Oxford Research Encyclopedia of Politics reached a major milestone by publishing our 1000th article! For more information visit our News page.

Show Summary Details

Page of

PRINTED FROM the OXFORD RESEARCH ENCYCLOPEDIA, POLITICS (oxfordre.com/politics). (c) Oxford University Press USA, 2020. All Rights Reserved. Personal use only; commercial use is strictly prohibited (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 10 August 2020

The Intersection of LGBT Rights and Religious Beliefs in the United States

Summary and Keywords

Tension has long existed in the United State between the equality claims of LGBT individuals, on the one hand, and free exercise claims by those who hold that compelling equal treatment violates their convictions, on the other. This tension increased, however, after the United States Supreme Court extended marriage equality to same-sex couples nationwide. Equality advocates hold that antidiscrimination laws simply allow LGBT individuals to enjoy the same rights as others. Many religious advocates, however, believe that they are being prohibited from living out the implications of their conscientious beliefs. Neutrality between these conflicting claims cannot be achieved, as policies that appear neutral to one group appear non-neutral to the other.

Private voluntary organizations are one site of conflict. Although private organizations should not typically be forced to reflect the values of the larger society, not all organizations are similarly situated within it. Groups such as the Boy Scouts should be able to exclude at will. Public authority does not itself always support the values of free and equal citizenship, and organizations may evolve over time as the Scouts itself has done. Organizations that exist within larger entities, however, fall into a different category. The Supreme Court was correct to uphold Hastings Law School in forcing the Christian Legal Society as a registered student organization to admit all comers. These groups also represent the values of a public entity and can continue to operate as independent entities if they so choose.

The provision of services in connection with same-sex weddings and commitment ceremonies has been another site of conflict. In Craig v. Masterpiece Cakeshop (2015), the Supreme Court found narrowly that bakery owner Phillips could refuse to create cakes for same-sex wedding celebrations, as the state of Colorado had displayed animus toward Phillips’s religious beliefs. Commercial establishments, however, are public accommodations and generally should not be allowed to discriminate against customers on the basis of their identities. Discrimination against the activity or conduct of formal commitment is also discrimination against the identity or status of a same-sex couple. These kinds of cases do not admit of neutral solutions. Some suggest that those with religious reservations could advertise that they do not serve same-sex couples, but this is reminiscent of Jim Crow in the post–Civil War South.

Jurisdictional pluralists suggest that the government designate a sphere of noninterference as a jurisdictional boundary that it will not cross. Thus individuals and associations with religious commitments would be free to pursue these interests with minimal interference. However, a prior authoritative structure must exist to define the nature and scope of this jurisdiction, just as the Constitution defines the relationships between the national government and the states. Applications for religious exemptions should not be treated more generously when they conflict with LGBT equality concerns than with equality claims based on race or gender. Although religious individuals and groups should be able to exercise their religious convictions within their areas of competence, in a liberal society and state they cannot define the limits of these areas.

Keywords: LGBT equality rights, religious exercise rights, neutrality, Boy Scouts, Christian Legal Society, Masterpiece Cakeshop, equal citizenship, LGBT politics


Tension has long existed in the United States between equality claims by LGBT (lesbian, gay, bisexual, transgender) individuals, on the one hand, and free exercise claims by those who hold that compelling equal treatment violates their religious convictions, on the other. This tension increased, however, after the Supreme Court extended marriage equality for same-sex couples nationwide (Obergefell v. Hodges, 2015). Equality advocates hold that antidiscrimination laws simply allow LGBT individuals to enjoy the same rights as others. Many religious advocates, however, believe that they are stigmatized for attempting to live out the implications of their conscientious beliefs (Caramagno, 2002, p. 193; see pp. 190–197). For the government, protection for one position appears to disrespect the opposite position. Although the government could achieve neutrality by refusing to protect either sexual orientation and gender identity or religious belief in this context, this solution would allow discrimination by individuals and entities of both viewpoints.

The First Amendment to the United States Constitution both forbids the establishment of religion and guarantees its free exercise. Yet the principles of non-establishment and free exercise do not apply to sexual orientation and gender identify. Non-establishment dictates that the nature of American laws and institutions should not be dictated by religious beliefs. Yet the free exercise guarantee suggests that both religious and nonreligious citizens with strong moral convictions are entitled not only to live in accordance with their convictions, absent harm to others, but also to shape civil institutions to reflect their sincere beliefs.

Until the early part of the 21st century, heteronormativity, or the assumption that most people are heterosexual, was both the social and the legal norm, resulting in a type of establishment. Gains made by those with nontraditional sexual orientations and gender identities constitute a disestablishment. In the case of a majority religion, disestablishment means greater free exercise opportunities for adherents of minority religions. Regarding sexual orientation and gender identity, however, traditionalists often believe that the elevation of these competing “faiths” through equal access to an institution as central as marriage constitutes a hostile and competing establishment. It does not merely exist alongside the traditional institution. Rather, it disadvantages traditionalists in their own religious practices when it prevents their living out the implications of their own beliefs and moral teachings, teachings that they perceive as superior to those of others. For LGBT individuals, on the other hand, traditionalists’ attempts to prevent their access to institutions and accommodations on the same terms as others relegates them to a status of civic inequality, or to a discriminatory exclusion that distinguishes between favored and disfavored classes of citizens (Gutmann, 2003, p. 97). To LGBT individuals and their allies, sexual orientation and gender identity as well as religious affiliation should be respected, not necessarily because they are unchosen, but because they are a central facet of personal identity (Gill, 2001, pp. 175–217).

Several issue areas have become prominent regarding these sorts of conflicts. None of these are amenable to truly neutral dispositions. Voluntary associations such as student groups and community organizations like the Boy Scouts of America have been embroiled in controversies regarding the inclusiveness of their membership. More salient are conflicts surrounding the inception of marriage equality, as in the rights and responsibilities of private businesses that supply services connected with same-sex weddings or commitment ceremonies. Those whose stances stem from traditional religious or ethical beliefs are those most likely to question LGBT claims to equal treatment. If the state is supposed to be neutral toward religion and its exercise, for religious traditionalists it should not be discouraging or prohibiting practices that flow from sincere religious commitments. If, however, the state is to honor equality claims to ensure that all citizens stand similarly in their relationships to the state and to one another, it should not uphold religious claims that threaten to undermine this status. These conflicts are often the equivalent of zero sum games: one side wins while the other side loses.

The Neutrality Conundrum

Many traditionalists reject the idea of racial discrimination hypothetically grounded on religious belief but justify a degree of similar discrimination toward LGBT individuals, suggesting that the latter have not suffered through slavery, a civil war, thousands of deaths, and decades of further struggle. Clear religious teachings against same-sex intimacy exist, they suggest, that do not have parallels in teachings supporting subordination on the basis of race (Laycock, 2016, pp. 252–253). Others, however, point to white Americans, with or without slaves, who believed that biblical references to slavery meant that it was divinely sanctioned (Griffith, 2017, pp. 85–87) and that post-emancipation segregation was justified (Babst & Compton, 2016, p. 101). If racists or segregationists had received religious exemptions from civil rights laws, these regulations would have been toothless in many instances (Koppelman, 2015, pp. 642, 653).

Moreover, although some believe that white supremacy was arguably not a religious teaching (Laycock, 2016, pp. 252–253), it seems clear that both white supremacy, preserved in part through anti-miscegenation laws, and heterosexism, or the privileging of straight over LGBT people, have been ways of life supported by complex systems of belief. Where Loving v. Virginia (1967), by legitimizing interracial marriage, overturned the preservation of a caste system that enshrined white supremacy, Obergefell, by legitimizing same-sex marriage, overturned a caste system that has privileged heterosexuality over non-heterosexuality. Each hierarchy targets one group of people as inferior to another (Gill, 2012, pp. 18–25, 60–65). If one is rooted in religious principles while the other was perhaps less so, both have been destructive. It is questionable, to say the least, for a society to pull its punches in aiding one group because its members may not have suffered as much or as long as the members of a different group.

These observations do not yet, however, address the difficulty of achieving governmental neutrality in its response toward both the equality claims of LGBT individuals and the free exercise claims of religious traditionalists. To begin with, no state can be neutral across the board, as neutrality only carries meaning when it can be measured in terms of some independent standard. Even polities such as the liberal state that strive for neutrality on some definition must espouse public policies that do not appear neutral to all. Policies that appear neutral to one group, such as the protection of LGBT equality claims, appear non-neutral to another, such as religious traditionalists who hold that they are then prevented from living out the practices that flow from their religious beliefs. One’s definition of injury or harm depends on one’s beliefs and perspectives and is therefore problematic. “The civil discourse of facticity,” or civil discourse, is characterized by conflicting interpretations of what kinds of conditions result in “the sort of ‘harm’ appropriately subject to political jurisdiction” (McClure, 1990, p. 383; see pp. 382–384).

On one interpretation, moral neutrality is achieved when legal norms reflect the conviction that a particular way of life harms neither its practitioners nor those exposed to them. In the LGBT context, the government implies that same-sex relationships are neither good nor bad. Civil rights laws that prohibit discrimination against these individuals, however, convey an additional moral assessment that therefore private parties are unjustified in engaging in discrimination against LGBT individuals (Feldblum, 2008, p. 131). Policies toward religious practice implicitly take this approach also. Although government protection against religious discrimination does not take a stance on the moral status of religion itself, it does suggest a prior moral assessment that religious practice generally is not harmful and that therefore discrimination on the basis of religion is wrong. On what might be termed a first-order level, this approach is neutral: it does not prefer one religion or sexual orientation to another. On a second-order level, however, it is non-neutral. That is, it is not neutral between the view that a particular religious or sexual orientation is superior and uniquely worthy of protection, on the one hand, and the view that all are on an equal plane and thus all eligible for antidiscrimination protection, on the other.

The absence of antidiscrimination laws in the face of discrimination, however, also sends a message about morality. The government is implying that a minority sexual orientation—or gender identity—does not possess a neutral valence but is morally questionable (Feldblum, 2008, pp. 131–133). Similarly, if the record shows that some religions are traditionally despised and the government does nothing in the face of majority sentiment, it has taken a moral position. Although some private actors will always view particular orientations or religions as problematic, the government’s failure to act signals that public authority itself perceives these orientations or religions as problematic. The liberal state must step up to protect ways of life within the range deemed not morally problematic according to its own prior moral assessment. For example, if the Supreme Court had not struck down the Colorado constitutional amendment forbidding antidiscrimination laws concerning housing and employment based on sexual orientation in Romer v. Evans, this inaction would have constituted a moral assessment that revealing a minority sexual orientation was problematic and should therefore be discouraged. Tension between LGBT equality claims and religious exercise claims is unavoidable, and therefore neutrality is unachievable. That is, each side in such disputes possesses a different conceptualization of what it conceives to be the proper relationship between religion and the state and therefore holds a different vision of neutrality.

Because no state can be neutral across the board, liberalism and the neutrality of broad value pluralism can be incompatible. The practice of some types of diversity can conflict with other kinds of diversity, and thus a defense of the liberal society and state cannot rest on hospitality to diversity across the board. It must instead be grounded in the particular range of values or preferences that a given interpretation of liberalism puts forward. This interpretation, moreover, must be civilly determined by the government—albeit with democratic input—rather than determined by the imperatives of religious belief (McClure, 1990, pp. 373–381; see also Gill, 2012, pp. 34–48; Neal, 1997, pp. 28, 34–47). Religious believers disagree among themselves about the legitimacy of some manifestations of LGBT equality rights. More broadly, the American constitutional commitment to civic equality may, depending on circumstance, require not only a qualified noninterference with religious belief and practice, but also public action that may prevent the creation of two classes of citizens.

Private Voluntary Organizations

According to the logic of congruence (Rosenblum, 1998), the health of liberal democracy requires voluntary associations to reflect democratic principles in their own organizations and internal life. All voluntary organizations would be inclusive, so that anyone who wanted to join or to remain a member could do so. These groups, however, might then lose their defining characteristics and their messages as expressive associations. For others, freedom of association does not mean the liberty to enter associations of one’s choice, but rather the freedom to leave these organizations and perhaps to form new ones. This approach protects both the consciences of those who remain and also of those who leave to form new associations (Kukathas, 2003). It is crucial in this context, however, that other opportunities for association exist, or the slack without which freedom to leave is not meaningful. Two contrasting kinds of cases may illustrate this point concerning sexual orientation and gender identity. Despite attempts at fairness, in both areas there is always a loser as there can be no neutral solution.

Associations in the Larger Community

After the New Jersey Supreme Court had ruled that the Boy Scouts of America was a public accommodation under state law and therefore could not discriminate against gay scouts, the United States Supreme Court in 2000 overruled that decision in Boy Scouts of America v. Dale to allow the Boy Scouts to expel James Dale, an openly gay scoutmaster, on grounds of associational freedom of expression. Although the Scouts had said little publicly about sexual orientation, the Scout Oath had long affirmed a belief in God. Like religious organizations, the Scouts could defend their own beliefs even absent prior publicity. Eventually, the Scouts changed its policy first to permit openly gay boys to be members, later to allow gay scoutmasters, and finally to accept transgender boys. Is this sort of organization a public accommodation that must therefore be open to all comers, or is it a private association that can establish its criteria for membership without interference?

Regarding theism, the Boy Scouts was clearly an expressive association, and its theistic orientation was clear and public. Its status was murkier regarding sexual orientation. Although the Scout Oath contained a promise to be “morally straight,” this quality was defined as requiring strong character, honesty, and justice. Only when controversy regarding sexual orientation arose did the Scouts claim that this meant that boys should be straight rather than gay. Meanwhile the organization operated with a de facto “don’t ask, don’t tell” policy. Sexuality issues were not discussed in scouting programs; Dale’s orientation was revealed only when a newspaper article revealed that he was a gay activist. Nevertheless, the Supreme Court ruled that forcing the Scouts to accept or retain members it did not desire infringed on its freedom of expressive association. The fact that Scout officials interpreted “morally straight” to exclude known gays was reason enough to allow their exclusion. Moreover, the existence of disagreements on this subject within the Scouts did not preclude the legitimacy of an official position.

The court majority in Boy Scouts v. Dale seems correct, given certain caveats. In a liberal polity, like-minded individuals should be able to associate to express and practice their beliefs in privately funded organizations, religious or nonreligious, without being forced to accept those who disagree. This is especially true for those whose beliefs may be out of favor in mainstream organizations, as the Scouts’ exclusivity increasingly became. A society that requires all organizations to be equally diverse can be a society without diversity (Alexander, 2008, p. 15). A major qualification exists, however. The liberty to leave organizations and to form new ones, especially when one is forced to leave, depends on a society that affords other opportunities for association. Otherwise, exclusion may be punitive. If enough voluntary organizations were exclusive in the same way—excluding agnostics, atheists, or LGBT individuals honest about their status—the combined effect would drastically curtail the breadth of the larger forum within which individuals exercise their freedom, thus compromising diversity in a different way. The combined effect of parallel criteria for exclusion by a majority of voluntary associations would amount to a public expression of civic inequality (Gutmann, 2003, p. 97). Like the regime of Jim Crow in the post–Civil War South, it would undermine the equal standing of citizens in their relationships both to the state and to one another.

Some commentators hold that the state itself should not even attempt to be neutral, but should instead engage in democratic persuasion through its policies by expressing or sometimes by enforcing the public values of freedom and equality that underlie the right to freedom of association, expression, and conscience in the first place. According to what Corey Brettschneider terms value democracy, the state should engage in deliberate efforts to change or transform beliefs that would undermine these core values. Value democracy formally guarantees freedom of expression and association, but it should also use its own expressive capacities, he argues, to indicate that it does not itself support viewpoints that contravene free and equal citizenship (Brettschneider, 2012, pp. 8, 12–15). Brettschneider unsurprisingly believes that Boy Scouts v. Dale was wrongly decided. The court on his view there endorsed an organization advocating antigay policies and thereby denied the public ideal of free and equal citizenship.

Others support the values of free and equal citizenship, but are uneasy about this means of promoting them (Gill, 2013, 2014). Courts can uphold a constitutional right to advocate for particular policies without advocating for these policies themselves. First, public authority itself cannot be counted on to support these values in its speech. Many past court decisions come to mind that have disregarded the equality rights of LGBT persons. Second, which organizations espouse values that sufficiently oppose free and equal citizenship that they should be censured, and who is to decide? Brettschneider argues that the government may criticize particular viewpoints by the selective withdrawal of tax exemptions. The Supreme Court, for example, in 1983 upheld the removal of tax exemptions from Bob Jones University (Bob Jones University v. United States) because the institution not only forbade interracial dating, but also public support for interracial marriage and membership in the NAACP. On the other hand, he thinks that the Roman Catholic Church is different, despite the fact that it explicitly bars women—and implicitly bars gays—from the priesthood on religious grounds. It does not suggest that women and gays are not equal citizens in the larger society. Others wonder, however, if people want the Internal Revenue Service to decide the meaning of equality under the direction of political appointees (Spinner-Halev, 2011, 7p. 82; see also Inazu, 2016, pp. 69–71, 74–80).

Finally, Brettschneider’s quasi-coercive, viewpoint-based recommendations arguably exert a chilling impact on these entities. To him such measures are not coercive, because these organizations may still choose to maintain their present courses and to forego tax exemptions. But given the existing baseline of long-standing exemptions, the threat of revocation would certainly be experienced as coercive. Moreover, organizations may over time be influenced by the egalitarian ideals of the larger society to rethink some of their own policies and to embody more fully the values of free and equal citizenship. Although Bob Jones University altered its policies seventeen years after the court decision, it is impossible to determine the extent to which the loss of the tax exemption influenced this change. The Boy Scouts, however, eventually decided to allow gay and transgender participants, without any threat of coercion. Brettschneider argues that religious entities can be pressured to alter their tenets and still remain intact. Religions have existed for centuries, he maintains, because of their ability to evolve in diverse cultural contexts. He simultaneously appears too impatient, however, to allow them to evolve on their own. In most cases, the logic of congruence, or the principle that all voluntary associations should reflect the same egalitarian values, is mistaken. Pluralism can be characterized by a diversity of associations not all of which adhere to pluralist values internally. No contested policy can exhibit neutrality toward both sides. Achieving even a modicum of neutrality is difficult enough without deliberate governmental efforts to transform people’s beliefs.

Associations Within Larger Entities

A different kind of dilemma emerges in the case of Christian Legal Society v. Martinez, in which the Supreme Court in 2010 upheld Hastings College of Law of the University of California when it deprived the Christian Legal Society (CLS) of its registered student organization (RSO) status when it refused to follow the school’s accept-all-comers policy. Following this policy was required to access school funds and certain channels of communication, and it required that all students be allowed to participate as members or leaders in RSOs regardless of their status or personal beliefs. CLS, however, was a national organization requiring members and officers to sign and to adhere to a statement of faith, including the belief that sexual activity should occur only within marriage between a man and a woman. CLS therefore excluded anyone engaging in “unrepentant homosexual conduct” or holding opposing religious convictions. CLS challenged its loss of RSO status on grounds that the RSO policy violated its right to free speech, expressive association, and the free exercise of religion.

The Supreme Court found that CLS was not a victim of compelled association. Unlike the Boy Scouts, it could exclude anyone it chose by giving up the benefits of RSO status. That is, the Scouts could not simply forego institutional support and continue as before. The court moreover ruled that the all-comers policy was viewpoint neutral, as Hastings students were thereby not compelled through mandatory student activity fees to fund groups that would reject any of them as members. Educational institutions must have the ability to maintain standards consistent with their missions, and the RSO policy was neutral despite potentially burdening some groups more than others. The dissenters, however, argued that the putative non-discrimination policy in fact discriminated against religious groups with a particular viewpoint.

This case invites analysis from several perspectives. Unlike the Boy Scouts, CLS received funding from a public institution. One reaction supports the decision, but on different grounds from those presented by the court. From this vantage point, the all-comers policy was not viewpoint neutral, but this was as it should be. In fact, granting state funds to an entity should be premised on its support for the values of free and equal citizenship. An organization’s rights can be respected without its viewpoint being funded (Brettschneider, 2012, pp. 118–120; Tebbe, 2017, pp. 255–256, n. 19). Although this latter point is correct, it seems less extreme, on the other hand, to ground the decision in Martinez on ensuring that all students can benefits from RSOs that they help to fund rather than on the instantiation of correct state speech.

On a second interpretation, the Martinez decision is justifiable because it was not based on CLS’s viewpoint or its members’ beliefs, but on its exclusionary conduct. Antidiscrimination policies in general, and Hastings Law School in particular, do not aim at the silencing of expression but at the prohibition of conduct. Therefore, they do not contravene the First Amendment; any effects on expression are incidental (Ball, 2017, pp. 213–218). Groups like CLS, however, argue that these “incidental” effects are precisely the point: groups are burdened when their views diverge from the mainstream and this result is illegitimate. As Carlos Ball notes, however, if the courts were to exempt dissenters from the application of antidiscrimination laws because of their “disparate impact” on those who wish to discriminate, “it would bring the application of antidiscrimination law, as we know it, to a grinding halt” (Ball, 2017, p. 219; see pp. 218–221). When universities in earlier cases refused to recognize gay student groups, it was their beliefs and the content of their speech that were being targeted, not their conduct. From an ethical perspective, moreover, in their equality claims LGBT groups sought the same treatment as other groups, rather than the preferential treatment embodied in exemptions. Interestingly, Ball suggests that although CLS did not have a constitutional entitlement to a religious exemption in Martinez, the law school should nevertheless have permitted an exemption. Few LGBT students probably wanted to join a group that condemned their relationships. Moreover, if some of these students sought to join and were rejected, the harm would have been less than that suffered by LGBT individuals who are denied employment or services in public accommodations. Finally, it was unlikely that other student groups would take their cues from CLS and decide themselves to exclude sexual minorities. The exemption, however, would have set a precedent that could prove to be unwelcome in the future.

On a third interpretation, Martinez was correctly decided because CLS was not an independent organization within the larger society, but an organization within a larger entity, Hastings Law School of the University of California. Religious associations, like political parties, ordinarily have the right to include or exclude members and leaders for their own reasons because this autonomy is necessary to the very definition of the group’s identity. The court in this case, however, regarded CLS not as an independent voluntary group, but “as an association-within-a-university” (Flanders, 2016, p. 109; see also Lupu & Tuttle, 2014, pp. 188, 202–203). It was in part expressive of Hastings’s values as well as its own. CLS was not simply a private actor but also a quasi-public one. By way of comparison, political parties are not only private organizations that pick nominees for public office, but also associations within a publicly run electoral system. The courts have ruled that because the states hold party primaries that select nominees for general elections, discriminatory party membership criteria constitute state action, not private activity (Flanders, 2016, p. 113; see pp. 105–115). Chad Flanders concludes that the right to freedom of association does not automatically preclude a degree of regulation; the function of the group must be examined. The more that a group exercises public functions as well as private ones, the weaker its claim to complete autonomy. In Martinez, CLS’s aims competed with the law school’s goals.

Overall, the decision seems correct on several grounds, bearing added weight because public funds were at stake. An organization’s ability to exercise a recognized liberty right is not contradicted by a government refusal to support its activity, although the refusal may increase the cost of its exercise. The government has some discretion in refusing to support discriminatory groups even when some of these groups may be religious in nature (Tebbe, 2017, pp. 184–191). This point may appear to contradict the argument that the withdrawal of tax exemptions is coercive. Although even this step may be constitutionally permissible (Tebbe, 2017, pp. 191–193), the termination of long-standing indirect support for nonprofit associations deemed to promote the public interest in general is a graver move, especially in view of real possibilities that these organizations may over time make changes of their own volition.

Small Businesses as Public Accommodations

Masterpiece Cakeshop and Related Cases

A primary site of conflict between LGBT equality claims and religious free exercise claims has been the provision of services connected with same-sex weddings and commitment ceremonies. Some observe that same-sex couples and their advocates, on the one hand, and those with religious objections to marriage equality, on the other, are parallel in their demands. Both sexual orientation and religious faith are fundamental aspects of identity, they claim, and the living out of this identity should be subject to only minimal regulation (Koppelman, 2015, p. 626; Laycock, 2008, p. 198; Wilson, 2016, p. 263). The truth of this parallel appears in the fact that while these scholars are drawing attention to it in the service of religious freedom, others make the same claim on behalf of LGBT equality (Ball, 2003; Gill, 2012; Richards, 1999, 2005). The parallel also illustrates the impossibility of neutrality in that both sides of this controversy cite the same defense of their viewpoints.

Fewer people in the early 20th century view marriage as a primarily religious institution. Nevertheless, some religious wedding planners, photographers, caterers, bakers, and florists decline to participate in what they view as the validation of a forbidden relationship. Their advocates believe, therefore, that vendors in very small businesses and relationship counselors should be exempt from providing services as long as alternatives are available to same-sex couples. They suggest that the emotional harm of being coerced to act against one’s religious beliefs outweighs the dignity harm to same-sex couples of being turned away by their desired providers (Laycock, 2016, pp. 242–246). Even scholars who would elevate equality rights over religious rights in cases of irresolvable conflict concede that individuals rightly expect to be able to express these core aspects of identity through behavior and practice (Feldblum, 2008, pp. 123–124, 142–143, 152–156).

Until 2018, the tide appeared to favor LGBT equality rights (Mezey, 2017, pp. 161–163). In the Elane Photography case, for example, a New Mexico studio’s co-owner and chief photographer cited her religious beliefs in refusing to photograph a lesbian couple’s commitment ceremony. She said she did not object to their status as lesbians, but to their conduct in solemnizing their commitment. Overall, the lower courts rejected this distinction between status and conduct as vitiating the state human rights act and ruled that the state Religious Freedom Restoration Act did not apply to private citizens; the United States Supreme Court denied certiorari (Elane Photography v. Willock, 2013; Elane Photography v. Willock, 2014).

In 2012, however, devout Colorado Christian bakery owner Jack Phillips refused to create a wedding cake for the reception of a same-sex couple previously married in Massachusetts, claiming that doing so would force him to express a message with which he disagreed and would violate his religious freedom. He was willing, however, to sell ready-made baked goods to anyone. An administrative law judge, the Civil Rights Commission, and the Colorado Court of Appeals all ruled that the state antidiscrimination law was applicable. Like the Elane court, the latter in 2015 rejected the distinction between status and conduct (Craig v. Masterpiece Cakeshop, Inc.), as it was the couple’s sexual orientation that prompted their union in the first place and their desire to celebrate it. On appeal, the United States Supreme Court ruled in 2018 that although gay individuals must be protected in exercising their civil rights, religious objections to marriage equality must also be protected forms of expression (Masterpiece Cakeshop v. Colorado Civil Rights Commission).

Writing for the five-justice majority, Justice Anthony Kennedy stated that members of the Colorado Civil Rights Commission (CCRC) had displayed animus or hostility toward religion by characterizing Phillips’s beliefs as despicable, rhetorical, and by implication insincere. In previous cases, it had supported other bakers who would not decorate cakes in ways that denigrated gays or same-sex marriages, apparently deciding that the requested messages offended their own possibly secular convictions. In Phillips’s case, however, it argued that any message on a cake would be that of the customer, not of the baker, thereby treating his religious objections as illegitimate, a judgment in which the government should play no role. The CCRC’s decision was therefore inconsistent with the government’s constitutionally required neutrality. It was not unreasonable, Kennedy wrote, for Phillips to refuse as he did, as same-sex marriages were not even recognized in Colorado at the time. In sum, the state of Colorado did not in the court’s view come close to weighing fairly the equality rights of same-sex couples against Phillips’s religious objections. The animus displayed toward the latter was enough. The permissibility of a religious exemption need not even be considered when this hostility precluded the enforcement of the antidiscrimination law in the first place. Otherwise, the court could have created a broad exemption for those with religious objections to providing services for the celebration of same-sex weddings or commitment ceremonies.

The CCRC might have found against Phillips in an unbiased fashion, however. All of the bakers refused to create particular types of cakes, but only Phillips denied a cake to one type of couple, a same-sex one, that he would have made for another type of couple, an opposite-sex one. Presumably those other bakers would have refused to make cakes denigrating gays for any customer who requested them, whereas Phillips denied a cake to a same-sex couple that he would have made for an opposite-sex couple. As the dissent argued, whereas the bakers in the earlier cases would refuse to make the requested cakes for any customer at all, however they might use these cakes, Phillips would not sell to his would-be customers because of their sexual orientations and/or the use to which the cakes would be put.

To some, the outcome in Masterpiece Cakeshop was a victory for religious liberty. To others, its narrowness indicated that protections for the civil rights of LGBT individuals were not generally threatened. The benefit to religion resulted not from deliberate intent to favor religious practice, but from the negative liberty of freedom from interference with it. To still others, the outcome encouraged others to bring similar cases that may broaden protections for religious exercise at the expense of the civil rights of others. In general, commercial establishments are public accommodations and should not be able to discriminate against customers on the basis of activities that may result from their identities, even if those who run those establishments may have sincere religious objections to doing so. A request for a cake celebrating a same-sex marriage will only be made by or on behalf of a same-sex couple. Therefore, the implicit distinction made in these types of cases between status—the type of customer—and conduct—the type of activity that only that sort of customer would engage in—is illegitimate. The dissent was thus correct in arguing that Phillips’s problem with the cake was determined by the identity of the customer, a gay couple.

Given that the court ruled for Phillips, its justification also seemed correct. The government must not establish, sponsor, or support religion, but also it must not denigrate or exhibit hostility to religion or to its free exercise. The court could have ruled for Phillips more expansively, alternatively, but without creating a religious exemption. The other Colorado bakers were allowed to refuse cake orders that would be derogatory toward gays or same-sex couples because this request offended their beliefs, but Phillips was not afforded the same latitude to decline an order that would have offended his beliefs. If the focus had been on expressive conduct, it would not have mattered that some bakers refused objectionable cakes to any requesting customer, whereas Phillips denied a cake for a same-sex wedding that he would have provided for a traditional wedding. All that would matter would be sincere religious or conscientious objections to compelled expressive conduct. This line of argument, however, does not evade the difficulty with the status-conduct distinction: only same-sex couples request cakes for same-sex weddings. To understand why this resolution would not require a religious exemption, one could look to West Virginia Board of Education v. Barnette. In this 1943 case, the Supreme Court did not grant a religious exemption to Jehovah’s Witness children so that they could not be compelled to salute the American flag, a practice that contravened their religion. Instead, the decision protected all students from any compulsion to demonstrate their political loyalty if this violated their religious or conscientious beliefs. For Masterpiece Cakeshop, this would have meant that neither Phillips nor the other bakers could be forced to contravene their sincere conscientious beliefs, religious or otherwise. This potential justification is broad and would have been questionable in this case, however, whereas the actual justification was narrow.


Masterpiece Cakeshop clearly exemplifies the challenge of neutrality between LGBT equality claims and religious exercise claims. Whatever the grounds for decisions in these types of cases, either service providers such as Phillips must create cakes for same-sex couples or these couples must look elsewhere for these services. Before this tension arose, in 1993 the United States Congress passed the Religious Freedom Restoration Act (RFRA), stipulating that when a law substantially burdens religious freedom, the government must show that it is justified by a compelling state interest and that the law requires the least restrictive means of pursuing that interest. Partly as a result of the legal instantiation of marriage equality nationwide, many states passed their own RFRAs, requiring the same compelling governmental interest achieved by the least restrictive means as the federal RFRA for public policies to survive a religious liberty challenge. Although some of these laws do not allude to marriage equality, others, known as First Amendment Defense Acts (FADAs), do explicitly protect individuals and organizations with religious objections to same-sex marriage (Mezey, 2017, pp. 164–172). Many include a religious freedom defense in private civil suits, a protection that Elane Photography did not have. Controversies over these laws have been far-reaching.

Although Masterpiece Cakeshop was not decided on the basis of compelled expression, wedding cake design, photography, floral arranging, and other services do contain expressive aspects that convey messages. On grounds of the Kentucky RFRA and the threat of compelled speech, a trial judge in 2015 upheld the refusal of a Christian printer to provide T-shirts for a gay pride event, ruling that the printer had discriminated because of the requested message, not sexual orientation. He had previously refused messages promoting a strip club and a pornographic video (Hands On Originals v. Lexington-Fayette County Human Rights Commission, 2015; Laycock, 2016, p. 246). If the Elane Photography and Masterpiece Cakeshop refusals were rooted in sexual orientation, one could argue similarly that sexual orientation grounded the Kentucky printer’s refusal to print gay pride messages on T-shirts. When are these refusals ethically legitimate? And must they be based on religious belief? Two points are clear.

First, vendors may discriminate on the basis of the requested message in areas other than sexual orientation and gender identity. A kosher bakery would undoubtedly refuse to decorate a cake with swastikas. Such refusals should be legitimate if the cake is custom designed. The owner would be refusing this service to any individuals or groups that requested it, whether for religious or nonreligious reasons, rather than to a particular group of people or for particular uses. Second, however, same-sex weddings connect directly to sexual orientation and the living out of that identity in ways that other messages are not. Such refusals operate against a particular identity group (Gill, 2012, pp. 160–169). Same-sex marriage is legitimate nationwide. If the government condones discrimination against same-sex couples when they attempt to arrange their weddings, it is legally privileging straight couples by making it more difficult for same-sex couples to live out the implications of their identities and of their perhaps conscientious beliefs that they should marry, a belief that is a government-supported right. Moreover, if the government condones this kind of discrimination, it is allowing discrimination against some on the basis of the religious beliefs of others. Under the First Amendment establishment clause, this indirectly condones one type of conscientious belief over another, despite the fact that private parties, not the government, are engaged in discrimination. Under the free exercise clause, it impedes the ability of same-sex couples to celebrate according to their own preference their decision to take up a legal status open to all couples.

Somewhat surprisingly, more than one view exists with regard to the religious rights of public officials. On one view, public officials should never be insulated from the consequences of refusing to issue marriage licenses to same-sex couples. The most well-known example here is county clerk Kim Davis of Rowan County, Kentucky, who, the day after Obergefell was handed down, ordered her deputy clerks to withhold all marriage licenses—to avoid discrimination against same-sex couples—because issuing them would validate same-sex marriage. To summarize, after Davis spent time in jail on civil contempt charges, the Kentucky legislature eventually created a uniform marriage license form that does not bear the names of county clerks (Mezey, 2017, pp. 174–178; Tebbe, 2017, pp. 166, 174–175). Davis, like other public officials, agreed to follow the law when she took office as a deputy clerk 26 years before becoming county clerk in 2014. She undoubtedly never glimpsed the possibility of marriage equality. Laws change over time, however, and public officials should adjust or seek some other career. Overall, a public official who registers disapproval of same-sex marriage by refusing to issue licenses exemplifies Gutmann’s public expression of civic inequality (Gutmann, 2003, p. 97).

On another view, however, public officials are immediately and directly affected by marriage equality in ways that bakers, photographers, florists, and such are not. County clerks must not only issue marriage licenses but must also be ready to preside at actual civil marriage ceremonies. On this line of reasoning, although the government should not be legally required to accommodate public officials, accommodation is permissible under certain circumstances (Tebbe, 2017, pp. 164–181). Davis initially altered the license form to state that it was being issued on a federal court order, to omit even the name of her office of county clerk, and to be issued by her deputies not as deputy clerks, but as notaries public. This altered form, however, communicated official resistance to marriage equality and degraded the citizenship status of same-sex couples. The final form, however, provided that deputies sign it in their capacities as deputies and did not include the earlier disclaimer. If a distinction is made between private individuals and the offices that they hold, Kim Davis was an individual with deeply held beliefs, but she did not constitute the office itself. If one or another deputy clerk can seamlessly and invisibly process all comers by issuing marriage licenses—and by presiding at marriages when necessary—then a state office as a public entity cannot be charged with aligning itself with a particular religion. Because it is an informal arrangement, its “expressive impact” is negligible and is invisible to the public (Tebbe, 2017, p. 170). Aside from the mechanics involved in ensuring that a willing issuer or celebrant is available at all times, the question arises as to whether individuals or couples may be harmed or have their status degraded even without awareness that this is the case, an issue beyond the scope of this article. As a further consideration, should accommodations also be available to county clerks or deputies who object to remarriage that contravenes their own religious beliefs or to interfaith unions? If public offices wish to use informal arrangements to staff around objectors so that same-sex couples can experience a seamless transaction, that is their prerogative. But they should not be required to do so.

Some owners or employees of private businesses in the line of commerce, such as pharmacies, have objected to dispensing emergency contraception if they believe that it functions as an abortifacient. Although sending patients to another pharmacy is unacceptable, some favor the solution that a willing pharmacist always be available at any pharmacy. Pharmacies might be forced at additional expense to retain a second pharmacist to staff around the first one, however. This might be difficult in rural areas or at odd times of day or night. In the case of emergency contraception, timing is essential. Moreover, contraceptives, including emergency contraception, are often prescribed for reasons other than birth control. Some women have health reasons for not becoming pregnant. What if a pharmacist has conscientious objections to selective serotonin reuptake inhibitors, such as Prozac for attention deficit hyperactivity disorder, because she believes that too many children are mistakenly medicated for this problem? What if a pharmacy employs several pharmacists, but they cover different times of the day or week? And what if they all entertain conscientious objections to dispensing certain medications, but for each pharmacist, different medications are objectionable?

Compromise Approaches

Although no solution is neutral, the conflict between the equality rights of same-sex couples and the religious rights asserted by those who do not wish to serve them is open to several possible compromises. Bakers, for example, could decide not to provide wedding cakes to anyone. Or they could offer ready-made baked goods, including wedding cakes, to all comers, but custom-designed wedding cakes to no one. This compromise might also apply to florists. Couples who want custom cakes or floral arrangements from particular businesses would not be pleased, and this solution would not work for photographers, as photographs are necessarily custom designed. Alternatively, small businesses could merge with larger ones where at least some employees would not object to providing these services. Or a small business could decide not to advertise or to hold itself out as a public accommodation, relying on social networks for customers (Koppelman, 2015, p. 630). Finally, if exemptions exist for small businesses, couples can seek services elsewhere.

This last point raises the dignitary issue. If same-sex couples have a narrower range of options than traditional couples, how is this different from legal segregation that narrowed the options of African Americans who sought services at restaurants, hotels, public swimming pools, and schools? Of course, a great many vendors are happy to provide services for same-sex couples, unlike the situation for African Americans. As discussed, private voluntary organizations such as the Boy Scouts should be free to set their own membership rules—unless most voluntary organizations discriminated against the same groups of individuals in the same way, therefore curtailing their opportunities overall. Similarly, it could be argued that exemptions based on religious belief or conscience should exist unless no other vendors are willing to provide services for the weddings of same-sex couples. These vendors, however, are not nonprofit voluntary associations, but for-profit businesses that are public accommodations.

If discrimination is most serious where it is most pervasive, perhaps it is less serious when it is the exception rather than the rule. The law could exempt small businesses from providing services to same-sex couples if and only if they agree to identify themselves publicly as discriminatory. First, the cost they would engender for doing so would soon make discrimination rare. This sort of narrow religious exemption would not encourage more widespread discrimination against LGBT individuals, as social attitudes have greatly changed, and even religious people generally do not refuse simply to interact with LGBT people (Koppelman, 2015). Some libertarians suggest, moreover, that serving a same-sex wedding is an expressive action that implicitly supports same-sex marriage, unlike renting hotel rooms to African Americans (Levy, 2016). Second, although refusal of service is a dignitary harm, it may be considered less serious than the possibility of wedding vendors closing their businesses if they cannot sustain the fines for flouting antidiscrimination laws. Moreover, if discriminatory practices are contingent on public statements, same-sex couples are not faced with direct refusals (Koppelman, 2015; Laycock, 2008, pp. 198–199; Wilson, 2016, pp. 271–274). Finally, antidiscrimination laws aim not so much at preventing discrete injuries to given individuals as at social reconstruction, at transforming shared values and the resulting practices that contribute to inequality in the larger society. For many, the culture has been transformed so thoroughly that those who persist in discrimination against same-sex couples will be marginalized. “The gay rights movement has won. It will not be stopped by a few exemptions. It should be magnanimous in victory” (Koppelman, 2015, p. 628; see also p. 658). The question remains, however, as to why the options of same-sex couples should be curtailed while the rest of the social transformation occurs.

Despite marriage equality’s availability nationwide, some argue that compromises are necessary to protect LGBT rights in states without antidiscrimination protection where the aforementioned victory has been elusive. Utah legislators, for example, in 2015 constructed a compromise to protect both civil rights and religious interests. Most employers may not discriminate against LGBT individuals, must accommodate transgender individuals, must allow political and religious expression outside the workplace, and must allow equal treatment in the workplace for both political and religious speech, although this speech may also be prohibited across the board. Religious organizations with housing units can prefer those of their own faith, and landlords with up to four units have complete freedom in choosing tenants. Churches and their affiliates, religious schools, the Boy Scouts, and small family businesses with fewer than 15 employees may follow their religious values in hiring.

The Utah Compromise, as it is termed, also permitted marriage equality, although it was fashioned a few months before Obergefell was handed down. County clerks’ offices were required to designate “a willing celebrant,” which provided seamless access to marriage for all couples (Wilson, 2016, p. 267). Both religious believers and LGBT individuals gained protections here that neither had previously (Laycock, 2016, pp. 252–253; Wilson, 2016, pp. 265–268). For some, religious freedom advances most securely when it also protects others (Wilson, 2016, pp. 268, 262–264). For others, the protection of LGBT individuals can be promoted most reliably when it accompanies religious exemptions (Laycock, 2016, p. 253) and earns the cooperation of religious allies (Koppelman, 2015, p. 658). In sum, on these views defenders of religious freedom and LGBT equality advocates will each gain more through compromise than without it.

Although religious objections to providing services for the weddings of same-sex couples may be narrowly focused (Koppelman, 2015, p. 643), entities receiving any public funding should be held strictly to non-discrimination provisions even if they are religiously affiliated, as with agencies such as Catholic Charities. The states regulate adoption agencies, and most states require that these agencies be licensed. Although the license requirement may allow states to prohibit discrimination by Catholic Charities against same-sex couples, some states have passed various exemptions that permit religious adoption agencies to reject same-sex spouses. Connecticut, for example, allows adoption agencies to refuse same-sex couples but only if they receive public funding (Tebbe, 2017, pp. 124–126). Otherwise, in 2006 Catholic Charities of Boston terminated child placement altogether rather than place children with same-sex couples. Catholic Charities of Illinois, on the other hand, terminated its relationship with the Catholic Church in 2011 in order to continue its child placement services on a non-discriminatory basis. Overall, religious adoption agencies have never been permitted to discriminate against couples on other theologically based grounds, such as remarried, interfaith, or interracial couples. Why should they be able to exclude same-sex couples, perhaps even if they do not receive public funds? (Ball, 2017, pp. 225–226; Tebbe, 2017, pp. 136–137).

Regarding photographers, bakers, and florists, arguments may go in both directions. First, as public accommodations these businesses should serve all comers. Second, once again, their services accompany entry into a legal status actually encouraged by the state through the material benefits that married couples receive. Even those advocating generosity toward religious opponents of marriage equality admit that in a worst-case scenario, the landscape might be littered with public notices by those who will not serve same-sex couples, a scene reminiscent of “Whites only” signage in the South before desegregation. Douglas Laycock notes that religious dissenters may not occupy choke points that would prevent same-sex couples from receiving services overall, although ironically, there might be more choke points in the same places, such as the Bible Belt, where there are also more objectors. This would result in an odd situation: where merchants are most likely to be able to delay or prevent same-sex couples from receiving services, they would also be most likely compelled to provide them (Laycock, 2008, p. 200; Wilson, 2008, pp. 97–102).

How would this work? In a small town or rural area, for example, there might be only one photographer, bakery, or florist shop—or pharmacies with only one pharmacist, for that matter. Would the law stipulate that an exemption would apply if there were at least x number of photographers in a geographic area of y square miles containing a population of z? Such a law could in theory also threaten civil rights protections for the religious freedom of customers if, for example, a Muslim florist declined to sell flowers for a Jewish wedding, or a caterer declined to provide refreshments to celebrate a wedding where the officiant was a woman (Flynn, 2010, pp. 237, 243–247). Contrary to the claim that religious refusals have centered narrowly on marriage ceremonies and related issues (Koppelman, 2015, p. 643), Taylor Flynn argues that concerns for religious liberty emerged well before marriage equality and that the latter has been used as a pretext to attempt to curtail existing protections for sexual orientation. The claim that religious animosity is not directed toward LGBT persons, but only toward particular activities, is once again the status-conduct distinction, typically rejected in the area of religious discrimination (Feldblum, 2008, pp. 123–124, 142–143; Flynn, 2010, p. 252; see pp. 251–254; Gill, 2012, pp. 22–23). Although some might argue that allowing sign posting only respects religious beliefs and would not be a state endorsement of discrimination, it would bestow the sanction of law on vendors in the public realm in their discrimination against a single group. “The state-provided exemption . . . is the sine qua non for sign-posting; it could not exist without the imprimatur of the state” (Flynn, 2010, p. 256; see pp. 254–257). It would in effect create a second-class citizenship and therefore a public expression of civic inequality.

For comparative purposes, in 1948 when the Supreme Court decided Shelley v. Kraemer, it did not prohibit homeowners from agreeing through restrictive covenants with neighbors not to rent or sell their homes to African Americans. It did, however, prevent state enforcement of these covenants, as this state action would violate the equal protection clause of the Fourteenth Amendment by legitimating discrimination based on race. Similarly, state support for refusals of service pursuant to marriage equality would legitimate discrimination based on sexual orientation or in some cases on gender identity. This is all the more so because the Cakeshop case suggests the possibility of new laws offering exemptions conditioned on public notice, whereas with the earlier restrictive covenants, only the enforcement of private agreements was at issue. The possibility that signage permission might incentivize some businesses to discriminate on additional religious grounds against remarriages, interfaith weddings, and so forth lends added force to the inadvisability of allowing this sort of exemption.

This article has suggested that bakers are entitled to decline to design cakes with white supremacist messages, that kosher bakeries may refuse to decorate cakes with swastikas, and by implication that the Kentucky printer at Hands On Originals was correctly upheld in declining to print T-shirts for a gay pride event. In the latter case, the court held that it was the message itself, not sexual orientation, that in part justified the printer’s refusal (Laycock, 2016, p. 248). Can an LGBT rights supporter argue that underlying this refusal is sexual orientation, just as in the photography and the cake shop cases? No, because the connection is more attenuated. In the same-sex union cases, objectors are targeting couples who would not be seeking these services if it were not for their sexual orientations. In the other cases, objectors are withholding their perceived endorsement from messages they do not want to convey regardless of the identities or orientations of those who seek their services.

Although some suggest that antidiscrimination prohibitions are directed toward social reconstruction rather than against individual injury (Koppelman, 2015, p. 652), condoning instances of individual injury can allow social reconstruction to regress. As stated by Kenneth Karst,

The most heartrending deprivation of all is the inequality of status that excludes people from full membership in the community, degrading them by labeling them as outsiders, denying them their very selves. The harms of exclusion unquestionably happen to people one by one, but those individual harms result from the subordination of groups.

(Karst, 1989, p. 4)

Although some religious opponents to LGBT equality rights themselves feel excluded by being forbidden to discriminate, their religious exercise rights are curtailed only as to how they may treat others, not in terms of their own opportunities. If the purpose of antidiscrimination law is to change society, then the legal exclusion of the members of some groups should not be allowed, even “one by one.”

Broader Issues

Tension between the free exercise of religion and equality rights in general has attracted the attention of jurisdictional pluralists. Although typically their focus is on conflicts between religious organizations and the state, the issues they raise also apply to the conflicts discussed here. Some consideration of their claims demonstrates more generally the impossibility of neutrality in the disposition of many religious freedom claims.

Although the state is sovereign within its territory, the government has the power to designate a sphere of noninterference as a jurisdictional boundary that it will not cross. Federalism and the separation of powers in the Constitution designate spheres of authority, and courts commonly dismiss cases for lack of jurisdiction, whatever the merits of a case. Even though the government holds the ultimate power, in the law it can decide to defer to alternative authorities (Horwitz & Tebbe, 2016, pp. 222–223; Smith, 2016, pp. 20–31). Concerning religion, the state must provide an infrastructure or a protected framework as the nourishing conditions within which individuals and groups may pursue their religious interests (Garnett, 2016, pp. 44–54).

Whenever autonomous jurisdiction is accorded to particular entities, however, a prior authoritative structure must exist to define the nature and scope of this jurisdiction, just as the Constitution defines the relationship between the national government and the states. Unless an ultimate source of legitimate sovereignty is present, individuals “remain structurally vulnerable to the arbitrary will of others” (Laborde, 2017, p. 161; see pp. 161–171). That is, liberal democratic states hold final authority to represent the interests of individuals as such, shorn of their contingent identities and including but not limited to their religious identities. This point applies to those who wonder not what makes religion special, but what makes the state so special that religion must fit into it as simply one component (Dane, 2018, p. 147; see also Cohen, 2017, pp. 85–93, 99–100).

Even a state that is neutral about religious truth cannot be neutral in determining the line between what is civil and what is religious. That is, even when it defers to other authorities or to individuals, it must have “meta-jurisdictional sovereignty” (Laborde, 2017, p. 162). Private voluntary associations and even for-profit entities have unique expertise or competence in interpreting and applying their own purposes and commitments, and in these areas they should exercise jurisdictional autonomy, although not always complete immunity. They do not, however, possess “competence interests to identify their own area of competence . . . they do not have a right to jurisdictional self-definition” (Laborde, 2017, p. 196; see pp. 190–196; Creppell, 1996, p. 224; McClure, 1990, pp. 373–381). In this sense the state must be secular, but only in the sense that it must determine the line of demarcation between the civil and the religious (Laborde, 2017, p. 162). The claims of jurisdictional pluralists who object to this assertion actually center not on whether the state possesses this authority, but on “whether legitimate state power is exercised justly” (Laborde, 2017, p. 168).

Associations possess interests in interpreting and applying their own purposes and commitments, or their own ethos, sometimes termed their coherence interests. On one approach to the way these interests function, groups that desire religious or ethical exemptions must be formally organized as voluntary associations and must focus on specific doctrines or purposes. Furthermore, its members or those who associate with or patronize it must identify with its core projects and commitments. A small personal or family business might meet these criteria, but not large for-profit corporations such as Hobby Lobby. Religious bookstores or kosher markets with self-selecting customers might qualify, but not public accommodations that want exemptions from serving certain classes of individuals. What matters is not the religious nature of an association or entity, but the type or mode of association (Laborde, 2017, pp. 182–187; see also pp. 178–180).

Moreover, protected coherence interests may require a clearly articulated public message, a criterion that the Boy Scouts did not meet in Boy Scouts of America v. Dale (2000). As mentioned, Scout officials only belatedly explained that requiring boys to be “morally straight” meant that they could not be gay, thereby raising the suspicion that this new interpretation might be pretextual despite the Supreme’s Court’s decision otherwise. On Cécile Laborde’s view, discrimination should be expensive in a liberal society that promotes free and equal citizenship, and requiring public dissemination of a discriminatory message raises its cost (Laborde, 2017, pp. 188–190; see also Koppelman & Wolfe, 2009; Tebbe, 2017, pp. 80–97). The coherence requirement of a clear public message may in fact promote internal debate and eventual change, as happened with the Scouts. On the other hand, the Scouts changed their policies anyway after being allowed to discriminate without a clearly articulated public message. A compromise here might be a stipulation that the more clearly an association’s members or customers identify with its core projects and commitments, the more strictly it should be held to the requirement of a clear public message if it is allowed to discriminate, as in the Roman Catholic commitment to an all-male clergy.

Values organizations whose predominant purposes and activities are religious may still receive exemptions. The Civil Rights Act of 1964 provided that religious corporations and associations may limit their hiring to co-religionists, but did not allow them to discriminate on other grounds such as race or sex. It also stipulated that this exemption only applied to an organization’s religious activities, or those pertaining to directing its religious mission. In the 1972 amendments to Title VII of this law, however, Congress broadened the exemption to allow religious organizations to hire co-religionists for any and all positions, not limiting it to employees involved in religious activities. This move represented a compromise between those who would have allowed complete freedom to discriminate on any grounds and those who wanted to continue a limitation of the exemption to religious activities. To many observers, this compromise has worked well (Ball, 2017, pp. 174–180). It was upheld in 1987 by the Supreme Court in Corporation of the Presiding Bishop v. Amos, allowing a gym run by the Church of Jesus Christ of Latter-day Saints to fire a janitor for failing to qualify for the certificate required to attend the religion’s temples.

Those critical of Amos, however, are concerned about the economic power wielded by religious associations that may be the dominant economic force in certain regions. In this view, even religious organizations should at least be required to demonstrate a religious basis for their policies. Religious organizations should not be able “to define for themselves what falls within the scope of activity that is conceivably part of their self-definition” (Rosenblum, 2000, p. 189). In other words, although organizations should exercise a wide latitude within their areas of expertise and competence, they should not be able to define for themselves the limits of these areas. Moreover, evidence exists that religious individuals and organizations often receive license to discriminate on the basis of gender and sexual orientation, even in state antidiscrimination statutes when this would be forbidden regarding race (Babst & Compton, 2016, pp. 99–103).

In Kentucky in 1998, a lesbian family specialist was fired by a Baptist organization operating children’s homes. Although Kentucky did not protect sexual orientation, she was allowed to mount an establishment clause challenge, as the organization received state funding. A federal appeals court in 2009 ruled in part that although employers could not discriminate against employees because they did not follow their employer’s religion, the employee had not shown that discrimination because of her sexuality constituted discrimination based on religion (Pedreira et al. v. Kentucky Baptist Homes for Children, Inc.). The establishment challenge went forward, but the Supreme Court denied an appeal, and the case was settled out of court in 2013. What is meant by hiring co-religionists is open to interpretation. Some suggest that granting broad exemptions from civil rights laws violates the separation of church and state (Ball, 2017, p. 178). More specifically, allowing religious organizations wide discretion—at least in practice—when others do not have this option edges toward an establishment of religion.

However, although exemptions for those with religious objections to providing services in connection with same-sex weddings and commitment ceremonies may not be constitutionally required, under certain circumstances they may be permissible. For religious organizations and religious nonprofits in particular, exemptions from providing goods and services connected with same-sex weddings or commitment ceremonies would be analogous to the nearly universal understanding that clergy should not be forced to officiate at these marriages (Ball, 2017, pp. 265–266, 270)—or for that matter at any marriage for which clergy may deem a couple is unprepared. In any case, the purposes, commitments, and activities of these nonprofit organizations are values based as well as religious; their coherence interests afford them some latitude.


Despite the impossibility of neutrality, overall the law has tried to balance equality and religious liberty concerns in its application of antidiscrimination provisions. Accordingly, applications for religious exemptions should be treated neither more nor less generously when these conflict with LGBT equality claims than would be the case with equality claims based on race, gender, or other features of identity. Or at least the burden of proof should rest on those who claim otherwise. In most circumstances, one should be wary when religious (or ethical) liberty claims are made that would single out a particular class of individuals as the object of an exemption (Ball, 2017, pp. 263–266). This wariness appeared in Romer v. Evans (1996), which struck down the Colorado state constitutional amendment that withheld from LGB individuals, and only from them, the ability to pass antidiscrimination laws that would protect them. This motivation also infects the desires of those who would turn away same-sex couples seeking services pertaining to their weddings, but would not reject interracial, remarrying, or interfaith couples of whom they might disapprove.

This article began in part with a denial of the claim that although clear religious teachings exist against same-sex intimacy, this is not the case with white supremacy. On the contrary, both white supremacy and heterosexism or heteronormativity have been rooted in religious convictions, and each has established a caste system that privileges one group of people as superior to another. A more recent variation on this theme rests on claims that reasonable, good-faith religious disagreement still exists as to the legitimacy of same-sex marriage, whereas outdated expressions of white supremacy are rooted in pure race prejudice and are by implication insincere (Ball, 2017, pp. 271–273; Laycock, 2008, pp. 195–196; Wilson, 2008, p. 101). Violence against African Americans and Jews in 2017 and 2018 following the election of President Donald J. Trump, however, revealed good-faith beliefs by white supremacists that were obviously sincere. History may reveal that defenses of racial superiority and of heterosexual superiority still parallel each other more than one might like to think.

Finally, some individuals and groups who seek exemptions from serving same-sex couples in connection with weddings and commitment ceremonies may accuse defenders of LGBT equality rights of holding a double standard. The article has demonstrated that the exemption claims of business owners tended to distinguish between the status of being LGBT, on the one hand, and the conduct of LGBT couples who wanted to celebrate marriages. The status was fine; the conduct was not. The objection to this line of argument was that penalizing conduct in effect penalizes status as well, as only same-sex couples would seek to celebrate same-sex weddings or commitment ceremonies. Religious opponents of marriage equality may argue similarly against LGBT equality advocates. These advocates suggest that they do not object to religious convictions against same-sex marriage—the status—but only to discrimination against same-sex couples—the conduct. But only religious convictions as to the illegitimacy of same-sex marriage give rise to conscientious refusals to serve same-sex couples. Therefore, penalizing this conduct also penalizes status, or the beliefs that give rise to these refusals.

This analogy is correct in theory. But how it should be addressed in practice goes back to the discussion of jurisdictional pluralists and the weakness of their arguments. Religious individuals and groups have great autonomy within their areas of competence. In a liberal society and state, however, they cannot be accorded the competence to define the limits of these areas of competence (Laborde, 2017, pp. 161–171). Plenty of room for argument exists as to where this line should be drawn and whether it has been drawn correctly or justly. Without an entity with jurisdictional sovereignty, antidiscrimination laws would hold no authority, whether the discrimination they were to counter involved race, gender, sexual orientation and gender identity, or even religion itself. No dispensation is neutral in conflicts between LGBT equality claims and religious exercise claims. Although religious convictions should be considered, they cannot be accorded preeminence.


Alexander, L. (2008). What is freedom of association, and what is its denial? Social Philosophy and Policy, 25(2), 1–21.Find this resource:

    Babst, G. A., & Compton, J. W. (2016). Equal citizenship and religious liberty: An irresolvable tension. In A. Cudd & W. Lee (Eds.), Citizenship and immigration (pp. 91–104). Basel, Switzerland: Springer.Find this resource:

      Ball, C. A. (2003). The morality of gay rights: An exploration in political philosophy. New York, NY: Routledge.Find this resource:

        Ball, C. A. (2017). The First Amendment and LGBT equality: A contentious history. Cambridge, MA: Harvard University Press.Find this resource:

          Bob Jones University v. United States, 461 U.S. 574 (1983).Find this resource:

            Boy Scouts of America v. Dale, 530 U.S. 640 (2000).Find this resource:

              Brettschneider, C. (2012). When the state speaks, what should it say? How democracies can protect expression and promote equality. Princeton, NJ: Princeton University Press.Find this resource:

                Caramagno, T. C. (2002). Irreconcilable differences? Intellectual stalemate in the gay rights debate. Westport, CT: Praeger.Find this resource:

                  Cohen, J. (2017). Sovereignty, the corporate religious, and jurisdictional/political pluralism. In C. Laborde & A. Bardon (Eds.), Religion in liberal political philosophy (pp. 83–102). New York, NY: Oxford University Press.Find this resource:

                    Christian Legal Society v. Martinez, 561 U.S. 661 (2010).Find this resource:

                      Corporation of the Presiding Bishop of the Church of Latter-Day Saints v. Amos, 483 U.S. 327 (1987).Find this resource:

                        Craig v. Masterpiece Cakeshop, Inc., 370 P. 3d 272 (Colo. App., 2015).Find this resource:

                          Creppell, I. (1996). Locke on toleration: The transformation of constraint. Political Theory, 24(2), 200–240.Find this resource:

                            Dane, P. (2018). Scopes of religious exemption: A normative map. In K. Vallier & M. Weber (Eds.), Religious exemptions (pp. 138–164). New York, NY: Oxford University Press.Find this resource:

                              Elane Photography v. Willock, 309 P. 3d 53 (N.M. 2013).Find this resource:

                                Elane Photography v. Willock, 134 S. Ct. 1787 (2014).Find this resource:

                                  Feldblum, C. R. (2008). Moral conflict and conflicting liberties. In D. Laycock, A. R. Picarello, Jr., & R. F. Wilson (Eds.), Same-sex marriage and religious liberty: Emerging conflicts (pp. 123–156). Lanham, MD: Becket Fund for Religious Liberty/Rowman & Littlefield.Find this resource:

                                    Flanders, C. (2016). Religious organizations and the analogy to political parties. In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 103–122). New York, NY: Oxford University Press.Find this resource:

                                      Flynn, T. (2010). Clarion call or false alarm: Why proposed exemptions to equal marriage statutes return us to a religious understanding of the public marketplace. Northwestern Journal of Law and Social Policy, 5, 236–259.Find this resource:

                                        Garnett, R. W. (2016). The freedom of the church: (Toward) an exposition, translation, and defense. In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 39–62). New York, NY: Oxford University Press.Find this resource:

                                          Gill, E. R. (2001). Becoming free: Autonomy and diversity in the liberal polity. Lawrence: University Press of Kansas.Find this resource:

                                            Gill, E. R. (2012). An argument for same-sex marriage: Religious freedom, sexual freedom, and public expressions of civic equality. Washington, DC: Georgetown University Press.Find this resource:

                                              Gill, E. R. (2013). Religion, civic values, and equal citizenship in the liberal democratic polity. Politics and Religion, 7(2), 235–260.Find this resource:

                                                Gill, E. R. (2014). Democracy: A paradox of rights? In A. E. Cudd & S. J. Scholtz (Eds.), Philosophical perspectives on democracy in the 21st century (pp. 15–27). Dordrecht, The Netherlands: Springer.Find this resource:

                                                  Griffith, R. M. (2017). Moral combat: How sex divided American Christians and fractured American politics. New York, NY: Basic Books.Find this resource:

                                                    Gutmann, A. (2003). Identity in democracy. Princeton, NJ: Princeton University Press.Find this resource:

                                                      Hands On Originals v. Lexington-Fayette Urban Cnty. Human Rights Commission, No. 14-CI-04474 (Ky. Cir. Ct., Fayette Cnty) (April 27, 2015).Find this resource:

                                                        Horwitz, P., & Tebbe, N. (2016). Religious institutionalism—why now? In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 207–226). New York, NY: Oxford University Press.Find this resource:

                                                          Inazu, J. D. (2016). Confident pluralism: Surviving and thriving through deep difference. Chicago, IL: University of Chicago Press.Find this resource:

                                                            Karst, K. L. (1989). Belonging to America: Equal citizenship and the constitution. New Haven, CT: Yale University Press.Find this resource:

                                                              Koppelman, A. (2015). Gay rights, religious accommodations, and the purpose of an antidiscrimination law. Southern California Law Review, 88, 619–659.Find this resource:

                                                                Koppelman, A., & Wolff, T. B. (2009). A right to discriminate? How the case of Boy Scouts of America v. James Dale warped the law of free association. New Haven, CT: Yale University Press.Find this resource:

                                                                  Kukathas, C. (2003). The liberal archipelago: A theory of diversity and freedom. Oxford, U.K.: Oxford University Press.Find this resource:

                                                                    Laborde, C. (2017). Liberalism’s religion. Cambridge, MA: Harvard University Press.Find this resource:

                                                                      Laycock, D. (2008). Afterword. In D. Laycock, A. R. Picarello, Jr., & R. F. Wilson (Eds.), Same-sex marriage and religious liberty: Emerging conflicts (pp. 189–207). Lanham, MD: Becket Fund for Religious Liberty/Rowman & Littlefield.Find this resource:

                                                                        Laycock, D. (2016). The campaign against religious liberty. In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 231–255). New York, NY: Oxford University Press.Find this resource:

                                                                          Levy, R. A. (2016). Libertarianism and the right to discriminate. Cato Policy Report, 38 (March/April), 2.Find this resource:

                                                                            Loving v. Virginia, 388 U.S. 1 (1967).Find this resource:

                                                                              Lupu, I. C., & Tuttle, R. W. (2014). Secular government, religious people. Grand Rapids, MI: William B. Eerdmans.Find this resource:

                                                                                Masterpiece Cakeshop v. Colorado Civil Rights Commission, 584 U.S. (2018).Find this resource:

                                                                                  McClure, K. (1990). Difference, diversity, and the limits of toleration. Political Theory, 18(3), 361–391.Find this resource:

                                                                                    Mezey, S. G. (2017). Beyond marriage: Continuing battles for LGBT rights. Lanham, MD: Rowman & Littlefield.Find this resource:

                                                                                      Neal, P. (1997). Liberalism and its discontents. New York, NY: New York University Press.Find this resource:

                                                                                        Obergefell v. Hodges, 135 S. Ct. 2584 (2015).Find this resource:

                                                                                          Pedreira et al. v. Kentucky Baptist Homes for Children, Inc., 579 F. Supp. 3d (6th Circuit 2009).Find this resource:

                                                                                            Richards, D. A. J. (1999). Identity and the case for gay rights: Race, gender, and religion as analogies. Chicago, IL: University of Chicago Press.Find this resource:

                                                                                              Richards, D. A. J. (2005). From Bowers to Lawrence and beyond. Lawrence, KS: University Press of Kansas.Find this resource:

                                                                                                Romer v. Evans, 517 U.S. 620 (1996).Find this resource:

                                                                                                  Rosenblum, N. L. (1998). Membership and morals: The personal uses of pluralism in America. Princeton, NJ: Princeton University Press.Find this resource:

                                                                                                    Rosenblum, N. L. (2000). Amos: Religious autonomy and the moral uses of pluralism. In N. L. Rosenblum (Ed.), Obligations of citizenship and demands of faith (pp. 165–195). Princeton, NJ: Princeton University Press.Find this resource:

                                                                                                      Shelley v. Kraemer, 334 U.S. 1 (1948).Find this resource:

                                                                                                        Smith, S. D. (2016). The jurisdictional concept of church autonomy. In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 19–37). New York, NY: Oxford University Press.Find this resource:

                                                                                                          Spinner-Halev, J. (2011). A restrained view of transformation. Political Theory, 39(6), 777–784.Find this resource:

                                                                                                            Tebbe, N. (2017). Religious freedom in an egalitarian age. Cambridge, MA: Harvard University Press.Find this resource:

                                                                                                              Wilson, R. F. (2008). Matters of conscience: Lessons for same-sex marriage from the healthcare context. In D. Laycock, A. R. Picarello, Jr., & R. F. Wilson (Eds.), Same-sex marriage and religious liberty: Emerging conflicts (pp. 77–102). Lanham, MD: Becket Fund for Religious Liberty/Rowman & Littlefield.Find this resource:

                                                                                                                Wilson, R. F. (2016). Bargaining for religious accommodations. In M. Schwartzman, C. Flanders, & Z. Robinson (Eds.), The rise of corporate religious liberty (pp. 257–283). New York, NY: Oxford University Press.Find this resource: