Does the Constitution guarantee a habeas Privilege or not? Even though the Supreme Court appeared to answer this foundational habeas question in Boumediene v. Bush, it seemed to have unceremoniously rescinded that answer in DHS v. Thuraissigiam. This Piece, using Thuraissigiam as a starting point, links this remarkable doctrinal instability to deficits in the associated habeas theory: The legal community is short on persuasive accounts of how the Constitution “originates” the habeas Privilege.

The basic problem is rooted in the tension between a literalist reading of the Suspension Clause, which textually formulates nothing more than an antisuspension rule, and other indicia of constitutional meaning that suggest a broader habeas guarantee—things like history, Framers’ intent, public understanding, and constitutional structure. If one believes that the Constitution enshrines the Privilege, then a puzzle follows: What is the constitutional mechanism that creates it? The legal community has not coalesced around a coherent origination theory, and the Supreme Court reflexively assumes that the Suspension Clause must be the font of all constitutional habeas law.

That casual assumption undermines basic rule of law virtues associated with clarity and predictability. Because the Suspension Clause is so ill-suited to origination work (both textually and structurally), undercooked theories treating it as the exclusive source of constitutional habeas law have sown confusion. To facilitate some doctrinal stability, the legal community ought to generate stronger theories about whether and how the Constitution originates the habeas Privilege, and the Supreme Court ought to adopt one.


Boumediene v. Bush was a stinging loss for the George W. Bush Administration, 1 553 U.S. 723 (2008). during the twilight of its “war on terror.” 2 I dislike this term, but I am resigned to use it in the interests of brevity and clarity. Generally considered one of the most important habeas cases in American history, Boumediene held that the Constitution guaranteed the habeas corpus Privilege 3 I capitalize “Privilege” to make clear that I am talking about the privilege of the writ of habeas corpus. It is also capitalized in the Constitution. See U.S. Const. art. I, § 9, cl. 2. —a detained person’s right to judicial review of detention—to enemy combatant designees at the Guantanamo Bay naval facility (GTMO). 4 Boumediene, 553 U.S. at 771. Virtually everyone understood the Supreme Court to have decided that the Constitution contained an affirmative guarantee of habeas process to those detained under color of American law. The author of academe’s leading Federal Courts casebook certainly saw it that way. See Richard H. Fallon, Jr., The Supreme Court, Habeas Corpus, and the War on Terror: An Essay on Law and Political Science, 110 Colum. L. Rev. 352, 378 (2010) [hereinafter Fallon, Habeas Corpus] (“Boumediene clearly held, for the first time, that the Suspension Clause protects a right to habeas . . . , and that it does not merely prohibit complete withdrawals of whatever habeas rights Congress might have chosen to provide at any particular time.”). The decision was broadly received as a moment of profound constitutional importance; 5 See, e.g., Gerald L. Neuman, The Habeas Corpus Suspension Clause After Boumediene v. Bush, 110 Colum. L. Rev. 537, 538 (2010) [hereinafter Neuman, Habeas Corpus Suspension Clause] (“Boumediene v. Bush is a central pillar of constitutional law . . . because [it held] that Congress had violated the Suspension Clause by denying someone an adequate judicial remedy for unlawful detention.”). no less a figure than Ronald Dworkin authored Why It Was a Great Victory in the New York Review of Books. 6 Ronald Dworkin, Why It Was a Great Victory, N.Y. Rev. (Aug. 14, 2008), https://www.nybooks.com/articles/2008/08/14/why-it-was-a-great-victory [https://perma.cc/KV7A-T4SD]. Setting forth the Constitution’s framework for balancing liberty against security, Boumediene rather dramatically declared that “[t]he Framers decided that habeas corpus, a right of first importance, must be a part of that framework, a part of that law.” 7 Boumediene, 553 U.S. at 798.

Or maybe not? Twelve years later, in DHS v. Thuraissigiam, 8 140 S. Ct. 1959 (2020). the Supreme Court abruptly reversed course. 9 See Gerald Neuman, The Supreme Court’s Attack on Habeas Corpus in DHS v. Thuraissigiam, Just Sec. (Aug. 25, 2020), https://www.justsecurity.org/72104/the-supreme-courts-attack-on-habeas-corpus-in-dhs-v-thuraissigiam [https://perma.cc/C3F5-CHDE] [hereinafter Neuman, The Supreme Court’s Attack] (“Justice Alito . . . rewrote and marginalized prior precedent on habeas corpus . . . .”); Amanda L. Tyler, Thuraissigiam and the Future of the Suspension Clause, Lawfare (July 2, 2020), https://www.lawfareblog.com/thuraissigiam-and-future-suspension-clause [https://perma.cc/54SQ-F8WK] (“That the Thuraissigiam majority apparently reopens this issue begs the question of just what, if any, of the Boumediene majority opinion remains good law.”). The decision unceremoniously declared American law’s existential habeas question—whether detainees are constitutionally entitled to habeas process—unsettled. 10 See Thuraissigiam, 140 S. Ct. at 1969 n.12; id. at 1984 (Thomas, J., concurring); id. at 1997 n.1 (Sotomayor, J., dissenting). The litigation was a dispute over, among other things, whether the habeas Privilege covered certain undocumented entrants whom DHS had designated for expedited removal. The Court held that the Constitution does not require an Article III forum to test the detention at issue and, in the process, returned the existential question to a state of doctrinal uncertainty. 11 See id. at 1983 (majority opinion). Does the Constitution guarantee habeas process in an Article III court or not?

This Piece argues that this head-spinning decisional behavior reflects an underlying theoretical void: There are no broadly accepted accounts of how the Constitution “originates” the habeas Privilege. 12 When I refer to a unit of text A that “originates” constitutional power or restriction Z, I simply mean that Z can be sufficiently traced to A such that norms of legal discourse should allow a person to say that “A creates Z” or that “Z comes from A.” So the Sixth Amendment originates the right to counsel; Article I, Section 8, clause 3 originates the commerce power; Article II, Section 3 originates the President’s obligation to faithfully execute laws; and so forth. My argument here is that there is no coherent account of what originates the habeas Privilege. At the root of that problem is the casual assumption that the Constitution’s only express reference to habeas corpus, its Article I Suspension Clause, 13 U.S. Const. art. I, § 9, cl. 2. must do any and all origination. But the text of the Clause, as a literalist matter, does no such thing; it simply bars suspensions in the absence of rebellions and invasions that threaten public safety. 14 Id. If one must treat the Clause as the source of all constitutional habeas law, then norms of modern textualist interpretation force a search for magic words—express language of origination—that the Clause simply does not contain.

This Piece proceeds in four parts. Part I explains why constitutional history and structure undercut the assumption that the Suspension Clause originates the Privilege. Parts II and III show how deficient origination theory has destabilized habeas doctrine over the last twenty-five years, culminating in Thuraissigiam. The problem is not just that the dominant theory of Privilege origination is unsatisfying; it is that the Supreme Court, reflecting a blind spot in the academic literature, has failed to even put a stake in the ground. Finally, Part IV identifies the precise doctrinal questions that the Court must resolve in order to promote rule of law virtues in its constitutional habeas law and offers brief thoughts about what the answers to those questions ought to be.

I. Sourcing the Habeas Privilege

The habeas Privilege had a storied English history long before it passed into American law, and it refers generally to a detainee’s entitlement to have a judge decide whether the detention is lawful. 15 Paul D. Halliday, Habeas Corpus: From England to Empire 1–2 (2010). Professor Paul Halliday has written the superior modern history of the English common law Privilege, and Professor Amanda Tyler has done the same for its statutory counterpart. See id.; Amanda Tyler, Habeas Corpus in Wartime: From the Tower of London to Guantanamo Bay 100–56 (2017). The history of suspension is almost as long but considerably less  inspiring,  as  suspensions  typically  mean  that  jailers  do  not  have  to  justify custody. 16 See Paul D. Halliday & G. Edward White, The Suspension Clause: English Text, Imperial Contexts, and American Implications, 94 Va. L. Rev. 575, 613–23 (2008) (chronicling English suspension from the late seventeenth century through the American Revolution). The Constitution contains no magic words that create the Privilege; its existence is simply assumed by the Suspension Clause: “The Privilege of the Writ of Habeas Corpus shall not be suspended, unless when in Cases of Rebellion or Invasion the public Safety may require it.” 17 U.S. Const. art. I, § 9, cl. 2.

There is a latent but serious problem in how the American legal community, including its Supreme Court Justices, talk about the Privilege. That the Constitution guarantees habeas process does not mean that the Suspension Clause must be the exclusive instrument of that guarantee. Nevertheless, virtually everyone casually assumes that, if  the  Constitution  originates  a  Privilege,  the  Clause  must  be  the  operative provision. 18 See, e.g., Boumediene v. Bush, 553 U.S. 723, 768 (2008) (referring to “rights under the Suspension Clause”); Immigr. & Naturaliz. Serv. v. St. Cyr, 533 U.S. 289, 301 (2001) (discussing whether the Suspension Clause “protect[ed] the writ” as it existed in 1789); Paul Diller, Habeas and (Non-)Delegation, 77 U. Chi. L. Rev. 585, 602 (2010) (referring to the “habeas-providing power in the Suspension Clause”); Philip Hamburger, Unconstitutional Conditions: The Irrelevance of Consent, 98 Va. L. Rev. 479, 511 n.50 (2012) (suggesting that the Suspension Clause creates a “right of habeas”); Neuman, Habeas Corpus Suspension Clause, supra note 5, at 548–50 (observing that Boumediene ratified this accepted view). The Clause does not just impose an antisuspension rule, the thinking goes, but also forbids habeas restrictions that do not qualify as suspensions.

The assumption that the Clause originates the Privilege is a touch bewildering, however, because it is inconsistent with some pretty basic constitutional text, structure, and history. The Privilege and suspension are related but distinct phenomena. Generally speaking, the Privilege refers to a remedy in which a judge decides whether custody is lawful, and suspension refers to legislation that purports to withhold that remedy. The English suspension statutes animating the Clause usually authorized arrest and detention “without bail or mainprize” and displaced law and custom inconsistent with that authority. 19 See infra notes 21–27 and accompanying text. The notion that constitutional text restricting a legislative suspension power would also originate the Privilege to be suspended betrays confusion about important distinctions between the two phenomena.

Start with text. The words in the Suspension Clause limit the congressional authority to suspend, but not power to withhold or restrict the Privilege in other ways. The commonly understood definition of a suspension in 1789 was, in turn, quite narrow. 20 See generally Halliday, supra note 15, at 247–56 (detailing English suspension practice through the late eighteenth century). English suspensions were a defining revolutionary grievance against the Crown. 21 See id. at 253. Lord North suspended the Privilege six times during the American struggle with George III. 22 See An Act for Further Continuing an Act 1782, 22 Geo. 3 c. 1 (Gr. Brit.) (renewing the suspension); An Act for Further Continuing an Act 1781, 21 Geo. 3 c. 2 (Gr. Brit.) (renewing the suspension); An Act for Further Continuing an Act 1780, 20 Geo. 3 c. 5 (Gr. Brit.) (renewing the suspension); An Act for Further Continuing an Act 1779, 19 Geo. 3 c. 1 (Gr. Brit.) (renewing the suspension); An Act for Continuing an Act 1778, 18 Geo. 3 c. 1 (Gr. Brit.) (renewing the suspension); An Act to Impower His Majesty to Secure and Detain Persons Charged with, or Suspected of, the Crime of High Treason, Committed in Any of His Majesty’s Colonies or Plantations in America, or on the High Seas, or the Crime of Piracy 1777, 17 Geo. 3 c. 9 (Gr. Brit.) [hereinafter Treason Act 1777] (enacting the suspension in the first instance). These suspensions empowered the Crown, for a time not exceeding a year, to lawfully arrest and imprison colonists without “bail or mainprize.” 23 See Halliday, supra note 15, at 248–52 (showing that the average pre-1777 suspension statute expired after five months, and that authorization of arrest and detention “without bail or mainprize” in the suspension statutes from 1777–1783 was consistent with earlier suspension practice). Suspension statutes did not use magic words like “suspension” or “habeas corpus,” but nonetheless contained text displacing the statutory and common law judicial process that effectuated the Privilege. 24 See id. at 248–49. For example, the 1777 English suspension provided that all persons committed in the colonies or the high seas for treason or piracy “shall and may be thereupon secured and detained in safe Custody, without Bail or Mainprize . . . any Law, Statute, or Usage, to the contrary in anywise notwithstanding.” 25 Treason Act 1777, 17 Geo. 3 c. 9 (Gr. Brit.) (emphasis added). The same language typified preconstitutional suspension statutes in the American colonies, which “authorized and empowered” executives to arrest and detain certain categories of people for a limited period of time. 26 See Amanda L. Tyler, Suspension as an Emergency Power, 118 Yale L.J. 600, 665 (2009) [hereinafter Tyler, Suspension as Emergency Power] (internal quotation marks omitted) (quoting Act of Nov. 10, 1786, ch. 41, 1786 Mass. Acts 102, 102).

So the text of the Suspension Clause is an awkward fit for an argument that the Clause originates the Privilege, and constitutional structure compounds that awkwardness. The Clause appears in Article I, Section 9, which is a set of restrictions on legislative powers enumerated in Article I, Section 8, or necessary and proper thereto. Every single clause of Section 9 contains express text—a “no [X] shall be” or an “[X] shall not be”—restricting the legislature. In other words, basic constitutional structure indicates that Article I, Section 9 does not originate the suspension power, let alone the Privilege. The suspension power, which the Clause limits textually, must be auxiliary to one of the powers that Article I, Section 8 enumerates, although the provenance of that power is itself undertheorized. 27 The suspension power might be necessary and proper to the enumerated powers to provide for the common defense, to control naturalization, and to govern the land and naval forces. See U.S. Const. art. I, § 8, cls. 4, 12–16, 18. One scholar has argued that the suspension power might be auxiliary to the power to “suppress insurrection and repel invasion,” William F. Duker, A Constitutional History of Habeas Corpus 131 (1980), although the actual enumerated power refers to the calling forth of the militia to do those things, U.S. Const. art. I, § 8, cl. 15. Another argument might be that the suspension power is necessary and proper to the power to constitute the federal judiciary. See U.S. Const. art. I, § 8, cl. 9.

The Suspension Clause looks much more like a reflection of the Privilege than its source; text and structure indicate that the Clause performs no origination. In terms of what the Framers actually intended, understood, and communicated to their constituencies, the Privilege might not “come from” the Constitution at all—at least in a positivist sense. It is well established that the Framers were legal naturalists and believed that the Privilege, along with many other rights, preexisted acts of statutory or constitutional creation. 28 See Akhil Reed Amar, The Bill of Rights and the Fourteenth Amendment, 101 Yale L.J. 1193, 1206–08 (1992) (contrasting early American understanding of the Bill of Rights with that of “modern day legal positivists”); Frederick Mark Gedicks, An Originalist Defense of Substantive Due Process: Magna Carta, Higher-Law Constitutionalism, and the Fifth Amendment, 58 Emory L.J. 585, 636–38 (2009) (collecting Founding-Era authority regarding the relationship between the Constitution and “natural and customary rights”); Howard Jay Graham, Our “Declaratory” Fourteenth Amendment, 7 Stan. L. Rev. 3, 3–4 (1954) (“Constitutions and amendments, therefore, were declaratory, even though by definition they were also fundamental and supreme law.”). So extreme was this thinking with respect to the habeas Privilege that the Clause itself memorializes the naturalist assumption. The desire to identify a localized string of constitutional text that expressly originates a power or limit thereupon is an impulse of modern positivism.

The naturalism-versus-positivism point is not some creative gloss on the Privilege. The Framers very much believed the Privilege to be inviolable, except on conditions of suspension. 29 See Lee Kovarsky, Prisoners and Habeas Privileges Under the Fourteenth Amendment, 67 Vand. L. Rev. 609, 618–20 (2014). A basic theme emerges from the Convention’s drafting history: The Framers fought about whether the Privilege could ever be suspended, not over whether the Constitution guaranteed it. 30 See id. Georgia, North Carolina, and South Carolina lodged the initial objection that the Privilege was not sufficiently secured. 2 The Records of the Federal Convention of 1787, at 438 (Max Farrand ed., 1911). These states were ultimately mollified by assurances that, despite the peculiar wording of Article I, Section 9, the habeas Privilege was in fact constitutionally guaranteed. See Francis Paschal, The Constitution and Habeas Corpus, 1970 Duke L.J. 605, 608–11. Having emerged from the Room Where It Happened, 31 See Leslie Odom Jr., Lin-Manuel Miranda, Daveed Diggs, Okieriete Onaodowan & Original Broadway Cast of Hamilton, The Room Where It Happens, on Hamilton: An American Musical (Atlantic Records 2015). Alexander Hamilton devoted Federalist No. 84 largely to the proposition that certain entitlements lacking express text of creation nonetheless passed into American constitutional law, and he even underscored the special status of the Privilege on the ground that the Constitution provided for the “establishment of the writ of habeas corpus.” 32 See The Federalist No. 84 (Alexander Hamilton) (emphasis added). Hamilton also stated that the function of the English Habeas Corpus Act was “provided for . . . in the plan of the convention.” The Federalist No. 83 (Alexander Hamilton) (Jacob E. Cooke ed., 1961).

Even though the Suspension Clause does not textually originate the Privilege in the sense that many legal positivists contemplate, everyone still talks about it that way—especially the Supreme Court. 33 See, e.g., supra note 18 (citing Boumediene v. Bush, 553 U.S. 723, 768 (2008) and Immigration & Naturalization Service v. St. Cyr, 533 U.S. 289 (2001)). And therein lies the mischief. If one believes that the Suspension Clause is the exclusive source of constitutional habeas law, and if the text of the Clause does no more than restrict formal suspension, then one could faithfully ask whether there are any constitutional guarantees for habeas process beyond the antisuspension rule. Professor Rex Collings took that position in an influential law review article, 34 See Rex A. Collings, Jr., Habeas Corpus for Convicts—Constitutional Right or Legislative Grace?, 40 Calif. L. Rev. 335, 341–42 (1952) (arguing that the Constitution does not guarantee any habeas process during periods of nonsuspension). and it has wormed its way into the U.S. Reports. Parts II and III document that process.

II. Pre-Thuraissigiam Decisional Law

Because the Suspension Clause is (1) the only express reference to the Privilege in the Constitution and also (2) comprised of text that is facially nonoriginating, the assumption that the Clause must do any and all origination work has created uncertainty around whether the Constitution’s habeas guarantee amounts to more than an antisuspension rule. Part II links this basic puzzle to the Supreme Court’s on-again, off-again relationship with the doctrine of “unconstitutional suspension,” a bedeviling construct behind much of the decisional instability that concerns me here.

The modern problems started with two decisions about statutory substitutes for postconviction process. United States v. Hayman was a 1952 decision involving postconviction process for those convicted of federal crimes. 35 342 U.S. 205 (1952). The contested statute required process functionally identical to a habeas proceeding, but required that process to take place in the sentencing court rather than in the court with jurisdiction over the jailer. 36 See 28 U.S.C. § 2255 (2018). Hayman held that, because the statute at issue had a safety valve applicable when substitute process was “inadequate or ineffective,” there was no constitutional issue presented. 37 See Hayman, 342 U.S. at 223. Swain v. Pressley 38 430 U.S. 372 (1977). was a 1977 case involving similar legislation that moved postconviction process in the District of Columbia to a federal court with local jurisdiction. 39 See D.C. Code § 23-110(g) (2020). Pressley, however, used a new formulation suggesting that habeas restrictions are forbidden only if they amount to suspension: “[T]he only constitutional question presented is whether [the substitute] should be regarded as a suspension of the Great Writ . . . .” 40 See Pressley, 430 U.S. at 381. This construct—the “unconstitutional suspension”—became quite destabilizing.

In Felker v. Turpin, 41 518 U.S. 651 (1996). the Supreme Court considered the constitutionality of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). 42 Antiterrorism and Effective Death Penalty Act of 1996, Pub. L. No. 104-132, 110 Stat. 1214 (codified as amended in scattered titles of the U.S.C.). More specifically, Felker involved severe restrictions on “successive” habeas petitions filed on behalf of people convicted of crimes in state courts—and the Court had to decide whether those restrictions were constitutional. The provisions at issue did things like bar successive litigation of constitutional claims presented in earlier petitions 43 See 28 U.S.C. § 2244(b)(1) (2018). and restrict relief for adjudication of new claims. 44 See id. § 2244(b)(2). In short, these provisions were nothing like traditional suspension statutes that granted jailers a time-limited immunity from judicial review during rebellion or invasion.

Nevertheless, and following Pressley’s lead, Felker framed the issue not as a question about whether Congress impermissibly restricted the Privilege during a period of nonsuspension, but instead as whether the successive petition rules “suspend[ed] the Writ of Habeas Corpus in violation of Article I, [Section] 9, clause 2, of the Constitution.” 45 Felker, 518 U.S. at 663. The Supreme Court answered in the negative, holding that the contested provisions did “not amount to a ‘suspension’ of the writ.” 46 Id. at 664. Felker, then, vividly exposes the basic problem. It assumed that the Suspension Clause originates any and all constitutional habeas law, read the Clause as a pure antisuspension rule, and thereby indicated that the Supreme Court could invalidate a legislative restriction only when that restriction qualifies as a suspension.

The next major constitutional habeas case was, like Thuraissigiam, a dispute over restrictions on noncitizen removability in 1996 immigration legislation. In INS v. St. Cyr, the Supreme Court explored the constitutional problems associated with a provision that arguably stripped judicial review over certain deportation orders. 47 533 U.S. 289, 297 (2001). Rather than striking the provision down, the Court engaged in some constitutional avoidance, determining that the provisions did not actually strip federal courts of habeas jurisdiction to hear the noncitizen’s challenge. 48 See id. at 314. In so doing, however, the Court necessarily analyzed the constitutional question avoided.

Quoting Felker, the Supreme Court expressly identified the Suspension Clause as the source of the Privilege: “[A]t the absolute minimum, the Suspension Clause protects the writ as it existed in 1789.” 49 Id. at 301 (internal quotation marks omitted) (quoting Felker, 518 U.S. at 663–64). But whereas Felker contains phrasing indicating that a habeas restriction must amount to a suspension before it may be declared unconstitutional, St. Cyr contains no such language. Instead of considering whether the restrictions amounted to “unconstitutional suspension,” the Court simply analyzed whether the “Suspension Clause protects” Article III jurisdiction over certain types of challenges. 50 See id. at 304; see also id. at 305 (observing that “a serious Suspension Clause issue would be presented” were it to interpret the provisions at issue to strip habeas jurisdiction over St. Cyr’s claim). In short, St. Cyr assumed that the Suspension Clause originated all constitutional habeas law, but did not indicate that such law was limited to an antisuspension rule.

In St. Cyr, Justice Scalia penned what has become a famous dissent, making some of the textual (but not the structural) points Part I makes. Citing Rex Collings, he observed that the Suspension Clause does not “guarantee any content to” or “the existence of” the habeas Privilege. 51 Id. at 337 (Scalia, J., dissenting). Noting the near universal Founding-era understanding of what a suspension statute looked like, he explained that the contested habeas amendments did not resemble a suspension contemplated by the Clause. 52 See id. at 337–38. From that proposition, Justice Scalia concluded that the Constitution guaranteed no habeas Privilege at all. 53 There is, as explained above, overwhelming evidence that the reason that the Suspension Clause omits words purporting to declare the Privilege itself inviolable is that naturalist Framers believed the Privilege was constitutionally protected without such textual origination. See supra note 29 and accompanying text. In other words, Justice Scalia agreed that the Suspension Clause originated all constitutional habeas law, but believed such law to include only an antisuspension rule.

Justice Scalia appeared to abandon his St. Cyr position three years later, in Hamdi v. Rumsfeld. 54 542 U.S. 507 (2004). In Hamdi, the Supreme Court had to decide whether the military could hold an American citizen indefinitely under the 2001 Authorization for Use of Military Force (AUMF). 55 Authorization for Use of Military Force of 2001, Pub. L. No. 107-40, 115 Stat. 224 (codified at 50 U.S.C. § 1541 note (2018)). The decision was confusing and there was no majority opinion, but Justice Scalia’s dissent is the subject of my interest here. He would have held that the executive cannot indefinitely detain American citizens without criminal charges, full stop. 56 See Hamdi, 542 U.S. at 554 (Scalia, J., dissenting).

In Hamdi, Justice Scalia wrote that due process was “the right secured” and that habeas corpus was “the instrument by which due process could be insisted upon by” an imprisoned citizen. 57 Id. at 555–56. In extolling the Privilege as a means of checking the political branches, he derided the idea that Congress could moot it through legislation that amounted to something other than suspension:

If the Suspension Clause does not guarantee [that a citizen will be tried or released] unless the [suspension conditions] exist and the grave action of suspending the writ has been taken; if it merely guarantees the citizen that he will not be detained unless Congress by ordinary legislation says he can be detained; it guarantees him very little indeed. 58 Id. at 575.

This language is inconsistent with the thrust of his St. Cyr dissent, in which he had maintained that the Privilege was a matter of legislative grace. 59 See supra notes 52–54 and accompanying text. In Hamdi and St. Cyr, therefore, Justice Scalia appears to take opposite positions on the question of whether a statute that does not amount to a suspension can restrict or eliminate the Privilege. Justice Scalia’s repositioning demonstrates how doctrinal instability is not just about changed Supreme Court composition or the fact that different Justices write majority opinions in different cases.

Boumediene shows similar repositioning—within a single Supreme Court opinion. In the run up to that decision, Congress had used Section 7 of the Military Commissions Act of 2006 (MCA) to strip habeas jurisdiction over combatant status determinations for noncitizen GTMO detainees, thereby curtailing the scope of issues reviewable in Article III courts. 60 Pub. L. No. 109-366, 120 Stat. 2600 (codified as amended in scattered titles of the U.S.C.). Boumediene held that the restrictions on Article III jurisdiction were unconstitutional 61 See Boumediene v. Bush, 553 U.S. 723, 733 (2008). —the first holding that a statutory restriction on Article III jurisdiction violated the Constitution since the Supreme Court decided United States v. Klein in 1871. 62 80 U.S. 128 (1871). Boumediene is an intricate opinion with many moving parts, but this Piece focuses on its internally inconsistent discussion of the constitutional habeas guarantee.

As with any modern case assuming that the Suspension Clause originates the Privilege, even the best of Boumediene’s phrasing is awkward. The Supreme Court referred to detainees “invoking the protections of the Suspension Clause,” 63 Boumediene, 553 U.S. at 739. remarked that “the protections of the Suspension Clause” might have expanded after 1789, 64 Id. at 746. and formulated the question presented as whether, in restricting habeas jurisdiction without purporting to suspend, Congress “acted in violation of” that same provision. 65 Id. This language suggests that the Court believed that the Clause originates the Privilege and that the Constitution contains more than an antisuspension rule. After all, Boumediene noted that “[t]he MCA does not purport to be a formal suspension of the writ; and the Government, in its submissions to us, has not argued that it is.” 66 Id. at 771.

In other places, however, Boumediene reverts to language indicating that a habeas restriction must amount to a suspension in order to run afoul of the Constitution—a framework that does not entail an affirmative constitutional guarantee. 67 Perhaps I am undercrediting Boumediene. One might read its references to suspension as shorthand for the idea that, once Congress has ordained and established Article III courts with habeas jurisdiction, any restriction—permanent or temporary—is constitutionally verboten. In that case, the Court would be imprudently relying on a context-independent definition of suspension but would not be guilty of analytic inconsistency. For example, the global introduction to the opinion declares that the MCA “operates as an unconstitutional suspension of the writ.” 68 Boumediene, 533 U.S. at 733. After determining that the Article III process that was available to detainees was an insufficient habeas substitute, the Court similarly explained that the MCA “thus effects an unconstitutional suspension of the writ.” 69 Id. at 792. Writing for four dissenters, Justice Scalia seemingly returned to his St. Cyr position: The Constitution restricts only acts of suspension and the MCA was not a suspension statute. 70 Id. at 827 (Scalia, J., dissenting). But see Fallon, Habeas Corpus, supra note 4, at 378 (reading Justice Scalia to have abandoned this argument in his Boumediene dissent).

Awkwardness and internal inconsistency notwithstanding, Boumediene was not entirely ambiguous. The dissenting opinions clearly believed the Supreme Court’s holding to be that the Constitution enshrined more than an antisuspension rule. 71 Boumediene, 553 U.S. at 804, 805, 815 (Roberts, C.J., dissenting); id. at 827 (Scalia, J., dissenting). The majority opinion includes a lengthy discussion of whether the legislatively abridged Article III process was an adequate substitute for habeas corpus, 72 Id. at 771–92 (majority opinion). and such material would make little sense unless the Court believed that the Constitution required substitutional adequacy. Instead, what the Court seemed to have in mind was the proposition that any violation of the Privilege was an “unconstitutional suspension”—which is an odd way to refer to a restriction that was not in fact litigated as, or judicially determined to be, a suspension. 73 See id. at 771. Observations like these, however—about a case’s essential holding, in light of all considerations—mean little when formulated through decisional language that is so varied and imprecise.

Boumediene follows the modern trend insofar as it sources all constitutional habeas law to the Suspension Clause. This linkage creates a problem because, if one is a literalist about the text, the Clause forbids only suspensions. A doctrinal test under which the Clause originates all constitutional habeas law will return unpredictably to the “unconstitutional suspension” construct—the idea that only restrictions amounting to suspensions are subject to constitutional limits. And because a properly contextualized definition of suspension is quite narrow, 74 See supra notes 20–27 and accompanying text. the failure to coherently articulate a theory of Privilege origination will tend to undermine the writ’s liberty-enhancing and power-separating functions. 75 See, e.g., 1 William Blackstone, Commentaries 120–45 (referring to the Privilege as an “inherent right” of the “liberty of the subject” and the “bulwark of our liberties”); Neuman, Habeas Corpus Suspension Clause, supra note 5, at 548–51 (discussing Boumediene’s views about how the Privilege separates powers).

III. Thuraissigiam

A differently composed Supreme Court decided Thuraissigiam twelve years later, with Justice Kavanaugh replacing Boumediene’s author, Justice Kennedy. 76 See Robert Barnes, Justice Kennedy Asked Trump to Put Kavanaugh on Supreme Court List, Book Says, Wash. Post (Nov. 21, 2019), https://www.washingtonpost.com/politics/courts_law/justice-kennedy-asked-trump-to-put-kavanaugh-on-supreme-court-list-book-says/2019/11/21/3495f684-0b0f-11ea-8397-a955cd542d00_story.html (on file with the Columbia Law Review). Thuraissigiam’s fraught relationship with Boumediene illustrates how the unconstitutional suspension construct obscures answers about whether (and how) the Constitution protects the Privilege when it is not suspended. 77 I acknowledge the possibility that the Boumediene Court was not actually attempting to parse a distinction between an “unconstitutional suspension” and an unconstitutional restriction on the Privilege in the absence of suspension—and that the presence of the former term in the opinion, alongside the latter, was just an unfortunate choice of language. See supra note 68 and accompanying text. If the Justices joining the Thuraissigiam majority understood that Boumediene intended no such distinction, then its deliberate exploitation is even more troubling.

A. Thuraissigiam and the Existential Question

At issue in Thuraissigiam were certain provisions from the Illegal Immigration Reform and Immigrant Responsibility Act of 1996 (IIRIRA). 78 Pub. L. No. 104-208, 110 Stat. 3009-546 (codified as amended in scattered sections of 8 U.S.C.). Before IIRIRA, noncitizens subject to exclusion or deportation could generally seek habeas review of the orders removing them. 79 See David Cole, Jurisdiction and Liberty: Habeas Corpus and Due Process as Limits on Congress’s Control of Federal Jurisdiction, 86 Geo. L.J. 2481, 2496 (1998). A precise version of the history is a little more complicated, and the subjects of certain deportation orders were, for a time, funneled to the federal courts of appeal for judicial review under a nonhabeas statute. See David M. McConnell, Judicial Review Under the Immigration and Nationality Act: Habeas Corpus and the Coming of REAL ID (1996–2005), 51 N.Y.L. Sch. L. Rev. 75, 81–82 (2006). IIRIRA, however, created an expedited removal process for certain undocumented entrants and border arrivals, and the expedited removal process severely restricted recourse to Article III courts for asylum seekers who found themselves in such proceedings. 80 See 8 U.S.C. § 1252(e)(2) (2018). For a general summary of the expedited removal provisions at issue, see generally Neuman, The Supreme Court’s Attack, supra note 9. IIRIRA had created a global term, “removal,” that covered both deportation of noncitizens physically present inside the country and exclusion of noncitizens seeking entry. 81 See 8 U.S.C. § 1101.

Officials apprehended Thuraissigiam about twenty-five yards past the U.S. side of the United States–Mexico border, after he had entered unlawfully. 82 Dep’t of Homeland Sec. v. Thuraissigiam, 140 S. Ct. 1959, 1967 (2020). DHS officials designated him for expedited removal, at which point Thuraissigiam claimed asylum and asserted that he feared returning to Sri Lanka because a group of men had abducted and beaten him, and that he was unsure the Sri Lankan government would be able to protect him in the future. 83 Id. After an asylum screening interview conducted through an interpreter, an asylum officer and a supervisor both determined that Thuraissigiam had failed to demonstrate a “credible fear” of persecution on a protected ground. 84 See id. at 1968. Thuraissigiam then received de novo asylum screening before an (administrative) immigration judge, who agreed with the asylum officers’ decision and determined that Thuraissigiam had not satisfied the credible fear requirement. 85 Id. Thuraissigiam was again designated for expedited removal, which precluded him from presenting the merits of his asylum claim to an immigration judge in a full removal proceeding. 86 Id. Pursuant to statute, officials detained Thuraissigiam throughout the process. 87 See 8 U.S.C. § 1225(b)(1)(B)(iii)(IV) (2018); Thuraissigiam, 140 S. Ct. at 1967–68.

IIRIRA limited what Thuraissigiam could argue when he moved the litigation to an Article III court. Pursuant to 8 U.S.C. § 1252(e)(2), he could obtain relief only if: (1) he was in fact a citizen, (2) he was not in fact subject to and removed under the expedited removal statute, or (3) he had existing permanent immigration status as a lawful permanent resident, refugee, or asylee. 88 See Neuman, The Supreme Court’s Attack, supra note 9. He was not permitted, in other words, to contest the lawfulness of Section 1252(e)(2) itself, to argue that the procedures used to decide the credible fear screening question failed to comply with Section 1252(e)(2) or the Constitution, or to dispute facts and the application of asylum law thereto.

Thuraissigiam held that the habeas Privilege did not cover noncitizens “claim[ing] the right to enter or remain in a country or obtain administrative review potentially leading to that result.” 89 Thuraissigiam, 140 S. Ct. at 1969. The breadth of the Thuraissigiam holding will be unusually dependent on subsequent interpretation. The Court did not seem to contemplate that the decision would reach (1) asylum seekers challenging their detention directly, in a case where they actually sought release, or (2) migrants with a longer time in, and more connections to, the United States. Justice Alito wrote for the Court and leaned heavily on the absence of any preconstitutional case authorizing a noncitizen “to remain in a country other than his own or to obtain administrative or judicial review leading to that result,” and he waved off preconstitutional history showing that English judges used the habeas writ to relieve a broad spectrum of unlawful custody. 90 See id. at 1971–75.

The parallels between Boumediene and Thuraissigiam were obvious. Both involved noncitizens bearing unclear bundles of substantive rights and seeking Article III remedies. And both involved statutory provisions expressly designed to restrict Article III review to a set of narrow legal questions, with tightly circumscribed authority to review or decide factual predicates that executive officers determined in the first instance. That the outcomes differed was not altogether surprising, given the Kennedy-to-Roberts change in the Supreme Court’s median voter. 91 Alicia Parlapiano & Jugal K. Patel, With Kennedy’s Retirement, the Supreme Court Loses Its Center, N.Y. Times (June 27, 2018), https://www.nytimes.com/interactive/2018/06/27/us/politics/kennedy-retirement-supreme-court-median.html (on file with the Columbia Law Review).

Thuraissigiam’s language on Privilege origination, however, was shocking. The Supreme Court expressly reserved the question that Boumediene had appeared to decide, and observed that whether the Constitution “guarantees the availability of the writ” remains the “subject of controversy.” 92 Thuraissigiam, 140 S. Ct. at 1969 n.12. Without mentioning Boumediene at all, Thuraissigiam noted that St. Cyr contained spirited debate about “whether the Clause independently guarantees the availability of the writ or simply restricts the temporary withholding of its operation.” 93 Id. The Court declared that it would not “revisit the question,” failing to mention that almost everyone had read Boumediene to have resolved it. 94 See id.; supra note 4 and accompanying text. Thuraissigiam reasoned that there was no reason to decide whether the Constitution’s habeas guarantee was more than an antisuspension rule, given its determination  that  the  expedited  removal  order  would  lie  outside  any  such guarantee. 95 Thuraissigiam, 140 S. Ct. at 1969 n.12.

Lest one was to read the reference to St. Cyr and omission of Boumediene as something other than a notation that the existential question remained undecided, the Thuraissigiam dissent extinguishes most doubt. The dissenters believed that the Constitution should be interpreted so as to originate the Privilege—otherwise they could not dissent on the habeas issue—but they did not assert that Boumediene or St. Cyr had actually decided the question. 96 See id. at 1997 (Sotomayor, J., dissenting). By contrast, Justice Breyer’s concurrence, joined by Justice Ginsburg, eschews the language of “unconstitutional suspensions” and abides by St. Cyr’s more careful phrasing indicating that the Suspension Clause could originate the Privilege. See, e.g., id. at 1989–90 (Breyer, J., concurring) (referring to what the Suspension Clause “protects” or “assures” or “guarantees” during periods of nonsuspension). Justice Sotomayor’s opinion, which Justice Kagan joined, states that “[t]he Court wisely declines to explore whether the Suspension Clause independently guarantees the availability of the writ or simply restricts the temporary withholding of its operation.” 97 See id. at 1997 n.1 (Sotomayor, J., dissenting).

Perhaps unsurprisingly, Justice Thomas was prepared to decide the existential question, using his concurrence to champion a version of the logic expressed in Justice Scalia’s St. Cyr dissent. 98 See id. at 1983 (Thomas, J., concurring). Justice Thomas noted that the naked text of the Suspension Clause does not itself authorize courts to grant habeas relief. 99 See id. at 1984. Given his textualist priors, however, his inference was not that the Privilege referred to in the Clause originates elsewhere in the Constitution, but that the Privilege is not in fact constitutionally guaranteed. 100 See id. at 1987–88. Instead, he reasoned, the constitutional guarantee is no more than an antisuspension rule, with suspension defined very narrowly as a statute authorizing the executive “to detain without bail or trial based on [mere] suspicion.” 101 Id. at 1988. Justice Thomas correctly asserts that English suspension statutes legalized otherwise unlawful preventative detention, but he does not seem to acknowledge that they also displaced inconsistent statutory and common law that would interfere with that detention. See supra note 26 and accompanying text.

After Thuraissigiam, the existential question appears to remain open. Seven Justices—the five in the majority, plus Justices Sotomayor and Kagan—signed opinions that seem to indicate that Boumediene did not in fact resolve it. Having refused to read Boumediene as more than a rule against “unconstitutional suspensions,” Thuraissigiam revives the decisional construct at the heart of the instability that this Piece catalogues. After Thuraissigiam, and with respect to the question whether the Constitution’s habeas guarantee is more than an antisuspension rule, we are back where we started.

B. Secondary Instability

Setting aside the existential question, Thuraissigiam illustrates lower-grade instability linked to the absence of a coherent theory of Privilege origination. Specifically, the Supreme Court escaped holdings that the Constitution guaranteed certain forms of habeas process—basically, the Privilege—by citing the earlier precedents’ failure to expressly designate the Suspension Clause as the originating source of the rule. Indeed, it was this slippage that allowed the Thuraissigiam majority to get out from under the so-called “finality-era” immigration cases holding that habeas review of removal orders persisted because the Constitution required it.

The finality era refers to the period during which Congress had declared certain immigration decisions, well, “final.” 102 See Thuraissigiam, 140 S. Ct. at 1977. The era began in 1891, before which time Congress permitted judicial review of all legal and factual questions resolved by exclusion and deportation orders. 103 See United States v. Jung Ah Lung, 124 U.S. 621, 632 (1888) (rejecting challenge to federal courts’ authority to review executive authority in immigration matters). Between 1891 and 1917, Congress passed a series of laws making administrative orders final and thereby immune from judicial review, 104 See Immigration Act of 1917, ch. 29, § 17, 39 Stat. 874, 887 (repealed 1952); Immigration Act of 1907, ch. 1134, § 25, 34 Stat. 898, 906–07 (repealed 1917); Act of Aug. 18, 1894, ch. 301, 28 Stat. 372, 390 (repealed 1966); Act of Mar. 3, 1891, ch. 551, § 8, 26 Stat. 1084, 1085 (in force). and the finality era continued until 1952. 105 See Shaughnessy v. Pedreiro, 349 U.S. 48, 51–52 (1955) (interpreting 1952 legislation to permit Article III review, other than habeas review, of deportation orders). Habeas review of legal issues arising out of the immigration decisions persisted nonetheless 106 See infra notes 108, 110–112 and accompanying text. —that is, notwithstanding statutory finality—and for fifty years the Court described the persistence of such habeas review as something required by the Constitution. 107 See infra notes 113, 116 (quoting 1953 and 2001 cases for the proposition).

In Nishimura Ekiu v. United States, the very first finality-era Supreme Court decision, the Court determined that Article III courts retained habeas jurisdiction to review legal questions, observing: “An alien immigrant, prevented from landing by any such officer claiming authority to do so under an act of [C]ongress, and thereby restrained of his liberty, is doubtless entitled to a writ of habeas corpus to ascertain whether the restraint is lawful.” 108 142 U.S. 651, 660 (1892). The Court held that the noncitizen seeking entry was “doubtless entitled” to habeas process notwithstanding the observation in the next sentence, that “the final determination of . . . [facts] may be entrusted by [C]ongress to executive officers . . . and no other tribunal, unless expressly authorized by law to do so, is at liberty to reexamine or controvert the sufficiency of the evidence on which he acted.” 109 Id. There is no lawyerly parsing necessary to discern the passage’s meaning: A noncitizen seeking entry was “doubtless entitled” to habeas process notwithstanding the observation that Congress was free to enact statutes granting executive officials authority to make final factual determinations.

The most famous statement about the provenance of finality-era Article III process appeared in Heikkila v. Barber, which declared that the finality-era statutes had  eliminated  all  judicial  review  “except  insofar  as  it  was  required  by  the Constitution.” 110 See 345 U.S. 229, 234–35 (1953) (emphasis added) (“Congress had intended to make these administrative decisions nonreviewable to the fullest extent possible under the Constitution.”). What was “required by the Constitution,” moreover, was understood to mean review of legal questions. 111 See, e.g., Shaughnessy v. United States ex rel. Mezei, 345 U.S. 206, 213 (1953) (“Concededly, his movements are restrained by authority of the United States, and he may by habeas corpus test the validity of his exclusion.”); United States ex rel. Knauff v. Shaughnessy, 338 U.S. 537, 540 (1950) (using habeas process to review the lawfulness of the Attorney General’s discretion to refuse admission to a spouse of a military servicemember); United States ex rel. Johnson v. Shaughnessy, 336 U.S. 806, 808 (1949) (using habeas process to review an allegation that a mental defect sufficient to bar admission was adjudicated without proper procedure); Yee Won v. White, 256 U.S. 399, 400 (1921) (using habeas process to review the legal question of whether a Chinese laborer legally residing in the United States may “demand that his wife and minor children be permitted to come into this country and reside with him notwithstanding they were born in China and have never resided elsewhere”); Gegiow v. Uhl, 239 U.S. 3, 9 (1915) (“And when the record shows that a commissioner of immigration is exceeding his power, the alien may demand his release upon habeas corpus.”); Gonzales v. Williams, 192 U.S. 1, 7 (1904) (using habeas to review a question about the legal status of Puerto Ricans under immigration laws). Heikkila therefore crystalized a half-century framework for judicial review of immigration orders. The notion that finality-era cases entailed constitutionally compelled Article III process for immigration orders was reaffirmed in St. Cyr, which cited Heikkila in conjunction with the observation that “[b]ecause of [the Suspension] Clause, some ‘judicial intervention  in  deportation  cases’  is  unquestionably  ‘required  by  the Constitution.’” 112 See Immigr. & Naturaliz. Serv. v. St. Cyr, 533 U.S. 289, 300 (2001) (quoting Heikkila, 345 U.S. at 235).

In order to slip the finality-era precedent holding that the Constitution guaranteed the Privilege to certain prospective entrants, the Thuraissigiam majority focused on what this Piece has called origination. Specifically, the majority emphasized that although the early cases had observed a constitutional habeas guarantee, they did not hold that the Suspension Clause was its originating source. 113 Dep’t of Homeland Sec. v. Thuraissigiam, 140 S. Ct. 1959, 1976 (2020). For example, Thuraissigiam elided Heikkila’s language that finality-era statutes had eliminated all judicial review “except insofar as it was required by the Constitution” by distinguishing between a Privilege originating in the Suspension Clause and a Privilege originating in some other constitutional provision. 114 Id. at 1980–81 (emphasis added).

With respect to Nishimura Ekiu, Thuraissigiam explained, “What is critical . . . is that the Court did not hold that the Suspension Clause imposed any limitations on the authority of Congress to restrict the issuance of writs of habeas corpus in immigration matters.” 115 Id. at 1977 (emphasis added). The Court was dismissive of the idea that Nishimura Ekiu had “based its decision on the Suspension Clause without even mentioning that provision.” 116 Id. at 1978 (emphasis added). It reasoned that some contemporaneous cases mentioned the Suspension Clause expressly, and the failure to do so in immigration cases meant that the associated habeas process was not constitutionally required. 117 See id. at 1979. The inferences the Thuraissigiam majority drew from this particular authority were suspect. None of the cases citing the Suspension Clause suggest that it was the constitutional source of the Privilege, so there is no way to draw any negative inference from the absence of such language in the noncitizen-admission cases. Specifically, in seeking to establish a negative inference from cases that did discuss the Suspension Clause, Thuraissigiam recited two episodes in which Congress expressly authorized suspension—not cases about the scope of constitutionally guaranteed Privilege during periods of nonsuspension. See id. (discussing episodes of suspension in the Philippines and Hawaii). Thuraissigiam also cited In re Yamashita, 327 U.S. 1 (1946), where a Japanese Imperial Army general, accused of permitting his troops to commit atrocities, was tried before an American military tribunal. Yamashita does not trace the Privilege specifically to the Suspension Clause, merely observing that Congress could not restrict habeas process to review the lawfulness of the military commission “unless there was suspension of the writ.” Id. at 9. Finally, Thuraissigiam observed that, in Ex parte Endo, 323 U.S. 283 (1944), “the Court invoked the Suspension Clause in holding that the Executive lacked authority to intern a Japanese-American citizen.” Thuraissigiam, 140 S. Ct. at 1979. The word “invoked” is doing an unfortunate amount of work in that sentence; while Endo mentioned the Suspension Clause, it did not use it to strike down the detention at issue or to interpret the pertinent statute. Endo mentioned the Suspension Clause only in passing; it made clear that the Clause was not the basis for the decision, and the passing reference did not suggest that the Clause was the source of the Privilege. See Endo, 323 U.S. at 299–300.

What the Thuraissigiam majority was really doing was taking the absence of decisional language citing the Suspension Clause as the originating source of the Privilege and interpreting that absence to mean that the Privilege did not in fact cover the custody at issue (the Court, however, was not especially candid about this interpretive choice). The failure to cite the Suspension Clause, Thuraissigiam ultimately reasoned, meant that the finality-era practice did not actually reflect a constitutional guarantee of habeas process, but instead the existence of a habeas statute. 118 The Court ultimately determined that the habeas review taking place during the finality era was pursuant to the habeas statute, albeit without discussing whether the statute effectuated a guarantee in the Constitution. See Thuraissigiam, 140 S. Ct. at 1980–81.

*    *    *

In sum, Thuraissigiam captures at least two major consequences of the origination dilemma. First, seizing on Boumediene’s references to a rule against “unconstitutional suspensions,” the Supreme Court declared the existential question—whether the Constitution’s habeas guarantee is more than an antisuspension rule—unsettled. Second, it used the uncertainty around origination to shrink the scope of whatever constitutional habeas guarantee there is, reading the Privilege out of cases that failed to identify the Suspension Clause as the originating text.

IV. Cleaning Things Up

Given the praise lavished on the habeas Privilege, the collective failure to coherently articulate an origination theory is puzzling. Part IV specifies the steps that the Court ought to take in order to clarify the scope of constitutional habeas protection. In so doing, and in the interest of fairness to readers, I do my best to disentangle the normative arguments for more doctrinal clarity from the normative arguments about what doctrinal choices I think the Supreme Court should make.

A. Existential and Second-Order Questions

This section’s objective is not to argue in favor or against any particular origination theory, but to urge decisional clarity that promotes rule of law virtues such as institutional certainty and doctrinal stability. 119 See Lon L. Fuller, The Morality of Law 33–94 (1964) (outlining the routes of legal failure and success in creating law); Rolf E. Sartorius, Individual Conduct and Social Norms 163–79 (1975) (questioning whether there are domains of human action and interaction that are not suited to governance of general rules). In order to lend some determinacy to the entire doctrinal enterprise, the Supreme Court ought to coherently resolve the whether and the how of origination. And the academy ought to help by formulating coherent origination accounts.

First, the existential question: whether the Constitution’s habeas guarantee is something more than an antisuspension rule. The Supreme Court has a history of avoiding the existential question by burying references in footnotes, engaging in creative statutory interpretation that moots the issue, or by assuming the answer for the sake of argument only. 120 See, e.g., Thuraissigiam, 140 S. Ct. at 1969 n.12 (“We do not revisit that question.”); Immigr. & Naturaliz. Serv. v. St. Cyr, 533 U.S. 289, 305 (2001) (“The necessity of resolving such a serious and difficult constitutional issue—and the desirability of avoiding that necessity—simply reinforce the reasons for requiring a clear and unambiguous statement of congressional intent.”). Decisions stretching back over a century have assumed the existence of a more robust constitutional guarantee, but most of those opinions do not qualify as formal holdings because the constitutional assumption was reflected in a statutory privilege. 121 See, e.g., St. Cyr, 533 U.S. at 314 (interpreting AEDPA and IIRIRA to preserve statutory habeas jurisdiction to review a deportation order); Heikkila v. Barber, 345 U.S. 229, 233–36 (1953) (explaining that immigration and habeas statutes had been construed so as to preserve habeas review over legal questions); Nishimura Ekiu v. United States, 142 U.S. 651, 660 (1892) (holding that Congress had permissibly stripped habeas jurisdiction only over questions of fact pertinent to an order excluding noncitizen entry).

Recall that Boumediene, a case that was a formal constitutional holding, used awkward phrasing indicating that the MCA was an “unconstitutional suspension.” 122 Boumediene v. Bush, 553 U.S. 723, 792 (2008). Specifically, Justice Scalia would have held that there was no constitutional problem because “there has thus been no suspension of the writ, and no basis exists for judicial intervention beyond what the [DTA] allows.” Id. at 827 (Scalia, J., dissenting). If the inquiry is indeed over upon the determination that the Act does not permit review and the restriction does not amount to a suspension, then it seems to exclude a separate determination as to whether the privilege guarantees the process in the absence of suspension. Four Boumediene Justices signed a dissent that an adventurous interpreter could read as endorsing  the  proposition  that  the  Constitution  only  contained  an  antisuspension rule 123 See id. at 827. —a position that Justices Scalia and Thomas took in other cases. 124 See Thuraissigiam, 140 S. Ct. at 1983–88 (Thomas, J., concurring). Seven Justices in Thuraissigiam joined opinions declaring the issue unresolved. 125 See supra notes 92–97 and accompanying text. Government institutions can discern neither the boundaries of lawful custody nor the requirements of Article III jurisdiction until the Court resolves it.

The answer to the existential question dictates which of two frameworks controls constitutional inquiry. The frameworks involve very different doctrinal questions, and they differ substantially in the degree of constitutional protection they afford. Start with the framework that follows from a determination that the constitutional guarantee includes the Privilege, and not just an antisuspension rule. That framework involves more robust protection of Article III access, and it would produce debate about the scope of constitutionally guaranteed habeas process without reference to whether a particular restriction qualifies as a suspension. Cases would pose tough questions about whether a habeas restriction actually impinges upon some constitutionally protected core—and whether that core has remained frozen since 1789—but there would be a constitutional core nonetheless. 126 The Supreme Court has routinely hedged on this particular issue, deciding cases in ways that allow it to avoid the question whether the scope of the Privilege was fixed in 1789 or has in some way expanded. See, e.g., Thuraissigiam, 140 S. Ct. at 1969 (explaining that it considered only the status of the Privilege in 1789 because Thuraissigiam had made only originalist arguments); St. Cyr, 533 U.S. at 300–01 (reasoning that the Clause protected, “at the absolute minimum,” the Privilege as it existed at the Founding); Felker v. Turpin, 518 U.S. 651, 663–64 (1996) (“But we assume, for purposes of decision here, that the Suspension Clause of the Constitution refers to the writ as it exists today, rather than as it existed in 1789.”).

Now consider the other framework, which would follow from a determination that the constitutional guarantee amounts to nothing more than an antisuspension rule. The scope of the Privilege would be irrelevant because the Constitution would not guarantee it. A habeas restriction would be unconstitutional only if it qualified as a suspension, and only if the suspension criteria—a threat to public safety created by rebellion or invasion—were unsatisfied. 127 See U.S. Const. art. I, § 9, cl. 2. To make matters even more complicated, the orthodox understanding is that satisfaction of the suspension conditions is nonjusticiable, so the Court would need to draw lines indicating which suspensions are suitable for Article III review and which are not. But see Tyler, Suspension as Emergency Power, supra note 27, at 687–93 (detailing a break with the conventional view that suspension is nonjusticiable). At least in the ways relevant to most modes of constitutional interpretation, a “suspension” describes legislative output that is far narrower than a “restriction.” 128 See supra notes 20–27 and accompanying text. For that reason, the scope of constitutional protection in such a scenario is thin. 129 The protection in the other scenario could be thin, too, if the core of a constitutionally guaranteed privilege is small. Of course, in the other scenario, the quantum of judicial review could not be taken to zero.

An affirmative answer to the whether of origination should also generate judicial activity necessary to bring clarity to the how 130 There may be something of a chicken-and-egg problem here. Deficits in the accounts of how have surely slowed the determination as to whether. —which constitutional provisions, combination thereof, or structural principles give rise to the Privilege. Different origination accounts will entail constitutional protections of different shapes and sizes. Those features will depend, in turn, on the interests that the originating constitutional phenomena express—a Privilege anchored to a theory of sovereign power would guarantee certain things, a Privilege anchored to a theory of individual rights would guarantee others, and so forth.

The theoretical possibility of variation notwithstanding, virtually everyone (academics included) assumes that, if the Constitution originates a habeas guarantee that goes beyond an antisuspension rule, then the Suspension Clause does the origination. 131 See, e.g., supra note 18. But that account is not the only option. For example, I have made an article-length case that the Privilege guarantee is a byproduct of Article III judicial power, springing to life after Congress ordains and establishes Article III courts. 132 Lee Kovarsky, A Constitutional Theory of Habeas Power, 99 Va. L. Rev. 753 (2013) [hereinafter Kovarsky, A Constitutional Theory]. On another account, the Clause recognizes a constitutional phenomenon that no narrow string of text  “creates,”  on  the  theory  that  the  Constitution  carried  forth  a  preexisting Privilege. 133 See supra notes 29–33 and accompanying text. The Court has constitutionalized state sovereign immunity on the theory that it was a constitutional background principle, 134 In refusing congressional abrogation of state sovereign immunity in federal court, Seminole Tribe of Florida v. Florida referred to “the background principle of state sovereign immunity embodied in the Eleventh Amendment.” 517 U.S. 44, 73 (1996). That language left it possible to infer that the Eleventh Amendment originated the sovereign immunity, rather than the “background principle” that it embodied. In Alden v. Maine, however, which involved a congressional abrogation of state sovereign immunity in state court, the Supreme Court made clear that sovereign immunity preexisted and was atextually incorporated into the Constitution: “[T]he sovereign immunity of the States neither derives from, nor is limited by, the terms of the Eleventh Amendment . . . . [T]he States’ immunity from suit is [instead] a fundamental aspect of the sovereignty which the States enjoyed before the ratification of the Constitution, and which they retain today . . . .” 527 U.S. 706, 713 (1999). See generally Ernest A. Young, Alden v. Maine and the Jurisprudence of Structure, 41 Wm. & Mary L. Rev. 1601 (2000) (discussing the structural jurisprudence of Alden, and observing that it “drops the textual fig leaf entirely, acknowledging that any principle of immunity applicable in state court can have no basis in the Eleventh Amendment” (emphasis omitted)). and, unlike the Privilege, the Constitution contains no textual reference to sovereign immunity.

The important point is that the academy and the bench contain complementary voids, with neither community producing an origination account persuasive enough to force a decisional commitment. Considerable doctrinal uncertainty flourishes in the absence of such commitment, leaving presidential administrations guessing about the scope of their detention powers, Congress guessing about the bounds of permissible legislative restrictions on Article III review, and lower courts guessing about what the rules of constitutional adjudication are supposed to be. All three branches of federal government get yanked around whenever the composition of the Supreme Court changes, because preexisting precedent sets no determinate guardrails on what habeas practices are constitutionally protected. In short, the failure to articulate a theory of origination dissipates many rule of law virtues that we should expect in the context of government detention. 135 Cf., e.g., Joseph Raz, The Rule of Law and Its Virtue, in The Authority of Law 210, 213 (1979) (describing as a basic rule of law principle “that the law should be such that people will be able to be guided by it”). See generally Richard H. Fallon, Jr., “The Rule of Law” as a Concept in Constitutional Discourse, 97 Colum. L. Rev. 1, 15–16 (1997) (discussing robustness of rule of law virtues).

B. The Best Path for Protection

Up to this point, I have encouraged doctrinal clarity, largely without reference to the merits of particular origination accounts—at least beyond what was necessary to explain the underlying instability. As Part I explains when setting forth the problem, however, the originalist and structural cases for recognizing an affirmative habeas guarantee are overwhelming. 136 See supra notes 20–32 and accompanying text. Those indicia of constitutional meaning are at odds with literalist arguments that the Constitution contains only an antisuspension rule because the Suspension Clause lacks express text of Privilege origination. That tension is the problem in a nutshell.

The dissonance notwithstanding, I do not consider the existential question particularly close. The Privilege was too important to the Founders and the generation to which they belonged, 137 See supra notes 30–33 and accompanying text. the Suspension Clause text too plainly reads as an assumption that the Constitution guarantees habeas process, and the Privilege plays too extraordinary an American role in enforcing liberty and separating powers. 138 See supra note 76. Those indicia of constitutional meaning should be enough to overcome the literalist inferences behind the theory that the Constitution originates only an antisuspension rule.

Nor does an antisuspension rule sufficiently protect the Privilege. The basic structure of the most interpretively significant suspension statutes discloses that these statutes were narrow—setting aside the Privilege only for a time and including language authorizing detention that would otherwise be unlawful. In (only) this respect, I agree with Justices Scalia and Thomas. 139 Recall that this is roughly the position that Justice Scalia took in St. Cyr and that Justice Thomas took in Thuraissigiam. See supra notes 52–54 and accompanying text (Justice Scalia’s position in St. Cyr); supra notes 99–102 (Justice Thomas’s position in Thuraissigiam). Attempting to reproduce the protection of a constitutional  Privilege  by  expanding  the  definition  of  suspension  is  a  fool’s errand. 140 One could conceivably use an expanded definition of suspension to achieve this result, if one were engaged in a textualist enterprise that was unconstrained by original meaning or precedent.

The restrictions at issue in all of the modern cases were nothing like suspension statutes. At issue in Felker were permanent limits on successive petitions filed by people convicted of crimes in state courts (not temporary suspensions); 141 See Felker v. Turpin, 518 U.S. 651, 656–57 (1996). in St. Cyr it was a provision permanently restricting judicial review of deportation orders; 142 See Immigr. & Naturaliz. Serv. v. St. Cyr, 533 U.S. 289, 297 (2001). in Boumediene it was a provision permanently barring otherwise-available judicial review for GTMO detainees classified as enemy combatants; 143 See Boumediene v. Bush, 553 U.S. 723, 733 (2008). and in Thuraissigiam it was a permanent restriction on habeas process for asylum claimants subject to expedited removal. 144 See Dep’t of Homeland Sec. v. Thuraissigiam, 140 S. Ct. 1959, 1966 (2020). If the sum total of constitutional habeas law is an antisuspension rule, then the Supreme Court will find itself between the horns of a dilemma. It will either have to sever the contextualized definition of suspension, or it will have to accept a thin habeas protection that is utterly inconsistent with the liberty-enforcing and power-separating roles that the Privilege performs.

If there is to be a constitutionally guaranteed Privilege, then the question of how it originates remains. Different origination accounts will dictate different Privilege parameters. As mentioned, I have elsewhere argued that the inviolability of the Privilege is an auxiliary feature of Article III judicial power and inheres in the very definition of a court. 145 See Kovarsky, A Constitutional Theory, supra note 136, at 773–78. Professor Eric Freedman has also made a robust argument along these lines. See Eric Freedman, Making Habeas Work: A Legal History 91 & 175 n.23 (2018). I will not rehash that position here, except to say that it moves the origination work outside of the Suspension Clause—thereby providing an account much more consistent with constitutional text and structure, as well as with the (rather clearly supported) original understanding of how the Privilege was to mediate institutional power. Assigning an origination function to the Clause is probably the worst interpretive approach—given its fraught relationship to text, history, and structure—but that approach remains superior to a framework that denies a constitutional guarantee altogether. 146 Given space limitations, I do not address the theory that the Suspension Clause bars federal suspension of a state privilege. This theory, set forth most prominently by William Duker, is based on the implausible notion that the Suspension Clause was meant to preclude Congress from suspending state judicial process that Congress would have never had authority to require. See Duker, supra note 28, at 126. Restrictions on suspension apply to the sovereign that guarantees the privilege. See Kovarsky, A Constitutional Theory, supra note 136, at 789–92 (rejecting Duker’s theory on grounds that it severs the necessary relationship between privilege and suspension, that the antisuspension rule would have appeared in a different part of the Constitution if it did what Duker suggested, and that Duker misreads both the Federalist Papers and other data in the historical record).


By some combination of accident and brokered indecision, the Supreme Court steadfastly refuses to resolve whether the Constitution’s habeas guarantee is more than an antisuspension rule. The scope of constitutional protection is therefore anyone’s guess—it is somehow both unpredictable and enormously sensitive to the composition of the Supreme Court. Such doctrinal uncertainty both reflects and produces a related theoretical deficit. Neither the Court nor the academy has produced a coherent, satisfactory account of whence the Privilege springs. These problems will not abate until the legal community produces a viable theory of Privilege origination and the judiciary embraces it.