Americans support religious liberty—in general. But they are deeply polarized about how far the natural and constitutional right of individuals to respond to their conceptions of the divine should extend. And unfortunately, Americans tend to be reluctant to extend religious liberty broadly to views they find unsympathetic.

I think that’s sad. Religious liberty is for everyone and should be cherished by all. It’s also ironic, as I argue in my new book, Religious Liberty in a Polarized Age, because historically, the central social purpose of religious liberty was to reduce the fear and anger people feel when they’re threatened with penalties for living according to their religious commitments.

Fear and anger produced cycles of violent retaliation in 16th and 17th century Europe among Protestants and between Catholics and Protestants. In response, Americans embraced principles of religious liberty. The founding father James Madison called it the “true remedy” for the “disease” of religious conflicts and their threat to “the health and prosperity” of the nation.

Today’s conflicts between progressives and conservatives are, thankfully, less violent. Yet we also see cycles of coercion, fear, resentment, and retaliation. We also live in an age when people’s response to “ultimate concerns” vary greatly and are often understood in opposition to each other. Progressives sometimes seek to compel conservative religious people or groups to support same-sex marriages or transgender procedures, in violation of their consciences. Conservative Christians sometimes seek to secure privileges for Christianity, forcing acknowledgements from those who aren’t Christian. And minority religions, notably Islam and Native American faiths, often find their religious practices sharply curtailed, surveilled, and treated with suspicion.

Religious liberty can be an answer to the polarization that pushes people further and further into opposing, hostile camps. It can counter cycles of fear and retaliation in a society where people believe different things. But to achieve this, religious liberty must have three features.

First, it must be strong. It must protect people’s ability not just to hold beliefs but to live them out consistently. This has to include public settings, including schools, social services, and the workplace.

Second, religious liberty must be equal for all faiths. Religious liberty is only meaningful if it protects those who need protection. This includes religious conservatives, when people are hostile towards them. And it includes Muslims, Sikhs, Jehovah’s Witnesses, and everyone else.

Article continues below

Finally, religious liberty must consider other important, competing interests, including nondiscrimination interests. Protecting religious liberty will reduce suffering and fear, but not if it goes so far as to allow serious impositions on others. With careful analysis, courts, legislatures, and administrative bodies can reach conclusions that maximize these goals.

There is developing evidence that strong religious liberty can command bipartisan support if it’s even-handed and sensibly defined and defended. Scholar and advocate Asma Uddin, who emphasizes that Muslims and evangelicals have common interests in religious liberty, describes how political scientist Andrew Lewis surveyed 1,100 respondents in 2018 on whether they’d support a provider’s right to refuse to design a cake for a same-sex wedding.

Liberals opposed that right, Lewis found—but less so when first presented (“primed”) with a Muslim truck driver’s claim to refuse to deliver beer or alcohol. They were more receptive to the evangelical claim, in Uddin’s words, “once they saw the issue through the perspective of the Muslims’ religious freedom case.” Once they saw, that is, that the same principle defends a minority with whom they sympathize.

Support for religious liberty claims also grows when the key limiting facts of cases are made clear. For example, with respect to small businesses and same-sex weddings, the Public Religion Research Institute asks respondents broadly whether businesses should be able “to refuse to provide products or services to gay or lesbian people” and finds, unsurprisingly, that clear majorities of up to 65 percent “have consistently opposed” that refusal. But when the Becket Fund asks whether a vendor can refuse a specific wedding-related activity—for example, an Orthodox Jewish restaurant declining to cater a reception or a website designer declining to create a wedding site—it finds that around 70 percent support religious liberty. And the claims in the real cases involve specific services. Those who went to court did, in fact, provide other “products and services to gay or lesbian people.”

There’s other evidence that religious liberty can command broad support even in this present period of polarization. The last year has seen two encouraging cases of religious-liberty bipartisanship.

Article continues below

First, bipartisanship prevailed when Congress passed the Respect for Marriage Act in December 2022. The act protects same-sex civil marriages, requiring states to recognize marriages contracted in another state. That’s an insurance policy for same-sex couples who feared the Supreme Court might overrule its Obergefell decision, which held that states must recognize same-sex marriages. Simultaneously, the act protects religious organizations with objections to gay marriage. It makes clear that it does not diminish their rights or their eligibility for government benefits; it also acknowledges that traditional beliefs about sex and marriage rest on “decent and honorable” premises and are “due proper respect.” Congress protected same-sex marriages and went on record rejecting the equation of beliefs in male-female marriage with racist beliefs against interracial marriage.

Some religious conservatives found these protections insufficient and opposed the act. But opponents, I predict, will end up using these provisions to bolster protection for their own organizations.

Second, a unanimous Supreme Court held this June that federal law requires employers to accommodate workers’ religious practices unless the accommodation causes “substantial” costs or disruption to employer operations. The decision, Groff v. Dejoy, rejected a much weaker standard that many courts had articulated for decades. Groff will strengthen protection for all faiths: Christians or Jews objecting to Sabbath work, Muslims seeking reasonable breaks to perform daily prayers, Sikhs or others seeking to wear beards or religious headdress. The case involved a Christian, but the holding will effectively protect the religious practice of many, many people, including those from the faiths most marginal in society. It’s exactly the kind of ruling that enables us to live together, though we have significant differences.

Of course, there remain many cases of sharp division. Bipartisanship on the Court broke down in 303 Creative v. Elenis, where the 63 majority held that website designer Lorie Smith could refuse to provide websites for same-sex weddings while providing them for male/female weddings. Because Smith’s customized wedding sites would celebrate the particular marriages, the Court said, requiring her to create sites for same-sex weddings would unconstitutionally compel her to speak inconsistently with her beliefs or else give up significant business.

Article continues below

The decision is correct, although analyzing it requires some care. In weighing First Amendment and nondiscrimination interests, my book argues, courts and legislatures should provide broad protection for religiously-grounded nonprofits. Such organizations give notice of their religious commitments to prospective clients or employees, and in most cases those persons can find alternatives.

Small businesses providing ordinary goods and services also deserve protection, because people have a strong interest in living consistently with their beliefs in their workplaces or professions, which occupy much of their time, energy, and identity. But protections must be narrower and carefully defined. It’s also important that all people have access to goods and services without the constant insecurity that providers may refuse them. And refusals may happen more often because ordinary commercial providers, unlike religious entities, do not inherently give notice that their beliefs may affect their provision of goods or services.

So commercial exemptions, I argue, should be limited to small businesses, personally involving the individual believer who objects only to a specific practice and not to serving LGBT people in general, where there are ample alternative providers. In these cases, the First Amendment interests are at their strongest and the nondiscrimination interests far from their strongest.

Smith fit that category. The state stipulated that she would provide expressive, customized websites celebrating the particular couple’s marriage; a nondiscrimination rule would force her to speak in conflict with her beliefs. The state also stipulated that she would provide non-wedding websites for LGBT clients, and that there were “numerous” alternative wedding-site providers.

Smith rightly prevailed. But commercial exemptions must be carefully defined, and the Court’s opinion fell short in a couple of ways. One significant example: Because Smith sued before she entered the wedding business, there was no full litigation record on the extent of expressiveness in her work, only the stipulations that it would be “expressive” and “custom-tailored.” That hampers drawing lessons for what’s “expressive” in future cases, such as for a wedding-gown or jewelry designer.

Article continues below

Moreover, although there were ample alternative providers, the Court did not expressly say this was necessary to the holding. The issue might matter in rural areas with limited providers, where one refusal could significantly limit a couple’s access to certain wedding services.

303 Creative is correct, but later decisions may have to qualify some passages to ensure it doesn’t undercut nondiscrimination laws broadly.

Nondiscrimination rules with significant religious exemptions can protect both sides in our polarized debates and reduce their sense of threat. The Respect for Marriage Act teaches that I’m more likely to achieve protection for my side’s claims if I also make room for the other side’s. Conservatives can support equal treatment of LGBT people in many contexts (employment, housing, commercial services) without agreeing with the behavior. Progressives can support religious freedom in many contexts without agreeing with conservative views.

Transgender-equality laws are complicated by other issues: effects on women’s sports and domestic-violence shelters, as well as medical interventions before adulthood. But religious-freedom concerns need not block the enactment of core nondiscrimination laws—if, and only if, there are significant religious exemptions.

In this and other ways, religious liberty can again serve its historic purpose of reducing suffering, fear, resentment, and conflict.

Thomas C. Berg has advocated for religious liberty for evangelical Christians and other faiths for three decades. He teaches at the University of St. Thomas School of Law (Minnesota), where he directs the Religious Liberty Appellate Clinic.

[ This article is also available in español. ]