Eroding Immigrants’ Rights Through the “New” New Textualism

Shalini Bhargava Ray*

The Supreme Court decided three immigration cases in the 2021-22 term, Johnson v. Arteaga-Martinez,^[1] Garland v. Aleman Gonzalez,^[2] and Patel v. Garland,^[3] handing the government three wins. These cases decided significant questions for noncitizens fighting deportation. They rejected statutory arguments to limit the government’s detention power and largely closed the courthouse door to noncitizens in two other high-stakes cases. Noncitizens ordered removed who face lengthy detention must now bring claims under the Due Process Clause to compel the government to grant them a bond hearing, for the Court decided that the Immigration and Nationality Act (“INA”) does not require one.^[4] They must bring these claims as individuals rather than as a class, for the Court decided that lower federal courts lack jurisdiction to issue class-wide injunctions to compel the Executive to comply with immigration law.^[5] They cannot obtain judicial review of an agency’s factual findings underlying a judgment granting some immigration benefits, leaving blatant factual errors irremediable, because the Court decided that federal courts lack jurisdiction over those questions.^[6]

These cases, in part, reveal the legacy of the harshest enactment in modern immigration law, the Illegal Immigration Reform and Immigration Responsibility Act of 1996 (“IIRIRA”).  This amendment to the INA expanded immigration detention, stripped judicial review in a variety of settings, and limited relief from removal. Unsurprisingly, the Court is regularly called upon to interpret IIRIRA or decide whether a challenged provision comports with the Constitution.^[8] The Court’s approach to statutory interpretation, therefore, matters to millions of immigrants, their allies, and their advocates.

The IIRIRA cases from the 2021-22 term showcase divergent approaches to textualism, the dominant theory of statutory interpretation in the United States. ^[9] Textualism today generally refers to “new textualism,” a view of the judge’s role in statutory interpretation as limited to discerning the plain meaning of statutory language without resort to extrinsic evidence, like legislative history.^[10] Immigration cases from this term, however, exposed distinctions within textualism. Some featured a textualism that analyzed plain meaning by considering the broader context of disputed statutory language, like neighboring provisions of the same statute.^[11] Justice Antonin Scalia, the figure most closely associated with new textualism, frequently endorsed structural arguments of this sort.^[12] Other immigration cases from this Term featured a more segmented textualism focused on individual words regardless of their effects.^[13] If the “new textualism” focused on discerning the plain meaning of text without resort to extrinsic sources of meaning, this segmented approach might be the “new” new textualism, which eschews structural arguments and value-laden substantive canons. Ultimately, these different interpretive approaches within textualism underscore the many discretionary choices that underlie the search for plain meaning.^[14]

Arteaga-Martinez illustrates a textualism cut off from the constitutional backdrop of legislation. Usually, when the government locks someone up for long periods of time without a criminal conviction, the Due Process Clause requires a bail hearing where the government must prove that the person poses a danger or a flight risk.^[15] Because the Due Process Clause protects “persons” rather than only “citizens,” lengthy detention without bond raises a profound constitutional doubt.^[16] In Arteaga-Martinez, however, the Court did not decide the constitutional question. Instead, it ruled that the INA on its own terms permitted lengthy detention without bond, reserving judgment on whether this scheme comports with due process.^[17] It declined Arteaga-Martinez’s invitation to adopt a plainly constitutional interpretation of the statute providing for a bond hearing while suggesting that such a move would amount to rewriting the statute rather than interpreting it.^[18]

The cases decided on jurisdictional grounds, Aleman Gonzalez and Patel, illustrate what Professors William Eskridge and Victoria Nourse call “statutory populism”^[19] and what Professors Anya Bernstein and Glen Staszewski call “judicial populism.”^[20] They feature a dictionary-driven analysis of discrete words, much like the controversial decision in Bostock v. Clayton County, Georgia last term.^[21] Pursuant to this approach, the Court breaks a sentence into its component words, looks them up in the dictionary, and stitches them back together to arrive at what it deems a clear meaning. Eskridge and Nourse call this a type of “textual gerrymander” known as “cracking-and-packing.”^[22] This textualist approach is “populist” in their view because it “lays claim to democratic legitimacy by invoking a search for the ‘true will’ of the people.”^[23] Bernstein and Staszewski similarly characterize judicial populism as anti-pluralist, seeking the one true answer to interpretive questions.^[24]

If these jurisdictional decisions reveal a general populist trend, the immigration context only heightens the stakes of the justices’ interpretive debates. Judicial populism has an exclusionary impulse, one typified by nativism and oversimplification.^[25] Unsurprisingly, it has devastating implications for noncitizens who, in an increasingly illiberal order, are not considered part of “We the People,” whose will the Court discerns.^[26] But courts minimizing immigrants’ interests and well-being is not a new phenomenon or solely the product of a particular strand of textualism. Rather, this essay argues that these cases are best understood as springing not only from a populist interpretive method, but from background norms that minimize immigrants’ standing in our legal culture.

That legal culture has been shaped by the plenary power doctrine. This nineteenth-century doctrine, announced in the Chinese Exclusion Case, recognized a nearly boundless congressional power to regulate immigration and a corresponding lack of judicial authority to review those judgments. ^[27] In the twentieth century, it shifted from a non-justiciability doctrine to a deferential approach to reviewing rights-based challenges to immigration statutes and executive policies implementing immigration law.^[28]

Although today courts recognize that the doctrine has limits, the doctrine has also enjoyed a renaissance in recent years, most notably in Trump v. Hawaii, where the Court applied only the mildest review to claims that the third version of the Trump Administration’s order suspending the entry of nationals from several majority-Muslim countries violated the Establishment Clause based on ample evidence of the President’s anti-Muslim animus.^[29] Today’s interpretive debates occur against this uncertain constitutional backdrop.

For much of the twentieth century, as theorized by Professor Hiroshi Motomura, federal courts used the constitutional avoidance canon to “offset the disadvantaged position of aliens in constitutional immigration law.” ^[30] Even in purely statutory cases, judges routinely invoked various presumptions channeling substantive values protecting individuals against government overreach and error.^[31] However, these protective interpretive practices lost out in the 2021-22 term’s IIRIRA cases.^[32] The dominant interpretive approach today prizes a segmented, “piecemeal”^[33] analysis of words and phrases, which seldom fails to provide a definitive answer.  This essay analyzes the three decisions, their implications for immigrants’ rights, and where we might go from here.

I.    Johnson v. Arteaga-Martinez: Clear Text Curbing Liberty

The Supreme Court has decided a string of immigration detention cases over the last twenty-plus years that pose some version of the question: What rules must the government follow when locking up immigrants? On the one hand, “Due process calls for an individual determination before someone is locked away.”^[35] But on the other, “Congress may make rules as to aliens that would be unacceptable if applied to citizens.”^[36] How should these rules be harmonized?  Periodically, immigrants ask the federal courts to decide. But the stakes of explicit constitutional adjudication are high, and skilled advocates routinely ask courts to avoid deciding the constitutional question by adopting a rights-protective interpretation of the INA. They push for use of the “constitutional avoidance” canon.^[37] This canon of statutory interpretation holds that, where a statute is susceptible to more than one interpretation, and one interpretation would violate the Constitution or raise a serious constitutional question, the court should avoid that interpretation and instead adopt an interpretation that respects the relevant constitutional interests.^[38] One formulation requires only that this alternative interpretation be “fairly possible,”^[39] where another requires that the alternative interpretation be plausible or even equally so. In recent years, the Supreme Court has limited the scope of that canon in immigration detention cases,^[40] adopting the plausibility formulation, and Johnson v. Arteaga reflects that limitation.

Johnson v. Arteaga-Martinez presented the latest challenge to the INA’s detention regime for noncitizens who have been ordered removed. Antonio Arteaga-Martinez, a Mexican national, had been deported after entering without inspection, but he reentered the United States sometime after 2012.^[41] In 2018, Immigration and Customs Enforcement (“ICE”) arrested him, reinstated his order of removal, and detained him.^[42] Arteaga-Martinez sought withholding of removal, a form of humanitarian relief that can take months or years to adjudicate, based on his fear of persecution or torture in Mexico.^[43] This means that when the government elects to detain someone pending resolution of a claim for withholding of removal, that decision can trigger prolonged detention. After four months of detention, Arteaga-Martinez filed a petition for a writ of habeas corpus.^[44]

The district court granted relief based on circuit precedent recognizing a right to a bond hearing after six months of detention.^[45] Based on this same circuit precedent, the Third Circuit affirmed.^[46] The government petitioned for a writ of certiorari, which the Court granted.^[47]

The INA authorizes the detention of noncitizens who have been ordered removed.^[48] It requires the government to detain a noncitizen ordered removed for 90 days after the order becomes administratively final, known as the “removal period.”^[49] This same provision, 8 U.S.C. § 1231(a)(6), authorizes, but does not require, detention beyond the removal period.^[50] It states that a noncitizen who is inadmissible under the INA, removable under certain crime-based provisions, or who poses a “risk to the community or [is] unlikely to comply with the order of removal, may be detained beyond the removal period . . . .”^[51] (emphasis added).

In the 2001 case of Zadvydas v. Davis, the Court read the word “may” in this very same provision to confer discretion on the immigration bureaucracy and to create ambiguity about the length of the detention authorized.^[52] In Zadvydas, the petitioner was stateless, his removal impracticable.^[53] He faced permanent detention. The Court determined that Congress did not authorize permanent detention through § 1231(a)(6); instead, the Court read the INA to authorize detention only for a presumptively reasonable period of six months.

Justice Stephen Breyer, writing for the 5-4 majority, concluded that civil detention beyond a presumptively reasonable period raised a constitutional question, for U.S. constitutional law has long required some “special justification” for continued deprivation of physical liberty.^[55] Justice Breyer’s analysis drew on a rich due process jurisprudence that begins with the text of the Fifth Amendment’s Due Process Clause. That clause states, “No person shall be . . . deprived of life, liberty, or property, without due process of law . . . .”^[56] Because the Constitution elsewhere specifies “citizens,” the reference to “person” here has typically led the Court to recognize the clause’s applicability to all persons within U.S. territory, regardless of immigration status.^[57] Justice Breyer then observed that freedom from imprisonment, whether labeled incarceration, custody, or detention, “lies at the heart of the liberty that Clause protects.”^[58] Even though Congress generally possesses the power to detain noncitizens for short periods pursuant to removal or exclusion, the Due Process Clause still imposes limits on the procedures used to implement that power.^[59]  With respect to the petitioner in Zadvydas, the government simply lacked the authority to detain him beyond six months when his removal was not reasonably foreseeable.

Arteaga-Martinez argued that his removal, too, was not reasonably foreseeable.^[60] It could be years until he received a decision on his withholding of removal claim, and, if decided favorably, he would not be removable at all.^[61] Under Zadvydas, he argued for outright release, or at the very least, a neutral bond hearing where the government would have to demonstrate a special justification for continuing to detain him.^[62] He argued that the existing ICE custody review process fell short because it assigned responsibility to ICE to review its own custody decisions after the removal period had expired.^[63] Read against a robust constitutional backdrop, he argued, the INA requires more—specifically, a bond hearing before an immigration judge.^[64] Noncitizen respondents in the companion case, Aleman Gonzalez, emphasized the longstanding requirement of bond hearings for prolonged civil detention.^[65] They argued that the Court could not deem the absence of the specific phrase “bond hearing” from Section 1231 dispositive of whether the statute implicitly required one.

The Court in Arteaga-Martinez did not engage this reasoning. Instead of starting with Zadvydas, the Court focused on the text of Section 1231. It acknowledged that the word “may” conferred discretion, as in Zadvydas.^[66] But the similarities ended there. Zadvydas supported a durational limit on the government’s detention power, but it did not require adopting an implied bond hearing requirement, the Court concluded.^[67] Why? It determined that such a reading lacked plausibility, regardless of potential constitutional infirmities of the alternative.^[68] Justice Sonia Sotomayor, writing for the Court, expressly noted that the Court reserved judgment on whether the Due Process Clause, rather than the INA, required a bond hearing given the substantial deprivation of liberty at issue.^[69]

The Court’s approach to statutory interpretation in recent years reflects an increasing emphasis on the plain meaning of words in a statute, with less consideration of context and mainstream constitutional norms.^[70] Where a Court might have once seized on a textual ambiguity to supply protections from harsh incursions on liberty through statutory interpretation, it now finds no ambiguity at all.

What was this old, rights-protective method, and where has it gone? Thirty years ago, Professor Hiroshi Motomura theorized that federal courts had a practice of invoking “phantom constitutional norms” to reach immigrant-protective statutory interpretations.^[71] When faced with a constitutional challenge to an immigration statute, courts would decline to openly decide the constitutional question. This allowed them to avoid the impact of the plenary power doctrine, which in a previous era made immigrants poor rights-holders and would typically compel a government victory.^[72] By instead deeming the statutory language ambiguous and then referencing an imagined constitutional norm that protected immigrants’ interests, the court could use the canon of constitutional avoidance to “undermine the plenary power doctrine.”^[73] Over time, however, immigrants’ status as rights-holders grew, turning phantom norms into real ones, at least in some settings. But Motomura also noted the distortions that resulted from this practice, one of which was the suppression of “dialogue en route to new constitutional doctrine.”^[74]

After Zadvydas v. Davis, this practice of using the constitutional avoidance canon to safeguard immigrants’ real or aspirational constitutional interests has waned. In recent years, the Court has expressly limited the constitutional avoidance canon’s applicability. In the 2018 case Jennings v. Rodriguez, the Court rejected immigrant detainees’ claims that the INA required periodic bond hearings for detention pending completion of removal proceedings.^[75] The dissenting justices viewed the lack of bail and bail hearings as constitutionally suspect, but the majority did not engage that question. Instead, it reviewed the relevant statutory language, declined to find it ambiguous as to a bond hearing requirement, and declined to apply the avoidance canon to adopt a rights-protective interpretation.^[76] Crucially, the Court deemed the plaintiffs’ (and the dissent’s) interpretation “implausible.”^[77] Under a textualist approach, the Court held, “the canon of constitutional avoidance ‘comes into play only … after the application of ordinary textual analysis’” when a court has exhausted the traditional tools of statutory interpretation.  In other words, constitutional avoidance cannot be used to adopt a friendly interpretation unless the statute is ambiguous and the friendly interpretation is first found plausible. Accordingly, in Jennings, the Court declined to read the INA to require bond hearings or to set an implicit time limitation on detention under the challenged provisions.^[79]

The role of “plausibility” as gatekeeper has only been amplified since Jennings, and each new case reveals more about its significance and implications. The rise of plausibility could simply mean that the Court sharply divides statutory interpretation from constitutional analysis, and that threshold questions of ambiguity and plausibility do not depend on the constitutional stakes. On this view, a court would be no more likely to find language ambiguous or open to more than one plausible interpretation simply because the alternative reading is blatantly unconstitutional. This stylized conception eschews the “sliding scales” likely shaping judgments of ambiguity and plausibility.^[80]

But the Court’s reluctance to use statutory interpretation as a vehicle for vindicating constitutional interests might, alternatively, reveal a poor view of the constitutional interests at stake—a de facto take on the merits. In Trump v. Hawaii, for example, the Court famously considered the scope of 8 U.S.C. § 1182(f), which authorizes the President to suspend the entry of any noncitizens or classes of noncitizens under certain circumstances.^[81] Challenging the third version of the exclusion order, the plaintiffs asserted statutory claims and an Establishment Clause claim.^[82] One option might have been to infer limits on the statutory power to spare the Court having to decide the Establishment Clause claim. But no justice deemed the challenged executive order to violate any textual limits contained in the statute.^[83] The Court similarly did not regard the statute’s place in an intricate scheme of inadmissibility and removability grounds to cabin the broad grant of exclusion power. The Court proceeded to decide the constitutional question. Ultimately, the Court found that the citizen plaintiffs were entitled only to a highly deferential review of the President’s order. Under that framework, the Court discerned no violation of the Establishment Clause, despite the President’s anti-Muslim animus.^[84] Hawaii might suggest that a refusal to cabin statutory power merely reveals skepticism about the claimed underlying constitutional right.

All of this takes us back to Arteaga-Martinez. The decision was not close, 8-1, with only Justice Breyer dissenting. Had the Court deemed the detention statute at issue ambiguous as to the procedural requirements for lengthy detention, the Court would have proceeded to consider whether the no-bond-hearings interpretation would present a constitutional problem. If enough justices thought so, at that point, the constitutional avoidance canon would counsel for adopting an alternative interpretation—one with bond hearings. But Justice Sotomayor concluded that Jennings foreclosed this result.

The implications of Arteaga-Martinez are numerous. The rise of a textualism that establishes a high bar for ambiguity and plausibility, regardless of the constitutional stakes, obviously makes it more difficult to adopt interpretations sensitive to constitutional interests. It reflects an attitude that “justif[ies] power irrespective of its effects.”^[85] It channels litigation toward constitutional claims in an uncertain or potentially hostile doctrinal environment.

A recent explicit constitutional decision raises questions as to the scope of protection for detained immigrants. After Zadvydas, the Court considered a due process challenge to a mandatory detention provision, 8 U.S.C. § 1226, which failed to provide bond hearings to noncitizens convicted of certain crimes, whom the government detained pending completion of removal proceedings. In Demore v. Kim, the Court held that Congress had substantial leeway to detain “deportable” noncitizens for a “limited” period pending proceedings.^[86] There, the Court emphasized that detention was both “short” and finite: Case completion took “about five months” and would end when the immigration judge issued a final order of removal (or relief). This empirical claim was false, and the Solicitor General submitted a letter to the Court admitting as much some years later.^[87] It turns out that the average length of time to complete a case in which the noncitizen appeals was closer to thirteen months.^[88] In the face of protracted proceedings and lengthy detention beyond what the Court sanctioned in Demore, immigrants similarly situated to Arteaga-Martinez will likely argue that Demore does not diminish their constitutional claims.

Unfortunately, constitutional rights are an increasingly unstable source of protection today. Although immigrants rights’ jurisprudence has long required courts to engage in a nuanced analysis of a range of factors, including the noncitizen’s status, length of residence, and ties to the United States, ^[89] the Court has upended this body of law in recent years. Given the current judicial mood and populist approach to interpretive questions, the open adjudication of noncitizens’ constitutional rights invites a restrictive reading of due process and potentially massive doctrinal upheaval.

A recent case serves as a cautionary tale. In a 2020 case, United States Agency for International Development v. Alliance for Open Society International, Inc. (“AOSI II”), the Court announced that it was “long settled as a matter of American constitutional law that foreign citizens outside U.S. territory do not possess rights under the U.S. Constitution.”^[90] This grossly mischaracterizes the state of the law.^[91] As explained above, federal courts have long assessed the strength of a noncitizen’s rights under the Constitution based on a range of factors.^[92] In recharacterizing decades of precedent, the Court sowed doubt about whether the principle was in fact a “holding” or merely “dicta.”^[93] When the Court forgoes a nuanced, longstanding, multi-factor analysis for complete erasure of the underlying right, the true terms of the “dogmatic textualist” deal are apparent.^[94] It offers a cleaner, minimalist approach, but it comes at the cost of potentially misrepresenting and upending settled law—in addition to massively eroding liberty for those on the losing end.

Against this backdrop of constitutional interpretation-by-fiat, litigation over immigrants’ rights today feels perilous. After Jennings and Arteaga-Martinez, however, the next phase of immigration detention litigation will necessarily present the Court with the question of whether the challenged INA provisions violate the Due Process Clause. With Justices Clarence Thomas and Neil Gorsuch questioning the application of the Due Process Clause to removable immigrants,^[95] and recent decisions like AOSI II, the danger of a radically restrictive constitutional decision looms. But with rights-protective statutory interpretation no longer viable, plaintiffs seeking release from allegedly illegal Executive detention will likely pursue constitutional claims, nevertheless.

II.  Garland v. Aleman Gonzalez: The Perils of “Piecemeal” Textualism^[96]

Immigrants’ rights advocates have used class actions in recent years to challenge the INA’s detention regime. Detained noncitizens typically lack the resources or capacity to retain individual counsel, making group litigation both appealing and necessary. But in Garland v. Aleman Gonzalez, the Court eliminated the lower federal courts’ power to award injunctive relief in class actions challenging specific INA provisions.

This case involved a class action brought by noncitizens similarly situated to Arteaga-Martinez, who challenged their prolonged detention without bond. As in Arteaga-Martinez, the Court rejected the plaintiffs’ statutory arguments, finding no implied bond hearing requirement. More significantly, however, the Court held that the INA deprived the lower federal courts of jurisdiction to issue injunctive relief in class actions based on a provision of the INA codified at 8 U.S.C. § 1252(f).  The government had not raised this issue below, but the Court ordered briefing on the question prior to oral argument.

Section 1252(f) reads:

Regardless of the nature of the action or claim or the identity of the party or parties bringing the action, no court (other than the Supreme Court) shall have jurisdiction or authority to enjoin or restrain the operation of [select INA provisions],…other than with respect to the application of such provisions to an individual alien against whom proceedings under such party have been initiated.^[97]

The parties agreed that the core text to interpret was “to enjoin or restrain the operation of...‘provisions of the INA.’”^[98] The parties also agreed on the relevance of the savings clause, “other than with respect to the application of such provisions to an individual alien . . . .”^[99] The disagreement emerged in how to analyze the core text.

Justice Samuel Alito “cracked” the provision into pieces,^[100] invoking dictionary definitions of each word, “enjoin,” “restrain,” and “operation.” He began by noting that “enjoin” means to “‘require,’ ‘command,’ or ‘positively direct’ an action or to ‘require a person to perform . . . or to abstain and desist from, some act,’” according to Black’s Law Dictionary and Webster’s Third New International Dictionary.^[101] “Restrain,” in contrast, means to “‘check, hold back, or prevent (a person or thing) from a course of action.’”^[102] Finally, “operation of” means the “functioning of or working of (that thing),” according to the Random House Dictionary of the English Language and Webster’s.^[103] Seizing on the terms “work” and “function,” Justice Alito then determined that, “[t]he way in which laws ordinarily ‘work’ or ‘function’ is through the actions of officials or other persons who implement them.”^[104] Arriving at the word “implementation,” he then “packed” the pieces back together to conclude that the core provision forbids injunctive relief mandating or prohibiting any functioning or implementation of the law.^[105] After finding that the core text favored the government, Justice Alito then interpreted the savings clause narrowly, requiring that a case be brought by an “individual alien” rather than a class.^[106]

Justice Sotomayor, in dissent, assailed this “piecemeal” approach. She analyzed the words “enjoin or restrain the operation of” together and in the broader context of the jurisdiction-stripping provision in which they appeared and the INA as a whole.^[107] First, she noted that an injunction compelling “the Executive Branch to comply with the specified provisions does not ‘enjoin or restrain’ the ‘operation’ of those provisions.”^[108] She agreed that “operation” meant “working” or “functioning,” but she did not accept an equivalence to “implementation.”^[109] Whereas other simultaneously enacted subsections of Section 1252 expressly limit jurisdiction over challenges to “implementation” of a statutory provision, this one doesn’t.^[110] She drew a negative inference from Congress’s choice not to use “implementation” here. She also drew on Congress’s use of “enjoin” throughout federal immigration law, codified in Title 8, and particularly in neighboring provisions, to conclude that “enjoin” in the disputed text means “a prohibition on the operation of a statute.”^[111] Its pairing with the word “restrain,” she argued, “cements a prohibitory reading of ‘enjoin.’”^[112] Finally, Justice Sotomayor invoked a clear statement rule requiring “the clearest command” before “displacing the courts’ traditional equitable authority.”^[113] In other words, any remaining ambiguity should be resolved in favor of preserving the lower courts’ equitable jurisdiction.

Unsurprisingly, Justice Sotomayor also disagreed with the majority’s reading of the savings clause. Noting that a class is a collection of individuals, she drew on “contextual and historical evidence” showing that the 1996 Congress would not have used the word “individual” to preclude class-wide injunctive relief.^[114] Furthermore, Congress had elsewhere in Section 1252 expressly divested jurisdiction to “certify a class under Rule 23 of the Federal Rules of Civil Procedure.”^[115] Had it meant to do so here, it very well could have expressly barred actions under Rule 23.^[116]

Justice Alito rejected the crux of the dissent’s argument, that “operation” refers to the statute’s operation when “properly interpreted,” with an appeal to “ordinary meaning.”^[117] He arrived at an understanding of the common usage of “operation” by casting a wide net to catch uses of the word in unrelated cases. Specifically, he noted that courts commonly refer to the “unlawful” or “improper” operation of cars, trucks, railroads, water utilities, drainage ditches, radios, and other forms of transportation or technology.^[118] For that reason, he argued, the common meaning of operation does not imply “proper” operation.^[119]

Ultimately, the majority and the dissent differed not on whether to privilege text over purpose, for example, but what text mattered, and how to go about discerning the meaning of that text. Justice Alito took an approach that emphasized each individual word’s definition in the dictionary, regardless of context or its implications. Justice Sotomayor’s textualism also remained faithful to textualist sources. But she offered an informative, contextually grounded analysis of the disputed provision and its role in Section 1252. She found the relevant text to encompass Section 1252 as a whole, as well as Title 8 of the U.S. Code, where the INA is codified. Whereas Justice Sotomayor conceived of the possibility of some “remaining ambiguity” at the end of her analysis, Justice Alito adopted a more absolutist stance consistent with judicial populism. The reality that thousands of detained immigrants would now lack the ability to obtain release from potentially illegal custody—or be forced to file individual actions, flooding the federal courts—did not factor in at all as the majority moved through the chain of definitions from “operation” to “functioning” to “implementation.” In contrast, the dissent appreciated the stakes.

III.        Patel v. Garland: Clear Text for Unchecked Agency Power

The Court also closed the door on judicial review of agency factual determinations underlying judgments regarding the granting of relief under specific INA provisions in Patel v. Garland. In that case, Pankajkumar Patel and his wife Jyotsnaben had applied for permanent residence (a “green card”) through a process known as adjustment of status.^[120] Patel contended that, when he applied for a driver’s license in Georgia, he mistakenly checked a box indicating that he was a U.S. citizen.^[121] Under Georgia law, however, one need not be a citizen to obtain a driver’s license: being a permanent resident or even having applied for permanent residence is sufficient for eligibility.^[122] Accordingly, the checking of the box did not secure him any benefits for which he was not already eligible.

In any event, the Georgia DMV contacted DHS about Patel having indicated he was a citizen, and DHS placed him in removal proceedings as someone who “falsely represents . . . himself or herself to be a citizen of the United States for any purpose or benefit under” state or federal law.^[123] Before an immigration judge, Patel argued that he had mistakenly checked the box. To support that argument, he noted that he would have been eligible for a driver’s license even had he not checked the box; however, the immigration judge found him “not credible” and ordered him removed.^[124] The immigration judge insisted that Patel had misrepresented his citizenship status intentionally to procure a benefit. Patel petitioned for review, arguing that no reasonable factfinder could find him not credible, based on the standard established in 8 U.S.C. § 1252(b)(4)(B).^[125]

In federal court, on a petition for review from the BIA, things took a surprising turn. Although no party raised the issue, the Eleventh Circuit found that it lacked jurisdiction to consider Patel’s claim under Section1252(a)(2)(B)(i), deepening a circuit split on the issue. Section 1252(a)(2)(B)(i) states:

(B) Denials of discretionary relief

Notwithstanding any other provision of law. . ., and regardless of whether the judgment, decision, or action is made in removal proceedings, no court shall have jurisdiction to review—

• any judgment regarding the granting of relief under section . . .1255 of this title . . . .^[126]

A separate provision preserves judicial review for constitutional claims or questions of law.^[127] Patel petitioned for certiorari, and upon granting, the Court asked attorney Taylor A.R. Meehan to submit an amicus brief arguing that Section 1252(a)(2)(B)(i) stripped the federal courts of jurisdiction over all factual claims involved in a decision to grant relief under the specified provisions.^[128]

The argument centered on the role of discretion in the challenged provision. Does “judgment” refer to any decision at all, or only to those requiring an exercise of discretion? Federal agencies make all kinds of factual determinations while adjudicating immigration benefits—some are discretionary, and some are not. For example, whether a noncitizen has “good moral character” is often considered a discretionary judgment for an immigration judge to make, and courts have long held that such determinations are unreviewable (although Patel disagreed). But relief in some cases can also depend on nondiscretionary facts, such as whether the noncitizen lived in the United States long enough to qualify for relief. And many courts have long held that those are reviewable.

The Court considered each of the words in Section 1252(a)(2)(B)(i), focusing on the word “judgment,” but also the words “any” and “regarding,” and found each to have a maximalist quality that supported a broad interpretation of the provision.^[129] Relying on Webster’s and the Oxford English Dictionary, Justice Amy Coney Barrett, writing for the majority, noted that “judgment” means an “authoritative decision.”^[130] She then noted that the word “any” worked to expand the realm of covered decisions, in keeping with a dictionary definition of “any” meaning “one or some indiscriminately of whatever kind.” Accordingly, judgments of “whatever kind” were covered. The majority found “regarding” to have a similar “broadening effect.” It then read the provision preserving judicial review for questions of law and constitutional questions as reinforcing this conclusion. With all the classic populist rhetoric of staying in their “lane” and not letting policy “trump the best interpretation of the statutory text,” the majority similarly rejected the relevance of any presumption in favor of judicial review. The Court exerted its power “without justifying its effects.”^[131]

In a stinging dissent, Justice Gorsuch advanced an equally textualist interpretation but began with the overall statutory scheme. Requests for adjustment of status involve two steps, he noted.^[132] First, the government determines whether the individual is statutorily eligible. Do they meet the criteria the INA establishes? If so, the Attorney General decides whether to grant adjustment “in his discretion.”^[133] That second decision is unreviewable under the challenged provision, all parties agreed. But Patel had argued for review of the first—the determination of statutory eligibility. The dissent then explained that “judgment regarding the granting of relief” clearly referred to this second step rather than both—the actual decision, at the second step, to grant relief.^[134] In contrast to the majority’s method of “cracking” open the relevant text into pieces and then reassembling,^[135] the dissent made sense of the disputed sentence based on knowledge of the statutory scheme. But this more informed, contextually grounded textualism lost.

IV. The Future of Immigrants’ Rights

These three cases all involved challenges to provisions of IIRIRA, but one reflected the changing relationship of constitutional norms to statutory interpretation, and two reflected a populist interpretive method. Stepping back, the three cases all involve a recalibrated relationship of statutory text to context, especially choice of relevant context.^[136] Arteaga-Martinez confirms the path charted in Jennings, divorcing plausibility analysis from the constitutional stakes, and declining to ask whether Congress would plausibly enact what might be a glaringly unconstitutional law.  ^[138] shields itself from relevant information, and creates the impression of finding the answers to interpretive questions, e.g., in dictionaries, rather than participating in the making of meaning through a series of (unavoidable) discretionary choices.^[139]

Put differently, the jurisdictional cases reveal a judicial minimalism in the service of maximalism. In a bid to diminish the role of the lower federal courts, Aleman Gonzalez has set them up for a deluge of individual actions challenging illegal Executive Branch implementation of the INA, assuming immigrants successfully navigate pro se litigation. Similarly, Patel cuts the federal courts off from reviewing agency factual errors, leaving untold numbers of immigrants with no shield against unjustified deportation. In this way, these cases have furthered the erasure of noncitizens from the legal system—the right to have their claims heard or the right to an effective remedy.

With statutory or judicial populism ascendant, immigrants and their advocates face an uphill effort to protect their interests and their well-being. Immigrants are not secure rightsholders in today’s legal environment, yet they will be pressed to assert constitutional claims. In previous decades, advocates worried about rights without remedies. Now, the rights themselves are in question, despite decades of progress, and without any compensating measures in the realm of statutory interpretation.

Advocates are not helpless, however. It remains to be seen if the Court will interpret Section 1252(f)(1) to bar claims for injunctive relief brought by multiple named plaintiffs, a question not resolved by Aleman Gonzalez. That decision also appears to permit class-wide declaratory relief. As ever, advocates might seek to leverage some justices’ anti-regulatory attitudes to limit government power in immigration regulation.^[140] Scholars have argued that justices alarmed by the prospect of unchecked agency power, like Justice Gorsuch in Patel, might serve as immigration allies at times.^[141] But skepticism toward regulation does not always translate into skepticism toward the regulation of noncitizens or a belief that they have rights under the Constitution.

Crucially, people can effect change outside the courts altogether. For example, interest group pressure on the Department of Homeland Security could induce the agency to issue protective regulations even if no legal authority requires it to.^[142] As commentators have noted, nothing precludes DHS from granting detained immigrants bond hearings after six months of detention. Similarly, a leadership team sensitive to immigrants’ well-being might set an agenda for the agency that rejects maximalism and views immigrants, to the extent feasible, as “clients” rather than “targets.” Finally, but perhaps fantastically, the people could pressure Congress to repeal the most draconian provisions of the IIRIRA. Calls for social change through the political process generally have an aspirational quality, premised on a belief in the power of “We the People” and the promise of collective action for a just society. But in these dark times, such calls can feel naïve and the effort Herculean. At the same time, if legal meaning is dynamic and contested , an adverse Supreme Court ruling is not the final word. People are dedicated to improving their communities and the broader world, from organizers and health-care providers to public-interest lawyers, teachers, and community leaders. We might hope that, in the words of composer Frederic Rzewski, “the people united will never be defeated.”

*Associate Professor, University of Alabama School of Law. Thanks to Lucie Dierikx for helpful research assistance.

[1] Johnson v. Arteaga-Martinez, 142 S. Ct. 1827 (2022).

[2] Garland v. Aleman Gonzalez, 142 S. Ct. 2057 (2022).

[3] Patel v. Garland, 142 S. Ct. 1614 (2022).

[4] See Arteaga-Martinez, 142 S. Ct. at 1833.

[5] See Aleman Gonzales, 142 S. Ct. at 2063.

[6] See Patel, 142 S. Ct at 1627.

[7] Illegal Immigration Reform and Immigration Responsibility Act of 1996, Pub. L. No. 104-208, 110 STAT. 3009 (codified as amended in scattered sections of 8 U.S.C.).

[8] See, e.g., Dep’t of Homeland Sec. v. Thuraissigiam, 140 S. Ct. 1959 (2020).

[9] See Tara Leigh Grove, Which Textualism?, 134 Harv. L. Rev. 265, 268 (2020) (noting “competing strands of textualism”).

[10] See William N. Eskridge, The New Textualism, 37 UCLA L. Rev. 621, 623 (1990).

[11] See Aleman Gonzalez, 142 S. Ct. at 2071 (Sotomayor, J., dissenting).

[12] See Eskridge, supra note 10, at 661–62 (describing Justice Scalia’s endorsement of structural textualist arguments, such as considering how the same language is used elsewhere in the statute and how “possible meanings fit within the statute as a whole”).

[13] See generally Aleman Gonzalez, 142 S. Ct. at 2063.

[14] See, e.g., Anya Bernstein, Before Interpretation, 84 U. Chi. L. Rev. 567, 644–647 (2017).

[15] United States v. Salerno, 481 US. 739, 755 (1987) (affirming the constitutionality of the Bail Reform Act of 1984, which authorized “detention prior to trial of arrestees charged with serious felonies who are found after an adversary hearing to pose a threat to the safety of individuals or to the community which no condition of release can dispel”); Foucha v. Louisiana, 504 U.S. 71, 80 (1992) (noting that detention “for any purpose constitutes a significant deprivation of liberty that requires due process protection”).

[16] Zadvydas v. Davis, 533 U.S. 678, 690 (2001).

[17] Johnson v. Arteaga-Martinez, 142 S. Ct. 1827, 1834 (2022).

[18] Id. at 1833.

[19] William N. Eskridge, Jr. & Victoria F. Nourse, Textual Gerrymandering: The Eclipse of Republican Government in an Era of Statutory Populism, 96 N.Y.U. L. Rev. 1718, 1722, 1738 (2021).

[20] Anya Bernstein & Glen Staszewski, Judicial Populism, 106 Minn. L. Rev. 283, 284 (2021).

[21] See Bostock v. Clayton Cnty., Ga., 140 S. Ct. 1731 (2021).

[22] Eskridge & Nourse, supra note 19, at 1770 (“[I]nterpreters should ‘not simply split statutory phrases into their component words, look up each in a dictionary, and then mechanically put them together again,’” which they dub “a cracking-and-packing gerrymander.”).

[23] Id. at 1722.

[24] Bernstein & Staszewski, supra note 20, at 289. They further characterize judicial populism as “anti-institutionalist” (disparaging deliberation by expert bodies), and “Manichean” (pitting morally pure or virtuous ordinary, marginalized people against corrupt elites or outsiders).

[25] Id. at 284.

[26] See Eskridge & Nourse, supra note 19, at 1793–94; see Geoffrey Heeren, Persons Who Are Not the People: The Changing Rights of Immigrants in the United States, 44 Colum. Hum. Rts. L. Rev. 367 (2013).

[27] Ping v. United States, 130 U.S. 581, 609–11 (1889).

[28] Kleindienst v. Mandel, 408 U.S. 753 (1972); Shalini Bhargava Ray, Plenary Power and Animus in Immigration Law, 80 Ohio St. L. J. 13, 72, n. 19 (2019).

[29] Trump v. Hawaii, 138 S. Ct. 2392, 2415­–23 (2018).

[30] Hiroshi Motomura, Immigration Law After a Century of Plenary Power: Phantom Constitutional Norms and Statutory Interpretation, 100 Yale L. J. 545, 568 (1990).

[31] See Garland v. Aleman Gonzalez, 142 S. Ct. 2057, 2071 (2022) (discussing requirement of clear statement to divest court of equitable authority to issue injunctive relief); William N. Eskridge & Philip P. Frickey, Quasi-Constitutional Law: Clear Statement Rules as Constitutional Lawmaking, 45 Vand. L. Rev. 593, 601 (1992) (discussing requirement of clear statement to overcome presumption of judicial review).

[32] These protective presumptions for individual rights lost out in the very same term that the Court doubled down on presumptions regarding structural constitutional norms. See W. Va. v. Environmental Protection Agency, 142 S. Ct. 2587, 2609–16 (2022) (invoking the “major questions doctrine” to hold that Congress’s power to delegate major regulatory questions to agencies, such as the EPA’s power to promulgate climate change rules under the Clean Air Act, requires a clear statement granting authority). For a discussion of this longstanding asymmetry between presumptions in favor of individual rights versus structural constitutional norms, see Eskridge & Frickey, Quasi-Constitutional Law, supra note 31, at 630–31.

[33] Aleman Gonzalez, 142 S. Ct. at 2068 (Sotomayor, J., dissenting).

[34] Bernstein & Staszewski, supra note 20, at 305.

[35] Demore v. Kim, 538 U.S. 510, 551 (2003) (Souter, J., dissenting).

[36] Id. at 522.

[37] Ashwander v. Tenn. Valley Auth., 297 U.S. 288, 348 (1936).

[38] Id.

[39] Jennings v. Rodriguez, 138 S. Ct. 830, 869 (2018) (Breyer, J., dissenting).

[40] See id. at 842 (2018) (majority opinion).

[41] Johnson v. Arteaga-Martinez, 142 S. Ct. 1827, 1828 (2022).

[42] Id.at 1831.

[43] Id.

[44] Id.

[45] Id.

[46] Id.

[47] Id.

[48] 8 U.S.C. § 1231.

[49] Id.

[50] Id.

[51] Id.

[52] Zadvydas v. Davis, 533 U.S. 678, 683 (2001).

[53] Id. at 685.

[54] Although commentators sometimes question where “six months” came from, it appears plausible that it comes from development of the writ of habeas corpus. At least one habeas corpus scholar has noted that habeas corpus draft legislation in England, laying the foundation for the Habeas Corpus Act of 1679, limited imprisonment to six months. Amanda Tyler, Habeas Corpus in Wartime 23 (Oxford Univ. Press 2017). Subsequently, the Massachusetts town of Milton indicated that a suspension of habeas in a time of war should not exceed “six months,” which is “fully sufficient for any Legislature to ascertain the precise crime, and to procure the evidence against any Individual, in order to bring him for Trial.” Id. at 116.

[55] Zadvydas, 533 U.S. at 690.

[56] U.S. Const. amend. V.

[57] See Wong Wing v. United States, 163 U.S. 228, 238 (1896).

[58] Zadvydas, 533 U.S. at 690.

[59] Demore v. Kim, 538 U.S. 510, 523-24 (2003)

[60] Brief for Respondent at 19, Johnson v. Arteaga-Martinez, 142 S. Ct. 1827 (2022), (No. 19-896), 2021 WL 5513651, at *2.

[61] Id. at *3.

[62] Id. at *17.

[63] Id. at *39–40.

[64] Id. at *9–10.

[65] Garland v. Aleman Gonzalez, 142 S. Ct. 2057, 2062 (2022).

[66] Arteaga-Martinez, 142 S. Ct. at 1832 (2022).

[67] Id. at 1832-33.

[68] Id. at 1833.

[69] Id. at 1836.

[70] See Motomura, supra note 30, at 549 (“[C]onstitutional norms provide the background context that informs our interpretation of statutes and other subconstitutional texts.”).

[71] Id.

[72] Id. at 564–65.

[73] Id. at 549.

[74] Id. at 612.

[75] Jennings v. Rodriguez, 138 S. Ct. 830, 842–44 (2018).

[76] Id. at 859 (Breyer, J., dissenting).

[77] Id. at 849 (majority opinion).

[78] Id. at 842 (citing Clark v. Martinez, 543 U.S. 371, 385 (2005)).

[79] Id. at 846.

[80] Neal Katyal & Thomas P. Schmidt, Active Avoidance: The Modern Supreme Court and Legal Change, 128 Harv. L. Rev. 2109, 2159 (2015).

[81] 8 U.S.C. § 1182(f); Trump v. Hawaii, 138 S. Ct. 2392, 2410 (2018).

[82] Trump v. Hawaii, 138 S. Ct. at 2416.

[83] Id. at 2410.

[84] For discussion of possible frameworks for analyzing presidential animus in an area where courts typically defer under the plenary power doctrine, see Bhargava Ray, supra note 28, at 61–71.

[85] Bernstein & Staszewski, supra note 20, at 305.

[86] See Kim, 538 U.S. 526 (2003).

[87] Letter from Jean C. King, Gen. Couns. at U.S. Dep’t of Just. Off. of the Solic. Gen., to Hon. Scott S. Harris, U.S. Sup. Ct. Clerk (Aug. 26, 2016) (available at: https://trac.syr.edu/immigration/reports/580/include/01-1491%20-%20Demore%20Letter%20-%20Signed%20Complete.pdf ).

[88] Letter from Ahilan T. Arulanantham, Legal Dir. of the Am. Civ. Liberties Union, to Hon. Scott S. Harris, U.S. Sup. Ct. Clerk (Oct. 17, 2016) (available at: https://www.supremecourt.gov/DocketPDF/15/15-1204/27054/20180108160849570_15-1204%20Letter%20of%20respondents.PDF.).

[89] Karen Nelson Moore, Aliens and the Constitution, 88 N.Y.U. L. Rev. 801, 804 (2013).

[90] U.S. Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc., 140 S. Ct. 2082, 2086 (2020).

[91] See Ahilan Arulanantham & Adam Cox, Immigration Maximalism at the Supreme Court, Just Sec., (Aug. 11, 2020), https://www.justsecurity.org/71939/immigration-maximalism-at-the-supreme-court/.

(“The Court characterized this as a bedrock principle of American constitutional law. It is anything but.”). For a discussion of the state of the law before AOSI II, see Fatma E. Marouf, Extraterritorial Rights in Border Enforcement, 77 Wash. & Lee L. Rev. 751 (2020).

[92] See Shalini Bhargava Ray, The Contested “Bright Line” of Territorial Presence, 56 U.G.A. L. Rev. 1511, 1524-26 ( 2022) (critiquing AOSI II on these grounds).

[93] See U.S. Agency for Int’l Dev. v. All. for Open Soc’y Int’l, Inc. (AOSI II), 140 S.Ct. 2082, 2099 (Breyer, J., dissenting) (“Even taken on its own terms, the majority’s blanket assertion about the extraterritorial reach of our Constitution does not reflect the current state of the law.”).

[94] See Eskridge & Nourse, supra note 19, at 1722.

[95] See. Arteaga-Martinez, 142 S. Ct. at 1835–36 (2022) (Thomas, J., concurring).

[96] Aleman Gonzalez, 142 S. Ct. at 2068 (2022) (Sotomayor, J., dissenting in part) (stating that the Court used “a purportedly textualist opinion that, in truth, elevates piecemeal dictionary definitions and policy concerns over plain meaning and context”). See also Shalini Bhargava Ray, SCOTUSBlog, The demise of rights-protective statutory interpretation for detained immigrants and the rise of ‘piecemeal’ textualism (June 14, 2022), https://www.scotusblog.com/2022/06/the-demise-of-rights-protective-statutory-interpretation-for-detained-immigrants-and-the-rise-of-piecemeal-textualism/.

[97] 8 U.S.C. § 1252(f)(1).

[98] Aleman Gonzalez, 142 S. Ct. at 2070.

[99] 8 U.S.C. § 1252(f)(1).

[100] Garland v. Aleman Gonzalez, 142 S. Ct. 2057, 2063–65 (2022). See Eskridge & Nourse, supra note 19, at 1740 (describing a similar approach Justice Scalia took in critiquing the Church of the Holy Trinity v. United States decision as “pack[ing] the entire statute’s meaning into six words: ‘labor or service of any kind.’ Then, these six words were cracked apart, each interpreted broadly, and reassembled into a plain meaning.”).

[101] Aleman Gonzalez, 142 S. Ct. at 2063 (citing Enjoin, Black’s Law Dictionary (6^th ed. 1990)) (also referencing Enjoin, Webster’s Third New International Dictionary (1993)).

[102] Id. (citing Restrain, Oxford English Dictionary (2d ed. 1989)) (also citing Restrain, Webster’s Third New International Dictionary (1993)).

[103] Id. (citing Operation Of, Random House Dictionary of the English Language (2d ed. 1987)) (also citing Operation Of, Webster’s Third New International Dictionary (1993)).

[104] Id.

[105] Garland v. Aleman Gonzalez, 142 S. Ct. 2057, 2065 (2022). See Eskridge & Nourse, supra note 19, at 1753 (describing how “cracking” and “packing” a statute omits “relevant text and (con)text” to allow one to claim that a statute has “plain meaning” that fits a preferred interpretation).

[106] Aleman Gonzalez, 142 S. Ct. at 2065.

[107] See id. at 2068–78 (Sotomayor, J., dissenting in part).

[108] Id. at 2070.

[109] Id.

[110] Id.

[111] Id. at 2070–71.

[112] Id. at 2071.

[113] Id.

[114] Id. at 2072.

[115] Id.

[116] For an analysis of Section 1252(f)(1)’s implications for class action litigation, see Jill E. Family, Another Limit on Federal Court Jurisdiction? Immigrant Access to Class-Wide Injunctive Relief, 53 Clev. St. L. Rev. 11 (2005).

[117] Aleman Gonzalez, 142 S. Ct. at 2066.

[118] Id.

[119] Id.

[120] Patel v. Garland, 142 S. Ct. 1614, 1620 (2022); 8 U.S.C. § 1255(i).

[121] Id. at 1628 (2022).

[122] Ga. Comp. Rules & Regs., Rules 375–3–1.02(3)(e), (7) (2022).

[123] Patel, 142 S. Ct. at 1620.

[124] Id. at 1628.

[125] 8 U.S.C. 1252(b)(4)(B) (“Except as provided [with respect to nationality claims]—(B) the administrative findings of fact are conclusive unless any reasonable adjudicator would be compelled to conclude to the contrary.”).

[126] 8 U.S.C. 1252(a)(2)(B)(i).

[127] 8 U.S.C. § 1252(a)(2)(D).

[128] Patel, 142 S. Ct. at 1622.

[129] Id.

[130] Id. at 1621.

[131] See Bernstein & Staszewski, supra note 20, at 307.

[132] Patel, 142 S. Ct. at 1630 (Gorsuch, J., dissenting).

[133] Id. at 1630­–31.

[134] Id. at 1631.

[135] Eskridge & Nourse, supra note 19, at 1732.

[136] See id.

[137] Id. at 1810 (describing textualism’s “false humility”).

[138] Patel, 142 S. Ct. at 1628 (Gorsuch, J., dissenting).

[139] See Eskridge & Nourse, supra note 19 at 1810-11; see also Bernstein, supra n. 14 at 571 (describing “selecting and situating [text] as key conceptual moments in statutory interpretation”).

[140] See generally Heeren, supra note 26 (arguing that advocates have found success in appealing to judges and the publics disaffection with the immigration bureaucracy); Catherine Y. Kim, Plenary Power in the Modern Administrative State, 96 N.C. L. Rev. 77, 115 (2017) (arguing that concerns about delegating power to “unelected agency officials” have shaped immigration jurisprudence, explaining why courts continue to defer in some areas of immigration law and not others); Shalini Bhargava Ray, The Emerging Lessons of Trump v. Hawaii, 29 Wm. & Mary Bill of Rts. J. 775 (2021) (arguing that immigrants’ interests are increasingly protected, if at all, through ordinary administrative law claims rather than constitutional claims).

[141] Jill E. Family, Immigration Law Allies and Administrative Law Adversaries, 32 Geo. Immigr. L. J. 99, 103 (2018).

[142] Cf. Michael G. Kagan, Regulatory Constitutional Law: Protecting Immigrant Free Speech Without Relying on the First Amendment, 56 Ga. L. Rev. 1417, 1429-31 (2022) (discussing possibility of regulatory protection of immigrants’ speech rights in the absence of a First Amendment prohibition on selective deportation).