My colleague Steve Goldberg helpfully identifies a couple of constitutional anomalies suggested by the Supreme Court’s 2005 decision in Cutter v. Wilkinson. In that case, the Court unanimously upheld the Religious Land Use and Institutionalized Persons Act of 2000 (RLUIPA), notwithstanding the fact that the statute requires state officials to accommodate religious exercise—to grant religious exemptions to otherwise generally applicable regulations—in prisons and other institutions. [Disclosure: I helped draft RLUIPA when I worked at the Department of Justice, and after I left the government, I filed an amicus brief in Cutter on behalf of Senators Hatch and Kennedy, defending the constitutionality of the statute.]
I think Steve is correct about one implication of Cutter; but I have some doubts about his broader reading of the decision.
First, Steve notes that although the Free Exercise Clause does not of its own accord compel such religious exemptions (not, anyway, under the Court’s modern FEC doctrine), the Free Exercise Clause must still be doing some work in “permissive accommodation” cases such as Cutter, because there is no other way to explain why such preferred treatment does not violate the Establishment Clause. As Steve writes in his sharp and provocative recent essay on Cutter (14 Wm. & Mary Bill of Rights J. 1403 (2006)), the Free Exercise Clause “is not passive. It pulls some legislation into a zone of safety that otherwise would be crushed by the Establishment Clause.”
I think this is absolutely right. Although the current Court is not inclined to read the Free Exercise Clause as mandating religious exemptions in many circumstances, it is willing to allow states to choose to grant such exemptions under limited circumstances—namely, where the exemption alleviates a substantial, government-imposed burden on religious exercise without imposing significant burdens on other private parties. In addition to Cutter, see Corporation of Presiding Bishop of Church of Jesus Christ of Latter-Day Saints v. Amos, 483 U.S. 327 (1987). This doctrine can only be explained as a sort of Free Exercise-generated “exception” to the “no religious preference” rule that the Establishment Clause otherwise prescribes.
Moreover, as Steve emphasizes in his essay, a state is permitted to thereby accommodate the exercise of one constitutional liberty—religious exercise—without simultaneously accommodating other constitutional liberties: Ohio can, for instance, alleviate prison-imposed burdens on inmates’ religious exercise by providing them with a special religiously appropriate diet, even if it does not also choose to alleviate the burdens that incarceration imposes on the exercise of other constitutional liberties, such as the rights of speech, privacy, marriage, petitioning, etc.
Steve’s second reading of Cutter, however, is much more far-reaching, and somewhat more contestable.
“The big news,” Steve writes in his essay, “is that Cutter empowers Congress to give religious expression a preferred position in relation to political expression.” (My emphasis.) In other words, Steve reads Cutter as holding that where a RLUIPA-generated accommodation of religious exercise takes the form of accommodating religious expression (or assembly), the state may grant such an accommodation even where it declines to similarly accommodate analogous nonreligious inmate expression or assembly.
Here’s Steve’s provocative example, drawn from the facts of Cutter itself: “Congress can require that a meeting time and place be made available for The Church of Jesus Christ Christian, a white supremacist religion, while not requiring that any provision be made for the Aryan Nation, a white supremacist political group.”
In short, on this reading, Cutter permits pro-religion viewpoint and content discrimination that the Free Speech Clause (or the Free Assembly Clause) would otherwise forbid.
I understand why Steve reaches this conclusion, but I don’t read Cutter to go nearly so far.
As Steve explains in his essay, in virtually any other context it would be very difficult to imagine the Court approving of a state preference for religious over nonreligious expression or assembly. Take, for instance, the long line of “equal access” cases beginning with Widmar v. Vincent. For example, in its 2001 decision in Good News Club, the Court held that if a school board made school facilities available for private groups discussing “character and morals,” the Free Speech Clause prohibits the school board from excluding a religious group that taught youngsters “moral lessons from a Christian perspective through live storytelling and prayer.” It is, I think, inconceivable that the Court would have allowed the school to give a preference in access to the prayer group. Similarly, in Rosenberger, the Court held that where the University of Virginia paid for the printing costs of a wide array of student journals of diverse viewpoints and perspectives, it could not deny such a subsidy to journals that “primarily” reflected views for or against religion: That would be impermissible viewpoint discrimination. Is it really possible, however, that the University of Virginia could choose to subsidize only the religious journals?
Of course not. And in fact, on the few occasions when the Court has spoken directly to the question, it has indicated that viewpoint-based preferences for religious expression (or publication, or speech-related assembly) are unconstitutional. See, e.g., Heffron v. International Soc’y for Krishna Consciousness, Inc., 452 U.S. 640, 652-653 (1981) (suggesting that if the state granted Krishnas an exemption from a state fair rule prohibiting distribution of leaflets, it would be required to extend such an exemption as well to other organizations seeking similar speech rights on the fairgrounds); Texas Monthly, Inc. v. Bullock, 489 U.S. 1, 25 (1989) (White, J., concurring in the judgment) (tax exemption specifically for religious publications discriminates on basis of content of publication and therefore violates Free Press Clause); id. at 28 (Blackmun, J., concurring in the judgment, joined by O’Connor, J.) (similar, although oddly resting on the Establishment Clause: “A statutory preference for the dissemination of religious ideas offends our most basic understanding of what the Establishment Clause is all about and hence is constitutionally intolerable.”).
Even the strongest and most eloquent of defenders of permissive religious accommodation appears to agree: “Favoring religious speakers over similarly situated nonreligious speakers would violate the viewpoint-neutrality requirement of the Free Speech Clause.” Michael McConnell, 50 DePaul L. Rev. at 40. (As far as I know, the only notable defense of viewpoint discrimination in favor of religion in the context of speech and assembly appears in Justice Scalia’s dissent in Texas Monthly, 489 U.S. at 44-45. But Justice Scalia’s analysis consists almost entirely of ipse dixit (e.g., “it is impossible to believe that the State is constitutionally prohibited from taxing Texas Monthly magazine more heavily than the Holy Bible”—Why is that so impossible to believe? It seems perfectly sensible to me.); and, in any event, the Court did not adopt it. (Justice Blackmun suggested that preferences for religious expression could survive Free Press Clause scrutiny where the Free Exercise Clause requires the exemption, id. at 26-27, but he did not indicate that a permissive accommodation could be content-based in a way that would otherwise violate the Press Clause.))
Indeed, precisely because such a preference for religious expression or assembly would raise serious constitutional concerns in its own right, I have argued (in my amicus brief in Cutter) that as a matter of statutory interpretation RLUIPA does not require states to grant such a preference: “[I]f in a particular case a prison official would violate someone else’s constitutional rights by granting a requested religious accommodation—such as, for example, by granting content-based preferences for prisoners’ religious expression in a manner that the Free Speech, Press or Assembly Clauses of the First Amendment would forbid [citing Heffron and Texas Monthly]—then the avoidance of that constitutional violation would itself be a 'compelling interest' that would, under RLUIPA’s own statutory standard, justify denial of the accommodation (at least in the form it was requested).”
(This is not to say that a state would never have to grant an expression- or assembly-based RLUIPA accommodation—only that if it does, it must do so in a manner that avoids impermissible viewpoint discrimination, such as by extending the exemption more broadly to nonreligious claimants, too. If there are compelling reasons for refusing to grant such a broadened extension, the state might satisfy its RLUIPA burden for denial of any exemption. But where such an argument is made, the choice should be between the broadened exemption (not limited to religion) or no exemption at all, because a viewpoint-based exemption for religious expression alone would appear to be constitutionally untenable.)
I’m not sure Steve himself disagrees with all of this. But he reads Cutter as calling it into serious doubt. In particular, Steve focuses on the fact, noted in Justice Ginsburg’s opinion for the Court, that Ohio itself had historically allowed prisoners “to assemble for worship, but not for political rallies.” 544 U.S. at 724-725. Steve views the Court as having noted this practice “with favor,” and as therefore indicating that “preferring religious assembly to political assembly is an acceptable accommodation.”
In my view, this singular reference in Cutter to Ohio’s practice of preferring religious assemblies in prison doesn’t quite bear the weight that Steve would give it, for three reasons.
Most importantly, I don't think it's right to read the passage in question as suggesting that the identified pre-existing Ohio religious exemptions were necessarily constitutional. Justice Ginsburg's point, as I read it, was merely that such practices undermined Ohio’s own constitutional argument to the Court (i.e, if, as Ohio contended, all religious accommodations must be matched with accommodations of other constitutional rights, then Ohio itself had been violating that principle all along). For example, Ginsburg also noted that Ohio had chosen to provide special treatment for “traditionally recognized” religions. But such a sect-based preference would almost certainly be unconstitutional (see Larson v. Valente), and there’s no reason to suppose that Ginsburg, or the Court, intended to suggest that a discriminatory accommodation in favor of “established” religions would be permissible. Likewise with her example of a preference for religious over political assemblies.
Second, even if Justice Ginsburg had intended to suggest that the state could draw the lines that it traditionally had enforced, the example she gives does not establish the broad principle that Steve suggests. There are, of course, quite dramatic practical differences between assembly for worship and a political rally—differences that might warrant disparate prison treatment, wholly apart from the content of the speech at the assembly. Thus, for example, a prison might be able to forbid prisoner “rallies” in general—of a political or of a religious (or other) nature—but still allow quiet and peaceful assemblies for worship or discussion. The passage in Cutter does not say anything about a state’s power to distinguish between religious and political rallies, or between assembly for worship and assembly for a political discussion. Therefore even on the strongest reading of this passage, it does not as a general matter “prefer religious assembly to political assembly is an acceptable accommodation,” where the two assemblies in question are indistinguishable in every respect save the content of what is spoken there.
Finally, it may simply be the case that the usual constitutional requirements of content- and viewpoint-neutrality do not fully apply in prisons. Somewhat surprisingly, the Court has never definitively resolved this question. In Turner v. Saffley, the Court noted that “[w]e have found it important to inquire whether prison regulations restricting inmates’ First Amendment rights operated in a neutral fashion, without regard to the content of the expression.” 482 U.S. at 90; see also Procunier v. Martinez, 416 U.S. at 415. In Thornbugh v. Abbott, the Court qualified this somewhat, by holding that where prisons draw distinctions between publications solely on the basis of their potential implications for prison security, “the regulations are ‘neutral’ in the technical sense in which we meant and used that term in Turner.” 490 U.S. at 415-416.
The Court could have further clarified the doctrine—and applied it to the context of a preference for religious content, in particular—in last Term’s decision in Beard v. Banks, where the Court upheld a policy whereby some prisoners were denied all access to newspapers, magazines, or personal photographs except that they were permitted to receive legal and personal correspondence and religious and legal materials. Pursuant to this policy, it appears that the prison denied prisoners the right to receive the Christian Science Monitor but allowed them to obtain the Jewish Daily Forward, based on the determination of a prison official that the latter qualifies as a religious publication and the former does not. 126 S. Ct. at 2593.
For for some reason, however, the prisoner respondents in Beard did not raise the content-discrimination issue, and therefore the Court did not address it.
If, as the result (if not the rationale) in Beard suggests, the Free Speech Clause does not in the first instance forbid prison officials from preferring religious expression in a content-based manner, then of course the application of RLUIPA in that setting simply does not raise any First Amendment questions once the Establishment Clause argument is overcome (as it was in Cutter). And, if that is the case, then Cutter should not be read to suggest that outside the prison walls—where viewpoint neutrality is the usual rule—religious accommodation may take the form of a preference for religious expression or assembly over analogous expression or assembly of a nonreligious nature.