This article considers the effects of secularization on the American constitutional law of religion. It argues that, in a secular age, religious belief becomes a matter of subjective opinion to be treated in much the same way as other opinions. Claims that religious beliefs occupy a privileged legal position become problematic, once those opinions are understood to be merely subjective. Free exercise jurisprudence tends to collapse into the jurisprudence of free speech: We protect religious speech because it is speech, not because it is religious. In establishment clause jurisprudence, endorsement becomes the focus of attention. If religious belief can never be more than a subjective opinion, then it is measured like other opinions: we want to know who supports it, not whether it is true. Thus, the endorsement test asks whether the state is asserting a religious belief as its own. Once religious belief is understood as subjective opinion, its natural place will appear to be within, not outside of, politics, for interest groups are defined by the opinions they pursue. Perhaps ironically, the increasingly visible role of religious groups in American politics is an expression not of how seriously we take religion, but of how much less serious it has become. No longer about the truth of the human condition, it is merely another opinion.
Charles Taylor points out that over the last several hundred years, Western societies have gone from a situation in which God was an acknowledged presence in everyday life to one in which it is not only possible to live with disbelief, it may be difficult to be too public about belief. Belief in God, once a self-evident truth, has become a mere opinion in what Taylor calls our “secular age.”  The modern American jurisprudence of religion has, in large part, been an effort to align legal norms with this shift in the status of religious propositions from representations of fact to expressions of opinion.
Is there anything special about religious opinions? Much of our recent jurisprudence suggests that the answer is “no.” The Supreme Court confronted the issue directly in several statutory interpretation cases in the 1960s and 70s, involving draft exemptions for those making claims to conscientious objection.  Despite seemingly clear statutory language meant to distinguish religious objections from other grounds of objection, the Court insisted that the statutory reference to “religious” belief had to be seen as a short hand for sincere opinion or conviction.  The Court seemed to be following a colloquial usage under which to hold to an opinion “religiously” is to be steadfast in that belief.
As a matter of constitutional law, the shift of religious claims from the sphere of facts (truth) to that of opinion contributed to the substantial collapse of the jurisprudence of religion into the jurisprudence of speech.  We protect religious speech not because it is religious, but because it is speech. Government practices that touch on religion are also viewed within the framework of speech: What speech may the government permissibly support? Coupled to this understanding of religion as opinion was a view that religious practices and beliefs serve a sort of ornamental function. Secularism, ornamentalism, and opinion are all linked in the modern jurisprudence of religion.
Alongside the changes in the reception of religious claims, there has been a parallel shift in the perception of political participation by religious groups. They are seen now as political interest groups that engage with others as active participants in America’s culture wars. Religious claims are often tactics in this much larger fight that now characterizes American politics. It would be too much to claim that the jurisprudential shift caused a political shift, but both shifts make sense within a broadly secular culture that understands religion as primarily a matter of subjective opinion.
Desacralization: A Secular Society
When I refer to a secular society, I use the term as Charles Taylor does, meaning not a society characterized by an absence of belief, but one in which the conditions of belief have shifted: “The shift to secularity … consists … of a move from a society where belief in God is unchallenged and indeed, unproblematic, to one in which it is understood to be one option among others, and frequently not the easiest to embrace.”  In a secular society, belief must compete with non-belief. Today, one can live an entire life without ever having any serious contact with organized religious faith. For many people, nonbelief is taken for granted. There is nothing problematic about living outside of religious practices. There is no sense that one is missing something necessary to a full and meaningful life.
This recognition of non-belief as an option has a profound effect on belief itself. Naïve belief becomes difficult to maintain, for one cannot escape awareness that the questioning and rejection of belief are always possible.  Religion now operates within a polity that accepts as a norm that the ends of individual and society need extend no further than fulfillment of the possibilities within life itself.
Fundamentalist and evangelical groups do have an outsized influence on American politics – an influence vastly inflated beyond their proportion of the population by gerrymandering and the organization of Republican Party primaries.  That political power does not change the basic dynamic of secularization as a cultural phenomenon. Nonbelievers simply do not care that believers disapprove of their secular life style; they are unmoved by assertions about their own prospects for life in the hereafter. Believers cannot escape the awareness of these views; they must take account of them in their own political efforts. The recent debate over gay marriage, leading to Obergafell v. Hodges, illustrates this dynamic of secularism: not just nonbelievers, but many believers came to acknowledge that a secular, political society must recognize possibilities of human flourishing outside the normative demands within a community of faith. 
This desacralization of our world is so deep that we usually take it for granted. It takes a significant effort to imagine what it was like to live in a world in which belief was not optional; in which to proclaim nonbelief would have been as odd as proclaiming today that one does not believe in medicine. This possibility of widespread disbelief was as unimaginable to our predecessors as their condition is unimaginable to us.  As modern a figure as Kant wrote that reason itself was compelled to ask ultimate questions – theological questions – even though it did not have the means to answer them. His notion of Enlightenment was not secularization but “religion within the limits of reason alone.” That religion included faith in God as a necessary corollary to our experience of the organized, systemic character of the world, and belief in the afterlife as a necessary corollary of the moral demand that reason makes upon the will.  Today, we do not find that our reason compels us to ask these questions and therefore we are not particularly bothered by the lack of answers. Faith is not a necessary condition of our experience, but something extra that we can choose to take up or put off. Its function is often ornamental.
While Kant thought that religion and science were complementary and reciprocally supportive, today we tend to see a natural animosity – or at least tension – between religious faith and science. Science permits no lacuna in its causal account that might allow for the possibility of divine intervention. There is literally no space or time for God. The assumptions of several hundred years ago ran just the other way. It was impossible to imagine the world without creation, and impossible to imagine creation without an active intellect of some form. This did not produce any tension with science, for it was an article of faith that the world that science explored was the world that God created. The same was true of morality: the natural law that reason discovered was the same law that God willed. Thus, reason and revelation worked in tandem. 
Right through the eighteenth century, nature was studied as a sign – a text – of God’s creative work.  This was, for example, the way that Newton understood his work. We like to think of Newton as a man of science: someone like us. He was a man of science, but he was not like us, for his world was still enchanted.  Jefferson reflects a similar view when he writes, at the beginning of the Declaration of Independence, of “the laws of nature and of nature’s God.” There was no break between nature and creation, between the truth of science and the truth of religious belief. Faith was not merely a matter of opinion, but of truth. The early colleges were places for the study of science and theology – not as distinct disciplines but as elements of a single plan.
Looking to the origin of American colleges suggests a more general caution: behind apparent institutional continuity there can be major discontinuity. Today’s churches may trace their origins back to the pre-modern era, but they are remnants lodged in a different world. They are like the monarchies of Europe: genealogical continuity tells us little about the architecture of contemporary belief.  Those monarchs were once the site of God’s presence; their bodies were the mystical corpus of the nation, just as Christ was the mystical corpus of the Church universal.  The descendants of these Christ-like kings and queens are today only ornaments in a thoroughly secular political project. Their continuation depends on practical judgments that have nothing to do with faith in the mystical corpus of the nation.
The same is true of many of our churches. Whether they continue is often a function of the social work they accomplish, and their ability to position themselves in response to family and community needs. We shop for churches today in much the same way in which we shop for anything else. A church that can be tried on and discarded is not one that stands guard over the destiny of the soul and meaning of life. Many of our churches, along with their ministers-administrators, belong in the category of the ornamental. Indeed, often churches and ministers seem most in demand at those celebrations marking significant life events.
The disjunction between contemporary disenchantment and the presence of God in the age of the Founders creates a situation ripe for anachronistic readings. When we look to the Founders’ actions for any sort of guidance as to the meaning of the constitutional norms respecting religion, we are looking into a foreign land. For example, the Supreme Court recently noted that Samuel Adams suggested starting the first session of the First Continental Congress with a prayer, thinking this might help to promote unity.  Apparently, it worked. The opinion, however, sounds a false note, a purposive anachronism, in suggesting a parallel with a prayer opening a contemporary town meeting.  After all, one of the reasons the Court gives for upholding the practice is that no one takes it very seriously – people continue to come and go without regard for the prayer. If we were really asked to take the ministerial voice at such moments as more than ornamental, many would take offense.
Ornamentalism, as I use the term, refers to an institution, position, or practice that continues even after the retreat of the supporting belief structure out of which it had arisen and within which it had a substantive meaning. Sometimes, the shift is geographic as when objects of a foreign culture become decorative. Objects that had occupied a rich network of meanings and practices are now displayed for their aesthetic value. That same shift can occur historically as objects or practices outlive the networks of meaning in which they had been embedded as part of a ready-to-hand world. The tool or the icon becomes an ornament.
Ornamentalism can express romanticism and nostalgia. It can attach to objects or practices that we keep before us as expressions of a longing that we have no intention of fulfilling. Ornamentalism can produce the “quaint.” We are not serious about these objects or groups on their own terms. Rather, the quaint is juxtaposed to our ordinary life as a way of responding to anxieties about the limits of that life. Holding on to an ornamental presence can relieve us of the burden of confronting just how much we have changed. The colonial-era church on the town green is quaint. Its presence deflects our attention from the commercialization that has otherwise occupied the surrounding streets.
There is always something potentially inauthentic about ornamentalism. We have stripped the traditional meaning from ornamental objects, groups, and practices. Yet, by keeping them present, we avow a kind of attachment. The British monarch is hardly the mystical corpus of the nation. She is not really the head of the government or the leader of the Anglican Church. She is an ornament, a remnant of all of those past meanings, to a nation that is not quite willing to confront what it has become – a financial center but a weak power, a subordinated ally, a problematic member of Europe, an aggregate of diverse ethnicities.
One might say of the Queen that she is a reminder of what Britain once was, except that hardly anybody actually knows that history. The very claim of a “reminder” is itself a practice of ornamentalism. Ornamentalism suggests something not deep, but on the surface. If we take it too seriously, we are embracing inauthenticity, for the ornament is distinctly not what it seems. The survival of the Queen is not an invitation for the revival of monarchism. The survival of the church on the New England town green is not an invitation for the revival of the theocratic governments of the early colonies.
Just as the Queen is an ornament, for many people the minister officiating at a wedding ceremony or giving an invocation at a public gathering is an ornament. Most people do not believe that marriage is a sacrament or that God has any concern with our public meetings. The minister used to stand in a place of authority within the community. No more. He has become ornamental. That does not mean he is irrelevant or unimportant. It does mean that his presence is no longer to be understood in a theological, but rather in a social, framework.
Practices and institutions have a sort of stickiness. They linger for reasons that have little to do with their origins. They become habitual, which may make them a source of comfort in an unstable world. They may remind us of other events or relationships; they signify home or childhood. They may serve as signs marking the significance of an occasion. People like to have celebratory occasions simply to set the day apart from others or to mark the cycles of a year. They inherit these celebrations and their marks; they continue to enjoy what were once rituals, long after they have forgotten their original significance. They may not know how to do things any other way.
In large parts of the United States, religion retains a space within our secular society by becoming ornamental. Ornaments can be signs of our own identity, as the Queen may symbolize national identity, or a colonial church may symbolize the history of a town. Ornamental practices can be used to suggest the importance of an event. Ornamentalism, accordingly, refers to a shift in the way in which religious practices and beliefs signify, not to an absence of significance.
Religious practices, of course, always served a variety of functions, apart from invoking God’s presence. Some might argue that these social and psychological functions were all that religion ever did – a proposition that might seem to follow from a denial of God’s existence.  Yet the constitutional protection of religion was not grounded on ornamentalism. Rather, at issue was a need to align a political project with a divinely infused world. Jefferson expresses this in the preamble he wrote to what was to become the Virginia Bill of Religious Freedom:
Almighty God hath created the mind free … all attempts to influence it by temporal punishments or burthens, or by civil incapacitations, tend only to beget habits of hypocrisy and meanness, and are a departure from the plan of the Holy author of our religions, who being Lord both of body and mind, yet chose not to propagate it by coercions on either … [T]o suffer the civil magistrate to intrude his powers into the field of opinion and to restrain the profession or propagation of principles, on the supposition of their ill tendency, is a dangerous fallacy, which at once destroys all religious liberty … truth is great and will prevail if left to herself; … she is the proper and sufficient antagonist to error. 
Jefferson speaks here of opinion, but he does not mean what we mean by that term. For us, opinion is a state of belief that cannot be grounded in truth. We have opinions where there can be no settlement of disputes by proof or scientific inquiry. People have religious beliefs in the same way that they have opinions about politics or about sports teams. It is opinion as an expression of personal belief. No one expects opinions about religion to change based on new evidence concerning the truth of God’s existence. The idea of a “proof” of God’s existence seems to us a confusion of the meaning of the concept of proof.
Jefferson means quite the opposite. Opinions about matters religious are, for him, like opinions in matters of science. Both should be allowed freedom from state coercion in order to work toward the truth of “the laws of nature and of nature’s God.” This is opinion in the classical sense as belief that reflects truth, but does not yet clearly grasp it: opinion is not distinct from knowledge, but the beginning of knowledge. 
Just as Jefferson’s idea of religious belief as opinion informs his idea of the appropriate relationship between state and church, so our idea of religious opinion informs our ideas of that relationship. Thus, ornamentalism has deeply influenced both Free Exercise and Establishment Clause doctrine. The central problem for the former has been to understand the status of a religious claim once religion is understood as opinion and opinion is a matter of subjective belief. The expression of opinion is generally protected by the Free Speech Clause of the First Amendment. Does free exercise add anything to free speech in a world of religious ornamentalism?  This will be my first theme.
Establishment Clause jurisprudence as a form of ornamentalism will be my second theme. If religion is speech without objective referent, then the law of establishment will tend to collapse into the law regulating government support for speech. The issue becomes what opinions can government “endorse.”  Endorsement is a concern in a world of opinions, where truth has no bearing. Ornamentalism takes a vivid form when the government speech is literally ornamental: the Christmas crèches, Chanukah decorations, and public benedictions upon which so many of the modern establishment cases have focused.
My final point is to observe an ironic counterpart to a constitutional jurisprudence of ornamentalism. While this body of law rests on the diminishing importance of religious belief in a secular age, the collapse of religion into opinion leads to an understanding of the religious as simply another interest group. Religious groups enter politics on the same footing as other groups, all of which are defined by the opinions they hold. These groups have now become vigorous participants – and sometimes winners – in the culture wars that characterize contemporary, American politics.
Free Exercise Ornamentalism
Modern free exercise jurisprudence begins with Gobitis, a 1940 case involving the refusal of children of Seventh Day Adventists to participate in the flag salute and pledge of allegiance in their public school.  For this, they were expelled. While the lower courts upheld their free exercise claim, the Supreme Court rejected it with only a single dissent. In 1940, of course, much of the world was at war and the United States was already participating indirectly in the war in Europe. The threat of war fills the opinion. Inculcating patriotism as a ground of national cohesion is elevated by the Court to a matter of national security. Against this compelling state interest, the free exercise claim fails:
National unity is the basis of national security. To deny the legislature the right to select appropriate means for its attainment presents a totally different order of problem from that of the propriety of subordinating the possible ugliness of littered streets to the free expression of opinion through handbills. Compare Schneider v. State of New Jersey. 
The flag salute and pledge arguably serve a double function: as a pedagogic practice of inculcating national unity and as a test to identify those whose patriotism is suspect. The Seventh Day Adventists fail this test, putting in question their relationship to the national project that begins with the security of the state. They are a fringe group located on the periphery of American religious and political practices. The question is whether they should be “indulged,” and the Court’s answer is no.
The Court sees in the case not so much an immediate national security concern, as a deeper competition for the children. The state has no less a claim on these children than do the parents; the children are both citizens and parishioners: “What the school authorities are really asserting is the right to awaken in the child’s mind considerations as to the significance of the flag contrary to those implanted by the parents.”  Just as there is training in religious faith, there is training in civic faith. The children are, especially in 1940, potential soldiers and the state can make sure they are ready to fill this role.
At stake for the government is the inculcation of civic virtues. The flag salute “will best promote in the minds of children who attend the common schools an attachment to the institutions of their country.”  This is, for the Court, not just a ritual of citizenship; it is training in citizenship.  Investing citizenship with a transcendent meaning is the basis of the nation’s security. Quite literally, these children must be willing to sacrifice for the country. There is a direct path from the classroom to the battlefield, and the presence of the flag marks the connection between them. Religion, the Court holds, does not relieve one of the burdens of citizenship.
The Gobitis family resists the claims of civil religion on the grounds of conflict with the biblical commands not to worship other gods or to make craven images. The practice of the pledge appears to them as a form of idolatry. This raises a deeper suspicion that they view the whole civil religion of the nation as a practice of idolatry. The Seventh Day Adventists claim that they live under the ultimate jurisdiction of God, not that of the popular sovereign. For their God, they are quite willing to sacrifice themselves. What we do not know is whether they will sacrifice themselves for the state. Is that sacrifice a form of forbidden worship?
Their willingness to sacrifice for their faith becomes a literal truth immediately after the decision – not, however, on a foreign battlefield. The Seventh Day Adventists are attacked by mobs around the country and become martyrs for their faith.  The members of this faith are viewed by many citizens through a sort of reverse image of Gobitis’s original claim: Declining to worship the flag, they are viewed as idolaters against the civil religion. They are, accordingly, aliens within the state, and there is only a short step from seeing someone as alien to seeing them as enemy. There is nothing ornamental about this confrontation of existential claims grounded on competing beliefs in a transcendent value.
To the eight justices who voted for the outcome, Gobitis seemed a reasonable measure directed at a necessary end. The decision, however, rapidly came to be seen as a mistake. Just two years later, in West Virginia State Board of Education v. Barnette, Gobitis is reversed. Increased intolerance and violence against Seventh Day Adventists was not what the Court had had in mind in Gobitis. Nor was it anything they wanted to be associated with. Something had gone wrong with the pedagogy of patriotism that had been defended in the decision.
Barnette shifts the ground of protection for those who refuse to participate in the pledge. Instead of a free exercise claim, the Court now focuses on free speech. Sectarian faith cannot stand against civil faith within the jurisdiction of the classroom, but free speech can stand against compelled speech. It no longer matters that the objection to the pledge was based on the belief that it was an idolatrous ritual. The offense of the state law is the intrusion on opinion. No one can be made to speak in support of an opinion with which they disagree. “If there is any fixed star in our constitutional constellation, it is that no official, high or petty, can prescribe what shall be orthodox in politics, nationalism, religion or other matters of opinion.”  Religion has become simply one ground of opinion; it has no special privilege. The protection of opinion is part of the protection of speech, which is the “fixed star” of our secular constitutional order.
If Gobitis was about the sovereign character of our civil religion – it can demand a life – then Barnette is about the law of that civil religion. Every faith, religious or political, must translate the experience of the transcendent into a normative practice; every religion has its ethical norms, every state has its law.  The American civil religion begins with belief in the claim of the popular sovereign, but American law begins with freedom of speech. A constitution that protects free speech as a matter of law is grounded on the belief that people of very different opinions can share a common faith in the popular sovereign. In terms of later cases, it hardly follows from the fact that one burns the flag in a moment of political protest that one will not defend it with one’s life. 
With this transformation of a claim for religious free exercise into a claim for free speech, there is also a fundamental shift in the Court’s characterization of our civil religion. That religion is now constructed in opposition to other civil religions – in particular, to the totalitarian practices of the actual enemies of the nation. The ritual of citizenship perceived in Barnette is no longer the pledge to the flag, but the open inquiry of the classroom. Debate has become the site of the American pedagogy of its civil faith. The classroom, the legislative chamber, and the courtroom are all “sacred spaces” under the protection of freedom of speech. The Seventh Day Adventists’ refusal to recite the pledge becomes a learning opportunity, not a sign of treasonous behavior.
For a long time, Barnette is misunderstood: it is read as a free exercise case reversing Gobitis. It is taken to express the idea that religious reasons have a special force in claiming an exception from generally applicable laws.  This was a proposition that had been specifically rejected in Gobitis: “Conscientious scruples have not, in the course of the long struggle for religious toleration, relieved the individual from obedience to a general law not aimed at the promotion or restriction of religious beliefs.”  This proposition was thought to have been overruled by Barnette.
There followed, accordingly, a series of cases in which the reach of a free exercise exception for “conscientious scruples” was tested. The thrust of these cases was that absent a compelling state interest, government must make a reasonable accommodation when its laws impose a substantial burden on religious practice, even when those laws are enacted for reasons that have nothing to do with religion. The cases granting this exception from generally applicable laws became the core of modern free exercise jurisprudence.
Religious exemption cases often arose in the context of application for unemployment benefits. Typically, an employment position required behavior inconsistent with the applicant’s religious beliefs and practices – for example, work on the Sabbath.  The Court generally upheld these free exercise claims for an exception. Thus, those who would not work for religious reasons had a privileged position compared, for example, to those who might have political objections to a particular employer. One has no claim to unemployment benefits, if one quits because of moral or ethical objections to the quality of an employer’s production or if one wants to spend Sunday at home with the family.
We can understand the unemployment cases as analogous to disability cases. The free exercise claim is perceived as a claim that no one should suffer a disability for his or her religious beliefs.  Just as government should accommodate physical disability, it should compensate for religious disability.  This idea of disability has a particular resonance in the employment context: unemployment insurance covers a work-preventing disability. On this reasoning, to be seized by the Holy Ghost is to suffer a sort of seizure. In a secular age, what else could we make of it? As with other forms of disability, we are likely to think that a disabling religious belief is no one’s “fault.”
In these unemployment cases, the Court would say that no one should have to choose between their faith and access to a government benefit. How far this principle would actually extend – beyond these cases – was never made clear. In practice, the limits of the principle reflected an understanding of religion as ornamental. Peripheral religions that reminded us of an America that served as a place of refuge for the religiously persecuted did better than others.  We could express care for such “unfortunate” groups put to the choice between state and faith. As a religious claim looked less like a disability and more like a claim of privilege against the modern state itself, the claim would fail. There was, for example, no sympathy for claims of religious disability with respect to paying taxes, the criminal law, bureaucratic administration, or regulation of the military. 
Particularly illustrative of the Court’s embrace of religious ornamentalism is Yoder, which upheld the right of the Amish not to comply with a law requiring school attendance until the age of 16. The Amish argued that their children should not have to attend beyond the eighth grade. After that, secular education was incompatible with their practices and beliefs.
To the modern reader, this sounds like a claim of right to keep one’s children ignorant. If the state can require everyone to pay their taxes, one would think it could require everyone to attain a minimal level of education. Education and tax policy are arguably the two pillars of the modern state. Yet, the Court holds for the Amish. The explanation, which is very much on the surface of Chief Justice Burger’s opinion for the Court, is the ornamental quality of the Amish community. They are quaint: a remnant of a different America and thus a nostalgic link with the past.
The Amish, of course, are not a threat to contemporary American culture or politics. They do not proselytize; they are themselves being worn down by the attractions of contemporary culture.  They are, for many, little more than a tourist attraction: a kind of living tableau of a premodern community. This ornamental quality is evident when the Court writes:
The record shows that the respondents’ religious beliefs and attitude toward life, family, and home have remained constant – perhaps some would say static – in a period of unparalleled progress in human knowledge generally and great changes in education. The respondents freely concede, and indeed assert as an article of faith, that their religious beliefs and what we would today call “lifestyle” have not altered in fundamentals for centuries. Their way of life in a church-oriented community, separated from the outside world and “worldly” influences, their attachment to nature, and the soil, is a way inherently simple and uncomplicated, albeit difficult to preserve against the pressure to conform. Their rejection of telephones, automobiles, radios, and television, their mode of dress, of speech, their habits of manual work do indeed set them apart from much of contemporary society; these customs are both symbolic and practical. 
The Court goes on to say:
[W]e must not forget that, in the Middle Ages, important values of the civilization of the Western World were preserved by members of religious orders who isolated themselves from all worldly influences against great obstacles. There can be no assumption that today’s majority is “right,” and the Amish and others like them are “wrong.” 
This is religious ornamentalism, for we have indeed forgotten about the Middle Ages and we very much assume that our beliefs are right and the Amish wrong. Who would consult the Amish on how to deal with global warming, infectious diseases, or international trade – if there were even anyone to consult? When was the last time a political leader or government official asked the Amish leadership for their advice?
None of these judicial remarks about the Amish is serious. They are constructing the Amish ornamentally.  Not surprisingly, Yoder’s subsequent history is very limited. It becomes less a case about free exercise claims against government than a case about a general right to home school one’s children. That right is not about preserving religious communities, but about controlling the opinions – or opinion formation – of one’s children. Yoder, in short, becomes another free speech case.
Ornamentalism as the ground for a constitutional doctrine protecting free exercise makes a very weak claim: Why exactly would we be so concerned with the ornamental? Protection of the quaint, the nostalgic, or the symbolic has little to do with the Constitution’s original idea of standing clear of God’s truth. Ornamentalism has no good, principled response to the argument that all opinions should be treated equally. Not surprisingly, this entire branch of free exercise jurisprudence, under which the state must accommodate religious practices and beliefs, is brought to an abrupt halt in Smith. 
Smith involved Native Americans who used peyote as part of their religious rituals. They were convicted of illegal drug use under state law. As a result of that conviction, they were denied state unemployment insurance. Citing Sherbert, Thomas, and Yoder, they claimed that the state could not put them to a choice between their religious practice and the provision of this public benefit. The Court, reading the issue as that of application of the criminal drug laws to religious objectors, concluded that there is no constitutional obligation upon government to make religious exceptions to generally applicable, neutral laws.  As long as the policy was not adopted with the purpose of discriminating against a religious group, the state did not have to accommodate that group’s contrary beliefs any more than it had to accommodate the contrary beliefs of any other group. The religious objector had no privileged claim.
Whatever the virtues of ornamentalism, they were not perceived in Smith. Peyote use by Native Americans came with none of the quaintness that the Court attached to the Amish. Nor was the Court sympathetic to an analogy between drug use and rest on the Sabbath, even though both are religious rituals.  Rather than distinguishing the case from the line of free-exercise precedents, the Court uses it to reevaluate those cases. Failing to see anything special in religious opinions, the Smith Court returns to the non-accommodationist position of Gobitis. Barnette is now read as a free speech case, leaving Gobitis’s free exercise position standing: “We have never held that an individual’s religious beliefs excuse him from compliance with an otherwise valid law prohibiting conduct that the State is free to regulate. On the contrary, the record of more than a century of our free exercise jurisprudence contradicts that proposition.” Thomas and Sherbert are limited to their facts; Yoder is read as a case that rests on parental rights to educate their children – a right grounded not in religion but speech. 
With Smith, a constitutional right to free exercise against laws of general applicability extends just so far as free speech extends. Religion in a secular age has become only a source of opinion. Free Exercise ornamentalism proved unstable because it views religion as opinion. If religion is only opinion protected as speech, its natural place is within, not outside of, politics. Smith explicitly tells religious groups that they must make their case in the fora of politics.  They did just that.
In a secular age, religion becomes opinion. Not opinion as a step on the way to truth, but opinion as a proposition that has no truth independent of the subjectivity of the person whose opinion it is. In the sciences, a proposition that cannot be falsified is one that is literally meaningless. The idea that an opinion stripped of an objective referent can nevertheless have value is a thoroughly modern idea. It is an idea that emerged first with respect to politics in the form of public opinion. The Court expresses this idea in Barnette:
There is no mysticism in the American concept of the State or of the nature or origin of its authority. We set up government by consent of the governed … Authority here is to be controlled by public opinion, not public opinion by authority. 
Accordingly, public opinion is understood as the foundation of the political order.  Its authority has nothing to do with whether it is “true.” It is right because it is believed.
The discovery of public opinion as a source of value inaccessible, and indeed incomprehensible, to science begins with politics, but spreads from there to religion. Constitutional provisions that were founded on belief in the supervening reality of God come to be read as if they protect opinions that can make no contact with any referent at all. In the 18th and 19th centuries, history was understood as progress from the primitive to the civilized. The religion of the civilized was Christianity.  History moved from false to true opinions; the truth of religious opinion was Christianity. These attitudes toward progress, opinion, and religious truth no longer exist. Religion has become something like an expression of taste. Opinions severed from any referent stand to each other as acts of speech. This is the “hermeneutic turn” in the jurisprudence of the religion clauses.
Establishment clause claims will now be assessed through the prism of government speech: What constitutional limits extend to government speech or government support of speech? Framing the question this way enables us to see why so much of modern Establishment Clause jurisprudence turns on the issue of endorsement. Endorsement steps into the place of truth in a world of opinion. Opinions gain strength not by virtue of their relationship to truth, but by virtue of their endorsement. 
Representative of this approach to the Establishment Clause jurisprudence is Rosenberger v. University of Virginia, which involved a student group seeking access to a campus activity fund to pay for publication of their newsletter.  The purpose of the publication was “to challenge Christians to live, in word and deed, according to the faith they proclaim and to encourage students to consider what a personal relationship with Jesus Christ means.”  The University declined payment, citing Establishment Clause concerns with spending public funds on religious activity. The Court rejected the state’s justification, holding that the student group was the victim of view-point discrimination by the university. Religious speech, including appeals to a personal relationship with Christ, has become simply speech – the expression of an opinion. The state cannot distinguish among the opinions of students by reference to their viewpoints; it especially has no ground to treat differently those opinions that express religious beliefs.
In Rosenberger, the issue is not what the state is actually doing – funding a religious publication – but what it appears to be saying. The Court recognizes that the state has a cognizable interest in not appearing to endorse the religious expression of the students. Under the circumstances, however, the Court concludes there is little likelihood of such an appearance. After all, the state subsidizes the activities of many different student groups with many different messages. Despite the funding, there is a sufficient gap between the student expression of opinion and the university to make any inference of endorsement unlikely.
Again, we see that there is nothing special about religious speech, even when it seeks a subsidy from the state. Instead of requiring the state to “steer clear” of such involvement as a matter of distinct jurisdictions of church and state, or even permitting the state to steer clear, the state must treat religious expression as simply more student speech. Whatever the wall of separation between religion and state was once imagined to be, it no longer prohibits – and may require – state funding of some religious activities. This is not because we take religion more seriously than in the past, but because we take it less seriously. It is just one opinion among many. It’s a matter of taste, choice, self-expression.
Rosenberger is a case about the actual payment of state funds.  The state argued that state funding is especially problematic, citing Madison’s argument that no person should ever be taxed to pay the expenses of a minister, regardless of whether it is someone of their own faith.  The modern Court rejects this argument, relying upon a series of cases that involved religious groups using state facilities. Religious speech was once seen as part of religious practice. In these cases, religious practice is more likely to be seen as speech that happens to be by religious groups.
For example, in Lamb’s Chapel, a local evangelical church wanted to use public school facilities on a Sunday to screen films about parenting from a Christian perspective.  The school declined to permit this, arguing that it was not engaged in viewpoint discrimination since it did not allow any religious speech in the facility. The Court rejects this argument, holding that the subject matter of the speech at issue was family, not religion. Accordingly, the state was discriminating against the expression of a religious viewpoint on family matters.
Since speakers always draw their positions on any particular issue from their broader beliefs, commitments, and understandings – including religious beliefs – it cannot be the case that religious groups are prohibited from engaging on public issues such as parenting. No doubt for this reason, the decision in Lamb’s Chapel was not only unanimous, but written by Justice White. White had dissented several years earlier in Widmar v. Vincent, which held that a university could not prohibit a student religious group from using school facilities for religious worship, once it had opened those facilities for use by student groups.  The Widmar opinion makes explicit the identification of religious worship and speech: “UMKC has discriminated against student groups and speakers based on their desire to use a generally open forum to engage in religious worship and discussion. These are forms of speech and association protected by the First Amendment.”  White objected to this identification of speech and worship:
A large part of respondents’ argument, accepted by the court below and accepted by the majority, is founded on the proposition that, because religious worship uses speech, it is protected by the Free Speech Clause of the First Amendment. Not only is it protected, they argue, but religious worship qua speech is not different from any other variety of protected speech as a matter of constitutional principle. I believe that this proposition is plainly wrong. Were it right, the Religion Clauses would be emptied of any independent meaning in circumstances in which religious practice took the form of speech. 
Juxtaposing Widmar and Lamb’s Chapel suggests just how difficult distinguishing religious worship from speech by religious groups can be: What exactly are calls for Christian parenting? Does it matter if the call is made from the pulpit or in a schoolroom? Does the identity of the speaker matter? Even in his dissent, White had recognized that the line would not be easy to draw. 
White was thinking particularly of the school prayer cases. If prayer is just speech, how is school prayer different from a civics class? The direction of the jurisprudence will, in fact, be to diminish that distance.  Once we begin down a road that sees worship as speech, there may be no stopping the movement toward collapse of any distinction.
The claim that worship is speech would have struck most people as a distinctly odd proposition until quite recently. It becomes convincing only in a secular age in which religious speech can have no referent. Worship has always been conducted partly in speech, but the point was neither to express an opinion nor to convey ordinary information. Speech was a calling forth and a calling upon: worshippers were called forth into the presence of God and God was called upon to extend His grace. Worship created and sustained a relationship between God and man. Sustaining this relationship was the ground and ultimate purpose of a church.
In this practice of worship, there was no reason to privilege or distinguish the moment of speech. There is, for example, a long tradition of silent worship. We miss the point, if we think that the parishioners are exercising a “right not to speak.” Similarly, it makes little sense to privilege the words of the Eucharist over the drinking of the wine and the eating of the bread: all are symbolic expressions. Religious speech functions within an overall ritual. Reading the Torah in the library and reading it in a service are not the same even though the spoken words may be the same. Yet, it has become increasingly difficult to maintain the distinction, apart from pointing to the ornamental character of the latter. We cannot say that God is present in one place, but not the other.
To create a forum for public expression of opinion and a mechanism for decision among opinions is the guiding idea of constitutional republicanism. As long as God was understood as a fact, not a matter of opinion, there were reasons for separation on both sides. Supervening truth claims could undermine the exchange of political opinions; the truth of religious claims is not a subject to be decided by counting opinions.  In political life, opinions are to stand on their own in a competition of persuasion; they cannot claim to rest on divine truth. A church service is not a forum for the exchange of opinions; a political debate is not a site of worship. Or, so it was thought until the rise of secularism in the late twentieth century.
When the religiously motivated intervene in public debates, are they participating in the political exchange of opinions or are they proselytizing? Those two things look the same once we come to believe that the latter is nothing beyond the expression of opinion. Once God has permanently withdrawn from our everyday world, we are left with speech that has no ground except as opinion. Pulpit and podium become dual fora for the expression of opinion. Instead of truth, we argue about endorsement. This is the context within which religious ornamentalism thrives.
Many of the endorsement cases demonstrate a literal ornamentalism as the Court considered challenges to such things as Christmas decorations, park displays, wall hangings, invocations, and benedictions. In the end, very few of the challenges succeed.  These ornaments are often so peripheral to the practices of an actual faith that the Justices find it hard to think that much turns on them – apart from the symbolism of separation itself. The Court deploys a fact sensitive test: It regularly concludes that no one accustomed to our social practices could think that the state was endorsing sectarian faith on the basis of these ornaments. 
Given the religious genealogy of so many of our political rituals, including the opening words of the Court’s own sessions and the closing words of every presidential address, endorsement is inevitably a high standard to meet. The Court is not about to declare our currency (“In God we trust”) or the Pledge of Allegiance (“one nation under God”) unconstitutional. For this reason, the measure of unconstitutional endorsement tends to move toward coercion. Just as free exercise becomes free speech, establishment becomes coerced speech. Schools, in particular are rank with coercion. They have, therefore, become one of the few places in which the Court has actually found endorsement.  Even there, when government aid moves through third party choices or is neutral as between religious and non-religious schools, claims of endorsement generally fail. 
The difficulty of finding endorsement in these ornamental religious practices is well illustrated in the recent case of Town of Greece.  Town practice was to invite local ministers to open town meetings with a prayer. Since the inception of the program, the ministers had been almost exclusively Christian. The prayers were often explicitly sectarian, appealing, for example, for the blessings of Jesus. Ministers asked those present to participate, whether by standing or bowing their heads. The ministers spoke from the front of the room and addressed the public as if it were a congregation. The people in attendance were those who had business to conduct before the town council, students there to study civics, and otherwise engaged citizens. After the prayer, the minister was awarded a commemorative certificate by the town supervisor.
The Court upheld this practice, despite its seeming explicit endorsement of religion: one of its purposes, after all, was to “invoke divine guidance.” It was, the Court held, part of our tradition to begin public deliberations with a prayer, even a sectarian prayer. The opinion emphasizes tradition, not belief. Ornamentalism is at stake, not divine providence. Indeed, one of the reasons to uphold the practice is that no one takes it too seriously. When asked to diversify ministers, the town found, as one of its four non-Christians, a “Wiccan priestess.” As in politics, where opinions are protected even when they are frivolous, religious ornamentalism can look silly when it is not quaint. The Court goes to some pains to describe a situation of rather lax attention and disregard – people coming and going – when the prayer is actually delivered.
Respondents complained that the practice was an endorsement that made them feel as if they were not full members of the community.  The Court is wholly unsympathetic:
In their declarations in the trial court, respondents stated that the prayers gave them offense and made them feel excluded and disrespected. Offense, however, does not equate to coercion. Adults often encounter speech they find disagreeable; and an Establishment Clause violation is not made out any time a person experiences a sense of affront from the expression of contrary religious views in a legislative forum, especially where, as here, any member of the public is welcome in turn to offer an invocation reflecting his or her own convictions. 
All of the elements of modern Establishment Clause jurisprudence are here. Worship is just speech; prayers to Jesus are only “disagreeable” speech. Government often speaks in ways that some citizens find disagreeable. That offense is taken is not a constitutional violation. A constitutional violation requires coercion, which means that speech alone is never an establishment. Of course, outside of the schools, coercive prayer is not likely to be on the agenda.
Ornamentalism and Interest Groups
If the Establishment Clause is not much of a restraint on government policy, then religious groups suffer no special disability when they enter the political fray. If the farm lobby can pursue government benefits through political coalition building, why not a religious group? On most issues, there is simply no way to prevent this. Religious groups have positions regarding health care, social welfare, the environment, education, and virtually everything else that comes up for consideration in the modern state. Religiously inclined voters will pursue those interests in their political choices and in their messages to government actors and candidates. This has always been the case. No one believes that the religiously based motives of voters can or should be removed from our political life. Religious belief motivated many abolitionists, for example, as well as those in the temperance movement a few decades later. A successful candidate is not rendered illegitimate if she owes her victory to the support of Catholics or Jews.
Religious groups not only seek positive benefits from government programs; they also seek exceptions from otherwise applicable requirements. This was at issue in Hobby Lobby, where the Court had to interpret the meaning of the Religious Freedom and Restoration Act (RFRA) as applied to a claim of exemption from some aspects of employer-mandated health insurance coverage.  Hobby Lobby, a corporation, argued that the health care law violated its owners’ religious convictions by requiring coverage of forms of contraception that they believed to be indistinguishable from abortions. The Court reads RFRA to support this claim for an exception from these otherwise applicable general requirements.
What had once been a constitutional issue of free exercise has now become an issue of statutory interpretation. RFRA was passed in response to Smith. Indeed, at the conclusion of Smith, the Court had said that politics, not the Free Exercise Clause, must be the source of any religious exemption from generally applicable laws. Politics, it turned out, was very favorably inclined toward religious groups. RFRA passed with virtually no opposition in the House or Senate. The express intent of the legislation was to return to the pre-Smith jurisprudence of balancing religious claims against government interests. It applied to both federal and state laws. The examples Congress had in mind when passing the Act were ornamental: Orthodox Jewish and Hmong objections to state autopsy requirements and zoning regulations that covered churches and synagogues.  These are religious practices at the fringe of the secular order: edifices and funeral rites.
When RFRA first came before the Court for review of its application to state laws, it was approached with unrelenting hostility. Congress had passed the statute claiming to act under Section 5 of the Fourteenth Amendment, which gives Congress the power to enforce the substantive provisions of the Amendment, including the due process clause.  The Court held that, with respect to free exercise claims against neutral rules of general application, there are no rights to be enforced, precisely because Smith was a complete statement of the constitutional right to free exercise. The Court aggressively defended its authority “to say what the law is.” It wholly rejected the idea that because Congress was explicitly granted the power to enforce the Amendment, it had the power reasonably to define the scope of the due-process right that it intended to enforce. The reasonableness of Congress’s definition, after all, followed from the fact that RFRA used the Court’s own standard prior to the reversal in Smith. 
Without the Fourteenth Amendment foundation, RFRA survives only as a self-imposed limit on federal regulations – an act of legislative grace or charity toward religious objectors. Hobby Lobby must decide, then, how far that charity extends. While earlier the Court had demonstrated an unremitting hostility toward the Act, Hobby Lobby takes just the opposite approach. It reads the Act as broadly as possible, as if Congress intended no limits at all on the restraints it was imposing on itself when a person – a category that includes corporations – claims that a federal policy interferes with religious practice or belief.
The Court rejects any suggestion that caution is appropriate in dealing with claims at the intersection of religion and government. It rejects arguments grounded in the third-party interests of the employees whose access to health insurance is at stake, and who are unlikely to share the religious beliefs of their employers. It rejects arguments that try to differentiate the concerns of a marketplace from those of a religious practice. It suggests no limits on the kinds of claims that may be made under RFRA or on the “artificial” persons – corporations – that may make them. It goes so far as to suggest that there can be no adequate countervailing interest on the government’s side because government can always assume the cost and practical burden of providing the controverted health benefit itself – an argument that would apply to virtually every conceivable religious claim for an exception from a regulation. Never before had the proposition seriously been made that a religious claim could require government to create and fund an alternative program. Congress had suggested no such intention.
All of this is a far cry from the sort of ornamental practices that had been the focus of congressional attention when it passed RFRA. Hobby Lobby is a major, national corporation with some 13,000 employees. The Affordable Care Act was an effort to move toward universal health insurance and it targeted large corporations in particular. Indeed, when the Act was before the Senate, an amendment was rejected that would have created a religiously based exemption to its requirements.  No one suggested that the amendment was not needed because RFRA had already done the work. The Court reads RFRA as if the religious groups that lobbied for the Act were wildly successful, winning an across the board change in federal policy extending well beyond a limited exception for the out-of-the-ordinary practices of marginal, religious groups. To all the warnings that this reading will lead to unprecedented interventions in government policies, the Justices respond that the Court is acting simply as the agent of congressional policy.
The Court reads RFRA as if Congress turned over to the courts all government policy that affects religious claimants. Under Hobby Lobby, the question for a court will be whether government can achieve its ends by assuming the regulatory burden itself. Accommodating religious beliefs becomes supervening government policy in every field of action. Instead of occupying the ornamental periphery, religious claims are treated as if they are as central to governance as federalism claims. Just as the federal government cannot generally commandeer state agents to carry out federal policy – they must do it themselves – now the religious objector cannot be forced to carry out federal policy.  The federal government must do it itself.
It is unlikely that anyone in Congress imagined that this was what they were doing when they passed RFRA. That does not, in itself, mean that this interpretation of the Act is wrong, for Congress often does not imagine the full reach of the laws it passes. It is wrong because it is an entirely implausible reading of government policy toward religious claims and an implausible reading of what the government was doing when it acted to relieve marginal groups of some of the burdens of the modern state. Society’s interest in the dead body upon which an autopsy might be performed is simply not the same sort of thing as society’s interest in universal health coverage for the very live bodies of company employees.
Conclusion: From Ornamentalism to the Culture Wars
Whatever religious groups lost in Smith was more than made up in Hobby Lobby. What change between the two cases can account for this radical change in fortune for religious groups? In a phrase, it was the rise of the culture wars as the organizing fact of American politics. Once religious groups are viewed as just another politically active interest group, religious claims can easily become tactics within the larger culture wars. Adjudication is doing the work of political warfare.
One sees this in Hobby Lobby when the Court rejects the argument that the burden on religious practice here cannot possibly meet the statutory test of being “substantial.”  The company objected to a minor element within a much larger insurance schedule. Even with respect to that element, its own involvement was as indirect as possible. Whether the contraceptive practices at issue would ever be pursued would depend on decisions by employees and their doctors. The company’s objection was the private equivalent of objecting to paying some portion of one’s taxes because of religious objections as to how that money might be used.  The latter claim has never succeeded. The tail now wags the dog, because the case had become a vehicle for conservative opposition to the social welfare state.
From the point of view of actual religious liberty, there is something not just odd but lacking in seriousness about Hobby Lobby. The pressing religious liberty issue is precisely the opposite of that which the Court recognizes. The serious claim is that of the employees who are being coerced to accept a health care plan limited by their employers’ religious beliefs.  They are being compelled by a structure of economic power to behave in ways determined by the faith of another. The focus should be on the religious coercion. This is especially true today when so much of private economic power is made possible by public policies and regulations. If we do not, and should not, let employers discriminate against employees on the basis of their religion, why should employee access to government mandated health benefits turn on religious beliefs they do not share? How can a liberal government give such coercive weight to anyone’s faith? It is hardly an answer to say that government can provide the benefit itself, for it is unimaginable that it could do so across the full range of possible religious objections.
Tolerance of religious ornamentalism had not included intrusions on dependent third parties.  Yet, one aspect of the culture wars is to shield more and more of the “private” from public regulation. Through RFRA, religious claims turn out to be one more method of erecting that shield. Thus, the religious claims of corporations are parallel to the speech claims of corporations upheld in Citizens United.  A corporation that engages in protected political speech can now also engage in protected religious activity. The problem of the concentration of power – political and economic – disappears from view. In Hobby Lobby a corporation operating on a national scale with thousands of employees is treated as if it is no different from a corner butcher store that wants to remain open on Sunday.  To ignore the effects of power has little to do with religion, but everything to do with the raging culture wars that are the site of a genuine political contest.
The Constitution of 1787 made provisions for two private practices of power that stood outside its understanding of a modern state founded on principles of reason and consent: slavery and religion. Many in the antebellum period argued that the Constitution had put slavery “on a course of natural extinction,” but that changes in the slave economy undermined that public purpose. It took a war to end the practice. In the latter half of the twentieth century, many might have argued that the Constitution also put religion on a course of natural extinction – not intentionally, but practically. Stripped of the support of the state, this would be religion’s fate in a secular age. That, of course, did not happen. Today, people write of a “post-secular” age. 
Ornamentalism treats religion as both marginal and quaint. There is a substantial minority in America that feels that the movement of society has been toward treating them as marginal and quaint. Remarkably, these are often the descendants, culturally and geographically, of the southern interests in the Civil War. Again, they feel as if they are “losing their country.” Cosmopolitanism has stepped into the place of industrialization, but the disruption to a social order of regionalism, anti-intellectualism, and economic hierarchy is strikingly similar. In the nineteenth century, southern populism was linked to slave interests. Today, that populism is linked to corporate interests. In both the hard and the soft war, religious groups have played a supporting role.  These are the odd alliances of American politics. We have been fighting and refighting the Civil War for well over 100 years. We still do not know who the winners are.
In the post-War years of the twentieth century, American law moved toward an understanding of religion as speech and of religious groups as interest groups. By the end of the millennium, religious claims had become yet another vehicle for and site of our unconstrained culture war. Religious ornaments became weapons: a public prayer was a symbolic act in the same way that flying a Confederate flag was a symbolic act; a claim of religious liberty was like a claim of a right-to-bear-arms.  Adjudication became only another front in a multi-front war – a process most vividly on display in our long battle over abortion. 
Hobby Lobby is a reminder that the Court too is a participant in this modern civil war. In that, too, our soft war is following the path of the earlier hard war. Hobby Lobby is hardly as morally repellant as Dred Scott, but someday we may look back and conclude that once again the Court was intervening in a political dispute in which it did not belong.
© 2016 Walter de Gruyter GmbH, Berlin/Boston