Printer Friendly

Religion and social coherentism.

Today, prominent academics are questioning the very possibility of a theory of free exercise or non-establishment. They argue that judgments in the area can only be conclusory or irrational. In contrast to such skeptics, this Essay argues that decisionmaking on questions of religious freedom can be morally justified. Two arguments constitute the Essay. Part I begins by acknowledging that skepticism has power. The skeptics rightly identify some inevitable indeterminacy, but they mistakenly argue that it necessarily signals decisionmaking that is irrational or unjustified. Their critique is especially striking because the skeptics ' prudential way of working on concrete problems actually shares much with the methods of others. Part II then argues that the best defense of religious freedom jurisprudence begins with an approach known as coherentism. In political philosophy, coherentism refers to the way legal actors compare new problems to existing principles and paradigms in order to identify solutions that are justified. The Essay then extracts and emphasizes the social aspects of this basic account. It contends that arguments about the meaning of the Constitution appropriately reflect social and political dynamics. The resulting approach, social coherentism, describes a powerful method for generating interpretations of the first Amendment that are justified, not conclusory. This matters at a moment when some defenders of religious traditionalism are suggesting that principled decisionmaking on questions of religious freedom is impossible, and therefore that such issues should be largely surrendered to political processes.


Writing on religious freedom often begins with an observation that the field is in crisis. (1) Scholars regularly say that the doctrine is deeply confused and has not been successfully rationalized by anyone. Even some judges have critiqued the law of religious freedom as unavoidably messy, (2) or they have predicted that systematization will continue to elude the courts. (3)

One response to these assessments has become influential. Skeptics are arguing not only that the existing law is confused, but that decisionmaking in the area is necessarily rudderless. (4) They believe that deep and persistent conditions of western thought doom religious freedom jurisprudence to contradiction, and that no defensible theory of religious freedom is possible. (5) Their contention that contemporary implementation of the Religion Clauses is unavoidably incoherent represents a significant obstacle to religious freedom jurisprudence today. A response is needed, and this Essay begins work on the most promising candidate.

Part I acknowledges that the skeptics' critique has substantial force. It focuses on an example, namely the skeptical attack on pluralist theories of religious freedom. Pluralists argue that the religion provisions of the First Amendment incorporate multiple values, such as individual autonomy to practice religion without government control, equality among religions and between religion and nonbelief, the separation of church and state properly interpreted, government agnosticism on questions of religious truth, and so forth. (6) Unlike some others, pluralists argue that these values cannot be reduced to any single principle or rubric. (7) Pluralism represents probably the leading perspective on religious freedom today--or certainly a common one. (8) I have applied a polyvalent method in previous work. (9)

Skeptics argue that pluralism cannot avoid significant indeterminacy of outcomes. Moreover and relatedly, they argue that it cannot resolve disputes without resorting to conclusory argumentation or ipse dixit-ism. And therefore they believe that decisions taken by pluralists will necessarily be unwarranted. The argument has power because the complexity of decisionmaking in this area does seem to allow for a range of conclusions in many cases. Of course, this is a problem for much of constitutional law, and for law generally, but it is especially salient in religious freedom law at the moment. And although the skeptics levy their critique against all forms of religious freedom theory, it has particular force as applied to the pluralists, both because those theorists admit greater complexity than others, and because they represent probably the most prevalent approach.

It is striking, and not usually noticed, that pluralists and skeptics actually employ similar ways of working on ground-level cases: both carefully assess the reasons on either side of an issue before making a judgment, both take into account a wide range of considerations, and both avoid deducing outcomes from singular formulas. (10) And this Essay envisions a method that is not entirely dissimilar. What distinguishes skeptics is their critical evaluation of this process. They believe that it cannot generate legal principles, and that it can only result in haphazard solutions.

So far, pluralists have not formulated an entirely satisfying answer to the skeptics' critique. Kent Greenawalt, who is among the leading pluralists and a primary target of skeptics like Steven Smith, has said only that in hard cases reasons run out. (11) In those cases, Greenawalt chooses a particular outcome because the reasons behind it simply "seem" stronger than reasons supporting the alternatives. (12) Part II aims to supplement that response. It argues for an approach to religious freedom that is capable of generating conclusions that are rationally justified, even in hard cases. To defend against the skeptical attack, no more is required--it is not necessary to go further and show that determinate outcomes can be identified in all cases. (13)

The best defense of religious freedom begins with coherentism. Inspired by moral philosophy, where it is common if not mainstream, coherentism offers an account of how people properly resolve new problems by comparing them to existing cases and principles. (14) When constitutional actors encounter a new question or situation, they ask whether a given result fits together with their considered convictions, both about correct resolutions of more familiar cases and about abstract tenets. Working back and forth among these elements, they seek a solution that achieves coherence. If they succeed, then the solution qualifies as warranted.

This Essay articulates a version of reflective equilibrium or coherentism that offers an attractive alternative to skepticism. Coherentism also rationalizes and justifies what many theorists are already doing in practice. Moreover, it has a recognized application to law--it shows how constitutional determinations, in particular, can count as morally warranted. It has conceptual bite because (and insofar as) it defends against the skeptical claim that conclusions on questions of religious freedom law will be necessarily conclusory or ad hoc. Though the Essay offers only a method for thinking about these disputes, and not a substantive theory of religious freedom, it needs to do nothing more to accomplish its purpose, which, again, is to show that thoroughgoing skepticism regarding religious freedom is wrongheaded. (15)

Underemphasized in the philosophical literature on coherentism, however, is a full appreciation for social and political dynamics. Citizens reason through problems in a situated way. Paradigms and precepts, to which actors compare the conflict before them, are made salient by cultural developments, which also limit the range of appropriate outcomes. Especially for constitutional law, these influences are salutary because they work to vindicate a commitment to democratic responsiveness. They also help to limit the problem of indeterminacy. So while a social aspect may not be appropriate for all fields of inquiry--scientific disciplines, for example--it should be included in constitutional methods.

Social coherentism does not eliminate unpredictability of outcomes, however. Nor does it offer a moral ontology--an account of what is real or true. Nor does it count as a theory of moral epistemology--of how we know what is good or true. (16) It has a different and more modest aspiration than any of these, namely to defend a common way of solving legal problems. It argues that complex decisionmaking on questions of legal interpretation can count as justified, where "justified" means backed by reasons, rather than conclusory or ad hoc. A moral conclusion that is justified or warranted (17) has persuasive power because of the reasoning behind it. Sometimes arguments must be confronted, even though power politics and private interests are also at play, simply because they carry force. Convincing each other, not just contestation, becomes necessary.

And social coherentism does generate criteria for critiquing some judgments. If a legal conclusion cannot be harmonized with principles or precedents that are authoritative within constitutional practice, then it fails to satisfy social coherentism and it is unwarranted. Below, Burwell v. Hobby Lobby Stores, Inc. (18) serves as an example of a decision that was discontinuous with the Court's precedent and therefore could not count as morally or legally justified. (19)

Of course, people who use this approach will still disagree about whether a particular outcome is defensible or not. This is a feature, rather than a defect, of the approach. Defenders of Hobby Lobby have arguments that they believe make the Court's conclusion consistent with precedent and principle. Conclusions can be conflicting but still justifiable, meaning backed by reasons. And people can disagree about whether a result is even warranted. Social coherentism cannot and does not resolve those conflicts--only ongoing efforts to persuade each other can do that. Its aim is simply to show that defensible arguments can be made, and to provide an understanding of their criteria and conditions. (20)

Skeptics might still object that social coherentism cannot count as a theory because it cannot eliminate irrationality. When someone chooses outcome X over Y in a close case, she cannot give a reason for that choice itself, even if she can give reasons that support both X and Y. Part II concludes by offering three responses. First, we should distinguish between two meanings of irrationality (at the very least). While no approach can escape the first, which demands determinate outcomes for every case, social coherentism successfully avoids the other meaning, which asks for justified results. (21) Second, the Essay argues that instances of arguable irrationality are rare--several of the skeptics' own examples do not actually present such situations, and even their core illustrations do not involve unreason. (22)

Finally, Part II contends that we should focus on conceptual yield and ask what is really at stake in the skeptics' claims of irrationality. Two fears turn out to underlie those charges: bias and arbitrariness. But social coherentism is not disproportionately exposed to either of those dangers. In fact, coherentism understood from a social perspective is less vulnerable to those risks than it is without that perspective. Reflecting on how our beliefs fit together works to combat bias and arbitrariness. Moreover, the historical and political contestation that shapes our reflection limits both risks.

In sum, Part I of this Essay shows that the skeptics have a point: leading theorists of religious freedom can do more to defend their recommendations as not conclusory or unwarranted. Part II then argues that social coherentism provides an account of how constitutional decisions can justifiably be made under such conditions.

The Conclusion argues that this account is needed today, when skeptics are arguing that irrationality surrounding the Religion Clauses provides a reason to leave more religious freedom disputes to political processes, including in local jurisdictions. (23) If they succeed in that argument, minorities could be exposed to majoritarian forces in regions of the country that are dominated by religious traditionalists. And that could happen without resistance from constitutional law, whether in courts or outside them. (24) Although social coherentism sometimes generates results that do not match particular political impulses--for example, it provides some support for the notion that religion sometimes does and should have a special place in constitutional law (25)--it offers a needed account of how citizens can and should argue about religious freedom.

I. The Skeptics' Critique

This Part acknowledges that the skeptics have identified a genuine problem with a leading theory of religious freedom, pluralism. Although they argue more broadly that all theories necessarily will fail, their attack is especially significant when directed against pluralism, which is common among scholars across the political spectrum. Section A describes pluralism by focusing on Kent Greenawalt, who is one of its most prominent advocates. A focus on Greenawalt makes sense not just because of his stature, but because his work has been a primary target for skeptics.

Pluralism has drawn the critique that it underdetermines outcomes, as Section B explains. Once the relevant concerns have been identified, there is no obvious way to determine which should govern in a given case, especially when it conflicts with others. Decisionmaking thus can seem patternless and even biased. Personal preferences and social interests are thought to have worrisome latitude to influence results where legal authorities do not dictate discrete outcomes. As noted above, Greenawalt has been critiqued along these lines by Steven Smith, the influential and impressive skeptic. (26) Smith admires the way that Greenawalt carefully analyzes the arguments on either side of a difficult question, but he thinks that ultimately Greenawalt simply chooses the side of the argument that seems strongest to him. (27) Smith describes Greenawalt's approach as conclusory, ad "hoc, or an exercise in ipse dixit- ism. (28)

Section C argues that this critique of pluralism has power. Greenawalt himself has not provided the strongest possible response. Without a convincing account, the process of identifying and applying the values involved in a constitutional decision can seem incomplete or rudderless. That sets up Part II, which aims to provide a more compelling account of how the jurisprudence properly works in practice.

A. Pluralism

Pluralists recognize that religious freedom doctrine cannot be explained or guided by reference to a single value, and consequently they incorporate multiple commitments. (29) Among those may be: evenhandedness among faiths and between belief and nonbelief, liberty of individual belief and observance from government control, separation between religion and government, latitude for religious institutions to conduct internal affairs, preventing religion and government from corrupting one another, and avoiding political division along religious lines. (30) Not all of these values are implicated in every case, nor does every dispute necessarily involve more than one, but neither can any single principle generate results that are attractive and workable across the full range of religious freedom questions. Managing competing considerations is as necessary as it is difficult for them. (31)

As an example of pluralism, consider how Greenawalt addresses the question of whether government may isolate religious actors for denials of support without offending the Constitution--the issue of excluding religion. (32) Sometimes courts consider such denials to be harmless non-subsidies, (33) but other times they find them to involve impermissible discrimination on the basis of religion. (34) Characteristically, Greenawalt does not adopt any single approach to the entire category, and even within subcategories he offers recommendations that are nuanced. Consider his treatment of just one issue: equal access by private religious groups to public educational facilities.

Greenawalt endorses most of the equal access cases, which prohibit the government from barring religious speakers from public facilities once it opens them up for private speech. (35) But he stops short of fully embracing Rosenberger v. Rector & Visitors of the University of Virginia, a decision that required the University of Virginia to fund an evangelical newspaper in the same way as other student publications. (36) Greenawalt questions the Court's decision in that case because he believes that cash aid for core religious advocacy, which was at issue in Rosenberger, ought to be distinguished from equal access to state facilities, which was the subject of previous cases. Even though cash aid is a form of government support like access to public facilities, it presents a more serious danger to non-establishment because of its fungibility. Had the Court appreciated that difference, it would have permitted the university to defund the religious publication--and it might even have required it to do so under the Establishment Clause. (37)

Greenawalt also criticizes Good News Club v. Milford Central School, where the Supreme Court required a school to include a Christian club when it opened up the school building after hours to other groups that promoted "moral and character development." (38) For him, three differences separate that situation from other equal access cases: the case concerned elementary students, who are particularly impressionable; it involved evangelization by an outside group; and it protected meetings that began directly after the school day. (39) Under those circumstances, school administrators should have had discretion to conclude that students would feel pressured to attend.

Whether his conclusions are correct is not the issue here. Rather, these examples serve to show how pluralism works in practice. In Greenawalt's discussions, much depends on the particular circumstances, because different facts can trigger distinctive constitutional values, and in divergent ways. In fact, his sensitivity to complexity extends well beyond the ideals that he outlines in his theoretical introduction to the Establishment Clause. (40) In practice, it includes many non-ideal factors such as pragmatic and political considerations. Greenawalt also takes an eclectic approach to methods of constitutional interpretation without addressing the subject extensively. (41)

Given this complexity, it is not surprising that the skeptics have critiqued the approach as not just eclectic but erratic, as detailed in the next Section. What is genuinely striking is that skeptics and pluralists end up applying a similar method in concrete cases. (42) What separates skeptics from pluralists is their normative assessment of that method.

B. The Skeptics' Critique

Skeptics believe that religious freedom doctrine is afflicted by an incoherence that is unavoidable. Problems endemic to the area doom all efforts to rationalize the jurisprudence. Constitutional decision makers can only muddle through, seeking modus vivendi solutions to particular problems. Necessarily, these decisions will be indefensible on the level of reasoned principle.

Interestingly, skeptics can be found among political progressives as well as among political conservatives. Some skeptics envision an expanded role for religion in public life and robust protection for observance, (43) while others support a muscular interpretation of non-establishment and robust protection for religious minorities and nonbelievers. (44) What unites them is a conviction that no theory of religious freedom is possible. (45)

Steven Smith has offered a sustained and significant articulation of the skeptical position. Smith believes that the search for a coherent theory of religious freedom jurisprudence is hopeless. By "theory," he means "an internally consistent set of principles capable of generating answers to questions of religious freedom." (46) Ad hoc argumentation does not count, nor does reasoning that generates only workable compromises. To say that there cannot be a theory of religious freedom is also to say that there can be no principled approach, for Smith. (47) The two phrases appear to mean the same thing.

Smith offers many reasons for this inevitability. One is historical. On that account, the western intellectual tradition includes two political moralities that are fundamentally at odds with one another. (48) An older way of thinking is theological and jurisdictional--it divides authority into sacred and profane realms on religious grounds. A newer conceptualization is secular--it foregrounds aspirations of justice or fairness toward citizens. (49) Because both moralities are represented in American constitutionalism, and because they are incompatible, attempts to interpret and apply the law to concrete disputes cannot achieve consistency. For example, religion receives special constitutional treatment, following a belief that is widespread and traditional, but the theological justifications for that distinctiveness have been eschewed and cannot be replaced by a secular conception of justice. (50)

A related difficulty is that the principles offered by modern theories to anchor the doctrine are not up to the task. Equality among religions and neutrality of government toward them are the principles most often invoked, but Smith believes they are empty concepts. (51) As an example, he criticizes Eisgruber and Sager's principle of equal liberty. (52) Their theory "smuggles in" criteria by which beliefs ought to be compared to one another--it holds that we should equalize constitutional treatment of all commitments that are "deep" and "valuable" without specifying what would satisfy each of those criteria. (53) Smith is not surprised that the results then conform to Eisgruber's and Sager's substantive commitments, which remain unarticulated. That disqualifies equal liberty as a theory, for him.

Pluralism of values is no solution, for Smith. Although theorists who advocate pluralism admit to complexity and elevate it to the level of a theoretical commitment, they still purport to resolve disputes using a reasoned process of interpretation and application of principles. Greenawalt, for instance, ultimately chooses the result that seems best to him, without justifying the final decision. Smith selects Greenawalt's discussion of Good News Club as an example, saying that it weighs various values on each side before simply declaring which position is most convincing. (54) That method uses the language of principle to describe decisionmaking that must be driven by something else, according to Smith. (55)

So how should constitutional actors make decisions? Smith proposes a prudential approach. Some of what makes it distinctive is rhetorical--he believes that dropping the pretense of principle will lower the stakes of cultural conflict. (56) Another part goes to the institutional differentiation between law and politics, because many conflicts would be resolved by political branches of government under his approach, rather than as a function of constitutional law administered by courts, and that would allow both for greater public say and for local variation. Overall, Smith urges constitutional actors to confront the sort of decisionmaking that they are already practicing, under which disparate factors are taken into account in a way that is intuitive, ill-defined, inclusive, and implicit. (57)

Yet another part of what distinguishes the skeptics' prudentialism comes from the substantive values that it ultimately endorses, even though they cannot be required by the approach itself. First, Smith advocates the return of tolerance. (58) Of course, the broad idea is that government should permit activities of which it disapproves, but that disapproval can be vigorous and can sometimes be accompanied by regulation, as long as it does not amount to intolerance. (59)

Second and related, he allows open government recognition of the view that America is not only deeply religious but also a "Christian nation." (60) Nonendorsement of religion by government therefore is a mistaken idea, and expressions like the national motto, "In God We Trust," are at once fully religious and constitutional. (61) Do these commitments follow from the prudential approach itself? Smith admits, and even insists, that they do not--that is part of what it means for the approach to be pragmatic rather than principled. (62) He argues that if people view his recommendations as idiosyncratic, that perception should strengthen his main argument that principled solutions are unavailable, and that instead we should allow "the 'living practices of the American people [to] bespeak our basic constitutional commitment.'" (63)

A notable conclusion from this summary is that skeptics and pluralists adopt similar ways of working in practice, despite their deep differences on matters of theory. Both evaluate the strongest arguments on both sides of an issue before issuing an all-things-considered ruling. Both take into account a wide range of considerations, principled as well as pragmatic. (64) Both think that religion should be counted as special only some of the time, and that the difference will depend on careful analysis that is situational and specific.

Of course there are important differences, too. For one thing, Smith tends to recommend results that favor religion, (65) while Greenawalt generally falls on the secular or progressive side of more constitutional disputes.66 And the methodological disputes are real, including whether variegated decision-making of this type is irrational. How should pluralists react to that charge?

C. Pluralism's Response

So far, pluralists have not offered a satisfying response to the skeptics. Greenawalt himself, considering Smith's attack, has said only that at a certain point, in certain hard cases, reasons may run out. All we can do is "survey the reasons on each side" and make a decision. Greenawalt says, "I do not believe there is some process of reasons that settles which side is stronger." (67) In those hard cases, we can only "rely on our intuitive judgments--judgments influenced by our cultural heritage, particular upbringing, and professional training." (68) In other words, reasons cannot support the ultimate choice in every case, and therefore a theorist must rely on his or her intuition about which set of reasons is more convincing. Greenawalt puts it this way in conclusion: "[W]hat a critic might label 'bald pronouncements,' [quoting Smith] I prefer to regard as my intuitive sense of the stronger side of an argument, one that I admittedly cannot show convincingly to be correct." (69)

This is not quite the same thing as conceding that reasons can never be given for outcomes in religious freedom cases--which is one reading of the skeptical attack--but it does come notably close to conceding that in some hard cases, presumably including the most hotly contested ones, reasons cannot support the ultimate choice among outcomes. So at least for those cases, the skeptical critique goes unanswered. Greenawalt does not explicitly acknowledge that this means his judgments in such cases are unwarranted, but he does not attempt to refute that charge, either. (70)

An intriguing variation on the pluralist response comes from self-styled "tragedians." These thinkers believe both that religious freedom cases can only be decided based on plural values, which can and often do conflict, and that there is no principled way of resolving such conflicts. So they agree with the pluralists about the complexity of considerations that are often in play, and they agree with the skeptics that the result risks irrationality or arbitrariness. As a result, they believe, American jurists are in a situation of "tragedy," meaning that some values will have to be sacrificed in order to vindicate other values in real cases, and no good reason can be given for those choices. What makes the situation tragic, in fact, is precisely that some values must be sacrificed to vindicate others, but without justification from some master commitment. (71)

Is there a better response available to the pluralists? Can a wide range of considerations be brought to bear in ways that count as warranted? Or are the skeptics right that no rational approach to religious freedom is possible under modern conditions?

That question is not unique to religious freedom law--it affects much of constitutional doctrine. Perhaps religious freedom law is experiencing more than its share of internal tensions for reasons that skeptics have identified, including especially the shifting place of religion in American society and the sharp divisions reflected in the so-called "culture wars." But it is also possible that the dynamics around religion are not so different that they cannot be analyzed using available tools of political morality and constitutional theory.


Part I acknowledged that the skeptics' critique of religious freedom jurisprudence has power. In particular, it contended that pluralists face a difficulty in that they ultimately rely only on intuition, not reason, at least in certain hard cases. But the problem is not confined to pluralists. If there is no way to recommend outcomes that is not conclusory or ad hoc, that endangers all efforts to account for the law of religious freedom. So far, theorists have offered no compelling answer.

To depict a plausible and attractive account of how people can and should operate under modern conditions, this Part begins with an element of moral philosophy, variously known as the method of reflective equilibrium or coherentism. (72) At root, that way of thinking holds that moral actors properly seek harmony among their considered convictions regarding concrete disputes and among the principles that they have abstracted from those situations. Solutions to new problems are justified if they cohere with those existing convictions. As this Part will argue, coherentism accurately describes how moral reasoning often works in the real world, including in legal domains. Because it rationalizes what theorists are already doing in practice, it supplements rather than supplants existing substantive theories of religious freedom. (73) Most important, it defends against the skeptical claim that decisionmaking in this area can only be conclusory and therefore unwarranted. (74)

Yet this Part will also argue that coherentism should reflect the insight that moral decisionmaking on questions of religious freedom happens in a contextual way, influenced by developments in politics and the broader society. It calls this method social coherentism. Actors do not randomly call to mind real world situations that generate fixed moral conclusions. Instead, they select moral reference points that are made salient by influences such as social movements, political parties, legislators, executive officers, media outlets, and civil society groups. They arrive at conclusions about actual experiences through conversation and conflict with other interpreters who also are socially located.

After Sections A and B lay out those affirmative arguments, Sections C and D confront objections. In particular, Section C anticipates the response that social coherentism does not do any better than previous "theories" at avoiding patternless thinking, ipse dixit-ism, and unreason. It answers that whether it qualifies as a theory, or as rational, depends on what one demands of a theory or of rationality. In particular, the Section distinguishes between two meanings of irrationality. While no approach can escape the first, which demands determinate outcomes for every case, social coherentism successfully avoids irrationality of the other type, namely the failure of moral justification. Outcomes are rationally justified or warranted if they cohere with our considered convictions.

Section D then shifts the focus to payoffs. What exactly is at stake in the charge of irrationality? At bottom, the worry seems to surround two possible defects in religious freedom doctrine and scholarship: bias and arbitrariness. Focusing on those concerns, the Section concludes that social coherentism manages them relatively well. Nothing about the approach is distinctively vulnerable to either of them, and much about it suggests distinctively strong protection against both.

Finally, Section D argues that the social perspective ameliorates several problems with coherentism, familiar from the philosophy literature, including the difficulty that the approach is too individualized or relativistic for use in legal practice. Commonalities of culture and history provide a base for these conversations, so that legal actors are intelligible to one another when they argue over the meaning of a particular constitution in a given society at a particular moment in time.

A. Introduction to Coherentism.

Coherentism is a term with a range of possible meanings; here, it describes a strain of philosophy that asks how moral judgments can be rationally justified. Against skeptics who believe that such conclusions must be driven by intuitions or interests that operate outside reason, coherentism provides a way of understanding how even complex moral judgments can be defended as a matter of reasoned morality, not just pragmatics or politics.

Start with moral reasoning in daily life. People try to make sense of their views--that is, they try to harmonize their judgments about existing moral situations with each other and with principles that they abstract from those situations. They try to articulate general principles that can explain their judgments about specific situations, and they try to ensure that those judgments resonate with judgments about analogous situations. They seek coherence, which bolsters all their conclusions.

When people encounter a new moral problem, they consider their initial impulse about the proper outcome. They test it against principles and paradigmatic cases about which they have come to have some confidence, after examination. If the judgment conflicts with those beliefs, it does not count as warranted. But that is not necessarily the end of the matter. Existing conclusions may need to be reexamined in light of new situations or new understandings of existing positions. All elements of moral analysis are open to revision in an ongoing process of reflection and testing. When commitments fit together, then they can claim to be warranted or justified because they are backed by reasons.

Coherentism owes much to John Rawls's method of reflective equilibrium. Rawls famously said that reasoning about justice should begin with particular convictions about which there is a relatively high degree of confidence, after reflection. (75) These considered convictions can exist on any level of generality, from abstract principles to judgments about concrete cases. (76) They are commitments that appear to be correct, after they have been carefully scrutinized. (77)

Actors proceed by working back and forth between principles and considered beliefs about real world situations in an effort to reach a reflective equilibrium. (78) As Rawls put it: "A conception of justice cannot be deduced from self-evident premises or conditions on principles; instead, its justification is a matter of the mutual support of many considerations, of everything fitting together into one coherent view." (79) Although Rawls was developing a method for thinking about the basic morality of a political system--a "conception of justice"--the idea of reflective equilibrium can be applied also to moral reasoning by individual actors about particular problems. If our sense of the best solution to a problem withstands reflection, so that it fits together with the rest of our considered convictions and the principles abstracted from them, then it is warranted.

Although coherentism begins with Rawls, it does not entail his entire theory. This Essay defends only the idea that warranted judgments on questions of religious freedom are possible.

What kind of claim can be made if a state of coherence or equilibrium is reached or approached on a given topic? First, actors can argue with some confidence that their judgments on problems within that topic are reasonable, warranted, or justified. That is, they can argue based on the process of reaching coherence that their conclusions are not just conclusory or irrational, but that they are grounded in reasons. (80) Moreover, identification of principles that have survived examination can give them a reason to conclude that "at least some questions about the subject matter have determinate answers," as Scanlon puts it. (81) Therefore, the method of coherentism or reflective equilibrium provides an answer to charges of irrationality, and to related concerns that moral decisions are driven solely by interests or intuitions. Providing a defensible account of moral reasoning is, in fact, its whole purpose. (82)

With that general account as background, consider eight clarifications and specifications about the form of reflective equilibrium or coherentism that is under construction in this Essay.

First, "fitting together" means not just consistency, but a stronger relationship of mutual support. (83) Joseph Raz uses the example of someone who holds two empirical beliefs: first, that John was seen going into Emily's house, and second, that John has long wanted to visit Emily. Those two beliefs are not just consistent; they are mutually reinforcing. If one is challenged, the other can be cited as support. (84)

Second, the notion of coherence is not static. All commitments are revisable in light of the others--none is foundational, and the system is dynamic. Old commitments can be reexamined in light of new ones.

Third and related, coherentism allows for sharp critiques of prevailing legal arrangements. It is not limited to rationalizing established rules, and it should not be understood to have a conservative bias. (85) Critique can be based on a claim that a particular doctrine is unfaithful to a principle. Criticism can also take the form of a claim that established understandings do not sufficiently account for right outcomes in concrete examples that might have been overlooked or misapprehended. Additional forms of critique are doubtless possible, and they too could lead to a convincing charge of incompleteness or incoherence. So it is a mistake to assume that coherence theories favor existing rules.

Fourth, coherence or reflective equilibrium probably is not best understood as a state that anyone is likely to reach. Rather, it is an ideal that shapes efforts to arrive at conclusions that are reasonably justified. (86) As someone's commitments begin to cohere with each other, that person becomes more confident that they are reasonable or warranted. People are more likely to approach coherent views on a particular subject matter, rather than on the whole panoply of their moral views. That coherence can bolster their conclusions on that subject, even if those conclusions remain local and provisional. Moreover, what Raz calls "the untidiness of politics" is likely to frustrate efforts at global coherence. (87) But again, that does not mean that the methodology itself cannot drive justified results in real cases. (88)

Fifth, principles play an important part in the method. They can help to bring together judgments about concrete cases, both because they can cast those judgments in a new light, and because they can help to resolve new problems. (89) Nothing guarantees that principles can be found to account for particular judgments in a particular area, in which case those judgments should cohere merely as a matter of horizontal analogy. Nor is there any guarantee that there will not be multiple principles that govern a given area--in which case a situation of moral pluralism will exist. (90) But overall, principles work to rationalize discrete commitments.

Sixth, reflective equilibrium has a recognized application to law and legal problems. (91) Laws contain ambiguities that either allow or require legal actors to engage in moral reasoning in order to arrive at the best understandings of those authorities and the best application of them to cases. (92) What coherentism adds is a way of understanding how legal actors can arrive at a justified conclusion about some new problem, without resort to intuitionism or irrationalism. If a solution fits their reflective convictions (about legal principles) and conclusions (about analogous cases) then it is rationally warranted.

Seventh, coherentism in this Essay grounds moral justification only. It does not provide an ontology, or an explanation of moral reality or truth. (93) Nor is it necessarily best understood as an epistemology, meaning an account of the conditions under which it is possible to know what is morally right. (94) While it may be possible to build up coherence theories of epistemology or ontology, nothing here depends on them. An account of moral justification is all that is necessary to defend the view that complex decisionmaking about questions of religious freedom can qualify as warranted. Skepticism would require a deeper response if moral conclusions could only claim to be reasonable based on some ontological or metaphysical basis, but there is no reason to think that is the case. (95) In fact, the method of coherentism is designed to show how moral positions have force simply because of their relationship of mutual support to specific judgments and general principles.

The final point requires further elaboration, because it concerns the question of disagreement. What happens when we reach (or approach) an equilibrium that conflicts with someone else's? That possibility is not an embarrassment for the approach. (96) To the contrary, it is among its key features. T.M. Scanlon helps to show why disagreement is not debilitating in a recent book, where he adopts a version of reflective equilibrium. When we find ourselves in a situation of disagreement, we should ask whether the other person began with different considered judgments, or considered different principles, or resolved conflicts between principles and judgments differently, and we should consider whether we should have made those choices, too. (97) If the answer is yes, then revisions are necessary.

If the answer is no, however, then the fact of disagreement is no reason to change our views. We can try to persuade the other person that our judgments were better or we can accept our differences. (98) Scanlon emphasizes that what matters is not chiefly whether someone has reached coherence, but rather the quality of the reasoning that led to that state. (99) Their thinking process is what gives their conclusions justificatory force. Moreover, Scanlon concludes from this depiction of disagreement that seeking coherence is not just a way of clarifying what we think, (100) but it is a "method for deciding what to think." (101)

Is it possible for conflicting conclusions to convincingly claim to be coherent? Of course that is possible--it characterizes situations of reasonable disagreement. (102) In a close case, some people will think that one side should win, while others believe the other side should. Each side may have reasons for choosing its outcome over the alternative, because each may be able to make a case for how that result resonates with its other commitments. Each side can try to convince the other that it has neglected important principles or paradigms, or each can argue that the other has drawn the wrong conclusions from the common authorities. Moreover, each side can argue not only that the other side has made a mistake in reasoning, but further that its reasoning is so discontinuous with precedents or principles that it is incoherent and therefore unwarranted. Disagreement can happen on any level of argument.

Whether ultimately only one side can be correct--whether there is, in the end, only one right answer to problems of law or political morality--is a question that is beyond the scope of this Essay, and quite far beyond it. What matters here is simply that both sides can come to conclusions that are backed by reasons. Section D will return to this deep problem, anticipating the objection that although reasons can be given for both outcomes, no reason can be given for choosing one outcome over another in hard cases, and therefore the approach ultimately does devolve into irrationality. (103) It will refute that idea, suggesting that any understanding of rationality that has practical yield in this context can be satisfied by a method of reflective equilibrium or coherentism.

As an example of how all this works, consider Greenawalt's discussion of a basic precept, namely that governments cannot take positions on questions of religious truth. That commitment comes in for attack from Smith, as Section D describes more fully. (104) But in fact Greenawalt's argument for its existence provides one of the clearest examples of how seeking coherence can result in warranted conclusions.

How does he establish that tenet? He methodically shows it to follow from other principles that are widely observed, and he argues that it resonates with actual case outcomes that have survived reflection. (105) His first move is to say that if the Establishment Clause prohibits the government from recognizing a particular denomination as the official faith of the state, something that is virtually uncontroverted in American constitutionalism, then it follows that the state may not embrace a theological teaching of that faith as true. Based on that syllogism, he judges the rule against government declaration of religious truth to be perhaps the best example of a principle that is supportable by reasons. (106) It coheres with a basic non-establishment tenet, while fitting with correct case outcomes, and it is therefore is justified.

Greenawalt goes further, however. (107) He also asks whether this same rule (against government endorsement of theological truth) extends even to official actions that are purely expressive and do not carry any coercive consequences. To test the answer, he offers a hypothetical in which a legislature makes an explicit statement denouncing abortion and reproving women who choose to terminate their pregnancies, all on theological grounds. (108) Abortion remains legal. He thinks that statement is designed to "condemn" those women, and he concludes, "legislatures could definitely be engaged in establishing religious ideas even if the laws themselves impose no adverse consequences on dissenters." (109) This exercise then supports the general principle.

Moreover, his discussion of the rule draws on the values he articulates in the book's introduction and throughout--they are mutually reinforcing. Confidence in the principle supports those values and vice versa. Government pronouncements on religious truth conflict with virtually every non-establishment value, including the concern about government incompetence on such matters, the warning about corruption of both religion and state, the danger of harm to equal dignity for adherents of disfavored views, and so forth. (110) Repeatedly throughout his discussions of particular controversies, Greenawalt observes that the prohibition on government endorsement of religious truth is driving at least part of the analysis. (111) Far from worrying that the principle is underdetermined by reasons--the skeptics' concern--he believes it is demanded by them. (112) Greenawalt never mentions reflective equilibrium. But if this approach to supporting the prohibition on government theologizing is not textbook coherentism, then it is at least compatible with that method, and it serves as a nice example of how coherentism can work to justify conclusions about principles and outcomes even in a complex moral environment like the one surrounding the First Amendment.

B. Social Coherentism

Underemphasized in coherentist accounts in moral philosophy, however, is the role of social and political dynamics. Actors do not randomly recall real world situations that generate fixed moral conclusions, but instead they tend to choose reference points that are made salient by social and political processes. Moreover, their reasoning process is influenced by the arguments of others, in the sense that it most readily confronts judgments that are prominent in the culture at that historical moment.

Why emphasize a social dimension? There are at least two reasons. One objective is to anticipate and disarm the argument that because moral reasoning is influenced by politics and interests, it cannot be justified. Social coherentism vigorously resists this conclusion, showing that arguments can be shot through with power dynamics and nevertheless carry the force of reasonableness. People think from within social conflicts, not outside them, and yet their reasons can carry the authority of principle. Another reason to highlight the social dimension is to make the point that the social embeddedness of arguments over constitutional understandings can help bolster those interpretations' popular responsiveness, and hence their democratic legitimacy.

None of this means that constitutional actors must actively poll others before coming to a conclusion. It is not prescriptive in that way. Rather, the social aspect of the argument describes and defends how the method works implicitly.

To continue with the example in the last Section, Greenawalt draws his abortion hypothetical from the Zeitgeist, and his conclusion reflects prevailing mores. He himself senses this influence and he observes that our judgments are "influenced by our cultural heritage, particular upbringing, and professional training." (113) Nothing about this is embarrassing, nor does it necessarily entail bias or intuitionism. Americans fix the contours of constitutional morality in debate and dialogue with one another. Of course disagreement persists, but it persists within parameters set by social, cultural, and political forces. And within those parameters as well, it is influenced by social dynamics.

Social coherentism, as the term is used here, provides a method for arriving at warranted conclusions to problems of religious freedom in the real world. (114) Moreover, social coherentism depicts encounters in which people try to convince each other about correct moral and legal outcomes in the context of contemporary life, with its complexity and contingency. (115)

Strong support for a social form of coherentism comes not only from moral philosophy, where such considerations are still relatively underdeveloped in discussions of reflective equilibrium, but additionally from current constitutional theory. Looking there not only provides needed perspective to philosophical accounts of coherentism or reflective equilibrium. It is also appropriate because, after all, religious freedom is a constitutional doctrine. And the social aspect of this method reinforces constitutional law's democratic legitimacy.

Popular or democratic constitutionalism describes a family of approaches that embrace the role of a wide range of groups and institutions in shaping constitutional understandings. (116) Courts are important to constitutional meaning, of course, but interpretation also happens in legislatures, in executive branch offices, in political parties, in the media, in civil society, and in social movements, among other places. Court doctrines are often shaped by interpretive arguments made in these other institutions and through these other processes.

Moreover, their influence is appropriate because constitutional law derives its legitimacy, in the long run, from democratic assent by the people. That relationship is established not only through the formal ratification process, and perhaps not even primarily that way, but also and instead through popular mobilizations and political movements. Similarly, constitutional change happens not only through formal amendment following the procedures laid out in Article V, but through myriad mechanisms of social and cultural contestation that form and express the constitutional understandings of the people themselves. (117)

Social coherentism assimilates this account of constitutional construction and change to moral reasoning that seeks a reflective equilibrium. When legal actors ask themselves whether their position on a particular issue fits together with their other convictions, they do not do so in a vacuum. Instead, they select examples that are made salient by social and political developments. Moreover, legal actors encounter arguments from others about how principles and judgments ought to fit together. Finally, they realize that their constitutional understandings must take into account doctrines that are widely accepted, if their arguments are to count as interpretations of the Constitution of the United States, rather than the charter for some other system. (To take a basic example, any interpretation of the First Amendment that allows an official church cannot count as an interpretation of American constitutional law, even if it might work for a constitutional democracy that is otherwise similar, like Great Britain's.) (118)

What happens when people disagree about the meaning of the Constitution, according to this perspective? Crude political contestation may ensue, of course, but efforts to persuade each other about the proper meaning of the Constitution may also play an important role. (119) Someone may argue that her opponents have not considered all the relevant principles or paradigmatic cases, that they have included irrelevant considerations, or that they have reasoned in the wrong way--without sufficient reflection, or in a manner that is motivated or interested. Considerations of coherence may not determine the outcome to any particular disagreement, of course, but they can prove important.

C. An Example of Incoherence

As an example of a religious freedom decision that does not cohere with basic principles, consider the Court's decision in Burwell v. Hobby Lobby Stores, Inc. (120) There, the Court held that a business corporation was exempt from the contraception mandate imposed under authority of the Affordable Care Act. Because Hobby Lobby had a religious basis for its objection to providing insurance coverage for contraception, it was protected by the federal Religious Freedom Restoration Act (RFRA), according to the Court. (121) As a result, Hobby Lobby was relieved of the obligation to provide health insurance coverage that included contraception for its employees.

But this holding violated a longstanding principle. Accommodations cannot shift meaningful burdens from some private citizens, because of their religious beliefs, to other private citizens who may have different beliefs. (122) That rule against burden shifting, which is also known as the third party harm doctrine, has been articulated in free exercise cases. For example, the Court denied an exemption from social security taxes for an Amish employer because of concern for the impact on his employees. (123) There, the Court explained that "[w]hen followers of a particular sect enter into commercial activity as a matter of choice" they accept limits on their activity that cannot be lifted where doing so would "impose the employer's religious faith on the employees." (124)

The principle has also been articulated under the Establishment Clause. In Estate of Thornton v. Caldor, Inc., the Court reasoned that "The First Amendment ... gives no one the right to insist that in pursuit of their own interests others must conform their conduct to his own religious necessities." (125) It therefore held that Connecticut could not require employers to accommodate the Sabbath observance of every employee, no matter what the impact on others. (126) The principle against harm to others has limits--it only applies to harm that constitutes an undue hardship--but within those limits it represents an important tenet of religious freedom. (127)

Although the Hobby Lobby Court acknowledged the potential harm to employees, nothing in its opinion conditioned the result on an actual absence of harm to employees. (128) The Court ruled only that exempting the company from the contraception mandate need not shift burdens to third parties, because the government retained the power to craft a solution that would extend coverage to employees another way. (129)

And in fact, the employees of Hobby Lobby were harmed. Although the Obama Administration subsequently formulated regulations that covered the employees of Hobby Lobby, those regulations were not issued until over a year after the decision was handed down. (130) During that period, significant harm was shifted to third parties. Employees went without coverage, and they likely suffered irreparable harm. (131) Therefore, the Court's decision conflicts with a basic principle of religious freedom and is unwarranted.

Of course, there is significant disagreement about whether the Court's reasoning in Hobby Lobby is supportable. (132) Proponents of the majority opinion will argue that it is not only coherent, but correct. Again, the fact of disagreement is no embarrassment to social coherentism. Its object is only to identify the conditions under which moral and legal actors can claim that their conclusions are justified and not conclusory or irrational. What we are seeing now, in the wake of Hobby Lobby, is a robust conversation about whether the decision is reasonable and correct. That debate is happening in a thick social context that is riven with conflicts over women's equality, reproductive freedom, and LGBT rights. (133) All of that is expected and salutary. It helps to ensure that the resulting settlements about the meaning of religious freedom are shaped by democratic engagement.

Still, doubts may remain about whether social coherentism really answers skeptics who claim that no theory of religious freedom is possible, and that legal decisions in this area will inevitably be irrational. Answering them depends on exactly how those doubts are understood.

D. Irrationality ?

Section I.B explained how existing theories of religious freedom are being charged with indeterminacy, ipse dixit-ism, and even irrationality. Does social coherentism fully answer those charges? This Section begins by examining the meaning of the question about irrationality. It isolates two possibilities: one understanding of the question demands too much of any theory, and the other is answered by social coherentism. Next, the Section examines an example. A sophisticated and influential version of the critique has been made by Steven Smith against Kent Greenawalt, as detailed above. But it turns out that instances of even arguable irrationality are rare. Several of Smith's charges against Greenawalt do not qualify. Even where they do arguably pertain, moreover, accusations of irrationality miss their mark--the decisions can claim to be rationally warranted in any sense that matters. Finally, the Section urges that a profitable way to assess whether a method is irrational is to ask about payoffs. When that is done, two concerns turn out to underlie the charge: bias and elitism. Yet social coherentism is not disproportionately subject to either of these, and actually it is relatively resistant to both of them.

Whether irrationality afflicts social coherentism depends on what exactly is meant by that charge. Smith seems to have in mind a situation where two resolutions of a constitutional problem, X and Y, are possible. People differ as to which they think is more defensible. They give arguments for X and arguments for Y, both of which seem reasonable, and in the end people simply choose the one they think better. Does this actually happen? And if so, does it evince irrationality?

There are at least two ways to understand the accusation of irrationality in religious freedom law. First, you could understand it to be arguing that the jurisprudence is not capable of determining discrete outcomes for all religious freedom disputes. So here the issue is whether social coherentism gives us a way to choose X over Y definitively and for every actor in every situation.

But that cannot be the question. No theory of religious freedom, or of constitutional law more generally, can determine every outcome in practice. Not even theological approaches can do that. Interpretation will be required, and it will produce some disagreement simply because people differently bear "the burdens of judgment." (134) Achieving or approaching an equilibrium should give us some good reason to think that we can identify determinate results for at least some disputes. But it is unrealistic and unnecessary to ask for determinate outcomes in all cases. (135)

A second way to understand the question of rationality is whether constitutional decisionmaking in this area can count as morally warranted or justified. This is the sense in which the challenge of the skeptics is serious. When they say that legal actors make the choice between X and Y in a conclusory way, they may mean that the choice itself is indefensible, however sophisticated the preceding arguments supporting X and Y individually happen to be. And this difficulty is especially troubling, for the skeptics, when the constitutional actor has the power to impose that choice on others who may not share it. Mere preferences or intuitions are not enough to justify a constitutional understanding, especially when it is backed by state power.

Social coherentism answers this form of the charge of irrationality by showing that the choice of X over Y can count as warranted. When someone decides that X rather than Y resonates with her other convictions about proper resolutions to legal disputes, and with constitutional principles fairly abstracted from those cases, then her choice of X over Y is rationally justified. We can say that the person has reason to make that decision. This is not intuitionism--it grounds the choice between X and Y on a conviction of coherence, and therefore on a claim of reasonableness that is subject to argument.

Again, it is always possible that someone else could object that Y rather than X coheres with all the relevant constitutional commitments. That objection could take the form that support for X is reasonable but mistaken, perhaps because it overlooks some precedent or principle, or it could take the stronger form that a defense of X is so discontinuous with considered judgments and established principles that it is in fact unwarranted. Either way, the two actors could have a conversation about how to best understand the web of considered convictions, whether all the pertinent scenarios or principles are being taken into account, and whether each other's claims of coherence hold up. They will try to convince one another.

However, the fact of debate itself does not defeat the argument that justification is possible. That debaters can work to convince one another suggests just the opposite, in fact. People with conflicting intuitions can do little to persuade one another, whereas people with different assessments of how their judgments fit together can continue the conversation.

Moreover, the approach's social aspect allows that Americans have an opportunity to influence the government's decisions on questions of constitutional meaning. Working through democratic mechanisms over time, citizens can shift the range of accepted outcomes--the "Overton Window" of constitutional meanings. (136) Political responsiveness as to the meaning of the Constitution helps to justify the imposition of one choice over another by government. Of course, lots of qualifiers must accompany this claim. People do not have equal access to political channels, and political or economic interests can improperly influence their judgments. (137) These distortions must be resisted strenuously. But effective popular influence also eases one concern, namely that the reality of government coercion makes intolerable the problem of indeterminacy. Not only does a legal determination not need to be unwarranted, but also it does not need to be unresponsive to popular will.

A search for examples reveals that instances of arguable irrationality are rare, and even Smith's key illustrations do not involve unwarranted conclusions. Smith believes that Greenawalt is complacent "with respect to the requirements of reason" and that he "makes virtually no effort to justify either his general precepts or his array of particular conclusions." (138) Given Greenawalt's commitment to reason, that failure is remarkable and it says more about the state of the general discourse on religious freedom than it does about Greenawalt himself, according to Smith. (139) All theories of religious freedom work under these discursive conditions, which generate proposals that "look like ... thinly veiled exercise[s] in ipse dixit.' (140)

Smith gives a few examples of this failing in Greenawalt, but clear instances are difficult to find. One of Smith's principal targets is Greenawalt's treatment of Good News Club, (141) as mentioned above. (142) Greenawalt critiques the invalidation of an elementary school policy that disallowed a Christian club for young children from meeting directly after school in the building. (143) Justice Thomas, writing for the Court, relied on earlier cases holding that religious groups could not be singled out for exclusion from public facilities because doing so constituted impermissible viewpoint discrimination against speech. (144) Once the school opened its premises after hours to community groups, on this approach, it could not deny access to religious groups. (145) Recall that Greenawalt distinguishes Good News Club from these earlier cases on three grounds: it concerned impressionable elementary school students; Good News was an outside group seeking to evangelize children, not an internal school club; and the meetings were to take place immediately after the school day ended. (146) Calling these "disturbing features," he concludes that the Court should have allowed school officials the discretion to conclude that young students could have perceived official endorsement of the club, and that children would have felt pressured to attend such meetings, and might have urged their parents to allow them to go, particularly if large numbers of their peers were attending. (147)

Does this decision count as irrational or conclusory? Smith calls it a "bald pronouncement[]" or something that "at least look[s] like" a bald pronouncement. (148) Given Greenawalt's detailed reasoning, that characterization is puzzling. Earlier in the same book, Greenawalt lays out the non-establishment values that inform his thinking about such cases. Among them is individual autonomy, by which he means freedom to choose free of official compulsion, whether or not the choice is informed by conscience. Government interferes with this choice, and thus burdens autonomy, whenever it '"stacks the deck' in favor of one religion or all religions." (149) Autonomy of this sort explains his conclusions regarding Good News Club, and the case helps to confirm the principle in a mutually reinforcing way. (150) One might not agree that non-establishment concerns should have prevailed in the case. But mere disagreement is not enough to show that the conclusion amounts to a bald pronouncement.

Probably Smith's clearest example of intuitionism is Greenawalt's discussion of Lee v. Weisman, (151) which involves prayer at a high school graduation ceremony. (152) On the one hand, free exercise values are implicated because religious students and their families wish to solemnize a life event of tremendous personal significance and to include a "transcendental dimension" to the ceremony. (153) On the other hand, dissenting students may experience coercion precisely because the event is so significant that attendance is not experienced as optional. Even if standing or sitting silently does not communicate assent to the prayer, it could cause offense. (154) Moreover, Justice Scalia is wrong to think that such prayers are unifying, according to Greenawalt, because Hindus and atheists may feel excluded even by the bland monotheistic prayer that was in fact offered in the case, as might people who are Christian but who wish the prayer to be more overtly religious. (155) Ultimately, Greenawalt expresses "regret" that graduation prayer cannot be permitted, but he concludes that "the principle of nonsponsorship of religion is important enough to justify the Court's determination that even these prayers are unconstitutional," and that conclusion holds regardless of whether they are officially sponsored by the school. (156) In a later piece, Greenawalt returns to school graduation, offering it as his leading example of a genuine value conflict. (157)

So the case of graduation prayer is hard. After asking "[h]ow does our reason really work?," Greenawalt contends, again, that reason cannot get him all the way to his conclusion in that case--he says, "I do not believe there is some process of reasons that settles which side is stronger.... [A] 11 I can honestly say is that the reasons on one side seem to be stronger than the reasons on the other." (158) Here, he sounds strikingly like Smith himself, as noted above. (159) Both thinkers believe that reasons cannot dictate the outcome. But is that all there is to say?

Actually, reasons do support Greenawalt's conclusion about Lee. First of all, the holding complements what is probably the central non-establishment tenet for Greenawalt, namely the principle against government endorsement of religious truth. And, again, that principle itself harmonizes with several constitutional commitments that are virtually uncontroverted among Americans. (160) Moreover, the ban on graduation prayer derives support from Greenawalt's conclusion that other ceremonial invocations of religion are similarly inappropriate, such as inclusion of the phrase "under God" in the Pledge of Allegiance--a practice that is acceptable, for him, only because overturning it at this stage would be impracticable. (161) A footnote points the reader to other discussions to bolster the proposition that favoring religious believers generally over nonbelievers can constitute endorsement of religion. (162) Finally, as Justice Kennedy himself emphasizes, the holding is consistent with the general principle barring coercive observance. (163) From the perspective of social coherentism, a conclusion in favor of Lee v. Weisman is warranted. And that is true even if the method could not eliminate indeterminacy in the case, because other commentators could find reasons to reach the opposite conclusion.

Greenawalt acknowledges the impact of his own social location on his judgments--factors like his "cultural heritage, particular upbringing, and professional training," as well his personal religious affiliation and his father's liberalism both on internal church governance and on church-state relations. (164) He mentions these factors in order to confess the influence of subjective factors on his conclusions. Yet from the perspective of social constitutionalism, those influences are benign. They allow public conversations about constitutional principles to shape the understandings of free exercise and non-establishment that matter in consequential moments of decision on questions like the constitutional propriety of graduation prayer. Yet, they are appropriate only if and after they are subject to rigorous testing through a process like the one Greenawalt actually uses.

Of course this discussion will not settle the question for everyone. A large literature exists on the nature of rationality, and it provides resources for arguing that such conclusions do not qualify as rational. (165) In addition to engaging that debate, as this Section has, we also should ask about payoffs for a theory of religious freedom. What turns on whether a decision is rational? Digging deeper, two concerns seem to be driving the charge that theories of religious freedom trade in unreason.

One concern is bias. That must be what is driving Smith when he observes that some pluralists consistently generate progressive outcomes. (166) Elsewhere he says, "the tradition seems to be ailing" and "[p]olemic, invective, and sophistry abound." (167) Reason has the potential to check those evils, he implies, but it is failing to do so under contemporary conditions. (168) Second, there is a concern with arbitrariness, which compromises the rule of law. A combination of these two worries may be driving the claim that there is no available theory of religious freedom--that making sense of decisions in this area is impossible. (169) Consider them in turn.

Bias is a familiar charge against coherentism, actually. Critics writing in philosophy likewise have worried that coherentism is voluntaristic, meaning it places too much faith in the ability of individuals to construct coherent worldviews. Not only bias, but also epistemic defects such as memory failure or incompetence can affect that process. (170) Doubtless that is right. Yet acknowledgement of those problems does not detract from the claim that coherence describes the conditions for justification--it only affects how easy it is to achieve that state. Moreover, there is no reason to think that the approach is disproportionately subject to bias or any other cognitive defect. (171) At least coherentist methods work to systematically expose bias and error, subjecting them to critique and reflection. (172)

Furthermore, there is good reason to think that coherentism in its social form is even less susceptible to bias than the conventional version. If the argument in Section B is correct, then social and political forces work not just to warp conclusions arrived at from a perspective of power or with particular interests, but they also limit the range of available understandings of religious freedom that may be seen as authoritative in any given time and place, and they give social movements opportunities to impact constitutional meanings.

Second, coherentism does not necessarily generate arbitrariness, even if different actors will arrive at different conclusions. (173) There is a difference between noticing the diversity of individual judgments and saying that their choices are arbitrary. (174) Given the choice, Greenawalt will rule against the government in the graduation case every time. Moreover, to say that it is hard to know how someone else will come down is not necessarily to say that person's decision is arbitrary. Difficult cases will persist--it is impossible to imagine any approach that could eliminate them, including theological ones. Rule of law concerns will not disappear. (175) But recognizing the social constraints on decisions mitigates those dangers substantially. And coherentism gives us substantial confidence that at least some decisions are determinate, and that many others can be defended as rationally justified and in that sense not arbitrary.

E. Individualism or Inconclusiveness?

Coherentism has drawn several characteristic objections, some of which pertain specifically to its use in law. First, the method is thought to be too individualized. On this argument, each person can arrive at his or her own understanding, which he or she can claim is justified as long as it hangs together. The theory may therefore lack the resources to condemn positions that are internally consistent but wrongheaded. (176) Law in particular cannot operate in a fragmented manner--it must be generally applicable in order to serve rule-of-law values. (177) Second is the fear that coherentism is inconclusive. (178) Because conceptions of law are always open to new conclusions based on new disputes in the real world, people's equilibriums are continually in flux. Third and last, coherentism is accused of excessive voluntarism, meaning that it puts too much faith in the ability of individuals to make sense of their moral worlds--here, specifically, the legal domain. (179) Social coherentism offers responses to all three of these objections.

In the philosophy literature, there are two typical responses to those problems. First is to say that coherentism offers only a theory of moral justification and not a theory of moral truth or knowledge. Therefore, it does not say anything about whether an individual's worldview is true or worthwhile in some absolute sense, only that the person is justified in holding it. This response is intellectually defensible--and this Essay has adopted the distinction (180)--but also somewhat unsatisfying for religious freedom or constitutional law. (181) When we say that someone's reading of free exercise or non-establishment is justified, we mean something more than that it is coherent internally. For example, a white supremacist interpretation of the Constitution might be believed to hang together but is unwarranted. (182) So this answer is too modest.

A second response among philosophers is to say that coherentism does provide a theory of truth. Someone's claim that their interpretation of religious freedom is coherent means not only that it is justified given their perspective but also that it must be true for everyone, everywhere. Yet a difficulty with absolute or ethical-realist philosophies of religious freedom is that none has succeeded in winning a consensus. Moreover, most people do not expect truth from a legal system--they understand that there are multiple ways to arrange a constitutional order, and that saying such an order makes sense (183) is not the same thing as saying that its conception of political morality is the only true one. (184) Finally, absolute justification is not needed to construct a theory of legal reasoning that can attractively depict how constitutional actors come to understand and construct doctrines. Ungrounded accounts can do the job. So, moral absolutism does not seem to be the best way to answer the criticisms of coherentism.

Social coherentism offers an intermediate solution. It holds that in order for an actor's conception of law to be coherent, it must account for uncontroverted features of a constitutional culture. (185) Constitutional law arises out of conversations and contests that citizens have with one another in a particular social location and moment. Those contingencies set outer boundaries on the understandings of religious freedom, even ones that are internally consistent. So while someone from Great Britain could construct a coherent understanding of religious freedom that included an established religion, it would not count as an interpretation of American constitutionalism. (186) Nor would the white supremacist's understanding of equality count.

Social coherentism thus avoids both the individualism of formal coherentism and the absolutism of moral ontology. Rather than either of these, it describes a way of interpreting and constructing law that is intersubjective. (187) Moreover, it is dynamic and historical, contemplating constant revision. At the same time, social coherentism modulates rapid oscillations from passing trends or quotidian politics. (188)

This method is particularly appropriate for constitutional law. American constitutionalism aspires to be democratically responsive, so that a key component of its legitimacy is that it resonates with society, politics, and culture. It stakes its legitimacy chiefly on its acceptability to the political body over time. That responsiveness means that it is revisable (rather than static) and it means that it allows for specificity (rather than suitability to all nations). Yet it also preserves the rule of law.

Putting all of this together, there emerges a depiction of complex deliberation in constitutional law. Constitutional actors properly strive for coherence. They seek to make decisions that fit with their principles and paradigm outcomes. Working back and forth, they analyze new situations in light of that existing constellation, and they revise elements of it when necessary. They do so informed by, and in conversation with, the wider constitutional culture, assessing its features and seeking to convince one another to alter them where necessary to achieve a more workable and just society.


This Essay has argued that decisions on questions of religious freedom can be rationally justified. Social coherentism offers a method for resolving disputes in a way that can generate warranted outcomes, even though it may require assimilating plural principles and analogizing to multiple judgments in other cases. Moreover, these decisions can be rationally justified despite the fact that they are embedded within, and shaped by, social and political contests over the proper meaning and implementation of the First Amendment.

Without an account like the one provided by social coherentism, skeptics may be inclined to convince decisionmakers that questions of religious freedom are incapable of principled resolution and should be left to the political process, at least to a greater degree than they currently are. (189) And the Supreme Court seems to be heeding that call, partly by eliminating the standing of plaintiffs to bring Establishment Clause challenges. (190)

Beyond the question of institutional competence, skeptics are pressing against the possibility of making arguments of constitutional principle in any forum--not only in the courts, but also in legislatures, executive agencies, and elsewhere. (191) That would abandon religious minorities to the vagaries of local power across large swaths of America. Pluralist thinkers have not yet provided a defense against charges of irrationality and indeterminacy.

This Essay has provided such a defense in the method of social coherentism. Although nothing about the approach demands progressive outcomes, it grounds them in an attractive approach. First, it resists the argument, which is surprisingly widespread, that only theological understandings of free exercise and non-establishment work to support the doctrine. (192) Moreover, it provides an understanding of moral justification that can be free from religious precepts.

And finally, its social aspect keeps the approach responsive to constitutional arguments from across the full range of political actors, and to diverse sorts of historical dynamics. It allows coherentism to incorporate the insight that constitutional meanings cohere and change in conversation with democratic politics, not outside them. And it provides a way of appreciating exactly how those social influences properly impact decisions by individual legal actors. For all these reasons, social coherentism provides an approach to religious freedom that is capable of responding to skeptics.

[c] 2015 Nelson Tebbe. Individuals and nonprofit institutions may reproduce and distribute copies of this Ardele in any format at or below cost, for educational purposes, so long as each copy identifies the author, provides a citation to the Notre Dame Law Review, and includes this provision in the copyright notice.

Nelson Tebbe, Professor of Law, Brooklyn Law School; Visiting Professor of Law, Cornell Law School. Thanks for help on this project to Gregory Alexander, Richard Bierschbach, Frederic Bloom, Corey Brettschneider, Nathan Chapman, Marc DeGirolami, John Deigh, Michael Dorf, Stanley Fish, Chad Flanders, R. Craig Green, Kent Greenawalt, Eric Gregory, Ben Griffith, Alan Hurst, John Inazu, Michael Kessler, Adam Kolber, Brian Lee, Ira Lupu, Gregory Magadan, Jens Ohlin, Sabeel Rahman, Aziz Rana, Larry Sager, Richard Schragger, Micah Schwartzman, Christopher Serkin, Steven Shiffrin, Steven Smith, Brian Tamanaha, and Robert Tuttle, as well as participants at the Annual Law and Religion Roundtable and in workshops at Brooklyn Law School, BYU Law School, Cornell Law School, University of St. Thomas Law School, Temple University James E. Beasley School of Law, University of Texas Law School, and Washington University in St. Louis. Excellent research assistance was provided by John Moore and Alex Goldman. Generous support came from the Dean's Summer Research Grant at Brooklyn Law School.

(1) See, e.g., Paul Horwitz, The Agnostic Age xiii (2011) (addressing the "incoherence of religious freedom jurisprudence"); Andrew Koppelman, Defending American Religious Neutrality 1 (2013) ("The American law of freedom of religion is in trouble."); Steven D. Smith, The Rise and Decline of American Religious Freedom 10 (2014) (describing "a constitutional jurisprudence widely regarded as incoherent").

(2) See, e.g., Van Orden v. Perry, 545 U.S. 677, 697 (2005) (Thomas, J., concurring) (describing "[t]he unintelligibility of this Court's precedent" on the Establishment Clause).

(3) See id. at 700 (Breyer, J., concurring in the judgment) ("[In borderline Establishment Clause cases,] I see no test-related substitute for the exercise of legal judgment."); Galloway v. Town of Greece, 681 F.3d 20, 30 (2d Cir. 2012) (Calabresi, J.) (noting that hard religion cases "defy exact legal formulas"), rev'd, 134 S. Ct. 1811 (2014).

(4) See, e.g., Smith, supra note 1, at 1-13; Winnifred Fallers Sullivan, The Impossibility of Religious Freedom 8 (2005) (arguing that protecting religious freedom is "arguably impossible" without intolerable inequality to nonreligious people); see also Stanley Fish, The Trouble With Principle 11-12 (1999); Frederick Mark Gedicks, The Rhetoric of Church and State 123 (1995) (arguing that constructive efforts are "doomed" and citing Steven Smith); Stanley Fish, Where's the Beep, 51 San Diego L. Rev. 1037, 1043 (2014) ("[C]ases involving free exercise exemptions and the danger of establishment continue to arise and must be dealt with, and there is no satisfactorily rational way of dealing with them."); cf. Paul Horwitz, The Hobby Lobby Moment, 128 Harv. L. Rev. 154, 155 (2014) ("[T]he very notion of religious liberty--its terms and its value--has become an increasingly contested subject [at least outside courts]."). Smith is particularly influential in this regard. He has been called "the most penetrating and thoughtful scholar of religious freedom of our generation." Marc DeGirolami, Review of Steve Smith's Rise and Decline of American Religious Freedom, Mirror of Justice (July 21, 2014), can-religious-freedom.html.

(5) Skepticism about the possibility of a coherent approach to the Religion Clauses is contemporaneous with a broader claim that the commitment to religious freedom, and to human rights more generally, only makes sense on theological grounds. See, e.g., Michael Stokes Paulsen, Is Religious Freedom Irrational?, 112 Mich. L. Rev. 1043, 1053 (2014) (book review) ("[R]eligious liberty only makes entire sense on essentially religious philosophical presuppositions."); see also Michael J. Perry, The Morality of Human Rights: A Nonreligious Ground?, 54 Emory L.J. 97, 126 (2005) ("The point is that the ground one who is not a religious believer can give for the claim that every human being has inherent dignity is obscure."); see also id. at 129 ("I am inclined to concur in [this] view[:] ... 'The essence of [our] morality is this: to believe that every human being is of infinite importance, and therefore that no consideration of expediency can justify the oppression of one by another. But to believe this it is necessary to believe in God.'"). However, the skepticism addressed in this Essay focuses on implementation of religious freedom rather than its basic justification.

(6) For examples and citations, see infra note 29.

(7) See, e.g., 1 Kent Greenawalt, Religion and the Constitution: Free Exercise and Fairness 6 (2006) (critiquing approaches to religious freedom that treat equality as the "single overarching value").

(8) See infra note 29. The term pluralism has been adopted by several experts, including some who identify with the label. It sometimes forms part of a typology that divides theories of religious freedom into three categories: singularism, which proposes one rubric or value that is capable of rationalizing the jurisprudence; pluralism, which rejects that possibility and argues that only plural values can make sense of the doctrine in an attractive way; and skepticism, which rejects the possibility of rationalizing the doctrine. See, e.g., Marc O. DeGirolami, The Tragedy of Religious Freedom 11-12 (2013) (adopting a pluralist approach and setting it off against monist and skeptical theories); see also Greenawalt, supra note 7, at 6-7 (embracing a pluralist approach, while also noting that many "condemn judicial work as incoherent or irrational"); Horwitz, supra note 1, at xxvii (contrasting theories of religious freedom that "focus [] on a single liberal principle" with "more eclectic" approaches); id. at 59 (describing "anti-theories," which deny that any theory is possible). Although that typology is common, it is not clear that it is defensible--many singularists end up relying on polyvalent considerations, at least in practice, and many skeptics look like pluralists when they decide cases.

(9) Nelson Tebbe, Nonbelievers, 97 Va. L. Rev. 1111, 1127-30 (2011).

(10) See infra text accompanying note 64.

(11) Kent Greenawalt, Fundamental Questions About the Religion Clauses: Reflections on Some Critiques, 47 San Diego L. Rev. 1131, 1144 (2010) ("[V]ery often, reason ... does not get us to the end of the line.").

(12) Id.

(13) A great deal depends on how the charge of conclusory argumentation, ipse dixit-ism, or irrationality is understood. Here, it is taken to mean that outcomes are not morally justifiable, meaning that they are not supportable by reasons. Section II.C considers other possible meanings.

(14) See Norman Daniels, Reflective Equilibrium, in Stanford Encyclopedia of Philosophy (Edward N. Zalta ed., 2011) ("The method of reflective equilibrium has been advocated as a coherence account of justification (as contrasted with an account of truth) in several areas of inquiry, including inductive and deductive logic as well as ethics and political philosophy."). As used in this Essay, the term "coherentism" describes a process designed to generate a conclusion is justified or warranted, rather than one that is true. For more on the distinction between claims of truth, knowledge, and moral justification see infra text accompanying notes 93-95.

(15) For sources, see infra notes 134-35.

(16) For more on this distinction, see text accompanying notes 93-95.

(17) The adjectives "justified" and "warranted" are used interchangeably here.

(18) 134 S. Ct. 2751 (2014).

(19) See infra Section II.C.

(20) Specifying the boundaries of social coherence in religious freedom law generally, and within its subdivisions, exceeds the scope of this Essay, which first sets out to propose a methodological defense and framework. A subsequent book will take up the more detailed work of setting out ramifications in specific legal domains. See Nelson Tebbe, Religious Freedom in an Egalitarian Age (in contract, Harvard University Press).

(21) Here, it may be helpful to distinguish between intra-personal indeterminacy, which coherentism denies is inevitable, and inter-personal disagreement, which coherentism recognizes as a source of unpredictability in legal outcomes. See Micah Schwartzman, The Completeness of Public Reason, 3 Pol. Phil. & Econ. 191, 192-93 (2004) (following Gerald Gaus in distinguishing between "indeterminacy," which is internal to a particular decisionmaker and is avoidable, and "inconclusiveness," which results from rational disagreement among actors and is "a permanent feature of liberal politics").

(22) See infra Section II.D.

(23) See infra note 189 and accompanying text.

(24) See infra note 191 and accompanying text.

(25) For critiques of the notion that religion deserves special constitutional solicitude, see Christopher L. Eisgruber & Lawrence G. Sager, Religious Freedom and the Constitution 52 (2007), Brian Leiter, Why Tolerate Religion? (2013), and Micah Schwartzman, What If Religion Is Not Special?, 79 U. Chi. L. Rev. 1351 (2012).

(26) Steven D. Smith, Discourse in the Dusk: The Twilight of Religious Freedom?, 122 Harv. L. Rev. 1869, 1893 (2009) (book review) (calling Greenawalt's prescriptions "highly conclusory" before softening that charge).

(27) Id. at 1895 (calling Greenawalt's defense of basic values "conclusory").

(28) Id. at 1906 ("[T]he various pronouncements of judges and scholars in this domain come to look like a thinly veiled exercise in ipse dixit"). But Smith is far from alone in this sort of critique of pluralism. Koppelman, for example, also thinks that Greenawalt is driven partly by unstated values. Andrew Koppelman, Religious Establishment and Autonomy, 25 Const. Comment. 291, 291 (2008) ("[A]utonomy is a mask for other concerns that Greenawalt is reluctant, for respectable but ultimately unpersuasive reasons, to spell out.").

(29) See, e.g., DeGirolami, supra note 8, at 3 ("[T]he values which swirl around the conflicts of religious liberty are incompatible and incommensurable."); 2 Kent Greenawalt, Religion and the Constitution: Establishment and Fairness 1 (2008) ("Neither free exercise nor nonestablishment is reducible to any single value; many values count."); Horwitz, supra note 1, at xxvii; Steven H. Shiffrin, The Religious Left and Church-State Relations 2-3 (2009) ("[T]he religion clauses are supported by pluralistic foundations."); Alan Brownstein, Why Conservatives, and Others, Have Trouble Supporting the Meaningful Enforcement of Free Exercise Rights, 33 Harv. J.L. & Pub. Pol'y 925, 929 (2010) ("[R]eligion is a multidimensional constitutional interest which subsumes and implicates several independently recognized constitutional values.").

(30) See, e.g., Greenawalt, supra note 29, at 7-13 (listing and describing non-establishment values); Shiffrin, supra note 29, at 20-23 (delineating free exercise values); id. at 28-40 (describing Establishment Clause values).

(31) Because that choice can entail costs, some pluralists have described themselves as pessimists or tragedians. See, e.g., DeGirolami, supra note 8, at 3 (emphasizing the "tragic" aspects of many religious freedom disputes); Horwitz, supra note 1, at xxv (noting the "tragic choices" that religious freedom jurisprudence entails). They believe that trading off values against one another can involve regret, because something of worth will be sacrificed in difficult cases. Other pluralists are more sanguine. See, e.g., Greenawalt, supra note 29, at 6-7 (noting that his pluralism makes him hesitant to condemn judicial work).

(32) For an analysis of the general problem, see Nelson Tebbe, Excluding Religion, 156 U. Pa. L. Rev. 1263 (2008). For further discussion of Greenawalt's handling of the topic, see Nelson Tebbe, Eclecticism, 25 Const. Comment. 317, 318-19 (2008).

(33) See, e.g., Locke v. Davey, 540 U.S. 712 (2004).

(34) See, e.g., Rosenberger v. Rector & Visitors of Univ. of Va., 515 U.S. 819 (1995).

(35) Greenawalt, sufrra note 29, at 194-206 (discussing Rosenberger, 515 U.S. 819; Capitol Square Review & Advisory Bd. v. Pinette, 515 U.S. 753 (1995); Lamb's Chapel v. Ctr. Moriches Union Free Sch. Dist., 508 U.S. 384 (1993); Widmar v. Vincent, 454 U.S. 263 (1981)). But see Bronx Household of Faith v. Bd. of Educ., 650 F.3d 30 (2d Cir. 2011) (upholding New York City's exclusion of worship services from public school buildings after hours, a policy that had been challenged on free speech grounds).

(36) Rosenberger, 515 U.S. 819.

(37) Greenawalt, supra note 29, at 203-04.

(38) Good News Club v. Milford Cent. Sch., 533 U.S. 98, 105, 108, 111 (2001).

(39) Greenawalt, supra note 29, at 206.

(40) Beyond ideals, Greenawalt seems interested in factors such as public opinion--recall his reference to political feasibility--and feasibility of implementation, given the current makeup of the Court. Id. at 432.

(41) Greenawalt's decision not to emphasize constitutional interpretation in the religion volumes left commentators wondering. See, e.g., Steven D. Smith, Kent Greenawalt's Elusive Constitution, 25 Const. Comment. 301, 301-07 (2008) (asking what method of constitutional interpretation Greenawalt is using). Since then, Greenawalt has embarked on a multi-volume project on interpretation. Kent Greenawalt, Interpreting the Constitution (2015); Kent Greenawalt, Legal Interpretation (2010); Kent Greenawalt, Statutory and Common Law Interpretation (2013).

(42) See infra text accompanying notes 64-66.

(43) See, e.g., Steven D. Smith, Getting over Equality 82 (2001) (pressing for stronger free exercise protection and weaker non-establishment barriers); Paul Horwitz, More "Vitiating Paradoxes": A Response to Steven D. Smith--and Smith, 41 Pepp. L. Rev. 943, 945-46 (2014) (comparing Smith's brand of skeptical religious traditionalism to a conservative version of critical legal studies).

(44) See, e.g., Sullivan, supra note 4, at 150 (stating a preference for egalitarian outcomes); see also Gedicks, supra note 4, at 7 (noting the incoherence of secular reasons for religion exemptions); Frederick Mark Gedicks, An Unfirm Foundation: The Regrettable Indefensibility of Religious Exemptions, 20 U. Ark. Little Rock L.J. 555, 556-57 (1998) (arguing against religion exemptions).

(45) See, e.g., Gedicks, supra note 4, at 123 (arguing that constructive efforts in the area of religious freedom are "doomed" and citing Steven D. Smith, Foreordained Failure 68 (1995)); id. at 124 (calling the effort to manage conflicting religious freedom values "futil[e]"); Sullivan, supra note 4, at 8, 150-51 (arguing that the category of religion is so unstable that the endeavor of protecting freedom of religion as such is "impossible").

(46) Smith, supra note 43, at 45.

(47) See, e.g., id. at 46-47; id. at 55 (arguing that separationism cannot offer "any theory or principle defining the proper scope of religious freedom"); id. at 57 (arguing that a modus vivendi is the best that can be hoped for).

(48) See id. at 52; see generally Smith, supra note 1.

(49) See Smith, supra note 26, at 1882. This categorization resembles Gedicks's distinction between secular individualism and religious communitarianism; Gedicks too believes that the discourses coexist in the doctrine and yet are contradictory. See Gedicks, supra note 4, at 11-13.

(50) Smith, supra note 26, at 1882; id. at 1883; id. at 1904-05. Smith draws on MacIntyre for some aspects of this critique:
   MacIntyre shows how as older moral and religious traditions broke
   down, debates about issues such as the justness of war, the
   permissibility of abortion, and the tension between liberty and
   economic equality became what they are today: a chaos of
   conflicting and incommensurable assertions. In this context
   MacIntyre argues that genuine political community is impossible,
   and that politics, losing the dignity it enjoyed in classical
   thought, becomes a form of "civil war carried on by other means."

Steven D. Smith, The Rise and Fall of Religious Freedom in Constitutional Discourse, 140 U. Pa. L. Rev. 149, 206-07 (1991) (citing Alasdair MacIntyre, After Virtue 6-7, 244, 253 (2d ed. 1984)).

(51) Smith, supra note 43, at 13-14 (discussing and endorsing Peter Westen, The Empty Idea of Equality, 95 Harv. L. Rev. 537 (1982)).

(52) See Eisgruber & Sager, supra note 25, at 52 (describing the core of "equal liberty").

(53) Smith, supra note 43, at 18, 34-38 (revisiting "smuggling in").

(54) Smith, supra note 26, at 1893.

(55) Smith, supra note 43, at 65.

(56) Id. at 67. Framing a dispute in terms of principle raises the discursive stakes, for Smith. Talking that way makes the losers feel that they are on the wrong side of morality, and it converts what might be squabbles into emotional conflagrations. It also raises the practical stakes, because resolutions of small fights articulate rubrics that are then applicable much more widely--perhaps nationally, if the federal Constitution is being interpreted and applied.

(57) Id. at 56-64.

(58) Id. at 96; see also Steven D. Smith, The Restoration of Tolerance, 78 Calif. L. Rev. 305, 306 (1990).

(59) Smith, supra note 43, at 96-98.

(60) Id. at 68-69.

(61) Id. at 69.

(62) Id. at 82.

(63) Id. (alteration in original) (quoting Mark DeWolfe Howe, The Garden and the Wilderness 174 (1965)).

(64) Greenawalt considers his approach to be more prudential than rigidly principled, although he confesses to relying more strongly on ideal reasoning than Smith advises. Greenawalt, supra note 29, at 438 n.19; id. at 450 (seeming to endorse "Smith's most important practical conclusions" in favor of a prudential approach "in the face of doubts about many of his conceptual claims"). What bothers Smith is not so much that the results differ as that Greenawalt purports to derive his conclusions from principles using a rational methodology, whereas Smith himself admits to the influence of non-ideal considerations on a rudderless method. Smith, supra note 26, at 1904.

(65) See supra text accompanying notes 48-53.

(66) See Smith, supra note 26, at 1890-91.

(67) Greenawalt, supra note 11, at 1144.

(68) Greenawalt has responded to Smith only by admitting that at some point in legal decision making reasons run out and all he can honestly say is that the arguments for one outcome seem stronger than the arguments for the other. Id. at 1144-45.

(69) Id. at 1145.

(70) He does suggest that relying on religious reasons would not lead to more determinate answers. Id. But that argument leaves intact the skeptical critique of his position.

(71) See DeGirolami, supra note 8, at 6 (2013); Horwitz, supra note 1, at 306-07.

(72) For citations, see infra Section II.A.

(73) Social coherentism differs from typical pluralism in several respects. First, it is a method rather than a substantive account. Second, it does not deduce the answer to new problems from multiple values one-directionally. Instead, it looks to ground-level scenarios about which there is a high degree of confidence. Also, although abstract principles or values also must cohere with any warranted conclusion about a particular scenario, they are subject to revision.

(74) Much depends on how the skeptics' critique is understood. That existing theories cannot generate outcomes that are justified is only one way of understanding that critique. Section II.C will examine others.

(75) John Rawls, A Theory of Justice 20, 46-53 (1971) [hereinafter Rawls, TOJ];John Rawls, The. Independence of Moral Theory, in Collected Papers 286, 288 (1999) [hereinafter Rawls, Independence]', John Rawls, Outline of a Decision Procedure for Ethics, 60 Phil. Rev. 177, 187-89 (1951) [hereinafter Rawls, Outline] (setting out a precursor to the method of reflective equilibrium); see also Daniels, supra note 14 (same), archives/spr2011/entries/reflective-equilibrium/; David Lyons, Nature and Soundness of the. Contract and Coherence Arguments, in Reading Rawls 141, 145-46 (Norman Daniels ed., 1989); Geoffrey Sayre-McCord, Coherentist Epistemology and Moral Theory, in Moral Knowledge? 137, 141 (Walter Sinnott-Armstrong & Mark Timmons eds., 1996) (describing the influence of Rawls on coherentism).

(76) Rawls, Independence, supra note 75, at 289.

(77) A critique of reflective equilibrium is that it depends for its power on these initial normative judgments, which cannot themselves be justified by the method itself. See Thomas Kelly & Sarah McGrath, Is Reflective Equilibrium Enough?, 24 Phil. Persp. 325, 353-54 (2010). But that critique misunderstands the provisional nature of those judgments and their vulnerability to revision through the process of seeking coherence. See T.M. Scanlon, Being Realistic About Reasons 83-84 (2014).

(78) Scanlon, supra note 77, at 76-80; see also Rawls, Outline, supra note 69, at 187-89.

(79) Rawls, TOJ, supra note 75, at 21.

(80) See Scanlon, supra note 77, at 80-81 (providing arguments that the method of reflective equilibrium can ground claims that resulting conclusions are both subjectively and objectively rational); Rawls, Outline, supra note 75, at 187-89 (describing conditions under which the method of reflective equilibrium can ground claims that moral judgments are "reasonable" or "rational" or "justified").

(81) Scanlon, supra note 77, at 78, 84 ("[A]n overall account of a subject matter can provide assurance that some judgments about it have determinate truth values.").

(82) Charles Taylor's concept of complementarity bears some similarities to the argument here. Taylor argues that it is possible to defend judgments as nonarbitrary, even if they involve conflicting values or goods, if the judgments "fit together" with a person's conception of an overall life plan, of what kind of person he or she is becoming. Charles Taylor, Leading A Life, in Incommensurability, Incomparability, and Practical Reason 181-83 (Ruth Chang, ed. 1997). What might separate the two approaches is that fitting together, for Taylor, is somewhat mysterious: it means a "sense" that the chosen path resonates with "the shape of our lives." Id. at 183. It also is not altogether clear that Taylor means to claim that complementarity bolsters a claim that the decision is warranted, although he does argue that complementarity can ground a claim that the decision is not arbitrary.

(83) Scanlon, supra note 77, at 78, 84; see also Daniels, supra note 14, at 2-3; Geoffrey Sayre-McCord, Moral Realism, in Stanford Encyclopedia of Philosophy 16 (Edward N. Zalta ed., 2011), ("[T]he specific judgments are taken as evidence for the principles and the principles reciprocate by helping to justify the thought that the specific judgments are accurate.").

(84) Joseph Raz, The Relevance of Coherence, 72 B.U. L. Rev. 273, 277-78 (1992).

(85) See Rawls, Independence, supra note 75, at 288-89 (arguing that the method of reflective equilibrium is not conservative); cf. Ronald Dworkin, Law's Empire 99 (1986) (noting the critical potential of his interpretive approach).

(86) John Rawls, Political Liberalism 97 (expanded ed. 2005) [hereinafter Rawls, PL] ("The struggle for reflective equilibrium continues indefinitely."); id. at 385 ("[Reflective equilibrium] is a point at infinity we can never reach, though we may get closer to it in the sense that through discussion, our ideals, principles, and judgments seem more reasonable to us and we regard them as better founded than they were before."); Scanlon, supra note 77, at 77 & n.14 (noting that, for Rawls, the method of reflective equilibrium is an ideal). This observation is compatible with Raz's view that coherence can be achieved locally, but likely not globally. Raz, supra note 84, at 310.

(87) Raz, supra note 84, at 310.

(88) That reflective equilibrium is an ideal state we are not likely to reach also does not mean that it is "a utopian vision toward which we cannot make progress." Tom L. Beauchamp & James F. Childress, Principles of Biomedical Ethics 383 (6th ed. 2009).

(89) Scanlon, supra note 77, at 78 (adopting the method of reflective equilibrium and describing the role of principles within it).

(90) Id. at 78-79; cf. Raz, supra note 84, at 310 ("[G]lobal coherence accounts underestimate the degree and implications of value pluralism, the degree to which morality itself is not a system but a plurality of irreducibly independent principles.").

(91) See, e.g., J.M. Balkin, Understanding Legal Understanding: The Legal Subject and the Problem of Legal Coherence, 103 Yale L.J. 105, 106 (1993) ("Coherence is more than a property of law; it is the result of a particular way of thinking about the law."); Raz, supra note 84, at 273-74 (exploring the relevance of coherence theories to legal reasoning).

(92) See Dworkin, supra note 85, at 52, 65-68, 96-98 (describing an interpretive approach to legal argumentation that allows for the influence of conceptions of justice). Of course, there is controversy about this basic proposition, that legal interpretation allows room for (or requires) moral argument, but this Essay simply assumes it to be correct. Defending that idea would take the argument too far afield, and it would mean rehearsing arguments in jurisprudence that have become familiar.

Is Dworkin a coherentist? On the one hand, Dworkin's approach shares some attributes with coherentism. Famously, he emphasizes fit and justification as criteria for constitutional interpretation. Dworkin, supra note 85, at 65-68 (defending fit and justification as independent criteria); id. at 254-58 (same); Nicos Stavropoulos, Legal Interpretivism, in Stanford Encyclopedia of Philosophy (Edward N. Zalta ed., 2014), On the other hand, his approach seems to differ from social coherentism in crucial respects. First, Dworkin sometimes says that interpretation should seek to reconstruct a single unified view of law that can be shared by the entire political community, Dworkin, supra note 85, at 225, while social coherentism--at least as defended in this Essay--demands only that a given constitutional actor achieve coherence from his or her perspective, which is likely to be shaped by social location. Social coherentism also assimilates disagreement as a fixed feature of democratic politics. Second, Dworkin usually treats justification as a top-down form of reasoning, from principles to results (with some subsequent revision of principles), id. at 65-68, while coherentism seeks mutual support among principles and paradigmatic outcomes. For assessments of whether Dworkin applies a coherence approach, see S.L. Hurley, Coherence, Hypothetical Cases, and Precedent, 10 Oxford J. Legal Stud. 221 (1990); Kenneth J. Kress, Legal Reasoning and Coherence Theories: Dworkin's Rights Thesis, Retroactivity, and the Linear Order of Decisions, 72 Calif. L. Rev. 369, 398-402 (1984); Raz, supra note 84, at 317 (arguing that Dworkin is not a coherentist); Mike Dorf, Requiem for a Hedgehog: Ronald Dworkin R.I.P., Dore on Law (Feb. 15, 2013), hog.html (Dworkin was a coherentist, not only as to justification but also as to truth).

(93) See Rawls, Independence, supra note 75, at 289; Daniels, supra note 14, at 25 (noting Rawls's distinction between a coherence theory of truth and a coherence theory of justification).

(94) Daniels, supra note 14, at 24-25 (exploring the differences between reflective equilibrium and epistemology).

(95) Dworkin, supra note 85, at 82 (arguing that "external skepticism" based on metaphysics does not threaten moral claims); Scanlon, supra note 77, at 86 & n.27; cf. Richard Rorty, Contingency, Irony, and Solidarity xv (1989) ("For liberal ironists, there is no answer to the question 'Why not be cruel?'--no noncircular theoretical backup for the belief that cruelty is horrible.... Anybody who thinks that there are well-grounded theoretical answers to this sort of question... is still, in his heart, a theologian or a metaphysician.").

(96) Rawls himself thought that the question of whether a reflective equilibrium was unique was beyond the scope of his project. He assumed that a reflective equilibrium would be the same across persons, or that disagreement would reflect familiar divides in moral philosophy. Rawls, TOJ, supra note 75, at 50. Later, Rawls developed the idea of a reflective equilibrium that was "general and wide." In a well-ordered society, every citizen has achieved a reflective equilibrium that is general and wide, and therefore "full" or "fully intersubjective," meaning that each person takes into account the arguments of everyone else. Rawls, PL, supra note 86, at 384 n.16 (in the Reply to Habermas).

(97) Scanlon, supra note 77, at 79.

(98) Here again it might be helpful to distinguish between intra-personal coherence and inter-personal conflict. See supra note 21. That we disagree with someone else is no necessary reason to conclude that our views are unwarranted. For similar reasons, inter-personal indeterminacy is not inevitable, whereas inter-personal indeterminacy is a feature of political life that we can expect to persist. See Schwartzman, supra note 21, at 192-93.

(99) Scanlon, supra note 77, at 79-80, 82.

(100) As Rawls himself sometimes suggests. See Rawls, TOJ, supra note 75, at 53.

(101) Scanlon, supra note 77, at 78; see also Sayre-McCord, supra note 75, at 143-45 (arguing that coherence is useful not just for discovering our beliefs that are justified some other way, but for moral justification itself).

(102) See, e.g., Beauchamp & Childress, supra note 88, at 25 ("[T]he phenomenon of moral disagreement provides no basis for skepticism about morality or moral thinking.").

(103) See infra Section II.D.

(104) Smith, supra note 26, at 1893-95 (arguing that Greenawalt "never attempts any genuine defense" of non-establishment precepts, including the ban on government endorsement of religious truth).

(105) Greenawalt himself does not claim to be a coherentist. He explains that he reasons "from the 'bottom up,'" starting with discrete conflicts and "investigating conflicting values over a range of issues." Greenawalt, supra note 29, at 1. Smith compares Greenawalt's method to Gratian's "dialectical" strategy, although that approach has more to do with considering arguments for and against a particular decision than it does seeking coherence. Smith, supra note 26, at 1888.

(106) Greenawalt, supra note 11, at 1144.

(107) See Greenawalt, supra note 29, at 53. The surrounding chapter of Greenawalt's book serves to support one other principle as well, namely the rule against government preferentialism among religions.

(108) Id. at 59-60.

(109) Id. at 60.

(110) Id. at 10-13. When Greenawalt lists these non-establishment values in the introduction to his book, he does so as a way of previewing what his "bottom-up" method will yield. During the course of the discussion, he then induces those values from ground-level cases. He also draws upon them to support his conclusions about other cases. That initial discussion is supported by the book's whole apparatus.

(111) See, e.g., id. at 15, 53, 57-58, 195, 445, 467 (stressing incompetence); id. at 493.

(112) At times, Greenawalt seems to articulate something like a coherence grounding for the principle and for religious freedom generally. "[D]ifferent justifications [might] coalesce behind a common principle that itself can be employed to reach decisions." Greenawalt, supra note 29, at 494. Other times, the defense sounds more like the identification of an overlapping consensus. For example, "very different underlying justifications for the religion clauses support the principle of nonsponsorship." Id. at 494-95.

(113) Greenawalt, supra note 11, at 1144. He mentions his own parentage, his religious background, and his personal experiences of kindness among loved ones as important influences. Id.; see also Nancy Levit, Practically Unreasonable: A Critique of Practical Reason, 85 Nw. U. L. Rev. 494, 502 (1991) (reviewing Richard A. Posner, The Problems of Jurisprudence (1990)) (noting, albeit in a critical way, that practical reason is "dependent on cultural fabric, on social, ethnic, and geographic variations, and on historical traditions").

(114) It does not depend on "the priority of the moral over the political," to use Bernard Williams's language, but instead it speaks to legal actors who are fully situated within social dynamics and political struggles. Bernard Williams, In the Beginning Was the Deed 8 (2005). Williams is right that it would be foolish to think that our moral convictions are not the products of social and cultural conditions. Id. at 13 ("[W]e would be merely naive if we took our convictions, and those of our opponents, as simply autonomous products of moral reason rather than as another product of historical conditions.... This does not mean that we throw our political convictions away ....").

(115) Compare Rorty's argument that liberalism "is content to call 'true' whatever the upshot of [persuasive] encounters turns out to be." Rorty, supra note 95, at 52. Rorty says that Rawls and other writers would happily grant that a circular justification of our practices, a justification which makes one feature of our culture look good by citing still another ... is the only sort of justification we are going to get. I am suggesting that we see such writers as these as the self-canceling and self-fulfilling triumph of the Enlightenment.

Id. at 57. It is instructive to compare social coherentism to the shift from epistemology to politics urged by both Rorty and Williams, albeit in different ways. Rorty, supra, at 95; cf. Williams, supra note 114, at 7-8. Social coherentism simply takes no position on questions of truth, at this stage in its development, and instead it insists only that conclusions arrived at through the method deserve to be called reasonable or warranted.

(116) For a definition of popular constitutionalism and an overview of its varieties see Larry D. Kramer, Popular Constitutionalism, Circa 2004, 92 Calif. L. Rev. 959, 959 (2004) ("In a system of popular constitutionalism, the role of the people is not confined to occasional acts of constitution making, but includes active and ongoing control over the interpretation and enforcement of constitutional law."); see also Jack M. Balkin, Living Originalism 277-82 (2011); Robert C. Post & Reva B. Siegel, Democratic Constitutionalism, in The Constitution in 2020 25 (Jack M. Balkin & Reva B. Siegel eds., 2009).

(117) Cf. McCulloch v. Maryland, 17U.S. (4 Wheat.) 316,403 (1819) (distinguishing measures adopted by states from measures enacted by "the people themselves"); Larry Kramer, The People Themselves: Popular Constitutionalism and Judicial Review (2004).

(118) This is not the same as Dworkin's criterion of fit, though it is similar in some respects. See supra note 92 (describing the debate over whether Dworkin is a coherentist).

(119) See Balkin, supra note 116, at 331 ("[N]ormative argument about the Constitution is hardly futile.... Arguments about what the Constitution means and who has the authority to say what it means are important because they can persuade the actors in the system to think differently."); Post & Siegel, supra note 116, at 27 ("Americans believe in the possibility of persuading others--and therefore ultimately the Court--to embrace their views about constitutional meaning.").

(120) 134 S. Ct. 2751 (2014).

(121) Id. at 2759 (applying 42 U.S.C. [section][section] 2000bb-bb4).

(122) For a proposal regarding exactly when religion accommodations should be permitted to harm third parties, see Micah Schwartzman, Richard Schragger & Nelson Tebbe, When Do Religion Accommodations Burden Others? (unpublished manuscript, on file with author).

(123) United States v. Lee, 455 U.S. 252 (1982).

(124) Id. at 261.

(125) 472 U.S. 703, 710 (1985) (quoting Otten v. Balt. & O.R. Co., 205 F.2d 58, 61 (2d Cir. 1953) (L. Hand, J.)).

(126) See also Religious Freedom Restoration Act and the Religious Land Use and Institutionalized Persons Act: Hearing Before the Subcomm. on the Constitution and Civil Justice of the H. Comm, on the Judiciary, 114th Cong. 36-46 (2015) (testimony of Nelson Tebbe, Prof, of Law, Brooklyn Law Sch.); Frederick Mark Gedicks & Rebecca G. Van Tassell, RFRA Exemptions from the Contraception Mandate: An Unconstitutional Accommodation of Religion, 49 Harv. C.R. & C.L. L. Rev. 343 (2014); Micah Schwartzman, Richard Schragger & Nelson Tebbe, The Establishment Clause and the Contraception Mandate, Balkinization (Nov. 27, 2013), http://

(127) See Schwartzman et al., supra note 122 (arguing for an undue hardship limit on the principle against harm to others).

(128) The Court cast doubt on the third party harm doctrine in a footnote. Burwell v. Hobby Lobby Stores, Inc., 134 S. Ct. 2751, 2781 n.37 (2014). But that language is dicta, given the Court's (erroneous) view that no harm to third parties would result from its ruling. Moreover, Justice Kennedy, whose vote was necessary to assemble the majority, was clearer that the absence of a harm to third parties was a necessary predicate of the ruling in Hobby Lobby. Id. at 2786-87 (Kennedy, J., concurring) ("[In America,] no person may be restricted or demeaned by government in exercising his or her religion. Yet neither may that same exercise unduly restrict other persons, such as employees, in protecting their own interests, interests the law deems compelling. In these cases the means to reconcile those two priorities are at hand in the existing accommodation the Government has designed, identified, and used for circumstances closely parallel to those presented here. RFRA requires the Government to use this less restrictive means."). For a detailed critique of footnote 37, see Micah Schwartzman et al., supra note 122, at 9-13.

(129) 134 S. Ct. at 2759-60.

(130) Coverage of Certain Preventative Services Under the Affordable Care Act, 80 Fed. Reg. 41318 (July 14, 2015); see also Nelson Tebbe, Richard Schragger & Micah Schwartzman, Update on the Establishment Clause and Third Party Harms: One Ongoing Violation and One Constitutional Accommodation, Balkinization (Oct. 16, 2014), 2014/10/update-on-establishment-clause-and.html.

(131) See Notre Dame v. Burwell, 786 F.3d 606, 607-08 (7th Cir. 2015) (detailing the health benefits to women of inexpensive contraception coverage, including the avoidance of unintended pregnancies, which are associated with other health problems and reduced participation in economic, social, and political life); Priests for Life v. U.S. Dep't of Health & Human Servs., 772 F.3d 229, 257-64 (D.C. Cir. 2014) (same); Marty Lederman, Hobby Lobby Part VII: Hobby Lobby's Arguments on Compelling Interest and the Alleged Exemption "Honeycomb", Balkinization (Feb. 17, 2014), (arguing that depriving employees of coverage even for claimed abortifacients will result in unwanted pregnancies and citing an amicus brief from the Guttmacher Institute); Nelson Tebbe, Richard Schragger & Micah Schwartzman, Hobby Lobby's Bitter Anniversary, Balkinization (June 30, 2015), http://balkin.blogspot. com/2015/06/hobby-lobbys-bitter-anniversary.html.

(132) See, e.g., Horwitz, supra note 4, at 165 (calling Hobby Lobby an "easy case" and the Court's reasoning "highly straightforward" but focusing on the social phenomenon of disagreement over the holding).

(133) See id. (situating the Hobby Lobby decision in wider culture-war debates); see also Elizabeth Sepper, Reports of Accommodation's Death Have Been Greatly Exaggerated, 128 Harv. L. Rev. F. 24 (2014) (responding to Horwitz).

(134) Rawls, PL, supra note 86, at 55-56. For similar reasons, Dworkin was right to reject what he called "external" skepticism and to take seriously only "internal" skepticism. Dworkin, supra note 85, at 79; see also Scanlon, supra note 77, at 86 n.27 ("Here I am in agreement with Ronald Dworkin that only internal skepticism is worth worrying about.").

(135) Scanlon, supra note 77, at 78, 84, 104 (discussing reflective equilibrium and arguing that "an overall account of a subject matter can provide assurance that some judgments about it have determinate truth values"). But see id. at 85, 104 (arguing that it is not completely plausible to think that the domain of practical reasoning can be characterized so that every normative claim within it has determinate truth value).

(136) The Overton Window: A Model of Policy Change, Mackinac. Ctr. for Pub. Pol'y, ("Joseph Overton observed that in a given public policy area, such as education, only a relatively narrow range of potential policies will be considered politically acceptable."); see also Lawrence Lessig, The Puzzling Persistence of Bellbottom Theory: What a Constitutional Theory Should Be, 85 Geo. L.J. 1837, 1837 (1997) (describing the difference between "taken for granted" constitutional ideas and "up for grabs" arguments); Jack M. Balkin, From off the Wall to on the Wall: How the Mandate Challenge Went Mainstream, The Atlantic (June 4, 2012), http://www.theatlantic.coin/national/ archive/2012/06/from-off-the-wall-to-on-the-wall-how-the-mandate-challenge-went-mainstream/258040 (describing how arguments once considered "off the wall" become "on the wall" through contestation and argument).

(137) Rawls was mostly worried about the influence of power and interests on judgments of politics and morality, particularly judgments made on the basis of intuition. His methodology sought to expunge such influences, or at least dampen them. Rawls, TOJ, supra note 75, at 35.

(138) Smith, supra note 26, at 1872.

(139) Id.

(140) Id. at 1906.

(141) Id. at 1893 (discussing Good News Club v. Milford Cent. Sch., 533 U.S. 98 (2001)).

(142) See supra text accompanying notes 38, 54.

(143) Greenawalt, supra note 29, at 206.

(144) See, e.g., Lamb's Chapel v. Ctr. Moriches Union Free Sch. Dist., 508 U.S. 384, 394 (1993).

(145) Id. at 396-97.

(146) Greenawalt, supra note 29, at 206.

(147) Id.

(148) Smith, supra note 26, at 1893.

(149) Greenawalt, supra note 29, at 9.

(150) Another example Smith admits is more complicated, and it involves not a judgment about a case but a principle, is that government may not take positions on questions of religious truth. Smith, supra note 26, at 1893, 1894-95. When he first announces it, Greenawalt does not offer much to support this principle. Greenawalt, supra note 29, at 57-58. But much else in the book does undergird it, as argued above. See supra text accompanying notes 99-107. And in fact, Greenawalt cites it as perhaps the best example of a precept that can in fact be supported by reasons. Greenawalt, supra note 11, at 1144. Still another of Smith's examples is inapt. He criticizes the argument that being forced to support religion through taxation violates non-endorsement. Smith, supra note 26, at 1897. But Greenawalt acknowledges that the argument is weak when the taxes are not specific to religious support, and he therefore does not rely on it. See Greenawalt, supra note 29, at 5, 8, 196.

(151) Greenawalt, supra note 29, at 111.

(152) Given the strength of the example, it is notable that Smith actually offers it not as an example of conclusory argument, but for a different observation, namely that Greenawalt is charitable to religious people who wind up on the other side of his conclusions. Smith, supra note 26, at 1894 (quoting Greenawalt, supra note 29, at 113 n.40). Greenawalt notes that religious students and their families want "to acknowledge God and seek God's blessing." Greenawalt, supra note 29, at 113 n.40.

(153) Greenawalt, supra note 29, at 111.

(154) Id. at 112.

(155) Id. at 114.

(156) Id. at 115.

(157) Greenawalt, supra note 11, at 1144.

(158) Id.

(159) See supra text accompanying note 64.

(160) See supra text accompanying notes 100-08.

(161) Greenawalt, supra note 11, at 1143.

(162) Greenawalt, supra note 29, at 113 n.38.

(163) See Lee v. Weisman, 505 U.S. 577, 587-88 (1992) ("It is beyond dispute that, at a minimum, the Constitution guarantees that government may not coerce anyone to support or participate in religion or its exercise ... [and here] subtle coercive pressures exist[ed,].... [T]he student had no real alternative which would have allowed her to avoid the fact or appearance of participation.").

(164) Greenawalt, supra note 11, at 1144-45.

(165) Cf., e.g., Raz, supra note 84, at 313 n.71 (rejecting the idea that a moral choice is unreasoned simply because a reason cannot be given for preferring that choice over all others, because there are reasons to support the chosen policy); see also Elinor Mason, Value Pluralism, Stanford Encyclopedia of Philosophy (Edward N. Zalta ed., 2011), (describing this debate and citing sources).

(166) Smith, supra note 26, at 1890-91.

(167) Id. at 1871 ("[T]he tradition [of discourse on religious freedom] seems to be ailing. Probably the most common adjective used in descriptions of the contemporary jurisprudence of religious freedom is incoherent. Polemic, invective, and sophistry abound." (internal quotation marks omitted)).

(168) Id. ("What the tradition desperately needs, it seems, is ... a careful, systematic demonstration that controversies over religious freedom can actually be resolved through 'reasoned analysis, as distinguished from rhetoric.'"). At one point, Smith is discussing the historical pedigree of the rule against government pronouncements on questions of religious truth. After noting that Washington, Jefferson, and Lincoln all violated the rule, he asks whether perhaps more recent history supports it:
   But tradition is an evolving matter, and in our more secular and
   diverse society, we understand that such expressions are divisive
   and inappropriate. Don't we?

      Well, actually, no: we don't--not unless the "we" is understood to
   refer to a smaller and more select fellowship (like, say, devout
   readers of the New York Times?).

Id. at 1900 (footnote omitted).

(169) Sullivan, supra note 4, at 8; Smith, supra note 26, at 1901.

(170) Daniels, supra note 14; Raz, supra note 84, at 282.

(171) And this is not just a product of secular modernity. Theological approaches also suffer from such limitations. Greenawalt, supra note 11, at 1145 ("1 would not do better if I directly employed my religious convictions. They do not typically yield more decisive answers ....").

(172) See Daniels, supra note 14; Sayre-McCord, supra note 75, at 141 (arguing that the "underlying idea" is that we "subject our evaluative attitudes to the pressures of reflection").

(173) Rawls, PL, supra note 86, at 55-56 ("[T]he sources of reasonable disagreement--the burdens of judgment--among reasonable persons are the many hazards involved in the correct (and conscientious) exercise of our powers of reason and judgment in the ordinary course of political life.").

(174) Raz, supra note 84, at 313 n.71. Again, this difference tracks the distinction urged by Schwartzman and Gaus between intra-personal indeterminacy, which is avoidable, and inter-personal indeterminacy, which is unavoidable and generates uncertainty in legal outcomes. See supra note 21.

(175) Compare the view of Raz, who insists that "social politics" will result in moral inconsistency, because more than one perspective will be written into doctrine. Raz, supra note 84, at 310-11. As shown in the next Section, social coherentism is compatible with much of what Raz says, although it is agnostic on his objection to views that the law should be conceptualized as a coherent whole. Whatever the truth of those views, this Essay argues only that constitutional actors are engaged in a shared project of arguing over the meaning of the First Amendment.

(176) Id. at 280 ("[I]t is simply false that we hold as justified beliefs conceived in prejudice and superstition, or entertained because of gullibility, obstinacy, or similar cognitive defects of the believer. The racist's belief in the untrustworthiness of members of a certain race, bred of prejudice, is not justified even if it coheres best with all the racist's other (mostly racist) beliefs."); see also Sayre-McCord, supra note 75, at 170 (noting the prevalence of a version of this objection).

(177) See Raz, supra note 84, at 313.

(178) See Lyons, supra note 75, at 146.

(179) See Raz, supra note 84, at 282.

(180) See supra text accompanying notes 93-95.

(181) For example, Raz thinks that coherentism cannot work for law because it lacks a "base" that can draw together individuals' coherent views together in one legal system. Raz, supra note 84, at 284. But see Balkin, supra note 91, at 128 (critiquing Raz on the ground that there are many different perspectives from which to seek coherence).

(182) See Raz, supra note 84, at 280. Jack Balkin believes that truth claims place wide boundaries on coherentist accounts. Balkin, supra note 91, at 120-21. Even if a legal understanding is consistent, for him, it cannot claim to be reasonable if it relies on an interpretation of principle that is false in an absolute sense.

(183) Cf. Williams, supra note 114, at 10-11 (defining "makes sense" as a category of historical understanding).

(184) Balkin, supra note 91, at 118.

(185) Cf. Dworkin, supra note 85, at 66-68 (articulating a criterion of fit); see also supra note 92 (noting differences between social coherentism and Dworkin's approach).

(186) Thanks to Adam Kolber for this hypothetical. Cf. Leiter, supra note 25, at 116-30 (arguing that an established church might be acceptable, if one argues from first principles of liberal democracy).

(187) Beauchamp and Childress adopt the method of reflective equilibrium for moral justification, but they believe that reflective equilibrium cannot defend against the charge of individualism or relativism without some grounding that does not itself depend on coherence. They say,
   We cannot justify every moral judgment in terms of another moral
   judgment without generating an infinite regress or vicious circle
   of justification in which no judgment is justified. The way to
   escape this regress is to accept some judgments as justified
   without dependence on other judgments.

Beauchamp & Childress, supra note 88, at 385. They suggest that a "common morality" can ground the method of reflective equilibrium in this way. Id. Social coherentism, by contrast, proceeds without any such foundation. It argues that justification can avoid the problem of individualism by reference to intersubjective convictions, without more.

(188) Dynamism is an asset of the theory, not a liability. Cf. Lyons, supra note 75, at 146 ("[N]o coherence argument can be conclusive."). Democratic responsiveness demands that closely contested constitutional principles remain in flux, at least in principle (even though others will be relatively entrenched).

(189) See, e.g., Smith, supra note 43, at 68 (arguing that the rejection of a principled approach to religious freedom suggests that courts should play a more modest role, generally, though acknowledging that determining how modest will require additional, complicated jurisprudential considerations); Richard W. Garnett, Judicial Review, Local Values, and Pluralism, 32 Harv. J.L. 8c Pub. Pol'y 5, 6 (2009) (critiquing judicial review for "homogenize[ing] community norms").

(190) Sri Hein v. Freedom From Religion Found., Inc., 551 U.S. 587 (2007).

(191) See Smith, supra note 43, at 67-68 ("[A] prudential approach could admit that constitutional law--and thus judges, lawyers, and legal scholars--probably have no more than a modest contribution to make to the realization of religious freedom."); id. at 79 ("I have suggested that there are no satisfactory principles of religious freedom. If I am right, the most obvious consequence is that in this area courts cannot act on the basis of genuine principles. But there is another less obvious but perhaps more interesting consequence: no one else--not a legislature, not a school board--can act on the basis of genuine principles of religious freedom either.").

(192) See supra note 4 (citing sources).
COPYRIGHT 2015 University of Notre Dame Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2015 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Tebbe, Nelson
Publication:Notre Dame Law Review
Date:Nov 1, 2015
Previous Article:Collateral consequences and the preventive state.
Next Article:Judicial candor and extralegal reasoning: why extralegal reasons require legal justifications (and no more).

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters