Boumediene v. Bush
Annotate this Case
553 U.S. 723 (2008)
- Syllabus |
- Opinion (Anthony M. Kennedy) |
- Concurrence (David H. Souter) |
- Dissent (Antonin Scalia) |
- Dissent (John G. Roberts, Jr.)
OPINION OF THE COURT
BOUMEDIENE V. BUSH
553 U. S. ____ (2008)
SUPREME COURT OF THE UNITED STATES
NOS. 06-1195 AND 06-1196
LAKHDAR BOUMEDIENE, et al., PETITIONERS
GEORGE W. BUSH, PRESIDENT OF THE UNITED STATES, et al.
KHALED A. F. AL ODAH, next friend of FAWZI KHALID ABDULLAH FAHAD AL ODAH, et al., PETITIONERS
UNITED STATES et al.
on writs of certiorari to the united states court of appeals for the district of columbia circuit
[June 12, 2008]
Justice Kennedy delivered the opinion of the Court.
Petitioners are aliens designated as enemy combatants and detained at the United States Naval Station at Guantanamo Bay, Cuba. There are others detained there, also aliens, who are not parties to this suit.
Petitioners present a question not resolved by our earlier cases relating to the detention of aliens at Guantanamo: whether they have the constitutional privilege of habeas corpus, a privilege not to be withdrawn except in conformance with the Suspension Clause, Art. I, §9, cl. 2. We hold these petitioners do have the habeas corpus privilege. Congress has enacted a statute, the Detainee Treatment Act of 2005 (DTA), 119 Stat. 2739, that provides certain procedures for review of the detainees’ status. We hold that those procedures are not an adequate and effective substitute for habeas corpus. Therefore §7 of the Military Commissions Act of 2006 (MCA), 28 U. S. C. A. §2241(e) (Supp. 2007), operates as an unconstitutional suspension of the writ. We do not address whether the President has authority to detain these petitioners nor do we hold that the writ must issue. These and other questions regarding the legality of the detention are to be resolved in the first instance by the District Court.
Under the Authorization for Use of Military Force (AUMF), §2(a), 115 Stat. 224, note following 50 U. S. C. §1541 (2000 ed., Supp. V), the President is authorized “to use all necessary and appropriate force against those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons, in order to prevent any future acts of international terrorism against the United States by such nations, organizations or persons.”
In Hamdi v. Rumsfeld, 542 U. S. 507 (2004), five Members of the Court recognized that detention of individuals who fought against the United States in Afghanistan “for the duration of the particular conflict in which they were captured, is so fundamental and accepted an incident to war as to be an exercise of the ‘necessary and appropriate force’ Congress has authorized the President to use.” Id., at 518 (plurality opinion of O’Connor, J.), id., at 588–589 (Thomas, J., dissenting). After Hamdi, the Deputy Secretary of Defense established Combatant Status Review Tribunals (CSRTs) to determine whether individuals detained at Guantanamo were “enemy combatants,” as the Department defines that term. See App. to Pet. for Cert. in No. 06–1195, p. 81a. A later memorandum established procedures to implement the CSRTs. See App. to Pet. for Cert. in No. 06–1196, p. 147. The Government maintains these procedures were designed to comply with the due process requirements identified by the plurality in Hamdi. See Brief for Respondents 10.
Interpreting the AUMF, the Department of Defense ordered the detention of these petitioners, and they were transferred to Guantanamo. Some of these individuals were apprehended on the battlefield in Afghanistan, others in places as far away from there as Bosnia and Gambia. All are foreign nationals, but none is a citizen of a nation now at war with the United States. Each denies he is a member of the al Qaeda terrorist network that carried out the September 11 attacks or of the Taliban regime that provided sanctuary for al Qaeda. Each petitioner appeared before a separate CSRT; was determined to be an enemy combatant; and has sought a writ of habeas corpus in the United States District Court for the District of Columbia.
The first actions commenced in February 2002. The District Court ordered the cases dismissed for lack of jurisdiction because the naval station is outside the sovereign territory of the United States. See Rasul v. Bush, 215 F. Supp. 2d 55 (2002). The Court of Appeals for the District of Columbia Circuit affirmed. See Al Odah v. United States, 321 F. 3d 1134, 1145 (2003). We granted certiorari and reversed, holding that 28 U. S. C. §2241 extended statutory habeas corpus jurisdiction to Guantanamo. See Rasul v. Bush, 542 U. S. 466, 473 (2004). The constitutional issue presented in the instant cases was not reached in Rasul. Id., at 476.
After Rasul, petitioners’ cases were consolidated and entertained in two separate proceedings. In the first set of cases, Judge Richard J. Leon granted the Government’s motion to dismiss, holding that the detainees had no rights that could be vindicated in a habeas corpus action. In the second set of cases Judge Joyce Hens Green reached the opposite conclusion, holding the detainees had rights under the Due Process Clause of the Fifth Amendment. See Khalid v. Bush, 355 F. Supp. 2d 311, 314 (DC 2005); In re Guantanamo Detainee Cases, 355 F. Supp. 2d 443, 464 (DC 2005).
While appeals were pending from the District Court decisions, Congress passed the DTA. Subsection (e) of §1005 of the DTA amended 28 U. S. C. §2241 to provide that “no court, justice, or judge shall have jurisdiction to hear or consider … an application for a writ of habeas corpus filed by or on behalf of an alien detained by the Department of Defense at Guantanamo Bay, Cuba.” 119 Stat. 2742. Section 1005 further provides that the Court of Appeals for the District of Columbia Circuit shall have “exclusive” jurisdiction to review decisions of the CSRTs. Ibid.
In Hamdan v. Rumsfeld, 548 U. S. 557, 576–577 (2006), the Court held this provision did not apply to cases (like petitioners’) pending when the DTA was enacted. Congress responded by passing the MCA, 10 U. S. C. A. §948a et seq. (Supp. 2007), which again amended §2241. The text of the statutory amendment is discussed below. See Part II, infra. (Four Members of the Hamdan majority noted that “[n]othing prevent[ed] the President from returning to Congress to seek the authority he believes necessary.” 548 U. S., at 636 (Breyer, J., concurring). The authority to which the concurring opinion referred was the authority to “create military commissions of the kind at issue” in the case. Ibid. Nothing in that opinion can be construed as an invitation for Congress to suspend the writ.)
Petitioners’ cases were consolidated on appeal, and the parties filed supplemental briefs in light of our decision in Hamdan. The Court of Appeals’ ruling, 476 F. 3d 981 (CADC 2007), is the subject of our present review and today’s decision.
The Court of Appeals concluded that MCA §7 must be read to strip from it, and all federal courts, jurisdiction to consider petitioners’ habeas corpus applications, id., at 987; that petitioners are not entitled to the privilege of the writ or the protections of the Suspension Clause, id., at 990–991; and, as a result, that it was unnecessary to consider whether Congress provided an adequate and effective substitute for habeas corpus in the DTA.
We granted certiorari. 551 U. S. ___ (2007).
As a threshold matter, we must decide whether MCA §7 denies the federal courts jurisdiction to hear habeas corpus actions pending at the time of its enactment. We hold the statute does deny that jurisdiction, so that, if the statute is valid, petitioners’ cases must be dismissed.
As amended by the terms of the MCA, 28 U. S. C. A. §2241(e) (Supp. 2007) now provides:
“(1) No court, justice, or judge shall have jurisdiction to hear or consider an application for a writ of habeas corpus filed by or on behalf of an alien detained by the United States who has been determined by the United States to have been properly detained as an enemy combatant or is awaiting such determination.
“(2) Except as provided in [§§1005(e)(2) and (e)(3) of the DTA] no court, justice, or judge shall have jurisdiction to hear or consider any other action against the United States or its agents relating to any aspect of the detention, transfer, treatment, trial, or conditions of confinement of an alien who is or was detained by the United States and has been determined by the United States to have been properly detained as an enemy combatant or is awaiting such determination.”
Section 7(b) of the MCA provides the effective date for the amendment of §2241(e). It states:
“The amendment made by [MCA §7(a)] shall take effect on the date of the enactment of this Act, and shall apply to all cases, without exception, pending on or after the date of the enactment of this Act which relate to any aspect of the detention, transfer, treatment, trial, or conditions of detention of an alien detained by the United States since September 11, 2001.” 120 Stat. 2636.
There is little doubt that the effective date provision applies to habeas corpus actions. Those actions, by definition, are cases “which relate to … detention.” See Black’s Law Dictionary 728 (8th ed. 2004) (defining habeas corpus as “[a] writ employed to bring a person before a court, most frequently to ensure that the party’s imprisonment or detention is not illegal”). Petitioners argue, nevertheless, that MCA §7(b) is not a sufficiently clear statement of congressional intent to strip the federal courts of jurisdiction in pending cases. See Ex parte Yerger, 8 Wall. 85, 102–103 (1869). We disagree.
Their argument is as follows: Section 2241(e)(1) refers to “a writ of habeas corpus.” The next paragraph, §2241(e)(2), refers to “any other action … relating to any aspect of the detention, transfer, treatment, trial, or conditions of confinement of an alien who … [has] been properly detained as an enemy combatant or is awaiting such determination.” There are two separate paragraphs, the argument continues, so there must be two distinct classes of cases. And the effective date subsection, MCA §7(b), it is said, refers only to the second class of cases, for it largely repeats the language of §2241(e)(2) by referring to “cases … which relate to any aspect of the detention, transfer, treatment, trial, or conditions of detention of an alien detained by the United States.”
Petitioners’ textual argument would have more force were it not for the phrase “other action” in §2241(e)(2). The phrase cannot be understood without referring back to the paragraph that precedes it, §2241(e)(1), which explicitly mentions the term “writ of habeas corpus.” The structure of the two paragraphs implies that habeas actions are a type of action “relating to any aspect of the detention, transfer, treatment, trial, or conditions of confinement of an alien who is or was detained … as an enemy combatant.” Pending habeas actions, then, are in the category of cases subject to the statute’s jurisdictional bar.
We acknowledge, moreover, the litigation history that prompted Congress to enact the MCA. In Hamdan the Court found it unnecessary to address the petitioner’s Suspension Clause arguments but noted the relevance of the clear statement rule in deciding whether Congress intended to reach pending habeas corpus cases. See 548 U. S., at 575 (Congress should “not be presumed to have effected such denial [of habeas relief] absent an unmistakably clear statement to the contrary”). This interpretive rule facilitates a dialogue between Congress and the Court. Cf. Hilton v. South Carolina Public Railways Comm’n, 502 U. S. 197, 206 (1991); H. Hart & A. Sacks, The Legal Process: Basic Problems in the Making and Application of Law 1209–1210 (W. Eskridge & P. Frickey eds. 1994). If the Court invokes a clear statement rule to advise that certain statutory interpretations are favored in order to avoid constitutional difficulties, Congress can make an informed legislative choice either to amend the statute or to retain its existing text. If Congress amends, its intent must be respected even if a difficult constitutional question is presented. The usual presumption is that Members of Congress, in accord with their oath of office, considered the constitutional issue and determined the amended statute to be a lawful one; and the Judiciary, in light of that determination, proceeds to its own independent judgment on the constitutional question when required to do so in a proper case.
If this ongoing dialogue between and among the branches of Government is to be respected, we cannot ignore that the MCA was a direct response to Hamdan’s holding that the DTA’s jurisdiction-stripping provision had no application to pending cases. The Court of Appeals was correct to take note of the legislative history when construing the statute, see 476 F. 3d, at 986, n. 2 (citing relevant floor statements); and we agree with its conclusion that the MCA deprives the federal courts of jurisdiction to entertain the habeas corpus actions now before us.
In deciding the constitutional questions now presented we must determine whether petitioners are barred from seeking the writ or invoking the protections of the Suspension Clause either because of their status, i.e., petitioners’ designation by the Executive Branch as enemy combatants, or their physical location, i.e., their presence at Guantanamo Bay. The Government contends that noncitizens designated as enemy combatants and detained in territory located outside our Nation’s borders have no constitutional rights and no privilege of habeas corpus. Petitioners contend they do have cognizable constitutional rights and that Congress, in seeking to eliminate recourse to habeas corpus as a means to assert those rights, acted in violation of the Suspension Clause.
We begin with a brief account of the history and origins of the writ. Our account proceeds from two propositions. First, protection for the privilege of habeas corpus was one of the few safeguards of liberty specified in a Constitution that, at the outset, had no Bill of Rights. In the system conceived by the Framers the writ had a centrality that must inform proper interpretation of the Suspension Clause. Second, to the extent there were settled precedents or legal commentaries in 1789 regarding the extraterritorial scope of the writ or its application to enemy aliens, those authorities can be instructive for the present cases.
The Framers viewed freedom from unlawful restraint as a fundamental precept of liberty, and they understood the writ of habeas corpus as a vital instrument to secure that freedom. Experience taught, however, that the common-law writ all too often had been insufficient to guard against the abuse of monarchial power. That history counseled the necessity for specific language in the Constitution to secure the writ and ensure its place in our legal system.
Magna Carta decreed that no man would be imprisoned contrary to the law of the land. Art. 39, in Sources of Our Liberties 17 (R. Perry & J. Cooper eds. 1959) (“No free man shall be taken or imprisoned or dispossessed, or outlawed, or banished, or in any way destroyed, nor will we go upon him, nor send upon him, except by the legal judgment of his peers or by the law of the land”). Important as the principle was, the Barons at Runnymede prescribed no specific legal process to enforce it. Holdsworth tells us, however, that gradually the writ of habeas corpus became the means by which the promise of Magna Carta was fulfilled. 9 W. Holdsworth, A History of English Law 112 (1926) (hereinafter Holdsworth).
The development was painstaking, even by the centuries-long measures of English constitutional history. The writ was known and used in some form at least as early as the reign of Edward I. Id., at 108–125. Yet at the outset it was used to protect not the rights of citizens but those of the King and his courts. The early courts were considered agents of the Crown, designed to assist the King in the exercise of his power. See J. Baker, An Introduction to English Legal History 38–39 (4th ed. 2002). Thus the writ, while it would become part of the foundation of liberty for the King’s subjects, was in its earliest use a mechanism for securing compliance with the King’s laws. See Halliday & White, The Suspension Clause: English Text, Imperial Contexts, and American Implications, 94 Va. L. Rev. (forthcoming 2008) (hereinafter Halliday & White) (manuscript, at 11, online at http://papers.ssrn.com/sol3 /papers.cfm?abstract_id=1008252 (all Internet materials as visited June 9, 2008, and available in Clerk of Court’s case file) (noting that “conceptually the writ arose from a theory of power rather than a theory of liberty”)). Over time it became clear that by issuing the writ of habeas corpus common-law courts sought to enforce the King’s prerogative to inquire into the authority of a jailer to hold a prisoner. See M. Hale, Prerogatives of the King 229 (D. Yale ed. 1976); 2 J. Story, Commentaries on the Constitution of the United States §1341, p. 237 (3d ed. 1858) (noting that the writ ran “into all parts of the king’s dominions; for it is said, that the king is entitled, at all times, to have an account, why the liberty of any of his subjects is restrained”).
Even so, from an early date it was understood that the King, too, was subject to the law. As the writers said of Magna Carta, “it means this, that the king is and shall be below the law.” 1 F. Pollock & F. Maitland, History of English Law 173 (2d ed. 1909); see also 2 Bracton On the Laws and Customs of England 33 (S. Thorne transl. 1968) (“The king must not be under man but under God and under the law, because law makes the king”). And, by the 1600’s, the writ was deemed less an instrument of the King’s power and more a restraint upon it. See Collings, Habeas Corpus for Convicts—Constitutional Right or Legislative Grace, 40 Calif. L. Rev. 335, 336 (1952) (noting that by this point the writ was “the appropriate process for checking illegal imprisonment by public officials”).
Still, the writ proved to be an imperfect check. Even when the importance of the writ was well understood in England, habeas relief often was denied by the courts or suspended by Parliament. Denial or suspension occurred in times of political unrest, to the anguish of the imprisoned and the outrage of those in sympathy with them.
A notable example from this period was Darnel’s Case, 3 How. St. Tr. 1 (K. B. 1627). The events giving rise to the case began when, in a display of the Stuart penchant for authoritarian excess, Charles I demanded that Darnel and at least four others lend him money. Upon their refusal, they were imprisoned. The prisoners sought a writ of habeas corpus; and the King filed a return in the form of a warrant signed by the Attorney General. Ibid. The court held this was a sufficient answer and justified the subjects’ continued imprisonment. Id., at 59.
There was an immediate outcry of protest. The House of Commons promptly passed the Petition of Right, 3 Car. 1, ch. 1 (1627), 5 Statutes of the Realm 23, 24 (reprint 1963), which condemned executive “imprison[ment] without any cause” shown, and declared that “no freeman in any such manner as is before mencioned [shall] be imprisoned or deteined.” Yet a full legislative response was long delayed. The King soon began to abuse his authority again, and Parliament was dissolved. See W. Hall & R. Albion, A History of England and the British Empire 328 (3d ed. 1953) (hereinafter Hall & Albion). When Parliament reconvened in 1640, it sought to secure access to the writ by statute. The Act of 1640, 16 Car. 1, ch. 10, 5 Statutes of the Realm, at 110, expressly authorized use of the writ to test the legality of commitment by command or warrant of the King or the Privy Council. Civil strife and the Interregnum soon followed, and not until 1679 did Parliament try once more to secure the writ, this time through the Habeas Corpus Act of 1679, 31 Car. 2, ch. 2, id., at 935. The Act, which later would be described by Blackstone as the “stable bulwark of our liberties,” 1 W. Blackstone, Commentaries *137 (hereinafter Blackstone), established procedures for issuing the writ; and it was the model upon which the habeas statutes of the 13 American Colonies were based, see Collings, supra, at 338–339.
This history was known to the Framers. It no doubt confirmed their view that pendular swings to and away from individual liberty were endemic to undivided, uncontrolled power. The Framers’ inherent distrust of governmental power was the driving force behind the constitutional plan that allocated powers among three independent branches. This design serves not only to make Government accountable but also to secure individual liberty. See Loving v. United States, 517 U. S. 748, 756 (1996) (noting that “[e]ven before the birth of this country, separation of powers was known to be a defense against tyranny”); cf. Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579, 635 (1952) (Jackson, J., concurring) (“[T]he Constitution diffuses power the better to secure liberty”); Clinton v. City of New York, 524 U. S. 417, 450 (1998) (Kennedy, J., concurring) (“Liberty is always at stake when one or more of the branches seek to transgress the separation of powers”). Because the Constitution’s separation-of-powers structure, like the substantive guarantees of the Fifth and Fourteenth Amendments, see Yick Wo v. Hopkins, 118 U. S. 356, 374 (1886), protects persons as well as citizens, foreign nationals who have the privilege of litigating in our courts can seek to enforce separation-of-powers principles, see, e.g., INS v. Chadha, 462 U. S. 919, 958–959 (1983).
That the Framers considered the writ a vital instrument for the protection of individual liberty is evident from the care taken to specify the limited grounds for its suspension: “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.” Art. I, §9, cl. 2; see Amar, Of Sovereignty and Federalism, 96 Yale L. J. 1425, 1509, n. 329 (1987) (“[T]he non-suspension clause is the original Constitution’s most explicit reference to remedies”). The word “privilege” was used, perhaps, to avoid mentioning some rights to the exclusion of others. (Indeed, the only mention of the term “right” in the Constitution, as ratified, is in its clause giving Congress the power to protect the rights of authors and inventors. See Art. I, §8, cl. 8.)
Surviving accounts of the ratification debates provide additional evidence that the Framers deemed the writ to be an essential mechanism in the separation-of-powers scheme. In a critical exchange with Patrick Henry at the Virginia ratifying convention Edmund Randolph referred to the Suspension Clause as an “exception” to the “power given to Congress to regulate courts.” See 3 Debates in the Several State Conventions on the Adoption of the Federal Constitution 460–464 (J. Elliot 2d ed. 1876) (hereinafter Elliot’s Debates). A resolution passed by the New York ratifying convention made clear its understanding that the Clause not only protects against arbitrary suspensions of the writ but also guarantees an affirmative right to judicial inquiry into the causes of detention. See Resolution of the New York Ratifying Convention (July 26, 1788), in 1 Elliot’s Debates 328 (noting the convention’s understanding “[t]hat every person restrained of his liberty is entitled to an inquiry into the lawfulness of such restraint, and to a removal thereof if unlawful; and that such inquiry or removal ought not to be denied or delayed, except when, on account of public danger, the Congress shall suspend the privilege of the writ of habeas corpus”). Alexander Hamilton likewise explained that by providing the detainee a judicial forum to challenge detention, the writ preserves limited government. As he explained in The Federalist No. 84:
“[T]he practice of arbitrary imprisonments, have been, in all ages, the favorite and most formidable instruments of tyranny. The observations of the judicious Blackstone … are well worthy of recital: ‘To bereave a man of life … or by violence to confiscate his estate, without accusation or trial, would be so gross and notorious an act of despotism as must at once convey the alarm of tyranny throughout the whole nation; but confinement of the person, by secretly hurrying him to jail, where his sufferings are unknown or forgotten, is a less public, a less striking, and therefore a more dangerous engine of arbitrary government.’ And as a remedy for this fatal evil he is everywhere peculiarly emphatical in his encomiums on the habeas corpus act, which in one place he calls ‘the bulwark of the British Constitution.’ ” C. Rossiter ed., p. 512 (1961) (quoting 1 Blackstone *136, 4 id., at *438).
Post-1789 habeas developments in England, though not bearing upon the Framers’ intent, do verify their foresight. Those later events would underscore the need for structural barriers against arbitrary suspensions of the writ. Just as the writ had been vulnerable to executive and parliamentary encroachment on both sides of the Atlantic before the American Revolution, despite the Habeas Corpus Act of 1679, the writ was suspended with frequency in England during times of political unrest after 1789. Parliament suspended the writ for much of the period from 1792 to 1801, resulting in rampant arbitrary imprisonment. See Hall & Albion 550. Even as late as World War I, at least one prominent English jurist complained that the Defence of the Realm Act, 1914, 4 & 5 Geo. 5, ch. 29(1)(a), effectively had suspended the privilege of habeas corpus for any person suspected of “communicating with the enemy.” See King v. Halliday,  A. C. 260, 299 (Lord Shaw, dissenting); see generally A. Simpson, In the Highest Degree Odious: Detention Without Trial in Wartime Britain 6–7, 24–25 (1992).
In our own system the Suspension Clause is designed to protect against these cyclical abuses. The Clause protects the rights of the detained by a means consistent with the essential design of the Constitution. It ensures that, except during periods of formal suspension, the Judiciary will have a time-tested device, the writ, to maintain the “delicate balance of governance” that is itself the surest safeguard of liberty. See Hamdi, 542 U. S., at 536 (plurality opinion). The Clause protects the rights of the detained by affirming the duty and authority of the Judiciary to call the jailer to account. See Preiser v. Rodriguez, 411 U. S. 475, 484 (1973) (“[T]he essence of habeas corpus is an attack by a person in custody upon the legality of that custody”); cf. In re Jackson, 15 Mich. 417, 439–440 (1867) (Cooley, J., concurring) (“The important fact to be observed in regard to the mode of procedure upon this [habeas] writ is, that it is directed to, and served upon, not the person confined, but his jailer”). The separation-of-powers doctrine, and the history that influenced its design, therefore must inform the reach and purpose of the Suspension Clause.
The broad historical narrative of the writ and its function is central to our analysis, but we seek guidance as well from founding-era authorities addressing the specific question before us: whether foreign nationals, apprehended and detained in distant countries during a time of serious threats to our Nation’s security, may assert the privilege of the writ and seek its protection. The Court has been careful not to foreclose the possibility that the protections of the Suspension Clause have expanded along with post-1789 developments that define the present scope of the writ. See INS v. St. Cyr, 533 U. S. 289, 300–301 (2001). But the analysis may begin with precedents as of 1789, for the Court has said that “at the absolute minimum” the Clause protects the writ as it existed when the Constitution was drafted and ratified. Id., at 301.
To support their arguments, the parties in these cases have examined historical sources to construct a view of the common-law writ as it existed in 1789—as have amici whose expertise in legal history the Court has relied upon in the past. See Brief for Legal Historians as Amici Curiae; see also St. Cyr, supra, at 302, n. 16. The Government argues the common-law writ ran only to those territories over which the Crown was sovereign. See Brief for Respondents 27. Petitioners argue that jurisdiction followed the King’s officers. See Brief for Petitioner Boumediene et al. 11. Diligent search by all parties reveals no certain conclusions. In none of the cases cited do we find that a common-law court would or would not have granted, or refused to hear for lack of jurisdiction, a petition for a writ of habeas corpus brought by a prisoner deemed an enemy combatant, under a standard like the one the Department of Defense has used in these cases, and when held in a territory, like Guantanamo, over which the Government has total military and civil control.
We know that at common law a petitioner’s status as an alien was not a categorical bar to habeas corpus relief. See, e.g., Sommersett’s Case, 20 How. St. Tr. 1, 80–82 (1772) (ordering an African slave freed upon finding the custodian’s return insufficient); see generally Khera v. Secretary of State for the Home Dept.,  A. C. 74, 111 (“Habeas corpus protection is often expressed as limited to ‘British subjects.’ Is it really limited to British nationals? Suffice it to say that the case law has given an emphatic ‘no’ to the question”). We know as well that common-law courts entertained habeas petitions brought by enemy aliens detained in England—“entertained” at least in the sense that the courts held hearings to determine the threshold question of entitlement to the writ. See Case of Three Spanish Sailors, 2 Black. W. 1324, 96 Eng. Rep. 775 (C. P. 1779); King v. Schiever, 2 Burr. 765, 97 Eng. Rep. 551 (K. B. 1759); Du Castro’s Case, Fort. 195, 92 Eng. Rep. 816 (K. B. 1697).
In Schiever and the Spanish Sailors’ case, the courts denied relief to the petitioners. Whether the holdings in these cases were jurisdictional or based upon the courts’ ruling that the petitioners were detained lawfully as prisoners of war is unclear. See Spanish Sailors, supra, at 1324, 96 Eng. Rep., at 776; Schiever, supra, at 766, 97 Eng. Rep., at 552. In Du Castro’s Case, the court granted relief, but that case is not analogous to petitioners’ because the prisoner there appears to have been detained in England. See Halliday & White 27, n. 72. To the extent these authorities suggest the common-law courts abstained altogether from matters involving prisoners of war, there was greater justification for doing so in the context of declared wars with other nation states. Judicial intervention might have complicated the military’s ability to negotiate exchange of prisoners with the enemy, a wartime practice well known to the Framers. See Resolution of Mar. 30, 1778, 10 Journals of the Continental Congress 1774–1789, p. 295 (W. Ford ed. 1908) (directing General Washington not to exchange prisoners with the British unless the enemy agreed to exempt citizens from capture).
We find the evidence as to the geographic scope of the writ at common law informative, but, again, not dispositive. Petitioners argue the site of their detention is analogous to two territories outside of England to which the writ did run: the so-called “exempt jurisdictions,” like the Channel Islands; and (in former times) India. There are critical differences between these places and Guantanamo, however.
As the Court noted in Rasul, 542 U. S., at 481–482, and nn. 11–12, common-law courts granted habeas corpus relief to prisoners detained in the exempt jurisdictions. But these areas, while not in theory part of the realm of England, were nonetheless under the Crown’s control. See 2 H. Hallam, Constitutional History of England: From the Accession of Henry VII to the Death of George II, pp. 232–233 (reprint 1989). And there is some indication that these jurisdictions were considered sovereign territory. King v. Cowle, 2 Burr. 834, 854, 855, 97 Eng. Rep. 587, 599 (K. B. 1759) (describing one of the exempt jurisdictions, Berwick-upon-Tweed, as under the “sovereign jurisdiction” and “subjection of the Crown of England”). Because the United States does not maintain formal sovereignty over Guantanamo Bay, see Part IV, infra, the naval station there and the exempt jurisdictions discussed in the English authorities are not similarly situated.
Petitioners and their amici further rely on cases in which British courts in India granted writs of habeas corpus to noncitizens detained in territory over which the Moghul Emperor retained formal sovereignty and control. See supra, at 12–13; Brief for Legal Historians as Amici Curiae 12–13. The analogy to the present cases breaks down, however, because of the geographic location of the courts in the Indian example. The Supreme Court of Judicature (the British Court) sat in Calcutta; but no federal court sits at Guantanamo. The Supreme Court of Judicature was, moreover, a special court set up by Parliament to monitor certain conduct during the British Raj. See Regulating Act of 1773, 13 Geo. 3, §§13–14. That it had the power to issue the writ in nonsovereign territory does not prove that common-law courts sitting in England had the same power. If petitioners were to have the better of the argument on this point, we would need some demonstration of a consistent practice of common-law courts sitting in England and entertaining petitions brought by alien prisoners detained abroad. We find little support for this conclusion.
The Government argues, in turn, that Guantanamo is more closely analogous to Scotland and Hanover, territories that were not part of England but nonetheless controlled by the English monarch (in his separate capacities as King of Scotland and Elector of Hanover). See Cowle, 2 Burr., at 856, 97 Eng. Rep., at 600. Lord Mansfield can be cited for the proposition that, at the time of the founding, English courts lacked the “power” to issue the writ to Scotland and Hanover, territories Lord Mansfield referred to as “foreign.” Ibid. But what matters for our purposes is why common-law courts lacked this power. Given the English Crown’s delicate and complicated relationships with Scotland and Hanover in the 1700’s, we cannot disregard the possibility that the common-law courts’ refusal to issue the writ to these places was motivated not by formal legal constructs but by what we would think of as prudential concerns. This appears to have been the case with regard to other British territories where the writ did not run. See 2 R. Chambers, A Course of Lectures on English Law 1767–1773, p. 8 (T. Curley ed. 1986) (quoting the view of Lord Mansfield in Cowle that “[n]otwithstanding the power which the judges have, yet where they cannot judge of the cause, or give relief upon it, they would not think proper to interpose; and therefore in the case of imprisonments in Guernsey, Jersey, Minorca, or the plantations, the most usual way is to complain to the king in Council” (internal quotation marks omitted)). And after the Act of Union in 1707, through which the kingdoms of England and Scotland were merged politically, Queen Anne and her successors, in their new capacity as sovereign of Great Britain, ruled the entire island as one kingdom. Accordingly, by the time Lord Mansfield penned his opinion in Cowle in 1759, Scotland was no longer a “foreign” country vis-À-vis England—at least not in the sense in which Cuba is a foreign country vis-À-vis the United States.
Scotland remained “foreign” in Lord Mansfield’s day in at least one important respect, however. Even after the Act of Union, Scotland (like Hanover) continued to maintain its own laws and court system. See 1 Blackstone *98, *109. Under these circumstances prudential considerations would have weighed heavily when courts sitting in England received habeas petitions from Scotland or the Electorate. Common-law decisions withholding the writ from prisoners detained in these places easily could be explained as efforts to avoid either or both of two embarrassments: conflict with the judgments of another court of competent jurisdiction; or the practical inability, by reason of distance, of the English courts to enforce their judgments outside their territorial jurisdiction. Cf. Munaf v. Geren, ante, at 15 (opinion of the Court) (recognizing that “ ‘prudential concerns’ … such as comity and the orderly administration of criminal justice” affect the appropriate exercise of habeas jurisdiction).
By the mid-19th century, British courts could issue the writ to Canada, notwithstanding the fact that Canadian courts also had the power to do so. See 9 Holdsworth 124 (citing Ex parte Anderson, 3 El. and El. 487 (1861)). This might be seen as evidence that the existence of a separate court system was no barrier to the running of the common-law writ. The Canada of the 1800’s, however, was in many respects more analogous to the exempt jurisdictions or to Ireland, where the writ ran, than to Scotland or Hanover in the 1700’s, where it did not. Unlike Scotland and Hanover, Canada followed English law. See B. Laskin, The British Tradition in Canadian Law 50–51 (1969).
In the end a categorical or formal conception of sovereignty does not provide a comprehensive or altogether satisfactory explanation for the general understanding that prevailed when Lord Mansfield considered issuance of the writ outside England. In 1759 the writ did not run to Scotland but did run to Ireland, even though, at that point, Scotland and England had merged under the rule of a single sovereign, whereas the Crowns of Great Britain and Ireland remained separate (at least in theory). See Cowle, supra, at 856–857, 97 Eng. Rep., 600; 1 Blackstone *100–101. But there was at least one major difference between Scotland’s and Ireland’s relationship with England during this period that might explain why the writ ran to Ireland but not to Scotland. English law did not generally apply in Scotland (even after the Act of Union) but it did apply in Ireland. Blackstone put it as follows: “[A]s Scotland and England are now one and the same kingdom, and yet differ in their municipal laws; so England and Ireland are, on the other hand, distinct kingdoms, and yet in general agree in their laws.” Id., at *100. This distinction, and not formal notions of sovereignty, may well explain why the writ did not run to Scotland (and Hanover) but would run to Ireland.
The prudential barriers that may have prevented the English courts from issuing the writ to Scotland and Hanover are not relevant here. We have no reason to believe an order from a federal court would be disobeyed at Guantanamo. No Cuban court has jurisdiction to hear these petitioners’ claims, and no law other than the laws of the United States applies at the naval station. The modern-day relations between the United States and Guantanamo thus differ in important respects from the 18th-century relations between England and the kingdoms of Scotland and Hanover. This is reason enough for us to discount the relevance of the Government’s analogy.
Each side in the present matter argues that the very lack of a precedent on point supports its position. The Government points out there is no evidence that a court sitting in England granted habeas relief to an enemy alien detained abroad; petitioners respond there is no evidence that a court refused to do so for lack of jurisdiction.
Both arguments are premised, however, upon the assumption that the historical record is complete and that the common law, if properly understood, yields a definite answer to the questions before us. There are reasons to doubt both assumptions. Recent scholarship points to the inherent shortcomings in the historical record. See Halliday & White 14–15 (noting that most reports of 18th-century habeas proceedings were not printed). And given the unique status of Guantanamo Bay and the particular dangers of terrorism in the modern age, the common-law courts simply may not have confronted cases with close parallels to this one. We decline, therefore, to infer too much, one way or the other, from the lack of historical evidence on point. Cf. Brown v. Board of Education, 347 U. S. 483, 489 (1954) (noting evidence concerning the circumstances surrounding the adoption of the Fourteenth Amendment, discussed in the parties’ briefs and uncovered through the Court’s own investigation, “convince us that, although these sources cast some light, it is not enough to resolve the problem with which we are faced. At best, they are inconclusive”); Reid v. Covert, 354 U. S. 1, 64 (1957) (Frankfurter, J., concurring in result) (arguing constitutional adjudication should not be based upon evidence that is “too episodic, too meager, to form a solid basis in history, preceding and contemporaneous with the framing of the Constitution”).
Drawing from its position that at common law the writ ran only to territories over which the Crown was sovereign, the Government says the Suspension Clause affords petitioners no rights because the United States does not claim sovereignty over the place of detention.
Guantanamo Bay is not formally part of the United States. See DTA §1005(g), 119 Stat. 2743. And under the terms of the lease between the United States and Cuba, Cuba retains “ultimate sovereignty” over the territory while the United States exercises “complete jurisdiction and control.” See Lease of Lands for Coaling and Naval Stations, Feb. 23, 1903, U. S.-Cuba, Art. III, T. S. No. 418 (hereinafter 1903 Lease Agreement); Rasul, 542 U. S., at 471. Under the terms of the 1934 Treaty, however, Cuba effectively has no rights as a sovereign until the parties agree to modification of the 1903 Lease Agreement or the United States abandons the base. See Treaty Defining Relations with Cuba, May 29, 1934, U. S.-Cuba, Art. III, 48 Stat. 1683, T. S. No. 866.
The United States contends, nevertheless, that Guantanamo is not within its sovereign control. This was the Government’s position well before the events of September 11, 2001. See, e.g., Brief for Petitioners in Sale v. Haitian Centers Council, Inc., O. T. 1992, No. 92–344, p. 31 (arguing that Guantanamo is territory “outside the United States”). And in other contexts the Court has held that questions of sovereignty are for the political branches to decide. See Vermilya-Brown Co. v. Connell, 335 U. S. 377, 380 (1948) (“[D]etermination of sovereignty over an area is for the legislative and executive departments”); see also Jones v. United States, 137 U. S. 202 (1890); Williams v. Suffolk Ins. Co., 13 Pet. 415, 420 (1839). Even if this were a treaty interpretation case that did not involve a political question, the President’s construction of the lease agreement would be entitled to great respect. See Sumitomo Shoji America, Inc. v. Avagliano, 457 U. S. 176, 184–185 (1982).
We therefore do not question the Government’s position that Cuba, not the United States, maintains sovereignty, in the legal and technical sense of the term, over Guantanamo Bay. But this does not end the analysis. Our cases do not hold it is improper for us to inquire into the objective degree of control the Nation asserts over foreign territory. As commentators have noted, “ ‘[s]overeignty’ is a term used in many senses and is much abused. ” See 1 Restatement (Third) of Foreign Relations Law of the United States §206, Comment b, p. 94 (1986). When we have stated that sovereignty is a political question, we have referred not to sovereignty in the general, colloquial sense, meaning the exercise of dominion or power, see Webster’s New International Dictionary 2406 (2d ed. 1934) (“sovereignty,” definition 3), but sovereignty in the narrow, legal sense of the term, meaning a claim of right, see 1 Restatement (Third) of Foreign Relations, supra, §206, Comment b, at 94 (noting that sovereignty “implies a state’s lawful control over its territory generally to the exclusion of other states, authority to govern in that territory, and authority to apply law there”). Indeed, it is not altogether uncommon for a territory to be under the de jure sovereignty of one nation, while under the plenary control, or practical sovereignty, of another. This condition can occur when the territory is seized during war, as Guantanamo was during the Spanish-American War. See, e.g., Fleming v. Page, 9 How. 603, 614 (1850) (noting that the port of Tampico, conquered by the United States during the war with Mexico, was “undoubtedly … subject to the sovereignty and dominion of the United States,” but that it “does not follow that it was a part of the United States, or that it ceased to be a foreign country”); King v. Earl of Crewe ex parte Sekgome,  2 K. B. 576, 603–604 (C. A.) (opinion of Williams, L. J.) (arguing that the Bechuanaland Protectorate in South Africa was “under His Majesty’s dominion in the sense of power and jurisdiction, but is not under his dominion in the sense of territorial dominion”). Accordingly, for purposes of our analysis, we accept the Government’s position that Cuba, and not the United States, retains de jure sovereignty over Guantanamo Bay. As we did in Rasul, however, we take notice of the obvious and uncontested fact that the United States, by virtue of its complete jurisdiction and control over the base, maintains de facto sovereignty over this territory. See 542 U. S., at 480; id., at 487 (Kennedy, J., concurring in judgment).
Were we to hold that the present cases turn on the political question doctrine, we would be required first to accept the Government’s premise that de jure sovereignty is the touchstone of habeas corpus jurisdiction. This premise, however, is unfounded. For the reasons indicated above, the history of common-law habeas corpus provides scant support for this proposition; and, for the reasons indicated below, that position would be inconsistent with our precedents and contrary to fundamental separation-of-powers principles.
The Court has discussed the issue of the Constitution’s extraterritorial application on many occasions. These decisions undermine the Government’s argument that, at least as applied to noncitizens, the Constitution necessarily stops where de jure sovereignty ends.
The Framers foresaw that the United States would expand and acquire new territories. See American Ins. Co. v. 356 Bales of Cotton, 1 Pet. 511, 542 (1828). Article IV, §3, cl. 1, grants Congress the power to admit new States. Clause 2 of the same section grants Congress the “Power to dispose of and make all needful Rules and Regulations respecting the Territory or other Property belonging to the United States.” Save for a few notable (and notorious) exceptions, e.g., Dred Scott v. Sandford, 19 How. 393 (1857), throughout most of our history there was little need to explore the outer boundaries of the Constitution’s geographic reach. When Congress exercised its power to create new territories, it guaranteed constitutional protections to the inhabitants by statute. See, e.g., An Act: to establish a Territorial Government for Utah, 9 Stat. 458 (“[T]he Constitution and laws of the United States are hereby extended over and declared to be in force in said Territory of Utah”); Rev. Stat. §1891 (“The Constitution and all laws of the United States which are not locally inapplicable shall have the same force and effect within all the organized Territories, and in every Territory hereafter organized as elsewhere within the United States”); see generally Burnett, Untied States: American Expansion and Territorial Deannexation, 72 U. Chi. L. Rev. 797, 825–827 (2005). In particular, there was no need to test the limits of the Suspension Clause because, as early as 1789, Congress extended the writ to the Territories. See Act of Aug. 7, 1789, 1 Stat. 52 (reaffirming Art. II of Northwest Ordinance of 1787, which provided that “[t]he inhabitants of the said territory, shall always be entitled to the benefits of the writ of habeas corpus”).
Fundamental questions regarding the Constitution’s geographic scope first arose at the dawn of the 20th century when the Nation acquired noncontiguous Territories: Puerto Rico, Guam, and the Philippines—ceded to the United States by Spain at the conclusion of the Spanish-American War—and Hawaii—annexed by the United States in 1898. At this point Congress chose to discontinue its previous practice of extending constitutional rights to the territories by statute. See, e.g., An Act Temporarily to provide for the administration of the affairs of civil government in the Philippine Islands, and for other purposes, 32 Stat. 692 (noting that Rev. Stat. §1891 did not apply to the Philippines).
In a series of opinions later known as the Insular Cases, the Court addressed whether the Constitution, by its own force, applies in any territory that is not a State. See De Lima v. Bidwell, 182 U. S. 1 (1901); Dooley v. United States, 182 U. S. 222 (1901); Armstrong v. United States, 182 U. S. 243 (1901); Downes v. Bidwell, 182 U. S. 244 (1901); Hawaii v. Mankichi, 190 U. S. 197 (1903); Dorr v. United States, 195 U. S. 138 (1904). The Court held that the Constitution has independent force in these territories, a force not contingent upon acts of legislative grace. Yet it took note of the difficulties inherent in that position.
Prior to their cession to the United States, the former Spanish colonies operated under a civil-law system, without experience in the various aspects of the Anglo-American legal tradition, for instance the use of grand and petit juries. At least with regard to the Philippines, a complete transformation of the prevailing legal culture would have been not only disruptive but also unnecessary, as the United States intended to grant independence to that Territory. See An Act To declare the purpose of the people of the United States as to the future political status of the people of the Philippine Islands, and to provide a more autonomous government for those islands (Jones Act), 39 Stat. 545 (noting that “it was never the intention of the people of the United States in the incipiency of the War with Spain to make it a war of conquest or for territorial aggrandizement” and that “it is, as it has always been, the purpose of the people of the United States to withdraw their sovereignty over the Philippine Islands and to recognize their independence as soon as a stable government can be established therein”). The Court thus was reluctant to risk the uncertainty and instability that could result from a rule that displaced altogether the existing legal systems in these newly acquired Territories. See Downes, supra, at 282 (“It is obvious that in the annexation of outlying and distant possessions grave questions will arise from differences of race, habits, laws and customs of the people, and from differences of soil, climate and production … ”).
These considerations resulted in the doctrine of territorial incorporation, under which the Constitution applies in full in incorporated Territories surely destined for statehood but only in part in unincorporated Territories. See Dorr, supra, at 143 (“Until Congress shall see fit to incorporate territory ceded by treaty into the United States, … the territory is to be governed under the power existing in Congress to make laws for such territories and subject to such constitutional restrictions upon the powers of that body as are applicable to the situation”); Downes, supra, at 293 (White, J., concurring) (“[T]he determination of what particular provision of the Constitution is applicable, generally speaking, in all cases, involves an inquiry into the situation of the territory and its relations to the United States”). As the Court later made clear, “the real issue in the Insular Cases was not whether the Constitution extended to the Philippines or Porto Rico when we went there, but which of its provisions were applicable by way of limitation upon the exercise of executive and legislative power in dealing with new conditions and requirements.” Balzac v. Porto Rico, 258 U. S. 298, 312 (1922). It may well be that over time the ties between the United States and any of its unincorporated Territories strengthen in ways that are of constitutional significance. Cf. Torres v. Puerto Rico, 442 U. S. 465, 475–476 (1979) (Brennan, J., concurring in judgment) (“Whatever the validity of the [Insular Cases] in the particular historical context in which they were decided, those cases are clearly not authority for questioning the application of the Fourth Amendment—or any other provision of the Bill of Rights—to the Commonwealth of Puerto Rico in the 1970’s”). But, as early as Balzac in 1922, the Court took for granted that even in unincorporated Territories the Government of the United States was bound to provide to noncitizen inhabitants “guaranties of certain fundamental personal rights declared in the Constitution.” 258 U. S., at 312; see also Late Corp. of Church of Jesus Christ of Latter-day Saints v. United States, 136 U. S. 1, 44 (1890) (“Doubtless Congress, in legislating for the Territories would be subject to those fundamental limitations in favor of personal rights which are formulated in the Constitution and its amendments”). Yet noting the inherent practical difficulties of enforcing all constitutional provisions “always and everywhere,” Balzac, supra, at 312, the Court devised in the Insular Cases a doctrine that allowed it to use its power sparingly and where it would be most needed. This century-old doctrine informs our analysis in the present matter.
Practical considerations likewise influenced the Court’s analysis a half-century later in Reid, 354 U. S. 1. The petitioners there, spouses of American servicemen, lived on American military bases in England and Japan. They were charged with crimes committed in those countries and tried before military courts, consistent with executive agreements the United States had entered into with the British and Japanese governments. Id., at 15–16, and nn. 29–30 (plurality opinion). Because the petitioners were not themselves military personnel, they argued they were entitled to trial by jury.
Justice Black, writing for the plurality, contrasted the cases before him with the Insular Cases, which involved territories “with wholly dissimilar traditions and institutions” that Congress intended to govern only “temporarily.” Id., at 14. Justice Frankfurter argued that the “specific circumstances of each particular case” are relevant in determining the geographic scope of the Constitution. Id., at 54 (opinion concurring in result). And Justice Harlan, who had joined an opinion reaching the opposite result in the case in the previous Term, Reid v. Covert, 351 U. S. 487 (1956), was most explicit in rejecting a “rigid and abstract rule” for determining where constitutional guarantees extend. Reid, 354 U. S., at 74 (opinion concurring in result). He read the Insular Cases to teach that whether a constitutional provision has extraterritorial effect depends upon the “particular circumstances, the practical necessities, and the possible alternatives which Congress had before it” and, in particular, whether judicial enforcement of the provision would be “impracticable and anomalous.” Id., at 74–75; see also United States v. Verdugo-Urquidez, 494 U. S. 259, 277–278 (1990) (Kennedy, J., concurring) (applying the “impracticable and anomalous” extraterritoriality test in the Fourth Amendment context).
That the petitioners in Reid were American citizens was a key factor in the case and was central to the plurality’s conclusion that the Fifth and Sixth Amendments apply to American civilians tried outside the United States. But practical considerations, related not to the petitioners’ citizenship but to the place of their confinement and trial, were relevant to each Member of the Reid majority. And to Justices Harlan and Frankfurter (whose votes were necessary to the Court’s disposition) these considerations were the decisive factors in the case.
Indeed the majority splintered on this very point. The key disagreement between the plurality and the concurring Justices in Reid was over the continued precedential value of the Court’s previous opinion in In re Ross, 140 U. S. 453 (1891), which the Reid Court understood as holding that under some circumstances Americans abroad have no right to indictment and trial by jury. The petitioner in Ross was a sailor serving on an American merchant vessel in Japanese waters who was tried before an American consular tribunal for the murder of a fellow crewman. 140 U. S., at 459, 479. The Ross Court held that the petitioner, who was a British subject, had no rights under the Fifth and Sixth Amendments. Id., at 464. The petitioner’s citizenship played no role in the disposition of the case, however. The Court assumed (consistent with the maritime custom of the time) that Ross had all the rights of a similarly situated American citizen. Id., at 479 (noting that Ross was “under the protection and subject to the laws of the United States equally with the seaman who was native born”). The Justices in Reid therefore properly understood Ross as standing for the proposition that, at least in some circumstances, the jury provisions of the Fifth and Sixth Amendments have no application to American citizens tried by American authorities abroad. See 354 U. S., at 11–12 (plurality opinion) (describing Ross as holding that “constitutional protections applied ‘only to citizens and others within the United States … and not to residents or temporary sojourners abroad’ ” (quoting Ross, supra, at 464)); 354 U. S., at 64 (Frankfurter, J., concurring in result) (noting that the consular tribunals upheld in Ross “w[ere] based on long-established custom and they were justified as the best possible means for securing justice for the few Americans present in [foreign] countries”); 354 U. S., at 75 (Harlan, J., concurring in result) (“what Ross and the Insular Cases hold is that the particular local setting, the practical necessities, and the possible alternatives are relevant to a question of judgment, namely, whether jury trial should be deemed a necessary condition of the exercise of Congress’ power to provide for the trial of Americans overseas”).
The Reid plurality doubted that Ross was rightly decided, precisely because it believed the opinion was insufficiently protective of the rights of American citizens. See 354 U. S., at 10–12; see also id., at 78 (Clark, J., dissenting) (noting that “four of my brothers would specifically overrule and two would impair the long-recognized vitality of an old and respected precedent in our law, the case of In re Ross, 140 U. S. 453 (1891)”). But Justices Harlan and Frankfurter, while willing to hold that the American citizen petitioners in the cases before them were entitled to the protections of Fifth and Sixth Amendments, were unwilling to overturn Ross. 354 U. S., at 64 (Frankfurter, J., concurring in result); id., at 75 (Harlan, J., concurring in result). Instead, the two concurring Justices distinguished Ross from the cases before them, not on the basis of the citizenship of the petitioners, but on practical considerations that made jury trial a more feasible option for them than it was for the petitioner in Ross. If citizenship had been the only relevant factor in the case, it would have been necessary for the Court to overturn Ross, something Justices Harlan and Frankfurter were unwilling to do. See Verdugo-Urquidez, supra, at 277 (Kennedy, J., concurring) (noting that Ross had not been overruled).
Practical considerations weighed heavily as well in Johnson v. Eisentrager, 339 U. S. 763 (1950), where the Court addressed whether habeas corpus jurisdiction extended to enemy aliens who had been convicted of violating the laws of war. The prisoners were detained at Landsberg Prison in Germany during the Allied Powers’ postwar occupation. The Court stressed the difficulties of ordering the Government to produce the prisoners in a habeas corpus proceeding. It “would require allocation of shipping space, guarding personnel, billeting and rations” and would damage the prestige of military commanders at a sensitive time. Id., at 779. In considering these factors the Court sought to balance the constraints of military occupation with constitutional necessities. Id., at 769–779; see Rasul, 542 U. S., at 475–476 (discussing the factors relevant to Eisentrager’s constitutional holding); 542 U. S., at 486 (Kennedy, J., concurring in judgment) (same).
True, the Court in Eisentrager denied access to the writ, and it noted the prisoners “at no relevant time were within any territory over which the United States is sovereign, and [that] the scenes of their offense, their capture, their trial and their punishment were all beyond the territorial jurisdiction of any court of the United States.” 339 U. S., at 778. The Government seizes upon this language as proof positive that the Eisentrager Court adopted a formalistic, sovereignty-based test for determining the reach of the Suspension Clause. See Brief for Respondents 18–20. We reject this reading for three reasons.
First, we do not accept the idea that the above-quoted passage from Eisentrager is the only authoritative language in the opinion and that all the rest is dicta. The Court’s further determinations, based on practical considerations, were integral to Part II of its opinion and came before the decision announced its holding. See 339 U. S., at 781.
Second, because the United States lacked both de jure sovereignty and plenary control over Landsberg Prison, see infra, at 34–35, it is far from clear that the Eisentrager Court used the term sovereignty only in the narrow technical sense and not to connote the degree of control the military asserted over the facility. See supra, at 21. The Justices who decided Eisentrager would have understood sovereignty as a multifaceted concept. See Black’s Law Dictionary 1568 (4th ed. 1951) (defining “sovereignty” as “[t]he supreme, absolute, and uncontrollable power by which any independent state is governed”; “the international independence of a state, combined with the right and power of regulating its internal affairs without foreign dictation”; and “[t]he power to do everything in a state without accountability”); Ballentine’s Law Dictionary with Pronunciations 1216 (2d ed. 1948) (defining “sovereignty” as “[t]hat public authority which commands in civil society, and orders and directs what each citizen is to perform to obtain the end of its institution”). In its principal brief in Eisentrager, the Government advocated a bright-line test for determining the scope of the writ, similar to the one it advocates in these cases. See Brief for Petitioners in Johnson v. Eisentrager, O. T. 1949, No. 306, pp. 74–75. Yet the Court mentioned the concept of territorial sovereignty only twice in its opinion. See Eisentrager, supra, at 778, 780. That the Court devoted a significant portion of Part II to a discussion of practical barriers to the running of the writ suggests that the Court was not concerned exclusively with the formal legal status of Landsberg Prison but also with the objective degree of control the United States asserted over it. Even if we assume the Eisentrager Court considered the United States’ lack of formal legal sovereignty over Landsberg Prison as the decisive factor in that case, its holding is not inconsistent with a functional approach to questions of extraterritoriality. The formal legal status of a given territory affects, at least to some extent, the political branches’ control over that territory. De jure sovereignty is a factor that bears upon which constitutional guarantees apply there.
Third, if the Government’s reading of Eisentrager were correct, the opinion would have marked not only a change in, but a complete repudiation of, the Insular Cases’ (and later Reid’s) functional approach to questions of extraterritoriality. We cannot accept the Government’s view. Nothing in Eisentrager says that de jure sovereignty is or has ever been the only relevant consideration in determining the geographic reach of the Constitution or of habeas corpus. Were that the case, there would be considerable tension between Eisentrager, on the one hand, and the Insular Cases and Reid, on the other. Our cases need not be read to conflict in this manner. A constricted reading of Eisentrager overlooks what we see as a common thread uniting the Insular Cases, Eisentrager, and Reid: the idea that questions of extraterritoriality turn on objective factors and practical concerns, not formalism.
The Government’s formal sovereignty-based test raises troubling separation-of-powers concerns as well. The political history of Guantanamo illustrates the deficiencies of this approach. The United States has maintained complete and uninterrupted control of the bay for over 100 years. At the close of the Spanish-American War, Spain ceded control over the entire island of Cuba to the United States and specifically “relinquishe[d] all claim[s] of sovereignty … and title.” See Treaty of Paris, Dec. 10, 1898, U. S.-Spain, Art. I, 30 Stat. 1755, T. S. No. 343. From the date the treaty with Spain was signed until the Cuban Republic was established on May 20, 1902, the United States governed the territory “in trust” for the benefit of the Cuban people. Neely v. Henkel, 180 U. S. 109, 120 (1901); H. Thomas, Cuba or The Pursuit of Freedom 436, 460 (1998). And although it recognized, by entering into the 1903 Lease Agreement, that Cuba retained “ultimate sovereignty” over Guantanamo, the United States continued to maintain the same plenary control it had enjoyed since 1898. Yet the Government’s view is that the Constitution had no effect there, at least as to noncitizens, because the United States disclaimed sovereignty in the formal sense of the term. The necessary implication of the argument is that by surrendering formal sovereignty over any unincorporated territory to a third party, while at the same time entering into a lease that grants total control over the territory back to the United States, it would be possible for the political branches to govern without legal constraint.
Our basic charter cannot be contracted away like this. The Constitution grants Congress and the President the power to acquire, dispose of, and govern territory, not the power to decide when and where its terms apply. Even when the United States acts outside its borders, its powers are not “absolute and unlimited” but are subject “to such restrictions as are expressed in the Constitution.” Murphy v. Ramsey, 114 U. S. 15, 44 (1885). Abstaining from questions involving formal sovereignty and territorial governance is one thing. To hold the political branches have the power to switch the Constitution on or off at will is quite another. The former position reflects this Court’s recognition that certain matters requiring political judgments are best left to the political branches. The latter would permit a striking anomaly in our tripartite system of government, leading to a regime in which Congress and the President, not this Court, say “what the law is.” Marbury v. Madison, 1 Cranch 137, 177 (1803).
These concerns have particular bearing upon the Suspension Clause question in the cases now before us, for the writ of habeas corpus is itself an indispensable mechanism for monitoring the separation of powers. The test for determining the scope of this provision must not be subject to manipulation by those whose power it is designed to restrain.
As we recognized in Rasul, 542 U. S., at 476; id., at 487 (Kennedy, J., concurring in judgment), the outlines of a framework for determining the reach of the Suspension Clause are suggested by the factors the Court relied upon in Eisentrager. In addition to the practical concerns discussed above, the Eisentrager Court found relevant that each petitioner:
“(a) is an enemy alien; (b) has never been or resided in the United States; (c) was captured outside of our territory and there held in military custody as a prisoner of war; (d) was tried and convicted by a Military Commission sitting outside the United States; (e) for offenses against laws of war committed outside the United States; (f) and is at all times imprisoned outside the United States.” 339 U. S., at 777.
Based on this language from Eisentrager, and the reasoning in our other extraterritoriality opinions, we conclude that at least three factors are relevant in determining the reach of the Suspension Clause: (1) the citizenship and status of the detainee and the adequacy of the process through which that status determination was made; (2) the nature of the sites where apprehension and then detention took place; and (3) the practical obstacles inherent in resolving the prisoner’s entitlement to the writ.
Applying this framework, we note at the onset that the status of these detainees is a matter of dispute. The petitioners, like those in Eisentrager, are not American citizens. But the petitioners in Eisentrager did not contest, it seems, the Court’s assertion that they were “enemy alien[s].” Ibid. In the instant cases, by contrast, the detainees deny they are enemy combatants. They have been afforded some process in CSRT proceedings to determine their status; but, unlike in Eisentrager, supra, at 766, there has been no trial by military commission for violations of the laws of war. The difference is not trivial. The records from the Eisentrager trials suggest that, well before the petitioners brought their case to this Court, there had been a rigorous adversarial process to test the legality of their detention. The Eisentrager petitioners were charged by a bill of particulars that made detailed factual allegations against them. See 14 United Nations War Crimes Commission, Law Reports of Trials of War Criminals 8–10 (1949) (reprint 1997). To rebut the accusations, they were entitled to representation by counsel, allowed to introduce evidence on their own behalf, and permitted to cross-examine the prosecution’s witnesses. See Memorandum by Command of Lt. Gen. Wedemeyer, Jan. 21, 1946 (establishing “Regulations Governing the Trial of War Criminals” in the China Theater), in Tr. of Record in Johnson v. Eisentrager, O. T. 1949, No. 306, pp. 34–40.
In comparison the procedural protections afforded to the detainees in the CSRT hearings are far more limited, and, we conclude, fall well short of the procedures and adversarial mechanisms that would eliminate the need for habeas corpus review. Although the detainee is assigned a “Personal Representative” to assist him during CSRT proceedings, the Secretary of the Navy’s memorandum makes clear that person is not the detainee’s lawyer or even his “advocate.” See App. to Pet. for Cert. in No. 06–1196, at 155, 172. The Government’s evidence is accorded a presumption of validity. Id., at 159. The detainee is allowed to present “reasonably available” evidence, id., at 155, but his ability to rebut the Government’s evidence against him is limited by the circumstances of his confinement and his lack of counsel at this stage. And although the detainee can seek review of his status determination in the Court of Appeals, that review process cannot cure all defects in the earlier proceedings. See Part V, infra.
As to the second factor relevant to this analysis, the detainees here are similarly situated to the Eisentrager petitioners in that the sites of their apprehension and detention are technically outside the sovereign territory of the United States. As noted earlier, this is a factor that weighs against finding they have rights under the Suspension Clause. But there are critical differences between Landsberg Prison, circa 1950, and the United States Naval Station at Guantanamo Bay in 2008. Unlike its present control over the naval station, the United States’ control over the prison in Germany was neither absolute nor indefinite. Like all parts of occupied Germany, the prison was under the jurisdiction of the combined Allied Forces. See Declaration Regarding the Defeat of Germany and the Assumption of Supreme Authority with Respect to Germany, June 5, 1945, U. S.-U. S. S. R.-U. K.-Fr., 60 Stat. 1649, T. I. A. S. No. 1520. The United States was therefore answerable to its Allies for all activities occurring there. Cf. Hirota v. MacArthur, 338 U. S. 197, 198 (1948) (per curiam) (military tribunal set up by Gen. Douglas MacArthur, acting as “the agent of the Allied Powers,” was not a “tribunal of the United States”). The Allies had not planned a long-term occupation of Germany, nor did they intend to displace all German institutions even during the period of occupation. See Agreements Respecting Basic Principles for Merger of the Three Western German Zones of Occupation, and Other Matters, Apr. 8, 1949, U. S.-U. K.-Fr., Art. 1, 63 Stat. 2819, T. I. A. S. No. 2066 (establishing a governing framework “[d]uring the period in which it is necessary that the occupation continue” and expressing the desire “that the German people shall enjoy self-government to the maximum possible degree consistent with such occupation”). The Court’s holding in Eisentrager was thus consistent with the Insular Cases, where it had held there was no need to extend full constitutional protections to territories the United States did not intend to govern indefinitely. Guantanamo Bay, on the other hand, is no transient possession. In every practical sense Guantanamo is not abroad; it is within the constant jurisdiction of the United States. See Rasul, 542 U. S., at 480; id., at 487 (Kennedy, J., concurring in judgment).
As to the third factor, we recognize, as the Court did in Eisentrager, that there are costs to holding the Suspension Clause applicable in a case of military detention abroad. Habeas corpus proceedings may require expenditure of funds by the Government and may divert the attention of military personnel from other pressing tasks. While we are sensitive to these concerns, we do not find them dispositive. Compliance with any judicial process requires some incremental expenditure of resources. Yet civilian courts and the Armed Forces have functioned along side each other at various points in our history. See, e.g., Duncan v. Kahanamoku, 327 U. S. 304 (1946); Ex parte Milligan, 4 Wall. 2 (1866). The Government presents no credible arguments that the military mission at Guantanamo would be compromised if habeas corpus courts had jurisdiction to hear the detainees’ claims. And in light of the plenary control the United States asserts over the base, none are apparent to us.
The situation in Eisentrager was far different, given the historical context and nature of the military’s mission in post-War Germany. When hostilities in the European Theater came to an end, the United States became responsible for an occupation zone encompassing over 57,000 square miles with a population of 18 million. See Letter from President Truman to Secretary of State Byrnes, (Nov. 28, 1945), in 8 Documents on American Foreign Relations 257 (R. Dennett & R. Turner eds. 1948); Pollock, A Territorial Pattern for the Military Occupation of Germany, 38 Am. Pol. Sci. Rev. 970, 975 (1944). In addition to supervising massive reconstruction and aid efforts the American forces stationed in Germany faced potential security threats from a defeated enemy. In retrospect the post-War occupation may seem uneventful. But at the time Eisentrager was decided, the Court was right to be concerned about judicial interference with the military’s efforts to contain “enemy elements, guerilla fighters, and ‘were-wolves.’ ” 339 U. S., at 784.
Similar threats are not apparent here; nor does the Government argue that they are. The United States Naval Station at Guantanamo Bay consists of 45 square miles of land and water. The base has been used, at various points, to house migrants and refugees temporarily. At present, however, other than the detainees themselves, the only long-term residents are American military personnel, their families, and a small number of workers. See History of Guantanamo Bay online at https://www.cnic. navy.mil/Guantanamo/AboutGTMO/gtmohistorygeneral/ gtmohistgeneral. The detainees have been deemed enemies of the United States. At present, dangerous as they may be if released, they are contained in a secure prison facility located on an isolated and heavily fortified military base.
There is no indication, furthermore, that adjudicating a habeas corpus petition would cause friction with the host government. No Cuban court has jurisdiction over American military personnel at Guantanamo or the enemy combatants detained there. While obligated to abide by the terms of the lease, the United States is, for all practical purposes, answerable to no other sovereign for its acts on the base. Were that not the case, or if the detention facility were located in an active theater of war, arguments that issuing the writ would be “impracticable or anomalous” would have more weight. See Reid, 354 U. S., at 74 (Harlan, J., concurring in result). Under the facts presented here, however, there are few practical barriers to the running of the writ. To the extent barriers arise, habeas corpus procedures likely can be modified to address them. See Part VI–B, infra.
It is true that before today the Court has never held that noncitizens detained by our Government in territory over which another country maintains de jure sovereignty have any rights under our Constitution. But the cases before us lack any precise historical parallel. They involve individuals detained by executive order for the duration of a conflict that, if measured from September 11, 2001, to the present, is already among the longest wars in American history. See Oxford Companion to American Military History 849 (1999). The detainees, moreover, are held in a territory that, while technically not part of the United States, is under the complete and total control of our Government. Under these circumstances the lack of a precedent on point is no barrier to our holding.
We hold that Art. I, §9, cl. 2, of the Constitution has full effect at Guantanamo Bay. If the privilege of habeas corpus is to be denied to the detainees now before us, Congress must act in accordance with the requirements of the Suspension Clause. Cf. Hamdi, 542 U. S., at 564 (Scalia, J., dissenting) (“[I]ndefinite imprisonment on reasonable suspicion is not an available option of treatment for those accused of aiding the enemy, absent a suspension of the writ”). This Court may not impose a de facto suspension by abstaining from these controversies. See Hamdan, 548 U. S., at 585, n. 16 (“[A]bstention is not appropriate in cases … in which the legal challenge ‘turn[s] on the status of the persons as to whom the military asserted its power’ ” (quoting Schlesinger v. Councilman, 420 U. S. 738, 759 (1975))). The 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. Petitioners, therefore, are entitled to the privilege of habeas corpus to challenge the legality of their detention.
In light of this holding the question becomes whether the statute stripping jurisdiction to issue the writ avoids the Suspension Clause mandate because Congress has provided adequate substitute procedures for habeas corpus. The Government submits there has been compliance with the Suspension Clause because the DTA review process in the Court of Appeals, see DTA §1005(e), provides an adequate substitute. Congress has granted that court jurisdiction to consider
“(i) whether the status determination of the [CSRT] … was consistent with the standards and procedures specified by the Secretary of Defense … and (ii) to the extent the Constitution and laws of the United States are applicable, whether the use of such standards and procedures to make the determination is consistent with the Constitution and laws of the United States.” §1005(e)(2)(C), 119 Stat. 2742.
The Court of Appeals, having decided that the writ does not run to the detainees in any event, found it unnecessary to consider whether an adequate substitute has been provided. In the ordinary course we would remand to the Court of Appeals to consider this question in the first instance. See Youakim v. Miller, 425 U. S. 231, 234 (1976) (per curiam). It is well settled, however, that the Court’s practice of declining to address issues left unresolved in earlier proceedings is not an inflexible rule. Ibid. Departure from the rule is appropriate in “exceptional” circumstances. See Cooper Industries, Inc. v. Aviall Services, Inc., 543 U. S. 157, 169 (2004); Duignan v. United States, 274 U. S. 195, 200 (1927).
The gravity of the separation-of-powers issues raised by these cases and the fact that these detainees have been denied meaningful access to a judicial forum for a period of years render these cases exceptional. The parties before us have addressed the adequacy issue. While we would have found it informative to consider the reasoning of the Court of Appeals on this point, we must weigh that against the harms petitioners may endure from additional delay. And, given there are few precedents addressing what features an adequate substitute for habeas corpus must contain, in all likelihood a remand simply would delay ultimate resolution of the issue by this Court.
We do have the benefit of the Court of Appeals’ construction of key provisions of the DTA. When we granted certiorari in these cases, we noted “it would be of material assistance to consult any decision” in the parallel DTA review proceedings pending in the Court of Appeals, specifically any rulings in the matter of Bismullah v. Gates. 551 U. S. ___ (2007). Although the Court of Appeals has yet to complete a DTA review proceeding, the three-judge panel in Bismullah has issued an interim order giving guidance as to what evidence can be made part of the record on review and what access the detainees can have to counsel and to classified information. See 501 F. 3d 178 (CADC) (Bismullah I), reh’g denied, 503 F. 3d 137 (CADC 2007) (Bismullah II). In that matter the full court denied the Government’s motion for rehearing en banc, see Bismullah v. Gates, 514 F. 3d 1291 (CADC 2008) (Bismullah III). The order denying rehearing was accompanied by five separate statements from members of the court, which offer differing views as to scope of the judicial review Congress intended these detainees to have. Ibid.
Under the circumstances we believe the costs of further delay substantially outweigh any benefits of remanding to the Court of Appeals to consider the issue it did not address in these cases.
Our case law does not contain extensive discussion of standards defining suspension of the writ or of circumstances under which suspension has occurred. This simply confirms the care Congress has taken throughout our Nation’s history to preserve the writ and its function. Indeed, most of the major legislative enactments pertaining to habeas corpus have acted not to contract the writ’s protection but to expand it or to hasten resolution of prisoners’ claims. See, e.g., Habeas Corpus Act of 1867, ch. 28, §1, 14 Stat. 385 (current version codified at 28 U. S. C. §2241 (2000 ed. and Supp. V) (extending the federal writ to state prisoners)); Cf. Harris v. Nelson, 394 U. S. 286, 299–300 (1969) (interpreting the All Writs Act, 28 U. S. C. §1651, to allow discovery in habeas corpus proceedings); Peyton v. Rowe, 391 U. S. 54, 64–65 (1968) (interpreting the then-existing version of §2241 to allow petitioner to proceed with his habeas corpus action, even though he had not yet begun to serve his sentence).
There are exceptions, of course. Title I of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), §106, 110 Stat. 1220, contains certain gatekeeping provisions that restrict a prisoner’s ability to bring new and repetitive claims in “second or successive” habeas corpus actions. We upheld these provisions against a Suspension Clause challenge in Felker v. Turpin, 518 U. S. 651, 662–664 (1996). The provisions at issue in Felker, however, did not constitute a substantial departure from common-law habeas procedures. The provisions, for the most part, codified the longstanding abuse-of-the-writ doctrine. Id., at 664; see also McCleskey v. Zant, 499 U. S. 467, 489 (1991). AEDPA applies, moreover, to federal, postconviction review after criminal proceedings in state court have taken place. As of this point, cases discussing the implementation of that statute give little helpful instruction (save perhaps by contrast) for the instant cases, where no trial has been held.
The two leading cases addressing habeas substitutes, Swain v. Pressley, 430 U. S. 372 (1977), and United States v. Hayman, 342 U. S. 205 (1952), likewise provide little guidance here. The statutes at issue were attempts to streamline habeas corpus relief, not to cut it back.
The statute discussed in Hayman was 28 U. S. C. §2255. It replaced traditional habeas corpus for federal prisoners (at least in the first instance) with a process that allowed the prisoner to file a motion with the sentencing court on the ground that his sentence was, inter alia, “ ‘imposed in violation of the Constitution or laws of the United States.’ ” 342 U. S., at 207, n. 1. The purpose and effect of the statute was not to restrict access to the writ but to make postconviction proceedings more efficient. It directed claims not to the court that had territorial jurisdiction over the place of the petitioner’s confinement but to the sentencing court, a court already familiar with the facts of the case. As the Hayman Court explained
“Section 2255 … was passed at the instance of the Judicial Conference to meet practical difficulties that had arisen in administering the habeas corpus jurisdiction of the federal courts. Nowhere in the history of Section 2255 do we find any purpose to impinge upon prisoners’ rights of collateral attack upon their convictions. On the contrary, the sole purpose was to minimize the difficulties encountered in habeas corpus hearings by affording the same rights in another and more convenient forum.” Id., at 219.
See also Hill v. United States, 368 U. S. 424, 427, 428, and n. 5 (1962) (noting that §2255 provides a remedy in the sentencing court that is “exactly commensurate” with the pre-existing federal habeas corpus remedy).
The statute in Swain, D. C. Code Ann. §23–110(g) (1973), applied to prisoners in custody under sentence of the Superior Court of the District of Columbia. Before enactment of the District of Columbia Court Reform and Criminal Procedure Act of 1970 (D. C. Court Reform Act), 84 Stat. 473, those prisoners could file habeas petitions in the United States District Court for the District of Columbia. The Act, which was patterned on §2255, substituted a new collateral process in the Superior Court for the pre-existing habeas corpus procedure in the District Court. See Swain, 430 U. S., at 374–378. But, again, the purpose and effect of the statute was to expedite consideration of the prisoner’s claims, not to delay or frustrate it. See id., at 375, n. 4 (noting that the purpose of the D. C. Court Reform Act was to “alleviate” administrative burdens on the District Court).
That the statutes in Hayman and Swain were designed to strengthen, rather than dilute, the writ’s protections was evident, furthermore, from this significant fact: Neither statute eliminated traditional habeas corpus relief. In both cases the statute at issue had a saving clause, providing that a writ of habeas corpus would be available if the alternative process proved inadequate or ineffective. Swain, supra, at 381; Hayman, supra, at 223. The Court placed explicit reliance upon these provisions in upholding the statutes against constitutional challenges. See Swain, supra, at 381 (noting that the provision “avoid[ed] any serious question about the constitutionality of the statute”); Hayman, supra, at 223 (noting that, because habeas remained available as a last resort, it was unnecessary to “reach constitutional questions”).
Unlike in Hayman and Swain, here we confront statutes, the DTA and the MCA, that were intended to circumscribe habeas review. Congress’ purpose is evident not only from the unequivocal nature of MCA §7’s jurisdiction-stripping language, 28 U. S. C. A. §2241(e)(1) (Supp. 2007) (“No court, justice, or judge shall have jurisdiction to hear or consider an application for a writ of habeas corpus . . .”), but also from a comparison of the DTA to the statutes at issue in Hayman and Swain. When interpreting a statute, we examine related provisions in other parts of the U. S. Code. See, e.g., West Virginia Univ. Hospitals, Inc. v. Casey, 499 U. S. 83, 88–97 (1991); Babbitt v. Sweet Home Chapter, Communities for Great Ore., 515 U. S. 687, 717–718 (1995) (Scalia, J., dissenting); see generally W. Eskridge, P. Frickey, & E. Garrett, Cases and Materials on Legislation: Statutes and the Creation of Public Policy 1039 (3d ed. 2001). When Congress has intended to replace traditional habeas corpus with habeas-like substitutes, as was the case in Hayman and Swain, it has granted to the courts broad remedial powers to secure the historic office of the writ. In the §2255 context, for example, Congress has granted to the reviewing court power to “determine the issues and make findings of fact and conclusions of law” with respect to whether “the judgment [of conviction] was rendered without jurisdiction, or … the sentence imposed was not authorized by law or otherwise open to collateral attack.” 28 U. S. C. A. §2255(b) (Supp. 2008). The D. C. Court Reform Act, the statute upheld in Swain, contained a similar provision. §23–110(g), 84 Stat. 609.
In contrast the DTA’s jurisdictional grant is quite limited. The Court of Appeals has jurisdiction not to inquire into the legality of the detention generally but only to assess whether the CSRT complied with the “standards and procedures specified by the Secretary of Defense” and whether those standards and procedures are lawful. DTA §1005(e)(2)(C), 119 Stat. 2742. If Congress had envisioned DTA review as coextensive with traditional habeas corpus, it would not have drafted the statute in this manner. Instead, it would have used language similar to what it used in the statutes at issue in Hayman and Swain. Cf. Russello v. United States, 464 U. S. 16, 23 (1983) (“ ‘[W]here Congress includes particular language in one section of a statute but omits it in another section of the same Act, it is generally presumed that Congress acts intentionally and purposely in the disparate inclusion or exclusion’ ” (quoting United States v. Wong Kim Bo, 472 F. 2d 720, 722 (CA5 1972))). Unlike in Hayman and Swain, moreover, there has been no effort to preserve habeas corpus review as an avenue of last resort. No saving clause exists in either the MCA or the DTA. And MCA §7 eliminates habeas review for these petitioners.
The differences between the DTA and the habeas statute that would govern in MCA §7’s absence, 28 U. S. C. §2241 (2000 ed. and Supp. V), are likewise telling. In §2241 (2000 ed.) Congress confirmed the authority of “any justice” or “circuit judge” to issue the writ. Cf. Felker, 518 U. S., at 660–661 (interpreting Title I of AEDPA to not strip from this Court the power to entertain original habeas corpus petitions). That statute accommodates the necessity for factfinding that will arise in some cases by allowing the appellate judge or Justice to transfer the case to a district court of competent jurisdiction, whose institutional capacity for factfinding is superior to his or her own. See 28 U. S. C. §2241(b). By granting the Court of Appeals “exclusive” jurisdiction over petitioners’ cases, see DTA §1005(e)(2)(A), 119 Stat. 2742, Congress has foreclosed that option. This choice indicates Congress intended the Court of Appeals to have a more limited role in enemy combatant status determinations than a district court has in habeas corpus proceedings. The DTA should be interpreted to accord some latitude to the Court of Appeals to fashion procedures necessary to make its review function a meaningful one, but, if congressional intent is to be respected, the procedures adopted cannot be as extensive or as protective of the rights of the detainees as they would be in a §2241 proceeding. Otherwise there would have been no, or very little, purpose for enacting the DTA.
To the extent any doubt remains about Congress’ intent, the legislative history confirms what the plain text strongly suggests: In passing the DTA Congress did not intend to create a process that differs from traditional habeas corpus process in name only. It intended to create a more limited procedure. See, e.g., 151 Cong. Rec. S14263 (Dec. 21, 2005) (statement of Sen. Graham) (noting that the DTA “extinguish[es] these habeas and other actions in order to effect a transfer of jurisdiction over these cases to the DC Circuit Court” and agreeing that the bill “create[s] in their place a very limited judicial review of certain military administrative decisions”); id., at S14268 (statement of Sen. Kyl) (“It is important to note that the limited judicial review authorized by paragraphs 2 and 3 of subsection (e) [of DTA §1005] are not habeas-corpus review. It is a limited judicial review of its own nature”).
It is against this background that we must interpret the DTA and assess its adequacy as a substitute for habeas corpus. The present cases thus test the limits of the Suspension Clause in ways that Hayman and Swain did not.
We do not endeavor to offer a comprehensive summary of the requisites for an adequate substitute for habeas corpus. We do consider it uncontroversial, however, that the privilege of habeas corpus entitles the prisoner to a meaningful opportunity to demonstrate that he is being held pursuant to “the erroneous application or interpretation” of relevant law. St. Cyr, 533 U. S., at 302. And the habeas court must have the power to order the conditional release of an individual unlawfully detained—though release need not be the exclusive remedy and is not the appropriate one in every case in which the writ is granted. See Ex parte Bollman, 4 Cranch 75, 136 (1807) (where imprisonment is unlawful, the court “can only direct [the prisoner] to be discharged”); R. Hurd, Treatise on the Right of Personal Liberty, and On the Writ of Habeas Corpus and the Practice Connected with It: With a View of the Law of Extradition of Fugitives 222 (2d ed. 1876) (“It cannot be denied where ‘a probable ground is shown that the party is imprisoned without just cause, and therefore, hath a right to be delivered,’ for the writ then becomes a ‘writ of right, which may not be denied but ought to be granted to every man that is committed or detained in prison or otherwise restrained of his liberty’ ”). But see Chessman v. Teets, 354 U. S. 156, 165–166 (1957) (remanding in a habeas case for retrial within a “reasonable time”). These are the easily identified attributes of any constitutionally adequate habeas corpus proceeding. But, depending on the circumstances, more may be required.
Indeed, common-law habeas corpus was, above all, an adaptable remedy. Its precise application and scope changed depending upon the circumstances. See 3 Blackstone *131 (describing habeas as “the great and efficacious writ, in all manner of illegal confinement”); see also Schlup v. Delo, 513 U. S. 298, 319 (1995) (Habeas “is, at its core, an equitable remedy”); Jones v. Cunningham, 371 U. S. 236, 243 (1963) (Habeas is not “a static, narrow, formalistic remedy; its scope has grown to achieve its grand purpose”). It appears the common-law habeas court’s role was most extensive in cases of pretrial and noncriminal detention, where there had been little or no previous judicial review of the cause for detention. Notably, the black-letter rule that prisoners could not controvert facts in the jailer’s return was not followed (or at least not with consistency) in such cases. Hurd, supra, at 271 (noting that the general rule was “subject to exceptions” including cases of bail and impressment); Oakes, Legal History in the High Court—Habeas Corpus, 64 Mich. L. Rev. 451, 457 (1966) (“[W]hen a prisoner applied for habeas corpus before indictment or trial, some courts examined the written depositions on which he had been arrested or committed, and others even heard oral testimony to determine whether the evidence was sufficient to justifying holding him for trial” (footnotes omitted)); Fallon & Meltzer, Habeas Corpus Jurisdiction, Substantive Rights, and the War on Terror, 120 Harv. L. Rev. 2029, 2102 (2007) (“[T]he early practice was not consistent: courts occasionally permitted factual inquiries when no other opportunity for judicial review existed”).
There is evidence from 19th-century American sources indicating that, even in States that accorded strong res judicata effect to prior adjudications, habeas courts in this country routinely allowed prisoners to introduce exculpatory evidence that was either unknown or previously unavailable to the prisoner. See, e.g., Ex parte Pattison, 56 Miss. 161, 164 (1878) (noting that “[w]hile the former adjudication must be considered as conclusive on the testimony then adduced” “newly developed exculpatory evidence … may authorize the admission to bail”); Ex parte Foster, 5 Tex. Ct. App. 625, 644 (1879) (construing the State’s habeas statute to allow for the introduction of new evidence “where important testimony has been obtained, which, though not newly discovered, or which, though known to [the petitioner], it was not in his power to produce at the former hearing; [and] where the evidence was newly discovered”); People v. Martin, 7 N. Y. Leg. Obs. 49, 56 (1848) (“If in custody on criminal process before indictment, the prisoner has an absolute right to demand that the original depositions be looked into to see whether any crime is in fact imputed to him, and the inquiry will by no means be confined to the return. Facts out of the return may be gone into to ascertain whether the committing magistrate may not have arrived at an illogical conclusion upon the evidence given before him …”); see generally W. Church, Treatise on the Writ of Habeas Corpus §182, p. 235 1886) (hereinafter Church) (noting that habeas courts would “hear evidence anew if justice require it”). Justice McLean, on Circuit in 1855, expressed his view that a habeas court should consider a prior judgment conclusive “where there was clearly jurisdiction and a full and fair hearing; but that it might not be so considered when any of these requisites were wanting.” Ex parte Robinson, 20 F. Cas. 969, 971, (No. 11,935) (CC Ohio 1855). To illustrate the circumstances in which the prior adjudication did not bind the habeas court, he gave the example of a case in which “[s]everal unimpeached witnesses” provided new evidence to exculpate the prisoner. Ibid.
The idea that the necessary scope of habeas review in part depends upon the rigor of any earlier proceedings accords with our test for procedural adequacy in the due process context. See Mathews v. Eldridge, 424 U. S. 319, 335 (1976) (noting that the Due Process Clause requires an assessment of, inter alia, “the risk of an erroneous deprivation of [a liberty interest;] and the probable value, if any, of additional or substitute procedural safeguards”). This principle has an established foundation in habeas corpus jurisprudence as well, as Chief Justice Marshall’s opinion in Ex parte Watkins, 3 Pet. 193 (1830), demonstrates. Like the petitioner in Swain, Watkins sought a writ of habeas corpus after being imprisoned pursuant to a judgment of a District of Columbia court. In holding that the judgment stood on “high ground,” 3 Pet., at 209, the Chief Justice emphasized the character of the court that rendered the original judgment, noting it was a “court of record, having general jurisdiction over criminal cases.” Id., at 203. In contrast to “inferior” tribunals of limited jurisdiction, ibid., courts of record had broad remedial powers, which gave the habeas court greater confidence in the judgment’s validity. See generally Neuman, Habeas Corpus, Executive Detention, and the Removal of Aliens, 98 Colum. L. Rev. 961, 982–983 (1998).
Accordingly, where relief is sought from a sentence that resulted from the judgment of a court of record, as was the case in Watkins and indeed in most federal habeas cases, considerable deference is owed to the court that ordered confinement. See Brown v. Allen, 344 U. S. 443, 506 (1953) (opinion of Frankfurter, J.) (noting that a federal habeas court should accept a state court’s factual findings unless “a vital flaw be found in the process of ascertaining such facts in the State court”). Likewise in those cases the prisoner should exhaust adequate alternative remedies before filing for the writ in federal court. See Ex parte Royall, 117 U. S. 241, 251–252 (1886) (requiring exhaustion of state collateral processes). Both aspects of federal habeas corpus review are justified because it can be assumed that, in the usual course, a court of record provides defendants with a fair, adversary proceeding. In cases involving state convictions this framework also respects federalism; and in federal cases it has added justification because the prisoner already has had a chance to seek review of his conviction in a federal forum through a direct appeal. The present cases fall outside these categories, however; for here the detention is by executive order.
Where a person is detained by executive order, rather than, say, after being tried and convicted in a court, the need for collateral review is most pressing. A criminal conviction in the usual course occurs after a judicial hearing before a tribunal disinterested in the outcome and committed to procedures designed to ensure its own independence. These dynamics are not inherent in executive detention orders or executive review procedures. In this context the need for habeas corpus is more urgent. The intended duration of the detention and the reasons for it bear upon the precise scope of the inquiry. Habeas corpus proceedings need not resemble a criminal trial, even when the detention is by executive order. But the writ must be effective. The habeas court must have sufficient authority to conduct a meaningful review of both the cause for detention and the Executive’s power to detain.
To determine the necessary scope of habeas corpus review, therefore, we must assess the CSRT process, the mechanism through which petitioners’ designation as enemy combatants became final. Whether one characterizes the CSRT process as direct review of the Executive’s battlefield determination that the detainee is an enemy combatant—as the parties have and as we do—or as the first step in the collateral review of a battlefield determination makes no difference in a proper analysis of whether the procedures Congress put in place are an adequate substitute for habeas corpus. What matters is the sum total of procedural protections afforded to the detainee at all stages, direct and collateral.
Petitioners identify what they see as myriad deficiencies in the CSRTs. The most relevant for our purposes are the constraints upon the detainee’s ability to rebut the factual basis for the Government’s assertion that he is an enemy combatant. As already noted, see Part IV–C, supra, at the CSRT stage the detainee has limited means to find or present evidence to challenge the Government’s case against him. He does not have the assistance of counsel and may not be aware of the most critical allegations that the Government relied upon to order his detention. See App. to Pet. for Cert. in No. 06–1196, at 156, ¶F(8) (noting that the detainee can access only the “unclassified portion of the Government Information”). The detainee can confront witnesses that testify during the CSRT proceedings. Id., at 144, ¶g(8). But given that there are in effect no limits on the admission of hearsay evidence—the only requirement is that the tribunal deem the evidence “relevant and helpful,” ibid., ¶g(9)—the detainee’s opportunity to question witnesses is likely to be more theoretical than real.
The Government defends the CSRT process, arguing that it was designed to conform to the procedures suggested by the plurality in Hamdi. See 542 U. S., at 538. Setting aside the fact that the relevant language in Hamdi did not garner a majority of the Court, it does not control the matter at hand. None of the parties in Hamdi argued there had been a suspension of the writ. Nor could they. The §2241 habeas corpus process remained in place, id., at 525. Accordingly, the plurality concentrated on whether the Executive had the authority to detain and, if so, what rights the detainee had under the Due Process Clause. True, there are places in the Hamdi plurality opinion where it is difficult to tell where its extrapolation of §2241 ends and its analysis of the petitioner’s Due Process rights begins. But the Court had no occasion to define the necessary scope of habeas review, for Suspension Clause purposes, in the context of enemy combatant detentions. The closest the plurality came to doing so was in discussing whether, in light of separation-of-powers concerns, §2241 should be construed to forbid the District Court from inquiring beyond the affidavit Hamdi’s custodian provided in answer to the detainee’s habeas petition. The plurality answered this question with an emphatic “no.” Id., at 527 (labeling this argument as “extreme”); id., at 535–536.
Even if we were to assume that the CSRTs satisfy due process standards, it would not end our inquiry. Habeas corpus is a collateral process that exists, in Justice Holmes’ words, to “cu[t] through all forms and g[o] to the very tissue of the structure. It comes in from the outside, not in subordination to the proceedings, and although every form may have been preserved opens the inquiry whether they have been more than an empty shell.” Frank v. Mangum, 237 U. S. 309, 346 (1915) (dissenting opinion). Even when the procedures authorizing detention are structurally sound, the Suspension Clause remains applicable and the writ relevant. See 2 Chambers, Course of Lectures on English Law 1767–1773, at 6 (“Liberty may be violated either by arbitrary imprisonment without law or the appearance of law, or by a lawful magistrate for an unlawful reason”). This is so, as Hayman and Swain make clear, even where the prisoner is detained after a criminal trial conducted in full accordance with the protections of the Bill of Rights. Were this not the case, there would have been no reason for the Court to inquire into the adequacy of substitute habeas procedures in Hayman and Swain. That the prisoners were detained pursuant to the most rigorous proceedings imaginable, a full criminal trial, would have been enough to render any habeas substitute acceptable per se.
Although we make no judgment as to whether the CSRTs, as currently constituted, satisfy due process standards, we agree with petitioners that, even when all the parties involved in this process act with diligence and in good faith, there is considerable risk of error in the tribunal’s findings of fact. This is a risk inherent in any process that, in the words of the former Chief Judge of the Court of Appeals, is “closed and accusatorial.” See Bismullah III, 514 F. 3d, at 1296 (Ginsburg, C. J., concurring in denial of rehearing en banc). And given that the consequence of error may be detention of persons for the duration of hostilities that may last a generation or more, this is a risk too significant to ignore.
For the writ of habeas corpus, or its substitute, to function as an effective and proper remedy in this context, the court that conducts the habeas proceeding must have the means to correct errors that occurred during the CSRT proceedings. This includes some authority to assess the sufficiency of the Government’s evidence against the detainee. It also must have the authority to admit and consider relevant exculpatory evidence that was not introduced during the earlier proceeding. Federal habeas petitioners long have had the means to supplement the record on review, even in the postconviction habeas setting. See Townsend v. Sain, 372 U. S. 293, 313 (1963), overruled in part by Keeney v. Tamayo-Reyes, 504 U. S. 1, 5 (1992). Here that opportunity is constitutionally required.
Consistent with the historic function and province of the writ, habeas corpus review may be more circumscribed if the underlying detention proceedings are more thorough than they were here. In two habeas cases involving enemy aliens tried for war crimes, In re Yamashita, 327 U. S. 1 (1946), and Ex parte Quirin, 317 U. S. 1 (1942), for example, this Court limited its review to determining whether the Executive had legal authority to try the petitioners by military commission. See Yamashita, supra, at 8 (“[O]n application for habeas corpus we are not concerned with the guilt or innocence of the petitioners. We consider here only the lawful power of the commission to try the petitioner for the offense charged”); Quirin, supra, at 25 (“We are not here concerned with any question of the guilt or innocence of petitioners”). Military courts are not courts of record. See Watkins, 3 Pet., at 209; Church 513. And the procedures used to try General Yamashita have been sharply criticized by Members of this Court. See Hamdan, 548 U. S., at 617; Yamashita, supra, at 41–81 (Rutledge, J., dissenting). We need not revisit these cases, however. For on their own terms, the proceedings in Yamashita and Quirin, like those in Eisentrager, had an adversarial structure that is lacking here. See Yamashita, supra, at 5 (noting that General Yamashita was represented by six military lawyers and that “[t]hroughout the proceedings … defense counsel … demonstrated their professional skill and resourcefulness and their proper zeal for the defense with which they were charged”); Quirin, supra, at 23–24; Exec. Order No. 9185, 7 Fed. Reg. 5103 (1942) (appointing counsel to represent the German saboteurs).
The extent of the showing required of the Government in these cases is a matter to be determined. We need not explore it further at this stage. We do hold that when the judicial power to issue habeas corpus properly is invoked the judicial officer must have adequate authority to make a determination in light of the relevant law and facts and to formulate and issue appropriate orders for relief, including, if necessary, an order directing the prisoner’s release.
We now consider whether the DTA allows the Court of Appeals to conduct a proceeding meeting these standards. “[W]e are obligated to construe the statute to avoid [constitutional] problems” if it is “ ‘fairly possible’ ” to do so. St. Cyr, 533 U. S., at 299–300 (quoting Crowell v. Benson, 285 U. S. 22, 62 (1932)). There are limits to this principle, however. The canon of constitutional avoidance does not supplant traditional modes of statutory interpretation. See Clark v. Martinez, 543 U. S. 371, 385 (2005) (“The canon of constitutional avoidance comes into play only when, after the application of ordinary textual analysis, the statute is found to be susceptible of more than one construction; and the canon functions as a means of choosing between them”). We cannot ignore the text and purpose of a statute in order to save it.
The DTA does not explicitly empower the Court of Appeals to order the applicant in a DTA review proceeding released should the court find that the standards and procedures used at his CSRT hearing were insufficient to justify detention. This is troubling. Yet, for present purposes, we can assume congressional silence permits a constitutionally required remedy. In that case it would be possible to hold that a remedy of release is impliedly provided for. The DTA might be read, furthermore, to allow the petitioners to assert most, if not all, of the legal claims they seek to advance, including their most basic claim: that the President has no authority under the AUMF to detain them indefinitely. (Whether the President has such authority turns on whether the AUMF authorizes—and the Constitution permits—the indefinite detention of “enemy combatants” as the Department of Defense defines that term. Thus a challenge to the President’s authority to detain is, in essence, a challenge to the Department’s definition of enemy combatant, a “standard” used by the CSRTs in petitioners’ cases.) At oral argument, the Solicitor General urged us to adopt both these constructions, if doing so would allow MCA §7 to remain intact. See Tr. of Oral Arg. 37, 53.
The absence of a release remedy and specific language allowing AUMF challenges are not the only constitutional infirmities from which the statute potentially suffers, however. The more difficult question is whether the DTA permits the Court of Appeals to make requisite findings of fact. The DTA enables petitioners to request “review” of their CSRT determination in the Court of Appeals, DTA §1005(e)(2)(B)(i), 119 Stat. 2742; but the “Scope of Review” provision confines the Court of Appeals’ role to reviewing whether the CSRT followed the “standards and procedures” issued by the Department of Defense and assessing whether those “standards and procedures” are lawful. §1005(e)(C), ibid. Among these standards is “the requirement that the conclusion of the Tribunal be supported by a preponderance of the evidence … allowing a rebuttable presumption in favor of the Government’s evidence.” §1005(e)(C)(i), ibid.
Assuming the DTA can be construed to allow the Court of Appeals to review or correct the CSRT’s factual determinations, as opposed to merely certifying that the tribunal applied the correct standard of proof, we see no way to construe the statute to allow what is also constitutionally required in this context: an opportunity for the detainee to present relevant exculpatory evidence that was not made part of the record in the earlier proceedings.
On its face the statute allows the Court of Appeals to consider no evidence outside the CSRT record. In the parallel litigation, however, the Court of Appeals determined that the DTA allows it to order the production of all “ ‘reasonably available information in the possession of the U. S. Government bearing on the issue of whether the detainee meets the criteria to be designated as an enemy combatant,’ ” regardless of whether this evidence was put before the CSRT. See Bismullah I, 501 F. 3d, at 180. The Government, see Pet. for Cert. pending in Gates v. Bismullah, No. 07–1054 (hereinafter Bismullah Pet.), with support from five members of the Court of Appeals, see Bismullah III, 514 F. 3d, at 1299 (Henderson, J., dissenting from denial of rehearing en banc); id., at 1302 (opinion of Randolph, J.) (same); id., at 1306 (opinion of Brown, J.) (same), disagrees with this interpretation. For present purposes, however, we can assume that the Court of Appeals was correct that the DTA allows introduction and consideration of relevant exculpatory evidence that was “reasonably available” to the Government at the time of the CSRT but not made part of the record. Even so, the DTA review proceeding falls short of being a constitutionally adequate substitute, for the detainee still would have no opportunity to present evidence discovered after the CSRT proceedings concluded.
Under the DTA the Court of Appeals has the power to review CSRT determinations by assessing the legality of standards and procedures. This implies the power to inquire into what happened at the CSRT hearing and, perhaps, to remedy certain deficiencies in that proceeding. But should the Court of Appeals determine that the CSRT followed appropriate and lawful standards and procedures, it will have reached the limits of its jurisdiction. There is no language in the DTA that can be construed to allow the Court of Appeals to admit and consider newly discovered evidence that could not have been made part of the CSRT record because it was unavailable to either the Government or the detainee when the CSRT made its findings. This evidence, however, may be critical to the detainee’s argument that he is not an enemy combatant and there is no cause to detain him.
This is not a remote hypothetical. One of the petitioners, Mohamed Nechla, requested at his CSRT hearing that the Government contact his employer. The petitioner claimed the employer would corroborate Nechla’s contention he had no affiliation with al Qaeda. Although the CSRT determined this testimony would be relevant, it also found the witness was not reasonably available to testify at the time of the hearing. Petitioner’s counsel, however, now represents the witness is available to be heard. See Brief for Boumediene Petitioners 5. If a detainee can present reasonably available evidence demonstrating there is no basis for his continued detention, he must have the opportunity to present this evidence to a habeas corpus court. Even under the Court of Appeals’ generous construction of the DTA, however, the evidence identified by Nechla would be inadmissible in a DTA review proceeding. The role of an Article III court in the exercise of its habeas corpus function cannot be circumscribed in this manner.
By foreclosing consideration of evidence not presented or reasonably available to the detainee at the CSRT proceedings, the DTA disadvantages the detainee by limiting the scope of collateral review to a record that may not be accurate or complete. In other contexts, e.g., in post-trial habeas cases where the prisoner already has had a full and fair opportunity to develop the factual predicate of his claims, similar limitations on the scope of habeas review may be appropriate. See Williams v. Taylor, 529 U. S. 420, 436–437 (2000) (noting that §2254 “does not equate prisoners who exercise diligence in pursuing their claims with those who do not”). In this context, however, where the underlying detention proceedings lack the necessary adversarial character, the detainee cannot be held responsible for all deficiencies in the record.
The Government does not make the alternative argument that the DTA allows for the introduction of previously unavailable exculpatory evidence on appeal. It does point out, however, that if a detainee obtains such evidence, he can request that the Deputy Secretary of Defense convene a new CSRT. See Supp. Brief for Respondents 4. Whatever the merits of this procedure, it is an insufficient replacement for the factual review these detainees are entitled to receive through habeas corpus. The Deputy Secretary’s determination whether to initiate new proceedings is wholly a discretionary one. See Dept. of Defense, Office for the Administrative Review of the Detention of Enemy Combatants, Instruction 5421.1, Procedure for Review of “New Evidence” Relating to Enemy Combatant (EC) Status ¶5(d) (May 7, 2007) (Instruction 5421.1) (“The decision to convene a CSRT to reconsider the basis of the detainee’s [enemy combatant] status in light of ‘new evidence’ is a matter vested in the unreviewable discretion of the [Deputy Secretary of Defense]”). And we see no way to construe the DTA to allow a detainee to challenge the Deputy Secretary’s decision not to open a new CSRT pursuant to Instruction 5421.1. Congress directed the Secretary of Defense to devise procedures for considering new evidence, see DTA §1005(a)(3), but the detainee has no mechanism for ensuring that those procedures are followed. DTA §1005(e)(2)(C), 119 Stat. 2742, makes clear that the Court of Appeals’ jurisdiction is “limited to consideration of … whether the status determination of the Combatant Status Review Tribunal with regard to such alien was consistent with the standards and procedures specified by the Secretary of Defense … and … whether the use of such standards and procedures to make the determination is consistent with the Constitution and laws of the United States.” DTA §1005(e)(2)(A), ibid., further narrows the Court of Appeals’ jurisdiction to reviewing “any final decision of a Combatant Status Review Tribunal that an alien is properly detained as an enemy combatant.” The Deputy Secretary’s determination whether to convene a new CSRT is not a “status determination of the Combatant Status Review Tribunal,” much less a “final decision” of that body.
We do not imply DTA review would be a constitutionally sufficient replacement for habeas corpus but for these limitations on the detainee’s ability to present exculpatory evidence. For even if it were possible, as a textual matter, to read into the statute each of the necessary procedures we have identified, we could not overlook the cumulative effect of our doing so. To hold that the detainees at Guantanamo may, under the DTA, challenge the President’s legal authority to detain them, contest the CSRT’s findings of fact, supplement the record on review with exculpatory evidence, and request an order of release would come close to reinstating the §2241 habeas corpus process Congress sought to deny them. The language of the statute, read in light of Congress’ reasons for enacting it, cannot bear this interpretation. Petitioners have met their burden of establishing that the DTA review process is, on its face, an inadequate substitute for habeas corpus.
Although we do not hold that an adequate substitute must duplicate §2241 in all respects, it suffices that the Government has not established that the detainees’ access to the statutory review provisions at issue is an adequate substitute for the writ of habeas corpus. MCA §7 thus effects an unconstitutional suspension of the writ. In view of our holding we need not discuss the reach of the writ with respect to claims of unlawful conditions of treatment or confinement.
In light of our conclusion that there is no jurisdictional bar to the District Court’s entertaining petitioners’ claims the question remains whether there are prudential barriers to habeas corpus review under these circumstances.
The Government argues petitioners must seek review of their CSRT determinations in the Court of Appeals before they can proceed with their habeas corpus actions in the District Court. As noted earlier, in other contexts and for prudential reasons this Court has required exhaustion of alternative remedies before a prisoner can seek federal habeas relief. Most of these cases were brought by prisoners in state custody, e.g., Ex parte Royall, 117 U. S. 241, and thus involved federalism concerns that are not relevant here. But we have extended this rule to require defendants in courts-martial to exhaust their military appeals before proceeding with a federal habeas corpus action. See Schlesinger, 420 U. S., at 758.
The real risks, the real threats, of terrorist attacks are constant and not likely soon to abate. The ways to disrupt our life and laws are so many and unforeseen that the Court should not attempt even some general catalogue of crises that might occur. Certain principles are apparent, however. Practical considerations and exigent circumstances inform the definition and reach of the law’s writs, including habeas corpus. The cases and our tradition reflect this precept.
In cases involving foreign citizens detained abroad by the Executive, it likely would be both an impractical and unprecedented extension of judicial power to assume that habeas corpus would be available at the moment the prisoner is taken into custody. If and when habeas corpus jurisdiction applies, as it does in these cases, then proper deference can be accorded to reasonable procedures for screening and initial detention under lawful and proper conditions of confinement and treatment for a reasonable period of time. Domestic exigencies, furthermore, might also impose such onerous burdens on the Government that here, too, the Judicial Branch would be required to devise sensible rules for staying habeas corpus proceedings until the Government can comply with its requirements in a responsible way. Cf. Ex parte Milligan, 4 Wall., at 127 (“If, in foreign invasion or civil war, the courts are actually closed, and it is impossible to administer criminal justice according to law, then, on the theatre of active military operations, where war really prevails, there is a necessity to furnish a substitute for the civil authority, thus overthrown, to preserve the safety of the army and society; and as no power is left but the military, it is allowed to govern by martial rule until the laws can have their free course”). Here, as is true with detainees apprehended abroad, a relevant consideration in determining the courts’ role is whether there are suitable alternative processes in place to protect against the arbitrary exercise of governmental power.
The cases before us, however, do not involve detainees who have been held for a short period of time while awaiting their CSRT determinations. Were that the case, or were it probable that the Court of Appeals could complete a prompt review of their applications, the case for requiring temporary abstention or exhaustion of alternative remedies would be much stronger. These qualifications no longer pertain here. In some of these cases six years have elapsed without the judicial oversight that habeas corpus or an adequate substitute demands. And there has been no showing that the Executive faces such onerous burdens that it cannot respond to habeas corpus actions. To require these detainees to complete DTA review before proceeding with their habeas corpus actions would be to require additional months, if not years, of delay. The first DTA review applications were filed over a year ago, but no decisions on the merits have been issued. While some delay in fashioning new procedures is unavoidable, the costs of delay can no longer be borne by those who are held in custody. The detainees in these cases are entitled to a prompt habeas corpus hearing.
Our decision today holds only that the petitioners before us are entitled to seek the writ; that the DTA review procedures are an inadequate substitute for habeas corpus; and that the petitioners in these cases need not exhaust the review procedures in the Court of Appeals before proceeding with their habeas actions in the District Court. The only law we identify as unconstitutional is MCA §7, 28 U. S. C. A. §2241(e) (Supp. 2007). Accordingly, both the DTA and the CSRT process remain intact. Our holding with regard to exhaustion should not be read to imply that a habeas court should intervene the moment an enemy combatant steps foot in a territory where the writ runs. The Executive is entitled to a reasonable period of time to determine a detainee’s status before a court entertains that detainee’s habeas corpus petition. The CSRT process is the mechanism Congress and the President set up to deal with these issues. Except in cases of undue delay, federal courts should refrain from entertaining an enemy combatant’s habeas corpus petition at least until after the Department, acting via the CSRT, has had a chance to review his status.
Although we hold that the DTA is not an adequate and effective substitute for habeas corpus, it does not follow that a habeas corpus court may disregard the dangers the detention in these cases was intended to prevent. Felker, Swain, and Hayman stand for the proposition that the Suspension Clause does not resist innovation in the field of habeas corpus. Certain accommodations can be made to reduce the burden habeas corpus proceedings will place on the military without impermissibly diluting the protections of the writ.
In the DTA Congress sought to consolidate review of petitioners’ claims in the Court of Appeals. Channeling future cases to one district court would no doubt reduce administrative burdens on the Government. This is a legitimate objective that might be advanced even without an amendment to §2241. If, in a future case, a detainee files a habeas petition in another judicial district in which a proper respondent can be served, see Rumsfeld v. Padilla, 542 U. S. 426, 435–436 (2004), the Government can move for change of venue to the court that will hear these petitioners’ cases, the United States District Court for the District of Columbia. See 28 U. S. C. §1404(a); Braden v. 30th Judicial Circuit Court of Ky., 410 U. S. 484, 499, n. 15 (1973).
Another of Congress’ reasons for vesting exclusive jurisdiction in the Court of Appeals, perhaps, was to avoid the widespread dissemination of classified information. The Government has raised similar concerns here and elsewhere. See Brief for Respondents 55–56; Bismullah Pet. 30. We make no attempt to anticipate all of the evidentiary and access-to-counsel issues that will arise during the course of the detainees’ habeas corpus proceedings. We recognize, however, that the Government has a legitimate interest in protecting sources and methods of intelligence gathering; and we expect that the District Court will use its discretion to accommodate this interest to the greatest extent possible. Cf. United States v. Reynolds, 345 U. S. 1, 10 (1953) (recognizing an evidentiary privilege in a civil damages case where “there is a reasonable danger that compulsion of the evidence will expose military matters which, in the interest of national security, should not be divulged”).
These and the other remaining questions are within the expertise and competence of the District Court to address in the first instance.
* * *
In considering both the procedural and substantive standards used to impose detention to prevent acts of terrorism, proper deference must be accorded to the political branches. See United States v. Curtiss-Wright Export Corp., 299 U. S. 304, 320 (1936). Unlike the President and some designated Members of Congress, neither the Members of this Court nor most federal judges begin the day with briefings that may describe new and serious threats to our Nation and its people. The law must accord the Executive substantial authority to apprehend and detain those who pose a real danger to our security.
Officials charged with daily operational responsibility for our security may consider a judicial discourse on the history of the Habeas Corpus Act of 1679 and like matters to be far removed from the Nation’s present, urgent concerns. Established legal doctrine, however, must be consulted for its teaching. Remote in time it may be; irrelevant to the present it is not. Security depends upon a sophisticated intelligence apparatus and the ability of our Armed Forces to act and to interdict. There are further considerations, however. Security subsists, too, in fidelity to freedom’s first principles. Chief among these are freedom from arbitrary and unlawful restraint and the personal liberty that is secured by adherence to the separation of powers. It is from these principles that the judicial authority to consider petitions for habeas corpus relief derives.
Our opinion does not undermine the Executive’s powers as Commander in Chief. On the contrary, the exercise of those powers is vindicated, not eroded, when confirmed by the Judicial Branch. Within the Constitution’s separation-of-powers structure, few exercises of judicial power are as legitimate or as necessary as the responsibility to hear challenges to the authority of the Executive to imprison a person. Some of these petitioners have been in custody for six years with no definitive judicial determination as to the legality of their detention. Their access to the writ is a necessity to determine the lawfulness of their status, even if, in the end, they do not obtain the relief they seek.
Because our Nation’s past military conflicts have been of limited duration, it has been possible to leave the outer boundaries of war powers undefined. If, as some fear, terrorism continues to pose dangerous threats to us for years to come, the Court might not have this luxury. This result is not inevitable, however. The political branches, consistent with their independent obligations to interpret and uphold the Constitution, can engage in a genuine debate about how best to preserve constitutional values while protecting the Nation from terrorism. Cf. Hamdan, 548 U. S., at 636 (Breyer, J., concurring) (“[J]udicial insistence upon that consultation does not weaken our Nation’s ability to deal with danger. To the contrary, that insistence strengthens the Nation’s ability to determine—through democratic means—how best to do so”).
It bears repeating that our opinion does not address the content of the law that governs petitioners’ detention. That is a matter yet to be determined. We hold that petitioners may invoke the fundamental procedural protections of habeas corpus. The laws and Constitution are designed to survive, and remain in force, in extraordinary times. Liberty and security can be reconciled; and in our system they are reconciled within the framework of the law. The Framers decided that habeas corpus, a right of first importance, must be a part of that framework, a part of that law.
The determination by the Court of Appeals that the Suspension Clause and its protections are inapplicable to petitioners was in error. The judgment of the Court of Appeals is reversed. The cases are remanded to the Court of Appeals with instructions that it remand the cases to the District Court for proceedings consistent with this opinion.
It is so ordered.