Jacques Maritain Center: Ethics Without God?

Reassessing the Role of Religious Belief in Democratic Deliberation:

Shifting the Focus from Belief Content to Moral Disposition

David Thunder

Doctoral Student,
Department of Political Science,
University of Notre Dame

Delivered at the Ethics Without God? Conference,
University of Notre Dame,
July 18, 2003

According to Richard Rorty, the American nation was founded on a "happy, Jeffersonian, compromise" reached between the Enlightenment and the religious, "making it seem bad taste to bring religion into discussions of public policy."{2} Whatever we may think about the historical accuracy of this claim, it seems fair to say that not only in the US, but also in most European nations, it is often considered at best inappropriate, at worst offensive and ignorant, to bring up religion or religious considerations in the context of a discussion of law or public policy. In recent decades, liberal political philosophers have offered a variety of moral rationales for principles of religious restraint in public discourse,{3} and if their arguments succeed, then the popular prejudice against religious interventions in debates about public policy may turn out to be more or less well grounded after all.

Yet in this paper, an examination of arguments for religious restraint in public discourse will show that such arguments consistently prove to be logically self-defeating and morally untenable (§1).{4} I will further argue that this persistent failure of arguments for religious restraint in public discourse is partly attributable to a larger failure by liberals to elaborate a compelling account of the nature, dynamics, and goals of public discourse (§2). I will argue that it is only by shifting the theoretical focus away from principles of restraint on speech content and towards a fuller account of the virtues required for a vigorous and civil public conversation that we can hope to develop a compelling, realistic, and normatively robust account of public discourse (§3). Finally, I will offer some remarks on the role of religiously informed discourse in a liberal democracy in light of the virtue-centered account of discourse (§4).

Before I proceed with the argument, I would like to make a few remarks on the significance and scope of my argument. Firstly, my argument against the principle of religious restraint (§1) is informed by a rich literature which has already exposed the incoherence of the liberal argument more eloquently and persuasively than I can here. In other words, this is well-worn ground and so I only propose to convey as succinctly as possible a sense of the liberal argument and its inadequacy rather than do it full justice here. The contribution claimed by this paper, then, is not the refutation of liberal arguments for restraint as such, but rather, (a) a deeper diagnosis of the failure of the liberal arguments for restraint in general, including arguments for religious restraint, and (b) a suggestion about a constructive way forward both for restraint-theorists and for their opponents, now that the restraint-centered position has proved at best deeply controversial, at worst (as I believe) untenable. In these two respects, I have avoided the pitfall of resting satisfied with a deconstruction of the liberal model of discourse and stopping short of interrogating its root causes and investigating alternative substantive approaches to discourse.

§1 The Failure of Liberal Arguments for Religious Restraint

As we consider the role of religion in (democratic) public discourse, it is useful to begin by underlining some of the goals that can be plausibly associated with public discourse in a democracy. For our purposes, we will discuss three such goals or functions: (a) a legitimating function, ensuring that laws are publicly justifiable and rendering lawmakers and citizens mutually accountable for their political decisions; (b) a stabilizing and unifying function, fostering mutual understanding and some degree of social solidarity, as citizens come to realize their common citizenship through a shared discourse; (c) an epistemic function, as the overall stock of knowledge is increased and the epistemic horizons of citizens meet and expand. Arguments for principles of discursive restraint tend to focus on the legitimating and stabilizing functions of discourse rather than its epistemic function. However, in this essay I will consider the implications of the liberal model of discourse for all three functions. Rather than cataloguing all the arguments that have been made on behalf of principles of religious restraint, I will construct an argument that draws on some of the most representative, convincing, and well-articulated positions in the literature. The view motivating the quest for principles of restraint is a broadly liberal view, which has two main components: first, a liberal conception of respect for persons; second, a liberal view of the preconditions of social stability.

1a. The liberal conception of respect for persons

According to the liberal conception of respect, no reasonable and rational adult should be required to do something with the backing of physical force unless the coercion can be justified to him in terms he can understand and could, in principle, accept. That is, coercive measures must be justified on terms that are both rational and accessible to the coerced.{5} This is because to treat a person with respect is to address oneself to the person's reason and free will rather than forcing the person to conform to a scheme or purpose that is entirely foreign to her and irrelevant to her own subjective purposes and interests. This notion of respect might be thought of as a radicalization of the principle of the "consent of the governed," which finds expression both in Locke's theory of revolution and in the American Declaration of Independence.{6}

Obviously, it is notoriously difficult to pin down precisely what is meant by terms such as "rational," "accessible," and "reasonable." To start with the term, "rational," the modern understanding of this term, or at least the understanding adopted by most liberal authors, is largely instrumental: does a proposed policy or law meet or approximate the goals or purposes for which it is intended? Is the proposal coherent and intelligible? Does it have any plausible link with the most obvious and acknowledged purposes of law? Call this the "bare rationality" requirement. Undoubtedly, the bare-rationality requirement does assume a common epistemic horizon, at least as far as instrumental rationality and logic are concerned, but this does not seem unduly optimistic or ambitious.

A stronger version of the rationality requirement is adopted by Robert Audi. Call this the "strong rationality" requirement. On this view, all laws, or at least all significantly coercive laws, must pass through a counterfactual test, which Audi calls the "surrogancy conception of justified coercion"{7}: would a person perform the required or coerced action "autonomously" "if adequately informed and fully rational ["hence willing to imagine a reversal of positions or roles between oneself and others"]"?{8} This version of the rationality requirement is far too strong to be plausible, since there is no reason to believe that even ideally adequate information and full rationality would dispel all of the disagreements that arise in the lawmaking process. For example, how could differences in prudential evaluations and moral choices be dispelled by such a test? In short, there are very few significant laws that require actions which all "fully rational and adequately informed" people would perform "autonomously" or absent the coercive legal measure.

The accessibility requirement seems to me almost as problematic as Audi's strong rationality requirement.{9} What would it be mean to say that the grounds for a coercive law must be "accessible" to the coerced? If it meant merely that they must be intelligible, this would seem so weak as to have no teeth at all. After all, even the most totalitarian justifications of laws are "intelligible" to most thoughtful and intelligent adults. If it meant that the grounds for law must be actually accepted as valid, here and now, by the citizen -- even if only by the "reasonable" or cooperative citizen -- then this requirement would be so strong that it would rule out a whole host of laws that are widely viewed as reasonable and necessary for the common good. After all, citizens and politicians commonly attack laws that are on the books without for a moment bringing into question their legitimacy or binding force.

If accessibility neither means actual acceptance nor bare comprehension of the grounds of a law, what can it mean? I think we must seek for an answer in Rawls's notion of the "burdens of judgment," the fact that given people's differing views of the world and their diverse personal experiences or life histories, their moral and political judgments -- even when reasonable and motivated by upright moral intentions -- can vary dramatically. Such variation is not due to faulty reasoning, malice, or "unreasonableness" but rather, due to the "burdens of judgment" -- the fact that people are inescapably conditioned in their judgments by a background of habits and experiences, many of which are beyond their control.{10}

But if this is the case, then on what basis am I to override another person's judgment about what is in his best interests, or how he is to live his life? It must be on some basis that a reasonable person, even with his present epistemic "baggage," so to speak, could come to see and accept. Does this mean that I must secure the actual agreement of every reasonable person? No, although this may often be a healthy aspiration, it cannot be an absolute requirement. The point of finding accessible grounds is not to secure immediate and unqualified approval of those grounds or the law they are supporting, but to offer grounds that "hook in" fairly obviously to the worldview or shared reason of my interlocutor, and do not blatantly bypass his epistemic horizon. For example, if I argue that such-and-such a vaccination should be required since I believe public health is a top priority at this time, a person may reject my prioritization of public health and may even think that requiring vaccinations does not follow from such a priority, and yet consider my proposal as "accessible" to her, as addressed to our shared reason, and as accepting, rather than overriding, the burdens of judgment. This example shows that the accessibility requirement interpreted in light of Rawls's "burdens of judgment" need not require a high degree of consensus on actual policy outcomes, and is therefore not as politically na´ve as it might at first appear.

But who are arguments to be accessible to? To all citizens? No, for that would be far too strong a requirement. Rather, to all reasonable citizens. But what is a "reasonable" citizen? One helpful definition of a "reasonable" person is a person who is willing to live a spirit of reciprocity in his dealings with other citizens.{11} This means that he is willing to offer his or her fellow citizens fair terms of social cooperation, and is sensitive to the burdens of judgment. Sensitivity to the burdens of judgment entails an awareness that people can reach different positions based on different experiences and judgment calls, without necessarily opening themselves to (justifiable) moral condemnation. If a person has abandoned the notion of seeking fair terms of cooperation or demands that everyone else see the world through her own epistemic horizon, that person has essentially rejected the conditions under which a liberal democracy can flourish and endure over time. That person has rejected the liberal conception of respect and therefore lacks the moral requisites of liberal citizenship. He or she can safely be considered "beyond the pale" of liberal citizenship and need not be extended any special accommodation by lawmakers and citizens within the "pale."

If the liberal conception of respect is right, then it does seem to follow that arguments that depend for their validity on the acceptance of religious truths are inaccessible and therefore provide insufficient grounds for justifying law and public policy -- at least where there is no equivalent argument available in purely "accessible" terms. Let us take religion here to include propositions that assume or assert the existence of God or the truth of a religious creed, or attribute some special, revelatory quality to Sacred Scripture or to some religious leader (e.g. papal infallibility).{12} All such arguments fail the accessibility test, since there are many intelligent people who accept the principle of reciprocity (and are thus reasonable) and for whom religious propositions of this kind would be foreign to their worldviews or incompatible with their present epistemic commitments. In light of this failure, liberals will argue not that the law should restrict citizens' freedom of expression, but that citizens ought to voluntarily abstain from publicly advocating laws or policies on the basis of religious or religion-dependent reasons.{13} Of course, some liberals, such as Rawls, extend this restraint beyond religious reasons to reasons derived from "comprehensive doctrines" more generally conceived.{14}

1b. Social Stability

So far, we have described the argument for restraint from the point of view of liberal respect for persons. There is, however, also an argument that is often made from social stability or consequentialist concerns. What are the consequences or likely consequences of permitting, versus restricting, religious reasons within policy-oriented public discourse? Here, liberals like to put forward what Eberle has referred to as the "argument from Bosnia," which is essentially that the introduction of religious issues into public discourse is likely to generate acrimony and cycles of distrust and mutual hostility, and that it is even liable to provoke a new round of religious persecution. As Eberle points out, this argument derives much of its force from the destabilizing force that religion had in the seventeenth century wars of religion in Britain and Europe.{15}

The argument from stability goes something like this: in light of the devastating effect of mixing religion and politics during the wars of religion, and in light of our own experiences of the acrimony, intolerance, and hostility aroused by religious discourse, we ought to restrict it at least when it comes to serious discussions about policy and law. The evil consequences averted by such a state of affairs would easily outweigh any positive benefits that religious discourse might bring to society. This sort of argument has been advanced by both Robert Audi and Richard Rorty.{16}

1c. Why the liberal arguments for restraint fail

Let us consider the liberal arguments for restraint on their merits. First, consider the consequentialist argument. I believe this is the weaker argument, for two reasons: first, because I am skeptical that anyone can compile sufficient historical and sociological evidence to prove the case either way, i.e. to prove that religious discourse has generated more good than bad or more bad than good consequences overall. Second, liberals such as Audi who make the consequentialist argument for restraint invariably fail to advert to the significant differences between early modern Europe or present-day Bosnia, and present-day developed liberal democratic cultures such as the United States and Western Europe. Even if religious discourse was a force for repression and a force to be feared in early modern Europe, the notion that religion poses similar threats in, say, Britain or France or the United States in the year 2003 requires a special argument and cannot be assumed as a given.{17}

The liberal argument from respect is more tricky and cannot be dispensed with quite so easily as the argument from stability. Here, I will challenge the liberal argument on two grounds: first, by way of a reductio ad absurdum, pointing out its absurd logical consequences which seem to have escaped its proponents. Second, by proposing a rival conception of respect that I believe is both more true-to-life and more compelling all round.

Let us begin with the reductio ad absurdum. Assuming an accessibility requirement roughly along the lines sketched above, what sorts of reasons can be accepted as legitimate grounds for advocating a policy or law? We have already seen that religious reasons, broadly understood, are likely to fail the accessibility test. But we have not yet considered what sorts of secular or non-religious reasons are likely to pass it. Recall that accessibility does not require actual acceptance of one's reason(s) for supporting a law, but rather, that the said reason must hook up in a plausible way with some aspect of the actual epistemic horizon of any "reasonable" citizen. It must be possible to embrace that reason without making some radical shift in one's belief system, without overcoming the burdens of judgment by overhauling one's worldview or radically revising one's moral commitments in life. One might even say that reasons offered for coercive laws should not be unduly epistemically challenging, or if they are challenging, they must be challenging from within rather than outside the epistemic horizon of the citizen who is to be persuaded.

But I am confident that this accessibility requirement is so demanding that it would immediately exclude a whole host of secular or non-religious reasons that seem to constitute the bread-and-butter of democratic deliberation. Here are a few sorts of reasons that I can think of that might be used as a basis for arguing for laws: the view that it is the responsibility of society to re-distribute wealth based on need rather than desert or performance; the view that human beings are morally inviolable at every stage of life; the view that the right to self-determination or autonomy trumps moral flourishing or well-being even to the point of justifying assisted suicide; the view that certain forms of art are worthy of state support and promotion more than, say, people's use of pornography. The list can be easily expanded, but the point to note here is that the set of legitimate laws as well as the set of legitimate contributions to public discourse must shrink considerably if the accessibility requirement is taken seriously, since none of the reasons I have just mentioned meets it. All of these reasons, and many others besides, fail to hook into the epistemic horizons of many actual reasonable people, i.e., in many cases, they could not be accepted without a radical shift in a person's worldview. As such, they violate liberal respect for a person's autonomy.

Can we seriously entertain the view that the legitimacy or acceptability of a public argument for a policy or law depends on its advocate's success in adducing grounds that are "accessible," in this strong Rawlsian sense, to all reasonable citizens? I think the consequences of this view are so absurd as to rule it out as a plausible interpretation of legitimacy. Wherever we find deep disagreement about the grounds for law we are likely to find disagreement based on parting worldviews or epistemic horizons. Should such a deep disagreement vitiate the legitimacy of the democratic outcome? Perhaps in some cases, yes. But not necessarily. This will become clearer, I hope, as I suggest an alternative conception of respect.

1d. Towards a more realistic and compelling conception of respect for persons

Faced with deep disagreement concerning the appropriate grounds of a law or public policy, what behavior best reflects genuine respect for my fellow citizen? I have already considered the liberal answer: seek some ground that both of you, given your present (reasonable) epistemic horizons, could potentially accept. Now, consider a different answer, that is also liberal in its own way: Seek some ground that you consider morally legitimate and that the other person could find persuasive, given her present epistemic horizon, or, failing that, only coerce another person if that coercion can be justified based on some aspect of her welfare compatible with the welfare of her fellow citizens that legitimately falls under the purview of government and public policy. To seek unanimous consent among citizens on what the legitimate purview of government is in advance of the policy and lawmaking process would be futile and would bring an end to effective government. To insist that unanimity on justificatory standards be secured during the policy and lawmaking process but in advance of the passage of any particular law would similarly bring democratic lawmaking procedures to a deadlock on important areas of policy and would arbitrarily favor the status quo ante over democratic reform.

There is no reason, as far as I can tell, to think that just because my reasons for coercing another are presently inaccessible to that person in light of their accumulated habits, lifestyle, and beliefs, I thereby do that person a wrong or fail to respect her. It may be that I do not even anticipate that the person will ever come round to accepting the grounds I offer. However, I must believe that the reasons I offer are grounded in genuine interests of the coerced and of the society in question, and I must give serious weight to the interest of a person in her own liberty or ability to shape her life in an individual and sometimes idiosyncratic way. Personal freedom and creativity can be quelled by an all-encompassing, heavily paternalistic state.

Of course, it may be objected that giving "serious weight" to considerations of personal liberty is altogether too vague a principle to guide legislators' and citizens' deliberations as they forge laws. However, this objection is almost certainly misplaced, arising out of a penchant for simplistic and "trumping" principles that is shared by many liberal theorists. The liberal conception of respect suffers from this simplistic absolutization of certain desiderata, most especially the desideratum of rational consent to coercive laws. While endeavoring to secure general consent to law, or at least as general a consent as can be secured, is a noble liberal goal, respect for citizens by no means requires universal consent to reasons for law as a matter of course.{18} There may be some extraordinary situations in which unanimous consent is morally required, but these do not concern us here. What concerns us here is how the central case of law is to be understood; what the conditions of legitimate coercion are in the most typical cases.

To recap, respect for persons does not require unanimous consent to coercive laws or universal accessibility of the grounds of such laws. It does require, however, that the proponents of such laws ground their arguments on the rational interests, broadly conceived,{19} of citizens and of the society at large, including the interest of citizens in their own liberty to shape their lives according to their individual purposes even if this means they may make mistakes from time to time. Some interests, such as the interest of citizens in having good moral motivations, only fall indirectly under the purview of the state, since the state cannot know the inner workings of people's minds and any such intervention would be wide open to abuse. However, insofar as good moral motivations are a requisite for good moral actions, which include acts of justice and bravery, the state is responsible for facilitating rather than undermining a social environment in which people develop social and moral virtues conducive to the common good.{20}

§2 Why the Liberal Model of Public Discourse is Deficient

The failure of arguments for religious restraint is rooted in a larger failure among liberals to develop a convincing account of the nature, dynamics, and goals of public discourse. Rawls, the paradigmatic anglo-American liberal, conceives discourse as a narrowly rationalistic, proposition-driven practice. This is also reflected in the work of Habermas, albeit under a somewhat different and arguably richer guise, as well as in the work of many recent liberal authors including Charles Larmore, Bruce Ackerman, Thomas Nagel, and Robert Audi.{21} According to this view, the theorist can helpfully focus in on those aspects of public discourse that consist of claims about how we ought to order our polity or what laws we ought to pass, and why -- what Habermas would call "validity claims." Essentially, public discourse consists of a set of normative or moral claims, along with more narrowly self-interested claims, which are justified publicly based on rational arguments that are assessed and accepted or rejected on their merits. This, at least, is the ideal picture offered by liberals. They do not deny that public discourse in reality contains many rhetorical and even manipulative appeals, but the pervasive assumption seems to be that we can best understand public discourse as a moral practice in terms of the interaction of equal and rational agents offering each other propositional arguments for policies and laws, arguments which are accepted or rejected on their merits.

This is what I will call the "propositional" model of public discourse. This liberal model of discourse is compatible either with the "constrained conversation" ideal of discourse advocated (under slightly different guises) by Ackerman, Larmore and Rawls, or with the relatively "unconstrained" conversation ideal of Habermas and some deliberative democrats. I believe that both versions of the liberal conception of public discourse are seriously flawed. However, in this paper I will focus exclusively on the more constrained model, since this is more a propos of our topic.

According to the constrained-propositional model, when, in the course of a public discussion, we disagree on a coercive measure, we should bracket out our disagreement and proceed on the basis of what we share in common. Ackerman has a fairly simple version of this model, Rawls a more sophisticated one. But the central idea of proceeding on the basis of shared premises is common to both.{22} Of course, the liberal believes that this bracketing strategy will in fact enhance the legitimating and stabilizing functions of discourse. The epistemic or knowledge-enhancing function is not uppermost on the liberal's mind, so he does not tend to have much to say about it. I, on the other hand, will argue that the liberal model of discourse, in this instance the constrained-conversation model, is deeply inimical to all three goals ob public discourse (viz., its legitimating, stabilizing and epistemic functions), and that this failure is due to the liberal's blindness to the rich complexity of discursive practices, whose significance can only be understood by moving beyond the realm of propositional arguments.{23}

Consider, first, the legitimating function of public discourse: a suitably structured public sphere (or spheres) of discourse is supposed to make it more likely that the laws of a polity be publicly justifiable to all or most of its citizens. But how are laws justified to citizens? Is it through a set of probative propositions abstracted from particular worldviews? Sometimes, perhaps, but more often than not, it is either by appeal to one's interlocutor's particular belief-system -- and this appeal can be propositional in form -- or by appeal to the authority, knowledge, or expertise of this or that person. The legitimacy of laws is largely parasitic upon the legitimacy of the political system as a whole, and this in turn depends on a relationship of trust developing between rulers and the people. Bonds of trust go far further in reassuring people that (often complex) laws can be justified than propositional arguments. In short, the moral relationship and mutual trust expressed through public discourse between citizens, and between citizens and legislators, is often at least as important as rational argumentation in legitimating the democratic process. And this moral relationship depends on an attitude of respect and understanding far wider and deeper than the basic liberal conception of respect that we have alluded to previously.

Consider now the stabilizing function of public discourse: this is closely related to its legitimating function. Although it obviously involves the building of bonds of trust and mutual respect between citizens and between citizens and legislators, it also involves a willingness to abide by the laws of one's polity even when they strike one as misconceived or poorly thought out. It involves a sense of belonging and responsibility and ownership of one's nation. All of these can be fostered in important ways by a respectful and vigorous public discourse. In particular, public discourse fosters mutual understanding among citizens, and this in turn can sow the seeds of cooperation and mutual trust. None of this is merely a matter of propositional correctness or the justification of laws-it involves various modes of conduct and dispositions that occur within the discursive encounter.

And finally, consider the epistemic function of discourse: clearly, if a Gadamerian "fusion of horizons" is a viable way of acquiring new knowledge, then any bracketing strategy-especially a relatively ambitious and unconditional one-has the potential to suppress new avenues of epistemic growth. If citizens are not confronted with their differences in a fairly open and occasionally provocative way, they are deprived of opportunities to understand better the uniqueness and strengths and weaknesses of their own positions, and possibly adopt new beliefs that at first appeared foreign to them. Jeremy Waldron makes a compelling argument against the liberal bracketing strategy, appealing instead to a more dynamic conception.

Even if people are exposed in argument to ideas over which they are bound to disagree -- and how could any doctrine of public deliberation preclude that? -- it does not follow that such exposure is pointless or oppressive. For one thing, it is important for people to be acquainted with the views that others hold. Even more important, however, is the possibility that my view may be improved, in its subtlety and depth, by exposure to a religion or a metaphysics that I am initially inclined to reject.I mean to draw attention to an experience we all have had at one time or another, of having argued with someone whose world view was quite at odds with our own, and of having come away thinking, "I'm sure he's wrong, and I can't follow much of it, but still, it makes you think." The prospect of losing that sort of effect in public discourse is, frankly, frightening - terrifying, even, if we are to imagine its being replaced by a form of "deliberation" that, in the name of "fairness" or "reasonableness" (or worse still, "balance") consists of bland appeals to harmless nostrums that are accepted without question on all sides. That is to imagine open-ended public debate reduced to the formal trivia of American television networks.{24}

The legitimating, stabilizing, and epistemic functions of discourse are of course all inter-related. Any concern with justification of the law as a social practice rather than a merely theoretical endeavor, must take very seriously the wider social context and conditions for such a practice to flourish and succeed, and these conditions not only cannot be met by the propositional model; the propositional model actively frustrates them.

§3 Remedying the Deficiency in Liberal Conceptions of Discourse: Shifting the Spotlight from Speech-Governing Rules to Action-Governing Virtues

Having considered some of the deficiencies of the constrained-propositional conception of public discourse, I would like to suggest that we re-think public discourse in more holistic and less narrowly rationalist and propositional terms. Public discourse is a space in which embodied human beings (not merely rational beings) encounter one another, relate to one another, endeavor to understand one another, and explore different ways of living together and of understanding their shared and individual lives. There is no single segment of public discourse that deals with the justification of laws and is hermetically sealed off from all others.{25} On the contrary, public discourse "flicks" from topic to topic almost effortlessly, and its various subject-matters cannot but influence one another -- indeed, sometimes profoundly. There are obviously many different loci or fora of public discourse, not just one national sphere, as Iris Young helpfully indicates.{26}

As an activity, public discourse involves real-live, flesh-and-blood human beings, and it is constituted by their more or less successful efforts to communicate with each other. Like all human practices, this is a practice that is subject to manipulation and abuse, but the student of discourse can develop an account that helps us identify ways in which discourse can fail or succeed from a moral and practical point of view. If we are to come to an adequate understanding of the legitimating, stabilizing, and epistemic functions of discourse and how best they can be realized, we must contend with the reality of discourse in a holistic rather than piecemeal fashion. Therefore, instead of focusing on the bare rational content of discourse -- an approach which, as I have suggested, has proven problematic and self-defeating -- I believe it is now time to focus our attention on the moral habits, dispositions, and relationships that constitute public discourse. For convenience, I will refer to this as the virtue-centered, as opposed to the rule-centered or propositional conception of discourse. Here, I merely want to sketch some of the central elements of such an account; in the next section, I will consider some of its implications for the role of religion in public discourse.

In calling for a shift from content-oriented rules to action-oriented virtues in the study of public discourse, I am not of course suggesting that rules have no place at all in such an account. It is undoubtedly the case that given a certain understanding of appropriate and inappropriate, healthy and unhealthy, ways of relating to our fellow discussants, certain forms of discourse can be ruled out as disrespectful from the outset. However, what I am arguing against is the notion that whole categories of discourse, that we know can be potentially conducted in good faith and with the interests of society and individuals in view, can be ruled out of court as modes of assessing and justifying laws. This attempt at formulating ambitious principles of restraint to govern the content of discourse has proven impractical and self-defeating time and time again. It either results in trivial principles that almost all can agree to already, e.g., that nobody should publicly call for the indiscriminate imprisonment of this or that race, or it results in controversial principles that can only be justified by recourse to controversial worldviews, e.g., secularist epistemologies, which pre-emptively assume for themselves the epistemic function of discourse.

Indeed, it may be that one of the biggest selling-points of the virtue-centered account is that it is much more deferential towards the epistemic function of discourse, i.e., it recognizes that many difficult issues of epistemology and morality must be worked out through an open and honest discussion rather than settled prior to discussion by the "armchair theorist."{27} The virtue-centered account can accept Waldron's dynamic, agonistic model of discourse outlined above and argue that only such an open-ended model can respect the epistemic function of discourse. Of course, the theorist has every right to share her ideas about social and political morality with the rest of the world, or with her academic colleagues, but she cannot pre-emptively settle heated moral disputes in advance of the discursive process. Rather than fixing the rules of discourse once and for all and in advance, she is joining her voice in the extended conversation which we are calling "public discourse." Of course, she may suggest that certain rules are desirable or undesirable, but many issues cannot be appropriately or realistically settled in advance of discourse itself. These include the appropriate political role, and the epistemic status, of religious propositions. Through discourse, this sort of issue can be gradually illuminated even if never settled definitively. This is part of the epistemic or knowledge-enhancing function of discourse.

In short, the virtue-centered account does not establish rigid or comprehensive rules governing the content of discourse and thus tends to allow a lot more room for a quasi-Socratic quest for the truth among discussants, while structuring this quest with interpersonal dispositions and virtues that can be accepted as important by a very large proportion of thoughtful citizens.

What, then, are the main elements of a virtue-centered account of discourse? They consist primarily of what I will call communicative virtues, virtues that one naturally associates with communication and especially with communication in a morally and culturally pluralistic polity. These virtues are not equivalent to Habermas's communicative ethics, which tends to treat discourse in a much more rationalist and propositional way, and also involves the claim, which I find dubious but need not discuss further here, that manipulative or disrespectful discourse is effectively a performative contradiction.{28} Some of the basic communicative virtues or moral excellences are the capacity to listen, the desire to understand one's interlocutor, truthfulness, magnanimity, humility, generosity, patience, perseverance, a love of the common good, a genuine concern for the interests of one's neighbors, respect for personal freedom, tact, sensitivity and openness to the concerns and experiences of others, a willingness to back down when one is proven wrong, and a willingness to compromise within the bounds of moral integrity. As you can see, the list could be expanded quite easily. I believe a lot of energy has been wasted trying to work out hard-and-fast exclusionary rules to govern the content of discourse, when political philosophers' energies would be much more fruitfully expended exploring the content and implications of the communicative virtues.

One of the most promising virtue-centered accounts of discourse that I have encountered can be found in Mark Kingwell's A Civil Tongue: Justice, Dialogue, and the Politics of Pluralism (1995).{29} Kingwell claims that "just talk" is one of the central, if not the central, virtue of a just liberal society. Although he may exaggerate the importance and centrality of communication and discourse in a liberal democracy, there is no denying that today, perhaps more than ever before, the justice of a polity and the quality of its shared life are deeply dependent on the justice, fairness, and quality of its public discourse. Kingwell summarizes the virtues of discourse in the terms "politeness" and "civility," and suggests that these need not be conceived narrowly as the slavish pursuit of convention and etiquette, but as phronetic virtues that help a discussant adapt his or her behavior to any given discursive context. Although this requires a special kind of prudence, it is not merely tactical or strategic or narrowly self-interested or instrumental prudence. Rather, I would suggest that it is a prudence oriented towards the good of the other as well as one's own good and the good of one's polity, as these are realized through speech.

We have already considered briefly the advantages of the virtue-centered account as far as the epistemic function of discourse goes. Now, I would like to comment briefly on its advantages for discourse's legitimating and stabilizing functions. First, by opening laws and policies to a wide range of challenges and interpretations, and permitting citizens to voice their interests and perspectives in a dynamic, give-and-take, process, the virtue-centered account opens up the lawmaking process to a wider array of inputs and critical voices. This has two consequences for legitimacy: first, more citizens feel they can have some legitimate influence on the lawmaking process, and this enhances the democratic legitimacy of that process in their eyes and minimizes alienation and resentment towards political authority; second, the laws and policies are subjected to a wider-ranging critique and discussion, and insofar as criticisms are aired rather than artificially suppressed, lawmakers are made more publicly accountable and the legitimacy of laws is enhanced.

Now, the stabilizing function is closely allied to its legitimating function. The relationship can be captured roughly by saying that legitimacy or at least perceived legitimacy often has far-reaching consequences for the stability of a political regime, especially in a modern liberal democracy. Other things being equal, the greater the perceived legitimacy of a regime, the less likely that citizens will become disgruntled, rebellious, uncooperative, and violent. The virtue-centered account, by rejecting artificial theoretical restrictions on the content of discourse, facilitates discursive outlets or channels for minority and majority views alike. Thus, views that more restrictive accounts would drive out of public discourse and into sub-cultures may be aired and tested in the to-and-fro of public discourse. This draws citizens into a shared or common public life, and thus fosters a sense of belonging and a sense that one is taken seriously by one's fellow citizens rather than dismissed out of hand. Bonds of mutual understanding and respect may develop in these circumstances if citizens are sensitive to the interpersonal and communicative virtues required to make this process work well. It is only in the context of such an inclusive and civil conversation about the political common good that citizens can overcome their sense of alienation and mutual suspicion and develop a strong sense of loyalty to this common enterprise which we call the "nation."

§4 Re-assessing Religious Interventions within a Richer Model of Discourse

I would now like to consider some of the implications of the virtue-centered account for the role of religion in public discourse. Although this is not the place to embark on a comprehensive assessment of the question, a few of the more obvious points bear mentioning. First, a phronetic or virtue-centered account, although rejecting in general the restraint-centered strategy and calling for a much more dynamic and agonistic model of discourse, does not issue a license to participants to bring up whatever they feel strongly about, whenever they feel like it, in whatever discursive context they find themselves in. On the contrary, it calls for an attitude of responsibility and a phronetic sensitivity to context among discussants. This means that intelligent discussants should recognize that discourse is a multi-faceted practice and that even a single person can assume different roles depending on the discursive context. How one behaves must be determined not only by the sum of one's convictions, but by the role one occupies (e.g., am I speaking as a son, a spouse, a colleague, a parent, a close friend, a fellow train passenger, a brother in faith, a legislator, a judge, a citizen, a campaigner? etc.) and the particular person or group one is addressing (e.g., am I speaking to a person who is responsive to "tough talk" or somebody who is very sensitive and only requires gentle understatement in order to be responsive to my message?). Tact as well as legitimate strategic considerations often counsel that I exercise a sort of temperance in my speech and either remain silent or re-formulate my point in a manner that is will be more effective and constructive under the circumstances.

Although moral excellence in discourse can positively require deliberate restraint in certain contexts, there may well be other situations in which one finds that disagreement runs so deep and an issue is so pressing and important that the only realistic contribution one can hope to make is to witness to one's moral beliefs with courage and hope that at least some people will see the cogency of, be influenced by, or at least become aware of, one's position.{30} This is a far cry from Ackerman's "constrained conversation"{31} or Rawls's notion of public reason, in which one is (morally) required to systematically silence one's comprehensive doctrines or deeply contentious views when talking about important political issues such as coercive law (for Ackerman) or fundamental justice (for Rawls).

What does this tell us about the role of religion in public discourse? Well, it does not tell us that religion is per se an illegitimate basis for public argumentation about the law. Of course, neither does it suggest that any and every religiously-based argument for a law or policy is cogent, valid, or just. Rather, it suggests that religious discussants must determine themselves, moved by phronetic considerations and the communicative virtues as well as the contributions of other discussants, when restraint is and is not appropriate. At the limit, the legitimacy of their reliance on religion may be challenged by other discussants, but this challenge, if it is made in good faith rather than used as a trick to dismiss adversaries, may be met on its merits.

If religious believers exercise the virtues of communication, all of which I believe are consistent with most religions, then they will undoubtedly find that public discourse requires a strong moral character and a generous dose of fortitude and humility. However, they, along with those either opposed to or suspicious of religious discourse during policy discussions, will not need to flounder under any intellectually despotic or arbitrary exclusionary rules; instead, they should endeavor to understand rather than merely be understood; to listen rather than merely be heard; to address the concerns of others and not only their own concerns; and to allay the fears of others regarding the implications of religious (or secular) discourse where this seems appropriate.

I do not claim that my account of discourse offers any neat resolution of the issues surrounding religious and moral diversity in liberal polities, nor do I think any realistic account could claim to offer such a resolution. However, I do believe that it offers a much more accurate description of the nature, dynamics, and goals of discourse than more traditional rule-centered models have offered, and I think it illuminates some important normative dimensions of discourse that most participants, religious and non-religious alike, can accept without compromising their moral integrity and deepest moral convictions. In short, it has the dual advantage of both being truer to the reality of public discourse as a social practice, and of offering a convincing basis for continuing the extended conversation about the common good in an atmosphere of freedom and mutual respect.


Ackerman, Bruce, Social Justice in the Liberal State (New Haven and London: Yale University Press, 1980)

---, "Why Dialogue?" Journal of Philosophy 86, 1 (January 1989)

Alexander, Larry, "Liberalism, Religion, and the Unity of Epistemology", San Diego Law Review 30, (Fall 1993): 763-797

Audi, Robert, Religious Commitment and Secular Reason (Cambridge: Cambridge University Press, 2000)

Audi, Robert and Nicholas Wolterstorff, Religion in the Public Square: The Place of Religious Convictions in Political Debate (London: Rowman & Littlefield Publishers, Inc., 1997)

Eberle, Christopher, Religious Convictions in Liberal Politics (Cambridge: Cambridge University Press, 2002)

Galston, William, Liberal purposes: Goods, virtues and diversity in the liberal state (Cambridge: Cambridge University Press, 1991)

Gaus, Gerald F., Justificatory Liberalism: An Essay on Epistemology and Political Theory (Oxford and New York: Oxford University Press, 1996)

George, Robert P., Making Men Moral (Oxford: Clarendon Press, 1993)

Greenawalt, Kent, Private Consciences and Public Reasons (New York: Oxford University Press, 1995)

Gutmann, Amy and Dennis Thompson, Democracy and Disagreement: Why moral conflict cannot be avoided in politics, and what should be done about it (Cambridge, Massachusetts and London, England: The Belknap Press of Harvard University Press, 1996)

Habermas, Jurgen, The Theory of Communicative Action, Vol. 1, trans. T. McCarthy (Boston: Beacon Press, 1984)

---, The Theory of Communicative Action, Vol. 2, trans. T. McCarthy (Boston: Beacon Press, 1987)

Hollenbach, David, "Contexts of the Political Role of Religion: Civil Society and Culture", San Diego Law Review 30, (Fall 1993): 877-901

Kingwell, Mark, A Civil Tongue: Justice, Dialogue, and the Politics of Pluralism (University Park, Pennsylvania: The Pennsylvania State University Press, 1995)

Larmore, Charles, Patterns of Moral Complexity (Cambridge: Cambridge University Press, 1987)

---, "Political Liberalism", Political Theory 18, 3 (August 1990 1990): 339-360

Nagel, Thomas, "Moral Conflict and Political Legitimacy", Philosophy and Public Affairs 16, 3 (Summer 1987): 215-240

Perry, Michael J., Love and Power: The Role of Religion and Morality in American Politics (New York and Oxford: Oxford University Press, 1991)

---, "Religious Morality and Political Choice: Further Thoughts -- And Second Thoughts -- on Love and Power", San Diego Law Review 30, (Fall 1993): 703-727

---, "Why Political Reliance on Religiously Grounded Morality is not Illegitimate in a Liberal Democracy", Wake Forest Law Review 36, (Summer 2001): 217-249

Rawls, John, Political Liberalism, The John Dewey Essays in Philosophy (New York: Columbia University Press, 1996)

Raz, Joseph, The Morality of Freedom (Oxford: Clarendon Press, 1986)

Rorty, Richard, "Religion as Conversation-Stopper", Common Knowledge 3, (Spring 1994): ??

Schwarzschild, Maimon, "Religion and Public Debate in a Liberal Society: Always Oil and Water or Sometimes More Like Rum and Coca-Cola?" San Diego Law Review 30, (Fall 1993): 903-915

Waldron, Jeremy, "Religious Contributions in Public Deliberation", San Diego Law Review 30, (Fall 1993): 817-848

Weithman, Paul, Religion and the Obligations of Citizenship (Cambridge: Cambridge University Press, 2002)

Young, Iris, Inclusion and Democracy (Oxford: Oxford University Press, 2000)

{1} I would like to thank my colleague, Matt Mendham, for some helpful editorial and philosophical comments on an earlier draft of this paper. Thanks are also due to Prof. Phil Quinn for leading some very stimulating discussions on this and related subjects in a philosophy graduate seminar at the University of Notre Dame in the Spring semester of 2003.

{2} Richard Rorty, "Religion as Conversation-Stopper", Common Knowledge 3, (Spring 1994): ??, 2.

{3} See, inter alia, Robert Audi, Religious Commitment and Secular Reason (Cambridge: Cambridge University Press, 2000); Kent Greenawalt, Private Consciences and Public Reasons (New York: Oxford University Press, 1995); Bruce Ackerman, Social Justice in the Liberal State (New Haven and London: Yale University Press, 1980); John Rawls, Political Liberalism, The John Dewey Essays in Philosophy (New York: Columbia University Press, 1996); Rorty, "Religion as Conversation-Stopper"; Gerald F. Gaus, Justificatory Liberalism: An Essay on Epistemology and Political Theory (Oxford and New York: Oxford University Press, 1996); Thomas Nagel, "Moral Conflict and Political Legitimacy", Philosophy and Public Affairs 16, 3 (Summer 1987): 215-240.

{4} My arguments against religious restraint are inspired by a rich literature and it would be unduly cumbersome to acknowledge my intellectual debts at every step of the argument. The following have been particularly helpful: Nicholas Wolterstorff's arguments in the book-long debate, Robert Audi and Nicholas Wolterstorff, Religion in the Public Square: The Place of Religious Convictions in Political Debate (London: Rowman & Littlefield Publishers, Inc., 1997); Michael J. Perry, Love and Power: The Role of Religion and Morality in American Politics (New York and Oxford: Oxford University Press, 1991), and Michael J. Perry, "Why Political Reliance on Religiously Grounded Morality is not Illegitimate in a Liberal Democracy", Wake Forest Law Review 36, (Summer 2001): 217-249. For an excellent argument against Robert Audi's principles of secular rationale and secular motivation (Audi, Religious Commitment and Secular Reason), see Quinn, P.L., "Religion and Politics, Fear and Duty" (forthcoming?, 2003). I would especially refer the reader to the enlightening debate conducted in the San Diego Law Review in Fall 1993. Arguments against blanket principles of religious restraint published there include Jeremy Waldron, "Religious Contributions in Public Deliberation", San Diego Law Review 30, (Fall 1993): 817-848; Michael J. Perry, "Religious Morality and Political Choice: Further Thoughts -- And Second Thoughts -- on Love and Power", San Diego Law Review 30, (Fall 1993): 703-727; Maimon Schwarzschild, "Religion and Public Debate in a Liberal Society: Always Oil and Water or Sometimes More Like Rum and Coca-Cola?" San Diego Law Review 30, (Fall 1993): 903-915; David Hollenbach, "Contexts of the Political Role of Religion: Civil Society and Culture", San Diego Law Review 30, (Fall 1993): 877-901; and Larry Alexander, "Liberalism, Religion, and the Unity of Epistemology", San Diego Law Review 30, (Fall 1993): 763-797.

{5} Terms such as "rational," "reasonable," and "accessible" are notoriously problematic, but for now I only wish to convey a sense of the liberal conception of respect.

{6} The principle of consent in Locke and in the Declaration concerns the general ends of government and law rather than the specific grounds of discrete laws. It is also the consent of society at large rather than the consent of each and every individual that is primarily at stake in Locke's and the Founders' discussions of legitimacy. Locke argues in Two Treatises of Government, 3rd ed, Cambridge Texts in the History of Political Thought, ed. P. Laslett (Cambridge: Cambridge University Press, 1988), Second Treatise, chap. XIX, §222 (pp.412-13) that "[f]or since it can never be supposed to be the Will of the Society, that the Legislative should have a power to destroy that, which every one designs to secure, by entering into Society, and for which the People submitted themselves to the Legislators of their own making; whenever the Legislators endeavorto grasp themselves, or put into the hands of any other an Absolute Power over their Lives, Liberties, and Estates of the People; by this breach of Trust they forfeit the Power, the People had put into their hands, for quite contrary ends, and it devolves to the People, who have a Right to resume their original Liberty, and, by the Establishment of a new Legislative (such as they shall think fit) provide for their own Safety and Security, which is the end for which they are in Society." Similarly, the American Declaration of Independence famously proclaims that among the "Truths" held to be "self-evident" is the truth that "to secure these Rights [to Life, Liberty, and the Pursuit of Happiness], Governments are instituted among Men, deriving their just Powers from the Consent of the Governed, that whenever any Form of Government becomes destructive of these Ends, it is the Right of the People to alter or to abolish it, and to institute new Government, laying its Foundation on such Principles, and organizing its Powers in such Form, as to them shall seem most likely to effect their Safety and Happiness."

{7} See Audi, Religious Commitment and Secular Reason, 65-69.

{8} Ibid., 66-67.

{9} I am using the term "accessible," where Rawls and some other liberals might use the term "public," or "acceptable by a reasonable person." The exact term we use is immaterial to the task at hand: namely, to determine what epistemic quality the grounds for a law must have in order to render the law morally legitimate from a broadly liberal perspective. According to Robert Audi, "[a]t least in the vast majority of cases where there are adequate reasons for the relevant kind of coercion, these reasons are suitably accessible (say, with some explanation and perhaps some instruction) to normal rational persons" (Ibid., 67). Kent Greenawalt attempts to draw a distinction between "accessible" and "nonaccessible" grounds of political decisions in Greenawalt, Private Consciences and Public Reasons, 23 ff. Similarly, Amy Gutman and Dennis Thompson believe that citizens should "press their public claims in terms accessible to their fellow citizens" (Amy Gutmann and Dennis Thompson, Democracy and Disagreement: Why moral conflict cannot be avoided in politics, and what should be done about it (Cambridge, Massachusetts and London, England: The Belknap Press of Harvard University Press, 1996), 57. Paul Weithman makes a compelling argument to the effect that "the notion of accessibility cannot be adequately spelled out . . . because the counterfactuals needed to make the notion clear [do not] express plausible conditions on the reasons citizens should be ready to offer" (Paul Weithman, Religion and the Obligations of Citizenship (Cambridge: Cambridge University Press, 2002), 178).

{10} For a more detailed account of the burdens of judgment, see Rawls, Political Liberalism (henceforth PL), 56-57. The burdens of judgment include complexity and scarcity of evidence; deciding what weight to assign to various considerations; the problem of interpreting or applying abstract concepts and criteria; unique life experiences that inform our judgments; settling on and interpreting correct moral standards; and deciding how to prioritize competing goods.

{11} Rawls is one of the most well-known architects of the principle of reciprocity as it is used by contemporary liberal theorists. According to the principle or reciprocity as understood by Rawls, "our exercise of political power is proper only when we sincerely believe that the reasons we offer for our political action may reasonably be accepted by other citizens as a justification of those actions" (Introduction to PL, xlvi). Later, Rawls remarks, "To make more explicit the role of the criterion of reciprocity as expressed in public reason, I note that its role is to specify the nature of the political relation in a constitutional democratic regime as one of civic friendship. For this criterion, when citizens follow it in their public reasoning, shapes the form of their fundamental institutions. For example -- I cite easy cases -- if we argue that the religious liberty of some citizens is to be denied, we must give them reasons they can not only understand -- as Servetus could understand why Calvin wanted to burn him at the stake -- but reasons we might reasonably expect that they as free and equal might reasonably also accept. The criterion of reciprocity is normally violated whenever basic liberties are denied. For what reasons can both satisfy the criterion of reciprocity and justify holding some as slaves, or imposing a property qualification on the right to vote, or denying the right of suffrage to women?" (ibid., li).

{12} For definitions of religious reasons, see Perry, Love and Power: The Role of Religion and Morality in American Politics, 66- 82; Audi, Religious Commitment and Secular Reason, 34-35; Greenawalt, Private Consciences and Public Reasons, 39; Weithman, Religion and the Obligations of Citizenship, 122.

{13} We need not settle here which laws or which category of laws deserves special protection from religious grounding. Some, such as Audi, earmark significantly coercive laws for special treatment, while others, such as Rawls, earmark laws involving "constitutional essentials and basic justice."

{14} Rawls stipulates in that by a "fully comprehensive" doctrine is meant a philosophical or religious view that "covers all recognized values and virtues within one rather precisely articulated system; whereas a "partially comprehensive" doctrine "comprises a number of nonpolitical values and virtues and is rather loosely articulated" (PL, 152, n. 17).

{15} For a comprehensive treatment and critique of the "argument from Bosnia" and of a more moderate form of this argument, see Christopher Eberle, Religious Convictions in Liberal Politics (Cambridge: Cambridge University Press, 2002), 152-186.

{16} See Audi, Religious Commitment and Secular Reason, 100-103; and Rorty, "Minimalist Liberalism," ??

{17} This point is made forcefully by Eberle in Religious Convictions in Liberal Politics, 158-166: "the argument from Bosnia founders on the fact that there are crucially important differences between the conditions that obtained in the confessional states party to the wars of religion. In particular, the confessional states' denial of religious freedom gave to religion an incendiary potential it lacks when citizens are free from the threat of persecution for their religious commitments and practices" (166). A similar argument is made by Maimon Schwarzschild in "Religion and Public Debate in a Liberal Society": "In fact, the Enlightenment stance towards religion seems to me largely an anachronism in developed countries today, and to that extent provides a quaint guide to the ethics of public debate in a modern society" (904).

{18} William Galston shows that the violation of negative liberty in the interests of the individual affected cannot be considered wrong in principle: "To invade negative freedom in the name of the real good is to promote the individual's benefit over his or her harm, rationality over irrationality, truth over error. In practice, such invasions can be wrong in principle only if the mere fact that the impetus toward the good is external somehow negates the worth of the good end so achieved, that is, only if the consciously willed pursuit of a goal is a necessary condition of the value of attaining it. As recent discussions of paternalism have shown, this proposition cannot be defended. Freeing an individual from heroin addiction is good even though the afflicted individual may not consciously will his or her liberation. Indeed, it may well be that th individual cannot affirm the worth of non-addiction before having been coerced to attain it." (William Galston, Liberal purposes: Goods, virtues and diversity in the liberal state (Cambridge: Cambridge University Press, 1991), 86-87. Also, Joseph Raz brings into question the notion that respect requires one to bracket out "comprehensive" or so-called "perfectionist" reasons for coercing another person: "The spring from which anti-perfectionism flows is the feeling that foisting one's conception of the good on people offends their dignity and does not treat them with respect. Is one treating another with respect if one treats him in accordance with sound moral principles, or does respect for persons require ignoring morality (or parts of it) in our relation with others? There can be little doubt that stated in this way the question admits of only one answer. One would be showing disrespect to another if one ignored moral considerations in treating him" (Joseph Raz, The Morality of Freedom (Oxford: Clarendon Press, 1986), 157).

{19} When I say, "rational interests," I do not mean interests conceived within a purely egoistic calculus nor interests in some purely subjective sense. Nor do I suggest that the concept of "rational interest" can be completely separated from a person's subjective sense of well-being.

{20} An argument along these lines is advanced in Robert P. George, Making Men Moral (Oxford: Clarendon Press, 1993).

{21} See Jurgen Habermas, The Theory of Communicative Action, Vol. 1, trans. Thomas McCarthy (Boston: Beacon Press, 1984); and Jurgen Habermas, The Theory of Communicative Action, Vol. 2, trans. Thomas McCarthy (Boston: Beacon Press, 1987); Charles Larmore, "Political Liberalism", Political Theory 18, 3 (August 1990 1990): 339-360; Bruce Ackerman, "Why Dialogue?" Journal of Philosophy 86, 1 (January 1989); Audi, Religious Commitment and Secular Reason ; Nagel, "Moral Conflict and Political Legitimacy".

{22} This idea is well-captured by Rawls's notion of finding an "overlapping consensus" upon which to build a conception of political justice. Similarly, it is expressed negatively by the "bracketing" strategy advocated by Ackerman to deal with fundamental moral conflict in political discourse: "When you and I learn that we disagree about one or another dimension of the moral truth, we should not search for some common value that will trump this disagreement; nor should we try to translate it into some putatively neutral framework; nor should we seek to transcend it by talking about how some unearthly creature might resolve it. We should simply say nothing at all about this disagreement and put the moral ideals that divide us off the conversational agenda of the liberal state" (Ackerman, "Why Dialogue?", 16). This bracketing strategy is also favored by Charles Larmore, who says, "When two people disagree about some specific point, but wish to continue talking about the more general problem they wish to solve, each should prescind from the beliefs that the other rejects, (1) in order to construct an argument on the basis of his other beliefs that will convince the other of the truth of the disputed belief, or (2) in order to shift to another aspect of the problem, where the possibilities of agreement seem greater" (Charles Larmore, Patterns of Moral Complexity (Cambridge: Cambridge University Press, 1987), 53).

{23} For arguments against the liberal model of discourse and in favor of more complex and dynamic models, see, inter alia, Mark Kingwell, A Civil Tongue: Justice, Dialogue, and the Politics of Pluralism (University Park, Pennsylvania: The Pennsylvania State University Press, 1995) ; Waldron, "Religious Contributions in Public Deliberation"; Hollenbach, "Contexts of the Political Role of Religion"; Iris Young, Inclusion and Democracy (Oxford: Oxford University Press, 2000); and Galston, Liberal Purposes, 98-117.

{24} Waldron, "Religious Contributions in Public Deliberation", 841-2.

{25} The complexity of public discourse is pointed to by David Hollenbach's simple observation of the obvious yet so often neglected fact that "political debate is not simply argument about whether to adopt or reject certain policies" (David Hollenbach, "Contexts of the Political Role of Religion: Civil Society and Culture," 878.

{26} Elaborating on a Habermasian theme, Young argues that "it is a mistake to think of the deliberatively democratic process as one that engages a unified people making decisions for society as a whole. Instead, processes of deliberation in complex mass society must be understood as subjectless and decentred. Among other things, this implies abandoning traces of face-to-face interaction as the model of public discussion, and instead reinterpreting public debate as mediated among people dispersed in space and time" (Young, Inclusion and Democracy, 167). While I think Young goes too far in suggesting that the public sphere is "subjectless," she is surely correct to point out that it is "decentered" and "dispersed in space and time."

{27} This is the thrust of Kingwell's criticism of proponents of what he calls "constrained liberal dialogue": "instead of stopping short with a defensible clearing of dialogic space, and letting the citizens themselves decide what rules are just, Ackerman (like Rawls, Nozick, and others) wants to work out, in some detail, a set of justified rules" (Kingwell, A Civil Tongue, 51).

{28} See "Discourse Ethics: Notes on a Program of Philosophical Justification," in Habermas's Moral Consciousness and Communicative Action, trans. Christian Lenhardt and Shierry Weber Nicholsen (Cambridge, Massachussetts: MIT Press, 1990), 88-89: "Participants in argumentation cannot avoid the presupposition that, owing to certain characteristics that require formal description, the structure of their communication rules out all external or internal coercion other than the force of the better argument and thereby also neutralizes all motives other than that of the cooperative search for truth." Quoted in Kingwell, A Civil Tongue, 158-159.

{29} Mark Kingwell, A Civil Tongue: Justice, Dialogue, and the Politics of Pluralism (University Park, Pennsylvania: The Pennsylvania State University Press, 1995). Thanks to my colleague, Matt Mendham, for bringing this book to my attention.

{30} Liberals often underestimate the power and potential contribution of bluntly expounding one's worldview rather than constructing an argument based on shared premises. On the one hand, it may greatly enhance mutual understanding and honest dialogue between discussants. On the other hand, on certain occasions it may be even more persuasive than consensus-based discourse. As Galston astutely points out, "The point of much dialogue is to invite one's interlocutor to see the world the way you do, or at least to understand what it is like to see the world the way you do. One way of doing that is the reverse of 'prescinding' form disputed issues: namely, stubbornly bearing witness to one's stance at the precise point of difference . . . . Now clearly the critic is not coercing me to change but, on the contrary, is appealing to something we have in common. But that something is not a premise; it is, rather, an experience" (Galston, Liberal Purposes, 106).

{31} See Ackerman, "Why Dialogue?"