There and Back Again: The Supreme Court’s Establishment Clause Journey

‘Tis the season to talk about the Establishment Clause. This time each year, lawyers for aggrieved grinches sue towns and school districts to stifle evidence of Christmas. As one federal judge wrote in 2018: 

Since ancient times, people have been celebrating the winter solstice, which occurs around the third week of December in the Northern Hemisphere. Many of these celebrations are religious in nature, and so in the modern United States they have led to a depressingly steady stream of First Amendment challenges, in which one party wishes to express its religious views in the public sphere and the other party asserts that the Establishment Clause would be violated by the display.

The wellspring of the “steady stream” has been the Supreme Court’s radically secularist interpretation of the First Amendment provision against “laws respecting an establishment of religion.” The Court’s master rule since around 1960 was this: no public authority may aid or promote, or “endorse,” religion as an objective good, even if doing so involved no favoritism towards any faith. The Court said in case after case that the government must be scrupulously “neutral,” not only among the various religions but between religion and what the Court called “non-religion” or “irreligion.” The desired practical effect was unmistakable. It was the naked public square that Father Richard Neuhaus cogently critiqued in his 1984 book by that title.

The Court took up its first Christmas case that year, a fight over the nativity scene erected by Pawtucket, Rhode Island. A majority of the Justices blanched at the implication that the Constitution would evict Baby Jesus. Speaking through Chief Justice Burger, the Court instead chose to recognize “an unbroken history of official acknowledgment by all three branches of government of the role of religion in American life from at least 1789.” The Court relied upon a historical exception to its secularistic uber-norm. The creche survived.

Pawtucket was not the first case where a page of history triumphed over volumes of Establishment Clause doctrine. In 1961, the Court upheld Sabbath “blue laws,” and in 1970, it validated tax exemptions for church properties against doctrinal headwinds, because both were fixtures of American history. Later, the Court upheld on historical grounds legislative prayer and public religious monuments. 

Thus, the great paradox, or tragedy, of Establishment Clause jurisprudence: the Court justified its secularist rules by deep dives into our nation’s history, and proclaimed itself humble herald of what the founders unequivocally commanded. Yet the justices recognized simultaneously that their secularism was not ours, that their rules were utterly incompatible with American history.  

In 2022, the Court finally acknowledged that the exceptions swallow the rule. Upholding a public high-school football coach’s post-game prayer routine in Kennedy v. Bremerton School District, the Court wrote that its doctrinal “approach” to the Establishment Clause was too “abstract, and ahistorical.” The Court said that it had “long ago abandoned Lemon,” the aptly named 1971 opinion where the justices codified their secularist ideology. This is surely untrue. As Judge Jay Richardson of the Fourth Circuit recently wrote, the cases that allegedly “abandoned” Lemon said no such thing. Nonetheless, “it is now clear,” Richardson added, that “Lemon and its ilk are not good law.” This is surely true.

Now it is to be all history and no doctrine. The Kennedy Court wrote that “the Establishment Clause must be interpreted by reference to historical practices and understandings” and “The line that courts and governments must draw between the permissible and the impermissible has to accord with history and faithfully reflect the understanding of the Founding Fathers.” (Internal quotation marks and brackets omitted.) 

This about-face is unprecedented, as Michael McConnell and Nathan Chapman observe in their timely and invaluable new book, Agreeing to Disagree. “No other constitutional provision has experienced such interpretive tergiversation.” These authors write that from now on Establishment Clause cases are to be “governed by a collection of context-specific rules—many of them the virtual opposite of the prior understandings” now “abandoned.” They are right.

We stand at the dawn of a new era in an important realm of constitutional law. As we step into this new dispensation, Agreeing to Disagree will serve well as a road map and guidebook to what comes.

Agreeing to Disagree is a synoptic treatment of the history of (dis)establishment in America and the Court’s wayward interpretations of the non-establishment clause. It is nearly encyclopedic in its coverage. Yet it is succinct, clear, and remarkably jargon-free. The authors are accomplished law professors; McConnell is also a former federal appellate judge. They admirably eschew the guild’s practice of endless footnoting. The book is perhaps best grasped as an extraordinarily good primer, if that description could be predicated of a work of such analytic sophistication. It is that rare book that repays careful reading as much by the intelligent non-specialist as by the expert.

One reason why the authors deliver so much in so few pages is that they leave aside the currents of philosophical literature (“Locke, Hobbes, Spinoza, Rousseau, Bayle, Mill, and other[s]”) in which so many works on the Establishment Clause founder. “[T]his book will focus on the establishment of religion as law” (my emphasis).

To the authors’ great credit, they scrupulously flag their editorial asides. For example: in the chapter about judicial resolution of church-property disputes, they describe the competing paradigms fairly. But after comparing the “deference” approach (the one more solicitous of hierarchical churches) to the more egalitarian (in their view) “neutral principles” test, they say frankly: “[Y]our authors believe that [the latter] approach is most consistent with Establishment Clause principles.”

McConnell and Chapman rightly say that “the correct interpretation of the Establishment Clause . . . matter[s],” for it “tells us much about the character of the American republic and our political culture.” Indeed, it does. The clause matters because religion matters to persons, to communities, and to our polity. Are we “one nation, under God” or not? Are we a “religious people whose institutions presuppose a Supreme Being,” as Justice William Douglas wrote for the Supreme Court in 1952, or not? Does our “happiness” as well as our “good government” depend on religion, as the Founders wrote in the Northwest Ordinance?

The first part of Agreeing comprises short chapters chronologically sequenced on the colonial inheritance of “establishments,” the making of the Establishment Clause, state disestablishment, and the “incorporation” of the Clause; that is, whether the Fourteenth Amendment was meant to apply it to the states, as opposed to the federal government alone, which was the antebellum understanding. (The Court “incorporated” it in 1947.)

The bulkier second part is topically organized. After a very short chapter on the “Rise and Fall of the Lemon Test,” the authors turn to “accommodation of religion.” Think of an exception to “must carry” contraception laws for, say, a Catholic pharmacist. McConnell and Chapman include a careful analysis of an increasingly prominent objection to such accommodations: “third-party harm.” The basic idea here is that it violates the Establishment Clause if “accommodating” that pharmacist imposes more than minimal inconvenience on someone else. This “non-beneficiary” could be a woman seeking the pill unsuccessfully from the local Walgreens. Why should she have to bear any cost for the sake of someone else’s religion?

The authors’ reply to this objection concedes that it is correct at “a high level of abstraction.” I rather think not; at least, it is incorrect at every level if one affirms (as one should) that religion is a basic good for everyone and therefore part of the polity’s common good. Then prima facie the community may rightly call upon us to make sacrifices for others’ benefit vis-à-vis such basic goods. After all, we pay for others’ kids’ educations and the elderly’s healthcare and for the poor’s food stamps. Why should not similar sacrifices be in principle justifiable if needed to help others to access and enjoy the good of religion? 

The authors affirm the permissibility under the Establishment Clause of legislative exemptions imposing “third-party harm.” They note rightly that the most obvious example has been the tradition of draft exemptions for conscientious objectors. The authors say that current doctrine requires only that governments that grant exceptions take “adequate account” of third-party burdens. I should think it better to express the moral calculus here as a duty of fairness to all those affected by this (or any other) government action. 

McConnell and Chapman say also that “[b]ehind this [third-party harm] argument . . . is a confusion over baselines.” Maybe so. Their supporting claim that these “accommodations” amount to “weighing . . . private harms, [which] is a paradigmatic question of policy rather than a question of law” may be right in suggesting that exemptions are generally best left to legislative rather than to judicial resolution. But “policy,” “weighing,” and “private harms” suggest that the lawmaker would operate without the critical moral guidance supplied by the truth that religion is a paramount public good and that using contraceptives is not.

Agreeing includes a masterly chapter on “No-Aid Separation,” particularly on the Court’s blockade of public assistance to religious schools from 1971 until 1985. The authors note that the Court’s 1973 Nyquist decision marked a turning point. There the Court rejected what the justices conceded were lawmakers’ sincere attempts to combat poverty and creeping illiteracy in largely minority urban schools in urban centers. No matter: even a penny in aid of a religious enterprise was one more than the Constitution allowed.

The Court’s key analytical move here was just as the authors describe it: “from whether the aid is itself secular in nature [for example, previously approved bus rides to and from school or state-approved textbooks in non-religious subjects] to whether the aid benefits” an institution which has a religious component—what we might now label “identity.” They are right that this whole misguided project reached a “high-water mark” in 1985. “With astonishing speed, the Court repudiated this entire inconsistent and discriminatory line of doctrine. In all the annals of the U.S. Reports, there is no example of a more complete volte-face in constitutional doctrine.” 

One of the greater ironies in these cases was that the most liberal Justices—Brennan, Douglas, and Marshall—were the most relentless opponents of what were, in chief effect, Great Society anti-poverty government initiatives.

In their chapter on religion in public schools, the authors take up the 1962 Engel v. Vitale decision. There, the Court struck down a non-denominational school prayer in what remains the most unpopular church–state decision in the Court’s history. Justice Black wrote for the majority: “Neither the fact that the prayer may be denominationally neutral nor the fact that its observance on the part of the students is voluntary can serve to free it from the limitations of the Establishment Clause, as it might from the Free Exercise Clause.” The reach and point of the Establishment Clause are broader than coercion and sect-preferences. Having teachers lead the voluntary recitation of the truths “Almighty God, we acknowledge our dependence upon Thee, and we beg Thy blessings upon us, our parents, our teachers and our country,” was, for Black and the Engel majority, a “union of government and religion.” 

McConnell and Chapman cite Engel as “the Court’s finest hour with the Establishment Clause.” The result, in their view, was correct, important, and courageous. They do not think well, however, of Black’s opinion; in fact, they label the Court’s reasoning dictum. They maintain instead that the constitutional vice in Engel was its kinship with compulsory worship. They claim that schools are an “inherently coercive environment.” Maybe so. But it is surely a matter of varying degree, lest we be obliged to conclude that all education amounts to indoctrination that bypasses the student’s freedom and critical reasoning.

The authors say that, apart from coercion, the school prayer was an unconstitutional “governmentally sponsored religious activity.” But “government sponsored religious activity” has a very long historical pedigree. In any event, a complete critical analysis of Engel would have to consider that the prayer comprised propositions about divine realities affirmable on the basis of reason alone. It was more philosophy than religion. Leading willing kids to recite it is much more like leading them in saying the Pledge of Allegiance (“under God”) than it is like compelling every child to attend, say, Catholic Mass. The Court in 1943 held that no student could be compelled to recite the Pledge. So too it was in Engel.

The distinction between what reason can affirm as truths about the farther reaches of reality—visible and invisible—and what can be known about such matters only through revelation is the distinction between the natural religion of common sense and the positive religious practices and beliefs of the sects, churches, and denominations. Grasping this distinction is essential to understanding the founding and the entire constitutional tradition ever since. Unfortunately, it is absent from Agreeing to Disagree, as it has been from the church–state corpus since World War II.

In their chapter treating religious monuments, such as the Bladensburg, Maryland, cross commemorating those killed in action during the Great War, McConnell and Chapman write that the “key question” is whether these symbols “go beyond acknowledging the role that religion plays in the lives of the people to effectively declare one religion to be superior or officially privileged, or to declare a minority religious view to be false or unworthy of respect.” Note the normative ground here in popular belief: is the monument reflective of facts about the “lives of the people?” This consideration is pertinent. The mentioned norm against preferring one faith over others is constitutionally sound. But a full constitutional reckoning with monuments must also be informed strategically by natural religion. The existence of a transcendent Creator God who providentially guides human affairs and to whom we consign our dead is not only the ringing affirmation of the American people until the day before yesterday. It is true.

In this chapter, too, McConnell and Chapman avoid speaking of natural religion and the truth about divine realities. One reason may be their convictions about the political significance of religion at the founding. They write, for example, that by then “the arguments in favor of establishment were not based on theocratic purposes but on the secular concern to instill the civic virtue necessary for the success of the new republic.” The founders most certainly believed that the virtue that only religion could inculcate in people was essential to free government. But contrasting “theocratic” with “secular” is scarcely illuminating of the historical materials. In fact, you will find neither of those words in the leading relevant primary sources from the founding. 

The authors say that “by the 1780s,” the “official justification for government support of religion” “had ceased to have any real theological component. It was entirely political and republican in nature.” The meaning of “theological” here is unclear to this reader. But if the sentence is meant to suggest that the founders maintained what we would call a “civil religion” with accompanying “ceremonial deism,” it is surely mistaken. Both of those concepts would have been incomprehensible to the founders, for the simple reason that they firmly believed that the religion and virtue required for republican government were not just, or even primarily, instrumentally valuable. The founders firmly believed that they were true. It is rather more likely, moreover, that Protestantism in religion produced republican government than that the latter called into being, as a kind of supporting buttress, the former.

Take just one tenet of the natural religion affirmed by the founders. Anyone who believes in Providence in its standard Christian form would be confounded by modern usages of “political,” “theological,” “theocratic,” and “secular.” A lively sense that oneself and one’s community stand under the judgment of a righteous God cuts across and blunts all such meanings. As Lincoln warned in his Second Inaugural: “The Almighty has His own purposes. . . . Woe unto the world because of offenses! for it must needs be that offenses come; but woe to that man by whom the offense cometh!” Countless Americans since have affirmed similar convictions. Almost every state constitution ever ratified in this country, for instance, includes expression of the people’s gratitude to “Almighty God” or the “Sovereign Ruler of the Universe” for their political blessings.

Dwight Eisenhower once said that “our form of government has no sense unless it is founded in a deeply felt religious faith, and I don’t care what it is.” Ike was a great man and a good president. The founders would nonetheless have been appalled at his latitudinarianism.

Many passages in Agreeing indicate that the authors hold that religion just in itself—in contrast to the liberty of choosing to be religious or not—is a private, not a public, good. The authors’ central normative claim is that “the Religion Clauses work together to prohibit the government from using sticks or carrots to induce uniformity of religious belief and practice.” Just so. But “uniformity” here elides a crucial distinction. It is the difference between, on the one hand, the highly desirable affirmation by public authority that religion is good, that natural religion is true, that each of us would be better off fulfilling our moral duty to seek the truth about religious matters conscientiously and by living dutifully in accord with it and, on the other hand, a servile conformity to what Caesar directs, or demands.  We should thank God if the entire population, “uniformly,” if you wish, were to live in accord with the first. And we should be repulsed if they lived according to the second. 

The authors assert that “the fundamental error [of postwar jurisprudence] was to view the Establishment Clause as about benefiting religion, rather than enlarging the scope of individual religious choice.” They judge that the Court’s stringent no-aid policy “reduced religious choice and educational diversity.” The book’s subtitle is: How the Establishment Clause Protects Religious Diversity and Freedom of Conscience. They seem to adopt as their own one of three possible readings of the Establishment Clause: “[I]s the constitution agnostic—neutral—on the question [whether religion or secularism has a privileged place in our republic], leaving the degree of religious influence on the public order to be decided by the people themselves through decentralized and deregulated institutions”?  There is no place here, it seems, for the truth about religion—that it is a fundamental good for everyone—or for the truths of religion, such as those ruled out of constitutional bounds in Engel.

In another chapter McConnell and Chapman say that they prefer neutral principles over “deference” in property disputes between factions of a church congregation, partly because it puts a “thumb on the scale” of hierarchical churches, which “violates the ‘sect neutrality’ fundamentals of disestablishment.” Here and elsewhere in the book they describe equality among religions as part of a sound understanding of what the Establishment Clause forbids. In the first chapter, they list it as one among six “elements of historical establishment.” They repeat the point in Chapter 8. Here they join every Supreme Court justice of the last half-century: all of them have held that such preferences violate the clear, core command of the Establishment Clause. Every one of them, save possibly Justice Scalia and the late Chief Justice Rehnquist, also maintained that the Clause does more than that. The differences among the justices have been about what, and how much, more. 

The historical evidence overwhelmingly supports the conclusion that no-sect preference is the meaning of non-establishment. The Kennedy Court’s rejection of secularist suppression was the perfect moment for the justices to substitute this norm for its ahistorical secularist mandates. Sometimes, the antidote to bad doctrine is better doctrine, not no doctrine at all. But the Court has unfortunately chosen to proceed by dead historical reckoning. There is no better compass for that journey than Agreeing to Disagree.

Image by kavunchik and licensed via Adobe Stock. Image resized.