[3]*3Opinion for the Court filed by Circuit Judge TATEL.
Opinion dissenting in part filed by Circuit Judge BROWN.
TATEL, Circuit Judge:
In this case we have before us a petition for a writ of habeas corpus filed on behalf of Shawqi Ahmad Omar, an American citizen captured and detained in Iraq by United States military forces operating as part of the Multi-National Force-Iraq. Omar has been held under the control of United States forces for over two years, allegedly without legal process and with no meaningful access to counsel. When the district court learned of Omar’s imminent transfer to Iraqi authorities for trial on terrorism charges, it issued a preliminary injunction barring transfer in order to preserve its jurisdiction to entertain the habeas petition. The government appeals, arguing that the district court lacks jurisdiction to entertain the petition and that, in any event, it had no authority to enter the preliminary injunction because Omar’s transfer to Iraqi authorities would afford him all the relief he seeks, i.e., release from U.S. custody. For the reasons set forth in this opinion, we affirm.
I.
In late October 2004, United States military forces operating in Iraq arrested ap-pellee Shawqi Ahmad Omar, a dual American/Jordanian citizen, at his Baghdad home. Born in Kuwait, Omar became a naturalized American citizen following his marriage to the former Sandra Kay Sulzle. According to Omar, after the overthrow of the Saddam Hussein government, he traveled to Iraq seeking reconstruction-related work and would have left by November 2004 but for his arrest and detention.
The government paints a very different picture of Omar’s presence in Iraq. According to the government, U.S. military forces, operating in Iraq pursuant to U.N. Security Council Resolutions 1546 (2003) and 1637 (2004) as part of the Multi-Na-tional Force-Iraq (MNF-I), captured Omar during a raid on associates of Abu Musab al-Zarqawi. The government believes that Omar was part of Zarqawi’s network and that he facilitated terrorist activities both in and outside of Iraq. The government alleges that four Jordanian foreign fighters and an Iraqi insurgent were captured along with Omar, and that weapons and improvised explosive device-making materials were found in his home.
Following Omar’s arrest, an MNF-I panel of three American military officers conducted a hearing to resolve his status. According to the government, the process employed by the panel exceeded the requirements of Article 5 of the Third Geneva Convention. The record, however, reveals little about the panel’s operation. We know only that the panel permitted Omar to see the evidence against him, to make a statement, and to call “immediately available” witnesses. Declaration of John D. Gardner, Deputy Commanding General for Detainee Operations, MultiNational Force-Iraq, at 3-4 (Feb. 7, 2006), reprinted in Joint Appendix 138-39 (hereinafter “Gardner Deck”). After the hearing, the panel declared Omar to be a “security internee under the law of war” and an “ ‘enemy combatant’ in the war on terrorism.” Appellants’ Br. 9. The panel also found that Omar was not a prisoner of war for purposes of the Third Geneva Convention. Since the panel’s decision, American MNF-I officials have held Omar at various detention facilities in Iraq. According to Omar, the military has transferred him between Camp Cropper, the Abu Ghraib prison, and Camp Bucea. Omar has been in custody for over two years without for[4]*4mal charges and, he alleges, without access to counsel.
In August 2005, the MNF-I decided to refer Omar to the Central Criminal Court of Iraq (CCCI) for trial. The record indicates neither who made this decision nor what procedures were followed. The CCCI, a Baghdad-based Iraqi court, has national jurisdiction over an array of criminal offenses, including terrorism. According to the government, during the CCCI investigation and trial phases, the MNF-I maintains physical custody of detainees like Omar, turning them over to the Iraqi Ministry of Justice only after conviction.
On December 12, 2005, Omar’s wife, Sandra, and son, Ahmed, filed a petition for a writ of habeas corpus as Omar’s next friends. Brought in the United States District Court for the District of Columbia, the petition names as respondents Francis J. Harvey, Secretary of the Army; Major General William H. Brandenburg, then-Deputy Commanding General of Detainee Operations and Commanding General of Task Force 134, MNF-I; and Lieutenant Colonel Timothy Houser of the 105th Military Police Battalion, commanding officer at Camp Bucea. The petition asserts that Omar’s detention by the United States military violates numerous constitutional provisions, chief among them the right to due process guaranteed by the Fifth Amendment. The petition asks the district court to “[i]ssue a Writ of Habeas Corpus requiring Respondents to release Shawqi Omar from detention, and/or requiring Respondents to bring Shawqi Ahmad Omar before a court of competent jurisdiction in the United States to show just cause for his continued detention.” Habeas Pet. at 17. Also alleging that “the United States military may turn Mr. Omar over to the custody of Iraqi authorities in an effort to evade the strictures of United States law,” id. at 12-13, the petition asks the district court to “[e]njoin Respondents from transferring Mr. Omar to the authority of any other government, sovereign, country, or agency until [the district court] has an opportunity to consider and decide the merits of this Petition.” Id at 17.
Approximately two months after filing the petition, Omar’s attorney received an e-mail from the Department of Justice informing her of the MNF-I’s earlier decision to refer Omar to the CCCI. Believing that CCCI proceedings could interrupt American custody of Omar, thereby stripping the district court of jurisdiction, that transfer would amount to an illegal extradition, and that Omar would likely face torture by Iraqi authorities, the attorney sought and received an ex parte temporary restraining order requiring that Omar “not be removed from United States custody.” Order Granting the Ex Parte Motion for a TRO at 2.
In a memorandum filed shortly after entry of the TRO, the government challenged the district court’s jurisdiction to entertain the petition. The government relied principally on Hirota v. MacArthur, 338 U.S. 197, 69 S.Ct. 197, 93 L.Ed. 1902 (1948), in which the Supreme Court held that World War II Japanese officials could not invoke habeas to challenge their conviction by a multinational military tribunal. The government also argued that the district court had no authority to issue injunc-tive relief because doing so would “inject [the court] into an exclusive Executive function” and because adjudication of Omar’s potential referral to the CCCI “raises non-justiciable political questions.” Resp’ts’ Opp’n to Pet’rs’ Ex Parte Mot. for a TRO at 22, 25.
Following briefing, the district court converted the TRO into a preliminary injunction ordering that “the respondents ... and any persons acting in concert or participation with them, or having actual [5]*5or implicit knowledge of this Order ... shall not remove [Omar] from United States or MNF-I custody, or take any other action inconsistent with this court’s memorandum opinion.” Order Granting the Mot. for a Prelim. Inj. (hereinafter “Prelim. Inj. Order”). In the accompanying memorandum opinion, the court explained that the jurisdictional issues in the case presented questions “so serious, substantial, difficult and doubtful, as to make them fair ground for litigation and thus for more deliberative investigation.” Omar v. Harvey, 416 F.Supp.2d 19, 23-24 (D.D.C.2006) (internal quotation marks omitted) (quoting Wash. Metro. Area Transit Comm’n v. Holiday Tours, Inc., 559 F.2d 841, 844 (D.C.Cir.1977)). Fearing imminent referral to the CCCI would forever preclude a more deliberative investigation of the weighty jurisdictional questions, the court issued the injunction to freeze the status quo. In doing so, the court credited Omar’s contention that transfer could irreparably deprive him of this investigation by “undoing the] court’s jurisdiction.” Id. at 28. This, the court concluded, “would abuse the process now put in place for the purpose of adjudicating matters on their merits.” Id.
The government appeals, arguing (as it did in the district court) that Hirota controls and that Omar’s challenge presents non-justiciable political questions. The government also argues that even if the district court does have jurisdiction, its injunction was improper because, by prohibiting Omar’s removal from American or MNF-I custody, it bars the government from providing Omar “all of the relief to which he is entitled through a writ of habeas corpus.” Appellants’ Br. 20. We consider each issue in turn.
II.
The “great writ” of habeas corpus, as Blackstone called it, has for centuries functioned as the “symbol and guardian of individual liberty.” Peyton v. Rowe, 391 U.S. 54, 58, 88 S.Ct. 1549, 20 L.Ed.2d 426 (1968). By the seventeenth century, the writ had become “the highest remedy in law, for any man that is imprisoned.” WILLIAM F. DUKER, A CONSTITUTIONAL HISTORY OF HABEAS CORPUS 46-49 (1980) (internal quotation marks and citation omitted). In the eighteenth century, it was the only common law writ expressly mentioned in the United States Constitution. See U.S. CONST, art. I, § 9, cl. 2; Hamdi v. Rumsfeld, 542 U.S. 507, 558, 124 S.Ct. 2633, 159 L.Ed.2d 578 (2004) (Scalia, J., dissenting). And now, in the twenty-first century, the writ continues to protect fundamental rights as the United States confronts the challenge of international terrorism. Indeed, since September 11, the Supreme Court has considered habeas petitions filed on behalf of at least three accused terrorists, confirming “the federal courts’ power to review applications for habeas relief in a wide variety of cases involving executive detention, in wartime as well as in times of peace.” Rasul v. Bush, 542 U.S. 466, 474, 124 S.Ct. 2686, 159 L.Ed.2d 548 (2004); see also Hamdan v. Rumsfeld, 548 U.S. -, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006) (entertaining habeas petition of alien detained at Guantanamo Bay, Cuba); Hamdi, 542 U.S. 507, 124 S.Ct. 2633, 159 L.Ed.2d 578 (plurality opinion) (entertaining habeas petition of American citizen detained within the United States).
Notwithstanding the writ’s long and celebrated history, the government argues that the district court lacks jurisdiction to consider Omar’s petition. Although acknowledging both that U.S. military officials are holding Omar, Appellants’ Br. 15, and that those officials operate “subject to” no independent MNF-I authority, Oral [6]*6Arg. Tr. 11, the government contends that federal courts lack jurisdiction to entertain habeas corpus petitions filed by individuals detained by American military officials operating as part of a multinational force. In support, the government relies primarily on Hirota, in which the Supreme Court ruled that the courts of the United States had no jurisdiction to entertain habeas petitions filed by Japanese citizens convicted and sentenced by an American-led international military tribunal in Japan “set up by General MacArthur as the agent of the Allied Powers.” 388 U.S. at 198, 69 S.Ct. 197. From Hirota, the government draws the general principle that federal courts lack habeas jurisdiction over individuals held by “United States military personnel under the auspices of a multinational force that is distinct from the United States military and ultimately derives its existence from an international body.” Appellants’ Br. 32. Applying that principle to this case and pointing out that Omar, like the Hirota petitioners, is in the custody of American officials acting as part of a multinational force, the government argues that Hirota “compels the dismissal of this habeas petition.” Appellants’ Br. 25.
Omar disagrees. Pointing out that the Hirota petitioners filed directly in the Supreme Court, he argues that “Hirota’s holding concerns the scope of Supreme Court jurisdiction under Article III of the Constitution.” Appellees’ Br. 29. Yet just six months after Hirota, in Flick v. Johnson, 174 F.2d 983 (D.C.Cir.1949), we applied Hirota to a habeas corpus petition filed not in the Supreme Court, but in the district court by an individual who, like the Hirota petitioners, had been convicted by an international tribunal. In this circuit, then, Hirota applies to habeas proceedings in the district court.
Omar also argues that Hirota “lacks vitality today” given that the Supreme Court has since “clarified the broad availability of habeas corpus.” Appellees’ Br. 23-24. In support, Omar cites Madsen v. Kinsella, 343 U.S. 341, 72 S.Ct. 699, 96 L.Ed. 988 (1952), and United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955). In both of those cases, however, the habeas petitioners had been convicted not by multinational tribunals, but rather by American tribunals sitting in foreign countries, i.e., Germany (Madsen) and Korea (Toth). Thus, neither case has anything to do with the issue the Supreme Court faced in Hirota. Omar also relies on Hamdi, 542 U.S. 507, 124 S.Ct. 2633, 159 L.Ed.2d 578, and Rasul, 542 U.S. 466, 124 S.Ct. 2686, 159 L.Ed.2d 548. Although we share Omar’s view that these decisions provide a basis for questioning Hirota’s vitality, the Supreme Court has never revisited the precise issue it confronted there, namely the availability of habeas to non-citizens convicted abroad by multinational tribunals. As the Supreme Court has cautioned, “[i]f a precedent of [the Supreme] Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the Court of Appeals should follow the case which directly controls, leaving to [the Supreme] Court the prerogative of overruling its own decisions.” Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U.S. 477, 484, 109 S.Ct. 1917, 104 L.Ed.2d 526 (1989).
That said — and setting aside Omar’s suggestion that Hirota is distinguishable because the MNF-I may not be as authentically multinational as the Allied Forces in Japan — we agree with Omar that Hirota is not nearly as broad as the government insists. A nine-sentence per curiam opinion, Hirota reads — in its entirety — as follows:
The petitioners, all residents and citizens of Japan, are being held in custody [7]*7pursuant to the judgment of a military tribunal in Japan. Two of the petitioners have been sentenced to death, the others to terms of imprisonment. They filed motions in this Court for leave to file petitions for habeas corpus. We set all the motions for hearing on the question of our power to grant the relief prayed and that issue has now been fully presented and argued.
We are satisfied that the tribunal sentencing these petitioners is not a tribunal of the United States. The United States and other allied countries conquered and now occupy and control Japan. General Douglas MacArthur has been selected and is acting as the Supreme Commander for the Allied Powers. The military tribunal sentencing these petitioners has been set up by General MacArthur as the agent of the Allied Powers.
Under the foregoing circumstances the courts of the United States have no power or authority to review, to affirm, set aside or annul the judgments and sentences imposed on these petitioners and for this reason the motions for leave to file petitions for writs of habeas corpus are denied.
338 U.S. at 198, 69 S.Ct. 197. As is apparent, Hirota nowhere explains which “circumstances” were controlling. Nor does anything in the opinion hold that federal courts lack habeas jurisdiction whenever, as the government insists, American officials detaining a petitioner are functioning as part of a multinational force. Indeed, the opinion articulates no general legal principle at all. The Court, moreover, has never cited Hirota for any substantive proposition, much less the one the government claims it supports. None of this should be surprising given that the Court heard Hirota not on a petition for certiora-ri granted to resolve an important question of law, see SUP. CT. R. 38(5) (1939) (“A review on writ of certiorari ... will be granted only where there are special and important reasons therefor.”), but rather as an original petition for habeas corpus. This, together with the terse per curiam opinion, reveals a Court determined to resolve the case on the narrowest possible grounds.
We thus take the Court at its word: it lacked habeas jurisdiction because of the “circumstances” of the case. As a matter of precedent, then, Hirota would “control” this case only if the “circumstances” significant to the Court’s decision are present here. Two circumstances are clearly the same: detention overseas and the existence of a multinational force. But two other circumstances — foreign citizenship and criminal conviction — are absent. Were we writing on a clean slate, we would thus have to determine which of these circumstances influenced the Court’s decision.
But our slate is not clean. In Flick, we considered a habeas petition filed by a German citizen held by American troops in Germany pursuant to his conviction by an entity known as the Military Tribunal IV. 174 F.2d 983. Given Hirota, we asked: ‘Was the court which tried and sentenced Flick a tribunal of the United States?” Id. at 984. “If it was not,” we explained, then under Hirota, “no court of this country has power or authority to review, affirm, set aside or annul the judgment and sentence imposed on Flick.” Id. Concluding that the Military Tribunal IV was not, in fact, an American tribunal, we dismissed the petition.
Flick thus holds that the critical factor in Hirota was the petitioners’ convictions by an international tribunal, and for good reason. Throughout its brief opinion, the Hirota Court repeatedly referred to the petitioners’ sentences, including imprison[8]*8ment and death, concluding it lacked habe-as jurisdiction “to review, to affirm, set aside or annul the judgments and sentences imposed on these petitioners.” Hirota, 338 U.S. at 198, 69 S.Ct. 197. Such language demonstrates that the Court’s primary concern was that the petitions represented a collateral attack on the final judgment of an international tribunal.
Viewed in this light, Hirota does not control this case. Unlike the Hirota and Flick petitioners, Omar has not been charged with a crime related to the allegations now lodged against him, much less convicted of one. Omar seeks not to collaterally attack a final international conviction, but only to test the lawfulness of his extrajudicial detention in Iraq, where he has remained in the control of U.S. forces for over two years without legal process. True, a panel of three military officers found him to be a “security internee” and an “enemy combatant,” but those determinations, based as they are on military considerations, are a far cry from trial, judgment, and sentencing. See Hamdi, 542 U.S. at 518-19, 124 S.Ct. 2633 (discussing enemy combatant status); Major General George R. Fay, AR 15-6 Investigation of the Abu Ghraib Detention Facility and 205th Military Intelligence Brigade 12 (Aug. 23, 2004), available at http://www. defenselink.mil/news/Aug2004/d20040825 fay.pdf (describing military definition of security internees as “[civilians interned during conflict or occupation for their own protection or because they pose a threat to the security of coalition forces, or its mission, or are of intelligence value”). Habe-as proceedings here run no risk, as they did in both Hirota and Flick, of judicial second-guessing of an international tribunal’s final determination of guilt.
The fact that Omar has never been convicted of criminal activity thus distinguishes this case from both Hirota and Flick, and rightly so, given that challenging extrajudicial detention is among the most fundamental purposes of habeas. “At its historical core,” the Supreme Court has explained, “the writ of habeas corpus has served as a means of reviewing the legality of Executive detention, and it is in that context that its protections have been strongest.” INS v. St. Cyr, 533 U.S. 289, 301, 121 S.Ct. 2271, 150 L.Ed.2d 347 (2001); see also Brown v. Allen, 344 U.S. 443, 533, 73 S.Ct. 397, 97 L.Ed. 469 (1953) (Jackson, J., concurring in the judgment) (“The historic purpose of the writ has been to relieve detention by executive authorities without judicial trial”). Acting in tandem with its partners-in-liberty—the Due Process Clauses of the Fifth and Fourteenth Amendments—the great writ is “the instrument by which due process [can] be insisted upon by a citizen illegally imprisoned.” Hamdi, 542 U.S. at 555-56, 124 S.Ct. 2633 (Scalia, J., dissenting). Where, as in Hirota and Flick, individuals have been convicted and sentenced by a criminal tribunal, some form of judicial process has occurred, reducing the risk of unlawful extrajudicial detention. But where, as here, the Executive detains an individual without trial, the risk of unlawful incarceration is at its apex.
In addition to Hirota, the government cites cases holding that federal courts may not grant habeas relief to Americans held by foreign governments, see, e.g., United States ex rel. Keefe v. Dulles, 222 F.2d 390 (D.C.Cir.1954), and that habeas is unavailable to Americans held in U.S. custody pursuant to a foreign conviction, see, e.g., Bishop v. Reno, 210 F.3d 1295 (11th Cir.2000). Such propositions, however, have nothing to do with this case since Omar is neither detained nor convicted by a foreign nation.
[9]*9With Hirota and the other cases the government cites thus distinguished, Omar’s petition fits comfortably within the terms of the modern habeas statute — a proposition the government nowhere contests. Under 28 U.S.C. § 2241, federal courts have authority to issue the writ “within their respective jurisdictions” to prisoners “in custody under or by color of the authority of the United States.” 28 U.S.C. § 2241(a), (c)(1). Omar’s petition satisfies both requirements. First, the petition is “within the jurisdiction” of the district court because respondents, the Secretary of the Army and two high-ranking Army officers, are amenable to service in the District of Columbia. See Rasul, 542 U.S. at 478-79, 124 S.Ct. 2686 (“[A] district court acts within [its] respective jurisdiction within the meaning of § 2241 as long as the custodian can be reached by service of process.” (second alteration in original) (internal quotation marks omitted)). Second, although American personnel in Iraq operate as part of the MNF-I, the government concedes that Omar is “held” by U.S. forces, Appellants’ Br. 15, and that those forces operate “subject to” no independent MNF-I authority, Oral Arg. Tr. 11. Omar is thus “in custody under or by color of the authority of the United States.” As a consequence, the district court has jurisdiction to entertain Omar’s habeas petition.
III.
The government next argues that the district court lacked jurisdiction to enter the preliminary injunction because this case “raise[s] quintessential political questions beyond the authority or competence of the judiciary to answer.” Appellants’ Br. 41. The political question doctrine puts beyond judicial cognizance “political decisions that are by their nature committed to the political branches.” Schneider v. Kissinger, 412 F.3d 190, 193 (D.C.Cir.2005) (internal quotation mark omitted). For example, and relevant to this case, the doctrine bars courts from considering claims whose adjudication would require judicial wading into foreign policy or military waters. Thus, in Schneider we invoked the political question doctrine to dismiss a claim that would have required us to second-guess U.S. policy towards Chile. Id. at 197-98. Similarly, in Bancoult v. McNamara, 445 F.3d 427, 436-37 (D.C.Cir.2006), we dismissed a complaint that would have required us to review the manner in which the United States established a military base in the Indian Ocean. Here, the government argues that Omar’s petition would likewise require the court to interfere with the “Executive’s textual constitutional authority to implement foreign policy and military functions for the purpose of protecting national security.” Appellants’ Br. 43.
As the political question doctrine is one “of ‘political questions,’ not one of ‘political cases,’” Baker v. Carr, 369 U.S. 186, 217, 82 S.Ct. 691, 7 L.Ed.2d 663 (1962), we must focus on Omar’s specific claims. First, he challenges his detention, claiming that U.S. military officials are holding him in violation of the Constitution, federal law, Army regulations, and international law. Critically for present purposes, Omar alleges that he is held in violation of the Due Process Clause of the Fifth Amendment because his “arrest and arbitrary, indefinite detention without process ... violates ... [his] ‘interest in being free from physical detention by one’s own government.’ ” Habeas Pet. at 13 (quoting Hamdi, 542 U.S. at 529, 124 S.Ct. 2633). Second, he challenges his transfer, arguing both that the military lacks treaty or statutory authorization to transfer him to Iraqi authorities and that the U.S. Constitution forbids transfer to a government likely to torture him.
[10]*10The Supreme Court’s recent decision in Hamdi makes abundantly clear that Omar’s challenge to his detention is justiciable. In Hamdi, as here, the petitioner challenged his detention by U.S. military authorities pursuant to an “enemy combatant” determination. Although the government never directly invoked the political question doctrine, it argued that separation of powers concerns — the very concerns underlying the political question doctrine — preclude courts from inquiring into the factual basis of an enemy combatant designation. “A commander’s wartime determination that an individual is an enemy combatant,” the government urged, “is a quintessentially military judgment representing a core exercise of the Commander-in-Chief authority.” Br. for the Resp’ts at 25, Hamdi v. Rumsfeld, 542 U.S. 507, 124 S.Ct. 2633, 159 L.Ed.2d 578 (2004) (No. 03-6696). Unequivocally rejecting this contention, the Hamdi plurality explained that “it does not infringe on the core role of the military for the courts to exercise their own time-honored and constitutionally mandated roles of reviewing and resolving claims like those presented here.” Hamdi, 542 U.S. at 535, 124 S.Ct. 2633.
Omar’s challenge to his transfer is equally justiciable. He argues (1) that the military may not transfer him to Iraqi authorities without treaty or statutory authorization, and (2) that the military lacks such authorization. In the extradition context, of course, treaty or statutory authorization has long been required, Valentine v. United States ex rel. Neidecker, 299 U.S. 5, 8, 57 S.Ct. 100, 81 L.Ed. 5 (1936) (“[A]lbeit a national power, [authority to extradite] is not confided to the Executive in the absence of treaty or legislative provision”), and we have some reason to believe this rule applies beyond the extradition context, see Wilson v. Girard, 354 U.S. 524, 528-30, 77 S.Ct. 1409, 1 L.Ed.2d 1544 (1957) (per curiam) (permitting in-country transfer of American service member to Japanese custody after noting that an agreement authorized by a treaty provided for transfer).
Our decisions in Holmes v. Laird, 459 F.2d 1211 (D.C.Cir.1972), and in countless other cases make clear that courts may determine whether the Executive possesses the necessary authority for transfer— the second of the two questions Omar raises. In Holmes, American soldiers convicted by a German court of attempted rape escaped U.S. military custody, returned to the United States, and then sought to prevent the military from transferring them to German custody. Id. at 1214. In support, they argued (among other things) that they were “surrenderable only pursuant to the terms of an extradition treaty.” Id. at 1219 n. 59. Although ultimately rejecting the petitioners’ authorization claim, we first satisfied ourselves that a valid treaty in fact authorized the transfers. Id.; see also Neidecker, 299 U.S. at 18, 57 S.Ct. 100 (denying extradition after determining that “the treaty with France fails to grant the necessary authority”).
The antecedent question — whether Omar’s transfer even requires treaty or statutory authorization — is also fully justi-ciable. On the merits, the government will surely argue that under Article II of the Constitution, the military needs no express authority to transfer detainees like Omar. Resolving this claim will involve difficult questions of constitutional law — questions which, significantly for our purposes, will require no judicial intrusion into the exclusive domain of the political branches. To be sure, a decision on the merits might well have implications for military and foreign policy, but that alone hardly makes the issue non-justiciable. For example, in Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 72 S.Ct. 863, 96 L.Ed. 1153 [11]*11(1952), the government’s assertion that wartime seizure of steel mills was “necessary to avert a national catastrophe which would inevitably result from a stoppage of steel production,” id. at 582, 72 S.Ct. 863, failed to prevent the Supreme Court from deciding whether the President may seize private property without congressional authorization. Likewise, in United States v. Curtiss-Wright Export Corp., 299 U.S. 304, 57 S.Ct. 216, 81 L.Ed. 255 (1936), the Court, despite the obvious implications for U.S. foreign policy, decided whether the President could constitutionally issue export controls on munitions absent specific statutory authorization. In each of these cases, the Court resolved a fundamental question of Executive authority without making any foreign policy or military judgments of its own. This case, like Youngstown and Curtiss-Wright and unlike Schneider and Bancoult, supra, presents constitutional issues that courts can resolve without making any judgments about foreign policy or the war in Iraq.
Finally, the “rule of non-inquiry,” which bars courts from “investigating the fairness of a requesting nation’s justice system,” In re Extradition of Howard, 996 F.2d 1320, 1329 (1st Cir.1993), does not require a different result. According to the government, Omar’s allegation that he faces torture at the hands of the Iraqis is precisely the type of claim the rule of non-inquiry bars courts from considering. This may be correct. See, e.g., In re Extradition of Manzi, 888 F.2d 204, 205-06 (1st Cir.1989) (permitting extradition to Italy and refusing to allow evidence on the petitioner’s claim that his life would be threatened by the transfer). But see Gallina v. Fraser, 278 F.2d 77, 79 (2d Cir.1960) (suggesting in dicta that there may one day arise a transfer so “antipathetic to a federal court’s sense of decency as to require reexamination of’ the rule of non-inquiry). But since the only question before us at this stage of the litigation relates to the district court’s jurisdiction, and given our earlier conclusion that the political question doctrine presents no jurisdictional bar to Omar’s challenge to his detention and transfer, we need not address his torture claims. The rule of non-inquiry therefore has no relevance to our disposition of the matter before us at this stage of the litigation.
IV.
Having established that the district court has jurisdiction, we turn to the propriety of the preliminary injunction. Applying the standard four-factor analysis, see Mova Pharm. Corp. v. Shalala, 140 F.3d 1060, 1066 (D.C.Cir.1998) (listing requirements for preliminary injunctions), the district court enjoined Omar’s removal from American or MNF-I custody in order to preserve its jurisdiction to address the case’s “serious, substantial, difficult, and doubtful” issues. 416 F.Supp.2d at 28 (internal quotation marks omitted) (quoting Holiday Tours, 559 F.2d at 844).
The government challenges the injunction, claiming that the district court improperly barred Omar’s outright release even though such release is precisely what his “petition ultimately seeks.” Appellants’ Br. 57. The injunction, in relevant part, orders that the respondents “shall not remove the petitioner from United States or MNF-I custody.” Prelim. Inj. Order (emphasis added). Although “remove” could include outright release, given the circumstances of this case, we interpret the word differently. Omar did not seek an injunction barring his outright release, nor could he have; he sought an injunction prohibiting his transfer to Iraqi authorities in order to preserve the district court’s jurisdiction to entertain his habeas petition. We thus understand the court to [12]*12have used the word “remove” to prevent Omar’s transfer in any form, whether by an official handoff or otherwise. Viewed this way, the injunction does not bar a bona fide release of Omar, even if the military releases him inside Iraq.
The government’s primary challenge to the injunction (other than the jurisdictional arguments we have rejected in parts II and III), is its claim that transfer to Iraqi authorities constitutes release from American/MNF-I custody — “all of the relief to which [Omar] is entitled through a writ of habeas corpus.” Appellants’ Br. 20. The government asserts that transfer is release because transfer “would end his detention by the MNF-I, which is the sole arguable basis for the district court’s jurisdiction.” Appellants’ Br. 58. In other words, the government sees transfer as a subset of release; one can be “released” either by being let go into the open, or by being-transferred to a different authority.
There is, however, an obvious and quite significant difference between transferring Omar to Iraqi authorities and releasing him to walk free from his current detention. If transferred, Omar would remain in custody and detention; if released, he might not. Indeed, were the government correct, federal courts would have no authority to stay an extradition long enough to test its validity since transfer to the foreign authority would, as the government sees it, be the same as release. Yet courts routinely stay extraditions, see, e.g., Ntakirutimana v. Reno, 184 F.3d 419, 423 n. 7 (5th Cir.1999) (stay of extradition pending appeal); Then v. Melendez, 92 F.3d 851, 853 n. 1 (9th Cir.1996) (same), and for good reason: transferring a petitioner to the foreign country seeking extradition is obviously not the same as releasing him.
To be sure, as the government argues, Iraqi authorities might arrest Omar the moment U.S. forces release him. Expanding on this point, the dissent speculates that “if the government simply releases Omar and allows him to walk out of Camp Bucea, he might well find a dozen armed Iraqi soldiers waiting for him.” Dissenting Op. at 6. The dissent also thinks that U.S. military officials could “notify Iraqi authorities as to the exact time and place of [his] release, thereby effectively ensuring his immediate recapture and detention.” Id. As a result, the dissent maintains, Omar has failed to demonstrate irreparable injury because even if he prevails at his habeas hearing, he may nonetheless end up in Iraqi custody. We disagree.
To begin with, such speculation at the appellate level cannot defeat Omar’s right to a habeas hearing on the lawfulness of his detention and transfer. See CityFed Fin. Corp. v. Office of Thrift Supervision, 58 F.3d 738, 746 (D.C.Cir.1995) (“We review a district court decision regarding a preliminary injunction for abuse of discretion .... ”). At this point in time, we have no way of knowing how the U.S. military would release Omar if the district court ultimately rules in his favor, much less whether and to what extent the military would communicate with Iraqi authorities. Nor do we have any idea what would happen to Omar once released. Perhaps he would end up in Iraqi custody, but perhaps he would not. For example, perhaps because of developments at the habeas hearing, such as the appearance of defects in the government’s case or the introduction of exculpatory evidence, the Iraqis would decide that Omar is no longer worth prosecuting. Or perhaps by the time the district court ordered Omar’s release, Iraqi priorities would have changed, leaving Iraqi authorities uninterested in allocating scarce military resources — much less the dissent’s dozen soldiers — to his [13]*13arrest. The point is that on the record before us at this stage of these proceedings, neither the government nor the dissent nor we can possibly know what would happen to Omar if the district court barred his transfer and ordered his release. Given this uncertainty, a preliminary injunction protecting Omar from the certainty of transfer now is hardly an “empty gesture.” See Dissenting Op. at 18.
The dissent’s speculation about a U.S. military “tip-off’ to the Iraqis suffers from a second defect. If the district court ultimately rules that the U.S. military lacks authority to transfer Omar, the military will be unable to transfer him either directly through a formal handoff or indirectly by “releasing” him with a wink-and-a-nod to the Iraqis. The United States may certainly share information with other sovereigns, see id. at 4-5, but it may not do so in a way that converts Omar’s “release” into a transfer that violates a court order. The district court has jurisdiction to hear Omar’s habeas petition, see part III supra, and federal courts have authority to enforce their orders; contrary to the dissent, see Dissenting Op. at 17, the political question doctrine is not implicated. In any event, we think it exceedingly unlikely that American military officers, sworn to uphold the law and represented by the Justice Department, would evade an order of a United States district court. Indeed, if the district court orders Omar’s release, we are confident that military officials and their lawyers will work in good faith with the district court to fashion an order that, based on then-existing circumstances, ensures his lawful release from American custody.
The government’s observation that “a court may not artificially prolong a case or controversy by issuing an injunction the effect of which is to prevent the Government from rendering the petition moot by granting relief,” Appellants’ Br. 57, though undoubtedly correct, has nothing to do with the issue before us. Because the military plans to transfer Omar to Iraqi authorities, not to release him, the preliminary injunction, far from “prevent[ing] the Government from rendering the petition moot by granting relief,” preserves the district court’s jurisdiction to review the lawfulness of that transfer.
The government cites Spencer v. Kemna, 523 U.S. 1, 118 S.Ct. 978, 140 L.Ed.2d 43 (1998), for the proposition that a prisoner’s final release from prison moots his habeas petition. It also cites a series of cases, including Yohey v. Collins, 985 F.2d 222 (5th Cir.1993), for the proposition that conviction moots a pre-trial habeas petition. Omar, however, has not been released nor has he been convicted of a crime, so the cited decisions have no bearing on this case.
Taking a different tack, the dissent believes Omar has failed to demonstrate likelihood of success on the merits because “to make an injunction against transfer to Iraqi authorities a viable form of preliminary relief, Omar would need to show .... [what] we might call ‘release plus,”’ i.e., “release combined with immunity to Iraqi prosecution, release following surreptitious transport out of Iraq, or release with a promise to conceal the time and place of the release.” Dissenting Op. at 18. Like the dissent’s other arguments, this argument is wholly speculative, for it assumes, without record support, that Omar will, once released, require protection from Iraqi authorities. But as we indicate above, no one knows at this time what will happen to Omar if the district court orders his release. Speculating about the conditions under which the military might release Omar or the lawfulness of those conditions is thus not only premature — the matter [14]*14may never arise — but irrelevant to the issue before us: whether the district court abused its discretion in issuing the injunction. In any event, Omar’s petition does not seek “release plus”; rather, alleging that Omar is being held by the U.S. military in violation of his constitutional rights, the petition seeks his release from military custody.
The dissent asserts that “interference by a court in the decisions of sovereigns acting jointly within the same territory is unprecedented,” Dissenting Op. at 19-20, citing only Floyd v. Henderson, 456 F.2d 1117 (5th Cir.1972), in support. In Floyd, a federal prisoner argued that the federal government could no longer exercise jurisdiction over him because the Attorney General had transferred him to a state prison, where he served a state sentence concurrently with his federal sentence. Because a statute, 18 U.S.C. § 4082, authorized the Attorney General to assign federal prisoners to state prisons, the court ruled that “[the statutory] authority is sufficient to permit the transfer of petitioner from one institution to another prison.” 456 F.2d at 1119. But as we have explained, whether the military has authority to transfer Omar is one of the central questions in this habeas litigation. Floyd and the proposition for which the dissent cites it would thus become relevant only if the district court rules either that the military possesses the requisite statutory or treaty authority to transfer Omar, or that no such authority is necessary.
According to the dissent, “the bar on Omar’s presentation before the CCCI while in United States custody is improper.” Dissenting Op. at 17. Although we agree with the dissent that the injunction prohibits the military from presenting Omar to the CCCI for trial, we think this an appropriate exercise of the district court’s discretion. The dissent’s argument hinges on the assumption, unsupported in the record, that Omar’s “presentation does not hamper the ability of the government to order Omar’s release should the district court rule in his favor.” Id. Although the government has advised us that the “MNF-I maintains physical custody of detainees while their cases are being heard by the CCCI,” Gardner Decl. at 6, we cannot determine from the record whether this means legal custody. We are thus uncertain whether once Omar is in a CCCI courtroom, the Iraqi judge could remand him to Iraqi custody, an action that would obviously defeat the district court’s habeas jurisdiction. Nor can we determine whether Omar’s presentation for trial might by itself amount to the very “transfer” that Omar argues the military lacks authority to execute. Given these uncertainties and their potential implications for habeas jurisdiction, the district court, by enjoining Omar’s presentation for trial, clearly acted within its discretion.
Pointing out that “American citizenship is not a grant of immunity to commit crimes in other countries with impunity,” the dissent thinks our decision has the “remarkable effect of enabling a court sitting in Washington, D.C., to block the efforts of a foreign sovereign to make an arrest on its own soil.” Dissenting Op. at 19. U.S. courts, of course, have no authority to constrain the actions of Iraqi authorities. But in this case the government concedes that Omar is in the custody of United States officials, and Omar’s petition merely calls on the district court to determine whether those officials are complying with American law — an altogether unremarkable action for a United States district court.
V.
In sum, neither Hirota nor the political question doctrine deprives the district [15]*15court of jurisdiction to entertain Omar’s petition for a writ of habeas corpus. Because transfer would not afford Omar all the relief he could obtain through a writ of habeas corpus and because the district court’s preliminary injunction properly preserves its jurisdiction to entertain his petition, we affirm.
So ordered.