement or the fact of reasonable pluralism, then what possible moral reason could you have that would motivate you to accept the ideal of public justification?’ ³⁴ For Quong, a person may have no moral motivation to behave reasonably if she or he is philosophically unreasonable. Unreasonable persons should be accommodated in a ‘mere’ or Hobbesian modus vivendi; period. No other more stable and positive form of toleration is available to include them. Quong agrees with Kelly and McPherson on just one point: on lamenting Rawls’s too narrow idea of toleration. For Rawls, the ‘unreasonable’ are merely not to be interfered with, that is, they may be merely tolerated; but, Rawls adds, the ‘unreasonable’ may be controlled if not coerced (and surely coercion amounts to a form of interference) when they become a menace for the social order. Like Kelly and McPherson, Quong maintains that, although unreasonable citizens are rightfully excluded from the constituency of public justification, this does not entail their exclusion from the benefits of citizenship.

As argued, I agree with Kelly and McPherson that in a liberal-democratic society we should make room for what I have called the not ‘unreasonable’, that is, persons who despite rejecting the burdens of judgement and the subsequent fact of pluralism do not simultaneously refuse any kind of cooperation. Although such people endorse unreasonable beliefs (hence they are philosophically unreasonable in Kelly and McPherson’s terminology), they possess an attitude to behave in a special non-unreasonable way. They may not be ready to appeal only to public reason in politics, but they share the ideal of fair social cooperation by committing themselves to political stability. In this sense, unlike the Rawlsian ‘unreasonable’, non-reasonable people do not pose any threat to the social order. Therefore, I agree with Quong, but disagree with Kelly and McPherson, regarding the opportunity to consider philosophically unreasonable people as politically reasonable. As they reject the equal terms of a fair system of cooperation, the philosophically ‘unreasonable’ may not count as politically reasonable, let alone reasonable tout court. However, unlike Quong, I share the inclusive spirit of Kelly and McPherson’s proposal and suggest that we ‘expand’ the Rawlsian idea of public reason. Instead of requiring that all citizens share the same public reason – including only those reasons-that-all-can-accept – as a prerequisite to being included in a practice of public justification, I conceive of public justification as a wider practice in which the ‘non-reasonable’ may participate by putting forward their special reasons that are unlikely to be acceptable to all.

The issue here is not whether non-reasonable citizens should for example have the right to vote or to support policies based on their convictions; of course they should. The issue is whether they fall short of the ideal of citizenship towards their fellow citizens. My suggestion is that, although the reasons of the ‘non-reasonable’ are likely to be unacceptable to reasonable people, they may nonetheless be made intelligible to them, that is, they may be exchanged in public debate. ³⁵ What I mean is that the reasons of non-reasonable people should not be dismissed from public debate in advance of an argument simply on the grounds that reasonable people disagree with them. ³⁶ As in many other complex matters, the use of human reason in free institutions leads us to disagree; the political debate cannot be insulated from this fact, nor can the realm of public reason be convincingly constricted so as not to be affected by it. ³⁷

If Rawls is correct in believing that a conclusive rational consensus may be reached on the basics of a liberal regime, this consensus only characterizes liberal politics at the most abstract level. The Rawlsian approach to public justification seems to be too restrictive in real societies where the fact of pluralism is not tamed by the assumption of reasonableness. In realistically plural societies several groups advocate policies for their non-sharable personal reasons. This does not imply that such persons are ready to impose their views on others (as Rawls’s ‘unreasonable’ are by definition expected to do); many such persons, whom I have called the ‘non-reasonable’, may be willing to participate in public debate, but they would do so by offering arguments that depend on reasons drawn from their unreasonable doctrines. Pace Rawls, it is totally plausible to think that persons may participate in public debate and offer reasons to others not because such reasons are expected to persuade them, but because they are the reasons the holders sincerely believe that matter from their specific perspective, be it religious or not. An enlarged public debate could include the ‘non-reasonable’, who would be deprived instead of such an opportunity in a Rawlsian society. Such an enlarged public debate is not expected to lead necessarily to an overlapping consensus on the ‘right reasons’, that is, the reasons-that-all-can-accept. Instead, it allows all citizens, including the ‘non-reasonable’, to start from their respective points of view and exhibit their reasons publicly. Such a process may well lead the participants to the conclusion that a consensus-based public justification is unattainable.

My point here is that the conditions of real politics should not be neglected by appealing to an ideal overlapping consensus. Not all citizens are reasonable in the Rawlsian sense and the problem of how to deal with the ‘non-reasonable’ is not a mere accidental matter, but is a crucial fact of politics with which our theories should concern themselves. Theories need not be utopian to be good theories; distance from reality may be ‘dangerously utopian’ in the sense that it risks overlooking the commitment to designing and not only imagining a just society. ³⁸ I agree with John Horton when he claims that thinking about politics normatively does not exclude also thinking about it realistically. ³⁹

The example of the Jehovah’s Witnesses, to which I referred above, may help to illustrate this point. They would count as unreasonable in the Rawlsian sense as they reject the premises of public justification. They do not recognize others as free and equal, nor do they acknowledge the burdens of judgement or the fact of reasonable pluralism; they do not conceive of society as a fair system of social cooperation, even though they are willing to cooperate in it on their terms. The unreasonableness of their beliefs does not prevent them from complying with liberal-democratic institutions. Non-reasonable citizens have their own reasons to sustain such institutions; their reasons, however, are not Rawls’s ‘right reasons’ as they do not cohere with liberal values that are basic in a fair society. Assuming that a liberal society is actually plural and that there is more than one way to take part in the public debate, non-reasonable people could not rightfully be excluded from any practice of public justification. Although I agree with Quong when he says that liberal institutions do not have the duty to find further justifications for the ‘unreasonable’, ⁴⁰ I distance myself from him in so far as I bestow on these people the right to advance their instances in the political debate, aware that it may result in disagreement and not in a consensus. What I mean is that the ‘non-reasonable’ can be thought to share something like a project of cooperation with other citizens, although they do that on other grounds than those belonging to a liberal political culture (for instance, as they value social cooperation as a means to accomplish peacefully their preaching mission). In other words, they are ready to endorse liberal institutions in so far as these guarantee them the conditions for pursuing their mission. Their engagement in society is therefore a kind of modus vivendi.

In summary: I have identified the category of the ‘non-reasonable’, who commit themselves to a project of social cooperation while rejecting the notion that fellow citizens are all equal and free as well as the fact of reasonable pluralism. I see their way of living in a fair society not as a ‘mere’ modus vivendi, as Quong claims; their acceptance of social cooperation reveals a commitment to political stability even though – in Rawlsian terms – for reasons that are not the ‘right’ ones. I shall develop my account of a modus vivendi in what follows.

Not a ‘mere’ or Hobbesian modus vivendi

My idea of modus vivendi is an alternative both to the ‘mere’ modus vivendi as an ‘unstable’ balance of forces and opposed interests – the Hobbesian modus vivendi – and the modus vivendi that Rawls mentions when he speaks of the first station on the route to inclusion of people who do not hold reasonable comprehensive doctrines. Indeed, for Rawls, experience of the benefits of living under liberal institutions is likely to transform the ‘unreasonable’ into reasonable people. My sense is that such a conception of modus vivendi is better understood in terms of a ‘not-stable’ modus vivendi. It is unstable as it connotes a transient phase of an evolutionary process towards the achievement of a full overlapping consensus in society.

Alongside these two ways of understanding a modus vivendi, I conceive my alternative as a stable modus vivendi: it is the way in which non-reasonable people may permanently inhabit a liberal society, their political conduct being inspired by reasons that cannot be accepted by all. Their reasons for taking part in the political debate are not the ‘right reasons’, rather they are reasons that are good only for them, in the light of their life-plans. Their reasons may be motivating reasons for them, albeit neither public nor acceptable to all. Non-reasonable citizens, as I see them, comply with liberal-democratic institutions neither reluctantly nor as a mere form of acquiescence; they are ready to abide by the terms of cooperation for their own reasons, that is, they comply with liberal-democratic institutions although they do not share the moral values in which such institutions are grounded. ⁴¹ To put it differently, their support for liberal-democratic institutions derives from their agreeing to them, even though they do not agree with them. ⁴² By agreeing to liberal-democratic institutions, non-reasonable citizens put themselves willingly under an obligation to obey the laws enacted by such institutions; they are ready to agree to institutions on the basis of their non-sharable reasons. ⁴³

When I speak of the non-public reasons of non-reasonable citizens I do not necessarily mean them in terms of prudential reasons, motives of fear or opportunity. The reasons of non-reasonable citizens may be reasons that, for example, rely on a view of the world in which people are not free and equal, but in which persons are viewed as divided into saved or sinners, elect or damned: a world in which the former have a mission to convert the latter for the safety for their souls. As illustrated in the case of the Jehovah’s Witnesses, non-reasonable citizens show their willingness to comply with liberal-democratic institutions despite their not-subscribing to the basic values of such institutions. The way in which they see the ‘others on earth’ – evidently an unreasonable way as it does not appeal to the ideal of equal respect for persons – does not imply any direct infringement of the rights of others. While they see their fellow citizens as people to be saved and not as moral partners, they do nothing to enforce their ‘truth’ upon them. If to be reasonable means to recognize others as free and equal and to accept the fact of pluralism as the result of the free exercise of human reason, surely such non-reasonable persons as the Jehovah’s Witnesses may not be called reasonable. They live peacefully with others in a collaborative way, but not on equal terms of cooperation; collaboration with the neighbour is prescribed by a divine command to pray and work for their safety, but it is not backed by a commitment to the principle of equal respect. ⁴⁴ In this sense they are non-reasonable, not to be assimilated with the fanatic ‘unreasonable’ about whom Rawls is worried.

It may be objected that it is unlikely that non-reasonable citizens, as I have characterized them, will be accepted as fellow citizens to be included in a wider public debate. An important objection from this point of view may be that they are often organized in groups within which not all persons are treated as equals. For example, groups of the ‘non-reasonable people’ are often organized in hierarchies in which women are subjected and deprived of basic liberties. The subjection of women is certainly a strong reason to confirm a group’s non-reasonableness; however, in so far as women are not literally forced to remain in the group and in so far as they accept to live subjugated to family and community, there should be no room for interfering with their decisions. A liberal state is committed to protecting individual rights and liberties, not to enforcing freedom on those whose conception of the good does not entail autonomy as a non-negotiable value. Rawls himself is very clear about this point:

… a liberal conception of justice may have to allow for some traditional gendered division of labour within families – assume, say, that this division is based on religion – provided it is fully voluntary and does not result from or lead to injustice. To say that this division of labour is in this case fully voluntary means that it is adopted by people on the basis of their religion, which from a political point of view is voluntary. ⁴⁵

For political liberalism, not every division is forbidden.

Having said that, I am totally aware that to speak of women’s freedom within traditional groups is controversial. This is a crucial matter in the debate concerning multiculturalism, and specifically, the relations between cultures and women within cultures. Generally speaking, my idea is that women are usually, but not necessarily, the most vulnerable members within groups. They should not be considered as indisputably imprisoned in their community without being aware of their submission and without contemplating any exit-right or any opportunity to change their way of life. This may happen, and it probably often does, but this does not necessarily have to be the case. ⁴⁶ In this discourse it is crucial to recognize that cultures are not cages and people within cultures are not prisoners; ⁴⁷ some persons will endorse cultural rules, some probably behave unconsciously according to these rules and others may reject them in order to take control of their own lives. Similarly, some women may walk their road to emancipation, and others are likely to go on choosing to spend their lives within the community by playing an assigned role. ⁴⁸ A group in which people are not treated as equals is not reasonable, but it is not necessarily to be excluded from the political debate and from the practice of public justification, unless the most vulnerable members of the groups are deprived of their right to exit. In this case alone liberal institutions should intervene in order to contain practices that are unequivocally unreasonable and unjustified.

In conclusion: by proposing a third way to view modus vivendi not as a ‘mere’ modus vivendi, I intended to show that there is a more realistic way to conceive of non-reasonable groups within a fair society as participating in the public debate. Although their non-reasonableness precludes them from reaching an overlapping consensus with reasonable citizens, they may sustain liberal institutions for their own reasons, although these are not the ‘right reasons’ or the reasons-that-all-can-accept. Non-reasonable citizens show themselves able to comply with liberal-democratic institutions not reluctantly, but as a kind of commitment to behaving politically together with their fellow citizens. That commitment is not necessarily a prudential assessment dictated by fear or opportunity. It may rather ground a special kind of a stable modus vivendi as a way for the ‘non-reasonable’ to be included in social cooperation for reasons that are good only for them, but not for all. ⁴⁹ My revision of the idea of modus vivendi in the terms of a stable modus vivendi represents an alternative way in which non-reasonable citizens may take part in a liberal-democratic society and sustain its institutions in a different way.

My defence of this kind of a stable modus vivendi is not backed by scepticism towards the notion that a wholehearted enforcement of institutions should be based on moral reasons; quite the contrary. I urge, rather, that we recognize realistically that an overlapping consensus is too abstract an ideal to be reached by all citizens in a plural society, inhabited not only by reasonable, but also by non-reasonable people. Rawls insists on a partial but nevertheless significant degree of uniformity in the way people regard institutions and in this way he does not provide any reason or motivation for establishing institutions in societies in which not all citizens share the same core of liberal values. Pace Rawls, if political liberalism has nothing to offer to non-reasonable people, then its defence of liberal institutions looks intolerably weak. Political liberalism shows itself to be of little relevance in situations where the justification of liberalism matters the most: where liberalism must engage with those citizens who do not subscribe to liberal values and tradition. ⁵⁰

This is the sense in which I am going beyond Rawls. My idea of modus vivendi corresponds to an enlarged practice of public justification that is hospitable to various kinds of reasons that do not necessarily belong to a liberal political culture. Speaking of modus vivendi in this special way means assuming an anti-utopian approach to political theory. Instead of pursuing the utopian project of a shared, albeit minimal, morality, I suggest that we should content ourselves with striving for political accommodation and compromise. My idea of modus vivendi builds on two facts: that there are people within liberal-democratic societies whose lives are not inspired by liberal values; and that these people may nonetheless sustain liberal-democratic institutions. Settling for this kind of modus vivendi does not mean that the project of a stable liberal society is doomed.

Citizens who take part in a stable modus vivendi may entertain stable relations of reciprocity. By reciprocity one does not have to mean solely the reciprocity of justification, but also, for example, reciprocity of advantage. ⁵¹ The relations of reciprocity in place in a stable modus vivendi are grounded in the idea that nothing is more important to people than their freedom to live in accordance with their standards of the good life. In this sense, subscribing to a stable modus vivendi does not amount to a refusal of morality tout court. It rather requires subscribing to a kind of minimal moral universalism grounded in the presumption that the interests of all persons matter equally. ⁵² From this point of view, modus vivendi is not conceived as an unstable balance of interests and preferences, but as a stable equilibrium of irreconcilable differences. This modus vivendi represents a further way of endorsing institutions not for moral but ‘contingent’ reasons. It means that these institutions are not accepted for embodying values in themselves, but they are accepted for working as conditions for the individual’s own self-realization. Given that the assumption behind the modus vivendi is that citizens recognize the benefits of an ordered political life, it follows that they have reason to support measures that consolidate it regardless of other objections they may have. Thus, by styling these reasons as ‘contingent’ reasons I mean not just prudential reasons, but also reasons held by others that are different from mine, be they moral or not. ⁵³

Perhaps this conclusion looks unsatisfactory to those who believe that moral political values embodied by liberal institutions should be endorsed for their own sake. It is probably true that values embraced for themselves may lead to greater political stability than those adopted because of ‘discontinuous’ reasons that cannot in principle be good for all. ⁵⁴ In a non-ideal world, where people do not in fact share the same values, we should be prepared to settle for less than ideal and perhaps morally deficient compromises. This said, I do not wish to renounce political morality completely, nor do I think we are bound to do so. Following David McCabe, the key point is that, even if it is still necessary that citizens should have good reasons to sustain liberal institutions, these reasons need not themselves be exclusively universally moral. There is indeed not only one account of justification but, to quote McCabe, a ‘multivalent account’ of it. ⁵⁵

In conclusion: in a realistic approach to political theory it is crucial that we take into account the actual conditions in which persons may have moral and non-moral reasons to accept the liberal-democratic order. A stable modus vivendi captures much of what we care about when it comes to justifying the social and political order. It does so in a fair manner: the agreement to liberal terms of cooperation results from a process that does not privilege either some parties over others or some contested value commitments over others. Rather, it aims at reaching terms of political cooperation that all citizens, reasonable and non-reasonable, may find acceptable under real conditions. The only basic presumption is that all citizens in a stable modus vivendi accept that the interests of others also matter.