PART 3 OF 3 (Justice Stevens, Opinion of the Court)
1 An “enemy combatant” is defined by the military order as “an individual who was part of or supporting Taliban or al Qaeda forces, or associated forces that are engaged in hostilities against the United States or its coalition partners.” Memorandum from Deputy Secretary of Defense Paul Wolfowitz re: Order Establishing Combatant Status Review Tribunal §a (Jul. 7, 2004), available at http://www.defenselink.mil/news/Jul2004 ... review.pdf
(all Internet materials as visited June 26, 2006, and available in Clerk of Court’s case file).
2 The military order referenced in this section is discussed further in Parts III and VI, infra.
3 The penultimate subsections of §1005 emphasize that the provision does not “confer any constitutional right on an alien detained as an enemy combatant outside the United States” and that the “United States” does not, for purposes of §1005, include Guantanamo Bay. §§1005(f)–(g).
4 “ ‘And be it further enacted, That so much of the act approved February 5, 1867, entitled “An act to amend an act to establish the judicial courts of the United States, approved September 24, 1789,” as authorized an appeal from the judgment of the Circuit Court to the Supreme Court of the United States, or the exercise of any such jurisdiction by said Supreme Court, on appeals which have been, or may hereafter be taken, be, and the same is hereby repealed.’ ” 7 Wall., at 508.
5 See Hughes Aircraft Co. v. United States ex rel. Schumer, 520 U. S. 939, 951 (1997) (“The fact that courts often apply newly enacted jurisdiction-allocating statutes to pending cases merely evidences certain limited circumstances failing to meet the conditions for our generally applicable presumption against retroactivity …”).
6 Cf. Hughes Aircraft, 520 U. S., at 951 (“Statutes merely addressing which court shall have jurisdiction to entertain a particular cause of action can fairly be said merely to regulate the secondary conduct of litigation and not the underlying primary conduct of the parties” (emphasis in original)).
7 In his insistence to the contrary, Justice Scalia reads too much into Bruner v. United States, 343 U. S. 112 (1952) , Hallowell v. Commons, 239 U. S. 506 (1916) , and Insurance Co. v. Ritchie, 5Wall. 541 (1867). See post, at 2–4 (dissenting opinion). None of those cases says that the absence of an express provision reserving jurisdiction over pending cases trumps or renders irrelevant any other indications of congressional intent. Indeed, Bruner itself relied on such other indications–including a negative inference drawn from the statutory text, cf. infra, at 13–to support its conclusion that jurisdiction was not available. The Court observed that (1) Congress had been put on notice by prior lower court cases addressing the Tucker Act that it ought to specifically reserve jurisdiction over pending cases, see 343 U. S., at 115, and (2) in contrast to the congressional silence concerning reservation of jurisdiction, reservation had been made of “ ‘any rights or liabilities’ existing at the effective date of the Act” repealed by another provision of the Act, ibid., n. 7.
8 The question in Lindh was whether new limitations on the availability of habeas relief imposed by the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), 110 Stat. 1214, applied to habeas actions pending on the date of AEDPA’s enactment. We held that they did not. At the outset, we rejected the State’s argument that, in the absence of a clear congressional statement to the contrary, a “procedural” rule must apply to pending cases. 521 U. S., at 326.
9 That paragraph (1), along with paragraphs (2) and (3), is to “take effect on the date of enactment,” DTA §1005(h)(1), 119 Stat. 2743, is not dispositive; “a ‘statement that a statute will become effective on a certain date does not even arguably suggest that it has any application to conduct that occurred at an earlier date.’ ” INS v. St. Cyr, 533 U. S. 289, 317 (2001) (quoting Landgraf v. USI Film Products, 511 U. S. 244, 257 (1994) ). Certainly, the “effective date” provision cannot bear the weight Justice Scalia would place on it. See post, at 5, and n. 1. Congress deemed that provision insufficient, standing alone, to render subsections (e)(2) and (e)(3) applicable to pending cases; hence its adoption of subsection (h)(2). Justice Scalia seeks to avoid reducing subsection (h)(2) to a mere redundancy—a consequence he seems to acknowledge must otherwise follow from his interpretation—by speculating that Congress had special reasons, not also relevant to subsection (e)(1), to worry that subsections (e)(2) and (e)(3) would be ruled inapplicable to pending cases. As we explain infra, at 17, and n. 12, that attempt fails.
10 We note that statements made by Senators preceding passage of the Act lend further support to what the text of the DTA and its drafting history already make plain. Senator Levin, one of the sponsors of the final bill, objected to earlier versions of the Act’s “effective date” provision that would have made subsection (e)(1) applicable to pending cases. See, e.g., 151 Cong. Rec. S12667 (Nov. 10, 2005) (amendment proposed by Sen. Graham that would have rendered what is now subsection (e)(1) applicable to “any application or other action that is pending on or after the date of the enactment of this Act”). Senator Levin urged adoption of an alternative amendment that “would apply only to new habeas cases filed after the date of enactment.” Id., at S12802 (Nov. 15, 2005). That alternative amendment became the text of subsection (h)(2). (In light of the extensive discussion of the DTA’s effect on pending cases prior to passage of the Act, see, e.g., id., at S12664 (Nov. 10, 2005); id., at S12755 (Nov. 14, 2005); id., at S12799–S12802 (Nov. 15, 2005); id., at S14245, S14252–S14253, S14257–S14258, S14274–S14275 (Dec. 21, 2005), it cannot be said that the changes to subsection (h)(2) were inconsequential. Cf. post, at 14 (Scalia, J., dissenting).) While statements attributed to the final bill’s two other sponsors, Senators Graham and Kyl, arguably contradict Senator Levin’s contention that the final version of the Act preserved jurisdiction over pending habeas cases, see 151 Cong. Rec. S14263–S14264 (Dec. 21, 2005), those statements appear to have been inserted into the Congressional Record after the Senate debate. See Reply Brief for Petitioner 5, n. 6; see also 151 Cong. Rec. S14260 (statement of Sen. Kyl) (“I would like to say a few words about the now-completed National Defense Authorization Act for fiscal year 2006” (emphasis added)). All statements made during the debate itself support Senator Levin’s understanding that the final text of the DTA would not render subsection (e)(1) applicable to pending cases. See, e.g., id., at S14245, S14252–S14253, S14274–S14275 (Dec. 21, 2005). The statements that Justice Scalia cites as evidence to the contrary construe subsection (e)(3) to strip this Court of jurisdiction, see post, at 12, n. 4 (dissenting opinion) (quoting 151 Cong. Rec. S12796 (Nov. 15, 2005) (statement of Sen. Specter))—a construction that the Government has expressly disavowed in this litigation, see n. 11, infra. The inapposite November 14, 2005, statement of Senator Graham, which Justice Scalia cites as evidence of that Senator’s “assumption that pending cases are covered,” post, at 12, and n. 3 (citing 151 Cong. Rec. S12756 (Nov. 14, 2005)), follows directly after the uncontradicted statement of his co-sponsor, Senator Levin, assuring members of the Senate that “the amendment will not strip the courts of jurisdiction over [pending] cases.” Id., at S12755.
11 The District of Columbia Circuit’s jurisdiction, while “exclusive” in one sense, would not bar this Court’s review on appeal from a decision under the DTA. See Reply Brief in Support of Respondents’ Motion to Dismiss 16–17, n. 12 (“While the DTA does not expressly call for Supreme Court review of the District of Columbia Circuit’s decisions, Section 1005(e)(2) and (3) … do not remove this Court’s jurisdiction over such decisions under 28 U. S. C. §1254(1)”).
12 This assertion is itself highly questionable. The cases that Justice Scalia cites to support his distinction are Republic of Austria v. Altmann, 541 U. S. 677 (2004) , and Hughes Aircraft Co. v. United States ex rel. Schumer, 520 U. S. 939 (1997) . See post, at 8. While the Court in both of those cases recognized that statutes “creating” jurisdiction may have retroactive effect if they affect “substantive” rights, see Altmann, 541 U. S., at 695, and n. 15; Hughes Aircraft, 520 U. S., at 951, we have applied the same analysis to statutes that have jurisdiction-stripping effect, see Lindh v. Murphy, 521 U. S. 320, 327–328 (1997) ; id., at 342–343 (Rehnquist, C. J., dissenting) (construing AEDPA’s amendments as “ousting jurisdiction”).
13 See Landgraf, 511 U. S., at 271, n. 25 (observing that “the great majority of our decisions relying upon the antiretroactivity presumption have involved intervening statutes burdening private parties,” though “we have applied the presumption in cases involving new monetary obligations that fell only on the government” (emphasis added)); see also Altmann, 541 U. S., at 728–729 (Kennedy, J., dissenting) (explaining that if retroactivity concerns do not arise when a new monetary obligation is imposed on the United States it is because “Congress, by virtue of authoring the legislation, is itself fully capable of protecting the Federal Government from having its rights degraded by retroactive laws”).
14 There may be habeas cases that were pending in the lower courts at the time the DTA was enacted that do qualify as challenges to “final decision[s]” within the meaning of subsection (e)(2) or (e)(3). We express no view about whether the DTA would require transfer of such an action to the District of Columbia Circuit.
15 Because we conclude that §1005(e)(1) does not strip federal courts’ jurisdiction over cases pending on the date of the DTA’s enactment, we do not decide whether, if it were otherwise, this Court would nonetheless retain jurisdiction to hear Hamdan’s appeal. Cf. supra, at 10. Nor do we decide the manner in which the canon of constitutional avoidance should affect subsequent interpretation of the DTA. See, e.g., St. Cyr, 533 U. S., at 300 (a construction of a statute “that would entirely preclude review of a pure question of law by any court would give rise to substantial constitutional questions”).
16 Councilman distinguished service personnel from civilians, whose challenges to ongoing military proceedings are cognizable in federal court. See, e.g., United States ex rel. Toth v. Quarles, 350 U. S. 11 (1955) . As we explained in Councilman, abstention is not appropriate in cases in which individuals raise “ ‘substantial arguments denying the right of the military to try them at all,’ ” and in which the legal challenge “turn[s] on the status of the persons as to whom the military asserted its power.” 420 U. S., at 759 (quoting Noyd v. Bond, 395 U. S. 683 , n. 8 (1969)). In other words, we do not apply Councilman abstention when there is a substantial question whether a military tribunal has personal jurisdiction over the defendant. Because we conclude that abstention is inappropriate for a more basic reason, we need not consider whether the jurisdictional exception recognized in Councilman applies here.
17 See also Noyd, 395 U. S., at 694–696 (noting that the Court of Military Appeals consisted of “disinterested civilian judges,” and concluding that there was no reason for the Court to address an Air Force Captain’s argument that he was entitled to remain free from confinement pending appeal of his conviction by court-martial “when the highest military court stands ready to consider petitioner’s arguments”). Cf. Parisi v. Davidson, 405 U. S. 34, 41–43 (1972) (“Under accepted principles of comity, the court should stay its hand only if the relief the petitioner seeks … would also be available to him with reasonable promptness and certainty through the machinery of the military judicial system in its processing of the court-martial charge”).
18 If he chooses, the President may delegate this ultimate decisionmaking authority to the Secretary of Defense. See §6(H)(6).
19 Justice Scalia chides us for failing to include the District of Columbia Circuit’s review powers under the DTA in our description of the review mechanism erected by Commission Order No. 1. See post, at 22. Whether or not the limited review permitted under the DTA may be treated as akin to the plenary review exercised by the Court of Appeals for the Armed Forces, petitioner here is not afforded a right to such review. See infra, at 52; §1005(e)(3), 119 Stat. 2743.
20 Having correctly declined to abstain from addressing Hamdan’s challenge to the lawfulness of the military commission convened to try him, the Court of Appeals suggested that Councilman abstention nonetheless applied to bar its consideration of one of Hamdan’s arguments—namely, that his commission violated Article 3 of the Third Geneva Convention, 6 U. S. T. 3316, 3318. See Part VI, infra. Although the Court of Appeals rejected the Article 3 argument on the merits, it also stated that, because the challenge was not “jurisdictional,” it did not fall within the exception that Schlesinger v. Councilman, 420 U. S. 738 (1975) , recognized for defendants who raise substantial arguments that a military tribunal lacks personal jurisdiction over them. See 415 F. 3d, at 42. In reaching this conclusion, the Court of Appeals conflated two distinct inquiries: (1) whether Hamdan has raised a substantial argument that the military commission lacks authority to try him; and, more fundamentally, (2) whether the comity considerations underlying Councilman apply to trigger the abstention principle in the first place. As the Court of Appeals acknowledged at the beginning of its opinion, the first question warrants consideration only if the answer to the second is yes. See 415 F. 3d, at 36–37. Since, as the Court of Appeals properly concluded, the answer to the second question is in fact no, there is no need to consider any exception. At any rate, it appears that the exception would apply here. As discussed in Part VI, infra, Hamdan raises a substantial argument that, because the military commission that has been convened to try him is not a “ ‘regularly constituted court’ ” under the Geneva Conventions, it is ultra vires and thus lacks jurisdiction over him. Brief for Petitioner 5.
21 See also Winthrop 831 (“[I]n general, it is those provisions of the Constitution which empower Congress to ‘declare war’ and ‘raise armies,’ and which, in authorizing the initiation of war, authorize the employment of all necessary and proper agencies for its due prosecution, from which this tribunal derives its original sanction” (emphasis in original)).
22 Article 15 was first adopted as part of the Articles of War in 1916. See Act of Aug. 29, 1916, ch. 418, §3, Art. 15, 39 Stat. 652. When the Articles of War were codified and re-enacted as the UCMJ in 1950, Congress determined to retain Article 15 because it had been “construed by the Supreme Court (Ex Parte Quirin, 317 U. S. 1 (1942) ).” S. Rep. No. 486, 81st Cong., 1st Sess., 13 (1949).
23 Whether or not the President has independent power, absent congressional authorization, to convene military commissions, he may not disregard limitations that Congress has, in proper exercise of its own war powers, placed on his powers. See Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579, 637 (1952) (Jackson, J., concurring). The Government does not argue otherwise.
24 On this point, it is noteworthy that the Court in Ex parte Quirin, 317 U. S. 1 (1942) , looked beyond Congress’ declaration of war and accompanying authorization for use of force during World War II, and relied instead on Article of War 15 to find that Congress had authorized the use of military commissions in some circumstances. See id., at 26–29. Justice Thomas’ assertion that we commit “error” in reading Article 21 of the UCMJ to place limitations upon the President’s use of military commissions, see post, at 5 (dissenting opinion), ignores the reasoning in Quirin.
25 The justification for, and limitations on, these commissions were summarized in Milligan: “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. As necessity creates the rule, so it limits its duration; for, if this government is continued after the courts are reinstated, it is a gross usurpation of power. Martial rule can never exist where the courts are open, and in the proper and unobstructed exercise of their jurisdiction. It is also confined to the locality of actual war.” 4 Wall., at 127 (emphases in original).
26 The limitations on these occupied territory or military government commissions are tailored to the tribunals’ purpose and the exigencies that necessitate their use. They may be employed “pending the establishment of civil government,” Madsen, 343 U. S., at 354–355, which may in some cases extend beyond the “cessation of hostilities,” id., at 348.
27 So much may not be evident on cold review of the Civil War trials often cited as precedent for this kind of tribunal because the commissions established during that conflict operated as both martial law or military government tribunals and law-of-war commissions. Hence, “military commanders began the practice [during the Civil War] of using the same name, the same rules, and often the same tribunals” to try both ordinary crimes and war crimes. Bickers, 34 Tex. Tech. L. Rev., at 908. “For the first time, accused horse thieves and alleged saboteurs found themselves subject to trial by the same military commission.” Id., at 909. The Civil War precedents must therefore be considered with caution; as we recognized in Quirin, 317 U. S., at 29, and as further discussed below, commissions convened during time of war but under neither martial law nor military government may try only offenses against the law of war.
28 If the commission is established pursuant to martial law or military government, its jurisdiction extends to offenses committed within “the exercise of military government or martial law.” Winthrop 837.
29 Winthrop adds as a fifth, albeit not-always-complied-with, criterion that “the trial must be had within the theatre of war . . . ; that, if held elsewhere, and where the civil courts are open and available, the proceedings and sentence will be coram non judice.” Id., at 836. The Government does not assert that Guantanamo Bay is a theater of war, but instead suggests that neither Washington, D. C., in 1942 nor the Philippines in 1945 qualified as a “war zone” either. Brief for Respondents 27; cf. Quirin, 317 U. S. 1 ; In re Yamashita, 327 U. S. 1 (1946) .
30 The elements of this conspiracy charge have been defined not by Congress but by the President. See Military Commission Instruction No. 2, 32 CFR §11.6 (2005).
31 Justice Thomas would treat Osama bin Laden’s 1996 declaration of jihad against Americans as the inception of the war. See post, at 7–10 (dissenting opinion). But even the Government does not go so far; although the United States had for some time prior to the attacks of September 11, 2001, been aggressively pursuing al Qaeda, neither in the charging document nor in submissions before this Court has the Government asserted that the President’s war powers were activated prior to September 11, 2001. Cf. Brief for Respondents 25 (describing the events of September 11, 2001, as “an act of war” that “triggered a right to deploy military forces abroad to defend the United States by combating al Qaeda”). Justice Thomas’ further argument that the AUMF is “backward looking” and therefore authorizes trial by military commission of crimes that occurred prior to the inception of war is insupportable. See post, at 8, n. 3. If nothing else, Article 21 of the UCMJ requires that the President comply with the law of war in his use of military commissions. As explained in the text, the law of war permits trial only of offenses “committed within the period of the war.” Winthrop 837; see also Quirin, 317 U. S., at 28–29 (observing that law-of-war military commissions may be used to try “those enemies who in their attempt to thwart or impede our military effort have violated the law of war” (emphasis added)). The sources that Justice Thomas relies on to suggest otherwise simply do not support his position. Colonel Green’s short exegesis on military commissions cites Howland for the proposition that “[o]ffenses committed before a formal declaration of war or before the declaration of martial law may be tried by military commission.” The Military Commission, 42 Am. J. Int’l L. 832, 848 (1948) (emphases added) (cited post, at 9–10). Assuming that to be true, nothing in our analysis turns on the admitted absence of either a formal declaration of war or a declaration of martial law. Our focus instead is on the September 11, 2001 attacks that the Government characterizes as the relevant “act[s] of war,” and on the measure that authorized the President’s deployment of military force—the AUMF. Because we do not question the Government’s position that the war commenced with the events of September 11, 2001, the Prize Cases, 2Black 635 (1863) (cited post, at 2, 7, 8, and 10 (Thomas, J., dissenting)), are not germane to the analysis. Finally, Justice Thomas’ assertion that Julius Otto Kuehn’s trial by military commission “for conspiring with Japanese officials to betray the United States fleet to the Imperial Japanese Government prior to its attack on Pearl Harbor” stands as authoritative precedent for Hamdan’s trial by commission, post, at 9, misses the mark in three critical respects. First, Kuehn was tried for the federal espionage crimes under what were then 50 U. S C. §§31, 32, and 34, not with common-law violations of the law of war. See Hearings before the Joint Committee on the Investigation of the Pearl Harbor Attack, 79th Cong., 1st Sess., pt. 30, pp. 3067–3069 (1946). Second, he was tried by martial law commission (a kind of commission Justice Thomas acknowledges is not relevant to the analysis here, and whose jurisdiction extends to offenses committed within “the exercise of . . . martial law,” Winthrop 837, see supra, n. 28), not a commission established exclusively to try violations of the law of war. See ibid. Third, the martial law commissions established to try crimes in Hawaii were ultimately declared illegal by this Court. See Duncan v. Kahanamoku, 327 U. S. 304, 324 (1946) (“The phrase ‘martial law’ as employed in [the Hawaiian Organic Act], while intended to authorize the military to act vigorously for the maintenance of an orderly civil government and for the defense of the Islands against actual or threatened rebellion or invasion, was not intended to authorize the supplanting of courts by military tribunals”).
32 Justice Thomas adopts the remarkable view, not advocated by the Government, that the charging document in this case actually includes more than one charge: Conspiracy and several other ill-defined crimes, like “joining an organization” that has a criminal purpose, “ ‘[b]eing a guerilla,’ ” and aiding the enemy. See post, at 16–21, and n. 9. There are innumerable problems with this approach. First, the crimes Justice Thomas identifies were not actually charged. It is one thing to observe that charges before a military commission “ ‘need not be stated with the precision of a common law indictment,’ ” post, at 15, n. 7 (citation omitted); it is quite another to say that a crime not charged may nonetheless be read into an indictment. Second, the Government plainly had available to it the tools and the time it needed to charge petitioner with the various crimes Justice Thomas refers to, if it believed they were supported by the allegations. As Justice Thomas himself observes, see post, at 21, the crime of aiding the enemy may, in circumstances where the accused owes allegiance to the party whose enemy he is alleged to have aided, be triable by military commission pursuant to Article 104 of the UCMJ, 10 U. S. C. §904. Indeed, the Government has charged detainees under this provision when it has seen fit to do so. See Brief for David Hicks as Amicus Curiae 7. Third, the cases Justice Thomas relies on to show that Hamdan may be guilty of violations of the law of war not actually charged do not support his argument. Justice Thomas begins by blurring the distinction between those categories of “offender” who may be tried by military commission (e.g., jayhawkers and the like) with the “offenses” that may be so tried. Even when it comes to “ ‘being a guerilla,’ ” cf. post, at 18, n. 9 (citation omitted), a label alone does not render a person susceptible to execution or other criminal punishment; the charge of “ ‘being a guerilla’ ” invariably is accompanied by the allegation that the defendant “ ‘took up arms’ ” as such. This is because, as explained by Judge Advocate General Holt in a decision upholding the charge of “ ‘being a guerilla’ ” as one recognized by “the universal usage of the times,” the charge is simply shorthand (akin to “being a spy”) for “the perpetration of a succession of similar acts” of violence. Record Books of the Judge Advocate General Office, R. 3, 590. The sources cited by Justice Thomas confirm as much. See cases cited post, at 18, n. 9. Likewise, the suggestion that the Nuremberg precedents support Hamdan’s conviction for the (uncharged) crime of joining a criminal organization must fail. Cf. post, at 19–21. The convictions of certain high-level Nazi officials for “membership in a criminal organization” were secured pursuant to specific provisions of the Charter of the International Military Tribunal that permitted indictment of individual organization members following convictions of the organizations themselves. See Arts. 9 and 10, in 1 Trial of the Major War Criminals Before the International Military Tribunal 12 (1947). The initial plan to use organizations’ convictions as predicates for mass individual trials ultimately was abandoned. See T. Taylor, Anatomy of the Nuremberg Trials: A Personal Memoir 584–585, 638 (1992).
33 Cf. 10 U. S. C. §904 (making triable by military commission the crime of aiding the enemy); §906 (same for spying); War Crimes Act of 1996, 18 U. S. C. §2441 (2000 ed. and Supp. III) (listing war crimes); Foreign Operations, Export Financing, and Related Appropriations Act, 1998, §583, 111 Stat. 2436 (same).
34 While the common law necessarily is “evolutionary in nature,” post, at 13 (Thomas, J., dissenting), even in jurisdictions where common law crimes are still part of the penal framework, an act does not become a crime without its foundations having been firmly established in precedent. See, e.g., R. v. Rimmington,  2 All E. R. 257, 275–279 (House of Lords); id., at 279 (while “some degree of vagueness is inevitable and development of the law is a recognised feature of common law courts, … the law-making function of the courts must remain within reasonable limits”); see also Rogers v. Tennessee, 532 U. S. 451, 472–478 (2001) (Scalia, J., dissenting). The caution that must be exercised in the incremental development of common-law crimes by the judiciary is, for the reasons explained in the text, all the more critical when reviewing developments that stem from military action.
35 The 19th-century trial of the “Lincoln conspirators,” even if properly classified as a trial by law-of-war commission, cf. W. Rehnquist, All the Laws But One: Civil Liberties in Wartime 165–167 (1998) (analyzing the conspiracy charges in light of ordinary criminal law principles at the time), is at best an equivocal exception. Although the charge against the defendants in that case accused them of “combining, confederating, and conspiring together” to murder the President, they were also charged (as we read the indictment, cf. post, at 23, n. 14 (Thomas, J., dissenting)) with “maliciously, unlawfully, and traitorously murdering the said Abraham Lincoln.” H. R. Doc. No. 314, 55th Cong., 1st Sess., 696 (1899). Moreover, the Attorney General who wrote the opinion defending the trial by military commission treated the charge as if it alleged the substantive offense of assassination. See 11Op. Atty. Gen. 297 (1865) (analyzing the propriety of trying by military commission “the offence of having assassinated the President”); see also Mudd v. Caldera, 134 F. Supp. 2d 138, 140 (DC 2001).
36 By contrast, the Geneva Conventions do extend liability for substantive war crimes to those who “orde[r]” their commission, see Third Geneva Convention, Art. 129, 6 U. S. T., at 3418, and this Court has read the Fourth Hague Convention of 1907 to impose “command responsibility” on military commanders for acts of their subordinates, see Yamshita, 327 U. S., at 15–16.
37 The other examples Justice Thomas offers are no more availing. The Civil War indictment against Robert Louden, cited post, at 25, alleged a conspiracy, but not one in violation of the law of war. See War Dept., General Court Martial Order No. 41, p. 20 (1864). A separate charge of “ ‘[t]ransgression of the laws and customs of war’ ” made no mention of conspiracy. Id., at 17. The charge against Lenger Grenfel and others for conspiring to release rebel prisoners held in Chicago only supports the observation, made in the text, that the Civil War tribunals often charged hybrid crimes mixing elements of crimes ordinarily triable in civilian courts (like treason) and violations of the law of war. Judge Advocate General Holt, in recommending that Grenfel’s death sentence be upheld (it was in fact commuted by Presidential decree, see H. R. Doc. No. 314, at 725), explained that the accused “united himself with traitors and malefactors for the overthrow of our Republic in the interest of slavery.” Id., at 689.
38 The Court in Quirin “assume[d] that there are acts regarded in other countries, or by some writers on international law, as offenses against the law of war which would not be triable by military tribunal here, either because they are not recognized by our courts as violations of the law of war or because they are of that class of offenses constitutionally triable only by a jury.” 317 U. S., at 29. We need not test the validity of that assumption here because the international sources only corroborate the domestic ones.
39 Accordingly, the Tribunal determined to “disregard the charges … that the defendants conspired to commit War Crimes and Crimes against Humanity.” 22 Trial of the Major War Criminals Before the International Military Tribunal 469 (1947); see also ibid. (“[T]he Charter does not define as a separate crime any conspiracy except the one to commit acts of aggressive war”).
40 See also 15 United Nations War Crimes Commissions, Law Reports of Trials of War Criminals 90–91 (1949) (observing that, although a few individuals were charged with conspiracy under European domestic criminal codes following World War II, “the United States Military Tribunals” established at that time did not “recognis[e] as a separate offence conspiracy to commit war crimes or crimes against humanity”). The International Criminal Tribunal for the former Yugoslavia (ICTY), drawing on the Nuremberg precedents, has adopted a “joint criminal enterprise” theory of liability, but that is a species of liability for the substantive offense (akin to aiding and abetting), not a crime on its own. See Prosecutor v. Tadi, Judgment, Case No. IT–94–1–A (ICTY App. Chamber, July 15, 1999); see also Prosecutor v. Milutinovi, Decision on Dragoljub Ojdani’s Motion Challenging Jurisdiction—Joint Criminal Enterprise, Case No. IT–99–37–AR72, ¶26 (ICTY App. Chamber, May 21, 2003) (stating that “[c]riminal liability pursuant to a joint criminal enterprise is not a liability for … conspiring to commit crimes”).
41 Justice Thomas’ suggestion that our conclusion precludes the Government from bringing to justice those who conspire to commit acts of terrorism is therefore wide of the mark. See post, at 8, n. 3; 28–30. That conspiracy is not a violation of the law of war triable by military commission does not mean the Government may not, for example, prosecute by court-martial or in federal court those caught “plotting terrorist atrocities like the bombing of the Khobar Towers.” Post, at 29.
42 The accused also may be excluded from the proceedings if he “engages in disruptive conduct.” §5(K).
43 As the District Court observed, this section apparently permits reception of testimony from a confidential informant in circumstances where “Hamdan will not be permitted to hear the testimony, see the witness’s face, or learn his name. If the government has information developed by interrogation of witnesses in Afghanistan or elsewhere, it can offer such evidence in transcript form, or even as summaries of transcripts.” 344 F. Supp. 2d 152, 168 (DC 2004).
44 Any decision of the commission is not “final” until the President renders it so. See Commission Order No. 1 §6(H)(6).
45 See Winthrop 835, and n. 81 (“military commissions are constituted and composed, and their proceedings are conducted, similarly to general courts-martial”); id., at 841–842; S. Rep. No. 130, 64th Cong., 1st Sess., 40 (1916) (testimony of Gen. Crowder) (“Both classes of courts have the same procedure”); see also, e.g., H. Coppée, Field Manual of Courts-Martial, p. 104 (1863) (“[Military] commissions are appointed by the same authorities as those which may order courts-martial. They are constituted in a manner similar to such courts, and their proceedings are conducted in exactly the same way, as to form, examination of witnesses, etc.”).
46 The dissenters’ views are summarized in the following passage: “It is outside our basic scheme to condemn men without giving reasonable opportunity for preparing defense; in capital or other serious crimes to convict on ‘official documents …; affidavits; … documents or translations thereof; diaries …, photographs, motion picture films, and … newspapers” or on hearsay, once, twice or thrice removed, more particularly when the documentary evidence or some of it is prepared ex parte by the prosecuting authority and includes not only opinion but conclusions of guilt. Nor in such cases do we deny the rights of confrontation of witnesses and cross-examination.” Yamashita, 327 U. S., at 44 (footnotes omitted).
47 Article 2 of the UCMJ now reads: “(a) The following persons are subject to [the UCMJ]: “(9) Prisoners of war in custody of the armed forces. “(12) Subject to any treaty or agreement to which the United States is or may be a party or to any accepted rule of international law, persons within an area leased by or otherwise reserved or acquired for the use of the United States which is under the control of the Secretary concerned and which is outside the United States and outside the Commonwealth of Puerto Rico, Guam, and the Virgin Islands.” 10 U. S. C. §802(a). Guantanamo Bay is such a leased area. See Rasul v. Bush, 542 U. S. 466, 471 (2004) .
48 The International Committee of the Red Cross is referred to by name in several provisions of the 1949 Geneva Conventions and is the body that drafted and published the official commentary to the Conventions. Though not binding law, the commentary is, as the parties recognize, relevant in interpreting the Conventions’ provisions.
49 Aside from Articles 21 and 36, discussed at length in the text, the other seven Articles that expressly reference military commissions are: (1) 28 (requiring appointment of reporters and interpreters); (2) 47 (making it a crime to refuse to appear or testify “before a court-martial, military commission, court of inquiry, or any other military court or board”); (3) 48 (allowing a “court-martial, provost court, or military commission” to punish a person for contempt); (4) 49(d) (permitting admission into evidence of a “duly authenticated deposition taken upon reasonable notice to the other parties” only if “admissible under the rules of evidence” and only if the witness is otherwise unavailable); (5) 50 (permitting admission into evidence of records of courts of inquiry “if otherwise admissible under the rules of evidence,” and if certain other requirements are met); (6) 104 (providing that a person accused of aiding the enemy may be sentenced to death or other punishment by military commission or court-martial); and (7) 106 (mandating the death penalty for spies convicted before military commission or court-martial).
50 Justice Thomas relies on the legislative history of the UCMJ to argue that Congress’ adoption of Article 36(b) in the wake of World War II was “motivated” solely by a desire for “uniformity across the separate branches of the armed services.” Post, at 35. But even if Congress was concerned with ensuring uniformity across service branches, that does not mean it did not also intend to codify the longstanding practice of procedural parity between courts-martial and other military tribunals. Indeed, the suggestion that Congress did not intend uniformity across tribunal types is belied by the textual proximity of subsection (a) (which requires that the rules governing criminal trials in federal district courts apply, absent the President’s determination of impracticability, to courts-martial, provost courts, and military commissions alike) and subsection (b) (which imposes the uniformity requirement).
51 We may assume that such a determination would be entitled to a measure of deference. For the reasons given by Justice Kennedy, see post, at 5 (opinion concurring in part), however, the level of deference accorded to a determination made under subsection (b) presumably would not be as high as that accorded to a determination under subsection (a).
52 Justice Thomas looks not to the President’s official Article 36(a) determination, but instead to press statements made by the Secretary of Defense and the Under Secretary of Defense for Policy. See post, at 36–38 (dissenting opinion). We have not heretofore, in evaluating the legality of Executive action, deferred to comments made by such officials to the media. Moreover, the only additional reason the comments provide—aside from the general danger posed by international terrorism—for departures from court-martial procedures is the need to protect classified information. As we explain in the text, and as Justice Kennedy elaborates in his separate opinion, the structural and procedural defects of Hamdan’s commission extend far beyond rules preventing access to classified information.
53 Justice Thomas relies extensively on Madsen for the proposition that the President has free rein to set the procedures that govern military commissions. See post, at 30, 31, 33, n. 16, 34, and 45. That reliance is misplaced. Not only did Madsen not involve a law-of-war military commission, but (1) the petitioner there did not challenge the procedures used to try her, (2) the UCMJ, with its new Article 36(b), did not become effective until May 31, 1951, after the petitioner’s trial, see 343 U. S., at 345, n. 6, and (3) the procedures used to try the petitioner actually afforded more protection than those used in courts-martial, see id., at 358–360; see also id., at 358 (“[T]he Military Government Courts for Germany . . . have had a less military character than that of courts-martial”).
54 Prior to the enactment of Article 36(b), it may well have been the case that a deviation from the rules governing courts-martial would not have rendered the military commission “ ‘illegal.’ ” Post, at 30–31, n. 16 (Thomas, J., dissenting) (quoting Winthrop 841). Article 36(b), however, imposes a statutory command that must be heeded.
55 Justice Thomas makes the different argument that Hamdan’s Geneva Convention challenge is not yet “ripe” because he has yet to be sentenced. See post, at 43–45. This is really just a species of the abstention argument we have already rejected. See Part III, supra. The text of the Geneva Conventions does not direct an accused to wait until sentence is imposed to challenge the legality of the tribunal that is to try him.
56 As explained in Part VI–C, supra, that is no longer true under the 1949 Conventions.
57 But see, e.g., 4 Int’l Comm. of Red Cross, Commentary: Geneva Convention Relative to the Protection of Civilian Persons in Time of War 21 (1958) (hereinafter GCIV Commentary) (the 1949 Geneva Conventions were written “first and foremost to protect individuals, and not to serve State interests”); GCIII Commentary 91 (“It was not … until the Conventions of 1949 … that the existence of ‘rights’ conferred in prisoners of war was affirmed”).
58 But see generally Brief for Louis Henkin et al. as Amici Curiae; 1 Int’l Comm. for the Red Cross, Commentary: Geneva Convention for the Amelioration of the Condition of the Wounded and Sick in Armed Forces in the Field 84 (1952) (“It should be possible in States which are parties to the Convention … for the rules of the Convention to be evoked before an appropriate national court by the protected person who has suffered a violation”); GCII Commentary 92; GCIV Commentary 79.
59 For convenience’s sake, we use citations to the Third Geneva Convention only.
60 The President has stated that the conflict with the Taliban is a conflict to which the Geneva Conventions apply. See White House Memorandum, Humane Treatment of Taliban and al Qaeda Detainees 2(Feb. 7, 2002), available at http://www.justicescholars.org/pegc/arc ... 207_ed.pdf
(hereinafter White House Memorandum).
61 Hamdan observes that Article 5 of the Third Geneva Convention requires that if there be “any doubt” whether he is entitled to prisoner-of-war protections, he must be afforded those protections until his status is determined by a “competent tribunal.” 6 U. S. T., at 3324. See also Headquarters Depts. of Army, Navy, Air Force, and Marine Corps, Army Regulation 190–8, Enemy Prisoners of War, Retained Personnel, Civilian Internees and Other Detainees (1997), App. 116. Because we hold that Hamdan may not, in any event, be tried by the military commission the President has convened pursuant to the November 13 Order and Commission Order No. 1, the question whether his potential status as a prisoner of war independently renders illegal his trial by military commission may be reserved.
62 The term “Party” here has the broadest possible meaning; a Party need neither be a signatory of the Convention nor “even represent a legal entity capable of undertaking international obligations.” GCIII Commentary 37.
63 See also GCIII Commentary 35 (Common Article 3 “has the merit of being simple and clear… . Its observance does not depend upon preliminary discussions on the nature of the conflict”); GCIV Commentary 51 (“[N]obody in enemy hands can be outside the law”); U. S. Army Judge Advocate General’s Legal Center and School, Dept. of the Army, Law of War Handbook 144 (2004) (Common Article 3 “serves as a ‘minimum yardstick of protection in all conflicts, not just internal armed conflicts’ ” (quoting Nicaragua v. United States, 1986 I. C. J. 14, ¶218, 25 I. L. M. 1023)); Prosecutor v. Tadi, Case No. IT–94–1, Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction, ¶102 (ICTY App. Chamber, Oct. 2, 1995) (stating that “the character of the conflict is irrelevant” in deciding whether Common Article 3 applies).
64 The commentary’s assumption that the terms “properly constituted” and “regularly constituted” are interchangeable is beyond reproach; the French version of Article 66, which is equally authoritative, uses the term “régulirement constitués” in place of “properly constituted.”
65 Further evidence of this tribunal’s irregular constitution is the fact that its rules and procedures are subject to change midtrial, at the whim of the Executive. See Commission Order No. 1, §11 (providing that the Secretary of Defense may change the governing rules “from time to time”).
66 Other international instruments to which the United States is a signatory include the same basic protections set forth in Article 75. See, e.g., International Covenant on Civil and Political Rights, Art. 14, ¶3(d), Mar. 23, 1976, 999 U. N. T. S. 171 (setting forth the right of an accused “[t]o be tried in his presence, and to defend himself in person or through legal assistance of his own choosing”). Following World War II, several defendants were tried and convicted by military commission for violations of the law of war in their failure to afford captives fair trials before imposition and execution of sentence. In two such trials, the prosecutors argued that the defendants’ failure to apprise accused individuals of all evidence against them constituted violations of the law of war. See 5 U. N. War Crimes Commission 30 (trial of Sergeant-Major Shigeru Ohashi), 75 (trial of General Tanaka Hisakasu).
67 The Government offers no defense of these procedures other than to observe that the defendant may not be barred from access to evidence if such action would deprive him of a “full and fair trial.” Commission Order No. 1, §6(D)(5)(b). But the Government suggests no circumstances in which it would be “fair” to convict the accused based on evidence he has not seen or heard. Cf. Crawford v. Washington, 541 U. S. 36, 49 (2004) (“ ‘It is a rule of the common law, founded on natural justice, that no man shall be prejudiced by evidence which he had not the liberty to cross examine’ ” (quoting State v. Webb, 2 N. C. 103, 104 (Super. L. & Eq. 1794) (per curiam)); Diaz v. United States, 223 U. S. 442, 455 (1912) (describing the right to be present as “scarcely less important to the accused than the right of trial itself”); Lewis v. United States, 146 U. S. 370, 372 (1892) (exclusion of defendant from part of proceedings is “contrary to the dictates of humanity” (internal quotation marks omitted)); Joint Anti-Fascist Refugee Comm. v. McGrath, 341 U. S. 123 , n. 17, 171 (1951) (Frankfurter, J., concurring) (“[t]he plea that evidence of guilt must be secret is abhorrent to free men” (internal quotation marks omitted)). More fundamentally, the legality of a tribunal under Common Article 3 cannot be established by bare assurances that, whatever the character of the court or the procedures it follows, individual adjudicators will act fairly.