Peripheral Detention, Transfer, and Access to the Courts

In the last forty years, immigration detention in the U.S. has grown exponentially, largely concentrated in the southern states and outside of the country’s metropoles. In turn, federal immigration officials routinely transfer immigrants from their communities to remote jails and prisons hundreds, if not thousands, of miles away, often in jurisdictions where the law is more favorable to the government. These transfers are conducted without notice or process and frequently occur on weekends or in the predawn hours, when offices are closed and interested parties are lucky to access voicemail.

Federal immigration officials’ use of peripheral detention and transfer significantly affects immigrants’ access to the courts and their ability to raise detention challenges. Lurking beneath these issues lies a seemingly technical Supreme Court decision relied on by the government to seek dismissal of habeas actions filed by immigrant petitioners who have been ferried to faraway jails and prisons.

In Rumsfeld v. Padilla, the Supreme Court held that the “default rule” in a habeas action challenging present physical confinement is that it must be brought in a petitioner’s “district of confinement” and that a petitioner can only name a single respondent: their “immediate custodian.” However, the history and development of immigration detention and of the habeas statute offer important insights into present debates about the primacy of Padilla in the context of transfer. A mining of these histories unravels the foundational premises on which Padilla relied and encourages us to question mechanical rules that silo immigrant habeas actions in faraway fora, away from evidence, witnesses, community, counsel, and the events giving rise to the detentions themselves.


In June 2004, the Supreme Court issued its decision in Rumsfeld v. Padilla,1Rumsfeld v. Padilla, 542 U.S. 426 (2004). a post-9/11 national security case challenging the president’s authority to detain U.S. citizen José Padilla in military custody on U.S. soil by classifying him as an “enemy combatant.”2Id. at 430. The Court ruled on what at first glance was a “technical issue,”3David Stout, Supreme Court Affirms Detainees’ Right to Use Courts, N.Y. Times (June 28, 2004), []. holding that Padilla filed his habeas petition in the wrong court. Because Padilla was being held in a military brig in South Carolina, he should have filed his petition there, in his “district of confinement.”4Padilla, 542 U.S. at 446. Further, he should have named his “immediate custodian”—normally the warden of the jail, but in this case, the navy brig commander—as respondent.5Id. at 436. The Court found that the Southern District of New York, which had issued the material witness warrant leading to Padilla’s arrest and where Padilla had filed his habeas petition, did not have jurisdiction over the brig commander.6Id. at 442. Accordingly, it dismissed Padilla’s habeas petition without prejudice.7Id. at 451.

The Court’s decision in Padilla did not capture the public’s attention, in large part because it was issued on the same day as two others: Hamdi v. Rumsfeld8Hamdi v. Rumsfeld, 542 U.S. 507 (2004). and Rasul v. Bush.9Rasul v. Bush, 542 U.S. 466 (2004). Hailed as victories by civil liberties advocates, the decisions in Hamdi and Rasul rejected the executive’s claims of military necessity to deprive individuals of their access to the courts and due process rights. Yet, Padilla’s seemingly technical ruling was in fact a fraught issue on the immigration stage. When Congress passed the Antiterrorism and Effective Death Penalty Act (AEDPA) and the Illegal Immigration Reform & Immigrant Responsibility Act (IIRIRA) in 1996, it vastly expanded the criminal grounds for deportation—many of which triggered mandatory detention without bond—and limited judicial review. As a result, many immigrants, including long-time residents, were subject to arrest, detention, and potential deportation.10See Bobby Hunter & Victoria Yee, NYU Sch. of L. Immigrant Rts. Clinic, Dismantle, Don’t Expand: The 1996 Immigration Laws 7 (2017), [] (describing how, since passage of the 1996 laws, individuals can be subject to mandatory detention and deportation for certain criminal offenses, even if there was no jail sentence imposed, a sentence was suspended, or the offense has been expunged under state law). In tandem, the government began ramping up enforcement. As more and more immigrants were swept into immigration custody, many were transferred to isolated detention facilities, where they received pro forma hearings and were quickly ordered to be deported. Those immigrants who filed habeas petitions had to contend with the question at the heart of Padilla: Where was the proper forum to bring their habeas actions? Was it where they had been living for years; where they had built families; where they pursued ongoing immigration proceedings; where they had been arrested and initially placed into immigration custody? Or, was it where they were physically confined at the behest of immigration authorities? Moreover, who should they name as respondent? Was it the attorney general,11Prior to passage of the Homeland Security Act of 2002 (HSA), the authority to interpret, implement, enforce, and adjudicate immigration law lay virtually and exclusively with the attorney general of the United States. See Immigration and Nationality Act of 1952, Pub. L. No. 82-414, § 103, 66 Stat. 163, 173 (codified as amended at 8 U.S.C. § 1103). After passage of the HSA, the enforcement and service functions were largely transferred to the Department of Homeland Security. Despite this reallocation, the amendments “still apparently allow[] the Attorney General to exercise a significant amount of authority.” See Stephen R. Viña, Cong. Rsch. Serv., RL31997, Authority to Enforce the Immigration and Nationality Act (INA) in the Wake of the Homeland Security Act: Legal Issues 3 (2003). the secretary of the Department of Homeland Security, the immigration field office director, the warden of the jail where they were detained, or some combination? A circuit split emerged12Compare Armentero v. Immigr. & Naturalization Serv., 340 F.3d 1058, 1071 (9th Cir. 2003), reh’g granted, opinion withdrawn, 382 F.3d 1153 (9th Cir. 2004) (concluding that the most appropriate respondent is the “individual in charge of the national government agency under whose auspices the alien is detained,” and holding that the attorney general was the proper respondent), dismissed on other grounds, 412 F.3d 1088 (9th Cir. 2005), with Vasquez v. Reno, 233 F.3d 688 (1st Cir. 2000) (holding that, as a general rule, the attorney general is not the proper respondent to an immigrant habeas action and remanding for dismissal or transfer of the petition), Henderson v. Immigr. & Naturalization Serv., 157 F.3d 106 (2d Cir. 1998) (declining to decide whether the attorney general is the proper respondent and certifying the question of whether the INS district director in New Orleans, Louisiana is within reach of New York’s long-arm statute to the New York Court of Appeals), and Yi v. Maugans, 24 F.3d 500, 507 (3d Cir. 1994) (“It is the warden of the prison or the facility where the detainee is held that is considered the custodian for purposes of a habeas action.”). that Padilla declined to resolve in the immigration context.13Rumsfeld v. Padilla, 542 U.S. 426, 435 n.8 (2004).

Nearly twenty years later, the issue of the proper forum in a habeas action challenging immigration detention has returned to the fore. Once arrested by Immigration and Customs Enforcement (ICE) officers, many immigrants are transferred hundreds, if not thousands, of miles away from home, counsel, witnesses, evidence, loved ones, community, and the ICE field office or immigration court presiding over their cases. They are held in facilities pursuant to intergovernmental service agreements (IGSAs) with local and state governments, U.S. Marshall Service riders, or contracts with largely private facility operators,14 U.S. Gov’t Accountability Off., Report to the Chairman Committee on Homeland Security: Immigration Detention; Actions Needed to Improve Planning, Documentation, and Oversight of Detention Facility Contracts 7 tbl.1 (2021), []. which are predominantly staffed by personnel with no connection to the immigration system.

The remoteness of detention, its largely contractual nature, and the frequency of transfers pose significant questions about which courts can hear immigrant petitioners’ detention challenges. Immigrants who seek release from detention by filing habeas petitions in their home districts are met with government arguments to dismiss or transfer on grounds that Padilla’s “default rule” controls: that petitioners challenging immigration detention can only file habeas actions in the “district of confinement,” and that they can only name their “immediate custodian”—the warden of the jail—as respondent.15See, e.g., Respondents’ Memorandum of L. in Support of Their Motion to Dismiss or to Transfer the Case to the Dist. of N.J. at 5–15, Darboe v. Decker, 442 F. Supp. 3d 592 (S.D.N.Y. 2020) (No. 19-cv-11393); Defendants’ Opposition to Ex Parte Application for Temp. Restraining Ord. at 7–8, Saravia v. Sessions, 3:17-cv-03615 (N.D. Cal. June 27, 2017). The government often prevails.16See, e.g., Sidhu v. Wolf, No. CV-20-1189-PHX-MTL, 2020 WL 8458857 (D. Ariz. Dec. 17, 2020); Sanchez v. Decker, No. 19-cv-8354, 2019 WL 6311955 (S.D.N.Y. Nov. 25, 2019). As a result, habeas petitioners in immigration detention find their petitions outright dismissed, or they are forced to litigate their petitions far from the tools and community necessary to launch a viable detention challenge, often in courts that lack connections to the cases at hand. The Court’s mechanical rule, laid on top of an increasingly expansive and fluid carceral system, impedes immigrants’ access to the courts and their attempts to seek freedom from imprisonment. It also undermines confidence in the administration of justice because it prevents public monitoring by affected community members, which courts have found to be “an essential feature of democratic control.”17See, e.g., United States v. Amodeo (Amodeo II), 71 F.3d 1044, 1048 (2d Cir. 1995). While the right of public access to court proceedings is beyond the scope of this Article, it is worth noting that the executive’s practice of ferrying individuals to faraway detention centers, and then arguing for judicial proceedings in courts hundreds of miles from and untethered to the events giving rise to the actions at bar, strains the “legacy of open justice” crucial to a “free and democratic government.” See Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 589–92 (1980) (Brennan, J., concurring).

Since the nascent years of immigration law and policy, the U.S. government has employed peripheral detention—detention in oft-remote localities, distant from the political communities dedicated to enforcing justice and accountability for its members—as a means of deporting so-called “undesirable” immigrants. Often conducted in openly racialized and xenophobic terms, peripheral detention impedes access to the courts. Now no longer episodic, peripheral detention is an entrenched feature of the immigration system, which has grown its enforcement arm to include interior policing, and which disparately affects poor and minoritized immigrant communities. Transfers now serve as a primary mechanism to control the growing detained population. Professor A. Naomi Paik explains that where “inclusion in a political community stands as a precondition for rights . . . to have meaning,” those extricated from that community through the prison camp “do not have the essential ‘right to have rights.’ 18 A. Naomi Paik, Rightlessness: Testimony and Redress in U.S. Prison Camps Since World War II 5 (2016). Seemingly technical decisions that promise a mechanical and value-neutral outcome, like Padilla, reinforce this rightlessness by depriving immigrant petitioners of access to the courts best suited to adjudicate their liberty.

The adverse access-to-justice implications require a second look at the Padilla decision and the underlying rationales of the habeas statute it interpreted. In passing the Habeas Act of 1867 in the early years of Reconstruction, Congress sought to expand access to federal court review while ensuring that a petition was heard by a court with a connection to the case.19See, e.g., Emanuel Margolis, Habeas Corpus: The No-Longer Great Writ, 98 Dick. L. Rev. 557, 564–64 (1994). Contrary jurisprudence in the immigration context is both inconsistent with the habeas statute and its legislative history and anachronistic considering the current detention and transfer regime.

Numerous scholars have critiqued the impact of transfers on immigrants.20See, e.g., Sabrina Balgamwalla, ICE Transfers and the Detention Archipelago, 31 Brook. J.L. & Pol’y 1 (2022) (examining ICE detention transfer policies, with a specific eye toward the implications for government forum shopping and biased adjudications of immigrants’ claims); Adrienne Pon, Note, Identifying Limits to Immigration Detention Transfers and Venue, 71 Stan. L. Rev. 747 (2019) (describing the mechanics of immigration detention and transfer and exploring the possibility of a Fifth Amendment due process constraint where a particular forum is unfair); Roger C. Grantham, Jr., Note, Detainee Transfers and Immigration Judges: ICE Forum-Shopping Tactics in Removal Proceedings, 53 Ga. L. Rev. 281 (2018) (describing the government’s unfettered transfer authority and its effect on decisions in immigration court); Peter L. Markowitz & Lindsay C. Nash, Constitutional Venue, 66 Fla. L. Rev. 1153 (2014) (examining the history of personal jurisdiction and venue doctrine in the United States and asserting that courts should recognize the constitutional aspects of venue to protect litigants’ due process interests in fair locations for trials); César Cuauhtémoc García Hernández, Due Process and Immigrant Detainee Prison Transfers: Moving LPRs to Isolated Prisons Violates Their Right to Counsel, 21 Berkeley La Raza L.J. 17 (2011) (describing the government’s widespread use of transfers and its violation of the procedural due process rights of lawful permanent residents and offering policy options for reform); Nancy Morawetz, Detention Decisions and Access to Habeas Corpus for Immigrants Facing Deportation, 25 B.C. Third World L.J. 13 (2005) (exploring the potential problems of limiting habeas actions to the territorial site of prisons in the aftermath of Padilla, using transfers to Oakdale Federal Detention Facility as a case study). This Article adds to that scholarship by unearthing how the current immigration transfer regime problematizes jurisprudence around where a case should be heard, specifically in the context of immigration habeas actions. In Part I, I describe how the government has historically exercised its power to isolate minoritized immigrants as a tool of exclusion, from Chinese exclusion to the more recent interdiction and detention of Haitian asylum seekers. In Part II, I describe the present-day immigration detention system and situate current immigrant detention and transfer policies, which isolate minoritized immigrants and limit their access to the courts, as an extension of this history. In Part III, I analyze the debate in federal courts over the proper venue for challenges to detention following transfer, critiquing application of the Padilla decision to the immigration detention context. I mine the Reconstruction Era origins of the statutory language on which courts rely, which evince concerns sounding largely in venue and demonstrate Congress’s intent to radically expand access to habeas relief, not limit it. I present a theory for rejecting Padilla’s default rule in the immigration context and, instead, for embracing traditional venue factors.21In this Article, I refer to venue as the term is defined in federal civil actions: “[The] locality, the place within the relevant judicial system where a lawsuit should be heard according to the applicable statutes or rules.” 14D Charles Alan Wright et al., Federal Practice and Procedure § 3801 (4th ed. 2023); see also 28 U.S.C. § 1390 (“[T]he term ‘venue’ refers to the geographic specification of the proper court or courts for the litigation of a civil action . . . .”); Markowitz & Nash, supra note 20, at 1158 (describing the “core of venue” as a means of “protecting litigants against unfair locations for trial”); Neirbo Co. v. Bethlehem Shipbuilding Corp., 308 U.S. 165, 168 (1939) (“[T]he locality of a law suit—the place where judicial authority may be exercised—though defined by legislation relates to the convenience of litigants and as such is subject to their disposition.”). This is in contrast to “jurisdiction,” which defines the types of cases and persons falling under a court’s adjudicatory authority. However, as is demonstrated by this Article and expounded in Wright and Miller, the terms are often confused. Wright et al., supra, § 3801 (noting that “[s]ometimes venue is confused with subject matter jurisdiction” and describing similarities between personal jurisdiction and venue, including that both are “concerned with the territorial reach of the court, and not with its inherent adjudicative power,” but that there are “important distinctions” between the two concepts).

I. The Racialized Geography of U.S. Immigration Detention

On June 28, 2004, the Supreme Court issued its decision in Rumsfeld v. Padilla, a habeas action challenging U.S. citizen José Padilla’s detention in military custody at a navy brig in South Carolina. Padilla was arrested in Chicago two years earlier on a material witness warrant, relating to a grand jury investigation into the September 11th attacks, in the Southern District of New York.22Rumsfeld v. Padilla, 542 U.S. 426, 430 (2004). He was held in federal criminal custody in New York for one month before President Bush designated him an “enemy combatant” and ordered him transferred to Department of Defense custody in South Carolina.23Id. at 431–32.

The court in the Southern District of New York immediately vacated his warrant, thereby terminating the Department of Justice’s custody of Padilla.24Id. at 432 n.3. Subsequent to his transfer and two days after the New York-issued warrant had been vacated, Padilla filed a habeas petition in the Southern District alleging that his military detention violated the Constitution.25Id. at 432. The petition named the president, the secretary of defense, and the navy brig’s commander as respondents. The government moved to dismiss Padilla’s petition for want of jurisdiction, arguing that the only proper respondent in Padilla’s habeas action was the commander of the navy brig where Padilla was then detained. The government further argued that the court in the Southern District had no jurisdiction over the commander, nor could it issue a writ of habeas corpus outside of its “territorial jurisdiction.” The district court denied the government’s motion, as did the U.S. Court of Appeals for the Second Circuit. The government petitioned for writ of certiorari.26Id. at 432–34.

In a 5–4 decision written by Chief Justice Rehnquist, in which Justice Kennedy concurred, the Court dismissed Padilla’s habeas action without prejudice to future filing.27Id. at 451. The Court held that “the default rule is that the proper respondent is the warden of the facility where the prisoner is being held”28Id. at 435. and that “the traditional rule has always been that the Great Writ is ‘issuable only in the district of confinement.’ 29Id. at 442 (quoting Carbo v. United States, 364 U.S. 611, 618 (1961)). In support of its holding, the Court cited to 28 U.S.C. § 2241(a), which provides that federal court judges may grant writs of habeas corpus “within their respective jurisdictions.”30Id. at 442 (quoting 28 U.S.C. § 2241(a)). The majority stated that its decision—“derived from the terms of the habeas statute”31Id. at 447.—would prevent forum shopping by habeas petitioners and district courts with simultaneous jurisdiction over a petition and track consistently with the concerns of the Congress that passed the Habeas Act 137 years prior.32Id. Notably, the Court was silent regarding how its holding might apply in the immigration context. It noted in a footnote that its earlier decision in Ahrens v. Clark33Ahrens v. Clark, 335 U.S. 188 (1948). “left open” the question of whether the attorney general is a proper respondent to a habeas petition filed by an immigrant detained pending deportation, or whether the immediate custodian rule applied, but “decline[d] to resolve it.”34Padilla, 542 U.S. at 435 n.8.

Despite the Court’s careful avoidance of the immigration habeas context in Padilla, government lawyers regularly cite Padilla when seeking dismissal of habeas actions filed by immigrants in detention. More often than not, courts apply Padilla’s “default rule.”35See, e.g., Lopez-Marroquin v. Barr, 955 F.3d 759, 759–60 (9th Cir. 2020) (construing emergency motion to remand as a habeas challenge to “core” physical confinement and transferring the case to the “district of confinement” in the Southern District of California); Bode v. Kolitwenzew, No. 20-cv-2258, 2020 WL 12309500 (C.D. Ill. Dec. 28, 2020) (finding that petitioner had erroneously filed his petition in the Southern District of Illinois, when he should have filed in the Central District under the “district of confinement” rule, but transferring the case to the Northern District of Illinois because petitioner was transferred to a detention facility there on the same day he filed); Singh v. Wolf, No. CV-20-1169-PHX-SPL, 2020 WL 8083631 (D. Ariz. Dec. 16, 2020) (granting government motion to dismiss or transfer to the Southern District of Mississippi where petitioner was then detained, citing Padilla, even though credible fear interview and immigration court review were conducted in Arizona); Deng v. Crawford, No. 20-cv-199, 2020 WL 6387010, at *4 (E.D. Va. Sept. 30, 2020) (concluding that the immediate custodian rule bars petitioner from naming respondents other than the director of ICA Farmville, where he was then detained); Wamala ex rel. Malachi v. U.S. Immigr. & Customs Enf’t, No. 19-cv-00067, 2019 WL 6713246, at *1 (W.D.N.C. Dec. 9, 2019) (transferring petitioner’s action after concluding that petitioner’s immediate custodian was not in the Western District of North Carolina, where he was arrested, but in the Middle District of Georgia, where he was then detained at Stewart Detention Center); S.N.C. v. Sessions, 325 F. Supp. 3d 401, 410 (S.D.N.Y. 2018) (concluding that the warden of the detention facility in New Jersey was the proper respondent and that the court did not have jurisdiction over the habeas claims challenging detention because the claims were brought outside the “district of confinement”). But see Pham v. Becerra, No. 23-cv-01288, 2023 WL 2744397, at *3–4 (N.D. Cal. Mar. 31, 2023) (quoting Saravia v. Sessions, 280 F. Supp. 3d 1168, 1185 (N.D. Cal. 2017), aff’d sub nom. Saravia ex rel. A.H. v. Sessions, 905 F.3d 1137 (9th Cir. 2018)) (collecting cases to demonstrate a “general consensus” in the Northern District of California that Padilla does not reach the question of whether the court has jurisdiction where an individual is detained “in a facility run by an entity other than the federal government,” and assuming jurisdiction over the matter). However, federal courts that apply Padilla’s holding in a strict and inflexible manner that depends solely on the location of the petitioner elide the unique contours of immigration detention, including the federal government’s use—both historically and today—of peripheral detention and transfer to cut off access to the courts.

As many scholars have explained, federal immigration authorities use detention to exclude and control “undesirable” immigrants in the U.S.—with particular emphasis on minoritized racial and ethnic groups.36See, e.g., Daniel Kanstroom, Deportation Nation: Outsiders in American History (2007); César Cuauhtémoc García Hernández, Migrating to Prison: America’s Obsession with Locking up Immigrants (2019). Immigration detention impacts many groups of people. I emphasize the targeting of racial and ethnic groups because of how prominently and explicitly race factors into major developments in immigration detention policy. One undertheorized feature of this detention power is how the location of detention deepens subordination by limiting marginalized groups’ access to justice. In this Part, I trace the history of federal immigration authorities’ use of peripheral detention as a tool for racialized social control. Once run by states and localities, immigration detention grew in parallel with federal immigration law and echoed its targets during three significant periods of detention expansion: first, Chinese and other Asian immigrants and persons deemed “undesirable”; then, as the nation solidified its borders, Mexican migrants; and later, following the abolition of racial quotas in immigration law, the arrival of Haitian asylum seekers. Immigration detention functioned to jettison groups of immigrants to the periphery, creating what scholars Jenna Loyd and Alison Mountz term “contingency space”37Jenna M. Loyd & Alison Mountz, Boats, Borders, and Bases: Race, the Cold War, and the Rise of Migration Detention in the United States 106–07 (2018). in geographically remote locations, in an attempt to forcibly disappear immigrants, deprive them of community support, and inhibit access to the courts and due process rights.

A. The Period of Chinese Exclusion and Detentions at McNeil and Angel Islands

While early American history is characterized by some as reflecting an “open door” policy with respect to immigration,38This “open door” is not without caveat. The border was largely uncontrolled by federal immigration authorities, but only certain immigrants were welcomed. Moreover, prior to the federalization of immigration law, states controlled their own borders and prevented the settlement of “undesirable” people, including, in some cases, immigrants. See Gerald L. Neuman, The Lost Century of American Immigration Law (1776–1875), 93 Colum. L. Rev. 1833, 1841–84 (1993) (exploring five major categories of immigration policy implemented by state legislation before 1875); see also Adam Goodman, The Deportation Machine: America’s Long History of Expelling Immigrants 10–11 (2020). this policy shifted in the nineteenth century. In 1819, the federal government began to keep track of immigration,39Congress passed the Steerage Act of 1819 on March 2, 1819, which went into effect on January 1, 1820. An Act Regulating Passenger Ships and Vessels, Pub. L. No. 15-46, 3 Stat. 488 (1819). The Steerage Act required a “captain or master of any ship or vessel” arriving in the United States or one of its territories to provide a passenger list, including the age, sex, and occupation of all passengers, as well as “the country to which they severally belong, and that of which it is their intention to become inhabitants.” Id. § 4. and in 1875, it passed its first federal law regulating immigration.40An Act Supplementary to the Acts in Relation to Immigration (The Page Act), ch. 141, 18 Stat. 477, 477 (1875) (restricting immigration for the first time based on race, gender, and class, including by prohibiting immigration of laborers from China, Japan, “or any Oriental country” brought to the U.S. involuntarily, as well as women who entered for the purposes of sex work). The gradual closing of the door to immigrants was rooted in Anglo-American anti-Chinese sentiment in western states, as manifest destiny, the Gold Rush, and railroads brought Anglo-Americans and Chinese residents into proximity. While Chinese immigrants represented only a small fraction of immigrants during this period,41 Erika Lee, At America’s Gates: Chinese Immigration During the Exclusion Era, 1882–1943, at 25 (2003) (noting that, from 1870 to 1880, 138,941 Chinese immigrants—the majority of whom were men and laborers—entered the United States, representing only 4.3 percent of the total number of immigrants during the decade). they were nonetheless targets of racial hostility and violence. Anti-Chinese sentiment materialized in state legislation “broadly designed to consolidate Anglo-American power in the West” through voter disenfranchisement, bans on owning land, and bans on testifying or serving on a jury.42 Kelly Lytle Hernández, City of Inmates: Conquest, Rebellion, and the Rise of Human Caging in Los Angeles, 1771–1965, at 66 (2017). Chinese residents were not the only targets. Restrictions were enforced against Black and Indigenous people, together with the Chinese, entrenching racial hierarchy and Anglo-American dominance.43See, e.g., People v. Hall, 4 Cal. 399, 399 (1854) (interpreting § 394 of California’s Civil Practice Act—which prohibited “Indians or Negroes” from testifying as a witness in an action involving a white person—and § 14 of California’s Criminal Act—which prohibited “Black, or Mulatto, or Indian” persons from giving evidence in favor or against a white person—to include a prohibition against testimony by “Chinese and all other people not white”).

Not all Anglo-Americans, however, expressed anti-Chinese fervor. Many individuals with interests in labor, international trade, and Christianity supported immigration from China; others, scarred by the racial animus that fueled the Civil War, promoted an antiracist immigration agenda.44 Lucy E. Salyer, Laws Harsh as Tigers: Chinese Immigrants and the Shaping of Modern Immigration Law 9 (1995). These sentiments were reflected in the U.S. government’s adoption of the Burlingame Treaty with China in 1868.45Burlingame Treaty, China-U.S., art. V, July 28, 1868, 16 Stat. 739. The Treaty permitted Chinese and American individuals to immigrate to each other’s countries and required that visitors and residents “enjoy the same privileges, immunities, or exemptions in respect to travel or residence as may there be enjoyed by the citizens or subjects of the most favored nation.”46Id. art. VI. Notably, however, the Burlingame Treaty expressly did not “confer naturalization” on Chinese citizens in the U.S. or vice versa.47Id.

Anti-Chinese forces quickly responded to the Burlingame Treaty. Settlers lobbied Congress for restrictionist immigration policies. Their lobbying efforts culminated in the passage of the 1882 Chinese Exclusion Act, which prohibited Chinese laborers from entering the United States for a period of ten years,48Chinese Exclusion Act, Pub. L. No. 47-126, 22 Stat. 58 (1882) (repealed 1943). later extended by the Geary Act of 1892.49After enactment of the Chinese Exclusion Act, settlers remained dissatisfied because the Act had no effect on Chinese residents who were already present in the United States. They responded with years of lobbying for more intensive restrictions, not only on Chinese migration but also on Chinese residents’ very existence. In 1892, Thomas J. Geary, a U.S. congressman from California, introduced legislation in the House of Representatives to continue Chinese exclusion, with more stringent requirements. He secured enough support for passage of the Geary Act, which extended Chinese exclusion for an additional ten years until 1902. It became permanent thereafter until the Chinese Exclusion Act’s repeal in 1943. Beth Lew-Williams, The Chinese Must Go: Violence, Exclusion, and the Making of the Alien in America 202, 208, 232 (2018). Among its strictures, the Geary Act required Chinese residents to register for a “certificate of residence” with the Internal Revenue Service within one year of the Act’s passage. Geary Act, ch. 60, § 6, 27 Stat. 25, 25–26 (1892). Failure to do so would result in a finding that a Chinese person was unlawfully within the United States and subject them to arrest. Id. Moreover, a Chinese person who failed to register had to prove “that by reason of accident, sickness or other unavoidable cause, he ha[d] been unable to procure his certificate,” and prove residency in the United States at the time of passage of the Geary Act through the corroboration of “one credible white witness.” Id. The Geary Act further provided that “any Chinese person or person of Chinese descent convicted and adjudged to be not lawfully entitled to be or remain in the United States shall be imprisoned at hard labor for a period of not exceeding one year.” Id. § 4. In addition, vigilantes forcibly purged Chinese residents from western towns through brutal violence.50 Lew-Williams, supra note 49, at 1–3 (noting that, between 1885 and 1886 alone, no fewer than 168 communities in the western states drove out their Chinese residents, a fait accompli that led to erasure from the national memory).

After passage of the Chinese Exclusion Act in 1882, the U.S. government began ramping up immigration detention in earnest. It incarcerated Chinese immigrants on steamships in San Francisco Bay pending inspection and admission.51 Erika Lee & Judy Yung, Angel Island: Immigrant Gateway to America 10 (2010). And it locked up Chinese immigrants who sought clandestine entry across the Canadian border at a penitentiary on McNeil Island, two-and-a-half miles off the coast of Washington State near Tacoma.52 Elliott Young, Forever Prisons: How the United States Made the World’s Largest Immigrant Detention System 23–24 (2021). McNeil Island would later become one of the U.S.’s first federal prisons after passage of the Three Prisons Act in 1891. It remained the only federal prison on the West Coast until the 1930s. Id. at 24. In both instances, the U.S. government forced Chinese immigrants to the periphery, reflecting a national political agenda of exclusion.

While the penitentiary at McNeil Island opened its doors in 1875, U.S. officials did not begin sending large numbers of Chinese immigrants there until the mid-1880s, when they ramped up enforcement of the Chinese Exclusion Act. The Act did not criminalize presence in the U.S. in and of itself. Federal district courts, however, routinely sentenced Chinese immigrants found to be unlawfully present in the U.S. with six months of hard labor, often at McNeil Island.53See, e.g., Judgment and Sentence, United States v. Ah Jim, No. 2226 (Dist. Ct. Tacoma Mar. 15, 1888), reprinted in Elliott Young, Caging Immigrants at McNeil Island Federal Prison, 1880–1940, 88 Pac. Hist. Rev. 48, 55 (2019); see also Young, supra note 52, at 29 (noting that, in the five years between 1887 and 1892, 243 Chinese people were imprisoned on McNeil Island, almost all for illegal entry or smuggling, with those detained serving sentences of between one month and two-and-a-half years). Federal immigration detention at McNeil Island coincided with anti-Chinese violence, which hardened racial boundaries. Hardened boundaries signaled the improbability of peaceful coexistence in diverse communities in the interior and platformed exclusion as a solution to white terror.54 Lew-Williams, supra note 49, at 8 (describing how vigilante violence by white settlers displaced more than 20,000 Chinese residents, accelerating Chinese segregation in the western region and hastening the return of Chinese residents to China); see also Mapping Anti-Chinese Violence, Tacoma Method, [] (mapping anti-Chinese violence in the Puget Sound region of the Washington Territory between 1885 and 1886 and across the American West between 1849 and 1922).

Further south, in San Francisco, detention of immigrants at the port was moved from steamships to a “detention shed,” which became notorious for chronic overcrowding and unsanitary conditions. Many Chinese immigrants detained in the shed escaped; others planned revolt.55 Lee & Yung, supra note 51, at 10–11. Congress ultimately agreed to fund a detention center at Angel Island, prompting protests from Chinese community leaders due to the difficulty for witnesses to travel there.56Id. at 11. Witnesses were crucial to demonstrate that an individual was exempt from the exclusion laws.57Id. at 30. The concerns expressed by Chinese community leaders about isolation, distance, and impediments to accessing witnesses, evidence, and justice formed the very rationale for siting detention on the island: Hart Hyatt North, the San Francisco commissioner of immigration, made clear that Angel Island was chosen precisely because the island would be “the most effective means of keeping a watchful eye over the newly arriving Chinese. They would be separated from friends and family who might try to coach them on how to pass the interrogations, something that was common to the ‘wily Chinee [sic].’ 58Id. at 12. North further noted that the island was “ideal” because it was detached from the mainland, protecting Americans from contagion and danger and preventing escape.59Id. The immigration station at Angel Island opened for detention on January 21, 1910, amidst protests from Chinese community members.60Id. at 13–15.

Not everyone ended up on Angel Island, however. Ship passengers were filtered by race, nationality, and immigrant and economic status. Some, mainly in first-class but sometimes in second-class, were quickly examined by immigration officials between quarantine and the wharf, so they could disembark immediately.61Id. at 33. These individuals were largely white, wealthy U.S. citizens and European travelers.62Id. The remaining second- and third-class and steerage passengers were escorted onto a ferry to Angel Island; they were subjected to family separation, invasive health examinations, lengthy interrogations, and unsanitary conditions.63Id. at 35–49. The ferry was racially segregated, just as the island was, with white passengers on the upper deck and nonwhite passengers below.64Id. at 33–34. Race and nationality largely determined a person’s length of detention on the Island. From 1913 to 1919, empirical data shows that approximately 70 percent of passengers arriving to San Francisco were detained at Angel Island, and nearly 60 percent of Chinese passengers were detained for more than three days.65Robert Barde & Gustavo J. Bobonis, Detention at Angel Island: First Empirical Evidence, 30 Soc. Sci. Hist. 103, 107 tbl.1 (2006). This detention rate starkly contrasts Ellis Island’s rate, “where only 10 percent of all arrivals were detained for legal reasons and another 10 percent were detained for medical treatment.”66 Lee & Yung, supra note 51, at 57. Empirical data further shows that Chinese and Japanese passengers were detained longer than all other non-Asians. Specifically, 76 percent of Chinese passengers were detained, 19.4 percent of whom were detained for longer than two weeks; 89.6 percent of Japanese passengers were detained during the same period, but normally for a shorter duration.67Barde & Bobonis, supra note 65, at 113.

Largely a port of entry for Asian immigrants, Angel Island was characterized by immigration policies designed to exclude.68Id. at 8; see also Lee & Yung, supra note 51, at 8 (noting that Angel Island was far from the “Ellis Island of the West,” in that the Asian immigrants who entered spent days and weeks on the island—as compared to a few hours—and were often excluded and denied the ability to naturalize); Aristide R. Zolberg, A Nation by Design: Immigration Policy in the Fashioning of America 1–2 (2006) (describing how early settlers engaged in nation-building through elimination of the original dwellers, importation of African workers whom they excluded altogether and later tried to incentivize to self-remove, and active recruitment of settlers from Northern and Western Europe; in this scheme, immigration policy emerged as yet another tool of nation-building and “fostered the notion that the nation could be designed, stimulating the elevation of that belief into an article of national faith”). In addition to being detained for longer periods, Asian passengers detained on the island were subject to disparate treatment based on race. For example, in 1909, concessionaires were allowed to spend fourteen cents per meal for Asian passengers detained on the island, but fifteen cents for European passengers.69 Lee & Yung, supra note 51, at 61. Further, detained Chinese passengers were not permitted visitors of any sort before their cases were settled, while others were provided specific opportunities to see friends, family, and attorneys.70Id. at 63. Women and individuals detained from Europe were also permitted to go on walks outside the detention center and visit family; Chinese men were not.71Id. at 63–64. Unlike its counterpart on the East Coast, Angel Island was not designed to admit immigrants to American shores for purposes of assimilation and naturalization.72Not all immigrants who sought admission at Ellis Island were permitted entry. Notable exceptions were individuals with disabilities, who were increasingly excluded as the eugenics movement gained popularity. See generally Douglas C. Baynton, Defectives in the Land: Disability and American Immigration Policy, 1882–1924, J. Am. Ethnic Hist., Spring 2005, at 31.

Angel Island not only functioned to prevent entry but also played a central role in the deportation of individuals arrested in the interior, pursuant to newly minted immigration laws and policies that targeted Asian immigrants, sex workers, individuals likely to become a “public charge,” those with criminal convictions, and those with radical politics.73 Lee & Yung, supra note 51, at 6–9.

In this context of growing exclusion and detention, Chinese immigrants and community members resisted, raising legal challenges to U.S. law and policy and sometimes circumventing the law altogether. Many of these legal challenges were successful in the courts, thwarting nativists’ intentions to construct an American identity around ideas of racial purity. Yet, in response to one such challenge, the Supreme Court enshrined civil immigration detention into the fabric of American jurisprudence.

Wong Wing, Lee Poy, Lee You Tong, and Chan Wah Dong were arrested in Detroit on July 15, 1892, two months after the passage of the Geary Act.74See Gerald L. Neuman, Wong Wing v. United States: The Bill of Rights Protects Illegal Aliens, in Immigration Law Stories 31, 34 (David A. Martin & Peter H. Schuck eds., 2005). They were subsequently held to be in the country unlawfully, ordered deported, and sentenced to sixty days of hard labor.75Wong Wing v. United States, 163 U.S. 228, 229 (1896). The four men challenged the Geary Act and its attendant criminal penalties. Through counsel, they asserted that the Geary Act violated the Fifth, Sixth, and Thirteenth Amendments by subjecting petitioners to hard labor—a form of punishment and involuntary servitude—without criminal trial.76Neuman, supra note 74, at 36 (citing brief for appellants). The four men’s cases were consolidated;77Id. at 35. their challenge took over three years to percolate through the judicial system before arriving at the Supreme Court in 1896.78Id. at 36–37. On May 18, 1896, the same day the Court decided Plessy v. Ferguson, a decision that sanctioned de jure segregation and stood for nearly sixty years,79Plessy v. Ferguson, 163 U.S. 537, 551 (1896) (“We consider the underlying fallacy of the plaintiff’s argument to consist in the assumption that the enforced separation of the two races stamps the colored race with a badge of inferiority. If this be so, it is not by reason of anything found in the act, but solely because the colored race chooses to put that construction upon it.”), overruled by Brown v. Bd. of Educ., 347 U.S. 483 (1954). the Court also issued its decision in Wong Wing v. United States. The Court observed that while Chinese exclusion was lawful, the hard labor provision was not necessary to deportation, stating: “But when Congress sees fit to further promote such a policy by subjecting the persons of such aliens to infamous punishment at hard labor, or by confiscating their property, we think such legislation, to be valid, must provide for a judicial trial to establish the guilt of the accused.”80Wong Wing, 163 U.S. at 237. The Court further affirmed that constitutional protections under the Fifth and Sixth Amendments extend to “all persons within the territory of the United States,” citizen and noncitizen alike.81Id. at 238.

Yet, while Wong Wing is notable for protecting certain constitutional rights of non-U.S. citizens,82See Neuman, supra note 74, at 40. for present purposes, it is most notable for what it permits. In addition to upholding much of the Geary Act and affirming Congress’s power to control immigration,83Wong Wing, 163 U.S. at 237 (“We regard it as settled by our previous decisions that the United States can, as a matter of public policy, by Congressional enactment, forbid aliens or classes of aliens from coming within their borders, and expel aliens or classes of aliens from their territory, and can, in order to make effectual such decree of exclusion or expulsion, devolve the power and duty of identifying and arresting the persons included in such decree, and causing their deportation, upon executive or subordinate officials.”) the Court, in dicta, sanctioned immigration detention or temporary confinement “as part of the means necessary to give effect to the provisions for the exclusion or expulsion of aliens.”84Id. at 235. The Supreme Court expressly upheld its validation of detention pending deportation some fifty-six years later in Carlson v. Landon, 342 U.S. 524 (1952).

B. The Criminalization of Migration and Creation of Carceral Spaces for Mexican Migrants Along the Southwest Border

During the period of Chinese exclusion, immigration detention emerged as a means of geographically isolating and asserting control over immigrants deemed racially undesirable. This phenomenon continued in the southwestern borderlands, where immigration from Mexico threatened nativists’ hopes of a white-dominant American nationality. However, rather than constructing a civil detention system, legislation passed in the 1920s enabled the creation of carceral spaces along the border by criminalizing migration.

On May 26, 1924, President Calvin Coolidge signed into law the Immigration Act of 1924. The Act permanently entrenched a quota system for migration first implemented in 1921 in an effort to keep the nation’s “racial strains” majority Anglo-Saxon.85See Mae M. Ngai, Impossible Subjects: Illegal Aliens and the Making of Modern America 22–25 (2004). This quota system exempted Mexico and other countries in the Western Hemisphere, largely to support the labor interests of industry and agribusiness.86See Immigration Act of 1924, Pub. L. No. 68-139, § 4(c), 43 Stat. 153, 155 (amended 1965) (defining “non-quota immigrant” as “[a]n immigrant who was born in the Dominion of Canada, Newfoundland, the Republic of Mexico, the Republic of Cuba, the Republic of Haiti, the Dominican Republic, the Canal Zone, or an independent country of Central or South America, and his wife, and his unmarried children under 18 years of age, if accompanying or following to join him”); Ngai, supra note 85, at 50. However, the 1924 Act signaled a new approach to immigration control that would ultimately be wielded against immigrants from Mexico, leading to mass detentions and deportations along the border.

Prior to 1924, Congress created statutes of limitation for deportation. For example, the Immigration Act of 1907 provided that an immigrant who entered the United States in violation of the law be taken into custody and deported within three years of entry into the United States.87Immigration Act of 1907, Pub. L. No. 59-96, §§ 20–21, 34 Stat. 898, 904–05. Notably, Chinese immigrants could still be arrested and deported pursuant to the Chinese exclusion laws after the three-year period had lapsed. See, e.g., Wong Chung v. United States, 244 F. 410, 411 (9th Cir. 1917). This provision was extended but not eliminated in 1917, when Congress enacted a five-year statute of limitations.88Immigration Act of 1917, Pub. L. No. 64-301, §§ 19–20, 39 Stat. 874, 889–90. The Immigration Act of 1917 is further notable for its expansion of crime-based grounds of deportation. Id. § 19 (providing that an immigrant who is sentenced to a term of imprisonment for one year or more as a result of a conviction for an offense categorized as a crime involving moral turpitude is deportable); see also Alina Das, Inclusive Immigrant Justice: Racial Animus and the Origins of Crime-Based Deportation, 52 U.C. Davis L. Rev. 171, 179 (2018) (noting that, in 1891, Congress included conviction of a crime involving moral turpitude as a ground for exclusion, and this was expanded to a ground for deportation in 1917 as part of a long line of legislative acts using criminal history to deport immigrants). The Immigration Act of 1924 upended this scheme by eliminating the statute of limitation and providing that deportation could occur “at any time” after entry for any person entering after July 1, 1924 without a valid visa or inspection.89Immigration Act of 1924, Pub. L. 68-139, § 14, 43 Stat. 153, 162 (amended 1952). Notably, Congress did away with time limits by which the United States may deport a noncitizen through passage of the Immigration and Nationality Act of 1952. See Immigration and Nationality Act of 1952, Pub. L. No. 82-414, 66 Stat. 163 (amended 1965); see also Andrew Tae-Hyun Kim, Deportation Deadline, 95 Wash. U. L. Rev. 531, 569 (2017) (describing the historical contraction of statutes of limitation in immigration law). While Mexican immigrants were not subject to quotas, they were still required to obtain visas upon entry and adhere to strict entry requirements. As a result, many Mexican immigrants, migrating across the porous and often-ambiguous U.S.-Mexico border without inspection, became vulnerable to the new scheme.

At the same time, nativists interested in cultivating a Euro-American identity—often foiled in their endeavor to exclude non-European (specifically Mexican) immigrants by western industrialists and agriculturalists seeking sources of labor—reframed the immigration debate as one of “controlling unauthorized Mexican immigration rather than capping authorized migration.”90 Lytle Hernández, supra note 42, at 138. Highlighting the number of unauthorized border crossings made by Mexican immigrants each year, Congressman Coleman Livingston Blease of South Carolina, a known segregationist, proposed to criminalize unlawful entry.91Isaac Stanley-Becker, Who’s Behind the Law Making Undocumented Immigrants Criminals? An ‘Unrepentant White Supremacist.’, Wash. Post (June 27, 2019, 5:54 AM), []. His proposal ultimately passed without objection in 1929.92Undesirable Aliens Act of 1929, Pub. L. No. 70-1018, 45 Stat. 1551 (repealed 1952). The “Undesirable Aliens Act of 1929” made unlawful entry a misdemeanor, punishable by imprisonment of up to one year or a fine of up to $1,000 or both, and made a second unlawful entry punishable by imprisonment of up to two years or a fine of up to $1,000 or both.93Id. §§ 1(a)–2. As historian Mae Ngai contends, these new immigration laws “rearticulated the U.S.-Mexico border as a cultural and racial boundary, as a creator of illegal immigration.”94 Ngai, supra note 85, at 67.

The repercussions of the criminalization of migration under the 1929 Act on Mexican immigrants were virtually immediate. Historian Kelly Lytle Hernández documents that after only one year, U.S. attorneys had prosecuted over 7,000 cases and, by year ten, had prosecuted more than 44,000 cases, with a conviction rate that never fell below 93 percent.95 Lytle Hernández, supra note 42, at 138–39. These prosecutions ballooned the federal prison population overwhelmingly with Mexican immigrants, who comprised no less than 84.6 percent of all imprisoned immigrants.96Id. at 139. As more immigrants faced prosecution and detention, Congress responded with the construction of prisons along the U.S.-Mexico border.97Id. at 140.

While Blease’s law was ultimately repealed, its principles remain. The provisions were recodified in 8 U.S.C. § 1325988 U.S.C. § 1325(a) makes it a federal misdemeanor to enter the United States without authorization, punishable by up to six months incarceration for a first offense, and up to two years for a second offense. 8 U.S.C. § 1325(a). and 8 U.S.C. § 1326998 U.S.C. § 1326(a) makes it a federal felony for an individual to reenter the United States without authorization after having been deported from the United States, punishable by two years’ incarceration. Id. § 1326(a). If an individual was deported following a criminal conviction, they face up to twenty years’ incarceration. Id. § 1326(b). through passage of the Immigration and Nationality Act of 1952 (The McCarran-Walter Act). Although the laws lay dormant for much of the twentieth century,100 Jesse Franzblau, Nat’l Immigrant Just. Ctr., A Legacy of Injustice: The U.S. Criminalization of Migration 11 (Heidi Altman, Tara Tidwell Cullen, Jordyn Rozensky & Mary Meg McCarthy eds., 2020), [] (stating that, by 1957, there were under 2,000 convictions pursuant to 8 U.S.C. §§ 1325 and 1326, and that the statute was in limited use from the 1960s through the 1990s); see also Transactional Recs. Access Clearinghouse, Syracuse Univ., Illegal Reentry Becomes Top Criminal Charge, at fig.1 & tbl. 1 (2011), [] (demonstrating that criminal immigration prosecutions from 1986 to the late 1990s rarely exceeded 5,000 but began to increase steadily beginning in 1998 and dramatically beginning in 2004). in the past twenty years these laws have resurfaced and continue to target Mexican and Latine immigrants101See Declaration of Michael T. Light at 2–3, United States v. Machic-Xiap, No. 19-cr-00407-SI (D. Or. March 31, 2021), ECF No. 52-1 [hereinafter Light Declaration] (finding that 99 percent of all cases sentenced under § 2L1.2—Unlawfully entering or remaining in the United States—of the U.S. Sentencing Guidelines involved “Hispanic” defendants). by criminalizing migration, separating families, and concentrating detention along the southwest border.102 Franzblau, supra note 100, at 5, 9–10 (describing how prosecutions separate families by targeting individuals traveling in family units, those seeking to be reunited with their family members and loved ones in the United States, and those apprehended by the Department of Homeland Security after several years of living in the United States). Now, prosecutions for migration-related offenses are the second most numerous prosecutions on the federal docket103Light Declaration, supra note 101, at 2. and are concentrated in the southwest border districts.104 U.S. Sent’g Comm’n, Illegal Reentry Offenses 8 (2015), []. In fiscal year 2013, the top five districts for felony illegal reentry prosecutions were Southern Texas, Western Texas, New Mexico, Arizona, and Southern California. Id. A TRAC snapshot of prosecutions in June 2022 demonstrates that prosecutions continue to be concentrated overwhelmingly in the borderlands, with the Western District of Texas (San Antonio), the Southern District of Texas (Houston), and the District of Arizona registering the largest number of prosecutions. See Transactional Recs. Access Clearinghouse, Syracuse Univ., Immigration Convictions for June 2022, []. Individuals charged for migration-related offenses almost never go to trial.105Light Declaration, supra note 101, at 4 (“Although trials in federal court are generally rare, they are virtually non-existent among 2L1.2 cases . . . [O]f the nearly 88,000 2L1.2 cases sentenced over the last 5 years, less than 0.3% of them were convicted by trial.”). They are often prosecuted through fast-track, “zero-tolerance” criminal proceedings. Defendants are rarely appointed defense counsel until their court date, have little time to consult with an attorney to understand the charges or consequences of a guilty plea, and appear before the federal court shackled and in an assembly line of up to eighty people.106Brief of Amici Curiae Nonprofit Orgs., Immigrant Rts. & Cmty. Grps., & Mental Health Providers in Support of Defendant-Appellant, United States v. Hernandez-Becerra, 636 Fed. Appx. 943 (9th Cir. 2016) (No. 14-50257). Most plead guilty, often without understanding their constitutional rights and in violation of Federal Rule of Criminal Procedure 11.107 Franzblau, supra note 100, at 9. Yet, the data suggests that individuals prosecuted under these statutes “are among the most likely to receive a prison sentence despite having the lowest mandatory minimums and final offense levels.”108Light Declaration, supra note 101, at 4. The effect is that these criminalization provisions—rooted as they are in the racist and eugenicist ideology of the “Tribal Twenties”109 Lytle Hernández, supra note 42, at 132.—have expanded the “legal and material conditions for deportability,”110 Loyd & Mountz, supra note 37, at 199. isolating and controlling Mexican and Latine immigrants along the Southwest border through the use of assembly-line justice and carceral spaces.

The immigrants subject to these provisions often have deep ties to the United States. The average age of an individual prosecuted for reentry is thirty-six, whereas the average age at initial entry is seventeen, suggesting that individuals reenter over a prolonged period.111 U.S. Sent’g Comm’n, supra note 104, at 27–29. Most individuals prosecuted under the reentry provisions have relatives—other than children—in the United States,112See id. at 25 n.42 (defining relative as a spouse, sibling, parent, grandparent, aunt, uncle, or cousin). and virtually half (49.5 percent) have at least one child living in the United States.113Id. at 25. Moreover, nearly three quarters have worked in the United States for more than one year.114Id. at 26. Federal law and policy requires Mexican families “to choose between permanent separation and making an unauthorized crossing to be together with their loved ones.”115 Franzblau, supra note 100, at 6. The decision carries the possibility of ensnarement in the federal criminal legal system, prosecution, and detention, all with minimal process and all by design.

C. The Reinvention of Immigration Detention and the Creation of “Contingency Spaces” for Haitian Immigrants

Two years after the Supreme Court validated detention pending deportation in Carlson v. Landon,116Carlson v. Landon, 342 U.S. 524 (1952). the former Immigration and Naturalization Service (INS)—under the Eisenhower Administration and Attorney General Herbert Brownell, Jr.’s leadership—began a wind-down of immigration detention, resulting in its near abolition.117See García Hernández, supra note 36, at 46–48. In 1954, INS officials adopted a policy of “parole,” permitting the vast majority of immigrants to be released into community during the pendency of their immigration proceedings.118Ana Raquel Minian, America Didn’t Always Lock Up Immigrants, N.Y. Times (Dec. 1, 2018), []. In a 1956 statement before the Subcommittee on Immigration of the Committee on the Judiciary of the U.S. Senate, Attorney General Brownell reaffirmed this policy, stating,

A practice which had grown up over the years was the indiscriminate detention of aliens who were awaiting decisions as to admissibility or deportability. While detention of some was undoubtedly necessary, our studies showed that the majority could be released under reasonable supervision without harm to the public safety or security.119Herbert Brownell, Jr., Att’y Gen. of the U.S., Statement Prepared for Delivery Before Subcommittee on Immigration of Committee on the Judiciary of the U.S. Senate (Apr. 13, 1956), [].

This antidetention frame—espoused in the courts120See Leng May Ma v. Barber, 357 U.S. 185, 190 (1958) (“Physical detention of aliens is now the exception, not the rule, and is generally employed only as to security risks or those likely to abscond. Certainly this policy reflects the humane qualities of an enlightened civilization.” (citations omitted)). and in the mainstream, including in later years by individuals affiliated with the Nixon-created National Advisory Commission on Criminal Justice Standards—would last approximately a quarter century.121See García Hernández, supra note 36, at 55. This policy of nondetention also coincided with U.S. acceptance of refugees fleeing communist countries during the Cold War. Jonathan Simon, Refugees in a Carceral Age: The Rebirth of Immigration Prisons in the United States, 10 Pub. Culture 577, 582 (1998). The frame ended in the twilight of the civil rights movement with the arrival of Haitian immigrants fleeing the Duvalier regime who sought asylum in the United States. In response to Haitian immigrants and other nonwhite and poor immigrants from Central America and the Caribbean, the U.S. returned to detention as a method of geographic isolation, intentionally distancing immigrants from community and counsel to thwart access to the administrative process and judicial intervention.

After he ascended to power in Haiti in 1957, François “Papa Doc” Duvalier instituted an autocratic regime that stifled dissent, enforced by a mercenary guard known as the “Tonton Macoute.” In response, between 1950 and 1960, many of the island-nation’s elite and middle class left Haiti for Europe, North America, and the newly independent nations of West Africa.122 Millery Polyné, From Douglass to Duvalier: U.S. African Americans, Haiti, and Pan Americanism, 1870–1964, at 190 (2010). Those who entered the United States often entered on visas and were largely ignored by immigration officials, even when statuses lapsed. In the 1970s, however, more Haitians began fleeing the country, a significant number of whom could not afford visas or airfare. They fled, instead, by boat, arriving onto U.S. shores in Florida. In contrast to immigrants seeking safe haven from Communist regimes, who were largely paroled into the United States and afforded an opportunity to apply for asylum, Haitian asylum seekers were cast as economic migrants, subjected to a policy of detention, and issued blanket denials of their asylum claims.123 Carl Lindskoog, Detain and Punish: Haitian Refugees and the Rise of the World’s Largest Immigration Detention System 12–18 (2018). The stark difference between the treatment of Haitians and other immigrant groups led to outcry from community members and those subject to the U.S. government’s targeted detention policies. One group of Haitian asylum seekers jailed at Collier County Stockade Detention Center in Florida, identifying themselves as the “Haitian Refugees at Immokalee,” penned a letter to the INS district director, stating:

We left our native country and came here to solicit the political asylum, because of the injustices we encountered with the brutal police corps of Duvalier, the tontons-macoutes [sic]. But as you know, we have been put in jail upon our arrival in the United States. Now that we are retained here without clear knowledge of our crime, the jailers have forced us to do all kinds of heavy jobs . . . . When we try to protest, we have been punched and beaten with sticks or they even put us in those small cells (dungeon), where we cannot move.124Id. at 23.

Ultimately, through organizing by directly impacted Haitians and religious and civil rights organizations, and strategic legal challenges to the government’s Haitian detention policy, Haitians won a policy of release from detention and a hearing before an immigration judge, where they could be represented by an attorney.125Id. at 25; see also Sannon v. United States, 631 F.2d 1247, 1249–50 (5th Cir. 1980) (describing how, during litigation, the INS promulgated regulations ensuring that refugees seeking asylum in the United States would be able to raise those claims in exclusion proceedings).

Soon after, in response to protests from immigration officials regarding the government’s liberalized stance, rampant hostility and racism towards Haitian refugees, and draconian Bahamian immigration policies targeting Haitians for roundup and deportation—which U.S. government officials inferred would lead to the U.S. becoming the natural destination for those fleeing Haiti—the U.S. government instituted the Haitian Program. The Haitian Program reinstated a policy of detention for Haitian asylum seekers, denied Haitian asylum seekers access to work permits, and expedited asylum hearings for speedy denial and deportation.126 Lindskoog, supra note 123, at 26–27. The government under the Carter Administration also began circumventing judicial interventions designed to uphold the due process rights of Haitian asylum seekers by detaining them outside the jurisdiction of the Southern District of Florida at Fort Allen in Puerto Rico.127Id. at 31. Moreover, it aimed to consolidate detention and processing away from the mainland, where the incarceration of asylum seekers was a hot-button issue, and where organizers, activists, and elected officials decried U.S. violations of human rights and international law principles.128See id.

The Carter Administration never realized its aims of centralizing refugee processing outside of the forty-eight contiguous states. But the Reagan Administration would carry the torch forward, enacting stricter enforcement policies to detain and deter immigrants in response to migrations from the Caribbean and Latin America. The Administration began subjecting asylum seekers to mandatory detention and interdicting Haitians on the high seas in the early 1980s.129Kristina Shull, Reagan’s Cold War on Immigrants: Resistance and the Rise of a Detention Regime, 1981–1985, J. Am. Ethnic Hist., Winter 2021, at 5, 10; Proclamation No. 4865, 3 C.F.R. 50 (1981); Agreement Between the United States of America and Haiti, Haiti-U.S., Sept. 23, 1981, 33 U.S.T. 3559, 3559 [hereinafter U.S.-Haiti Treaty] (an agreement for “the establishment of a cooperative program of interdiction and selective return to Haiti of certain Haitian migrants and vessels involved in illegal transport of persons coming from Haiti”). Mandatory detention, aimed first at Haitian asylum seekers beginning in May 1981,130 U.S. Gen. Acct. Off., Detention Policies Affecting Haitian Nationals 5–6 (1983), []. was extended to all asylum seekers the following year, enabling the Reagan Administration to state that detention “is done even-handedly with aliens of all countries.”131 Lindskoog, supra note 123, at 64. The I.N.A. provides that where an immigration officer determines that an applicant for admission is not “clearly and beyond a doubt” entitled to be admitted, that individual “shall be detained.” 8 U.S.C. § 1225(b)(2)(A). While this provision has existed in the I.N.A. since 1952, the government applied a policy of paroling individuals through resolution of their cases prior to this sea change under the Reagan Administration. See Simon, supra note 121, at 581. Interdiction, for its part, permitted the U.S. Coast Guard to halt more than 180,000 migrants attempting to reach the United States by boat, 55 percent of whom were from Haiti.132 Lindskoog, supra note 123, at 59. The interdiction agreement between the United States and Haiti stated that the United States did not intend to return Haitian migrants whom U.S. officials determined qualified for “refugee” status. See U.S.-Haiti Treaty, supra note 129. As such, INS officers conducted “pre-screening” interviews on U.S. Coast Guard cutters to determine whether an individual had a credible fear of persecution. See Haitian Ctrs. Council, Inc. v. McNary, 969 F.2d 1326, 1330 (2d Cir. 1992), cert. granted, judgment vacated sub nom. Sale v. Haitian Ctrs. Council, Inc., 509 U.S. 912 (1993). However, the screening process “was honored in word more than deed”: of the 24,600 Haitians screened in the ten-year period between 1981 and 1991, only eleven people were brought to the U.S. for asylum hearings. See Michael Ratner, How We Closed the Guantanamo HIV Camp: The Intersection of Politics and Litigation, 11 Harv. Hum. Rts. J. 187, 190 (1998). Both of these programs reflected the Administration’s strategy of using detention as a deterrent,133 U.S. Gen. Acct. Off., supra note 130, at 3. coinciding with a new “war on crime” used by presidents and their administrations to strip courts of discretion, expand policing, and build up prisons.134See García Hernández, supra note 36, at 59–60.

As the Reagan Administration embraced civil immigration detention, hundreds of Haitian asylum seekers were detained in overcrowded facilities in states as far away as Texas, New York, and Missouri, isolated from family and community, in depressed conditions, and unable to communicate with authorities or medical personnel.135 Lindskoog, supra note 123, at 74–75. Those detained organized hunger strikes and other uprisings in response to their incarceration; those on the outside raised legal challenges. Yet, as Carl Lindskoog argues, resistance threw obstacles in the way of the government’s detention program but also “led to the development of a more resilient legal, political, and economic rationale for its existence.”136Id. at 73. Jenna Loyd and Allison Mountz theorize that, because the Reagan Administration advanced a rhetoric that detention drove enforcement, the only way to enable the federal government to fulfill its legal duties was through the expansion of detention space.137 Loyd & Mountz, supra note 37, at 102. The government then engaged in a lengthy and ongoing process of siting immigration prisons and jails. This process was informed by logistical concerns, including the existence of transportation infrastructure, as well as community responses to the construction or expansion of immigration detention.138See, e.g., id. at 98–100. The new detention centers were “contingency spaces” available in the event of mass migrations139Id. at 106.—they were sold politically and rhetorically as a furtherance of the agency’s “law enforcement” mission, which enabled the government to collapse border logic with interior enforcement.140Id. at 102. By creating contingency spaces, the government continued its embrace of remote locations to isolate immigrants of color away from community and resistance.

By April 1986, the federal government had opened its first high-capacity immigration detention center in Oakdale, Louisiana, roughly 200 miles from the nearest population hubs in Houston and New Orleans.141 Lindskoog, supra note 123, at 80–81. Geographically isolated, Oakdale became a transfer location, where people from as far away as California and New York were transported for deportation. Oakdale, then, “epitomized the production of remoteness: designed to be far away from attorneys, and thereby to ‘increase the speed and number of deportations.’ 142 Loyd & Mountz, supra note 37, at 110.

Under the Bush Administration, the government forged ahead with its detention blueprint. It targeted asylum seekers from Central America and built infrastructure for mandatory detention in the Lower Rio Grande Valley. It continued the interdiction program established by President Reagan, summarily returning Haitian refugees fleeing a military coup regime rather than screening them for a credible fear of persecution.143Id. at 81–82. It transferred other Haitian refugees to the U.S. naval base at Guantanamo Bay, Cuba, “where they were detained behind razor-barbed wire in makeshift military camps without due process rights.”144Harold Hongju Koh & Michael J. Wishnie, The Story of Sale v. Haitian Centers Council: Guantánamo and Refoulement, in Human Rights Advocacy Stories 385, 389 (Deena R. Hurwitz & Margaret L. Satterthwaite eds., 2009). These policies served to limit the rights of those in detention by stripping them of procedural safeguards, cutting off their ability to communicate with the outside world, and impeding their access to legal counsel and judicial review.145See Loyd & Mountz, supra note 37, at 109–10. The U.S. government has again engaged in the repatriation of Haitian asylum seekers through expulsions pursuant to Section 265 of Title 42, a public health provision invoked during the COVID-19 pandemic, which prohibits the “introduction” of an individual where “there is serious danger of the introduction of [a communicable] disease into the United States.” See Am. Immigr. Council, A Guide to Title 42 Expulsions at the Border 2–3 (2022), [].

This patchwork of detention would have lasting ramifications on immigrants’ and communities’ abilities to challenge detention through law and organizing.146See infra Section II.B. Thus, from its earliest instances, immigration detention has been characterized by its desire to geographically isolate and exclude—to eliminate nonwhite immigrants’ channels of communication and connection to community and, ultimately, to the courts and due process.

II. The Rise of the Modern Immigration Detention System and the Role of Transfers as a Tool of Social Control

As noted in Part I, the rise of immigration detention has coincided with the concentration of immigration jails and prisons in remote locations.147These locations are often hundreds of miles away from urban centers, where there are higher concentrations of immigration attorneys, and are known for high parole and bond denial rates. For example, based on an analysis of the geographic distribution of attorneys registered with the American Immigration Lawyers Association (AILA) and those with noted expertise in deportation offenses, the ACLU found that, “on average, there are four times as many immigration attorneys within a 100-mile radius of people detained” in facilities opened before January 2017 than after that date. Eunice Hyunhye Cho, Tara Tidwell Cullen & Clara Long, Justice-Free Zones: U.S. Immigration Detention Under the Trump Administration 20 (2020), []. It has also coincided with an increase in transfers, facilitated by a transportation infrastructure that connects ICE’s patchwork system of jails and prisons. Transfer supports ICE’s ability to coordinate bed space and effectuate deportations. It also serves to divorce immigrants facing deportation from community supports integral to vindicating their due process rights, and from organizing efforts for release and abolition.148See Loyd & Mountz, supra note 37, at 188–89.

This Part documents the rise of the modern detention state, the federal government’s seemingly unfettered discretion to transfer immigrants in its custody to jails and prisons far from their homes and communities, and the attendant harms of transfer. What is perhaps more prominent in the modern detention state post-9/11—though not unique to it—is its ability to seize individuals who have long settled in the United States through interior enforcement. Enforcement, traditionally located “at the territorial margins of the state,” has ossified into a hybrid inward- and outward-looking mechanism of immigration control.149Mathew Coleman & Austin Kocher, Detention, Deportation, Devolution, and Immigrant Incapacitation in the US, Post 9/11, 177 Geographical J. 228, 228, 230–31 (2011) (describing a shift in immigration enforcement, after passage of the 1996 laws and the events of September 11, 2001, to a hybrid of both interior and border enforcement). Individuals embedded in the social, cultural, and political fabric of communities are plucked from their homes and transferred—without notice to their attorney of record, their loved ones, or the community at large—hindering access to zealous representation and inside-outside organizing strategies that are essential to fighting for one’s liberty and bringing an end to immigration detention more generally.

A. The Rise of the Modern Detention State: Legislative Frameworks and the Growth of Detention

As detailed in Section I.A, the Supreme Court sanctioned immigration detention as “a constitutionally valid aspect of the deportation process” in dicta in Wong Wing.150Demore v. Kim, 538 U.S. 510, 523 (2003) (citing Wong Wing v. United States, 163 U.S. 228, 235 (1896)). It then affirmed its stance in Carlson v. Landon in 1952.151Carlson v. Landon, 342 U.S. 524 (1952). Congress additionally provided for discretionary detention pending removal proceedings when it passed the Immigration and Nationality Act in 1952, which is now codified at 8 U.S.C. § 1226(a). Yet, despite the steady construction of the legal infrastructure for immigration detention, for decades the attorney general tended to exercise this discretion in favor of liberty during the pendency of an immigrant’s deportation proceedings. That presumption was narrowed in 1988, when Congress passed the first mandatory detention immigration law and created a small category of individuals whom the attorney general was required to take into custody without the possibility of release on bond or parole.1528 U.S.C. § 1101(a)(43) (1988) (adding the term “aggravated felony” to the statute, defined as murder and certain drug trafficking and firearms trafficking offenses); 8 U.S.C. § 1252(a)(2) (1988) (providing that “[t]he Attorney General shall take into custody any alien convicted of an aggravated felony upon completion of the alien’s sentence for such conviction”). This legislation reflected Congress’s renewed interest in targeting immigrants with criminal convictions, a phenomenon that paralleled harsh legislation at the federal and state levels increasing sentencing for minor offenses and sowed the seeds for today’s carceral infrastructure, which disproportionately targets people of color and the poor.

Congress’s interest would be long-lasting. In 1994, Congress amended the 1988 statute, increasing the number of immigrants subject to mandatory detention without the possibility of bond or parole. Then, in 1996, Congress enacted two laws—the Antiterrorism and Effective Death Penalty Act153Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214 (codified in scattered sections of 8, 18, 22, 28, 40, and 42 U.S.C.). (AEDPA) and the Illegal Immigration Reform & Immigrant Responsibility Act154Illegal Immigration Reform & Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, div. C, 110 Stat. 3009–46 (codified in scattered sections of 8 and 18 U.S.C.). (IIRIRA)—which made sweeping changes to the nation’s immigration laws, including through dramatic limitations on immigrants’ substantive rights155The 1996 laws dramatically expanded criminal grounds for deportation, including for lawful permanent residents; created expedited removal provisions, making it easier to deport people; and limited the availability of judicial review. See Nancy Morawetz, Understanding the Impact of the 1996 Deportation Laws and the Limited Scope of Proposed Reforms, 113 Harv. L. Rev. 1936, 1936, 1943–49 (2000) (outlining the effects of the 1996 laws on lawful permanent residents); see also Coleman & Kocher, supra note 149, at 228, 230–31. and, as relevant here, increased instances in which immigrants are subject to mandatory, no-bond detention. Together, these laws embedded a broad detention mandate into the immigration laws.

While Congress did not alter the discretionary detention statute through IIRIRA, the INS implemented new regulations, creating a presumption of detention and placing the burden on an arrested immigrant to demonstrate that their release would not pose a bail risk or danger.1568 C.F.R. § 236.1(c)(8) (2016) (“Any officer authorized to issue a warrant of arrest may, in the officer’s discretion, release an alien not described in section 236(c)(1) of the Act, under the conditions at section 236(a)(2) and (3) of the Act; provided that the alien must demonstrate to the satisfaction of the officer that such release would not pose a danger to property or persons, and that the alien is likely to appear for any future proceeding.”); see also Alina Das, Immigration Detention: Information Gaps and Institutional Barriers to Reform, 80 U. Chi. L. Rev. 137, 156 (2013) (describing how the INS justified its rule promulgation by citing to Congress’s budget enhancements and legislation, referring to them as a “mandate” to increase detention to ensure removal). The Board of Immigration Appeals quickly adopted this burden shift in the context of bond proceedings conducted by immigration judges.157See In re Adeniji, 22 I. & N. Dec. *1102, *1112 (B.I.A. 1999) (interim decision) (“Pursuant to 8 C.F.R. § 236.1(c)(8), the respondent must demonstrate that his release would not pose a danger to property or persons, and that he is likely to appear for any future proceedings.”), abrogated by Hernandez-Lara v. Lyons, 10 F.4th 19 (1st Cir. 2021); In re Guerra, 24 I. & N. Dec. *37, *40 (B.I.A. 2006) (interim decision) (“The burden is on the alien to show to the satisfaction of the Immigration Judge that he or she merits release on bond.”), abrogated by Hernandez-Lara v. Lyons, 10 F.4th 19 (1st Cir. 2021). These regulatory and procedural changes have subjected countless immigrants to detention—even in the midst of ongoing immigration proceedings—unless they can show to the satisfaction of an immigration judge that they merit release on bond.158See, e.g., In re Guerra, 24 I. & N. Dec. at 40.

The combined effect of these laws and policies has been an overall increase in immigration detention. In 1985, the INS went from detaining 2,000 in a given day to 6,600 in 1995.159Philip L. Torrey, Rethinking Immigration’s Mandatory Detention Regime: Politics, Profit, and the Meaning of “Custody, 48 U. Mich. J.L. Reform 879, 896 (2015); see also Alison Siskin, Cong. Rsch. Serv., RL32369, Immigration-Related Detention: Current Legislative Issues 11–12 (2004). By the turn of the twenty-first century, detention was growing steadily, with the largest increase occurring from 1997 to 1998, when the 1996 laws became enforceable.160 Siskin, supra note 159, at 11–12. And when the Supreme Court decided Rumsfeld v. Padilla in 2004, the average daily detention population was 22,812.161Id. But this was only the beginning.

Immigration detention has continued climbing steadily with congressional approval. In 2004, Congress passed the Intelligence Reform and Terrorism Prevention Act, which required the Department of Homeland Security (DHS) to increase detention bed space by a minimum of 8,000 beds in each of the fiscal years from 2006 to 2010.162Intelligence Reform and Terrorism Prevention Act of 2004, Pub. L. No. 108-458, § 5204, 118 Stat. 3638, 373435. In 2009, Congress passed the DHS Appropriations Act of 2010, with language mandating that DHS maintain no fewer than 33,400 detention beds through September 30, 2010, and allocating $2,545,180,000 for detention and removal operations.163Department of Homeland Security Appropriations Act of 2010, Pub. L. No. 111-83, tit. 2, 123 Stat. 2142, 214456. The bed mandate remained in effect for years, with detention numbers ranging from 30,000 to upwards of 40,000 during the Obama Administration.164For a detailed account of congressional actions relating to the bed mandates, see Nat’l Immigrant Just. Ctr., Immigration Detention Bed Quota Timeline (2017), []. During this same period, internal enforcement expanded through programs like Secure Communities, which linked local policing and immigration enforcement, funneling more people of color into the immigration detention system.165Adam B. Cox & Thomas J. Miles, Policing Immigration, 80 U. Chi. L. Rev. 87, 11215 (2013) (finding that the “Secure Communities” immigration enforcement program focused on heavily Hispanic communities).

While the bed mandate ended during funding negotiations in May 2017,166Detention Quotas, Det. Watch Network, []. detention numbers continued to rise under the Trump Administration, with DHS at times detaining upwards of 56,000 individuals per day167 Cho et al., supra note 147, at 4, 14. in a network of over 220 jails and prisons,168Id. at 14; see also U.S. Gov’t Accountability Off., Immigration Detention: Actions Needed to Improve Planning, Documentation, and Oversight of Detention Facility Contracts 11 (2021), [] (noting that ICE had contracts or agreements in place with a total of 233 over-seventy-two-hour detention facilities). 40 of which were operationalized for immigration detention during the Trump presidency. The twenty largest of these facilities are located predominantly in the Southern states, with seven in Louisiana alone.169 Cho et al., supra note 147, at 14–15. Additionally, Louisiana has the second-highest incarceration rate of any state in the United States, per the Bureau of Justice Statistics, incarcerating 564 per 100,000 people. E. Ann Carson, Bureau of Just. Stat., U.S. Dep’t of Just., Prisoners in 2021 – Statistical Tables tbls.15–16 (2022), []. These jails and prisons are largely operated through contracts and agreements, with space in about 57 percent of these facilities obtained through intergovernmental service agreements (or IGSAs).170 U.S. Gov’t Accountability Off., supra note 168, at 11. Notably, ICE can further modify agreements with IGSA-holders, such that a local or county jail can subcontract detention services to a private company, an arrangement that allows the IGSA-holder to collect various fees from the private contractor. Id. at 15–17. In recent years, ICE has managed to expand immigration detention through overspending; by exceeding its budget, the agency justifies later requests to Congress for budget increases, fueling expansion.171See Cho et al., supra note 147, at 16. Since 2004, ICE’s budget has more than doubled, from $3.7 billion to $8.3 billion, largely in aid of the agency’s detention capabilities.172 Am. Immigr. Council, The Cost of Immigration Enforcement and Border Security 3 fig.2 (2021), [].

B. The Role of Transfers Within the Modern Detention System

Although individuals are typically arrested close to where they reside, they are often transferred hundreds, if not thousands, of miles away from home, loved ones, and community. This is not a new phenomenon.173See, e.g., U.S. ex rel. Circella v. Sahli, 216 F.2d 33, 36 (7th Cir. 1954) (arresting relator and immediately transferring him from Cook County, Illinois to Hammond, Indiana, and later New York City, for imminent deportation to Italy, despite representations to relator’s counsel that “it was not the purpose of the [Immigration and Naturalization] Service to arrest the relator for immediate deportation”); see also Yi v. Maugans, 24 F.3d 500 (3d Cir. 1994) (transferring approximately 120 Chinese nationals who arrived by ship at New York harbor to York County Prison in Pennsylvania); Nwankwo v. Reno, 828 F. Supp. 171 (E.D.N.Y. 1993) (transferring petitioner from New York State to I.N.S. custody at the Federal Deportation Center in Oakdale, Louisiana); Michael v. Immigr. Naturalization Servs., 870 F. Supp. 44 (S.D.N.Y. 1994) (same); Mojica v. Reno, 970 F. Supp. 130 (E.D.N.Y. 1997) (same), aff’d in part, dismissed in part sub nom. Henderson v. Immigr. Naturalization Servs., 157 F.3d 106 (2d Cir. 1998). According to a study conducted by Human Rights Watch, in a twelve-year span from 1998 to 2010, of the 2,271,911 noncitizens in immigration detention, federal immigration officials transferred 1,159,568 of them (or 40 percent) at least one time.174 Alison Parker, Hum. Rts. Watch, A Costly Move: Far and Frequent Transfers Impede Hearings for Immigrant Detainees in the United States 17 (2011), []. Over 46 percent of individuals in immigration detention were transferred at least twice, 16.7 percent were transferred at least three times, and 3,400 individuals were transferred ten times or more.175Id. Between 2004 and 2009, the number of transfers of immigrants in detention tripled. In 2004, ICE effectuated 124,899 transfers; in 2009, it conducted 405,544.176Id. at 18 tbl.1. By 2015, about 54 percent of individuals in immigration detention experienced at least one transfer.177Emily Ryo & Ian Peacock, A National Study of Immigration Detention in the United States, 92 S. Cal. L. Rev. 1, 39 (2018) (finding that “about 27% of these individuals experienced one transfer, about 15% experienced two transfers, and about 12% experienced three or more transfers”). On average, an individual was transferred almost 400 miles away,178See Parker, supra note 174, at 1. with a frequent transfer route between Pennsylvania and Texas that is over 1,500 miles.179Id. Human Rights Watch notes that other frequent routes include North Carolina to Georgia (to Stewart Detention Center), Pennsylvania to Louisiana, and Southern California to Arizona (to Eloy Federal Contract Facility). Id. at 19–20. Individuals subject to transfer spent, on average, three times longer in immigration detention than those who were not transferred.180Id. at 2, 14.

ICE’s transfer data has grown opaque in recent years.181Since Human Rights Watch released its study on transfer, ICE has stopped publishing transfer data. Elena Hodges, Building Power: Charting Recent Victories in the Movement to End Immigration Detention in the United States 106 n.488 (2022), []. While the agency provides end-of-the-month snapshots of detention, these numbers do not shed light on transfers, and ICE has made their databases off-limits to FOIA requestors. See Dennis Kuo, Noelle Smart, Zachary Lawrence & Adam Garcia, Vera Inst. of Just., The Hidden Curve: Estimating the Spread of COVID-19 Among People in ICE Detention 8–9 (2020), [] (“ICE continues to conceal its transfer activity from the public.”). Indeed, for its report on the spread of COVID-19 within ICE detention facilities, the Vera Institute of Justice created a compartmental epidemiology model based on fiscal year 2016 data, the most recent full fiscal year data available. Dennis Kuo, Noelle Smart, Zachary Lawrence & Adam Garcia, Vera Inst. Just., The Hidden Curve: Estimating the Spread of COVID-19 Among People in ICE Detention Technical Appendix (2020), []. However, both modeling and anecdotal evidence demonstrate that the practice continues unabated and is likely increasing.182Freedom for Immigrants notes that so-called “shuffle flights,” which move people in the U.S. to and from detention centers, increased by 94 percent between 2020 and 2022. See Freedom for Immigrants, Trafficked & Tortured: Mapping ICE Transfers 11 (2023), []. For example, in 2019, when ICE decided to recast Karnes County Residential Center—a private immigration prison outside of San Antonio, Texas run by the GEO Group—from a women’s detention center to a family detention center, 856 women were transferred from Karnes.183Isabela Dias, When ICE Emptied Out an All-Women Detention Center in Texas, Chaos Ensued, Tex. Observer (Nov. 6, 2019, 1:11 PM), []. Women—many of whom had fled gender-based violence in their home countries—were transferred to jails and prisons in other parts of Texas, Mississippi, and Louisiana, their locations unknown to counsel and their court dates rescheduled or canceled, thereby prolonging detention.184Id. In the midst of transfer, lawyers struggled to identify where to file bond requests. Women transferred to detention centers in Mississippi (where there is no immigration court) were forced to fight their cases via televideo before an immigration judge in New York State, represented by a lawyer in Texas.185Id.

As states and municipalities organize for the just closure of immigration jails and prisons,186California became the first state to pass legislation halting immigration detention growth, with the passage of the Dignity Not Detention Act, which is comprised of provisions from two bills, SB 29 and AB 103. SB 29 prohibits local government entities from entering into new jail contracts with private prison corporations and bars those with existing contracts from expanding bed space. AB 103 prevents local government entities from entering into new jail contracts with public entities and prohibits local governments from expanding existing contracts with public entities. See Press Release, Immigrant Legal Res. Ctr., California’s Passage of Dignity Not Detention Act to Bring Needed Accountability to Private Immigration Prisons (Oct. 5, 2017),’s-passage-dignity-not-detention-act-bring-needed-accountability-private-immigration []; see also Freedom for Immigrants, Guide to Dignity Not Detention, []. Since 2017, three more states have passed even more stringent legislation in an effort to end immigration detention: Washington in April 2021 (HB 1090), Illinois in August 2021 (Illinois Way Forward Act), and New Jersey in August 2021 (A-5207/S-3361). See Press Release, Det. Watch Network, Bill Signed into Law in Washington State Will Phase Out Private Prisons. Groups Call for Immediate Closure of the Notorious Northwest Detention Center (Apr. 15, 2021), []; Press Release, State of Ill., Gov. Pritzker Signs Legislation Further Establishing Illinois as the Most Welcoming State in the Nation (Aug. 2, 2021), []; Press Release, State of N.J., Governor Murphy Takes Action on Legislation (Aug. 20, 2021), [] (describing how Governor Murphy signed into law A-5207/S-3361, which “[p]rohibits State and local entities and private detention facilities from entering into agreement to detain noncitizens”). These laws were a response to sustained, coordinated activism and organizing by directly impacted individuals, their loved ones, and community groups. However, they have been subject to challenge. See, e.g., CoreCivic v. Murphy, No. 23-cv-967, 2023 WL 5556025 (D.N.J. Aug. 29, 2023) (holding A-5207/S-3361 unconstitutional as applied to CoreCivic’s contract to operate the Elizabeth Detention Center to detain immigrants on ICE’s behalf), appeal docketed, No. 23-2598 (3d Cir. Sept. 7, 2023). For a description of these movements, see generally Hodges, supra note 181. Additional legislation aiming to end immigration detention has been introduced in Maryland and New York. Id. ICE continues to double down in its commitment to transferring individuals hundreds, if not thousands, of miles away.187See Amba Guerguerian, Inside ICE’s Immigrant Transfer Roulette, Indypendent (July 7, 2021), []. When Bergen County Jail in New Jersey—where most recent available data shows that over 86 percent of individuals in detention are from Latin America and the Caribbean188Immigration and Customs Enforcement Detention: ICE Data Snapshots, up to July 2019, Transactional Recs. Access Clearinghouse, [] [hereinafter ICE Data Snapshots].—severed its contract with ICE in the wake of new legislation banning future ICE detention contracts in the state, the agency transferred those in detention to a federal prison 300 miles away in Batavia, New York.189Sophie Nieto-Munoz, Bergen County Jail Transfers Last ICE Detainees to Facility 300 Miles Away, N.J. Monitor (Nov. 17, 2021, 6:55 AM), []. When Illinois passed legislation requiring local officials to end their ICE contracts, some individuals detained in McHenry County Jail in Woodstock—where over 77 percent of individuals in detention are from Latin America and the Caribbean190See ICE Data Snapshots, supra note 188.—and at the Jerome Combs Detention Center in Kankakee—where 83 percent of individuals in detention are from Latin America and the Caribbean191See id.—were transferred to Indiana, Oklahoma, and Texas.192Elvia Malagón, Immigration Detention Ends in Illinois After ICE Transfers Those Awaiting Deportation to Out-of-State Jails, Chi. Sun-Times (Feb. 15, 2022, 6:28 PM), []. And when advocates alleged racist and abusive treatment of individuals detained at Orange County Jail in New York193Matt Katz, Allegations Surface About Treatment of ICE Detainees in Orange County, Gothamist (Dec. 2, 2021), [].—where over 80 percent of individuals in detention are from Latin America and the Caribbean194See ICE Data Snapshots, supra note 188.—ICE transferred dozens of detained immigrants without warning to facilities throughout the country.195Matt Katz, NY Area’s ICE Detention Facilities Are Emptying, with Local Immigrants Moved Across the Country, Gothamist (Jul. 27, 2022), []. Sixty-one immigrants were transferred over 1,300 miles away to the Adams County Detention Center in Natchez, Mississippi, a facility the national American Civil Liberties Union and the ACLU of Mississippi asked President Biden to shut down in light of allegations of medical neglect, physical abuse, and indefinite detention.196Press Release, Jerry Nadler, Congressman, House of Representatives, Rep. Nadler Co-Leads NYC Delegation Letter to DHS and ICE Chiefs on Recent Transfer of Immigrant Detainees (Aug. 2, 2022), []; Ashton Pittman, Close Adams County ICE Facility, Mississippi ACLU Tells Biden, Alleging ‘Torment, Miss. Free Press (Apr. 28, 2021), []. The Orange County undersheriff stated publicly that his agency asked ICE to reduce the jail population in part because of “the amount of time the county law department ha[d] spent addressing complaints filed by advocates over the treatment of detainees.”197Katz, supra note 195.

These transfers continue, despite their facilitation of the spread of infectious disease198See Yeganeh Torbati, Dara Lind & Jack Gillum, In a 10-Day Span, ICE Flew This Detainee Across the Country—Nine Times, ProPublica (Mar. 27, 2020, 10:33AM), [] (detailing ICE’s transfer of Sirous Asgari, a fifty-nine-year-old father and materials science and engineering professor, around the country on nine different flights as the COVID-19 pandemic raged in March 2020). ProPublica’s investigation found flight records showing that ICE operated at least sixteen flights between March 16, 2020, when U.S. officials urged travelers to refrain from taking unnecessary trips and avoid gathering in groups of more than ten, and March 27, 2020, when the article was published—largely on Swift Air (owned by iAero Airways) and World Atlantic Airlines, neither of which has any contracts appearing in federal spending databases that would evidence how much transfer operations cost taxpayers. Id. and other implications on public health.199Carlos Franco-Paredes, an infectious disease doctor, stated, “If you’re moving people, particularly from an area where there is an ongoing outbreak, even though you sequester them for two weeks or so, there is contact with people.” He added, “You’re basically spreading the problems.” See Mica Rosenberg, Kristina Cooke & Reade Levinson, U.S. Immigration Officials Spread Coronavirus with Detainee Transfers, Reuters (July 17, 2020, 11:45 AM), []. During the COVID-19 pandemic, for example, in a three-month period from April to June 2020 when COVID-19 cases were increasing exponentially in several states, immigration court data revealed at least 268 transfers between detention centers. Half of these transfers involved moving individuals from centers with known COVID-19 cases to centers with none, or from centers without any known cases to those with documented exposure.200Rosenberg et al., supra note 199. One of these early transfers caused a “super-spreader” event at a detention center in Farmville, Virginia. A transfer of seventy-four individuals from detention facilities in Florida and Arizona on June 2, 2020—more than half of whom tested positive for COVID-19—led to an outbreak, with at least 315 total cases.201Id. Individuals being transferred to Krome Processing Center in Florida were not medically evaluated upon arrival, and attorneys of record were not notified of their transfer, even after it occurred.202See Torbati et al, supra note 198 (“[O]n March 18, an ICE official told Mary Yanik, a lawyer with the New Orleans Workers’ Center for Racial Justice, that a detainee’s scheduled March 19 transfer had been postponed because they were canceling ‘all movement of anybody for at least another week’ . . . . On March 25, the detainee was transferred and Yanik was not notified.”). In California, an attorney reported a client’s account of his transfer from Otay Mesa Detention Center203The first recorded COVID-19-related death of an individual detained in ICE custody—a fifty-seven-year-old man from El Salvador—was recorded at Otay Mesa Detention Facility. Maria Santana & Catherine E. Shoichet, First ICE Detainee Dies from Coronavirus, CNN (May 6, 2020, 10:09 PM), []. to Adelanto: after sleeping eight to a cell at Otay Mesa, he was transferred by bus for an almost three-hour trip to Adelanto along with thirty other men, including two who were visibly sick and coughing. They were separated only after their arrival at Adelanto.204Declaration of Keren Zwick in Support of Motion for Preliminary Injunction & Class Certification ¶¶ 28–29, Fraihat v. U.S. Immigr. & Customs Enf’t, 445 F. Supp. 3d 709 (C.D. Cal. 2020) (No. 19-cv-01546-JGB), vacated, 2022 WL 20212706 (C.D. Cal. Sept. 16, 2022). The total number of transfers ICE has conducted in recent years is unknown.205The last aggregate transfer data released by ICE covers the period from October 2014 to September 2015. Notes of Telephone Interview by Elena Hodges, N.Y.U. Sch. of L., with Susan Long, Co-Director, Transactional Recs. Access Clearinghouse (June 4, 2021) (on file with author); see also Detention Facility Reports: Transfers, Transactional Recs. Access Clearinghouse (Sept. 2015), []. However, modeling conducted by the Vera Institute of Justice based on last-published transfer activity shows that across 500 simulations and by the end of a sixty-day period, “ICE would conduct a median cumulative 1,744 transfers of people with active COVID-19 cases to other detention facilities,” including repeat transfers of individuals across state lines and to different regions of the country.206 Kuo et al., supra note 181. ICE continued these transfers, despite early warnings by its own contracted medical subject-matter experts that the extensive use of transfers of individuals (often without symptoms) through ICE’s network of immigration jails and prisons could “rapidly disseminate the virus throughout the entire system with devastating consequences to public health.”207Letter from Scott A. Allen, Professor Emeritus, Clinical Med., U.C. Riverside Sch. of Med., & Josiah “Jody” Rich, Professor of Med. & Epidemiology, Brown Univ., to Bennie Thompson, Chairman, House Comm. on Homeland Sec., et al. (Mar. 19, 2020), []. These experts stated that ICE’s undeterred reliance on detention—a congregate setting, where social distancing is “an oxymoron”—created a grave public health risk in light of detention centers’ reliance on local hospital systems.208Id. at 3.

Take Yousef Kane,209“Yousef Kane” is a pseudonym to protect Mr. Kane’s identity. for example. In the winter of 2018, ICE officers arrested Mr. Kane at his home in front of his wife and U.S.-citizen children.210ICE officers arrested Mr. Kane despite the fact that he was on an Order of Supervision with which he had diligently complied for six years. Second Amended Petition for Writ of Habeas Corpus and Complaint for Declaratory and Injunctive Relief (Mar. 1, 2019) (on file with author). Mr. Kane, an asylum seeker from West Africa and a Bronx resident for nearly fifteen years, was processed at ICE’s New York field office in Manhattan. Officers took him to a jail in New Jersey for four days before transferring him to a detention center in Louisiana,211Id. over 1,000 miles away from his home, loved ones, faith community, and counsel.212Id.

Mr. Kane’s immigration lawyers filed a habeas petition and motion for temporary restraining order (TRO) in the Southern District of New York—which maintained jurisdiction over both ICE’s New York Headquarters and Mr. Kane’s home—requesting that the court halt any deportation and transfer.213Id. The court granted the TRO.214Id. Two days later, ICE nevertheless transferred Mr. Kane from Louisiana to Texas, Texas to Arizona, Arizona back to Louisiana, Louisiana to New York, and New York to New Jersey: five transfers in less than three weeks.215U.S. Immigr. & Customs Enf’t, Freedom of Information Act Request Response (May 8, 2019) (on file with author). Then, ICE moved to dismiss or transfer Mr. Kane’s habeas petition to the Western District of Louisiana.216Second Amended Petition for Writ of Habeas Corpus and Complaint for Declaratory and Injunctive Relief, supra note 210. The agency argued that Mr. Kane was in Louisiana when the habeas was filed, and that for “core” challenges to present physical confinement, a habeas petitioner must name his warden as respondent and file his petition in the district in which he is confined.217Id.

ICE asserts that anyone in its custody is subject to transfer at any time and without notice. It cites to 8 U.S.C. § 1231(g)(1) as the source of its transfer authority. The subsection provides:

The Attorney General shall arrange for appropriate places of detention for aliens detained pending removal or a decision on removal. When United States Government facilities are unavailable or facilities adapted or suitably located for detention are unavailable for rental, the Attorney General may expend . . . amounts necessary to acquire land and to acquire, build, remodel, repair, and operate facilities . . . necessary for detention.2188 U.S.C. § 1231(g)(1).

While the plain language of 8 U.S.C. § 1231(g)(1) contemplates the authority to site immigration jails and detention facilities, rather than to transfer individuals subject to immigration custody, ICE’s authority to perform the latter has been upheld by many courts.219Calla-Collado v. Att’y Gen. of the U.S., 663 F.3d 680, 685 (3d Cir. 2011) (holding that a noncitizen does not have the right to be detained in a location where counsel and evidence would be the most effective); Van Dinh v. Reno, 197 F.3d 427, 433 (10th Cir. 1999) (finding that the attorney general’s discretionary transfer authority to transfer noncitizens “from one locale to another” arises from 8 U.S.C. § 1231(g)). But see Aguilar v. U.S. Immigr. & Customs Enf’t Div. of the Dep’t of Homeland Sec., 510 F.3d 1, 20 (1st Cir. 2007) (finding that 8 U.S.C. § 1231(g) “fails to ‘specify’ that individual transfer decisions are in the Attorney General’s discretion”). ICE’s perception of its transfer authority is further reflected in its policies and detention standards, which emphasize that transfer decisions are made at the discretion of the ICE field office director or their designee and designed to be effectuated without notice or process.220The agency has further asserted that “[t]he INA contains no language limiting ICE’s ability to move detainees from one facility to another.” Alison Parker, Hum. Rts. Watch, Locked Up Far Away: The Transfer of Immigrants to Remote Detention Centers in the United States 20 (David Fathi et al. eds., 2009), [] (quoting Letter from Susan M. Cullen, Dir. of Pol’y, U.S. Immigr. & Customs Enf’t, to Hum. Rts. Watch (Aug. 11, 2008) (on file with Human Rights Watch)). It has additionally stated that “[the agency] transfer[s] where beds are available. It’s out of operational necessity.” Id. (quoting Human Rights Watch Interview with Tae Johnson, Acting Unit Chief, Det. Compliance Unit, Off. of Det. & Removal Operations, in Washington, D.C. (May 12, 2008)). For example, Section 7.4 of ICE’s Performance-Based National Detention Standards (PBNDS), as revised in December 2016, which governs the transfer of individuals detained in ICE custody, does not require that individuals subject to transfer be contacted until “immediately prior to transport,”221 Immigr. & Customs Enf’t, Performance-Based National Detention Standards 2011 § 7.4(II)(3), 7.4(V)(b)(2) (rev. 2016), [] [hereinafter 2011 ICE Standards]. nor does it require that the legal representative of record be notified until after transfer has occurred.222Id. § 7.4(II)(2) (“The legal representative-of-record shall be notified as soon as practicable, but no later than 24 hours after the detainee is transferred.” (emphasis added)); see also id. § 7.4(V)(A)(2). No provision directs ICE to contact loved ones, family members, or communities prior to transfer. Indeed, Section 7.4 expressly prohibits individuals subject to transfer from (1) learning of the specific plans and time schedules for transfer, (2) making or receiving telephone calls until they reach the destination facility, and (3) having contact with any other individual in immigration detention in the general population prior to reaching the destination.223Id. § 7.4(V)(B)(2)(a)(1)–(2); see also U.S. Immigr. & Customs Enf’t, National Detention Standards for Non-Dedicated Facilities § 7.2(A)(1) (rev. 2019), [] (“For security purposes, specific plans and time schedules shall never be discussed with the detainee involved. The detainee shall not be notified of the transfer until immediately prior to leaving the facility.”). Individuals subject to transfer are not permitted to make a phone call until they are admitted to the receiving facility.224 2011 ICE Standards, supra note 221, § 7.4(V)(B)(2)(c)(3), 7.4(V)(F)(1). These policies are expressly designed to shroud the agency’s detention and transfer decisions in secrecy, purportedly for “security” reasons.225Id. § 7.4(I).

Even policies designed to limit transfers are riddled with carveouts for discretionary decisions that render the policies near-nullities. Policy 11022.1 on “Detainee Transfers” was expressly written to minimize transfers of immigrants in detention outside the area of responsibility of the field office director.226 John Morton, U.S. Immigr. & Customs Enf’t, Policy 11022.1 § 1 (2012), []. The policy expressly excludes transfers to facilities that are authorized to detain individuals for seventy-two hours or less, hold rooms, U.S. Customs and Border Protection Border Patrol stations, or ports of entry, as well as transfers to staging areas or facilities for the purpose of deportation of individuals with a final order of removal. Id. § 3.2. However, it only establishes responsibilities and procedures for ICE staff and “does not govern contract staff,” who are referred to their respective vendor agreements.227Id. § 1. The policy provides that ICE supervisory immigration officers “will not” transfer an individual outside of the area of responsibility when there is documentary evidence of immediate family or an attorney of record in the area of responsibility, when there are pending or ongoing removal proceedings, or where an individual has been granted bond or has been scheduled for a bond hearing.228Id. § 5.2. A field office director or his or her designee, however, may effectuate a transfer when “deemed necessary.”229Id. This section provides that a transfer may be deemed necessary by a field office director or their designee for any of the following reasons:

a) To provide appropriate medical or mental health care to the detainee. b) To fulfill an approved transfer request by the detainee. c) For the safety and security of the detainee, other detainees, detention personnel or any ICE employee. d) At ICE’s discretion, for the convenience of the agency when the venue of EOIR proceedings is different than the venue in which the alien is detained. e) To transfer to a more appropriate detention facility based on the detainee’s individual circumstances and risk factors. f) Termination of facility use due to failure to meet ICE detention standards, lack of sufficient use of the facility by ICE, or emergent situations. [or] g) To relieve or prevent facility overcrowding; in such cases, efforts should first be made to identify for transfer those detainees who do not meet any of the criteria listed in section 5.2(1) [regarding instances in which transfer is inappropriate].

Notably, a field office director may deem transfer necessary based on an individual’s “circumstances and risk factors,”230Id. And while transfers are recorded in computerized databases in the criminal legal system, requiring that the reason for transfer and an individual’s eligibility for parole be documented in a centralized database, there is no such requirement in the immigration context. Indeed, federal immigration agencies are not required to track the reason for transfer or eligibility for bond. Parker, supra note 220, at 24 (citing U.S. Immigr. & Customs Enf’t, 2008 Operations Manual ICE Performance-Based National Detention Standards, pt. 7, ch. 41, at 5–11, 15 (2008), []). categories so broad as to render the field office director’s discretion near-infinite.

The costs of transfer for individuals in detention, their loved ones, and their communities are immeasurable.231See Ryo & Peacock, supra note 177, at 5 (“Transfers warrant a special scrutiny because they can substantially hinder access to legal representation, sever family ties and community support, and separate detainees from the evidence needed in their court proceedings.”). Individuals in detention often lose access to pro bono counsel, as distance makes representation difficult, if not impossible, and attorneys withdraw from cases.232No public defender programs provide representation to low-income individuals in immigration detention seeking their freedom through habeas actions. Many immigrants depend on the availability of nonprofit organizations that have limited resources and are unable to meet all of their habeas needs. Increasing the inconvenience of the forum drives down representation in a pro bono environment that is already resource-strained. See Declaration of Andrea Saenz at 3, Sajous v. Decker, No. 18-cv-02447-AJN, 2018 WL 2357266 (S.D.N.Y. Apr. 5, 2018) (noting that litigation of individual habeas petitions on behalf of clients is done “on top of . . . contractual obligations to carry high caseloads of . . . removal defense matters,” and that “[i]t is not yet clear whether [Brooklyn Defender Services] will have the resources to litigate individual habeas petitions on behalf of all or even a majority of our clients.”); Declaration of Sarah Deri Oshiro at 3, Sajous v. Decker, No. 18-cv-02447-AJN, 2018 WL 2357266 (S.D.N.Y. Apr. 25, 2018) (discussing “significant time and effort” of habeas litigation that is not funded under the public defender contract); Declaration of Sarah Gillman at 2–3, Sajous v. Decker, No. 18-cv-02447-AJN, 2018 WL 2357266 (S.D.N.Y. Apr. 25, 2018) (noting that Legal Aid has “fixed and limited resources” and cannot file individual habeas corpus petitions on behalf of all clients subject to prolonged detention). Even where attorneys are able to remain on cases, the continuity of representation is disrupted. Attorneys are uninformed of transfer in advance and may learn of transfer well after it occurs.233Petition for Writs of Habeas Corpus and Mandamus and Class Complaint for Declaratory and Injunctive Relief at 12–23, Juan R. v. D.H.S., No. 21-cv-13117 (D.N.J. June 30, 2021) [hereinafter Juan R. Petition]. Moreover, representational difficulties abound due to: (1) prohibitively expensive phone fees and telephone access restrictions that make remote representation on a large scale untenable;234Id. at 10–11. (2) impediments to attorneys’ ability to conduct confidential assessments for relief, interview clients regarding traumatic facts relating to physical, sexual, and other violence, counsel clients as to their legal options, obtain signatures, and prepare clients to testify in court;235Id. at 8. (3) adverse impacts on attorneys’ abilities to support individuals in detention with cases beyond individual removal proceedings, including family court and child welfare proceedings,236See Thomas D. Honan, U.S. Immigr. & Customs Enf’t, Detention and Removal of Alien Parents or Legal Guardians § 5.2–4 (2017), [] (providing that if the detained individual’s family court or child welfare proceedings are within the area of responsibility of initial arrest, the field office director “must refrain” from making an initial placement or transferring the individual outside of the area of responsibility of arrest unless deemed “operationally necessary”; that, where practicable, the field office director “must arrange” for a detained parent to appear in-person at a family court or child welfare proceeding when their appearance is required in order to maintain or regain custody of their minor child(ren)); and that, in the event a parent or legal guardian is detained, “ICE will facilitate a means of regular visitation between the parent and minor child(ren)”). state court proceedings for conservatorships for individuals who are gravely disabled, and state court proceedings for ongoing criminal cases or postconviction relief;237Juan R. Petition, supra note 233, at 11. and (4) changes in circuit law that affect the viability of legal claims.238See Parker, supra note 220, at 36; Ryo and Peacock, supra note 177, at 40 (noting that of the roughly 86 percent of adults detainees who experienced one transfer, 29 percent experienced an intercircuit transfer).

Those unable to procure counsel prior to transfer are often transferred to localities where the ratio of attorneys to those in detention is low. For example, according to Human Rights Watch, the largest number of interstate transfers go to Louisiana, Mississippi, and Texas—states which collectively have the worst ratio of attorneys to individuals in detention.239 Parker, supra note 174, at 2. Moreover, transfers over hundreds of miles separate immigrants from evidence and witnesses crucial to adequately presenting their cases before immigration courts.240Id. Transfers further separate individuals from needed support from loved ones and community members. This separation takes a psychological toll on the individual detained and their surrounding community. And because individuals in detention are often transferred before they receive a bond hearing—where an immigration judge decides whether to continue an individual’s detention, in no small part based on levels of familial and community support—those subject to transfer are deprived of the ability to advocate for their release through the administrative process.241Id. at 16.

III. Transfers, Immigrants’ Access to the Courts, and the Case for Rethinking the “Immediate Custodian” and “District of Confinement” Rules in the Immigration Context

This Part turns to federal courts’ role in ensuring access to justice for immigrants challenging their detention in light of transfer. First, I describe how peripheral detentions and transfers impede immigrants’ ability to challenge their detention in federal court, and how Padilla’s sister rules exacerbate this impediment. Second, I problematize the Padilla Court’s reasoning, emphasizing the Reconstruction Era purpose of the civil habeas statue to expand access to habeas relief for all people. Finally, I argue that, considering the questionable—and arguably incorrect—foundations on which the Court relied in Padilla, the expansive intent of the Reconstruction Congress that passed the habeas statute, and the obsolescence of a territorial rule in the context of peripheral detention and transfer, courts should apply traditional venue principles in determining the proper forum to adjudicate an immigration habeas action that challenges present physical confinement.

A. Transfers, Access to the Courts, and Rumsfeld v. Padilla

ICE’s transfer decisions affect an individual’s ability to challenge the agency’s custodial authority in the federal courts through petitions for writ of habeas corpus.

Section 2241 of Title 28 of the U.S. Code provides that the federal courts may consider claims that an individual is “in custody in violation of the Constitution or law or treaties of the United States.”24228 U.S.C. § 2241(c)(3). The writ of habeas corpus has long been available to individuals in noncriminal custody, both public and private, and has been a vehicle for immigrants challenging their confinement since the United States’ founding. See Jonathan L. Hafetz, The Untold Story of Noncriminal Habeas Corpus and the 1996 Immigration Acts, 107 Yale L.J. 2509, 2522–24 & n.115 (1998) (noting that habeas corpus was available to immigrants at common law and was statutorily available to immigrants in the United States twenty-five years before federal courts gained habeas jurisdiction over citizens in state custody pursuant to the Habeas Act of 1867). Individuals in custody have used the writ in varied circumstances: to challenge Japanese internment,243Ex parte Endo, 323 U.S. 283 (1944). to express conscientious objection to the Vietnam War and challenge induction into the armed forces,244Strait v. Laird, 406 U.S. 341 (1972). to contest parole decisions,245Jones v. Cunningham, 327 U.S. 236 (1963). and to challenge extradition abroad.246Demjanjuk v. Meese, 784 F.2d 1114 (D.C. Cir. 1986). A writ of habeas corpus is also the vehicle through which individuals in immigration custody challenge their detention.247As the Supreme Court noted in Immigr. & Naturalization Servs. v. St. Cyr, “[a]t its historical core, the writ of habeas corpus has served as a means of reviewing the legality of Executive detention, and it is in that context that its protections have been strongest.” Immigr. & Naturalization Servs. v. St. Cyr, 533 U.S. 289, 301 (2001) (citing Swain v. Pressley, 430 U.S. 372, 380 n.13 (1977)). Despite congressional attempts to curb immigrants’ access to federal courts through passage of IIRIRA,248Illegal Immigration Reform and Immigrant Responsibility Act of 1996, Pub. L. No. 104-208, 110 Stat. 3009–46. AEDPA,249Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214. and the REAL ID Act of 2005,250REAL ID Act of 2005, Pub. L. No. 109-13, Div. B, §  106, 119 Stat. 231, 310–11. none of these purports to limit immigrants’ use of the writ of habeas corpus to challenge the legality of executive detention.251See St. Cyr, 533 U.S. at 305 n.25 (“[Section] 2241 descends directly from § 14 of the Judiciary Act of 1789 and the 1867 Act. . . . Its text remained undisturbed by either AEDPA or IIRIRA.”); see also § 106(b), 119 Stat. at 311; Randy Hertz & James S. Liebman, Federal Habeas Corpus Practice and Procedure § 42.1 n.1 (7th ed. 2022) (citing cases).

Transfer, however, has become a means of inhibiting immigrants’ access to the federal courts in detention challenges. As noted in Section II.B, ICE agents routinely arrest individuals for immigration detention and transfer them without notice to loved ones, community members, or counsel. When individuals challenge that detention through petitions for writ of habeas corpus in their home districts, they are met with motions to dismiss or transfer filed by the government. These motions routinely cite the Supreme Court’s Padilla decision and its rationale. First, they argue that Section 2241(a) contemplates a territorial limitation on the district court’s power: if an individual has been moved outside the territorial limits of the district court, no writ can follow. Second, they argue that the use of the definite article in reference to the custodian in Section 2243 implies that there can only be one proper respondent to a habeas action—a petitioner’s “immediate custodian”—who is, conveniently for the government, the warden of the jail or detention facility.

Argueta Anariba v. Director Hudson County Correctional Center provides a helpful example.252Argueta Anariba v. Dir. Hudson Cnty. Corr. Ctr., 17 F.4th 434 (3d Cir. 2021). Angel Argueta Anariba, a father to two U.S. citizens who has lived in the United States for over twenty years,253Opening Brief of Appellant at 10, Argueta Anariba, 17 F.4th 434 (No. 20-2633). was detained for a seven-year period beginning in 2014 and subject to approximately fourteen transfers between five different ICE facilities in five different states, including Alabama, Louisiana, Mississippi, New Jersey, and New York.254Id. In March 2019, while he was detained in a New Jersey county jail, Argueta Anariba filed a habeas petition on the basis of prolonged detention—he had been detained for twenty-nine months since his last bond hearing—in the federal court for the District of New Jersey.255Id. at 13–14. After the habeas petition had been pending for seven months, the federal district court denied it on procedural grounds, without prejudice to future filing.256Id. at 14 (explaining that the district court found that Respondent’s authorization for detaining Argueta Anariba changed during the pendency of the habeas from 8 U.S.C. § 1226 to § 1231, after the BIA denied Argueta Anariba’s removal case; and that 8 U.S.C. § 1231 permitted respondents to detain Argueta Anariba for ninety days after issuance of the BIA’s decision, rendering Argueta Anariba’s habeas petition premature. Petitioner notes, however, that the court “expressly permitted” Argueta Anariba to seek reopening if his detention continued passed the ninety-day period.). Six months later, Argueta Anariba filed a motion to reopen the habeas petition, renewing his prolonged detention claims and additionally seeking release because his specific comorbidities placed him at risk of severe illness or death during the COVID-19 pandemic.257Id. at 15. The district court issued another procedural decision, this time denying the habeas petition based on a finding that the court lacked jurisdiction because Argueta Anariba had been transferred outside of the court’s jurisdiction to an ICE contract facility in Catahoula Parish, Louisiana.258Id. at 16; see also Brief for Respondent-Appellee at 4, Argueta Anariba, 17 F.4th 434 (No. 20-2633). Argueta Anariba appealed.

On appeal, Argueta Anariba argued that transfer to Louisiana should not affect his ability to pursue his habeas petition in New Jersey. He noted the impracticalities of requiring a petitioner to file in the district of confinement, since the logic of that holding would have required him “to file at least three separate habeas petitions in three different courts in the past year as the statute authorizing his detention and physical location of detention change[d].”259Opening Brief of Appellant, supra note 253, at 24 n.1. The government argued that the district court did not have jurisdiction over Argueta Anariba’s claims because Argueta Anariba was not detained in the District of New Jersey.260Brief for Respondent-Appellee, supra note 258, at 1–2. It further argued that Argueta Anariba’s claims—that his detention was unreasonably prolonged, and that he was subject to conditions of confinement putting him at risk of severe illness or death during the COVID-19 pandemic—arose while he had been in custody in Louisiana.261Id. at 6, 12–13.

At oral argument, Argueta Anariba’s counsel emphasized that a federal district court in Louisiana was not the proper forum, given that all briefing in the underlying habeas action occurred in the District of New Jersey, including the filing of evidence and documents under seal, and pro bono counsel was located in New Jersey.262Oral Argument at 16:27, Argueta Anariba, 17 F.4th 434 (No. 20-2633). The Third Circuit panel, however, viewed Argueta Anariba’s arguments with skepticism.263See id. It assumed that the sole respondent to the habeas action was the director of the county correctional facility in Louisiana, where Argueta Anariba was being held, and asserted that the federal district judge in New Jersey “can’t direct that person to do anything.”264Id. at 12:26–12:28. The panel further stated, “There’s no one in the [New Jersey] judge’s jurisdiction who he can direct to give the relief that you are seeking. So, as a practical matter, I’m trying to understand how this works,”265Id. at 12:29–12:40. adding, “the attorney general is typically not the proper defendant in an immigration habeas case. So, you don’t have anybody in New Jersey that can deal with Louisiana. The attorney general looks like it may not be the proper defendant. So, who is?”266Id. at 13:17–13:32.

The Argueta Anariba panel’s questions invoked Padilla’s presumptions that there is a single respondent to an immigration habeas action, and that the respondent must be the equivalent of the warden of the jail where the petitioner is detained. The panel did not question whether the federal immigration officials who effectuated the transfers and determined the locus of detention were potential respondents.267See id. It quickly dismissed the attorney general as a potential respondent, despite the fact that individuals seeking release from immigration detention through habeas petitions are often granted a bail hearing conducted by the Executive Office of Immigration Review, a component of the Department of Justice, rather than immediate release.268See id.; Alexandra Lampert & Zoey Jones, Brooklyn Def. Servs., Litigating Habeas Corpus for Immigration Detainees, Vera Inst. of Just. (Jan. 21, 2021), []. Indeed, this is exactly what happened in Argueta Anariba’s case. Angel A.A. v. Edwards, No. 19-9135, 2022 WL 4001247, at *1 (D.N.J. Mar. 22, 2022) (ordering post-remand that an immigration judge provide petitioner with an individualized bond hearing within twenty-one days after the date of the court’s order). Nor did the panel entertain the possibility of the secretary of homeland security as a potential respondent, despite the secretary’s general detention authority.269See Oral Argument, supra note 262. Congress reorganized the Immigration and Naturalization Service through passage of the Homeland Security Act of 2002. See Homeland Security Act of 2002, Pub. L. No. 107-296, 116 Stat. 2135. Through the Homeland Security Act, Congress left the immigration adjudicative functions (including removal and bond hearings) in the Department of Justice. However, it moved immigration law enforcement, including detention and removal, to the Department of Homeland Security, of which the secretary of homeland security is “the head” and has “direction, authority, and control.” 6 U.S.C. § 112(a)(2).

Argueta Anariba prevailed in his appeal. The Third Circuit panel found that his motion to reopen was properly construed as a Rule 60(b)(6) motion, which did not amount to a new habeas petition. Accordingly, the panel found that the New Jersey district court retained jurisdiction following Argueta Anariba’s transfer outside of the District of New Jersey because it had already obtained jurisdiction over his habeas petition, which named his “then-immediate custodian,” the director of the New Jersey county jail.270Argueta Anariba v. Dir. Hudson Cnty. Corr. Ctr., 17 F.4th 434, 444–48 (3d Cir. 2021). This reasoning is largely premised on the Supreme Court’s decision in Ex Parte Mitsuye Endo, which held that when a federal district court acquires jurisdiction in a habeas action, removal of the petitioner does not cause the court to lose its jurisdiction “where a person in whose custody she is remains within the district.” Ex parte Endo, 323 U.S. 283, 306 (1944). However, the immediate custodian rule, particularly in the immigration context, is hard to square with Endo’s holding. The immediate custodian rule relies on a localized custodian—for example, the warden of a county jail or private detention center—who could no longer be the custodian upon removal of the petitioner outside of the district, as they are nowhere in the immigration agency’s chain of command. Argueta Anariba, 17 F.4th at 445. But had Argueta Anariba filed a new habeas petition while detained in Louisiana—or in Alabama or Mississippi, for that matter—the Third Circuit panel made clear that under its interpretation of the law, the only proper forum would have been his district of confinement, and the only proper respondent the warden of the jail.271Id. at 444 (“We recognize ‘[w]henever a § 2241 habeas petition seeks to challenge his present physical custody within the United States, he should name his warden as respondent and file the petition in the district of confinement.’ ” (quoting Rumsfeld v. Padilla, 542 U.S. 426, 441 (2004))). It would not have mattered that his residence, witnesses, records, attorneys, loved ones, and community were all in the Northeast. Venue principles would not have applied.272See Neirbo Co. v. Bethlehem Shipbuilding Corp., U.S. 165, 167–68 (1939) (explaining that the locality of a lawsuit is based on the convenience of the litigants). In considering the potentially competing interests of opposing litigants, future Circuit Judge Diane P. Wood explained:

Ideally, the location of each trial would optimize the interests of protection of defendant, fairness to plaintiff, speed of trial, and availability of witnesses. Barring achievement of this ideal, if liberal transfer statutes can protect the defendant adequately, and modern transportation facilities can minimize evidentiary problems, then it makes sense to give effect to the plaintiff’s initial choice of venue, assuming that he chooses a forum with a logical relation to his claim. This solution facilitates reaching the merits without lingering on procedural details, and it permits plaintiffs to enforce their legal rights without undue hardship.

Diane Pamela Wood, Comment, Federal Venue: Locating the Place Where the Claim Arose, 54 Tex. L. Rev. 392, 393–94 (1976).

The Third Circuit is not the only federal court confronted with this issue. Today, federal courts are increasingly tasked with addressing the proper forum in an immigration habeas action. Where a petitioner has been transferred between jails, the government regularly files motions to dismiss or transfer, insisting that Padilla’s immediate custodian/district of confinement sister rules be treated as absolute. This is nothing new. However, the Padilla decision, in tandem with the government’s detention authority and the increase in available jails and prisons for immigration detention, has effectively given the government a means of preventing a district court with the most salient connections to a case from administering an effective remedy. Federal courts, for their part, have taken disparate approaches to the issue of proper forum, most of them applying Padilla but others feeling obligated to distinguish Padilla and its application in the immigration context in order to adjudicate habeas petitions.

A look at three recent decisions in the same federal court exemplifies the different approaches to the question of proper forum. In Darboe v. Ahrendt, Darboe was arrested and detained by New York ICE officials and held in a New Jersey state penal institution.273Darboe v. Ahrendt, 442 F. Supp. 3d 592 (S.D.N.Y. 2020). The government filed a motion to dismiss Darboe’s habeas action for lack of jurisdiction or, in the alternative, to transfer the case from the Southern District of New York to the District of New Jersey.274Id. at 594. The court agreed and transferred Darboe’s case.275Id. at 596. In its analysis, the district court found that it “must have jurisdiction over the petitioner’s custodian,”276Id. at 594. and under Padilla, the custodian in a “core challenge” to present physical confinement is “ ‘the warden of the facility where the prisoner is being held,’ and the writ is issuable only in ‘the district of confinement.’ 277Id. (quoting Rumsfeld v. Padilla, 542 U.S. 426, 435, 442 (2004)). The court rejected the petitioner’s argument that it was ICE’s New York City field office director who was the proper respondent because he was the individual with the power to “transport, transfer, or release [Darboe].”278Id. at 595. Instead, the court found that only the warden of the New Jersey jail where Darboe was detained was his “immediate physical custodian.”279Id. (quoting Sanchez v. Decker, No. 19-cv-8354, 2019 WL 6311955, at *4 (S.D.N.Y. Nov. 25, 2019)). In so doing, the district court was able to ignore venue factors, which would have cut in Darboe’s favor given that he (1) resided in the Southern District of New York, (2) was represented by an attorney located in the Southern District of New York, (3) had removal proceedings in the Southern District of New York, and (4) was initially arrested and detained by ICE in the Southern District of New York.280See Shamira Ibrahim, Ousman Darboe Could Be Deported Any Day. His Story Is a Common One for Black Immigrants, Vox (Feb. 5, 2020, 11:58 AM), [] (describing Darboe’s life growing up in the Bronx and his immigration proceedings before the New York City immigration court).

Contrast the Darboe court’s analysis with that of the court in Cruz v. Decker.281Cruz v. Decker, No. 18-CV-9948, 2019 WL 4038555 (S.D.N.Y. Aug. 27, 2019), aff’d, No. 18-CV-9948, 2019 WL 6318627 (S.D.N.Y. Nov. 26, 2019). In Cruz, the government filed a motion to dismiss or transfer to the District of New Jersey, arguing that, under Padilla, the only proper respondent to Cruz’s habeas action was the warden of the New Jersey state penal institution where he was then detained.282Id. at *1. The Cruz court disagreed, making clear that the question of proper forum in an immigration habeas action was one of venue, not subject-matter jurisdiction.283Id. at *2. The court then held that Cruz’s proper respondent was the ICE New York field office director because ICE, “whether solely or in conjunction with the warden, exercise[d] that control to an extent that satisfies the immediate custodian rule under Padilla.”284Id. at *3. Thus, the Cruz court adopted Padilla’s holding but rejected the district of confinement rule in favor of traditional venue considerations; it looked beyond the warden to determine the proper respondent in light of the unique contours of contract detention.

Finally, consider the court’s analysis in J.G. v. Decker,285J.G. v. Decker, No. 22 Civ. 6273, 2022 WL 17832542 (S.D.N.Y. Dec. 20, 2022), rev’d, No. 22 Civ. 6273, 2022 WL 17992198 (S.D.N.Y. Dec. 28, 2022). where petitioner, a fifty-nine-year-old lawful permanent resident of forty years, was held in prolonged detention in Goshen, New York and then abruptly transferred to Adams County Correctional Center in Natchez, Mississippi.286Id. at *1. J.G. filed a petition for writ of habeas corpus on the same day of his transfer.287Id. at *2. The court initially sua sponte transferred the petition to the Southern District of Mississippi, holding that, under Padilla, the warden of the Adams County Correctional Center was the proper respondent and the Southern District of New York did not have jurisdiction over them.288Id. at *3–4. However, it reconsidered its decision upon learning that J.G. had filed his petition at 12:21 AM., about two-and-a-half hours before ICE officers shackled him at 3:00 AM and put him in a van for transport to Mississippi.289J.G. v. Decker, No. 22 Civ. 6273, 2022 WL 17992198, at *1 (S.D.N.Y. Dec. 28, 2022). Finding that the petitioner was in custody in the Southern District of New York at the time of habeas filing, the court held that it had jurisdiction over the petition.290Id.

So, how should courts analyze the question of proper forum when confronted with habeas challenges to physical confinement by petitioners in immigration detention? Before addressing this question, it bears returning to the Supreme Court’s analysis in Rumsfeld v. Padilla and its articulation of the “immediate custodian” and “district of confinement” rules.

B. The “Immediate Custodian” and “District of Confinement” Rules: A Look at the Supreme Court’s Underlying Rationale

When the Court was confronted with José Padilla’s habeas action, it articulated the question at issue as follows: “[D]id Padilla properly file his habeas petition in the Southern District of New York?”291Rumsfeld v. Padilla, 542 U.S. 426, 430 (2004). This question sounds in venue, a question of the proper locality for the filing and adjudication of Padilla’s habeas action.292See id. at 446–47; see also id. at 451–53 (Kennedy, J., concurring) (“In my view, the question of the proper location for a habeas petition is best understood as a question of personal jurisdiction or venue.”). Yet, in order to answer the venue question, the Padilla court further inquired into “whether the Southern District ha[d] jurisdiction over Padilla’s habeas petition.”293Id. at 434 (majority opinion). Thus, while the venue question animated the Court’s inquiry, the Court immediately signaled that it could not address the question of proper forum without grappling with jurisdictional issues. But what kind of jurisdiction? Although the Court did not do so expressly, it implicitly addressed questions of personal jurisdiction when it distilled its inquiry into two prongs: (1) “[W]ho is the proper respondent to that petition?” and (2) “does the Southern District have jurisdiction over him or her?”294Id. These questions reference the court’s ability “to render judgment in personam” and its “de facto power over the [respondent]’s person.”295Int’l Shoe Co. v. Washington, 326 U.S. 310, 316 (1945).

The “Immediate Custodian” Rule. Padilla had named several respondents to his habeas action: President George W. Bush, as commander-in-chief of the U.S. armed forces; Secretary of Defense Donald Rumsfeld; Attorney General John Ashcroft; and Naval Brig Commander Melanie A. Marr.296Amended Petition for Writ of Habeas Corpus ¶¶ 8–11, Padilla ex rel. Newman v. Bush, 233 F. Supp. 2d 564 (S.D.N.Y. June 19, 2002) (No. 02 Civ. 4445). It seems apparent that the Southern District would have had personal jurisdiction over some of these respondents. Yet, the Court quickly dismissed all but one—Commander Marr—finding that there could be but one proper respondent to a habeas action, the so-called “immediate custodian.”297Padilla, 542 U.S. at 435–36.

How did the Court get here? How did it determine that, despite the district court’s issuance of a material witness warrant for Padilla’s arrest and the president’s issuance of an order to the secretary of defense to name Padilla an enemy combatant, Commander Marr was the sole proper respondent to Padilla’s habeas petition?298See id. at 430–31. It reasoned that because the habeas statute in Sections 2242 and 2243 uses the singular article “the” in reference to the custodian, it must contemplate a single respondent.299See id. at 434 (discussing 28 U.S.C. §§ 2242–43). For example, Section 2242 instructs the petitioner to “name the person who has custody over [them].”30028 U.S.C. § 2242 (emphasis added). Section 2243 reiterates that the writ or order to show cause “shall be directed to the person having custody of the person detained” and further requires that “[t]he person to whom the writ or order is directed shall make a return certifying the true cause of detention.”301Id. § 2243 (emphases added).

The Court’s reasoning might seem simple enough, yet the Dictionary Act instructs us to the contrary. The Dictionary Act directs federal courts to apply certain grammatical rules in the course of statutory interpretation “unless the context indicates otherwise.”3021 U.S.C. § 1. While “context” is ambiguous, the Supreme Court has, in the past, narrowly interpreted the term to mean “the text of the Act of Congress surrounding the word at issue, or the texts of other related congressional Acts.” Rowland v. Cal. Men’s Colony, Unit II Men’s Advisory Council, 506 U.S. 194, 199 (1993). One such grammatical rule is that, when reading the U.S. Code, “words importing the singular include and apply to several persons, parties, or things.”3031 U.S.C. § 1; see also Niz-Chavez v. Garland, 141 S. Ct. 1474, 1482 (2021) (providing that the Dictionary Act tells us that “a statute using the singular ‘a’ can apply to multiple persons, parties, or things”). The Dictionary Act’s instruction suggests that multiple persons may be custodians and therefore, respondents to a habeas action under the habeas statute.

Yet, the Padilla Court never mentioned the Dictionary Act in its analysis. Instead, the Court looked to its 1885 decision in Wales v. Whitney, a case involving a habeas petition filed by a medical director and former navy surgeon general, who sought habeas relief from an order of court martial levied against him by the secretary of the navy.304Wales v. Whitney, 114 U.S. 564, 566 (1885), overruled in part by Hensley v. Mun. Ct., 411 U.S. 345, 350 n.8 (1973). Wales v. Whitney is not a case about the proper respondent (or number of respondents) to a habeas action. Rather, it is a case about the requisite custody requirement. Finding that Wales was subject only to a “moral restraint” rather than “actual confinement,” the Court held that the order of arrest and court martial against him were insufficient restraints of liberty to warrant a habeas action.305Id. at 571–75. Wales v. Whitney was subsequently overturned, as the definition of “in custody” steadily expanded to its present terms.306See, e.g., Hensley v. Mun. Ct., 411 U.S. 345 (1973) (holding that a petitioner who was released on his own recognizance pending execution of sentence is “in custody” for purposes of the habeas corpus statute); Jones v. Cunningham, 371 U.S. 236 (1963) (holding that a person released on parole is “in custody” for purposes of the district courts’ habeas jurisdiction).

Despite the fact that the Court later overturned Wales’s holding and that Wales challenged something other than present physical confinement, the Padilla court relied on the case’s dicta to determine the proper respondent to a habeas action where physical restraint is at issue—what the Court terms a “core” habeas challenge.307Rumsfeld v. Padilla, 542 U.S. 426, 435 (2004). Quoting Wales, the Padilla court stated that the habeas provisions “contemplate a proceeding against some person who has the immediate custody of the party detained, with the power to produce the body of such party before the court or judge, that he may be liberated if no sufficient reason is shown to the contrary.”308Id. (alteration in original) (quoting Wales, 114 U.S. at 574). The Court additionally cites decisions in Braden v. 30th Jud. Cir. Ct., 410 U.S. 484, 494–95 (1973), and In re Jackson, 15 Mich. 417, 439–40 (1867), for the proposition that the writ acts upon the jailer or custodian. Id. While Wales provided no clear indication who the person with “the immediate custody” might be, the Padilla court concluded that it must be the warden or warden-equivalent of the facility where an individual is being held, rather than a “remote supervisory official” like the attorney general.309Padilla, 542 U.S. at 435. How the Court reached this conclusion is unclear. Petitioner Wales challenged his court martial, not present physical custody.310See Wales, 114 U.S. at 568–69. The Court denied Wales’s petition on the basis that he did not meet the now-defunct physical restraint requirement, a result informed by the fact that the Court inferred Wales was using the writ to circumvent military justice in favor of a civil tribunal.311Id. at 570, 575.

Perhaps, as some have argued, the Padilla Court’s analysis turned on its understanding of the chain of command.312See, e.g., Ameen v. Jennings, No. 22-cv-00140, 2022 WL 1157900, at *4 (N.D. Cal. Apr. 19, 2022); Cruz v. Decker, No. 18-CV-9948, 2019 WL 6318627, at *2 (S.D.N.Y. Nov. 26, 2019); Uljic v. Baker, No. 06-13106, 2006 WL 2811351, at *2 (E.D. Mich. Sept. 28, 2006). In Padilla, Padilla’s “immediate custodian,” Commander Marr, was in the direct chain of command of the supervisory official with confining authority, the secretary of defense. Therefore, Commander Marr could, practically and legally, produce the prisoner’s body before the habeas court,313Padilla, 542 U.S. at 435–36. “that he may be liberated if no sufficient reason is shown to the contrary.”314Wales, 114 U.S. at 574. But what about instances where the warden or warden-equivalent is outside the chain of command and has no power to address the true cause of detention? How does one rationalize the Court’s decision in the context of immigration detention, which as of 2021 counted 220 facilities, about half of which exist pursuant to intergovernmental service agreements with state or local governments or private contractors315 U.S. Gov’t Accountability Off., supra note 168, at 11. whose officials cannot substantiate the cause of detention?

Nor is the Court’s singular, immediate custodian interpretation firmly rooted in the legislative history of the habeas statute. The Habeas Act of 1867 forms the basis of the general grant of habeas corpus jurisdiction in 28 U.S.C. § 2241 to the federal courts.316Act of Feb. 5, 1867, ch. 28, § 1, 14 Stat. 385 (codified at 28 U.S.C. § 2241). It is widely accepted that while the statutory language of the Act of 1867 has been amended in detail, its substance has not been.317Louis H. Pollak, Proposals to Curtail Federal Habeas Corpus for State Prisoners: Collateral Attack on the Great Writ, 66 Yale L.J. 50, 52 n.9 (1956). A look at Chapter 28, Section 1 of the Act of 1867 evidences that the drafters did not conceive of a singular individual as responsive to the writ. For example, Section 28 provides:

if any person or persons to whom such writ of habeas corpus may be directed shall refuse to obey the same, or shall neglect or refuse to make return, or shall make a false return thereto, in addition to the remedies already given by law, he or they shall be deemed to be guilty of a misdemeanor . . . .318Act of Feb. 5, 1867 § 1 (emphasis added). Even if Section 28 contemplated that these individuals were part of a singular custodial entity, that leaves higher-level officials within the chain of command subject to suit.

This usage of the plural undermines claims that inclusion of the definite article in later versions of the statute shows Congress’s clear intent to identify a single respondent.319C.f. Rumsfeld v. Padilla, 542 U.S. 426, 434–35 (2004). It bears questioning whether the immediate custodian rule was chosen by the Court because of its interpretation of the statute and its legislative history, or for some other reason. Suffice it to say that the Padilla Court’s decision to manacle the respondent to a petitioner’s physical jailer—with little consideration for the original text of the statute or the shifting nature of the carceral state and the rise of contractual detention—about-faced a longstanding effort by the court to imbue the writ with the “initiative and flexibility essential to insure that miscarriages of justice within its reach are surfaced and corrected.”320Harris v. Nelson, 394 U.S. 286, 291 (1969).

The District of Confinement Rule. The Padilla Court then went further to limit the reach of habeas when it articulated the “district of confinement” rule.321Padilla, 542 U.S. at 442. Turning to the second prong of its inquiry, the majority held that the “general” rule is where a habeas petitioner challenges present physical confinement, the writ is only issuable in the “district of confinement.”322Id. at 442–43.

Here again, it bears unpacking the Court’s analysis. By its terms, 28 U.S.C. § 2241 provides that district courts may grant writs of habeas corpus “within their respective jurisdictions.”32328 U.S.C. § 2241(a). The issue the Court had to resolve is whether the term “respective jurisdictions” created a territorial limitation on a district court’s ability to issue the writ; whether it referenced the power of a district court to hear a case; or whether it served only to ensure that cases were brought before the right court—a question of proper forum or venue. Here, we are reminded that “jurisdiction . . . is a word of many, too many, meanings.”324Steel Co. v. Citizens for a Better Env’t, 523 U.S. 83, 90 (1998) (internal citations omitted) (citing United States v. Vanness, 85 F.3d 661, 663 n.2 (D.C. Cir. 1996)). Thus unable to rely on pure textualism, the Court turned to legislative history to discern the meaning of ambiguous text. The Court stated that Congress added the clause “within their respective jurisdictions” to avoid a situation in which “every judge anywhere [could] issue the Great Writ on behalf of applicants far distantly removed from the courts whereon they sat.”325Padilla, 542 U.S. at 442 (alteration in the original) (quoting Carbo v. United States, 364 U.S. 611, 617 (1961)). It concluded that for petitions seeking release from physical confinement, the writ is only issuable in the “district of confinement.”326Id. at 447.

However, the Court’s conclusion, couched in terms of legislative history, is insufficient. While the legislative history of the Habeas Act of 1867 itself is limited,327See Anthony G. Amsterdam, Criminal Prosecutions Affecting Federally Guaranteed Civil Rights: Federal Removal and Habeas Corpus Jurisdiction to Abort State Court Trial, 113 U. Pa. L. Rev. 793, 823 (1965) (“Floor discussion of the [Habeas] act of 1867 was . . . quite limited.”). Writing the same year as Amsterdam, Lewis Mayers also argued that Congressional intent around the passage of the February 5, 1867 Act is anything but clear. See Lewis Mayers, The Habeas Corpus Act of 1867: The Supreme Court as Legal Historian, 33 U. Chi. L. Rev. 31, 42 (1965). Paul Bator called the legislative history of the Act “strikingly sparse.” Paul M. Bator, Finality in Criminal Law and Federal Habeas Corpus for State Prisoners, 76 Harv. L. Rev. 441, 475 (1963). William Wiecek notes, “Congressional intent in passing the Habeas Corpus Act of 1867 is unusually murky, even for a piece of Reconstruction legislation.” William M. Wiecek, The Great Writ and Reconstruction: The Habeas Corpus Act of 1867, 36 J. S. Hist. 530, 539 (1970). the historical record clearly demonstrates the Reconstruction Congress’s consistent intent to enhance the powers of the federal courts.328Wiecek, supra note 327, at 531–32 (“As a matter of fact, the business and the powers of the federal courts were expanded by Congress in the Reconstruction period to an extent that has no parallel before or since.”).

During the Civil War, enslaved Black Americans crossed Union lines in pursuit of freedom. Congress passed consecutive Confiscation Acts, which legally freed those enslaved by Rebels who were using them to assist Southern troops.329E.g., Act of Aug. 6, 1861, ch. 60, 12 Stat. 319; Act of July 17, 1862, ch. 195, 12 Stat. 589. As W.E.B. Du Bois recounts:

The North started out with the idea of fighting the war without touching slavery. They faced the fact, after severe fighting, that Negroes seemed a valuable asset as laborers, and they therefore declared them “contraband of war.” It was but a step from that to attract and induce Negro labor to help the Northern armies. Slaves were urged and invited into the Northern armies; they became military laborers and spies; not simply military laborers, but laborers on the plantations, where the crops went to help the Federal army or were sold North. Thus wherever Northern armies appeared, Negro laborers came, and the North found itself actually freeing slaves before it had the slightest intention of doing so, indeed when it had every intention not to.

William Edward Burghardt Du Bois, Black Reconstruction in America: An Essay Toward a History of the Part Which Black Folk Played in the Attempt to Reconstruct Democracy in America, 1860–1880, at 68 (Russell & Russell 1962).
As more and more Black people freed themselves in the chaos of war, the United States government “followed the footsteps of the [B]lack slave,” acknowledging that the Civil War was a war against slavery.330 Du Bois, supra note 329, at 83–84. On January 1, 1863, President Abraham Lincoln signed the Emancipation Proclamation, declaring that the enslaved people of all persons in rebellion were “thenceforward, and forever, free.”331Proclamation No. 95 (Emancipation Proclamation), 12 Stat. 1268 (Jan. 1, 1863). Many of the people to whom the Emancipation Proclamation applied had already freed themselves or had been freed by invading armies, such that the Proclamation merely “added possible legal sanction to an accomplished fact.” Du Bois, supra note 329, at 87. However, the Proclamation served to rally the support of four-and-a-half million Black Americans, and to prevent recognition of the Confederacy by England and France. Id. at 89–93. Importantly, the Proclamation also provided that “such persons, of suitable condition, will be received into the armed service of the United States to garrison forts, positions, stations, and other places, and to man vessels of all sorts in said service.”332Proclamation No. 95 (Emancipation Proclamation), 12 Stat. 1268; see also Eric Foner, Reconstruction: America’s Unfinished Revolution, 1863–1877, at 7 (1988). Of the Proclamation’s provisions, those around military service were considered some of the most radical and essential because they “breath[ed] life into the promise of emancipation” through the massive enlistment of Black people into the military.333 Foner, supra note 329, at 7. By the Civil War’s end, approximately 180,000 Black Americans had served in the Union Army, comprising “over one fifth of the nation’s adult male [B]lack population under age forty-five,” the majority from border states, where enlistment was the road to freedom. Id. at 8.

These actions, however, failed to provide Black Americans with habeas protections. Questions regarding the legal status of the formerly enslaved paralyzed Congress, such that when it passed the Habeas Corpus Indemnity Act of 1863, it only implicitly contemplated the individual liberties of the newly free.334See Justin J. Wert, Habeas Corpus in America: The Politics of Individual Rights 85–87 (2011). Wert explains that, while the writ of habeas corpus was suspended during much of the Lincoln Administration, the writ continued to develop in important ways. As one example, he discusses how an earlier version of the Second Confiscation Act in the House of Representatives included a habeas provision that would have allowed newly freed Black Americans to seek habeas relief in the case of re-enslavement by former or pretended masters. He notes that this provision was rejected, in part by House members who did not want to treat the newly free as persons, but as confiscated property. Id. at 86; see also H.R. Rep. No. 37-120 (1862). Instead, it sought protections for federal officials subject to prosecutions in Rebel jurisdictions, a precursor to the removal provisions that would ultimately form part of the Habeas Act of 1867.335Act of Mar. 3, 1863, ch. 81, §§ 4–5, 12 Stat. 755; see also Amsterdam, supra note 327, at 823–25.

Then, on March 3, 1865, just before the War’s end, Congress passed a resolution “to encourage [e]nlistments and to provide the [e]fficiency of the military [f]orces of the United States.”336Act of Mar. 3, 1865, ch. 29, 13 Stat. 571. The resolution provided that the wives and children of any person who had been or may have been enlisted in the U.S. Army or Navy would be “forever free.”337Id. It further conferred freedom on those who were living in slave states not in rebellion and therefore did not fall under the provisions of the Emancipation Proclamation.

On December 19, 1865, Republican Congressman Samuel Shellabarger of Ohio offered an enforcement measure for Congress’s March 3rd resolution, which was immediately agreed upon by the House:

Resolved, That the Committee on the Judiciary be directed to inquire and report to this House, as soon as practicable, by bill or otherwise, what legislation is necessary to enable the courts of the United States to enforce the freedom of the wives and children of soldiers of the United States under the joint resolution of Congress of March 3, 1865, and also to enforce the liberty of all persons under the operation of the constitutional amendment abolishing slavery.338 Cong. Globe, 39th Cong., 1st Sess. 87 (1865).

On March 15, 1866, Congressman Shellabarger raised the issue of habeas in connection with federal removal legislation. The congressman asked what was to be done about “a very large class of similar cases,” wherein Union officers were subject to state court trials in regions still loyal to the Confederacy by virtue of acts committed in the line of duty.339Id. These federal officers would inevitably be convicted by recalcitrant state tribunals with anti-Union prejudices and therefore were compelled to flee the state.

Ultimately, Shellabarger’s concerns regarding emancipation and removal seem to have been woven together into the Habeas Corpus Act of 1867, which was reported out of the House of Representatives on July 25, 1866. In explaining the nature of the bill, Republican Congressman William Lawrence of Ohio summarized his colleague Shellabarger’s efforts, stating that the congressman sought legislation that would “enforce the liberty of all persons,” “enlarge the privilege of the writ of habeas corpus,” and “make the jurisdiction of the courts and judges of the United States coextensive with all the powers that can be conferred upon them.”340 Cong. Globe, 39th Cong., 1st Sess. 4151 (1866). He added, “It is a bill of the largest liberty, and does not . . . restrain the writ of habeas corpus at all.”341Id. Thus, in its earliest iteration, the Habeas Act of 1867 was not intended for narrow constructions that would limit access to the federal courts; instead, Congress introduced the Act to enlarge the habeas remedy by enlarging the federal jurisdiction of the courts and judges of the United States.342Fay v. Noia, 372 U.S. 391, 417 (1963). But see Mayers, supra note 327, at 35 (arguing that the Habeas Corpus Act of 1867 was intended solely to fulfill the dictates of the Thirteenth Amendment and liberate those formerly enslaved, who remained subjects through bondage and peonage, and did not contemplate federal removal, as the Supreme Court later concluded in Fay v. Noia).

In the Senate, Republican Senator Lyman Trumbull of Illinois introduced the Habeas Act of 1867 by echoing Congressman Lawrence’s sentiments. Trumbull stated, “It is a bill in aid of the rights of the people.”343 Cong. Globe, 39th Cong., 1st Sess. 4228–29 (1866). Discussions of the bill remained quite limited. Nevertheless, Democratic Senator Reverdy Johnson of Maryland introduced the question that would spark debates about choice of forum for decades to come, and on which the Supreme Court would later hinge its decisions to limit federal court jurisdiction in Rumsfeld v. Padilla.

In a statement to the full Senate, Senator Johnson argued that the senators should examine the language of the bill because, as he saw it, “an application might be made to a district judge in Florida to bring before him some men convicted and sentenced and held under imprisonment in the State of Vermont or in any of the further States.”344 Cong. Globe, 39th Cong., 2d Sess. 730 (1867). According to Charles Fairman, Senator Johnson’s query arose from his proximity to a habeas action filed on behalf of Dr. Samuel A. Mudd. See Charles Fairman, 6 History of the Supreme Court of the United States, Reconstruction and Reunion, 1864–88, Part One 237–39 (1971). Dr. Mudd had set John Wilkes Booth’s injured leg during his escape after he had assassinated President Abraham Lincoln. Dr. Mudd was later tried for complicity in the assassination and sentenced to hard labor for life on the island of Dry Tortugas. During his sentence (and before he was pardoned), he applied for a petition for writ of habeas corpus with Supreme Court Chief Justice Salmon Chase and was represented by Andrew Sterett Ridgely, Senator Johnson’s son-in-law. Id. at 237–38. The chief justice denied the petition without opinion. Id. at 239. Fairman states that Senator Johnson understood the denial to be based on Chief Justice Chase’s opinion that he did not have the power to exercise the writ outside of his circuit. Id. He stated further, “Any man who may be imprisoned in any part of the United States may be brought out by this writ issued by a district judge of the United States farthest from the place of imprisonment,” adding that this would be “exceedingly inconvenient, embarrassing, and expensive.”345 Cong. Globe, 39th Cong., 2d Sess. 730 (1867).

At the Senate’s next meeting, Senator Trumbull summarized Senator Johnson’s critique of the House bill, stating, “I do not think the bill is susceptible of that construction. I think the judges and courts would under the bill be confined to their respective jurisdictions.”346Id. at 790. It is no surprise that Senator Trumbull did not envision such a construction. At the time of the Senate’s debate, the notion that an individual subject to physical confinement could be transferred out of the territorial limits of a state was anathema to the “positive republican theory of crime” that had emerged in the early years of the Republic.347See Emma Kaufman, The Prisoner Trade, 133 Harv. L. Rev. 1815, 1823 (2020) (quoting Rebecca M. McLennan, The Crisis of Imprisonment 19 (2008) (detailing that between 1776 and 1845, eleven states had enacted constitutional bans or limits on criminal sanctions involving forced mobility)). Indeed, a central concern of the Framers of the Constitution was that federal court litigation would “carry me a thousand miles from home—from my family and business—to where, perhaps, it will be impossible for me to prove [my case].”348Markowitz & Nash, supra note 20, at 1165 (alteration in original) (quoting 3 The Debates in the Several State Conventions, on the Adoption of the Federal Constitution, as Recommended by the General Convention at Philadelphia, in 1787, at 526 (Jonathan Elliot ed., J.B. Lippincott & Co. 2d ed. 1996) (1891) (statement of Virginia delegate George Mason)); (citing 4 The Debates on the Several State Conventions, supra, at 136 (statement of North Carolina delegate Judge Samuel Spencer) (discussing the oppression and interference that would be caused by parties having to travel to distant federal courts)). Moreover, at the time, neither a federal prison system nor a federal immigration detention system existed,349See Paul W. Keve, Prisons and the American Conscience: A History of U.S. Federal Corrections 29 (1991) (noting that the Fifty-first Congress “took the first step” in moving towards a federal prison system with passage of the Three-Prisons Act). Notably, one of the three prisons incorporated into the new Federal Prison System through Congress’s passage of the Three-Prisons Act in 1891 was McNeil Island, the only federal prison on the West Coast until the 1930s, which was largely used to incarcerate immigrants. See also supra Section I.A. such that neither senator could have contemplated a federalized system of confinement and transfer.

Despite Senator Trumbull’s vocalized skepticism, he accommodated Senator Johnson’s concern that courts with no connection to a case should not have habeas power, stating:

I shall have no objection to amending [the Act] by inserting after the word ‘courts” in the fourth line of the first section the words “within their respective jurisdictions;” so that it would then read “that the several courts of the United States and the several justices and judges of such courts, within their respective jurisdictions,” &c.350 Cong. Globe, 39th Cong., 2d. Sess. 790 (1867).

Senator Johnson agreed to Senator Trumbull’s amendment, and it was thereafter enacted.351Id. at 790, 903.

The Reconstructionist Congress that passed the Act could not have contemplated a scenario in which a federal agency could arrest, detain, and transfer an individual through a patchwork network of jails and prisons, with the ability to carry them hundreds or thousands of miles away from home; that an agency could do so with no notice to the individual, their loved ones, or their counsel; and that it could do so while substantive proceedings were ongoing in another judicial forum. What is clear from a review of the legislative history is that an individual’s locus of confinement could not have had much meaning in the postbellum United States. The Reconstruction Congress’s focus was on providing a meaningful remedy in a court that had grounds for asserting jurisdiction. Therefore, an interpretation that the Reconstructionist Congress passed the Act intending to create strict territorial limits to accessing the writ is unsubstantiated.

Yet, despite this history, language from the Habeas Act of 1867 was cited by the Padilla Court to create a doctrine which ultimately curtails access to the writ by incentivizing transfer without notice or process—this legitimized a practice that in the colonial period was considered an act of banishment and punishment in and of itself and that was widely outlawed in many state constitutions as reminiscent of the harshness of the British crown.352Rumsfeld v. Padilla, 542 U.S. 426, 446–47 (2004). The Habeas Act of 1867 was passed in a period when the American penal system “lived within state lines”—as is reflected in the first cases that arise under the Act. Those incarcerated were not shipped out of state; instead, punishment remained hyperlocal. See Kaufman, supra note 347, at 1822. It was also passed largely before the development of federal immigration law, which did not really begin until the 1870s. See An Act Supplementary to the Acts in Relation to Immigration (Page Act), ch. 141, 18 Stat. 477 (1875) (repealed 1974). Rather than allowing immigrant petitioners to seek redress from physical confinement in a federal district court with a connection to the substance of a case, the “district of confinement” rule ferries immigration custody disputes into forums that unfairly disadvantage immigrant petitioners by requiring that they litigate in locations far from the evidence and witnesses necessary to argue for their liberty.

The Reconstruction Congress’s focus was on providing a meaningful remedy in a court that had grounds for asserting jurisdiction.353See Wiecek, supra note 327, at 531. Therefore, an interpretation that the Congress that passed the Act intended to create strict territorial limits to accessing the writ is both unsubstantiated and anachronistic.

The Concurrence. Justice Kennedy, while joining the Padilla majority, wrote a “pivotal concurrence,”354See Thomas B. Bennett, Barry Friedman, Andrew D. Martin & Susan Navarro Smelcer, Divide & Concur: Separate Opinions & Legal Change, 103 Cornell L. Rev. 817, 847 (2018) (defining a pivotal concurrence as one that arises where a justice concurs in the majority opinion, but also writes separately, and “that judge’s vote is numerically necessary to give the majority opinion enough votes to become binding precedent”). clarifying what is only implied in the majority opinion: the question of proper forum for a habeas action is one of personal jurisdiction or venue,355Padilla, 542 U.S. at 451 (Kennedy, J. concurring). not of subject-matter jurisdiction.356Id. He highlighted that it was clear these rules do not flow from questions of subject-matter jurisdiction, given that the Court has heard numerous cases on the merits “despite their noncompliance with either one or both of the rules.”357Id. (citing Braden v. 30th Jud. Cir. Ct. of Ky., 410 U.S. 484, 495 (1973); Strait v. Laird, 406 U.S. 341, 345 (1972); United States ex rel. Toth v. Quarles, 350 U.S. 11 (1955); Burns v. Wilson, 346 U.S. 137 (1953); Ex parte Endo, 323 U.S. 283 (1944)); see also The Resolute, 168 U.S. 437, 439 (1897) (“Jurisdiction is the power to adjudicate a case upon the merits, and dispose of it as justice may require.”). Nevertheless, he concluded that in “the ordinary case,” the “immediate-custodian rule” should prevail,358Padilla, 542 U.S. at 453 (Kennedy, J. concurring). and that Padilla’s habeas petition did not raise cause for exception:359Id. at 453–54. Padilla’s custody “should be challenged in the place the Government has brought [him] to bear and against the person who is the immediate representative of the military authority that is detaining him.”360Id. Again, however, we are confronted with a rationale premised on the “immediate custodian” being a singular respondent who is “representative” of the official exercising detention authority, a premise at odds with the text of the Habeas Act of 1867, its legislative history, and the realities of contractual detention. And while the dissent disagreed with the ultimate decision, it, too, reaffirmed the immediate custodian and district of confinement rules,361Id. at 458 (Stevens, J., dissenting) (“All Members of this Court agree that the immediate custodian rule should control in the ordinary case and that habeas petitioners should not be permitted to engage in forum shopping.”). merely chastising the majority for its “slavish application of a ‘bright-line rule.’ 362Id. at 455 (quoting id. at 449 (majority opinion)).

Despite a fractious opinion, the Padilla court was unanimous in its assertion of the general rules governing proper forum in a habeas action. Their disagreements were ultimately about when to carve out exceptions to the rules, rather than an interrogation of the rules themselves. According to the majority, the Court’s interpretation purportedly “derived from the terms of the habeas statute,” would prevent rampant forum shopping by habeas petitioners and district courts with overlapping jurisdiction, and track consistently with the concerns of the Reconstruction Congress that passed the Habeas Act of 1867.363Id. at 447 (majority opinion).

Yet, as outlined above, the certainty of the Court’s conclusion is belied by a review of the habeas statute’s original text and its legislative history. What’s more, the Court was silent on how its analysis might apply in the immigration context, noting in a footnote that its earlier decision in Ahrens v. Clark “left open” the question of whether the attorney general is a proper respondent in an immigration habeas action, but that the Court “decline[d] to resolve [it]”364Id. at 435 n.8. Notably, the Court had already overturned its 1948 Ahrens decision in 1973, in Braden v. 30th Jud. Cir. Ct., 410 U.S. 484 (1973), asserting that, “developments since Ahrens have had a profound impact on the continuing vitality of that decision,” and the Court could “no longer view that decision as establishing an inflexible jurisdictional rule, dictating the choice of an inconvenient forum.” Id. at 497–500.—indeed, the Court never has. Instead, in the immigration context, district courts have been left to interpret the Supreme Court’s decision in Rumsfeld v. Padilla amidst its earlier caselaw. As discussed below, many district courts have applied the so-called “immediate custodian” and “district of confinement” rules, forcing immigrants to raise challenges to their present physical confinement in the district courts governing the locus of their detention—often in geographically isolated towns, far from home, loved ones, community, witnesses, evidence, and actual or potential counsel.

C. Rethinking the “Immediate Custodian” and “District of Confinement” Rules in the Immigration Context

The habeas statute’s Reconstruction Era beginnings suggest that the Act does not compel an interpretation that a district court’s power begins and ends with its territorial jurisdiction. Indeed, that jurisdiction could be defeated by transfers often conducted in obscurity seems to undermine the interests in liberty and equality at play during legislative debates.365Even going back to its very beginnings, the writ would “transcend jurisdictions, championing substance (whether the jailer had a legal basis for confining the prisoner) over jurisdictionally varied procedural forms.” Stephen I. Vladeck, The New Habeas Revisionism, 124 Harv. L. Rev. 941, 948 (2011) (reviewing Paul D. Halliday, Habeas Corpus: From England to Empire (2010)). What’s more, even as the Padilla Court was motivated in large part by a desire to “prevent[] forum shopping by habeas petitioners,”366Padilla, 542 U.S. at 447. the Court also held that habeas corpus “must not be subject to manipulation by those whose power it is designed to restrain.”367Boumediene v. Bush, 553 U.S. 723, 765–66 (2008). Together, these ideas stand for the “logical principle that neither the petitioner nor the government should be able to forum shop or manipulate the availability of habeas relief.”368Brief of Immigrant Def. Project as Amicus Curiae in Support of Petitioner, Cruz v. Decker, No. 18-CV-9948, 2019 WL 4038555 (S.D.N.Y. Aug. 27, 2019), ECF No. 39 [hereinafter Immigrant Def. Project Brief]. Yet here we are, with government attorneys standing on Padilla to argue that immigration habeas petitioners must file their petitions in their districts of confinement, forcing immigrants to litigate their freedom wherever the government decides to confine them without regard to the inconvenience of the forum, the availability of evidence, witnesses, and counsel, and the discord it causes ongoing immigration litigation.

However, fairness in location has “little to do with jurisdiction and everything to do with due process and venue.”369Markowitz & Nash, supra note 20, at 1153. Venue is “litigant protective” because it “inquires into the propriety of the site of proceedings based on the parties’ connections to the place and obstacles to a fair proceeding”370Id. at 1160. and has historically served to protect a litigant against an opposing party attempting to select an inconvenient location for trial as a tactical advantage.371Id. at 1163. Stripping immigration habeas actions of traditional venue protections in favor of a district of confinement rule has left immigrant petitioners exposed to the whims of the government without any protections against litigation in unfair fora, particularly as ICE transfer infrastructure has become more sophisticated and expansive.372See supra Section II.B. Now, the government has the ability to transfer immigrants to the jurisdiction that will provide it with the most favorable precedent while potentially limiting an immigrant’s access to counsel, loved ones, and community support.373See Grantham, supra note 20, at 301–02. This potential for manipulation is exacerbated by new video teleconferencing policies and the government’s seemingly unfettered authority to decide whether to transport an individual to their immigration hearing.374Id.; Off. of the Inspector Gen., U.S. Dep’t of Just., Limited-Scope Inspection and Review of Video Teleconference Use for Immigration Hearings, at i (2022), [].

Some courts have recognized that traditional venue principles appropriately apply in immigration habeas actions. Harkening back to the Supreme Court’s decision in Braden and to Justice Kennedy’s concurrence in Padilla, these courts view challenges to jurisdiction raised by the government as questions of personal jurisdiction and venue375Braden v. 30th Jud. Cir. Ct., 410 U.S. 484, 500 (1973); Rumsfeld v. Padilla, 542 U.S. 426, 451 (2004) (Kennedy, J., concurring); see, e.g., Cruz v. Decker, No. 18-CV-9948, 2019 WL 4038555, at *2 (S.D.N.Y Aug. 27, 2019) (interpreting the government’s motion to dismiss or transfer as a venue motion).—rightfully so. Statutory venue protections under 28 U.S.C. § 1391—together with the availability of motions to transfer under 28 U.S.C. § 1404(a) and forum non conveniens—operate as a “safeguard against a plaintiff’s decision to forum shop or to use an inconvenient forum to prejudice a defendant.”376Markowitz & Nash, supra note 20, at 1183–86. Accordingly, by applying venue principles, courts can protect defendants without prejudicing immigration habeas petitioners through rigid application of the district of confinement rule. By bringing the protective function of venue back into the fold, federal courts can partially curb abuse by either party.

Yet, applying venue principles alone cannot remedy the injustice of litigation in an unfair forum. By mapping the warden as sole proper respondent—the so-called “immediate custodian” rule—onto immigration habeas actions, courts ensure that immigrant habeas petitioners litigate in unfair fora upon transfer, because local and state jail wardens largely do not have the “minimum contacts” necessary to satisfy in personam jurisdiction in courts outside the district of confinement.377See Int’l Shoe Co. v. Washington, 326 U.S. 310, 316 (1945); Padilla, 542 U.S. at 435–36. Immigration detention is largely carried out through contracts and agreements, most commonly IGSAs with local and state governments, which may then subcontract detention services to a private provider.378See supra note 168 and accompanying text. As applied in the immigration context, the default immediate custodian rule requires immigrants to name as a respondent a jail warden acting merely as an agent, rather than the secretary of the Department of Homeland Security, who oversees immigration detention operations. It is no wonder that the Padilla Court “decline[d] to resolve” the question of the proper respondent in the immigration habeas context,379Padilla, 542 U.S. at 435 n.8. as naming a warden completely detached from the national detention scheme serves as a meaningless procedural hurdle and functions exclusively to pigeonhole immigrant petitioners into unfair fora by requiring that they sue wherever the federal government dictates.

That the warden is inaptly named as a respondent in an immigration habeas action is evidenced by the fact that they are represented in proceedings not by a local or state attorney but by the United States Attorney’s Office (USAO).380See Immigrant Def. Project Brief, supra note 368, at 21. The USAO’s statutory responsibilities under 28 U.S.C. § 547 are “the prosecution of criminal cases brought by the Federal Government; the prosecution and defense of civil cases in which the United States is a party; and the collection of debts owed the Federal Government which are administratively uncollectible.”381Mission, Offs. of the U.S. Att’ys, U.S. Dep’t of Just., []. There is no exception in 28 U.S.C. § 547 for immigration cases. Conversely, in non-immigration-related habeas cases, the USAO does not represent wardens of state and local jails.382Immigrant Def. Project Brief, supra note 368, at 22.

Given the Habeas Act’s legislative history, the Padilla court’s own sidestep of the immigration question, and the asymmetries created in favor of the government in immigration habeas actions, federal courts should eschew the immediate custodian rule. Traditional venue principles serve as a sufficient protective measure against petitioner forum shopping, in the same way they have where Congress has authorized national service of process.383Markowitz & Nash, supra note 20, at 1190–91 (describing how the “parade of horribles” theorized in connection with national service of process—for example, prejudice to defendants and license to forum shop—did not occur because “a variety of other mechanisms protect defendants from being forced to defend themselves in far-flung locales.”).


As directly impacted immigrants, their communities, and their allies call for the abolition of immigration detention, detention’s harmful and coercive effects, as an extension of the U.S.’s larger carceral infrastructure, have gained prominence in immigration discourse. Indeed, multiple states have passed legislation to end contractual agreements with immigration agencies, creating barriers to the expansion of immigration detention in many localities. Yet, immigration detention continues to loom, reinforced by Supreme Court jurisprudence. Against this backdrop, the executive’s practice of peripheral detention and transfer—the extrication of individuals into largely remote facilities, away from the political community instrumental to vindicate rights—has adverse access-to-justice implications. This Article suggests that traditional venue factors provide a legal tool for expanding access to the courts for individuals seeking release from immigration detention, thereby further ensuring that immigrant petitioners’ claims are heard in the right fora.

*Deputy Director, Immigrant Rights Clinic, New York University School of Law. I gratefully acknowledge the invaluable comments of Mizue Aizeki, Adam Cox, Alina Das, Barry Friedman, Sarah Gillman, Emma Kaufman, Marie Mark, Shauna Marshall, David Mauer, Nancy Morawetz, Lindsay Nash, Isaac Pachulski, Margo Rofé, Sarah Vendzules, and participants in the Lutie on the Hudson Writing Workshop, the New York University School of Law Lawyering Scholarship Colloquium, and the Emerging Immigration Scholars Conference, as well as the helpful assistance of Noelia Rodriguez. I also extend my thanks to David Jimenez, Elena Hodges, and Nora Christiani for their excellent research assistance. I additionally thank the editorial staff of the Michigan Law Review for their incredibly thoughtful insights and perspectives. Notably, this Article would not have been possible without the camaraderie of the Sunrise Tribe, a group of Black women academics carrying the torch in diverse fields of study throughout the country, and without the support and understanding of my partner and children. It is dedicated to my late father, Jack Rofé, who always encouraged me to write.