In Part III, I will focus on the far more modest concurring opinions of Judges Keenan and Thacker, which were aimed right at the Supreme Court and the far less modest concurring opinion of Judge Wynn. Part IV will analyze the three dissents from Judges Niemeyer, Shedd, and Agee, which aimed at highlighting for the Supreme Court how their colleagues selectively cited the record and deviated from precedent.

Standing

In my writings on the travel ban cases, I have spent little time addressing justiciability. While the revised executive order drastically shrank the universe of people that were affected by the travel ban, that number is still greater than zero, so ultimately some party could articulate a cognizable injury. I do, however, disagree with the Fourth Circuit’s determination that the travel ban inflicts an Article III injury because it “sends a state-sanctioned message condemning [the plaintiff’s] religion and causing him to feel excluded and marginalized in his community” (p. 34). Specifically, citing McCreary County v. ALCU, Chief Judge Gregory explains that “flowing from [the executive order] is the alleged state-sanctioned message that foreign-born Muslims, a group to which Doe #1 belongs, are ‘outsiders, not full members of the political community.’” Further, the court notes “this harm is consistent with the ‘[f]eelings of marginalization and exclusion.’” But the injury requirement of Article III is not so inclusive. [Update: My analysis of what I originally termed “snowflake standing” focused on the court’s approach to finding Article III injury, and in no way reflected on the plaintiffs, who have suffered cognizable harms. I apologize and meant no disrespect to the plaintiffs and have reworded to better reflect that intention.]

As Judge Agee notes in dissent, there is often a “difficulty of determining injury in Establishment Clause cases.” (p. 191). The Fourth Circuit observed in Moss v. Spartanburg (2012) that “[m]any of the harms that Establishment Clause plaintiffs suffer are spiritual and value-laden, rather than tangible and economic.” Yet, Moss stressed that plaintiffs “possess standing when they are ‘spiritual[ly] affront[ed]’ as a result of ‘direct’ and ‘unwelcome’ contact with an alleged religious establishment within their community” (emphasis added). The majority opinion repeatedly cites Moss, but excluded the latter discussion requiring the affront to be “direct.”

In Moss, the Fourth Circuit found no injury for a student who had no “personal exposure” to a school’s policy promoting off-campus religious instruction, even though the “mere awareness of its existence” made him feel “‘like an outsider.’” In contrast, other students who actually received solicitations about the religious program suffered an injury because they “changed their conduct in adverse ways as a result of their perceived outsider status.” From this, Judge Agee summarizes that the injury cannot be purely sentimental, but must be “immediate and concrete,” and come “into direct contact with the alleged Establishment Clause violations.” Article III demands no less.

This fracture lies at the core of the travel ban case, affecting both standing and the merits: whose rights does the executive order violate? Judge Agee explains:

[T]he Executive Order here applies only to prospective immigrants. The order’s focus faces outward towards the alien residents of the subject countries, not inward towards persons in the United States like the plaintiffs . . . . What matters is whether the plaintiff came into direct contact with the religious establishment. And that is not the case here simply because the President is the party signing an order . . . . Plaintiffs do not have standing to allege violations of the Establishment Clause on behalf of their immigrant relatives. (pp. 199-204) (emphasis added).

The majority, which does not cite the entirety of Moss, stops short of asserting that the plaintiffs “have standing to allege violations of the Establishment Clause on behalf of their immigrant relatives.” Rather, Chief Judge Gregory relies on the threatened “prolonged family separation” as the basis for Article III standing. This proposition rests on some sort of constitutional right to familial relations, but does not say so explicitly, and makes no effort to comport with Moss’s “direct” requirement. Judge Wynn, however, is far more forward in his concurring opinion:

Here, aliens who are denied entry by virtue of the President’s exercise of his authority under Section 1182(f) can claim few, if any, rights under the Constitution. But when the President exercises that authority based solely on animus against a particular race, nationality, or religion, there is a grave risk—indeed, likelihood—that the constitutional harm will redound to citizens. For example, we hold today that the denial of entry to a class of aliens solely based on their adherence to a particular religion likely violates the Establishment Clause by sending “a state-sanctioned message that foreign-born Muslims . . . are ‘outsiders, not full members of the political community.’” Ante at 38 (quoting Moss v. Spartanburg Cty. Sch. Dist. Seven, 683 F.3d 599, 607 (4th Cir. 2012)) . . . . (pp. 109-10) (emphasis added).

Though aspirational, this reasoning has no grounding in Supreme Court precedent. Nor does this form of stigma-standing have a limiting principle, as Judge Agee notes:

Despite the majority’s holding, the stigma that plaintiffs claim to have suffered is not a cognizable injury because it is simply a subjective disagreement with a government action. To allow these plaintiffs to pursue their claims based on an idiosyncratic projection of stigmatization is to grant every would-be Establishment Clause plaintiff who develops negative feelings in response to some action by the Government a court proceeding in which to vent his subjective reactions as a legal claim . . . . The majority does not provide any principled instruction on how its sweeping standing ruling is cabined to this particular case, and thus its holding far oversteps the bounds of traditional judicial authority. (pp. 196-97).

On appeal, if the Supreme Court finds an Article III injury, it could do so without reaching the question of stigmatization. It should take the more limited approach, as the Fourth Circuit’s holding yields a truly limitless approach to standing.

Mandel and Din

The majority’s opinion hinges on the interplay between the Supreme Court’s decision in Kleindienst v. Mandel (1972) and Justice Kennedy’s concurring opinion in Kerry v. Din (2005). The former case concerned the reviewability of the Attorney General’s denial of a waiver to a Belgian Marxist who sought to lecture at American universities. The latter case involved the reviewability of a consular officer’s denial of a spousal visa for an Afghan national who had purportedly engaged in “terrorist activities.” (For more background on these cases, see this post). Most relevant for our purposes is how Justice Kennedy’s concurring opinion in Din interpreted Mandel.

The district court limited Mandel and Justice Kennedy’s opinion to reviewing decisions of a “consular officer” who acts in bad faith:

Finally, Defendants argue that because the Establishment Clause claim implicates Congress’s plenary power over immigration as delegated to the President, the Court need only consider whether the Government has offered a “facially legitimate and bona fide reason” for its action. See Mandel, 408 U.S. at 777. This standard is most typically applied when a court is asked to review an executive officer’s decision to deny a visa. See, e.g., Din, 135 S. Ct. at 2140 (Kennedy, J., concurring).

Following this reasoning, the opinion concluded that the Mandel test “does not apply to the ‘promulgation of sweeping immigration policy’ at the ‘highest levels of the political branches.’”

The Fourth Circuit, in contrast, rejects this narrow reading of Justice Kennedy’s opinion, instead determining that Justice Kennedy’s decision permits the court to look “behind” the President’s policy when there is evidence that the executive action is not “bona fide,” the second factor identified in Mandel. Where bad faith is alleged, the court need not afford deference to the executive action:

Where plaintiffs have seriously called into question whether the stated reason for the challenged action was provided in good faith, we understand Mandel, as construed by Justice Kennedy in his controlling concurrence in Din, to require that we step away from our deferential posture and look behind the stated reason for the challenged action. In other words, Mandel’s requirement that an immigration action be “bona fide” may in some instances compel more searching judicial review (p. 50).

Using this framework, Chief Judge Gregory dismisses the Attorney General and Homeland Security Secretary’s defense of the travel ban as “pretext” and instead concludes that the executive order is not “bona fide” (p. 52). As a result, the court opts to “look behind” the document.

This analysis misreads precedent. The operative phrase in Mandel is “facially legitimate and bona fide reason.” Both “legitimate” and “bona fide” are best read as being modified by “facially.” It is not “legitimate” on the face, but “bona fide” as a whole. The lack of good faith must be represented on the face of the action, not beyond its face. Indeed, this is how the Court applied the test in Mandel, Din, as well as in Fiallo v. Bell (1977). In each case the plaintiffs made specific allegations of bad faith that were extraneous to the policy at issue. But because the four corners of the policy did not suggest bad faith was in play, the Court refused to peek behind the curtain. Judge Niemeyer explains in dissent:

If the majority’s understanding had been shared by the Supreme Court, it would have compelled different results in each of Mandel, Fiallo, and Din, as in each of those cases the plaintiffs alleged bad faith with at least as much particularity as do the plaintiffs here. In Mandel, the allegations were such that Justice Marshall, writing in dissent, observed that “[e]ven the briefest peek behind the Attorney General’s reason for refusing a waiver in this case would reveal that it is a sham.” Id. at 778 (Marshall, J., dissenting). In Fiallo, Justice Marshall, again writing in dissent, pointed to the fact that the statute in question relied on “invidious classifications.” Fiallo, 430 U.S. at 810 (Marshall, J., dissenting). And in Din, the plaintiffs argued that the consular decision should be reviewed because it fell within the “limited circumstances where the government provides no reason, or where the reason on its face is illegitimate.” Brief for Respondent at 31, Din, 135 S. Ct. 2128 (No. 13-1402), 2015 WL 179409. But, as those cases hold, a lack of good faith must appear on the face of the government’s action, not from looking behind it.

Perhaps more troubling is that the en banc court treated Justice Kennedy’s opinion with the care of a Ginsu knife. As Judge Niemeyer notes in dissent, Chief Judge Gregory “carelesss[ly]” sliced and diced Din (p. 163):

As support for its dramatic departure from Supreme Court precedent the majority relies on a scattershot strain of quotations drawn out of context from one sentence in Din. The carelessness of the majority’s presentation is demonstrated simply by a comparison of its characterization of Din and the actual language of Din taken in context. Here is how the majority characterizes Din:

Justice Kennedy explained that where the plaintiff makes “an affirmative showing of bad faith” that is “plausibly alleged with sufficient particularity,” courts may “look behind” the challenged action to assess its “facially legitimate” justification. Ante at 50.

And here is what Justice Kennedy in Din actually said, with the language quoted by the majority in bold:

Absent an affirmative showing of bad faith on the part of the consular officer who denied Berashk a visa—which Din has not plausibly alleged with sufficient particularity—Mandel instructs us not to look behind the Government’s exclusion of Berashk for additional factual details beyond what its express reliance on § 1182(a)(3)(B) encompassed. Din, 135 S. Ct. at 2141 (emphasis added).

As Judge Niemeyer explains in dissent, Din allows the court to peek behind the curtain only when the policy, on its face, is not bona fide:

Nowhere did the Din Court authorize going behind the government’s notice for the purpose of showing bad faith. The plaintiff had to show facially that the notice was in bad faith, i.e., not bona fide. The majority’s selective quotations from Din, which conceal Din’s faithful application of Mandel, are simply misleading.

While Justice Kennedy is always free to change his mind, and reimagine his old opinions in one direction or the other, the lower courts may not do so for him.

Lemon and Town of Greece

The majority’s Establishment Clause analysis is very difficult to square with the Supreme Court’s precedents and Congress’s practices. Two decades ago, immigration scholar Enid F. Trucios-Haynes observed in the Georgetown Immigration Law Journal that applying the Supreme Court’s Establishment Clause jurisprudence to long-standing immigration laws “is particularly awkward.” Under either the Lemon test or the related “endorsement” test, a facially neutral law with a non-secular purpose is constitutionally suspect. A law that prefers religion over non-religion is very likely unconstitutional. A law that overtly prefers certain religious sects over others is almost certainly unconstitutional. Yet, immigration law routinely does all of the above, and neither Congress nor the courts have expressed the slightest concern for the Establishment Clause—that is, until President Trump’s executive orders. Further, as Judge Niemeyer noted in dissent, “the Supreme Court has never applied the Establishment Clause to matters of national security and foreign affairs.” There are many reasons to think the court’s “normal constitutional tools” are simply inapplicable in this context (pp. 52-53).

More pressingly, it is unclear precisely what test ought to govern the Establishment Clause, and whether the Lemon test survives. Once again, Judge Niemeyer splays our fragmented First Amendment jurisprudence: “While there is no question that it binds us, Lemon’s test, and particularly its inquiry into government purpose, has repeatedly been criticized as open-ended and manipulable.” He cautions the majority to be “wary of jumping when on thin ice.”

The majority, perhaps sensing Lemon is on a shaky footing, takes a cue from the constitutional law professors’ amicus brief. In a footnote, Chief Judge Gregory cites the Court’s recent decision in Town of Greece v. Galloway, which involved the constitutionality of prayers before legislative meetings. In that case, Justice Kennedy upheld the policy because “[i]n no instance did town leaders signal disfavor toward nonparticipants or suggest that their stature in the community was in any way diminished.” Relying on this observation, the Fourth Circuit determined that “EO-2 would likely fail any purpose test, for whether religious animus motivates a government action is a fundamental part of our Establishment Clause inquiry no matter the degree of scrutiny that applies . . . . There is simply too much evidence that EO-2 was motivated by religious animus for it to survive any measure of constitutional review.”

This analysis, once again, returns to the applicability of domestic Establishment Clause cases to foreign policy cases. Justice Kennedy majority opinion in Town of Greece, however, provides a critical distinction. He wrote:

In no instance did town leaders signal disfavor toward nonparticipants or suggest that their stature in the community was in any way diminished. A practice that classified citizens based on their religious views would violate the Constitution, but that is not the case before this Court (emphasis added).

The key word is citizens.

Congress routinely classifies aliens based on religious views. The non-discrimination statute (on which Judge Thacker’s concurring opinion relies) explicitly omits religion for the grounds on which immigrant visas can be withheld:

Except as specifically provided in paragraph (2) and in sections 1101(a)(27), 1151(b)(2)(A)(i), and 1153 of this title, no person shall receive any preference or priority or be discriminated against in the issuance of an immigrant visa because of the person’s race, sex, nationality, place of birth, or place of residence.

Another longstanding statute provides special visas for those “carrying on the vocation of a minister of that religious denomination.” The implementing regulations specify that “[e]xamples of vocations include nuns, monks, and religious brothers and sisters.” Through the Lautenberg Amendment, our immigration law has afforded special status to “[a]liens who are (or were) nationals and residents of the Soviet Union and who are Jews or Evangelical Christians shall be deemed” to be subject to persecution, unless there was sufficient evidence to the contrary. It is certainly true, as Justice Kennedy explained, “A practice that classified citizens based on their religious views would violate the Constitution.” But such is not true for non-citizens outside the United States.

In any event, the executive order does not classify aliens based on their religion. It is facially neutral and far less religious than actions that Court has upheld in the domestic context. Limited to the four-corners of the executive order, the majority musters only one item to suggest it is not facially bona fide. Section 11 of the March 6, 2017 executive order, which was not subject to challenge, is titled “Transparency and Data Collection.” Subsection (i) asks the Attorney General and Secretary of Homeland Security to collect “information regarding the number of foreign nationals in the United States who have been charged with terrorism-related offenses while in the United States.” Subsection (ii) asks for “information regarding the number of foreign nationals in the United States who have been radicalized after entry into the United States and who have engaged in terrorism-related acts.” And most relevant for our purposes, subsection (iii) seeks “information regarding the number and types of acts of gender-based violence against women, including so-called “honor killings,” in the United States by foreign nationals.” This is the best evidence the court can find of the fact that the executive order is not facially bona fide.

The majority explains in a footnote:

Plaintiffs suggest that EO-2 is not facially neutral, because by directing the Secretary of Homeland Security to collect data on “honor killings” committed in the United States by foreign nationals, EO-2 incorporates “a stereotype about Muslims that the President had invoked in the months preceding the Order.” . . . . The Amici Constitutional Law Scholars go so far as to call the reference to honor killings “anti-Islamic dog-whistling.” Brief for Constitutional Law Scholars 19 n.3. We find this text in EO-2 to be yet another marker that its national security purpose is secondary to its religious purpose. (p. 60).

I’ll concede that honor killings are a reference to Islam, and not other religions. But this reporting requirement comes nowhere near the standard needed to find a violation of the Establishment Clause, let alone bad faith. Imagine if the executive order included a copy of the Ten Commandments, adorned by “two Stars of David, and the superimposed Greek letters Chi and Rho as the familiar monogram of Christ,” as the monument outside the Texas capitol did in Van Orden v. Perry. Or imagine if the President had opened the executive order with the invocation, “We acknowledge the saving sacrifice of Jesus Christ on the cross,” as did the ministers who spoke in the Town of Greece. In both of these cases, the Court upheld as constitutional explicit invocations of religion. These precedents simply do not support the supposition that a requesting information about violence against women in the form of honor killings—even if it is a practice that is predominantly associated with Muslims—is not bona fide.

Finally, it is worth stressing that Justice Kennedy joined both the majorities in both Van Orden and Town of Greece. He also joined part II of Justice Scalia’s dissent in McCreary County, rejecting Justice Souter’s purposivist analysis. That said, in Church of the Lukumi Babalu v. City of Hilaeah, he wrote that “[i]n our Establishment Clause cases, we have often stated the principle that the First Amendment forbids an official purpose to disapprove of a particular religion or of religion in general.” Trying to read the tealeaves on Justice Kennedy’s jurisprudence is always a fraught endeavor.

The Fourth Circuit’s Establishment Clause analysis, alas, is a strange brew that will not stand on appeal.

Cross-Posted at Lawfare