How should citizens conduct themselves as they engage in political action in an established democracy? The debate provoked by this question has rarely been as politically or intellectually compelling as it is in many countries today. In the United States, the rise of the Tea Party, with its culture of mass mobilization, intimidation, and partisan polarization, has provoked intense discussion concerning the rightful limits of political action.Footnote ¹ In Europe, similar arguments have focused on the related threats of extremism among religious minorities and fierce nationalist backlash among majority populations.Footnote ² Politically engaged citizens throughout mature democracies thus find themselves faced with a series of pressing questions concerning the ways in which they should interact with others with whom they disagree. Is democratic politics properly understood as a battle for supremacy with few rightful restraints beyond the rule of law and the avoidance of violence? Or is it better conceptualized as a collective endeavor where citizens transcend their disagreements in pursuit of an emergent common good? And what do our conclusions there tell us about citizen behavior itself?

As political theorists seek answers to these questions, they are likely to turn to a substantial body of literature produced by key liberal thinkers over the last three decades: that concerned with the ideal of “public reason” which has given succor to the notion of a “deliberative” democracy. The advocates of these positions, including most notably John Rawls and Charles Larmore, insist that in a realistically perfect democracy citizens should see their fellows neither as rivals to be battled against nor as comrades in some deep project of collective solidarity.Footnote ³ Instead, they should approach them as people who inevitably disagree about many matters of substantive importance but who are due respect as free and equal fellow citizens capable of finding agreement on a range of core political issues, especially those concerning the basic rules of their social cooperation. Actual democratic political action in such a context is said to require a form of “public reason giving.”Footnote ⁴ Our action in politics, in Rawls's words, is “proper only when we sincerely believe that the reasons we would offer for our political actions … are sufficient and we also reasonably think that other citizens might also reasonably accept those reasons.”Footnote ⁵

This “public-reason-giving requirement” initially appears directly to address the contemporary debate concerning the proper limits of citizen behavior. Citizens, on such an account, should not see themselves as trying to impose laws on others who could not reasonably be expected to accept those laws, and neither should they imagine that they are participants in some deep project of sociopolitical reconciliation. Instead, they should see themselves as people who live alongside others with whom they will always reasonably disagree but with whom they must come to agree on fair and shared terms of social cooperation. Such a self-image is said to encourage a distinctive kind of political behavior, one at once more moderate than that associated with groups like the Tea Party and less optimistic than that advanced by political romantics who believe that a transformational national harmony can emerge through democratic politics. Political action, on this account, involves agreeing to disagree in some areas and providing clear and broadly acceptable reasons to others in areas where one feels that coercive law is essential.Footnote ⁶

Despite its obvious attractions, this position is far from uncontroversial, especially when it comes to concrete recommendations for political behavior in the present. For some, it is simply unrealistic. Political action, some critics charge, could never be so restrained or so compromising. Citizens are always going to want to impose their will on each other whether they can provide generally applicable reasons for their preferences or not. There simply never could be a democratic society in which citizens behaved in the required public-reason-giving way.Footnote ⁷

For others, the charge is not just one of lacking reason but also one of political conservatism. The historic heroes of politics, it is argued, and even of liberal politics, have always had to act in ways that appear to violate this public-reason-giving requirement. Those who railed against slavery in the nineteenth century, who campaigned for votes for women at the beginning of the twentieth century, who built a welfare state in the face of fierce corporate opposition, and who struggled for civil rights for African Americans, were never as restrained as this liberal orthodoxy appears to demand. They acted decisively, providing reasons for their actions, of course, as any political movement does, but with no importance being placed on any distinction between public and nonpublic reasons. As Larmore notes, “William Ellery Channing argued for the emancipation of slaves just as Martin Luther King, Jr. argued against racial segregation by appealing to the belief that all human beings were equally God's creatures.”Footnote ⁸ Abolitionists and civil rights activists were reliant as well upon coercive forms of political action, concentrating on building coalitions, levying costs on rivals, and mobilizing potential supporters in often the crudest of ways.Footnote ⁹

Critics of public reason giving as the primary mode of democratic citizen behavior thus argue that too firm a dedication to such a requirement in the past would (paradoxically) have undermined the political achievements of liberalism itself. Without a willingness to engage in what might be called “non-public-reason-giving political action,” democratic states could never have been built and improvements to their structures never wrought.Footnote ¹⁰

These charges understandably unsettle liberal political philosophers. For while most liberals are temperamentally drawn to moderation in political action—and their emphasis on the “burdens of judgment” often leads them to insist on it—they also deeply value the achievements of the more radical groups of the past, and most also believe that coercive political action is a crucial component of contemporary politics, too.Footnote ¹¹ To put this another way, liberals are convinced that radical forms of political action, up to and including civil war with all of its “attendant miseries,” have sometimes been necessary, and they know that if this is true of the past, it may be true today as well. Unless this problem is resolved, liberals will not be able to offer any distinctive advice to those who are pressing for solutions to the question of how they should act politically in the present.

Political liberals such as John Rawls, of course, believe that this problem can be overcome. The task of this essay, therefore, is to ascertain whether they are right. The first section begins our enquiry by asking whether the public-reason-giving recommendation that has emerged from recent liberal political reflections on the ideal of public reason really does, in fact, speak to the question of how liberals think that citizens should behave in democratic politics today at all. Drawing extensively on the work of the two leading liberal thinkers on public reason, John Rawls and Charles Larmore, we suggest that liberals are often worryingly vague on this very point. They seem sometimes, that is, to wish to derive clear lessons for contemporary political action from their underlying theory, while at other times insisting that it is mistaken to take behavioral lessons for “real” politics from ideal philosophizing.

The second section delves deeper into this ideal/nonideal distinction in search of a clearer answer to our question of how citizens should behave. As it does so, it identifies a distinctive temporal perspective employed in this type of liberal theory: a perspective we here call the “Janus-faced” approach to the justification of modes of political behavior. From this perspective, liberals argue that we can justify past violations of the usual prescriptions of public reason giving in politics by “reading back” into arguments and actions the furtherance of the ideals of public reason in the long run. Traditional liberal heroes, such as the civil rights movement or the abolitionists, are thus said to have been justified in engaging in politics in a non-public-reason-giving way because their actions made American democratic life more suitable for public reason giving politics in the longer run. It is, in other words, not the nature of the action itself that matters but its justification, especially its justification in terms of the consequences for public reason itself. As all liberals will know, it is this perspective that is said also to allow present behavioral violations of the public-reason-giving requirement by looking forwards to a future in which our actions lead to the fundamental ideals of public reason being more completely realized.Footnote ¹²

The third section of the paper presents a thoroughgoing critique of this position. We contend that although this effort to incorporate non-public-reason-giving political action is a welcome development in liberal theorizing, it still fails to respond to the most powerful charges against the public-reason-giving requirement. Our argument focuses, in particular, on the fact that this attempt to justify non-public-reason-giving political action in nonideal circumstances is entirely inaccessible to actual political actors themselves, either as they think about their actions at the time or even as they reflect on those actions retrospectively. This, we insist, is a fatal flaw. It leaves liberals entirely unable to address the question of how citizens should actually conduct themselves in the democratic politics of the present.

This is the crux of the argument presented in this essay. Whatever the facts of particular historical cases, the Janus-faced argument does not provide us with a sound basis for deciding how and when we should act politically in our real lives. Public-reason liberalism is thus the wrong place to start if we wish to develop an account of proper citizen behavior that speaks to us today.

The Ideal Citizen in Ideal and Nonideal Circumstances

Liberal democratic political thought frequently focuses on questions of the ideal form of political legitimacy. At the heart of such thought is the classic concern about oppression: the notion that in modern societies characterized by the existence of a diverse range of contrasting comprehensive moral and political conceptions it would be wrong for the state or any other coercive agency to impose one particular conception on citizens unless that agency reasonably believes that other citizens can reasonably accept the conception in question. When liberal philosophers reflect on the political speech and actions of citizens in the ideal, then, they tend to imagine that such speech and action must necessarily reflect this concern. In an ideal democracy, citizens should thus always seek to offer justifications for their preferred principles and policies in the common values shared between citizens who endorse otherwise contrasting conceptions and practically highlight those justifications to others during the process of political debate. As Rawls put it, whenever citizens advocate policies and principles that suit their own values and interests they must also explain why it is at least reasonable for others to accept them.Footnote ¹³

The close relationship suggested between public reason giving in political speech and action, on the one hand, and the legitimacy of political power, on the other, might strike those less familiar with professional political theory as surprising. Individual citizens, after all, rarely exert much actual influence on the design of fundamental political institutions, generally playing second fiddle to established legislative elites even at times of constitutional change. Yet despite this practical constraint, most recent liberal political philosophers, led by Rawls himself, understand the scope of the ideal of the public-reason-giving requirement to stretch both deep and wide. In an ideal democracy, after all, citizens would in fact enjoy an “equal share in the coercive power,” and many aspects of their political speech and behavior thus fall within the requisite realm.Footnote ¹⁴ When they participate in public debate on matters of fundamental import, or when they cast their vote in an election that might have ramifications on those matters, citizens are thus required to think of themselves not as relatively unimportant members of a vast political order but rather “as ideal legislators.”Footnote ¹⁵ Even though they may not in practice wield a significant degree of political power, the ideal of public reason itself requires that, when exercising even their small share of such coercive power, they speak and act as if they were contributing in the manner of a delegate at a constitutional convention or a member of a constitutional supreme court.Footnote ¹⁶

In ideal circumstances, public reason thus has strong behavioral implications. It would result in a strict moral injunction against the exclusive employment of many kinds of political speech and action, of which two are of particular interest here.Footnote ¹⁷ The first is explicitly coercive speech and action: forms of politics, that is, that seek to ensure that citizens behave in a particular way not by means of persuasion but by means of force or the threat of force broadly conceived. It is thus not acceptable for citizens to seek to influence the decisions of their fellows by seeking to obstruct other citizens from choosing their own less favored options or by manipulating their decisions through recourse to some social or economic advantage. The second kind does not involve the open use of force or the threat of force, but it does not fully abide by the public-reason-giving requirement, either. This is speech or action that derives from a controversial comprehensive doctrine or value system that is not, and could not reasonably be expected to be, shared by every reasonable citizen in the polity. Arguments that are grounded in religious belief, in controversial ideology, or in contestable social and economic theories should not provide the only attempted justification for any particular policy or principle, for it is a requirement of legitimacy that all fundamental policies and principles could be endorsed, at least hypothetically, by all reasonable citizens.Footnote ¹⁸

Combining the two, as Rawls put it, this condition implies that the ideal of public reason requires that when some citizens propose policies or principles, “those proposing them must also think it at least reasonable for others to accept them, as free and equal citizens, and not as dominated or manipulated, or under the pressure of an inferior political or social position.”Footnote ¹⁹

Self-described “realists” in political philosophy have rejected this public-reason-giving requirement on the simple grounds that it is based on wishful thinking. Anyone seeking to achieve a political goal will need to draw on a far wider repertoire of strategies, including popular but exclusive emotional or rhetorical appeals through negative advertising, coalition building, street demonstrations, sit-ins, lockouts, and boycotts. They further argue that even if the public-reason-giving requirement were a plausible route in political argument it would render political debate on fundamental concerns remarkably stagnant. If, after all, every argument for a fundamental policy or principle has to be made in terms that every reasonable citizen could potentially accept then there might be little room for serious political innovation.Footnote ²⁰ The public-reason-giving requirement just cannot apply in nonideal circumstances, whatever its merits in a utopian polity.

Advocates of the public-reason-giving requirement have responded to this critique in two ways. First, they insist that the diverse range of political strategies sketched above are not necessarily inimical to the ideal of public reason, in that at least some of them could be accompanied “in due course” with public reason giving of the requisite sort. Such advocates remind us, then, that the public-reason-giving requirement does not insist that political actors restrict themselves to public reason giving. Instead, citizens can act in politics in a number of different ways but they can only act in pursuit of fundamental political outcomes the justifications for which could at least potentially be shared by all other reasonable citizens. It is, therefore, perfectly acceptable for citizens to attempt to persuade each other in ways that do not constitute public reason giving in the first instance, as long as such attempts are eventually followed by reasons in which the proposed political outcome is also justified in terms of the values that all reasonable citizens could be expected to share. Liberal political ideals are thus protected insofar as non-reason-giving political arguments or actions are never the sole or decisive arguments for actions publicly advanced in favor of some particular position on matters of fundamental import.Footnote ²¹

Second, and more importantly, advocates of the ideal of public reason also accept that there may well be conditions under which the public-reason-giving requirement does not apply in its behavioral form at all. Charles Larmore, for example, is clear that the public-reason-giving requirement in its purest form applies “only to the ideal case in which everyone in … society already accepts the norms of rational dialogue and equal respect.”Footnote ²² The reason for this is straightforward and derives from the moral foundation for the public-reason-giving requirement itself. According to the cluster of values that constitute the ideal of public reason, citizens are required to justify their preferred policies and principles to each other in terms that all could reasonably be expected to endorse only because this provides the best guarantee that any resulting coercive acts will be fully legitimate. The requirement actually to provide reasons, and to structure one's behavior as a citizen in a way that emphasizes such public reason giving, is therefore merely a means by which to ensure that no citizen is governed by the whim of others or subjected to rules that they could not have been reasonably expected to endorse. It is not an end in itself.

Legitimacy, then, is not born from the act of public reason giving itself, but follows insofar as such public reason giving ensures that each and every citizen is treated with the requisite degree of respect in the governing process. What this implies, of course, is that the conditions under which public reason giving takes place are of considerable importance, for it will only serve its function in particular circumstances.

This explanation might surprise some casual interpreters of Rawlsian political liberalism who understand the essential justification for the fundamental ideal of public reason itself as being derived from currently prevailing political attitudes rather from any foundational moral commitment. There is, after all, a widespread misunderstanding of Rawlsian political liberalism that it is historically located and contextually sophisticated such that its moral tenets are taken only to apply to societies that already accept them. Such a misunderstanding is the consequence of the frequent Rawlsian suggestion that the underlying justifications for liberal political ideals are “available in the public political culture of a liberal constitutional regime.”Footnote ²³ But Rawls and other advocates of public reason giving are, in fact, clear that this does not mean that the strictures of the ideal of public reason should not be seen to apply to earlier generations where the values that underpin them were not widely shared. Rather, and directly to the contrary, they are insistent that the ideal is in fact derived from “intrinsically moral ideas” the value of which is not dependent on the political setting.Footnote ²⁴

As is well recognized, Charles Larmore is particularly insistent in this regard. He explains that “in political liberalism, at least as I conceive it, the norms of rational dialogue and equal respect … are understood to be correct and valid norms and not merely norms which people in a liberal order believe to be correct and valid.”Footnote ²⁵ In the debates between abolitionists and supporters of slavery, between women's suffragists and those who wished to restrict the vote to men, between welfare statists and corporate power, and the civil rights movement and the protectors of segregation, there was a right and a wrong from the perspective of the ideal of public reason itself, and that right and wrong was entirely independent of the fact that the (vast) majority of people at the time saw matters differently.Footnote ²⁶ As Larmore again puts it, “we must consider respect for persons as a norm binding on us independent of our will as citizens, enjoying a moral authority that we have not fashioned for ourselves.”Footnote ²⁷

Such a view has far-reaching implications for what advocates of public reason giving often call the nonideal case, or what Raymond Geuss more pointedly calls the case of “real politics,” that is, circumstances of significant inequalities of power, political access, and opportunity.Footnote ²⁸ Most straightforwardly, consideration of these restrictions makes it clear that the practical moral requirement of public reason giving did not apply to the abolitionists, suffragists, early welfare-state builders, or civil rights movements, or at least not in the same way that they would apply to citizens of an ideal society where acceptance of the ideal of equal respect enables a conversation to take place on terms that everyone could reasonably be expected to endorse. All of these groups, after all, inhabited a political world in which many of their fellow citizens neither shared essential values with them nor recognized the need to engage in a public-reason-giving conversation on terms that offered them the requisite degree of equal respect. Rather, they were forced to live in political orders with engrained inequalities of respect and to engage with fellow citizens many of whom affirmed comprehensive political values that celebrated those inequalities. It would clearly be perverse to assume that the ideal of public reason demands that such disadvantaged groups should not seek to change their position, or should be hindered from doing so by the need to provide reasons to their fellow citizens that those citizens could accept, given that the ideal is itself only justified by the role it plays in guaranteeing a form of political legitimacy that ensures that each citizen be treated with the requisite degree of equal respect.Footnote ²⁹ The appeal of public reason in its most straightforward sense is an appeal to reasonable fellow citizens alone. It is a form of justification, rather than a form of action.Footnote ³⁰

Public Reason Giving and Real Politics: Janus-Faced Justification

The question of what, if anything, liberal adherents of the ideal of public reason say about how citizens should behave in circumstances of real politics thus becomes far less clear. Indeed, on closer inspection it becomes apparent that most of the advocates of public reason giving in politics are all but silent on the issue. Charles Larmore is open about this, explicitly admitting that “what should be said about the less-than-ideal case is not entirely obvious.”Footnote ³¹ Despite these difficulties, there are elements of an answer within the literature on public reason. Liberal advocates of the public-reason-giving requirement do find common ground, that is, with the suggestion that whatever practical political strategies citizens in nonideal polities employ, they are morally acceptable only insofar as they advance the cause of the ideal of public reason itself. Which is to say that non-public-reason-giving political strategies are to be supported if and only if at least one of two conditions is satisfied. Either the arguments of those deploying these strategies can be reconstructed in terms that we, today, can recognize as being public reasons,Footnote ³² or they play a role in creating a political order in which the conditions required for a public-reason-giving polity could be established and where they could have understood their actions as performing that role.

Putting that another way, there are two necessary conditions that release politically active citizens from their public-reason-giving obligations in actual behavior. First, they do not have to interact with their opponents in a public-reason-giving way if those opponents refuse, or are unwilling, to recognize the free and equal status of their fellow citizens. And, second, they can act in non-public-reason-giving ways so long as whatever actions they do adopt play a role in creating the conditions where public reason giving would be both possible and required in the future.Footnote ³³

It is John Rawls who provides the fullest account of this argument. He does so in his discussion of the antislavery movement in the nineteenth-century United States, which he briefly addresses in Political Liberalism. There he asks whether there is evidence that “the abolitionists [went] … against the ideal of public reason” because they employed forms of political speech and action that did not meet the general public-reason-giving-requirement sketched above.Footnote ³⁴ His answer is that they did not do anything wrong if we “take for granted that their [non-public-reason-giving] political agitation was a necessary political force leading to the Civil War and so to the destruction of the great evil and curse of slavery.” For “surely they hoped for that result and they could have seen their actions as the best way to bring about a well-ordered and just society in which the ideal of public reason could eventually be honored.” In other words, given the “historical conditions” in which the abolitionists lived and worked, “it was not unreasonable of them to act as they did for the sake of public reason itself.”Footnote ³⁵

This is a backward- and forward-looking justification. It requires that we—early twenty-first-century evaluators—look backward to justify past patterns of political action from our own understanding of the consequences of such action. It is as if we are being asked to provide reasons to our own contemporaries on behalf of those who conducted past political actions. And it also requires that political actors themselves be hypothetically asked to look forward to the potential future implications of their own actions. They are asked, if you like, to provide reasons to future generations and not to their own contemporaries.Footnote ³⁶

What matters in circumstances of real politics, then, is not the extent to which disadvantaged political movements conduct themselves according to the public-reason-giving requirement in their own time, but the extent to which their political speech and action assists in the creation of a political order within which those requirements would actually be meaningful. As Rawls again summarizes, “on this account the abolitionists and the leaders of the civil rights movement did not go against the ideal of public reason; or rather, they did not provided they thought, or on reflection would have thought” that their political speech and actions “were required to give sufficient strength” for the conditions of the ideal of public reason “to be subsequently realized.”Footnote ³⁷

This does not, of course, mean that “anything goes” in the circumstances of real politics, or that the end always justifies the means. Most of these thinkers are conscious that the particular kind of strategy and tactics employed in any political struggle will shape the possible outcomes of that struggle. That has, after all, been a primary argument against violence in civil disputes ever since Tocqueville, who insisted that the legacy of violent revolution in France was a society incapable of living cooperatively at peace long into the future.Footnote ³⁸ It is at least possible, however, that liberals could believe that very dramatic forms of political action—even very occasionally violence and all of its attendant miseries—might be justifiable in order to rid ourselves of significant evils even in the absence of reasons actually provided to contemporaries in terms that they could be reasonably expected to endorse, on the condition that such conflict make the world more secure for the ideal of public reason in the long run. As Rawls concludes, “the appropriate limits of public reason vary depending on historical and social conditions.” Citizens “are to be moved to honor the ideal itself, in the present when circumstances permit, but often we may be forced to take a longer view.”Footnote ³⁹

All of this shows clearly that the ideal of public reason demands a Janus-faced attitude towards the justification of non-public-reason-giving political action. From the present we should look back and see how past political campaigns that were not accompanied by a politics of public reason giving were in fact justifiable because they could have or would have been articulated in terms acceptable to public reason had the concept been available. At the same time, it is necessary that those who do not feel that they can satisfy the public-reason-giving requirement today, or who at least feel that their fellow citizens would not understand them to be doing so, look forward and ask whether the violation of the practical recommendations of the ideal of public reason today may lead to greater adherence to the fundamentals of the ideal of public reason tomorrow. If, and only if, the answer to that question is affirmative may political action that others might interpret as violating public reason be justified today.

Intriguingly, this view about the justification of present-day political action is clearly articulated in the work of deliberative democrats Amy Gutmann and Dennis Thompson. Their argument about contemporary non-public-reason-giving political strategies is structurally identical to Rawls's Janus-faced approach.Footnote ⁴⁰ Gutmann and Thompson argue that when “nondeliberative politics—antiwar marches, sit-ins, and workers' strikes—are necessary to achieve deliberative ends, deliberative theory consistently suspends its requirements for deliberation. We should also observe that these activities often provoke more deliberation than would otherwise occur. But even when they do not, they can be justified if they lead to future occasions for deliberative criticism of injustice.”Footnote ⁴¹ This is cast in the language of deliberation rather than public reason, but the requirement has the same forward-looking justification. Activists engaged in a politics that goes beyond mere public reason giving must look to the future, and their activism is only justified if they help to make the world safe for deliberation in the longer run.

The Problems of Janus-Faced Justification

We might think that this Janus-faced perspective helps us to begin to provide a clearer answer to the question of how citizens should behave in actually existing democracies. Unfortunately, this is not the case. The initial problem with the Janus-faced approach concerns the essential ambiguity of its behavioral recommendations. That is, although it is clear that this approach in some way releases citizens from their immediate public-reason-giving requirements, it is not at all clear what actually replaces them. Rawls's own account of nonideal political behavior, after all, spans from advocacy of civil disobedience to civil war.Footnote ⁴² Even more worrying than this is the fact that the Janus-faced justifications are all but entirely unavailable to actual political actors themselves. In other words, they provide ways in which we can speak about the behavior of people who live in a nonideal democracy, but it remains incapable of speaking directly to those people themselves.

The nature of this difficulty first emerges when we ask what the actors of the past would have made of the defense of their political actions offered retrospectively by liberal advocates of public reason. The Janus-faced argument proceeds by suggesting that the citizens of the past could have been justified in their non-reason-giving actions only if those actions were actually (now, looking back) justified with reference to public reason, or if they could be said to help construct a society in which the ideal of public reason itself is more fully realized. Explaining that more fully, if, say, the abolitionists had held back from radical forms of political action, and sought to compromise with their fellow citizens on terms those citizens could agree with, they would never have been able to bring an end to the evil of slavery, but if slavery still existed then Americans would live in a nation where the public-reason-giving requirement would not apply. It is, after all, only when citizens acknowledge that they owe a certain kind of equal respect to one another that it is appropriate to talk of a public-reason-giving requirement in citizen behavior in the first place.

It is not, however, clear that today's terms would have made sense to citizens of a previous era, brought up as they had been on a wholly different set of ethical, political, social, and scientific assumptions.Footnote ⁴³ Our own early twenty-first-century political and ethical thought has, after all, been shaped by a set of experiences that these generations had not had, and that includes experiences that they themselves created. The ideal of public reason resonates with us because we have had to reflect on the impact of the American Civil War, the Holocaust and the Second World War, and the civil rights movement, among many other events. All of these have crafted a conceptual schema that prioritizes reasonable discussion over extremist political action and that places a high premium on the maintenance of both social order and egalitarian respect. As Bernard Williams once put it, we always need to remember that “the circumstances in which [current] liberal thought is possible have been created in part by actions that violate liberal ideals.”Footnote ⁴⁴

What this means, of course, is that we have no way of telling how earlier generations would have responded to our ideal had it been articulated to them. This is a fact that present-day advocates of public reason giving have openly acknowledged but failed to engage with adequately. In Political Liberalism, John Rawls explicitly argued that abolitionists and civil rights campaigners were justified in their non-public-reason-giving political action because of the service their actions played for the ideal of public reason. But when he enquires further whether the movements themselves could possibly have understood that condition, the answer either claims access to knowledge that is unavailable to us, or simply obfuscates. “I do not know whether the abolitionists or King thought of themselves as fulfilling the purpose of [public reason],” he thus suggests, “[b]ut whether they did or not, they could have. And had they known and accepted the idea of public reason, they would have.”Footnote ⁴⁵

Unusually for Rawls, the last sentence adds nothing to our understanding of the issue at hand, for had the movements accepted the ideal of public reason then moral logic would presumably have demanded they commit themselves to an understanding of their actions along Rawlsian lines. It is, of course, possible to suggest that aspects of the abolitionist and civil rights movements' actual arguments lent support to elements of the ideal of public reason, and some theorists have done just that.Footnote ⁴⁶ Alternatively, we could argue that even if past generations did not understand their actions as being justified because they were making the world safer for public reason, that was, in fact, the reason for their being justified. This may or may not be the case. But either way such a response does nothing to alter the fact that the ideal of public reason would have been effectively silent to those actors as they actually existed, in their own time and their own context, and as they grappled with their own questions concerning the rightful limits of political action. Whatever bundles of reasons political activists in the past have made for their political activities—and as Larmore notes, these reasons have often been strictly theological and thus nonpublicFootnote ⁴⁷—it is always possible that their political action, justified by reference to nonpublic reasons alone, would itself advance the cause of public reason in the long run.

This objection to the notion that the ideal of public reason provides insight into actual citizen behavior is thus a powerful one.Footnote ⁴⁸ The case, however, goes even further than this. For if we imagine that the conceptual constraints described above did not hold, then the ideal of public reason still does not take us very far in appreciating the acceptability or otherwise of the political actions of previous generations. That is because the claim that the ideal of public reason might support non-public-reason-giving political action in nonideal conditions is dependent not only on an understanding of the ideal itself but on an ability to predict the way in which one's political actions will actually affect future states. The standard set by Janus-faced justification is then either too weak or too strong.

The standard is too weak if all actors have to do is tell a narrative by which it may come to pass that the world is made more secure for public reason as a result of their actions. This weak version may strike the reader of Rawls as a plausible interpretation of this temporally sensitive form of justification, on the grounds that neither Rawls nor anyone else can have certainty about the future effects of political action. Perhaps the public-reason-giving requirement is best understood as a frame for political debate rather than as a practical guide for political action. However, if the Janus-faced approach to justification is to have any substance at all, even for meaningful political debate, it would have to provide some criteria for assessing the plausibility of these future narratives, and a credibility threshold that genuinely justificatory narratives would have to traverse. Even in the absence of public reason, fellow citizens still require reasons of some sort to accept the proposal that my current nonpublic reasons will make the world more secure for public reason in the long run. The Janus-faced approach does not itself have the wherewithal to supply such criteria, as it merely states what must be the case for nonpublic reasons to be justified. Thus any such criteria would be independent of the Janus-faced account, and this independent standard would be doing the justificatory work.

At the very least, then, the Janus-faced account requires an accompanying set of criteria for determining the plausibility of consequential argument concerning the future security of public reason. This it does not have, and for reasons set out below it is unlikely that it ever would. The alternative is that forward-looking requirement is far too strong, if what is required is certainty regarding that outcome. In other words, past actors must have been certain (and present-day actors must be certain) not only that the ideal of public reason enabled them to act in particular ways but also that their non-public-reason-giving actions were likely to play a significant part in creating a future polity that was safe for public reason giving.

The difficulty is that such a claim involves a kind of political knowledge that nobody could actually have. As Hannah Arendt famously noted, our knowledge and understanding of the future is just too hazy to allow for that sort of future-oriented justification of political action, especially of a radical kind.Footnote ⁴⁹ It would have been impossible for those who led America into civil war, for example, to know how the conflict would turn out. It could have led to the decimation of the nation, the supremacy of the confederacy, and the deeper entrenchment of slavery. It did not, of course, but simply no one could have known that.

These objections present serious difficulties for the liberal attempt to justify the behaviors of political actors in the past. It could still, nonetheless, be argued that none of these objections has much to say about the role the ideal of public reason might play in shaping our political behavior in the present. The ideal of public reason self-evidently is, after all, available to political actors now in a way that it has never been previously. It is an ideal that has emerged in our own time, and as a result of our own experiences. As such, it requires no sophisticated leap of imagination to suggest that it should be of direct relevance to political actors as they reflect on the kinds of political behaviors that they employ in the here and now.

Yet when we reflect further, it is still not at all apparent what behavioral recommendations can actually be said to flow from the ideal. This is partly because theorists disagree about how close the conditions that would make the public-reason-giving requirement entirely compelling are to being realized in a society such as our own.Footnote ⁵⁰ Putting that in Rawlsian terms, if it is unclear how far away our political order is from being just in the relevant sense, then it is also unclear how we should start thinking about political behavior. It is also, of course, not immediately apparent how one would assess such a situation in the first place. Rawls and his followers have always been insistent that formal constitutional rules are not a sufficient standard, but they have also asserted that full equality of resources and opportunities is far from necessary, too.

Even if this question could be resolved satisfactorily, it would also be highly debatable what kind of political action would be required to bring society further toward the conditions under which the public-reason-giving requirement might hold sway. Among those who have openly debated this theme, there is no consensus. Some have suggested that an exhaustively public-reason-giving political strategy is still desirable even in highly inopportune conditions because it provides the most effective practical means of redressing political injustices of the relevant sort. Amy Gutmann and Dennis Thompson, for example, suggest that public reason giving “can make a more positive contribution to the elimination of injustice than can alternative processes” because “the power of reason is less directly tied to the existing distribution of power, and therefore has the potential to challenge it.”Footnote ⁵¹ It is for this reason that a whole subfield of public-reason-giving promotion has appeared in political textbooks, much of which claims to take inspiration from liberal political philosophizing concerning the ideal of public reason.Footnote ⁵² Others, however, remain far more skeptical. They note that the empirical grounds for this case are shaky at best and insist that the time is not yet ripe for a focus on actual public reason giving in democratic politics.Footnote ⁵³

In responding to these concerns, some liberals might be tempted to suggest that it does not matter that the ideal of public reason is incapable of giving direct guidance on behavioral matters. It is, they might argue, important only that there is a matter of fact about the issue even if that fact is currently unclear. According to this argument, it is not actual public reason giving that matters but rather the fact that reasons could be given in defense of a particular policy or set of actions. Even if political actors themselves are unaware of how they should behave during the actual hurly-burly of political argument, then, it is still possible that proper philosophical reflection on the ideal of public reason can provide us, as philosophers, with an answer to the question whether a particular action was justified, or justifiable, or not.Footnote ⁵⁴

If this is the only argument available, however, it should give all but the most dogmatic advocates of the ideal of public reason grave cause for concern. For surely anyone committed to democratic politics in the present must believe that an understanding of the proper restraints on political behavior should be available to all, or at least a majority, of those who are likely to act politically. A position that is entirely theoretical, or that focuses solely on abstract and retrospective justification and never on behavioral guidance, cannot thus do what a genuinely political theory ought to be able to do and speak directly to the citizens themselves.Footnote ⁵⁵

What is more, however, this position appears to suggest that there are those among us with special knowledge of the timeless requirements of public reason, a knowledge that is not available to others. Such a view is, however, surely unpersuasive. It is even more so when we reflect on our earlier discussion about the past. For if previous generations can be taken to have been so mistakenly wrong when it came to understanding the legitimate terms of political cooperation, it is difficult to appreciate why we should possess any great confidence in the assessment of a current generation of philosophers standing the test of time any better. But if that is the case then some worrying consequences follow. If political movements of earlier generations were justified in their actions because such actions helped foster a political order that appeals to today's liberal philosophers, it is at least possible that other groups in today's society might one day be considered justified in their actions, if those actions help create a society that will appeal to philosophers of the future.

It is important here, of course, not to fall into complete relativism. No one should suggest that “anything goes” in contemporary political behavior because it is possible—just possible—that future generations will see the benefit in such action. That would be a wholly unacceptable form of potential victor's justice. But we are suggesting that the standard approach to actors of the past should give us pause when we seek to assess actors of the present. There may, that is, be movements today that wish to employ dramatic political means in order to advance causes that generally fall outside the purview of orthodox public reason, but that it is entirely plausible to imagine might not always be seen in such a way. Those who contend that animals should be extended some sense of political respect, that younger-aged children should be allowed some political role, that principles of justice should extend to the nonhuman world, or that the physically and mentally disabled are currently unjustly discriminated against in the basic social and political structure might plausibly be in that category.

By analogy with previous campaigners, then, it might be concluded that such movements do not have to provide reasons in terms that orthodox advocates of the ideal of public reason expect, as long as future generations will understand those tactics as essential to the promotion of their (newly developed) equivalent of the ideal of public reason. If we so overwhelmingly congratulate the struggles of abolitionists, suffragists, welfare statists, and the civil rights movement, after all, it is surely not beyond our imagination to suppose that future generations might do the same for some of these campaigners of today.

The very uncertainty to which these instances give rise completes our case against the Janus-faced approach to citizen behavior. We cannot look clearly forward to a future public reason to justify the actions of movements of the present, but neither can we presume that our current standard of public reason will rightly remain the standard forever. As we have it, then, the ideal of public reason leaves us unable to say anything definite to the actors of the past, the present, or the future. However much public reason appeals as a standard of justification, it is all but silent as a guide for actual citizen behavior.