The Yale Law Journal


Trump v. Hawaii: How the Supreme Court Simultaneously Overturned and Revived Korematsu

30 Jan 2019

abstract. Neal Katyal, who argued Trump v. Hawaii at each level of the federal court system, compares the Supreme Court’s decision to uphold President Trump’s travel ban to the Court’s decision nearly seventy-five years ago to affirm the internment of Japanese Americans in Korematsu. He concludes that while Hawaii overturned Korematsu, it essentially recreated the doctrine under another name. The phoenix of excessive deference to the executive unfortunately persists as an entrenched part of our jurisprudence.


In 1944, Justice Jackson delivered an ominous warning to his colleagues on the bench and to the nation. By sanctioning the internment of Japanese Americans and upholding an exclusion order based on a “mere declaration”1 of “reasonable military necessity,”2 the Supreme Court’s decision in Korematsu v. United States would “lie[] about like a loaded weapon ready for the hand of any authority that c[ould] bring forward a plausible claim of an urgent need.”3 For decades, Justice Jackson’s warning has been characterized as a fear of the past or an alternate history that never came to be.4 Indeed, the Court’s decision in Korematsu has joined the ranks of this country’s most notorious antiprecedents—textbook cases of judicial decision-making gone wrong that jurists of all stripes vow never to repeat.5 But on the eve of Korematsu’s seventy-fifth anniversary, a majority of the Court brought to fruition what Justice Jackson predicted so long ago. Despite overturning Korematsu, the Court’s decision in Trump v. Hawaii6 perpetuates the very-near-blind deference to the executive branch that led the Korematsu Court astray.

In her dissent, Justice Sotomayor explicitly spelled out the glaring similarities between the two cases7—a move that garnered sharp criticism from some of her colleagues on the Court as well as commentators.8 According to her critics, Justice Sotomayor’s Korematsu reference was not just inapposite but a rhetorical cheap shot.9 Yet a closer examination of the language and arguments in the two cases shows that Justice Sotomayor was on to something. The parallels between these two opinions are striking, both on the surface and in the underlying substance. And if the “court of history” tells us anything,10 these cases will share a similar legacy too. Hawaii will join Korematsu on the list of antiprecedents having “no place in law under the Constitution.”11

In the ensuing pages, I lay out these parallels. The parallels were not ones that Justice Sotomayor discovered at the conclusion of the litigation. Rather, at every step of the Hawaii case, I was aware of them; indeed, I had a unique perspective on how bad Korematsu was for our nation. Six years before Donald Trump was elected President, I confessed error on behalf of the Solicitor General’s Office for its conduct in litigating Korematsu. I knew firsthand the dangers of untrammeled executive power. Our amici and supporters knew this as well, so much so that my last physical contact before standing up to give the oral argument in Hawaii was the warm embraces in the Supreme Court chamber given to me by Senator Mazie Hirono and Karen Korematsu, the daughter of Fred Korematsu.

In the months since the Supreme Court’s five-four decision and in the wake of the criticism of Justice Sotomayor, I’ve wondered whether those of us who litigated the case were unduly sensitive to the commonalities between Hawaii and Korematsu. In the end, the answer is no.

This Essay explains why. In Part I, I explain many of the surface-level similarities between the cases, such as the way in which the history and purpose of both governmental actions were laundered. Readers of the Hawaii majority opinion would have little idea about the terrible history of Trump’s travel ban, including many of the anti-Muslim statements he had made over the years. As in Korematsu, it took the dissent to detail these buried facts at length. In Part II, the Essay details how the surface-level similarities are eclipsed by a much more fundamental commonality between the two decisions: they are, at their root, decisions that place their unbounded trust in the President when he asserts military necessity. One way of putting the point is this: it was not hard for Chief Justice Roberts in Hawaii to overrule Korematsu in name, since he merely recreated its reasoning under a different appellation. The Court still has the same tool in its toolkit—it’s just that the case now begins with a T.

I. surface-level similarities from a side-by-side comparison

Of course, Korematsu and Hawaii are not the same case. They involve orders of substantially different dimensions and scope. To put it briefly, in Korematsu, the Military Commander of the Western Defense Command, General John DeWitt, under the authority of the President, ordered the relocation of Japanese Americans—including United States citizens—to internment camps according to explicit race-based classifications.12 In Hawaii, the President of the United States, Donald Trump, ordered a ban on foreign nationals’ entry to the country using a facially neutral policy that predominantly impacts Muslim-majority nations.13 These cases are products of distinct time periods, and their facts are undisputedly different. But it’s the legal reasoning and the remarkable commonalities in the language and arguments—not the facts—that are telling.

Begin with the statements of the officials responsible for the respective orders—statements that were available to both the Korematsu and Hawaii Courts and provide the contextual backdrop against which these cases were decided.14 As Justice Murphy noted in his Korematsu dissent, General DeWitt described people of Japanese ancestry as “subversive,” part of “an “enemy race,” and members of “a large, unassimilated, tightly knit racial group, bound to an enemy nation by strong ties of race, culture, custom, and religion.”15 Similarly, President Trump called for a “total and complete shutdown of Muslims entering the United States.”16 He also announced that “Islam hates us,”17 retweeted “three anti-Muslim propaganda videos” just two months after issuing the third travel ban,18 and stated that “it was ‘very hard’ for Muslims to assimilate into Western culture.”19 If the symmetries were not obvious, President Trump connected the dots himself. He likened his proposed ban during the campaign to the internment at issue in Korematsu, “noting that President Franklin D. Roosevelt ‘did the same thing’ with respect to the internment of Japanese Americans during World War II.”20 And while his lawyers sometimes tried to suggest otherwise, President Trump never once disavowed those remarks or said he was not trying to impose a Muslim ban.21

While less blatant, the majority opinions in both cases share common arguments and rhetorical devices, too. In Korematsu, Justice Black purported to draw a hard constitutional line when protecting individual liberties, citing only the most extreme threats to national security as justifying deviation from core constitutional protections. In his words, only “circumstances of direst emergency and peril” present an exception to the general rule that “[c]ompulsory exclusion of large groups of citizens from their homes . . . is inconsistent with our basic governmental institutions.”22 In Hawaii, Chief Justice Roberts purported to establish a similar exception, noting that while “[t]he clearest command of the Establishment Clause is that one religious denomination cannot be officially preferred over another,”23 and “[w]hile we of course ‘do not defer to the Government’s reading of the First Amendment,’ the Executive’s evaluation of the underlying facts is entitled to appropriate weight,” especially in the context of national security.24

But in both cases, the majorities tempered these implicit promises of (albeit marginal) judicial oversight by hiding behind the shield of the executive branch’s institutional competence. Take, for instance, the Korematsu Court’s emphasis on its lack of military expertise, declaring that it could not “reject as unfounded the judgment of the military authorities” or “say that the war-making branches of the Government did not have ground for believing” that the actions they took were necessary.25 For its part, the Hawaii Court explained that whether or not President Trump’s policy is “overbroad” or “does little to serve national security interests,” the Court “cannot substitute [its] own assessment for the Executive’s predictive judgments on such matters, all of which ‘are delicate, complex, and involve large elements of prophecy.’”26

The most glaring similarity is the fervor with which the two majority opinions dismissively rejected the dissents’ respective characterizations of the facts. Consider Justice Black’s retort to his three dissenting colleagues:

Our task would be simple, our duty clear, were this a case involving the imprisonment of a loyal citizen in a concentration camp because of racial prejudice. Regardless of the true nature of the assembly and relocation centers—and we deem it unjustifiable to call them concentration camps with all the ugly connotations that term implies—we are dealing specifically with nothing but an exclusion order. To cast this case into outlines of racial prejudice, without reference to the real military dangers which were presented, merely confuses the issue.27

Chief Justice Roberts’s remarks in Hawaii read like a sequel:

Whatever rhetorical advantage the dissent may see in doing so, Korematsu has nothing to do with this case. The forcible relocation of U.S. citizens to concentration camps, solely and explicitly on the basis of race, is objectively unlawful and outside the scope of Pres­idential authority. But it is wholly inapt to liken that morally repugnant order to a facially neutral policy deny­ing certain foreign nationals the privilege of admission. The entry suspension is an act that is well within executive authority and could have been taken by any other President . . . .28

There’s a bit of protesting too much in both of these passages. Despite the insistence that Korematsu is utterly distinct from Hawaii, the language, style, and arguments of the two majority opinions suggest otherwise.

A similar picture emerges in the dissents. In a single line, Justice Roberts’s dissent in Korematsu captures a principal frustration of the dissenters in both cases. “Why,” he asked, “should we set up a figmentary and artificial situation instead of addressing ourselves to the actualities of the case?”29 For Justice Roberts, the Korematsu majority’s central failure was treating the exclusion order as a “hypothetical” extension of a curfew, when a thorough review of the facts (specifically, those known to the Court but not fully detailed in the majority opinion) revealed an “over-all plan for forceable detention.”30 This is essentially the same line of argument that Justice Breyer employed in his Hawaii dissent. To Justice Breyer, the case boiled down to the actual implementation of the ban and its waiver and exemption system.31 Just as Justice Roberts had chided the Korematsu majority for deciding a hypothetical version of the case,32 Justice Breyer framed the issue in Hawaii as involving two potential alternatives. In the first, the government properly applied the Proclamation’s waiver and exemption provisions—an assumption the majority took for granted. In the other, Justice Breyer described a world where waivers and exemptions were merely a form of “window dressing.”33 As was the case when Justice Roberts laid out a more complete picture of the facts in Korematsu, Justice Breyer’s survey of the evidence available in Hawaii shattered the facade that the majority had erected, calling into question a number of the majority’s central premises.34

And what goes for Justice Breyer goes for Justice Sotomayor’s dissent in spades. Just four sentences into her opinion, Justice Sotomayor charged the Court with “leav[ing] undisturbed a policy first advertised openly and unequivocally as a ‘total and complete shutdown of Muslims entering the United States’ because the policy now masquerades behind a façade of national-security concerns.”35 She went on to label the waiver program a “sham”36 and to criticize the majority for providing a “highly abridged account” of the facts instead of the “full record” that “paints a far more harrowing picture.”37

Justice Sotomayor’s opinion is eerily reminiscent of the remaining two Korematsu dissents, authored by Justices Murphy and Jackson, respectively. Not only did Justice Sotomayor directly quote from these dissents,38 but she also reiterated their rejection of national security exceptionalism and mirrored their concerns about doctrinal manipulation. To start, in Korematsu, Justice Murphy insisted on placing “definite limits [on] military discretion,” arguing that “like other claims conflicting with the asserted constitutional rights of [an] individual, [a] military claim must subject itself to the judicial process of having its reasonableness determined and its conflicts with other interests reconciled.”39 Justice Sotomayor urged the same, noting that “[a]lthough national security is unquestionably an issue of paramount public importance, it is not ‘a talisman’ that the Government can use ‘to ward off inconvenient claims—a label used to cover a multitude of sins.’”40

The parallels don’t end there. Both Justice Jackson and Justice Sotomayor accused the majorities of a bait-and-switch, calling into question what the Court had said just months earlier. In Hirabayashi v. United States,41 a case decided just six months before Korematsu, the Court had expressly limited its holding to the curfew requirements at issue, declining to pass judgment on other military orders.42 But according to Justice Jackson, by relying on Hirabayashi and claiming that it dictated the outcome in Korematsu, the majority was “now saying that in Hirabayashi we did decide the very things we there said we were not deciding.”43 Along similar lines, Justice Sotomayor condemned the Court’s refusal to follow the very principles it laid out “just weeks [before]”44 in Masterpiece Cakeshop, Ltd. v. Colorado Civil Rights Commission.45 “Unlike in Masterpiece, where the majority considered the state commissioner’s statements about religion to be persuasive evidence of unconstitutional government action, the majority here completely sets aside the President’s charged statements about Muslims as irrelevant.”46 Looking beyond recent precedent, both Justices took the Court to task for manipulating the broader constitutional inquiry. For instance, Justice Jackson noted that, “whether consciously or not,” the majority is “distort[ing] the Constitution to approve all that the military may deem expedient.”47 By the same token, Justice Sotomayor criticized the Court for “incorrectly appl[ying] a watered down legal standard” that vitiates Establishment Clause jurisprudence.48

Taken together, the opinions in Hawaii read like a modern-day adaptation of Korematsu. Normally, this would not be surprising—after all, the same institutional body decided both cases, and later decisions necessarily draw from the language and arguments of their earlier counterparts. But Korematsu is not just any case. It is one of the most widely rejected and disfavored decisions in the Court’s history. And it is altogether surprising that a decision that purported to overrule Korematsu in effect recreated its reasoning.

II. blind deference to partial truth: the failure to learn from korematsu

It is not simply the standard of review (strong deference) that creates the parallels in these cases, but also the way in which the majorities applied the standard to the facts at hand. In each case, the dissents condemn the Court’s unquestioning acceptance of the government’s “half-truths”49 and “self-serving” statements,50 “all in the name of a superficial claim of national security.”51 This shared criticism is much more than a surface-level similarity—it explains how, despite overturning Korematsu, the Court “merely replace[d] one ‘gravely wrong’ decision with another.”52 Both Courts adopted a posture of broad deference to the executive—even when individual constitutional rights are infringed—when the executive asserts that a policy is necessary to ensure the nation’s security.

Consider how the Korematsu majority reached its decision. It accepted at face value the government’s attestations of military necessity, never questioning whether this conclusion was justified by anything more than General DeWitt’s own declarations.53 It never considered whether the military’s pronouncements of urgency actually made sense—not as a matter of policy, but for the purpose of credibility given that a significant period of time had passed since Pearl Harbor.54 Nor did the majority mention the “unverified” and vague nature of multiple findings in General DeWitt’s report.55 In fact, the majority failed to even consider General DeWitt’s patently racist comments about people of Japanese ancestry “belonging to ‘an enemy race’ whose ‘racial strains are undiluted’”—statements that could not be based on any purported military expertise.56 And it surely did not recognize the “sharp controversy as to the credibility of the DeWitt report” itself.57

Rather, the ­majority cited General DeWitt’s report just once—without even naming the General in the main text58—and supplemented the report by citing two congressional hearings, even though those hearings actually provide considerable reason to doubt the justifiability of the military’s assertions.59 The Court explicitly refused to look at “the true nature of the assembly and relocation centers,”60 opting instead to consider the exclusion order in a vacuum, wholly separate from the context in which it was executed. By the end of its opinion, the majority appeared to have accepted the government’s position as not just sufficiently reasonable but also right. The Court simply declared that “Korematsu was not excluded from the Military Area because of hostility to him or his race,”61 without providing any analysis to support such a definitive conclusion. Really the decision amounts to this: constitutional protections can be put on hold if the government asserts a remotely plausible claim of military necessity, and the ugly real motivations for a government policy can be swept under the rug.62

Yet the majority’s blind deference and its abdication of judicial responsibility are not the only reasons why Korematsu’s legacy is so spoiled. The Solicitor General withheld critical information from the Court—information that, even under the Court’s extremely deferential approach, could not have sustained the majority’s holding.63 In the government’s brief for Hirabayashi, the case on which the Korematsu majority so heavily relied, the Solicitor General omitted any mention of the Office of Naval Intelligence’s report that concluded “the entire ‘Japanese Problem’ ha[d] been magnified out of its true proportion, largely because of the physical characteristics of the people.”64 By the time Korematsu reached the Court, the Federal Bureau of Investigation, the Federal Communications Commission, and two members of the Alien Enemy Control Unit had also concluded that General DeWitt’s report was not credible and that the internment orders were not justified.65 But the Solicitor General still stood by General DeWitt’s report, going so far as to tell the Court during oral argument that “no person in any responsible position has ever taken a contrary position” to the one the government “has always maintained since the Hirabayashi case.”66

At the time Korematsu was decided, little of this information was known to the Court. But it was certainly known by the time of Hawaii, as the true story had emerged in the 1980s.67 Indeed, when I served as Acting Solicitor General of the United States, in 2011, I officially confessed error on behalf of the Justice Department in both Hirabayashi and Korematsu.68

Against this backdrop of the judiciary’s blind deference and the executive’s wanton dishonesty, Korematsu has taken its place in history as a warning message for judges, jurists, and government officials alike. Even apart from the Solicitor General’s suppression of valuable information, Korematsu’s legacy is inextricably linked to the majority’s failure to look past the government’s blanket assertions and evaluate the exclusion order for what it really was.69 In light of this infamous legacy, it is all the more surprising to see the current Court “redeploy[] the same dangerous logic”70 seventy-four years after the initial decision, and just seven years after the confession of error.

Arguing the Hawaii case, I was constantly aware of the ways in which military necessity claims had been abused by the White House in the past. I am not someone who generally believes courts should be in the business of second-guessing national security decisions. But when those decisions are the product of animus, and when they are not fully vetted by the interagency process and the national security professionals trained to make such decisions,71 some judicial scrutiny is not only appropriate—it is necessary.

Unfortunately, just like the majority in Korematsu, the Hawaii Court refused to thoroughly examine the evidence. In fact, in setting the standard of review, the majority expressed reluctance to even “look beyond the face of the Proclamation” and ask “whether the entry policy [was] plausibly related to the Government’s stated objective[s].”72 Were it not for “the Government[’s] . . . suggest[ion] that it may be appropriate here for the inquiry to extend beyond the facial neutrality of the order,” a “conventional” look at the facial legitimacy of the policy would have “put an end” to the matter.73

Even grossly discriminatory actions can be written in facially neutral terms. Such a legal test does nothing but reward clever lawyering. The Hawaii majority tried to grapple with the extrinsic evidence, but did not engage with everything Trump said—it merely recited a select few examples.74 For instance, when then-candidate Trump reformulated his proposed Muslim ban to focus on immigration and terrorism, he explained that the new proposed policy was in fact an “expansion” of the Muslim ban, and that “he used different terminology because ‘[p]eople were so upset when [he] used the word Muslim.’”75 Despite this statement’s clear connection to the Proclamation at issue—which emphasized facially neutral concepts like immigration and terrorism after previously promising a Muslim ban—the majority did not even cite it. Nor did the Court mention President Trump’s statement about how, under the first version of the travel ban, “Christians would be given priority for entry as refugees to the United States” because it was “very unfair” that Muslims had historically enjoyed such privileges.76 Likewise, the majority failed to consider President Trump’s infatuation with General Pershing’s supposed massacre of Muslim terrorists in the 1900s, a topic he repeatedly discussed.77

According to the majority, President Trump’s statements were unnecessary to evaluate because the Proclamation contained sufficiently legitimate justifications to satisfy rational basis review.78 And so the Hawaii majority ultimately concluded that the Proclamation is “a facially neutral policy denying certain foreign nationals the privilege of admission.”79 But in focusing on the four corners of the Proclamation, the Court ignored the tainted influence of the President’s comments on the Proclamation itself. Though the majority recognized that President Trump had instructed his lawyers to craft a “legal[]” version of a Muslim ban,80 it assessed the Proclamation as if it had randomly dropped out of the sky.

Just like the majority in Korematsu, which ignored General Dewitt’s racism and pretended the government’s actions were not what they really were, the Hawaii Court refused to consider the broader context of the travel ban. It did not even question whether the waiver and exemption programs were actually applied in practice—and declined to engage with any of the data on the programs presented by amici­81—despite relying on the existence of such programs to support its holding.82 The Court did not ask whether the policy, if truly designed to meet its stated objectives, made plausible sense.83 Notwithstanding President Trump’s comments about Christians and Muslims in connection to the first travel ban, the majority countered Hawaii’s arguments about religious animosity by simply pointing to the Proclamation and noting that “the text says nothing about religion.”84 And just as the majority in Korematsu accepted General DeWitt’s findings at face value,85 the majority in Hawaii did the same with the Trump administration’s interagency review. That report ostensibly contains findings in support of the travel ban, but the Court never saw a copy of the document.86

Indeed, the Administration went to great lengths to suppress the worldwide review—the asserted basis for the ban—from being released to the public, both in the Hawaii litigation and elsewhere (such as FOIA litigation).87 The extent to which the Administration insisted that the report stay secret, and that not even portions of it be disclosed with redactions, is somewhat suspicious. Meanwhile, other documents that were released present greater cause for concern. In February 2017—following the President’s issuance of his first Executive Order—a draft report authored by the Department of Homeland Security was leaked, finding that “citizenship [is] likely an unreliable indicator of terrorist threat to the United States.”88 In other words, a mere six months before the Department helped produce the review that purportedly justified the third travel ban, it had drafted a report calling into question the entire rationale for the policy.

And here is where the connections between Korematsu and Hawaii become even eerier. Not only did the government resist providing the Court with any substantive evidence against which it could measure—or at least consider—the government’s claim of necessity; it even allowed for improper information to go uncorrected. At oral argument, Solicitor General Francisco assured the Court that President Trump had “‘made crystal-clear on September 25 that he had no intention of imposing the Muslim ban’ and ‘has praised Islam as one of the great countries [sic] of the world.’”89 Yet the President never made any statement along those lines on September 25 (the day after the third travel ban was announced), and the Solicitor General later informed the Court that the actual date of the President’s statement was January 2017, not September.90 That statement concerned the first ban, not the one before the Court. And whatever the statement meant, the fact remained that President Trump never formally renounced the Muslim ban, and neither had any spokesperson for the White House—not former press secretary Sean Spicer, not deputy press secretary Raj Shah, and not current press secretary Sarah Huckabee Sanders.91 The government’s claim was, well, a “half-truth.”92


For decades, Korematsu has been viewed as a cautionary tale. But until Hawaii, it had never been overturned—a fact that set it apart from its fellow antiprecedents.93 As a renounced decision still on the books, it served as a stark reminder of failed legal reasoning and the dangers of blindly adhering to the government’s assertions.94 Given this legacy, it seemed only natural that in overturning Korematsu, the Court would strictly adhere to these lessons. After all, as Richard Primus articulately put it, “the court is the chief narrator of American constitutional history, so an ugly chapter from the past can never be fully closed until the court itself writes the better ending.”95

Hawaii could have been that “better ending.” With hundreds of pages in the Joint Appendix, the Court had more than sufficient evidence to affirm the injunction on religious animus grounds.96 Unlike the hidden intelligence report in Korematsu, President Trump’s anti-Muslim comments were littered across various media outlets, “plastered on the candidate’s website,” and “staring everyone in the face.”97 And yet when given the chance to memorialize Korematsu’s lessons, the Court instead made almost every mistake in Korematsu’s playbook—it accepted the government’s arguments at face value, deferred to the executive branch without ensuring that deference was warranted, and confined itself to a narrow review of the Proclamation, examining a “figmentary and artificial”98 case instead of the one actually before it. For these reasons, it will come as no surprise when, one day in the future, Trump v. Hawaii is eventually overturned. But let us hope that when that happens, the Court ends this line of cases for good, rather than resurrect it by another name.

Neal Katyal is the Saunders Professor of Law at Georgetown University, a partner at Hogan Lovells, LLP, and lead counsel for the State of Hawaii and other plaintiffs in Trump v. Hawaii. Thanks to Natalie Salmanowitz for fabulous assistance in preparing this Essay, and to the incredible appellate team at Hogan Lovells (and especially Mitch Reich, Tom Schmidt, Colleen Roh Sinzdak, and Sara Solow) for their brilliant work in litigating the Hawaii case.