Why the Establishment Clause Has Emerged as the Chief Stumbling Block for Trump's Muslim Ban

by Michael Dorf

Although the lawsuits challenging President Trump's first and second Muslim bans have offered a variety of legal claims, the one that has gotten the most leverage thus far appears to be the Establishment Clause challenge. Federal district courts in Hawai'i and Maryland both relied on the Establishment Clause as the basis for their invalidation of Ban 2.0 in the last few days. So did the federal district court in Virginia in enjoining Ban 1.0 in February. Why the Establishment Clause rather than the other claims?


From its inception, the most offensive aspect of the Trump travel ban was the obvious fact that it was an effort to dress up in legal garb the anti-Muslim bigotry that was plainly visible in Trump's 2015 Pearl Harbor Day call "for a total and complete shutdown of Muslims entering the United States until our country's representatives can figure out what is going on." Trump's initial Muslim ban lives in infamy, on his campaign website (for now at least), and most importantly, in both versions 1.0 and 2.0 of his Executive Order limiting travel into the United States.

That demonstrating what a badass Trump is by sticking it to Muslims was and remains the purpose of both versions of the travel ban is obvious to anyone but an abject apologist for the Trump administration or someone in rough agreement with the five Ninth Circuit judges (Alex Kozinski and four of his colleagues) who think they can overrule the Supreme Court case law that makes subjective motive relevant to a variety of constitutional doctrines, including doctrines construing the Establishment Clause. Judge Kozinski apparently thinks he scores a knockout blow by asking rhetorically how a president can ever cure the taint of his previously expressed bias, apparently unaware that this is a criticism of all subjective motive tests and equally unaware of the fact that various approaches, such as the one I described here, have been offered to test whether and when such a taint has been purged.

It is possible that other judges and maybe even Supreme Court justices will pretend along with Judge Kozinski that the motive for the travel ban is unknowable, but assuming that most do not, they will need to consider how to take account of the illicit motivation for the ban. Why do so through the Establishment Clause in particular? The Establishment Clause forbids government favoritism of one religion over another, and so it provides a natural source for a constitutional claim. But the intentional anti-Muslim discrimination also contravenes other provisions invoked by the plaintiffs challenging the Trump travel ban--including the First Amendment's Free Exercise Clause, the equal protection component of the Fifth Amendment, and the Religious Freedom Restoration Act. Why has the Establishment Clause challenge thus far gotten greater traction?

The answer--as I suggested in a March 7 blog post--is that the Establishment Clause is "structural" in a way that various rights provisions might not be. True, as I noted in that post, there is a good argument to be made that other rights are likewise structural, but the proposition is clearer with respect to the Establishment Clause. The value of a structural constitutional limit in the current context is that it permits a person currently lawfully in the United States--such as a Muslim US citizen or Muslim lawful permanent resident--to object to what the government is doing even though what it is doing occurs overseas (at US embassies and consulates where applicants are denied visas) or at ports of entry. A claim that the US violates the Free Exercise Clause when it applies the travel ban to deny a visa to a Syrian national seems to require that Syrian nationals outside the US have rights under the First Amendment--a point that the federal government contests. By contrast, a claim that the US government violates the Establishment Clause does not require that a plaintiff in the US hook his, her, or its claim to the rights of anybody outside the country. It simply requires a showing that the plaintiff is injured by the government's violation of the Establishment Clause.

Now, as I noted in that March 7 post, even on its own terms, the government's argument can be challenged. As Justice Kennedy wrote in his concurrence in Kerry v. Din, a showing of bad faith in the denial of a visa should be enough to vitiate a facially valid government reason for that denial. That should be true whether the bad faith is that of a consular officer making a determination for a single case or the government acting en masse via executive order, as Daniel Hemel explains on the Take Care Blog. Thus, even if they were not relying on a structural provision, the plaintiffs would have the better of the argument over the applicability of the Constitution.

Suppose a court were to conclude otherwise, as five judges of the Ninth Circuit have concluded with respect to Muslim Ban 1.0 (despite lacking any case before them in which to make the point properly). Then the fact that the plaintiffs are relying on a structural provision ought to be enough for them to establish the Constitution's applicability. To be clear, I'm not saying the dissenting judges in the Ninth Circuit think there's a violation of the Establishment Clause. As noted above, they think that the evidence of President Trump's subjective bias is not sufficient to show bad faith. But again, let's assume they're wrong about that--because they are in fact wrong about that.

So, to sum up where we are so far, imagine a judge or justice who thinks: 1) rights provisions don't apply to aliens outside the US; but 2) the Establishment Clause does apply to actions taken outside the United States because it's structural; and 3) subjective evidence of bad faith in adopting Muslim Ban 2.0 shows an Establishment Clause violation. That's game over. Plaintiffs win.

And yet, might someone argue that the Establishment Clause isn't structural for all the justices? After all, in a series of cases beginning with County of Allegheny v. ACLU in 1989, Justice Kennedy took issue with the view that the Establishment Clause forbids government from endorsing one religion over another, arguing for the narrower view that the Clause forbids coercion. True, in later cases, Justice Kennedy applied a broad view of what counts as coercion, but he has never fully signed onto the so-called endorsement test. And it appears that the courts that have struck down Muslim Ban 1.0 and 2.0 on the basis of the Establishment Clause have relied on the endorsement test. So maybe Justice Kennedy is not on board with the test needed to invalidate the Muslim Ban under the Establishment Clause?

Not a chance. In the relevant cases, Justice Kennedy was objecting to the proposition that by recognizing various religious holidays and practices (with, e.g., a creche display at Christmas time), government was endorsing religion and thus sending a message of exclusion to nonbelievers in the religions at issue. Maybe he was right in that objection; maybe he was wrong. But whatever one thinks about the limits on government affirmative accommodation and recognition, nothing in anything Justice Kennedy wrote in any of those cases remotely suggests that he would think it permissible for the government specifically to denigrate a particular religion.

And of course that is precisely what Muslim Ban 2.0 does. Unlike a display with Christian themes, the Muslim ban does not merely tell Muslim US citizens and other lawful Muslim residents that their religion is not practiced by a majority of Americans. They know that already. Muslim Ban 2.0 tells Muslims in the US (and elsewhere but the ones in the US have undoubted standing) that they are not welcome here because of their religion. One need not accept the principle that the Establishment Clause bars endorsement of religion or a particular religion to see that a policy of disfavoring a religious minority violates the Establishment Clause.

Even if one thought that a government policy had to rise to the level of coercion to count as a violation of the Establishment Clause in a case like this, the Muslim Ban clearly does that. In County of Allegheny, Justice Kennedy gave an example of the sort of message of affirmative endorsement of one religion that would, in his view, count as coercive. He wrote that the Establishment
Clause forbids a city to permit the permanent erection of a large Latin cross on the roof of city hall. This is not because government speech about religion is per se suspect, as the majority would have it, but because such an obtrusive year-round religious display would place the government's weight behind an obvious effort to proselytize on behalf of a particular religion.  . . . Speech may coerce in some circumstances, but this does not justify a ban on all government recognition of religion.
Three points bear emphasizing here. First, Trump's Muslim ban does not merely "permit" a private party to take actions that bear on the Establishment Clause; the ban is the policy of the government itself. Second, the Muslim ban sends a much more powerful message that Muslims are affirmatively disfavored than would a Latin cross on city hall, and therefore would be much more effective in telling Muslims that if they want to fit in in America, they should not openly practice their faith--a message that many Muslims have already acted on. Third and finally, holding the Muslim ban inconsistent with the Establishment Clause would not create any tension whatsoever with the proposition that there are circumstances in which government may recognize the importance of religion--including majority religions--in people's lives.

The Establishment Clause argument is winning because it is a winning argument.