It is a pleasure to have the opportunity to engage with the work of Francis Beckwith, whose work I’ve admired for a long time. We often disagree, and I disagree with much of what he has written here, but I’m glad we are having this conversation.
He writes that “if applied rigorously, the Lemon Test institutionalizes both a governmental preference for secularism as well as a hostility toward religion.” He also writes, however, that the terms of the test “seem endlessly pliable.”
These claims are inconsistent. The second is the more persuasive. Lemon v. Kurtzman’s test can’t be applied rigorously, because its terms are too vague. During the period of its dominance, the Court got some things right and others wrong. The test had nothing to do with it. It is so mushy that it could have no effect unless supplemented by unstated premises that give concrete meaning to the test’s generalities. It was those (since abandoned) premises, and not the test that the Court articulated, that produced the specific result in Lemon.
What Made Lemon Sour?
Lemon decrees that, in order to withstand an Establishment Clause challenge, “First, the statute must have a secular legislative purpose; second, its principal or primary effect must be one that neither advances nor inhibits religion, … finally, the statute must not foster ‘an excessive government entanglement with religion.’”
The first prong is hard to violate, but violations have happened. If, as the framers thought, state manipulation tends to corrupt religion, then it obviously follows that the state cannot dictate religious truth, which is what laws without a secular purpose do. The third prong reflects the sensible demand that religious institutions operate free of state control—an idea reflected in the recent ministerial exception cases.
But the second prong is the one that does most of the actual work, and it is incoherent without some specification of the baseline against which one judges the advancement or inhibition of religion. Some aspects of the baseline are uncontroversial. The classic example is police and fire protection for churches. Such protection certainly benefits the churches, and may even benefit core religious activities, as when it is given to religious articles that have only religious uses. Other benefits, such as funding expressly directed at specifically religious activities, would violate the clause. The Court has never explained which benefits are innocuous and which impermissibly advance religion, or how courts should determine which is which. (Recently the Court created a different baseline problem when it declared that laws can’t be applied to religious objectors so long as the state “treats some comparable secular activities more favorably,” thus inviting extravagant exemption claims.)
Lemon itself provided a very definite result: a certain kind of school aid was forbidden. It did so by augmenting its test with the undefended presumption that, even if teachers’ salaries were being supplemented for the purpose of enabling them to teach secular subjects, the state could not avoid an Establishment Clause violation unless it intensely monitored the schools to make sure that no taint of religion entered their classrooms. That would create excessive entanglement, so the aid was barred.
The crucial premise here is that, when the state provides funding for secular purposes, the money must be traced to avoid religious uses. If that is right, then the state would also need to intensely micromanage any recipient of aid. That totalitarian implication was only drawn with respect to schools. Ira Lupu and Robert Tuttle are probably right that, in this context, strict separationism was a prophylactic rule, aiming to prevent surreptitious, targeted government support for religion. In retrospect, that kind of separationism went way too far, and has few defenders today. But the problem was not the three-prong test, which was too vague to dictate that result. The reformulation by Rehnquist in Zelman v. Simmons-Harris, that school aid is appropriately neutral if it is directed by private choice, is not inconsistent with the terms of the Lemon test, though it abandons Lemon’s result. Beckwith writes that, in Zelman, Lemon “plays no substantive role in the Court’s judgment.” He is right. But then, you can’t blame Lemon for what came before.
The Value of America’s Religious Neutrality
The fundamental question is whether the American law of religious liberty is worth preserving. I have argued that American religious neutrality is coherent and attractive. During the period when Lemon has nominally been the prevailing law, America has been unusually successful in dealing with religious diversity—more so than many other generally well-functioning democracies, such as France, Germany, and Italy. Even if the American law of religious liberty were entirely incoherent, it might still offer a sensible approach to this perennial human problem.
There is, however, a deep logic to the law that its critics have not understood. First Amendment doctrine treats religion as a good thing. It insists, however—and here it parts company with the religious traditionalists—that religion’s goodness be understood at a high enough level of abstraction that the state takes no position on any live religious dispute. It holds that religion’s value is best honored by prohibiting the state from trying to answer religious questions.
American religious neutrality has over time become vaguer as America has become more religiously diverse, so that today (with the exception of a few grandfathered practices), the state may not even affirm the existence of God. This kind of neutrality is not the kind of neutrality toward all conceptions of the good that many liberal political theorists have advocated, but it is the best response to the enormous variety of religious views in modern America. It is faithful to the belief, held by the leading framers of the First Amendment, that religion can be corrupted by state support.
Beckwith worries that Lemon demands that the state “be required to completely eliminate from the public square any symbols, words, or monuments that acknowledge the religious heritage and beliefs of the nation’s citizens.” Lemon never had that effect, not even at the height of its influence. I mentioned grandfathered practices. It has been clear for some time that the Court treats long-established manifestations of ceremonial deism with more deference than new ones, which are more obviously attempts by the state to intervene in live religious controversies.
Beckwith is clearly unhappy with the secularizing effect of this aspect of the Lemon test. I don’t see why. I’m not religious, but I regard the established religions with great respect. This “acknowledgement” is a poisoned gift. Why are religious Christians not sickened, as I am, by the modern Christmas display, paid for by tax dollars secured through the influence of the local merchants association, reminding us that Jesus suffered and died on the cross so that we could enjoy great holiday shopping?