Religious Freedom Under the First Amendment: Three Supreme Court Cases and the Ambiguous Term “Religion”

Throughout the years, and in various Supreme Court cases, the distinction between “religious/sectarian” and “nonreligious/secular” has been rather ambiguous. In this essay, I will examine three separate Court cases in which the Court had to defend its verdict by employing what I deem “ambiguous” uses of the term “religion.” Moreover, I will argue that “religion” as a phenomenon is virtually impossible to define in any concrete, rigid manner. Given this reality, the Court’s decisions, when attempting to demarcate the line between that which is religious and nonreligious, will always remain blurry. Hence, it is my position that ambiguity will remain ever present in the their decisions so long as the Court continues to deal with an ambiguous phenomenon[1] known as “religion.”

Before examining the three cases, I will first begin by looking at the First Amendment and the surrounding historical context in which it was shaped, a context, as we shall later see, that set the trend for the Court’s various positions on “religion.”

The First Amendment was shaped in the 18th century during a time when several principles were deemed essentially conducive to a peaceful, well-governed society. The principles were: (1) liberty of conscience; (2) free exercise of religion; (3) religious pluralism; (4) religious equality; (5) separation of church and state; and (6) disestablishment. “While many of these terms carried multiple meanings in the later eighteenth century and several other terms were under discussion, these six principles were foundational for the American founders.”[2] The First Amendment—an amendment originally governing only Congress—was first applied to states and local governments via the Fourteenth Amendment’s due process clause in the pioneering case of Cantwell v. Connecticut (1940).[3] “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.”[4] The Founding Fathers initially feared a particular religious institution so close to the State that it would use the State to persecute any dissenting voices.[5] However, in their attempts to articulate a form of government that allowed the flourishing of religion, the Fathers left one fatal void: they failed to define “religion.” What constitutes a religion? Witte writes, “Nowhere is the word ‘religion’ defined in the Constitution or Bill of Rights…”[6] In fact, if “original intent” is observed, it becomes relatively clear that by employing the term “religion” the Fathers meant “a plurality of Protestant Christian faiths.”[7] That is, they probably did not mean to defend the religious freedom rights of Muslims, Buddhists, Hindus, or even Catholics. Nonetheless, a few scattered remarks from this time period do exist which help us understand what “religion” was thought to be. In 1802, Thomas Jefferson wrote, “[R]eligion is a matter which lies solely between a man and his God.”[8] Here “religion” was thought to be (a) a private affair and (b) involving a person and some deity. On June 26, 1788, during the Virginia convention on the Constitution, the authors wrote revealingly: “Religion, or the duty which we owe our creator, and the manner of discharging it…”[9] Here it can be seen that “religion” was thought to be something between a person and his or her deity/Creator. What is alarming in these two remarks is the lack of precise terminology. As we shall see, the modern day Court—from the 1940s onwards—has continued to wrestle with its definition of “religion,” having inherited this ambiguous legacy. I now turn my attention to three modern-day cases in which the demarcation between religious and nonreligious has continued, as in the past, along ambiguous lines.

In Frazee v Illinois (1989), the Court decided a case that involved a certain William Frazee who refused to accept a retail position that involved work on Sundays. He claimed that, as a Christian, it was unlawful for him to work on the Lord’s Day. Frazee later applied for unemployment benefits and was denied. Consequentially, the Department of Employment Security’s Board of Review justified its refusal to grant unemployment benefits to Frazee by stating: “When a refusal of work is based on religious convictions, the refusal must be based upon some tenets or dogma accepted by the individual of some church, sect, or denomination, and such a refusal based solely on an individual’s personal belief is personal and noncompelling and does not render the work unsuitable” [489 U.S. 829, 831] App. 18-19.[10] In a rather fortunate series of events, the Supreme Court picked up this case and overturned the earlier decisions made by the lower courts. Justice White, arguing for the majority opinion, wrote:

“While membership in a sect would simplify the problem of identifying sincerely held beliefs, the notion that one must be responding to the commands of a particular religious organization to claim the protection of the Free Exercise Clause is rejected. The sincerity or religious nature of appellant’s belief was not questioned by the courts below and was conceded by the State, which offered no justification for the burden that the denial of benefits placed on appellant’s right to exercise his religion.”[11]

Essentially, the Court said that while it may be true that Frazee was not a part of any church or sect—for all they knew, he might have stayed home on Sundays only to watch Oprah and eat Bon-Bons—nonetheless, it was not the State’s job to verify the sincerity of religious beliefs, or, for that matter, to act as an arbiter in religious affairs. Put simply: if a person stated they were Christian, it was beyond the State’s power to attempt to prove or disprove the sincerity of those beliefs. The State was not a religious organization, and so could not pass judgment on the sincerity of any deeply held—or, for that matter, deeply faked—religious beliefs.

In the above case we see, once again, a continuation of ambiguity when it comes to the subject of religion. Mr. Frazee was not a part of any church or religious organization. And yet the Court overturned an earlier denial of unemployment benefits on the basis that work on Sundays, for Frazee, was an unnecessary burden on his allegedly religious conscience. Justice White wrote, regarding the difficult process of demarcation between religious and secular, “Nor do we underestimate the difficulty of distinguishing between religious and secular convictions and in determining whether a professed belief is sincerely held.” Could one create a religion out of thin air, claim a free exercise violation, and win? In the post-Frazee v Illinois world, it seems so. For here—as much as ever—the term “religion” is not clearly demarcated from the secular/nonreligious. If staying home and watching football on Sundays is, at some future point, considered to be a “religious act,” who would blame the Court for not knowing what to do? Nobody seems to know what religion/religious is to begin with. Next, I will look at yet another pesky issue: just how cozy could the secular State get with religious holiday displays?

Lynch v Donnelly (1984) was a case settled after the groundbreaking Lemon v Kurtzman (1971). In Lemon v Kurtzman the three-pronged “Lemon test,” a test used to determine whether a law had the effect of establishing a religion, was first formulated.[12] In the case we are now considering—namely, Lynch v Donnelly—the city of Pawtucket, R. I. came under fire for erecting a Christmas display on private property owned by a nonprofit organization, property located directly in the center of the city’s shopping district. Amongst the Santa Claus house, Christmas tree and other such holiday objects, there was also placed a crèche, or nativity scene. This crèche was challenged for being an “establishment clause” violation: the State, funded by diverse taxpayers, was using its funds to “promote” a single religion, Christianity. The case ended up going to the Supreme Court, where the Court concluded, “Pawtucket has not violated the Establishment Clause.”[13] What were the Court’s reasons for reaching this verdict?

The Court argued that the now-famous concept of a wall of separation between church and state was a “useful metaphor” but “not an accurate description of the practical aspects of the relationship that in fact exists.” In addition to this, the Court argued that the Constitution did not, in fact, “require complete separation of church and state”; rather, “it affirmatively mandates accommodation…”[14] The Court also recognized how ubiquitous religion was. Religion was a part of the “American life.” Because it was the Christmas season, and because the crèche alone was not the singular focus of the Christmas display, the Court—echoing the “Lemon test”—ruled that “the city has a secular purpose for including the crèche in its Christmas display and has not impermissibly advanced religion or created an excessive entanglement between religion and government.”

As can be seen, the Court argued that religion was already mixed into the American way of life, thus admitting that the line between religious and secular was to be found “in the mix” somewhere. In other words, there wasn’t much of a line to begin with. Nonetheless, the Court still attempted to make that line materialize. Somehow, because of the “national tradition” and our desire to “depict the origins of that Holiday,” Christmas had become a rather secular holiday, with displays serving “legitimate secular purposes.” While the Court was busy employing the terms “secular” and “religious” without defining them, they had also snuck in some Orwellian double-think by referring to Christmas as both a “religious” and, finally, a “secular” holiday. And so the ambiguity continues.

I now want to turn my eyes to my final case. In Employment Division v Smith (1990) the Court back peddled on the “accommodationist logic” it used in Lynch v Donnelly. In this case, the defendants were two members of the Native American Church fired from their place of employment for using peyote on religious grounds. Once fired, they applied for unemployment benefits and were denied. The Oregon Supreme Court initially ruled that denying them unemployment benefits for using peyote on religious grounds violated their right to exercise religion; however, the state refused to pay out the benefits because possession of peyote was deemed a crime—so the case went to the Supreme Court. The Court focused, citing Sherbert v Verner, on whether the employees had a “constitutional right to unemployment benefits on the part of all persons whose religious convictions are the cause of their unemployment.”[15] Smith, one of the members who appealed to the Supreme Court, argued that he was doing nothing different than what we saw done in Frazee. That is, “[i]f Frazee could get unemployment compensation for refusing to work on Sunday, his day of rest but not worship, Smith argued, surely he could get compensation for being fired for engaging in the arduous and ancient religious ritual of peyote ingestion.”[16] The Court, however, was not in agreement with Smith. On the contrary, they argued that this case should be treated not as a case dealing with unemployment per se but rather as a case dealing with “free exercise” and compliance with “criminal laws.” In fact, the Court argued that Oregon State’s law regarding the illegal use of drugs (or which peyote was one) was “neutral” and “generally applicable”; hence, differing from the prior cases such as Frazee, the Court now argued that it was possible for the State to cast a burden upon a religious person so long as it was doing so by means of a generally applicable law that did not single out any particular person or religion.[17] Using the Court’s logic in Smith and applying it to Frazee one could argue that Frazee did not deserve unemployment compensation since he refused to work on Sundays—and “mandatory Sunday-work is required of everyone, being generally applicable to all, religious or irreligious.”[18] Such a statement, however, was not made in Frazee. Why?

Returning to the second case I looked at—that is, Lynch v Donnelly—allow me to remind you that in that case that which was secular and that which was religious was comingled. In Lynch the religious became the secular by means of “tradition.” Since what was initially religious had been around so long, it was no longer really religious; it was, in fact, perfectly secular. “Christmas is not really a religious holiday; it is mostly a secular holiday with ancient, religious roots. But most of us don’t focus on the religious element, so it’s basically secular,” went the argument.

But not so in Smith. Here a couple of men, who were unquestionably religious, were not allowed to exercise their religious beliefs. Like Christians partaking of the Lord’s Supper—sipping on a toxin known as alcohol[19]—the men involved in the Smith case could not exercise their beliefs. Why? Because the state thought their use of peyote, even in what was deemed a purely religious ritual, to be illegal. The line between religious and secular was assumed throughout the Smith case; there was no question that the two men were participating in a religious act. However, the relationship the State had with their so-called “religious activities” was vastly different than its cozy relationship with the mostly Christian activities we saw in Frazee and Lynch. In these cases, whatever was found to be religious was either explained away as the mostly secular (Lynch) or deemed impossible to verify (Frazee)—in both cases the Court allowed the religious to exercise their religious beliefs, no matter how fake (Frazee) or how assimilated into the secular culture (Lynch). What we saw in Smith, however, was what appeared to be a rather concrete, underlying assumption that the Court understood what it meant for something to be “religious.” But even here the “religious” was never defined. And so, despite the dogmatic rhetoric, the Court has yet to define what it means for something to be a religious act or a religion.

In 1912, James H. Leuba published a seminal paper that included an oft-cited appendix listing more than fifty definitions of religion.[20] Today, more than ever, the religious is ubiquitous—we see it in law, in politics, in science classrooms, in our libraries, in our churches, etc. As then, so now, we don’t really know what religious really means—if anything at all. There are a multitude of definitions available to us. Some, like the Founding Fathers, may see religion as that which involves some deity/Creator. Others, like Buddhists, may argue that no such deity is required by religion. Still others may argue that no such thing as God exists. Some may think a church or synagogue plays an essential part in what it means for something to be deemed religious; others, like Frazee, argue that religious acts do not have to involve such structures. Some may argue that religion has so infiltrated our society, it is no longer possible to clearly separate the two (e.g., Lynch). Some may argue that religion is relatively straightforward, involving the use of chemical substances; practices that the State could, in theory, forbid (e.g., Smith). In all of these various cases, involving a plurality of definitions, the distinction between religious and nonreligious, sectarian and secular, remain forever indistinct to our eyes as we gaze into that abysmal sea of religious discourse “through a glass, darkly” (1 Cor. 13:12). So long as the Supreme Court continues to deal with this most notoriously difficult of issues—that is, the ambiguous phenomenon we call religion—so long will we be haunted by paradoxical court cases and unclear decision-making processes.


Written by Moses Y. Mikheyev

Dedicated to John Witte, Jr.




[1] It is entirely possible to argue that my use of the term “religion” in itself is already misleading; instead, it may be argued, that what I should have written should have been the plural “religions.” However, I use the term colloquially: it encompasses all and every “religion,” whether the various religions have anything in common or not. (Even here one detects a thorough-going ambiguity: what, in fact, do all religions have in common? Or do we just group various phenomena that appear to be ceremonial as being “religious”? What, then, is “religion”?)

[2] John Witte, Jr. and Joel A. Nichols, Religion and the American Constitutional Experiment, 4th ed. (New York: Oxford University Press, 2016), 62-3.

[3] Ibid., 98-9.

[4] Ibid., 1.

[5] Ibid., 30-1.

[6] Ibid., 95.

[7] Ibid.

[8] Ibid., 56.

[9] Ibid., 74.

[10] Frazee v. Illinois Dept. of Employment Security, 489 U.S. 829 (1989), URL=

[11] Ibid.

[12] The three-pronged approach is as follows: “a challenged law must (1) have a secular purpose, (2) have a primary effect that neither advances nor inhibits religion, and (3) foster no excessive entanglement between church and state” (Witte and Nichols, Religion, 163).

[13] Lynch v. Donnelly, 465 U.S. 668 (1984), URL=

[14] Italics mine.

[15] Employment Division, Department of Human Resources of Oregon v. Smith, 494 U.S. 872 (1990), URL=

[16] Witte and Nichols, Religion, 146.

[17] Ibid., 146-7.

[18] The words in quotation marks are theoretical, in case that was not made clear.

[19] “Respondents contend that the sacramental use of small quantities of peyote in the Native American Church is comparable to the sacramental use of small quantities of alcohol in Christian religious ceremonies” (Employment Division, Department of Human Resources of Oregon v. Smith, 494 U.S. 872 [1990], URL=

[20] Jonathan Z. Smith, “Religion, Religions, Religious,” in Critical Terms for Religious Studies, ed. Mark C. Taylor (Chicago: University of Chicago Press, 1998), 281.

Choose to create a comment on this existential blog...

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s