Analysis of IRAP v. Trump Part III: The Concurring Opinions of Judges Thacker, Keenan, and Wynn

May 31st, 2017

This post is the third part of a four-part series on the Fourth Circuit’s recent en banc decision in IRAP v. Trump. Part I analyzed how the court considered pre- and post-inauguration statements. Part II analyzed how the court marshals the Supreme Court’s precedents concerning standing, reviewability of immigration decisions, and the Establishment Clause.

In this post, 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 less modest concurring opinion of Judge Wynn. Part IV will analyze the three dissents from Judges Niemeyer, Shedd, and Agee, which were aimed at highlighting for the Supreme Court how their colleagues selectively cited the record and deviated from precedent.

Judge Thacker’s Statutory Argument

Judge Thacker’s concurring opinion (pp. 129-47) parts with the majority opinion on both the constitutional and statutory questions. On the Establishment Clause question, she “would not consider remarks made by candidate Trump before he took his presidential oath of office.” However, limiting her review to “remarks made or sentiments expressed after January 20, 2017,” she would still find a violation of the Establishment Clause. As I indicated in Part I of my series on this case, my view is that Judge Thacker reads certain statements from Trump and his associates out of context and reaches unsupported conclusions about the president’s post-inauguration purpose. I won’t rehash those arguments here. Instead, I’ll focus on the second place where Judge Thacker differs from her colleagues.

The majority opinion declines to reach the statutory question about whether the travel ban violates the Immigration and Nationality Act (INA). In contrast, Judge Thacker “conclude[s] Appellees have demonstrated a likelihood of success on the merits of their argument that Section 2(c), as it applies to immigrant visas, violates 8 U.S.C. § 1152(a)(1)(A)” of the INA (p. 129). This conclusion does not appreciate the important distinction between entry and visa issuance under the statute.

§§ 1152 and 1182

Recall that there are two relevant statutory provisions. The first provision, 8 U.S.C. § 1182(f), gives the president the power to deny or restrict entry to certain classes of aliens “[w]henever the President finds that the entry of any aliens or of any class of aliens into the United States would be detrimental to the interests of the United States.”

The second provision, 8 U.S.C. § 1152(a)(1)(A), prohibits the government from discriminating based on nationality (but not religion) with respect to the issuance of immigrant visas. It states that, with exceptions, “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.” 8 U.S.C. § 1152(a)(1)(B) provides that subparagraph (A) does not “limit the authority of the Secretary of State to determine the procedures for the processing of immigrant visa applications or the locations where such applications will be processed.”

As a threshold matter, Judge Thacker limits her analysis to immigrant visas—properly, in my view, but thereby rejecting the line of argument put forth in parallel litigation by Neal Katyal, who urged the Ninth Circuit to extend § 1152(a)’s prohibition to immigrant visas and non-immigrant visa alike. Under Judge Thacker’s reasoning, § 1152(a) could not provide relief to those seeking non-immigrant visas. For that reason, the statutory argument does not resolve the entire case and the plaintiffs’ constitutional arguments must be considered at least as they apply to non-immigrant visas. In that sense, Judge Thacker’s position, which cannot address the legality of the travel ban as a whole, could still impact other areas of immigration law (pp. 140-49).

The core question is how § 1152 and § 1182 interact. Since the outset of this litigation, I have written that the statutory provisions are not in conflict because they operate based on different areas of immigration law. (See Parts I, II, III, and IV). Section 1182 accords the president broad discretion to bar entry into the United States, and Section 1152, which prohibits nationality-based discrimination, concerns the issuance of immigrant visas. The government reads the statutes in the same way. In fact, Solicitor General Wall explained that this is how the State Department has interpreted these provisions for decades, a position that warrants some degree of deference.

Judge Thacker rejects the government’s construction of these statute. She writes that it is “circular logic” for the government to argue that “when an alien subject to [EO-2] is denied an immigrant visa, he is not suffering discrimination on the basis of nationality of the sort prohibited by Section 1152(a)(1)(A); instead, he is being denied a visa because he has been validly barred from entering the country.” Under this logic, as she sees it, “an alien is barred from entry because he does not have and cannot attain a visa, but he is denied a visa because he is barred from entry.” Thus, she concludes, “the visa issuance and entry concepts are intertwined to the point of indistinguishability.”

Contrary to Judge Thacker’s assertion, the government’s position is not circular. In immigration law, a visa is a necessary but not sufficient condition for entry. This point, that the two statutory provisions deal with “two distinct actions in the context of immigration” and that aliens with a valid visa may still not be allowed to enter, is one that Judge Keenan highlights in her concurring opinion (in the only part not joined by Judge Thacker). As Judge Keenan points out, an alien with a visa could be denied admission “for a variety of reasons set forth elsewhere in the INA.” For example, a person with a valid visa could be denied entry at the airport due to the fact that he has a communicable disease or by providing incorrect information to a federal. Entry and visas are certainly related, but are not “intertwined to the point of indistinguishability.” More importantly, as Judge Keenan notes, Judge Thacker’s analysis has the effect of disregarding the usage of different terms in the “plain language” of the statute itself: “entry” and “visas.”

Assuming that § 1152 controls the travel ban, Judge Thacker adduces two primary reasons why the non-discrimination provision prevails.

First, Judge Thacker asserts that “[r]eading § 1182(f) as bestowing upon the President blanket authority to carry out a suspension of entry, which involves rejecting a particular country’s immigrant visa applications as a matter of course, would effectively nullify the protections in § 1152(a)(1)(A) and create an end-run around its prohibitions against discrimination” (p. 145). But § 1152(a)(1), as a whole, need not be so read. Subparagraph (B) of the provision allows the Secretary to adopt “procedures for the processing of immigrant visa applications” that could explicitly be based on national origin. That is, subparagraph (B) states that subparagraph (A) “shall [not] be construed to limit” the Secretary’s authority. Judge Thacker argued that reading subparagraph (B) in this fashion should render subparagraph (A) a nullity.

In my view, this is incorrect. The general policy is that visas should be issued without concern for nationality. The history of this provision reveals that Congress’s primary concern was to eliminate nationality-based quotas, which were a staple of our immigration law before 1965. But when the Secretary deems it necessary, he can adopt special procedures for issuing visas that take into consideration nationality. Simple examples could include the adoption of different procedures for embassies in certain countries where the risk of fraud is higher. Broader examples could include, as here, the adoption of stronger vetting processes to screen out risks of terrorism. The prohibition is not nearly as strict as subparagraph (A) suggests. In any event, the best way to avoid reading subparagraph (A) as a nullity is to treat “entry” and the issuance of visas as separate concepts. This reading is buttressed by § 1182(f), which gives the Secretary the ultimate trump card: deny entry to classes of aliens, even if they have validly issued visas, based on national security concerns. The greater power to deny entry includes the far lesser power to deny a visa.

Second, Judge Thacker notes that when enacting § 1152 more than a decade after § 1182, Congress intended to curtail the president’s authority under the older provision. But there appears to be nothing in the legislative history of the 1965 INA to suggest that Congress was attempting to restrict the president’s power over denial of entry. In addition, as a matter of experience, Judge Thacker fails to account for historical practice, such as the fact that President Reagan expressly barred Cubans from entering the United States under § 1182. In his Proclamation 5517 (1986), he justified this exclusion on the ground that he “found that the unrestricted entry into the United States as immigrants by Cuban citizens would . . . be detrimental to the interests of the United States.”

Judge Thacker was no doubt aware of the proclamation, as Judge Wynn addresses it in his concurring opinion. Judge Wynn argues in footnote 11, however, that this single action was never reviewed by a court, may be unconstitutional, and is not sufficient to establish a new precedent. While I agree that past practice does not suggest an action is constitutional, presidential custom provides a helpful gloss on our laws and the parameters of executive power. But beyond this gloss, there is another canon of statutory interpretation that bears on the resolution of these questions. Indeed, it is the elephant in the room: Article II.

Inherent Article II Power

President Trump’s denial of entry to nationals of six countries, like President Reagan’s denial of entry to Cubans, is premised not entirely on § 1182 but also on the president’s Article II authority. That invocation of power has implications for how we should interpret the parameters of his power, and the relevant statutory framework.

As I noted in one of my earlier posts on the topic, § 1182 can be viewed as a reaffirmation of the president’s Article II authority to turn away those at the border he deems detrimental to American interests. On this view, the president would not even need statutory authority. Consider if in the immediate aftermath of Pearl Harbor, even before the Declaration of War, President Roosevelt unilaterally halted the entry of Japanese nationals into the United States until Congress could act. Even in the absence of congressional authorization, such a temporary measure would almost certainly be lawful. (FDR had such statutory authority to bar entry to aliens pursuant to his declaration of a “national emergency,” though I suspect this statute was a reaffirmation of what the Commander in Chief would already do in such a scenario).

Here, however, we do not need to worry about unilateral authority in this case, because Congress gave the President. explicit statutory authority. This delegation brings us within Justice Jackson’s first tier of Youngstown. Here, presidential power is at its maximum, and judicial scrutiny is at a minimum. Courts should avoid any construction of § 1152 or § 1182 that intrudes on the president’s authority to deny entry to those he deems dangerous. But there is no need to rely on an avoidance canon, because both statutes are best read to allow the President to deny entry to classes of aliens, an act separate and apart from the issuance of visas.

Note that Judge Wynn directly addresses the argument concerning inherent constitutional authority:

Additionally, because the Executive Order cites the Immigration Act as the sole statutory basis for the President’s authority to proclaim Section 2(c)’s suspension on entry, I need not, and thus do not, take any position on the scope of the President’s delegated power to deny entry to classes of aliens under other statutes. Likewise, because the claim at issue relates only to Section 2(c)’s compliance with the Immigration Act, I do not address whether, and in what circumstances, the President may deny entry to classes of aliens under his inherent powers as commander-in-chief, even absent express congressional authorization. See The Prize Cases, 67 U.S. 635 (1862) (p. 98).

The Obama administration was hesitant to defend their actions based on inherent Article II powers, preferring instead to rely on intricate and multi-factored statutory arguments, aimed at cabining the authority of the Presidency. (See Jack Goldsmith’s commentary about the release of detainees at Guantanamo Bay and the longer-than-sixty-day bombing campaign of Libya). The Trump administration, perhaps like the Bush administration before it, may not be so hesitant. To the extent these statutory arguments prevail on appeal, the government may raise constitutional avoidance arguments.

Along similar lines, Judge Wynn suggests that Congress (and by extension, the president) has greater powers to exclude aliens during times of war. By implication, we are not at war, and such power is therefore lacking. Citing Justice Frankfurter’s concurring opinion in Korematsu (a case that is worlds apart from IRAP v. Trump), Judge Wynn argues that “Congress’s constitutional power to control immigration—and authority to delegate that control—fundamentally differs in a time of war.” He adds, “The Supreme Court’s broadest statements regarding the scope of the president’s delegated powers over immigration—which are relied upon by the Government—are in cases in which Congress expressly declared war and authorized the president to deny entry to aliens as part of his prosecution of the conflict.”

Are we at war? This argument raises the question of whether the 2001 and 2002 Authorizations for Use of Military Force, which both the Obama and Trump administrations have used to fight ISIS, provide support for the president’s terrorism-based travel ban. (To be precise, I do not think either AUMF provides authority to cover conflicts with ISIS). This is yet another reason to read the statues to avoid reaching these thorny constitutional questions. The Supreme Court has shied away from this question, as reflected in its decision to reverse Justice Douglas’s holding in Schlesinger v. Holtzman that the bombing runs in Cambodia were part of an undeclared, unconstitutional war.

Judge Keenan’s Concurring Opinion

Judge Keenan’s concurring opinion (pp. 81-93) proposes a different statutory argument. Unlike Judge Thacker’s reading of the INA, which would provide relief only to those seeking immigrant visas, Judge Keenan’s position would be something of a silver-bullet that wins the case for the challengers without touching on the constitutional question. In short, she argues that President Trump’s proclamation did not make sufficient findings to meet the requirements of Section 1182(f), and thus lacks the authority to deny entry to aliens from the six selected nations.

Recall that the second executive order provides that the unrestricted entry of nationals from the six target countries “would be detrimental to the interests of the United States” and then “direct[s] that the entry into the United States of nationals of those six countries be suspended for 90 days from the effective date of this order, subject to the limitations, waivers, and exceptions set forth in sections 3 and 12 of this order.”

Judge Keenan explains that Trump’s findings are too speculative to meet the stringent requirements of § 1182(f). She writes that “[t]he plain language of Section 1182(f) permits a president to act only if he ‘finds’ that entry of the aliens in question ‘would be detrimental to the interests of the United States’” and that “an unsupported conclusion will not satisfy this “finding” requirement.” Accordingly, she walks through the four corners of the executive order—avoiding any campaign-trail missives—and concludes that the “the Second Executive Order does not state that any nationals of the six identified countries, by virtue of their nationality, intend to commit terrorist acts in the United States or otherwise pose a detriment to the interests of the United States” (p. 88).

Here is her key analytical move: § 1182(f) concerns the exclusion of individuals, not countries. In contrast, President Trump’s executive order focuses on countries, not individuals. Because the executive order “does not identify a basis for concluding that entry of any member of the particular class of aliens, namely, the more than 180 million nationals of the six identified countries, would be detrimental to the interests of the United States,” the order does not meet the requirements of Section 1182(f), the text of which “plainly requires more than vague uncertainty regarding whether their entry might be detrimental to our nation’s interests” (p. 89).

In short, § 1182(f) mandates that the president find that “the entry of any aliens or of any class of aliens into the United States would be detrimental to the interests of the United States” (emphasis added). Judge Keenan asserts, and the Solicitor General’s argument does not dispute, that the executive order only concerns aliens that may be detrimental. So the question is: would the Supreme Court halt the travel ban because of the difference between would and may?

This is a very sophisticated textual argument, which I flagged on Lawfare and in the NY Daily News following oral arguments. Critically, it offers the Supreme Court a way to rule against the executive without impairing other government policies—indeed only Presidents Reagan and Trump employed § 1182(f) against nationals of specific country, rather than nationals with offending affiliations. It also avoids reaching difficult constitutional questions concerning the Establishment and Due Process Clauses. (Though constitutional avoidance may caution against such a construction).

Were this a run-of-the-mill statutory interpretation case, I would be inclined to agree with Judge Keenan’s textualist approach. But such is not the case. I think Judge Keenan’s parsimonious construction of § 1182(f) is inconsistent with the loose mode of statutory interpretation employed by the Supreme Court in Dames & Moore v. Regan.

Dames & Moore v. Regan

After the Iran hostage crisis, President Reagan agreed to suspend court proceedings involving Iran and to nullify pending attachments of Iranian property. The Dames & Moore company, which had a valid judgment against Iran, argued that Reagan’s actions were not supported by statutory authority. The opinion by Justice Rehnquist reaffirmed Justice Jackson’s seminal concurring opinion three decades earlier in Youngstown Sheet & Tube Co. v. Sawyer explaining that the president’s powers are their height when he is acting pursuant to congressional authorization and at their “lowest ebb” when acting contravention of congressional will. (Fun fact: Justice Rehnquist clerked for Justice Jackson the year Youngstown was decided; and Chief Justice Roberts clerked for Justice Rehnquist the year Dames & Moore was decided).

In Youngstown, recall that Justice Jackson found that Congress had not given President Truman the statutory authority to seize the steel mills to avert a labor strike. Along similar lines Justice Frankfurter noted in his concurring opinion, that Congress expressly considered granting the president such authority, but withheld it: “By the Labor Management Relations Act of 1947, Congress said to the President, ‘You may not seize. Please report to us and ask for seizure power if you think it is needed in a specific situation.’”

Under the Youngstown rubric, § 1182(f) is worlds apart from the Labor Management Relations Act of 1947. Here, we have a direct delegation of authority for the president to deny entry to those he deems detrimental to the interests of the United States. Against this backdrop, reading too much into the distinction between may versus would is inconsistent with the framework of cases affecting presidential power and national security. Indeed, as I’ve argued since the outset of this case, the statutory analysis reveals that we are in Jackson’s first tier, and scrutiny must be at a minimum.

But were the inquiry difficult under Youngstown, the task becomes far easier under Justice Rehnquist’s extremely flexible gloss from Dames & Moore. To find the requisite statutory authority to uphold President Reagan’s suspension of claims against Iran, the unanimous Court employed a form of interpretive limbo, reaching to two statutes that bear only indirectly on what President Reagan did: the International Emergency Economic Powers Act (IEEPA) and the Hostage Act. Neither expressly permitted the suspension of claims, but the Court did not find that fact dispositive. Justice Rehnquist deemed “both statutes highly relevant in the looser sense of indicating congressional acceptance of a broad scope for executive action in circumstances such as those presented in this case.” He explained:

We think both statutes highly relevant in the looser sense of indicating congressional acceptance of a broad scope for executive action in circumstances such as those presented in this case. . . . . Congress cannot anticipate and legislate with regard to every possible action the President may find it necessary to take, or every possible situation in which he might act. Such failure of Congress specifically to delegate authority does not, “especially . . . in the areas of foreign policy and national security,” imply “congressional disapproval” of action taken by the Executive. Haig v. Agee, ante at 453 U. S. 291. On the contrary, the enactment of legislation closely related to the question of the President’s authority in a particular case which evinces legislative intent to accord the President broad discretion may be considered to “invite” “measures on independent presidential responsibility,” Youngstown, 343 U.S. at 343 U. S. 637 (Jackson, J., concurring). At least this is so where there is no contrary indication of legislative intent and when, as here, there is a history of congressional acquiescence in conduct of the sort engaged in by the President. It is to that history which we now turn. (emphasis added)

In contrast, President Trump did precisely what § 1182(f) authorized him to do: deny entry to classes of aliens he deems detrimental. You don’t need to turn to a “looser” approach, or consider “closely related” measures to find that President Trump had the requisite statutory authority to act. Congress said so in the statute with unequivocal language.

For these reasons, I don’t think Judge Keenan’s textual argument will prevail. The Court has never employed such a stringent and rigid approach to construing Congress’s delegations to the president in matters of national security. To the contrary, the Court in Dames & Moore adopted extremely dubious constructions of statutes to find support for executive actions—far more implausible than reading the word “would” in § 1182(f) more permissibly.

Finally, construing § 1182(f) in the narrow fashion Judge Keenan suggests requires reconsideration of two questions addressed earlier. First, President Reagan’s proclamation concerning Cuban nationals made no specific findings about why Cuban nationals “would” be detrimental to the interests of the United States, beyond generalized concerns about “normal migration procedures between the two countries.” To say Trump’s declaration is void is to say Reagan’s declaration was also void. Second, even in the absence of the statute, the president arguably has the power to exclude certain aliens based on Article II and the 2001 and 2002 Authorizations for Use of Military Force. (Again, I disagree with the Obama and Trump administration’s constructions of the AUMF, but they could advance this argument in court). Reading the statute so narrowly would require the Court to broach that issue. Or, the Court can simply read § 1182(f) as Dames & Moore compels, and find the president met the requirements.

Judge Wynn’s Concurring Opinion

I turn last to Judge Wynn’s concurring opinion (pp. 94-128), which sounds in the major question doctrine. He writes, “the Immigration Act provides no indication that Congress intended the ‘broad generalized’ delegation of authority in Section 1182(f) to allow the President ‘to trench . . . heavily on [fundamental] rights’” (p. 96). That is, if Congress had intended to allow the president to encroach on protected classes or classifications, it would have made the delegation explicit. Or, as Justice Scalia noted in MCI v. AT&T, Congress “does not, one might say, hide elephants in mouseholes.”

Under Judge Wynn’s self-styled “delegation of authority cannon,” when a statute does not provide explicit evidence of that delegation and the delegation potentially encroaches on fundamental rights, the delegated powers must be construed narrowly.

The Supreme Court’s 1958 decision in Kent v. Dulles is important to Judge Wynn’s point. As I noted in the Harvard Law Review, long before the articulation of the major question doctrine in cases like MCI v. AT&T and FDA v. Brown & Williamson Tobacco, “the Court recognized [in Kent] that Congress does not cryptically delegate to agencies unbounded discretion to burden constitutional rights.” Judge Wynn’s position parallels the brief I co-authored in the Little Sisters of the Poor case, contending that the Affordable Care Act did not delegate to the executive branch the authority to infringe on religious liberty. Our brief contended that Congress never delegated to the agencies the power to decide which religious organizations receive “exemptions,” and which receive mere “waivers.” As a result, the regulation in question was beyond the departments’ ken.

The theory has gained influence elsewhere. For example, Judge Kavanaugh’s recent dissent from denial of rehearing en banc in the net neutrality case calls for a reinvigoration of what he calls the major rule doctrine. I second that call—though I suspect IRAP v. Trump is not the appropriate case.

Judge Wynn’s approach offers the benefit of avoiding a constitutional ruling; that is, the court could strike down the proclamation as ultra vires, instead of as unconstitutional. But the problem with his approach is that where the challenged action is itself unconstitutional, the major question doctrine adds nothing to the equation. To apply the major question doctrine, the court would still have to reach the question of whether the proclamation is unconstitutional. As he notes on page 106, his analysis requires determining whether “Section 1182(f) would encroach on the core constitutional values set forth in the First, Fifth, and Fourteenth Amendments.” This framework is avoidance in name only. Indeed, the statutory argument in the concurring opinion is indistinguishable from the constitutional argument in the majority opinion, relying on the same sorts of evidence to find animus.

Cross-Posted at Lawfare