LAW BEYOND GOD AND KANT: A PRAGMATIST PATH

Abstract

The liberal principle of reciprocity requires that states maintain neutrality with respect to their citizens’ competing comprehensive worldviews (both religious and secular) while officially justifying the law and while adjudicating under it. But the very possibility of such liberal neutrality has come under attack in a post-Enlightenment world in which even foundational arguments for the principle of reciprocity itself are no longer taken for granted. This article offers a pragmatist path to the resolution of this liberal dilemma. It recommends a “default and challenge” model for legal justification and legitimation that is rooted in social-linguistic practice. By rooting justification and legitimation in practice, it is argued liberal neutrality can be preserved without need for appeals to controversial foundational commitments at any level of public political justification. The article closes with a fictional case study concerning abortion to show how politicians and courts can apply this method to preserve liberal neutrality while addressing even the most controversial issues.

Introduction: The Post-Enlightenment Liberal Dilemma

Near the close of the 2012 American vice-presidential debate between Vice President Joseph Biden and Representative Paul Ryan, the moderator asked the candidates, “Tell me what role your religion has played in your own personal views on abortion.” ¹ The question and the candidates’ responses throw into sharp relief an important dilemma for Western liberal democracies in the face of increasing pluralism.

Rep. Ryan responded, “I don't see how a person can separate their public life from their private life or from their faith. Our faith is in everything we do.” ² He then proceeded to explain the pro-life position demanded, at least in part, by his commitment to Catholicism. ³ Ryan left the impression that, should the opportunity present itself, he would use his political office to change the nation's abortion policy on religious grounds. But this presents a problem. Western liberal democracies have historically rejected religion as a source of justification for law. This is because their moral conception of persons as free and equal entails a commitment to the liberal principle of reciprocity: The exercise of political coercion is justified only when it is based on reasons all citizens can be expected to accept. ⁴ In a pluralistic society, public political justification by appeal to one religion inevitably violates this principle because it draws on reasons only believers in that particular religion can be expected to accept. Such religious justification fails to show equal respect to persons because it arbitrarily privileges those who share the relevant faith.

Vice President Biden answered the question by rejecting its premise. He explained that while he personally accepts the Catholic Church's anti-abortion doctrine as de fide, ⁵ he refuses “to impose [his personal religious beliefs] on equally devout Christians and Muslims and Jews” ⁶ who do not share those beliefs. Biden's response attempts to respect the liberal principle of reciprocity by arguing that his religious reasons, as matters of subjective faith, should be excluded from public political justification altogether. The claim that such discourse should be limited to secular reasons is implicit in this response. While Biden's answer is consistent with the liberal orthodoxy of striving for justification on secular grounds in order to preserve the neutrality demanded by the liberal principle of reciprocity, it is subject to criticism of a different sort: It presumes the existence of a secular vocabulary that is neutral in a way that religious vocabularies are not.

Historically, the Enlightenment liberal attempted to satisfy the test of reciprocity by enlisting secular principles of science and moral philosophy that were generally regarded as rationally demonstrable (and therefore objective and universal). The idea was that both Catholics and Protestants must agree that the velocity of a body remains constant unless an external force acts upon it. And both Protestants and Catholics must agree that, as free and equal, all persons should be treated as ends in themselves, and never as mere means. If the Enlightenment liberal's interlocutor challenged either of these propositions, then the challenge could be dismissed as a mistake due to some defect in reasoning. ⁷ However, a decade into the twenty-first century, any liberal bold enough to claim access to an “objective” or “universal” account of human nature and justice will be met with suspicion, and perhaps even scorn. The pluralism once thought to be limited to the religious sphere has bled into the secular, undermining secularism's privileged claim to neutrality in both science and morals. ⁸ With this in mind, Biden's answer appears no less problematic than Ryan's from the standpoint of reciprocity: unless he is prepared to offer a defense of his pro-choice position that is demonstrably “objective” and “universal” (that is, rationally compelling to all human beings as such), his secular justification is no less controversial and no less a matter of private faith than Ryan's appeal to religion. ⁹

This example illustrates the post-Enlightenment dilemma for pluralistic Western liberal democracies: The liberal principle of reciprocity demands that the use of coercive power be justified in terms all citizens can accept, but neither religious nor secular vocabularies can satisfy the test. Consistency therefore demands that the liberal either (1) lift the ban on religious reasons from the public political forum (henceforth the “open competition model”), or (2) find some way to exclude both religious and secular foundational commitments ¹⁰ from the public square, which I call the “foundation-neutral model.”

Neither option appears promising at first blush. I argue below that the open competition model risks fomenting political instability, stifling public political discourse, and, most importantly, running afoul of the liberal principle of reciprocity. But the foundation-neutral model has problems of its own. How can a liberal conception of justice (or any conception of justice) be articulated, relied upon for adjudication, and made normative for a society absent an express appeal to some foundational commitments? What would supply the normative force for a moral vocabulary deprived of foundational moorings? While the foundation-neutral model may respect the liberal principle of reciprocity and preserve the core commitments of liberalism, it may make theoretically and practically impossible demands.

In what follows, I offer a novel solution to this liberal dilemma. The strategy applies methods advanced by the recent philosophical movement of neopragmatism. The method achieves foundation-neutrality by grounding political justification in the social-linguistic practices of a state's public political culture. It conceives political discourse under a “default and challenge” model whereby certain of our key first-order moral commitments enjoy a default status in political discourse that shifts any justificatory burden to those who would challenge them. This default status is exploited to offer an authoritative public resolution of complex and controversial social issues without ever appealing to citizens’ controversial foundational commitments. By grounding political justification in shared social practices, the proposed model achieves normativity for pluralistic liberal societies while preserving reciprocity. Once the method is articulated, I offer a practical example of how politicians and courts can apply it to address even the most controversial social issues, such as abortion. But first, I motivate my defense of this foundation-neutral model by offering three reasons why its alternative, the open competition model, is, importantly, flawed.

Three Problems for the Open Competition Model

The open competition model envisions a politics of difference ¹¹ wherein adherents to irreconcilable comprehensive religious and secular worldviews will enjoy complete freedom to compete and win supremacy for their controversial foundational commitments as express justifications and sources of authority for a state's constitutional order. ¹² Political candidates, legislators, judges, and other government office holders may all join in. So, for example, the open competition model opens the door to official religious justifications for law (along the lines of that offered by Rep. Ryan in the context of abortion) by politicians and judges alike. In other words, it sets things up to allow the chips to fall as they may. Proponents of the open competition model tend to be motivated by the worthy ideals of equal respect, toleration, and celebration of diversity, but the model suffers from three important flaws: it leads to instability, violates reciprocity, and stifles discourse. ¹³

Leads to Instability

First, it is important to understand that, although proponents of the open competition model tend to be guided by the moral ideals of equal respect, toleration, and celebration of diversity, the model itself cannot ultimately be presented or justified in terms of these moral ideals. For to justify the open competition model itself by appeal to these moral ideals would require some foundational religious or secular story about why these ideals should be privileged and not others, and why open competition is the best means of recognizing these ideals. But if the open competition model itself were ultimately and finally justified by appeal to some such controversial religious or secular doctrine, then there would be no true open competition—for the model's very justification would take a side and foreclose competition. ¹⁴ The only way for the open competition model to avoid this paradox is to dispense with the idea of an overarching and shared moral justification for the model and instead conceive itself as a modus vivendi.

A modus vivendi represents an equilibrium point at which parties who are at odds agree to terms that will govern their relationship for no other reason than that observing those terms is in their strategic interest. ¹⁵ But a constitutional order based on a modus vivendi generates problems in terms of normativity and stability. Since the modus vivendi's terms do not represent a shared moral commitment, they are not independently normative for the parties. Drawing from H. L. A. Hart, the parties will regard the rules and principles supported by a modus vivendi from the “external point of view.” ¹⁶ Where rules are regarded from the external point of view, violating them is “merely a basis for the prediction that a hostile reaction will follow … [but it is not] a reason for hostility.” ¹⁷ The rules or principles do not themselves serve as “guides to the conduct of social life, as the basis for claims, demands, admissions, criticism, or punishment.” ¹⁸ This leads to instability.

Since a balance of interest rather than moral commitment supports the constitutional order as a modus vivendi, the parties recognize that if that balance changes, the terms will change as well. This recognition can be expected to plant suspicion that the other parties (different racial, ethnic, philosophical, cultural, or religious groups) will be looking for opportunities to increase their legitimate economic and political power in order to position themselves to force a more favorable renegotiation of the terms of the compromise. This fear incentivizes even those parties who are satisfied with the terms of the current modus vivendi to engage in preemptive strikes to prevent other parties from gaining sufficient power to upset the equilibrium. Such perverse incentives foment antagonism on all sides. ¹⁹ The public “us” and “them” orientation protected by the open competition model soon devolves into an “us” versus “them” mentality that undermines stability in the political arena.

In addition, where power is checked by nothing more than a modus vivendi, there are no restrictions on how much power one group may acquire or lose in open competition. Over time, there is a strong likelihood that the modus vivendi will change to reflect the vocabulary and demands of only the strongest group or groups. This leads to the second problem for a constitutional order based on the open competition model.

Violates Reciprocity

Imagine a pluralistic liberal society comprising the following groups: liberal Christians, 70 percent; liberal Jews, 10 percent; liberal Muslims, 10 percent; and liberal atheists, 10 percent. One might think that even if the Christian majority exerts its influence to place an express Christian stamp on the language of the modus vivendi reached among all the groups (for example, by expressing laws and judicial decisions in Christian terms), the substance of the compromise itself will nevertheless be liberal. As such, the constitutional order will be constrained by a healthy toleration for other worldviews and a respect for persons as free and equal. Thus, in practice, so long as the constitutional order qualifies as liberal, it should be a matter of indifference to the liberal non-Christian whether the laws are expressed in Christian terms. But we know this is not true.

A constitutional order that is officially articulated (by statutes, judges, legislators, the executive, and the like) in a language that 30 percent of the population cannot be reasonably expected to accept will marginalize this minority, alienating them from democratic discourse and from the democracy itself. Part of what keeps a society pluralistic is the crucial link between each member's self-description (her comprehensive worldview) and her self-respect. It may not change the scope of a constitutional right for legislators or judges to say that it derives from the Gospels, but is there any doubt that a Muslim, Jew, or atheist would suffer humiliation when forced to avail herself of a protection so articulated? ²⁰ Moreover, the non-Christian is justified in feeling used and disenfranchised by such a regime. The state derives its political power indiscriminately from the Christian and non-Christian citizen, but then it converts this power to the greater glory of the Christian god. As Richard Rorty puts it, “the best way to cause people long-lasting pain is to humiliate them by making the things that seemed most important to them look futile, obsolete, and powerless.” ²¹ If the aim of the liberal principle of reciprocity is to protect citizens from such marginalization, alienation, and insult, then to the extent that the open competition model allows for it, it must be regarded as the abandonment and not an adaptation of liberalism.

Stifles Discourse

The third problem with adopting the open competition model is that it can be expected to have a limiting effect on public political discourse. This is because it encourages public-political appeals to foundational commitments. As foundational, such commitments are themselves unarguable; they admit of no non-circular defense. This is not a problem for fellow believers, but, as Rorty argued, an appeal to foundational commitments in discourse with a non-believer is a “conversation-stopper.” ²²

For example, political discourse between a Christian fundamentalist and an atheist is much more likely to be terminated than advanced by appeal to their religious foundational commitments. When the Christian falls back on the position that pornography should be outlawed because it offends the Will of God, how is the atheist supposed to respond? ²³ At this point the Christian has signaled to the atheist that nothing short of arguing him out of his Christian faith will advance the discussion. Since the atheist will recognize that he is unlikely to succeed in this endeavor, he is more likely to walk away or change the topic than to continue the discussion. ²⁴ This is of course a problem for all appeals to foundational commitments, both secular and religious. The atheist will be just as likely to bring a conversation to an impasse by citing The Origin of Species to a Christian fundamentalist. ²⁵ In addition to making progress seem hopeless, appeals to foundational commitments may also be interpreted as signaling a lack of respect for one's interlocutor and an abandonment of political discourse as a cooperative endeavor. Thus, not only are appeals to foundational commitments under the open competition model often conversation-stoppers, they also tend to hamper the prospects of future productive discourse.

Jeffrey Stout challenges Rorty's claim that religion serves as a political conversation-stopper. According to Stout, Rorty's position reflects a “militant secularism” that overstates the perils of religious justification in politics and perpetuates a false (and socially harmful) political dichotomy between the strictly secular state on the one hand and theocracy on the other. ²⁶ Stout explains that, “[f]ar from persuading most religious people to confine their religious convictions to the private sphere, secularism gives them reason to conclude that liberal democracies are essentially inhospitable to their concerns.” ²⁷ This forces religious citizens to “retreat from public life into communities of like-mindedness, or attempt to use the electoral process to advance theocratic ends.” ²⁸ The result is a false dichotomy whereby secularism and theocracy “mirror one another, and feed one another's fears. The more power the theocrats acquire at home and abroad, the stronger the wording of the secularist manifestoes becomes—and vice versa.” ²⁹ For Stout, the best way to diffuse such efforts to dominate is to address the motivation by lifting the secular exclusion of religious justification from public political discourse.

Stout is right to point out that efforts to dominate public political discourse can come in both religious and secular flavors. Rorty often (though not always) wrote as if he missed this crucial point. ³⁰ But it is important to explain why Stout's critique of Rorty fails to answer the arguments offered here. First, the problems for the open competition model outlined above apply to reliance on all manner of foundational commitments in politics, whether they are religious, secular, or other. Second, open competition does not answer Stout's worry that the false dichotomy perpetuated by exclusively religious versus exclusively secular political visions will motivate attempts to use politics as a means of domination. It is true that militant secularism breeds antagonism and efforts at domination in politics, but I have argued that open competition does not improve upon secularism in this regard. As a modus vivendi, the open competition model will also breed an “us versus them” mentality that leads to suspicion, motivates preemptive strikes, and ultimately results in political instability. ³¹ Moreover, contrary to Stout's suggestion, open competition cannot avoid the vocabulary of domination in public political discourse because it licenses (and indeed encourages) official justification for the exercise of political power by appeal to some religious, secular, or other foundational commitment. Such justification is always expressed in a vocabulary of domination and marginalization to those who avow conflicting foundational commitments. ³²

For these reasons, open competition cannot resolve the post-Enlightenment liberal dilemma. The solution must be foundation-neutral.

A Neopragmatist Path to Foundation Neutrality

Paraphrasing the Italian philosopher, Giovanni Pappini, William James wrote:

[Pragmatism] lies in the midst of our theories, like a corridor in a hotel. Innumerable chambers open out of it. In one you may find a man writing an atheistic volume; in the next some one on his knees praying for faith and strength; in a third a chemist investigating a body's properties. In a fourth a system of idealistic metaphysics is being excogitated; in a fifth the impossibility of metaphysics is being shown. But they all own the corridor, and all must pass through it if they want a practicable way of getting into or out of their respective rooms. ³³

For the foundation-neutral model proposed here to offer a solution to the dilemma of pluralism in the post-Enlightenment liberal state, it must provide a space for public political justification that resembles this corridor. To succeed, the model must (1) identify a foundation-neutral vocabulary that is rich enough to engage complex and nuanced issues of constitutional import (such as abortion, race, wealth distribution, freedom of expression, capital punishment), and (2) employ this vocabulary within a foundation-neutral justificatory framework that can satisfy the principle of reciprocity.

Drawing inspiration from Pappini's simile, the solution offered here employs a model of rationality developed within the neopragmatist philosophical tradition. This model is pragmatist in that it builds on insights shared by the classical American pragmatists William James ³⁴ and John Dewey ³⁵ : for example, that expedience ought to be the goal of inquiry; that skeptical questions are usually bad questions; and that we should be dubious of any form of foundationalism. ³⁶ The model is neopragmatist to the extent it follows philosophers like Rorty ³⁷ (influenced by W. V. O. Quine, ³⁸ Wilfrid Sellars, ³⁹ Donald Davidson, ⁴⁰ Ludwig Wittgenstein, ⁴¹ Robert Brandom, ⁴² and others) in relying on vocabularies manifest in social-linguistic practices as the ultimate source of justification.

Default and Challenge Model for Rationality

The neopragmatist model of rationality relied upon here conceives of discourse as having a default and challenge structure. In its political application (our concern), it begins by recognizing that certain of our key first-order moral commitments (for example, that persons are worthy of respect as free and equal) enjoy a default status in political discourse. It then shows how citizens can work from these default commitments toward the authoritative public resolution of complex and controversial social issues without ever appealing to their foundational commitments. Default commitments can be challenged, but their default status shifts the burden of proof to the challenger. Before fleshing out this neopragmatist model of rationality, it will be helpful to first understand how it differs from the orthodox approach, which presumes that recourse to foundational commitments is a necessary condition for justification.

The modern assumption that public political discourse (or any discourse for that matter) presupposes recourse to foundational commitments to ground justification can be traced to the Enlightenment rationalism manifest in Kant's regulism about norms. ⁴³ For Kant, all normativity is a matter of rule-governedness. In other words, all norms of understanding and action are rules that can be made propositionally explicit, step by step, all the way down to their foundations. ⁴⁴ This model presumes that no argument is complete unless the proponent stands ready to offer its ultimate theoretic, moral, or religious foundational support.

But Wittgenstein and others have exposed this rule-governed model of rationality as susceptible to the problem of infinite regress. ⁴⁵ Any explicit rule regarding the correctness or incorrectness of an inference, might itself be applied correctly or incorrectly. This rule will therefore require a further rule to confirm the correctness of its application, and so on. The regress can only be halted by finding some way of licensing the application of a rule (or correctness of an inference) that need not itself be made propositionally explicit. This means there must be a way of obeying a rule that is not a matter of knowing that, but rather a matter of knowing how. For Wittgenstein, such a test is found in the social practice of language use, or, as others put it, the “game of giving and asking for reasons.” ⁴⁶

The neopragmatist model of rationality takes off from this point. For the neopragmatist, concept use is a matter of making moves one can get away with in the social-linguistic game of giving and asking for reasons. The most basic move in this game is that of making an assertion. In asserting a proposition, one puts it forward as a reason one takes oneself to be committed to and also offers it to others as one they are entitled to on the authority of this commitment. ⁴⁷ According to Brandom, keeping track of these various commitments undertaken by oneself and others is a matter of taking “deontic score.” ⁴⁸ Brandom explains that being rational is just a matter of “mastering in practice the evolution of the [deontic] score” while talking and thinking. ⁴⁹

As in any game, there must be consequences for violating one's commitments in the game of giving and asking for reasons. The penalty is loss of credibility. This is precisely what happened to the boy who cried “wolf.” As Brandom explains, the boy asserted and thereby committed himself to the claim that a wolf was present several times without owning entitlement to it. He was punished “by withdrawal of his franchise to have his performances treated as normatively significant.” ⁵⁰ This can be done from the standpoint of the first- or third-person, and indeed we can understand the self-imposed sanctions of guilt or self-reproach as internalizations of this social mechanism.

At this point one might wonder whether conceiving rationality as participation in the game of giving and asking for reasons does not generate its own regress. Entitlement to an assertion is demonstrated either expressly (that is, by showing entitlement is inherited by another commitment), or by deferral to authority (that is, by showing entitlement is inherited by another's commitment to the same claim). But inheritance can be challenged in both cases. If each challenge must be met by another demonstration of entitlement, then a regress is initiated. This is not, however, a problem for the neopragmatist model of rationality as social-linguistic practice. The grounding problem arises for foundationalism because its rationalism (manifest in its regulism about norms) assumes that entitlement to an assertion cannot be attributed until it is demonstrated. But, as Brandom explains, once rationality is understood in terms of social-linguistic practice, we see that, within the practice, many claims are treated as “innocent until proven guilty—taken to be entitled commitments until and unless someone is in a position to raise a legitimate question about them.” ⁵¹ Since such default commitments do not need to demonstrate entitlement until credibly challenged, the threat of global regress that is ever-present for the foundationalist dissolves for the neopragmatist.

Thus, under the neopragmatist account of rationality offered here, the question of which claims will require a demonstration of entitlement and which will not is itself ultimately a matter of social practice. Indeed, the meaning of a particular sentence is determined in part by whether a demonstration of its entitlement is required—this is a crucial aspect of its inferential role. Claims like “There have been black dogs” are treated as prima facia entitled. This means simply that though they are not immune from challenge, the challenge itself will be obliged to demonstrate its own entitlement. Brandom refers to this as the “default and challenge” structure of entitlement and points out that nothing could be recognizable as a game of giving and asking for reasons without it. ⁵² These claims to which we are entitled by default might also be thought of as “free moves” in the game of giving and asking for reasons. Such free moves will play a key explanatory role in the account of legitimation and justification in law and politics that follows.

To take stock: The liberal dilemma motivating this project starts with the inability to reconcile the liberal principle of reciprocity with the pluralism of comprehensive worldviews in Western liberal democracies. In the post-Enlightenment West, the foundation-neutral model appears to offer the only way to respect individual comprehensive worldviews while conforming to the principle of reciprocity. But there remains a concern that the foundation-neutral model will be impotent in resolving disputes over complex constitutional questions. This is because it forecloses recourse to indispensable, though controversial, philosophical and religious foundational commitments that ground the competing answers to these questions. In sum, the concern is that appealing to foundational commitments to resolve constitutional dispute is precluded by the liberal principle of reciprocity, but it is required by our commitment to rationality. By now, however, it should be clear that this is a false dilemma born of the Kantian model of rationality that was challenged in this section.

The default and challenge model of rationality proposed here allows us to conceive of a constructive discourse that does not demand a demonstration of entitlement to every assertion all the way down to foundational commitments, at least not to those assertions that are regarded as free moves from within the practice. Where beliefs are regarded as free moves in a community (for example, “slavery is wrong”), they are presumed innocent unless and until some challenge to those commitments itself earns entitlement. But the burden of demonstrating entitlement to a commitment in tension with a free move will be on the challenger. Thus, as long as the vocabulary of free moves in the public political cultures of Western liberal democracies is rich enough to sustain discourse and achieve resolution on difficult constitutional questions, the requirement of demonstration by appeal to foundational commitments need never arise in the public political forum. If foundational commitments can be avoided in public political discourse without loss, then this neopragmatist conception of rationality will clear the path to success for the foundation-neutral model.

Default and Challenge Account of the Foundation-Neutral Model: Constructing a Constitutional Conception of Justice

The first step in articulating a foundation-neutral model that employs the default and challenge structure is to identify the vocabulary of free moves available in the public political culture. There is nothing technical about the idea of a free move in political discourse. It is any claim one can reasonably expect one's audience will accept without argument. For example, the claim “We all agree that religious difference should be tolerated” could serve as the opening line of any American politician's speech (regardless of her party) without risk of losing her audience. In building off of such opening claims, the politician may go on to draw controversial conclusions, but so long as the narrative she weaves flows coherently from free moves (and does not undermine their neutrality by enlisting controversial foundational commitments in support of them), the audience should remain engaged, even if not ultimately in agreement.

Any challenges to a narrative that proceeds in this way will be to the inferences drawn from the free moves, not to the free moves themselves. Such disagreements are to be expected in a democratic society. But so long as free moves, and not citizens’ foundational commitments, are providing the socially recognized rational constraint on the discourse (for example, “Explain to me how you move from ‘Religious difference should be tolerated’ to ‘Sikh children should be permitted to wear a ceremonial dagger at a public school’”), ⁵³ the discourse continues in a spirit of cooperation and mutual respect.

Of course what counts as a free move will be audience specific. So offering a plausible foundation-neutral model for the basic constitutional structure of a society will require identifying a vocabulary of free moves for everyone in that society—and identifying them in such a way that the relevant society will regard them as shared, that is, as independent of any controversial foundational commitments that may anchor them in the comprehensive worldviews of individual citizens. It will therefore be helpful to sketch out a method that can be applied to identify such a shared source of free moves for so broad an audience.

One promising approach for identifying free moves as shared is to observe the political practices of a given society as the cultural anthropologist would, from the third-person perspective. Any citizen can engage in this exercise by identifying persistent themes in social media, the books she and her fellow citizens read, the movies they watch, their heroes and villains, their shared history, their laws, traditions of interpreting those laws, and the institutions they establish. The hope is that consistent first-order moral propositions will emerge from these practices as free moves. ⁵⁴ If such free moves are identified from the third-person perspective, they can be presented in public political discourse as swinging free from the controversial foundational commitments of any of the citizens of that society. For, by maintaining the third-person perspective, the method assumes nothing in advance about individual citizens’ comprehensive beliefs and values. If there is sufficient homogeneity in the relevant society, one should be able to identify a vocabulary of free moves that is rich enough to serve as the starting point for the construction of foundation-neutral rules and principles that can provide the shared basis for a constitutional order.

Free moves teased from a public political culture by taking on the third-person perspective of the cultural anthropologist are neutral in two respects. First, they are neutral because one who asserts such free moves will not be met with a demand for entitlement. But taking on the third-person perspective allows us to see the second respect in which these free moves are neutral. Since they are culled from cultural behavior, and not by looking inside the heads of any one or all of the individuals comprised by the culture, they swing free of any individual self-conception or comprehensive worldview. It is in this respect that they are truly foundation-neutral; their source (the public political culture) is, by definition, shared. And a society adopting the foundation-neutral model will find solidarity in the recognition that these free moves are identified from a shared source.

Once a critical mass of free moves has been identified, the next step is to draw on these free moves to construct candidate rules and principles that can begin to shape a constitutional order. This step is also descriptive. Returning to the analogy of the cultural anthropologist, one tries to tease out candidate norms for the society based on their predictable moves in the game of giving and asking for reasons. The only constraint on this constructive exercise is that of coherence with the vocabulary of free moves. If coherence is maintained, then the foundation-neutral quality of the free moves will be transferred to any rules or principles so constructed. At this stage, a citizen may construct a complete conception of justice or just rules and principles necessary to address a particular issue. But there is a problem. A vast (if not infinite) number of candidate rules and principles will cohere with even a rich set of free moves. So how does one select from among them? This leads us to the final phase of justification under the foundation-neutral model.

From Is to Ought

The successful construction of foundation-neutral rules, principles, or conceptions of justice, does not guarantee their normativity. ⁵⁵ The process of construction outlined thus far has been descriptive (all accomplished from the third-person perspective), but justification, and therefore normativity, is crucially ad hominem. In other words, a principle (or conception) of justice is not normative for citizens unless they are committed to it in some sense. A foundation-neutral principle of justice's normativity is therefore not achieved until a citizen avows it from within the wider context of her comprehensive worldview. To use Rawls's term, the citizen will attempt to achieve “reflective equilibrium” between the free moves teased from her public political culture (descriptive), a foundation-neutral principle or rule of justice constructed from these free moves (descriptive), and her private foundational commitments (normative). ⁵⁶ So, the answer to the problem posed above is that it is at this ad hominem phase of justification that one selects from the potentially infinite class of candidate foundation-neutral rules, principles, or conceptions. Once a citizen achieves reflective equilibrium with respect to a foundation-neutral rule, principle, or conception, it will provide her with a platform for public political discourse on questions of constitutional import to which she is morally committed from the standpoint of her foundational commitments, but which does not beg the question against other citizens’ comprehensive worldviews. Citizens’ foundation-neutral principles of justice will usually differ, and there will be disagreement, but since all citizens’ foundation-neutral principles are grounded in, and openly proceed from, free moves teased from the shared culture, their dialogue engages in an express spirit of cooperation and mutual respect. In short, the very effort of formulating and expressing one's ideas in terms of foundation-neutral principles of justice signal's this civic-mindedness and cultivates solidarity.

Each citizen's foundation-neutral rules and principles of justice represent, for that citizen, the ideal constitutional order for her society, at least pertaining to the issues she has addressed. These principles compose what I refer to as her “ideal constitutional conception of justice.” ⁵⁷ I refer to a society's actual constitutional order (that is, the current state of that society's positive constitutional law) as its “actual constitutional conception of justice.” A society's actual constitutional conception is simply the legal result (political, legislative, and judicial) of free and fair competition among that society's individual ideal constitutional conceptions at a given moment in time. It is not fixed. It is constantly changing according to the evolution of discourse among competing ideal conceptions represented in the society. The only restriction on this discourse is that it be limited to competition among constitutional conceptions; appeals to foundational commitments are entirely excluded.

Finally, it remains to explain how an actual constitutional conception of justice can be expected to achieve moral authority and legitimacy under the foundation-neutral model. Even though some (perhaps even most) citizens acknowledge that the actual constitutional conception of justice for their society is non-ideal (that is, it departs in varying degrees from their individual competing ideal constitutional conceptions), they will nevertheless recognize the actual conception to be legitimate and morally authoritative so long as it is justified and articulated in foundation-neutral terms that result from the ongoing free and fair competition among citizens’ different ideal constitutional conceptions. Thus, despite the lack of consensus as to its terms, the actual conception will not only avoid offending any citizen's foundational commitments, it can generate its own normativity by virtue of the fact that it reflects the result (at that moment in time) of free and fair democratic discourse conducted in a spirit of cooperation and on the basis of mutual respect. Citizens will, of course, continue to strive through legitimate political means to change the actual conception to make it better conform to their respective ideal conceptions, but this is the point of democratic discourse. And in a pluralistic society, citizens will recognize that the goal of consensus, or even convergence, on a single complete conception of justice (or even a limited family of conceptions) is unrealistic, and perhaps even undesirable. ⁵⁸

Throwing Away the Ladder of Neopragmatism

The foundation-neutral model outlined thus far draws on the neopragmatist's default and challenge account of rationality to clear the path to a conception of justice that can be articulated and made normative for a society without relying upon the authority of foundational commitments or comprehensive worldviews. It was noted earlier, however, that one of the hallmarks of neopragmatism is its anti-foundationalism. It might therefore be argued that, to the extent the model proposed here turns on a commitment to neopragmatism as a comprehensive worldview, it cannot be described as foundation-neutral. It must instead be recognized as incompatible and competing with the philosophically foundationalist secular and religious worldviews represented in the relevant society. So understood, the model would not resolve the post-Enlightenment liberal dilemma but would instead contribute to the problem by itself violating the liberal principle of reciprocity. Indeed, Robert Talisse makes just this criticism of Rorty's neopragmatic “antifoundationalist liberalism.” ⁵⁹

It is true that a necessary condition for the success of the neopragmatist's anti-foundationalist project is that she be able to offer an account of justification and normativity that swings free of any particular religious or philosophical foundationalist commitments. But success in offering such an account is not a sufficient condition for undermining all foundationalist comprehensive worldviews. A philosophical foundationalist is, for example, free to agree that appeals to foundational commitments are unnecessary for public political justification without being forced to deny that their foundational commitments represent an objective and universal moral order that is true from the standpoint of eternity. So, to the extent the default and challenge method can offer an account of justification and legitimation that avoids ultimate reduction to foundational commitments, there appears no obstacle to separating this method from the anti-foundationalist neopragmatist project and applying it to the more limited and modest foundation-neutral task of resolving the paradox of pluralism in contemporary Western liberal democracies. In other words, now that methods drawn from the neopragmatist's anti-foundationalist project have shown us how we can articulate and make normative a constitutional conception of justice without appeal to foundational commitments, neopragmatism itself is a ladder that can be tossed away. By adopting the foundation-neutral model, neopragmatists, Kantians, Christians, and Muslims can all draw on their own private reasons for recognizing a constitutional conception of justice as normative for them.

Feasibility of the Foundation-Neutral Model

One might object that the foundation-neutral model presented thus far simply assumes (without warrant) that, despite the increasing pluralism in our society, we can expect the public political culture to support a sufficient number of free moves to construct foundation-neutral rules and principles rich enough to establish a constitutional order. There are two points to be made in addressing this objection. First, the free moves required to construct a constitutional conception of justice under this foundation-neutral model need not be so numerous or complex as to force, on pain of inconsistency, agreement on a single conception (or even a limited family of conceptions) of justice that must be shared by all members of that society. For, again, unanimity or broad consensus on substantive rules and principles is not a goal under this model. Second, by embracing the foundation-neutral model as a solution to the liberal dilemma, one is guided by the liberal principle of reciprocity and by a spirit of cooperation and mutual respect. The foundation-neutral model presumes this spirit of cooperation and respect will motivate citizens to work together to cultivate solidarity and increase the number of free moves (and shared sources for their identification and articulation) in their public culture. In other words, they will be motivated by mutual respect and a spirit of cooperation to increase their shared public political space. To this end, citizens are expected to employ art, literature, music, film, social media, public and private education, and other means of enculturation to reinforce and supplement shared values and the cultural sources for their articulation. In fact, it is expected that exercises in reinforcing and expanding the stock of free moves and their shared historical sources will develop an increasing sense of national identity, shared purpose, and pride among citizens, which will only reinforce these values. ⁶⁰

Alternatively, one might object that since the foundation-neutral model advocated here builds on free moves that are latent in the public political culture of a given society, it will only have the power to articulate and reinforce commitments already reflected in that society; consequently, it will have no power to reform. First, this concern ignores the fact that the actual constitutional conception is expected to change over time as the democratic contest among competing individual ideal constitutional conceptions persists. Each citizen is reform minded in advocating for changes to the actual constitutional conception to better conform to her ideal constitutional conception, which is itself constantly evolving pursuant to the default and challenge structure of the game of giving and asking for reasons. Second, this objection ignores the role the actual constitutional conception will play as a paradigm for the public political culture. As one articulation of normative commitments implicit in the public political culture, the actual constitutional conception will function in a reciprocal relationship to that culture. As a paradigm, the actual constitutional conception will focus, refocus, and reshape the public political culture in addition to being shaped by it. It will function as a premise as well as a conclusion in public political debate.

Another potential criticism of this foundation-neutral model is that it simply assumes the normativity of the liberal principle of reciprocity and is therefore left impotent to address the non-liberal's objection. ⁶¹ But recall that the model adopts a default and challenge structure. The family of ideal constitutional conceptions that gives rise to the actual conception all share the same starting point: the free moves in the relevant public political culture. As free moves, they enjoy a default status that can only be undermined by a challenge that can itself demonstrate its entitlement. This means that for the non-liberal's challenge to succeed, it must meet its burden of persuasion to the liberal audience. The non-liberal challenger must give the liberal good reason for questioning her commitment to the idea that, for example, persons are worthy of respect as free and equal. Until this burden is satisfied, the non-liberal challenge can be justifiably rejected by the liberal citizen and, if necessary, responded to with force by the liberal state.

It is important to emphasize however, that while the liberal may with warrant reject the non-liberal challenge under the foundation-neutral model presented here, she will not ignore the non-liberal. Again, one's embrace of the foundation-neutral model is motivated by the liberal principle of reciprocity and a spirit of cooperation and mutual respect. As noted above, this same commitment should inspire liberals to test the limits of their imaginations to find new and creative ways of increasing the number of liberal free moves in their society and the shared sources for their identification and articulation. In doing so, they will draw on their shared history and traditions to identify new heroes and weave new narratives designed to engage and inspire non-liberals towards a liberal gestalt shift. And note that because this liberal advocacy will draw on cultural paradigms rather than controversial foundational commitments, it will avoid the open competition model's problems identified above and will therefore stand a greater chance at success. In sum, the proponent of the foundation-neutral model will continue to actively engage the non-liberal while at the same time recognize that the latter bears the burden of persuasion in public political discourse.

Distinguishing Rawls and Others

Though the foundation-neutral model defended here adopts a great deal from Rawls's project, there are at least two important differences. The first difference concerns the goal or aim of the theory; the second concerns the path to achieving it. First, Rawls's political liberalism strives to achieve an overlapping consensus of all reasonable citizens around a political conception of justice. ⁶² Such consensus is only achieved when those citizens hold that political conception in “wide and general reflective equilibrium.” ⁶³ For Rawls, wide and general reflective equilibrium is achieved by a citizen when

that citizen has carefully considered alternative conceptions of justice and the force of various arguments for them. More specifically, the citizen has considered the leading conceptions of political justice found in our philosophical tradition (including views critical of the concept of justice itself) and has weighed the force of the different philosophical and other reasons for them. We suppose this citizen's general convictions, first principles, and particular judgments are at last in line. ⁶⁴

This is a tall order. And a common criticism of Rawls is that such consensus over a political conception is just as unlikely as convergence on a comprehensive conception in a pluralistic democracy. ⁶⁵ But in fairness to Rawls, this account of justification for a political conception represents the ideal, one that he admits is unlikely to be achieved in practice. ⁶⁶ In the non-ideal setting, overlapping consensus upon a “family” of political conceptions, “which family includes the most reasonable conception,” is enough for public justification. ⁶⁷ But it is not clear that this concession renders Rawls's justificatory test more realistic. For example, how can we expect to know whether a given family of reasonable political conceptions contains the “most reasonable” among it? Debate over this question would, it seems, be never ending. Moreover, Rawls seems to limit membership in the family of “reasonable” political conceptions to a very narrow class that would likely preclude the US Republican and Libertarian Party platforms, and with them the political conceptions of about half of the voting population of the most prominent liberal constitutional democracy. ⁶⁸

The model offered here, by contrast, does not share Rawls's goal of consensus on a particular constitutional conception of justice (or even on a family of such conceptions); it expects individual citizens to reach reflective equilibrium with respect to their own ideal constitutional conceptions, but there is no expectation or even hope that these millions of ideal conceptions will somehow converge. They will constantly compete with and challenge one another. The actual constitutional conception is the product of this competition at any given moment in time—and it is constantly changing, never staying in place. The lack of convergence on the contents of the actual conception does not undermine its normativity because all citizens recognize that it is the result of free and fair competition among constitutional conceptions constructed from free moves drawn from shared sources in a spirit of cooperation and mutual respect, and that is quite enough. ⁶⁹ Moreover, since citizens are limited by nothing other than coherence with free moves drawn from the public political culture for the articulation of their constitutional conceptions, the foundation-neutral model here leaves ample public political space for the articulation of Republican, libertarian, and many other constitutional conceptions that would be precluded by Rawls's political liberalism.

Nevertheless, Rawls can be slippery on this issue, ⁷⁰ and some may remain convinced that, despite his language to the contrary, public justification for political liberalism can be interpreted as requiring nothing more stringent than the criteria demanded by the foundation-neutral model articulated here. Even if this were true, however, the pragmatist path taken to the foundation-neutral model offered here would be enough to distinguish it from Rawls's political liberalism. The term “pragmatism” does not appear once in the extensive index to the expanded edition of Political Liberalism. And if one is convinced that Rawls is best understood as a closet pragmatist, the exposition of the foundation-neutral model above shows how his implicit pragmatism might be made explicit, and how applying the default and challenge structure might supplement the tools available to Rawlsians in their explication, exposition, and defense of political liberalism. ⁷¹

It has already been noted that the foundation-neutral model defended here must be distinguished from expressly anti-foundational neopragmatist models, such as that often attributed to Rorty. But this project should also be distinguished from another constitutional model that builds on pragmatist antecedents. Cheryl Misak offers a justification for deliberative democracy that draws on the pragmatist epistemology of Charles Saunders Peirce. ⁷² A central claim of Peircean pragmatism is that truth is the goal of inquiry. ⁷³ One cannot believe something without believing it is true, and to take a belief as true is to regard it as one that “cannot be improved upon” and will “forever meet the challenges of reason, argument, and evidence.” ⁷⁴ In short, to maintain a belief one must constantly subject it to public scrutiny and inquiry. Thus, for Misak, to recognize the truth-seeking function of belief is to admit that “we must, broadly speaking, be democratic inquirers.” ⁷⁵ This, in a nutshell, is Misak's Peircean justification for the democratic state. Robert Talisse then builds on Misak's model by emphasizing the ways in which a properly functioning deliberative democracy must embrace a “civic republicanism” that cultivates the “deliberative virtues” of honesty, modesty, charity, and integrity. ⁷⁶

Despite its pragmatist roots, Misak's and Talisse's model for deliberative democracy is distinguishable from the foundation-neutral model defended here in important ways. First, its adoption of the Peircean goal of truth as the aim of inquiry leads it to welcome foundational commitments into public political discourse; it is not, therefore, foundation-neutral. Second, this Peircean approach is itself expressly justified by appeal to foundational commitments; it purports to rest on an objective inference from the nature of belief to deliberative democracy. Finally, its express reliance on foundational commitments drives the model to embrace a non-liberal conception of the state as playing a crucially “formative” role in the development of deliberative virtues among its citizens. ⁷⁷ Of course Misak and Talisse claim that their theory's foundational and formative commitments are modest and noncontroversial, but are they?

At first blush, Misak's argument from the “is” of truth as the goal of inquiry to the “ought” of deliberative democracy looks like a Habermasian transcendental deduction. ⁷⁸ But Misak and Talisse do not go so far. As Talisse puts it, “[i]nstead of seeking for a transcendental proof of the legitimacy of democracy, Misak remains a naturalist by appealing only to practices of belief and assertion.” ⁷⁹ The goal is to offer a non-metaphysical, but non-circular, “naturalist” justification for democracy that can answer the anti-democratic challenger. But does it succeed?

Even if the anti-democrat concedes that the practice of giving and asking for reasons presupposes a community of discourse, he needn't concede that this community must be so broad as society as a whole, or so narrow as this society. He need not even concede that it must be among real or living interlocutors; fictional heroes or historical figures can do just fine. In other words, Misak and Talisse's modest naturalism cannot get them from the thin practice of giving and asking for reasons to the very thick practice of deliberative democracy. To get from truth-seeking to democracy, they must rely on cultural, historical, and institutional commitments already latent in their societies. ⁸⁰ But if such commitments are ultimately driving their response to the anti-democrat, then they are guilty of the same circularity for which they are wont to criticize Rawls and Rorty. ⁸¹ But while Rawls and Rorty are prepared to own up to the contingency of their culturally-based justifications, Misak and Talisse's circularity exposes their model as a sheep in wolf's clothing—it purports to a foundational objectivity it cannot defend. Under their model, proponents of competing foundational political visions will, without warrant, be forced to feel the sting of marginalization, a sting that will only be amplified by the model's civic republican formative mandate. ⁸² Finally, since one practical effect of Misak and Talisse's approach is that it will admit open competition among foundational commitments in public political discourse, it is vulnerable to the problems of violating reciprocity and stifling discourse identified in the first part of this article.

Adjudication without Foundations: The Problem of Abortion

Up to this point the foundation-neutral model has been developed in the abstract. It remains to fill out its exposition by way of example. The problem of abortion offers a good test for the model because it addresses issues of ultimate importance (such as life and death, personhood, and moral status) that many have argued can only be decided on the basis of private foundational commitments. ⁸³ Moreover, abortion is one of the most divisive contemporary political controversies. So if the foundation-neutral model can clear a space for productive discourse on this issue, this will provide a measure of its prospects for offering a general solution to the post-Enlightenment liberal dilemma.

The principal problem rendering the abortion issue so intractable is that of determining the moral status of the fetus. ⁸⁴ One who thinks that, from the moment of conception, a zygote has the same moral rights as an adult will have a different view of the moral permissibility of abortion than one who is convinced that moral rights only attach after birth. ⁸⁵ The concern is that one simply cannot address such questions of ultimate importance in the public political forum without wearing one's controversial foundational commitments on one's sleeve. Kent Greenawalt puts the problem this way:

[C]ommonly accessible reasons do not settle the status of the fetus at various stages … [O]rdinary forms of reasoning leave open a number of possibilities about how far potentiality should count for the inherent worth of a being already living in some form, and thus leave open the way a fetus should be valued as it progresses. If this conclusion is sound, then people must resolve the status of the fetus on grounds that go beyond commonly accessible reasons. If this is inevitable, the religious believer has a powerful argument that he should be able to rely on his religiously informed bases for judgment if others are relying on other bases of judgment that reach beyond common premises and forms of reasoning. ⁸⁶

Greenawalt is correct in concluding that if there is no alternative to relying on foundational commitments in deciding the abortion issue, then secular foundational commitments should not be privileged over religious beliefs. The challenge for the foundation-neutral model, however, is to answer Greenawalt's claim that shared reasons are inadequate.

Fictional Case Study

The following fictional case study is designed to demonstrate how judges might work within the foundation-neutral model proposed here to decide hard cases concerning issues of ultimate concern. ⁸⁷ As such, it is not intended to convince the reader on the issue of abortion, but rather to show how a controversial decision on this divisive issue can be justified and articulated by judges in terms everyone can accept.

Imagine a pluralistic liberal society that has adopted the foundation-neutral model proposed here and whose actual constitutional conception of justice does not expressly address the constitutionality of laws proscribing abortion. A college freshman has unprotected sex with her boyfriend and becomes pregnant. The student does not want to quit school and raise a child, so she seeks an abortion. Her doctor informs her that abortions are illegal by statute. The relevant statute provides that it is a crime to procure or attempt an abortion except in the case of rape or when necessary to save the life of the mother. Still determined, the student challenges the constitutionality of the anti-abortion statute in court. Article 1 of the state's constitution provides that “Human dignity shall be inviolable.” Article 2 provides that “Everyone shall have the right to life and physical integrity.” And Article 3 provides that “All persons shall be equal before the law.” As this is a case of first impression, there is no prior precedent to bind the judge. She must rely on some other justificatory source. Our judge is Catholic, but she knows that, under the foundation-neutral model, she cannot rely on her religious foundational commitments to decide the case; instead, she must turn to her ideal constitutional conception of justice.

The judge begins by identifying relevant free moves available in the public political culture. She first looks to Articles 1, 2, and 3 of the constitution. She then turns to other free moves in public political discourse, for example, that public law should “properly regulate the institutions through which society reproduces itself over time,” and that it should “secure the full equality of women.” ⁸⁸ Moreover, one expects at least some consensus in the public political culture concerning moral questions directly related to abortion. For example, there is no controversy at the extremes of the spectrum of human reproduction. It is a free move that society does not owe specific duties of protection to possible lives (that is, prior even to conception), and that newborn babies have a moral status equal to adults. ⁸⁹

Having identified a critical mass of relevant free moves, the judge begins constructing an ideal constitutional conception of justice that provides an answer to the abortion question before her. In doing so, the judge must construct a candidate foundation-neutral conception (1) that coheres with this collection of free moves, and (2) that she can then bring into reflective equilibrium with her own foundational commitments (including her foundational commitment to Catholicism) at the ad hominem phase of justification.

The judge begins with the free move that a newborn baby has a moral status equivalent to an adult. From this uncontroversial starting point, the judge proceeds by exploring some generally shared reasons why it is considered to be wrong to kill an innocent adult. Following Don Marquis, the judge concludes that at least part of the harm that is done by killing an innocent adult is that it deprives the victim of the value of all the experiences, activities, projects, and enjoyments that would otherwise have constituted her future. ⁹⁰ Our judge reasons that even if one is not prepared to concede that deprivation of the victim's future is the only reason killing an innocent is wrong, it is uncontroversial to conclude it is a reason. ⁹¹ Moreover, our judge observes that the free move that the law should protect the lives of newborns the same as adults (despite the fact that they lack moral and rational capacities) coheres with the conclusion that at least part of what makes killing wrong is that it would deprive the newborn of a valuable future. ⁹²

But, our judge reasons, if at least one wrong done in killing newborns is that of depriving them of a future, then we must also recognize that the future of a fetus is identical in all pertinent respects to that of the newborn. Indeed, this reasoning leads our judge to the conclusion that, at least from the moment after which the fertilized egg might split to form twins, ⁹³ there is an individual entity with a valuable future like ours, such that (at least by reference to its valuable future) killing it would be a wrong equal in weight to that of killing a newborn or an adult.

But the judge's task is not complete. She has reasoned from free moves to the conclusion that, all things being equal, there are moral grounds for protecting the fetus against those who would deprive it of its life, but she has yet to balance this conclusion against other relevant free moves. For example, what about a woman's right to life? The judge must also consider the Article 3 demand of equality before the law, and, more specifically, the free move that the law should seek full equality for women.

Judith Jarvis Thomson has famously argued that even if one accepts the premise that a fetus enjoys a right to life from the moment of conception, this does not force the conclusion that abortion is wrong in all cases. ⁹⁴ For Thomson, the woman's competing right to life and ownership in her body will trump the fetus's right to life where the woman's health would be at risk in carrying the pregnancy to term. ⁹⁵ In the case of rape, Thomson argues, since the woman played no voluntary role in becoming pregnant, the most that can be said is that she stands in the position of one who comes upon a stranger in need and is in a unique position to aid. ⁹⁶ Since our laws do not currently criminalize the failure to aid in such situations, it would be unfair and arbitrary to demand this of pregnant women alone. ⁹⁷ Donald Regan develops a similar argument expressly in terms of equal protection. ⁹⁸ Both Thomson and Regan go further and argue that the moral relationship of all pregnant women to the fetus (not just those who have been raped) is simply the relationship of stranger to one in need of aid. ⁹⁹ They conclude it would be unfair and violate our commitment to equal protection to demand that the pregnant woman provide such aid (at great cost), while making no similar demands on others. ¹⁰⁰

Assume that, similar to the United States, there is no general duty to aid in this fictional jurisdiction. The judge follows Thomson and Regan in concluding that due respect for the woman's right to life (under Article 1), her body (under Article 2), and the commitment to equal protection (under Article 3) requires recognizing a proviso permitting abortion where the woman's health is in danger or in the case of rape. Having gone this far, however, our judge rejects the claim that equal protection dictates a broader proviso because she is convinced that the woman who becomes pregnant through voluntary intercourse thereby stands in a special relationship to the fetus that justifies an obligation to aid. ¹⁰¹ The judge therefore holds that the state's abortion statute is constitutional and offers the justification outlined above in her published decision.

Notice that in reaching this decision our judge proceeds exclusively from free moves in the public political culture (such as respect for human dignity and physical integrity, the moral impermissibility of killing a newborn, equal protection under law). She then organizes these free moves into a coherent foundation-neutral position on abortion by drawing inferences from these free moves that are reasonably based on common sense and accepted science. By proceeding in this way, the judge does not enlist her foundational commitments in her decision. The decision itself is therefore foundation-neutral.

To say that the constitutional conception reflected in the judge's decision is foundation-neutral is not, however, to say that the judge's foundational commitments did not play a role in the result. They did play a role, but it was a private role. As noted above, though constructed as foundation-neutral, an individual citizen's ideal constitutional conception is not completely justified for her—it does not become normative for her—unless it can be avowed from the standpoint of her foundational commitments. So, for example, our judge may have rejected many candidate foundation-neutral positions on the question of abortion because they could not be reconciled with her Catholic faith. Of course this works both ways. To the extent that the free moves in the public political culture cannot be arranged to support certain political consequences of one's foundational commitments, one may be forced to revise those commitments to comply with the principle of reciprocity. ¹⁰² There will always be some tension between a citizen's foundational commitments and the demands of reciprocity, but the value of reciprocity in liberal societies will almost always outweigh the urge to impose one's foundational commitments on others. ¹⁰³ In any event, the private role our judge's foundational commitments play in justifying her ideal constitutional conception as normative for her does not impugn the public neutrality of her official decision.

The parties before the court may not be satisfied. The decision may not reflect their individual ideal constitutional conceptions. But, to satisfy the demand of the foundation-neutral model, the judge does not need to convince the parties before her or the public at large. The idea is that the very effort of foundation-neutral justification will legitimize even a controversial statute or judicial decision by showing it was adopted in a spirit of cooperation and mutual respect.

Conclusion

I have suggested that any solution to the post-Enlightenment liberal dilemma must open a space for public political justification that resembles Giovanni Pappini's hotel corridor. Citizens with vastly different religious and philosophical worldviews must be able to share this space, and pass through it, with equal pride of ownership. I have appropriated the neopragmatist default and challenge structure for belief and justification to show how such a corridor might be constructed, occupied, and maintained. It is constructed of free moves in public political discourse. It is occupied by a multitude of ideal constitutional conceptions of justice, and one actual conception. And it is maintained by a spirit of cooperation and mutual respect that has been a hallmark of liberalism throughout its history.