Dissenting opinion filed by Circuit Judge TATEL.
GRIFFITH, Circuit Judge:Abd Al-Rahim Hussein Muhammed Al-Nashiri is the alleged mastermind of the bombings of the U.S.S. Cole and the French supertanker the M/V Limburg, as well as the attempted bombing of the U.S.S. The Sullivans. Together, the completed attacks killed 18 crew members and injured dozens more. The government charged Al-Nashiri with nine offenses for his role in the attacks' and convened a military commission to try him. His trial, and any subsequent appeals, will be governed by the Military Commissions Act, in which Congress strengthened the procedural protections and review mechanisms for military commissions in response to the Supreme Court’s guidance in Hamdan v. Rumsfeld, 548 U.S. 557, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006). Al-Nashiri now seeks to avoid the structure Congress has created. He petitions for a writ of mandamus to dissolve the military commission convened to try him and appeals the district court’s denial of his motion to preliminarily enjoin that trial. We deny the petition for mandamus relief and affirm the district court.
I
A
At this pretrial stage, we recount the details of Al-Nashiri’s alleged offenses based on the information provided in the government’s charges. Al-Nashiri, a Saudi national, is a member of al Qaeda who orchestrated the attempted bombing of The Sullivans in January 2000 and the successful bombings of the Cole in October 2000 and the Limburg in October 2002.
Al-Nashiri met with Osama bin Laden and other senior members of al Qaeda -in 1997 or 1998 to plan a “boats operation” that would attack ships in the Arabian Peninsula. The government argues that while bin Laden was planning the “boats operation,” he was also coordinating the “planes operation” that would unfold on September 11, 2001. At bin Laden’s direction, Al-Nashiri and his alleged co-eon-spirator, Walid bin Attash, traveled to Yemen around 1998 to prepare for the boats operation. Al-Nashiri scouted the region and monitored ship traffic. He and his co-conspirators ultimately focused on Aden Harbor and bought and stored explosives to carry out an attack there. In 1999, after bin Attash was arrested, bin Laden instructed Al-Nashiri to take control of the operation. Al-Nashiri and his co-conspirators recruited others to the cause, bought a boat, and obtained false identification documents.
Under Al-Nashiri’s direction, his co-conspirators steered an explosive-filled boat toward The Sullivans in January 2000 while the warship was refueling. But the boat carrying the explosives foundered in Yemen’s Aden Harbor, thwarting the plan. Al-Nashiri and his co-conspirators recovered the boat and confirmed that the explosives could be used in future attacks. Sometime after the failed attack, Al-Nashi-ri returned to Afghanistan to meet with bin Laden and other high-ranking members of al Qaeda and to receive explosives training from an al Qaeda expert.
By the summer of 2000, Al-Nashiri had returned to Yemen to carry out prepara*114tions for a second attack in Aden Harbor. He and his co-conspirators rented a house from which they could surveil the harbor, repaired and tested the attack boat, filled it with explosives, and arranged for the attack to be videotaped. Sometime around September 2000, Al-Nashiri reported to bin Attash — who by then had been released from jail and was in Afghanistan— that the operation was ready and that he had chosen suicide bombers to carry it out. Before the attack, Al-Nashiri returned to Afghanistan at bin Laden’s direction and told him the bombing was imminent.
Adhering to Al-Nashiri’s instructions, in October 2000 the suicide bombers launched the boat — again filled with explosives — and piloted it toward the Cole, which was refueling in Aden Harbor. The bombers gave friendly gestures to crew members and steered their boat alongside the Cole, where they detonated the explosives. The blast killed 17 crew members and injured at least 37, and left a hole in the Cole’s side measuring about 30 feet in diameter.
After the attack, Al-Nashiri began planning another bombing. He and his co-conspirators acquired another boat and explosives, with Al-Nashiri directing the transfer of money to fund the attack. In October 2002, suicide bombers under Al-Nashiri’s direction drew their explosive-filled boat alongside the French supertanker the Limburg near the port of A1 Mukallah, Yemen. The explosion blasted a hole in the ship’s hull, killing one crew member and injuring 12. Some 90,000 barrels of oil also spilled from the tanker into the Gulf of Aden.
Local authorities arrested Al-Nashiri in Dubai in 2002 and turned him over to U.S. custody. He was transferred to the Guantanamo Bay Naval Base in 2006. A year later, a Combatant Status Review Tribunal determined that Al-Nashiri was detainable as an “enemy combatant” under the Authorization for Use of Military Force that Congress had passed and the President had signed in response to the attacks of September 11, 2001. Al-Nashiri v. MacDonald, 741 F.3d 1002, 1005 (9th Cir. 2013). The AUMF permits the President to use “all necessary and appropriate force” against the “nations, organizations, or persons” he determines were responsible for the 9/11 attacks. Pub. L. No. 107-40, § 2(a), 115 Stat. 224, 224 (2001). Al-Nashiri filed a petition for a writ of habeas corpus in the United States District Court for the District of Columbia in 2008, challenging various aspects of his detention at Guantanamo. Three years later, with Al-Nashiri’s habeas petition still pending, the Defense Department convened a military commission to try him for offenses including terrorism, murder in violation of the law of war, and attacking civilians. In re Al-Nashiri, 791 F.3d 71, 75 (D.C. Cir. 2015). The government is seeking the death penalty.
B
The current system of military commissions at Guantanamo Bay “is the product of an extended dialogue among the President, the Congress, and the Supreme Court.” Al-Nashiri, 791 F.3d at 73. After the passage of the AUMF in September 2001, the President began detaining enemy combatants and trying them by military commission at Guantanamo. The Supreme Court considered the legality of the commissions established by the President in Hamdan v. Rumsfeld, 548 U.S. 557, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006), and held that they exceeded certain limits Congress had previously imposed on the President’s authority. Specifically, the Court concluded that the President’s commissions did not comply with procedural protections set out in the Uniform Code of *115Military Justice (UCMJ) and the Geneva Conventions. See id. at 613, 620-28, 126 S.Ct. 2749. But four Justices explained that “[b]ecause Congress [ ] prescribed these limits [on presidential authority], Congress can change them, requiring a new analysis consistent with the Constitution and other governing laws.” Id. at 653, 126 S.Ct. 2749 (Kennedy, J., concurring).
In response, Congress passed the Military Commissions Act (MCA), which established a system of military commissions and largely exempted them from the requirements of the UCMJ and the Geneva Conventions. The MCA created the Court of Military Commission Review (CMCR) and empowered it to review judgments of military commissions. Al-Nashiri, 791 F.3d at 74. Under the current version of the MCA, as revised in 2009, the CMCR is composed of military and civilian judges who sit in panels of at least three. See 10 U.S.C. §§ 950d, 950f. It reviews questions of both fact and law. See id. § 950f. Our court has authority under the MCA to review military-commission convictions, as approved by the CMCR. Id. § 950g(a). We may review the CMCR’s legal conclusions, including the sufficiency of the evidence supporting the verdict. Id. § 950g(d).
The MCA provides that military commissions have jurisdiction to try “alien unprivileged enemy belligerents],” id. § 948c, for “any offense made punishable” by the MCA, “whether such offense was committed before, on, or after September 11, 2001.” Id. § 948d. The statute then lists 32 offenses that are “triable by military commission.” Id. § 950t. It further provides that “[a]n offense specified in this subchapter is triable by military commission under this chapter only if the offense is committed in the context of and associated with hostilities.” Id. § 950p(c). Hostilities are defined as “any conflict subject to the laws of war.” Id. § 948a(9).
Al-Nashiri’s military-commission proceedings were placed on hold in early 2015, when the presiding military judge granted Al-Nashiri’s motion to abate the commission’s proceedings while the government pursued interlocutory appeals of two rulings. By statute, the government may take an interlocutory appeal of any ruling by a military judge that terminates commission proceedings on a charge or that “excludes evidence that is substantial proof of a fact material in the proceeding.” 10 U.S.C. § 950d(a)(l)-(2).
In the first interlocutory appeal, the government contested the military judge’s dismissal in 2014 of the charges stemming from the bombing of the Limburg. Al-Nashiri, 791 F.3d at 75. The military judge dismissed these charges because the government had not introduced evidence to support its claim that the military commission had jurisdiction over offenses related to an attack on a French vessel. Two military judges and one civilian judge were assigned to hear this appeal. In the second interlocutory appeal, the government challenged a 2015 ruling by the military judge that forbade it from introducing evidence that Al-Nashiri’s actions endangered the lives of foreign nationals not onboard the Cole.
Al-Nashiri sought a writ of mandamus from our court in late 2014 to halt the first of these interlocutory appeals. He argued in part that because the two military judges on his CMCR appellate panel were “principal” officers, they should have been appointed to the CMCR by the President and confirmed by the Senate. See U.S. Const. art. II, §2, cl. 2; Al-Nashiri, 791 F.3d at 82. Their assignment to the CMCR by the Secretary of Defense violated the Constitution, Al-Nashiri asserted. See Al-Nashiri, 791 F.3d at 82. We denied his petition because Al-Nashiri had not shown he was clearly and indisputably entitled to *116mandamus relief, but we observed that the President and Senate could “put to rest any Appointments Clause questions regarding the CMCR’s military judges” by nominating and confirming them. Id. at 86. The President chose to take that tack. At the government’s request — which Al-Nash-iri did not oppose — the CMCR stayed its proceedings in both interlocutory appeals in June 2015 while the confirmation process was underway.
The Senate confirmed two military judges in April 2016, and the CMCR lifted its stay at the government’s request, even though Al-Nashiri asked the CMCR to continue the stay. See Order, United States v. Al-Nashiri, No. 14-001, — F.Supp.3d -, 2016 WL 3769710 (U.S.C.M.C.R. May 18, 2016). The CMCR then ruled on Al-Nashiri’s interlocutory appeals in June and July 2016, reversing the military judge’s dismissal of the charges related to the Limburg and its order excluding evidence. After the resolution of these appeals, the government asked the military commission to proceed. The commission granted that request, and the government states that commission proceedings will resume in September 2016. See Rule 28(j) Letter of Resp’t (filed Aug. 5, 2016).
C
In the present case, Al-Nashiri does not challenge the structural or procedural features of the military commissions created by Congress. He does not assert that the commissions are unconstitutional or that he was improperly classified as an “alien unprivileged enemy belligerent” subject to their jurisdiction. 10 U.S.C. § 948c. Instead, he argues that the offenses for which he has been charged are not triable by a military commission under the MCA because they were not “committed in the context of and associated with hostilities.” Id. § 950p(c). Because his alleged offenses had no nexus to hostilities, he contends, they are not war crimes, the only type of crime over which a military commission has jurisdiction under the Constitution.
Al-Nashiri first advanced these arguments in a motion to dismiss in 2012, but the military judge denied the motion without prejudice. According to the military judge, the existence of hostilities was a mixed question of law and fact. To the extent that it was a pure question of law, he deferred to what he called the “implicit” determinations of the political branches that hostilities existed at the time of Al-Nashiri’s alleged offenses. To the extent that the existence of hostilities was a question of fact, the government would need to prove that at trial.
Before us, Al-Nashiri advances his claims in two separate actions, which are consolidated here. The first began in 2014, when Al-Nashiri received permission from our district court to amend the habeas petition he filed in 2008. His amended petition asked the district court to enjoin his trial by the military commission and enter a declaratory judgment that his conduct did not occur in the context of hostilities. He also moved for a preliminary injunction to prevent his trial before the military commission until the district court ruled on his habeas petition. The government opposed this motion and moved to hold the habeas action in abeyance to allow the commission proceedings and corresponding appeals to run their course. To support its motion to hold the case in abeyance, the government relied upon Schlesinger v. Councilman, 420 U.S. 738, 95 S.Ct. 1300, 43 L.Ed.2d 591 (1975), where the Supreme Court directed federal courts to generally refrain from enjoining ongoing courts-martial. See id. at 756-58, 95 S.Ct. 1300. According to the government, Councilman likewise supports ab*117staining from interfering with ongoing proceedings in a military commission.
The district court found that adjudicating Al-Nashiri’s habeas petition would unduly interfere with the proceedings of the military commission and accordingly granted the government’s motion to hold the case in abeyance pending the resolution of his military-commission trial and any subsequent appeals. Al-Nashiri v. Obama, 76 F.Supp.3d 218, 221-23 (D.D.C. 2014). The district court then denied as moot Al-Nashiri’s motion to preliminarily enjoin his military-commission trial pending the resolution of his habeas petition. Id. at 222 n.3. On appeal, Al-Nashiri challenges the district court’s denial of preliminary injunctive relief, arguing primarily that abstention was inappropriate and that the district court therefore should have decided his motion on the merits.
The second action before us is a petition for a writ of mandamus. Al-Nashiri asks us to dissolve the military commission convened to try him, also on the ground that his conduct did not take place in the context of hostilities.
We have jurisdiction to review the district court’s denial of preliminary injunc-tive relief under 28 U.S.C. § 1292(a)(1).1 We have jurisdiction to issue a writ of mandamus to a military commission under the All Writs Act, 28 U.S.C. § 1651(a), and the 2009 MCA, 10 U.S.C. § 950g(a). See Al-Nashiri, 791 F.3d at 76-78 (“[TJhis Court has jurisdiction to issue a writ of mandamus in aid of our appellate jurisdiction of military commissions and the CMCR.”). We affirm the district court and deny Al-Nashiri’s petition for mandamus relief.
II
We first consider Al-Nashiri’s claim that the district court erred in denying his motion to preliminarily.enjoin his trial before the military commission pending the resolution of his habeas petition. The district court denied the motion based on its decision to hold Al-Nashiri’s habeas petition in abeyance pending the resolution of his case in the commission. Thus, to determine whether this denial was proper, we must examine whether the district court erred in staying Al-Nashiri’s habeas case.2
We emphasize at the outset that the question in this case is not whether Al-Nashiri will be able to make his “hostilities” argument to an Article III court. The MCA provides an appeal as of right to our *118court. The question in this case is when that argument to us may occur. The district court decided that Article III review should occur at the time that Congress contemplated: after any conviction and accompanying appeal in the military system. We generally review such decisions to stay a case “in favor of an ongoing proceeding” for abuse of discretion. Handy v. Shaw, Bransford, Veilleux & Roth, 325 F.8d 846, 349 (D.C. Cir. 2003). “Whether the lower court applied the proper legal standard in exercising that discretion, however, is a question of law reviewed de novo.” Id. We assume these standards apply here. We first ask whether the district court “applied the proper legal standard” in deciding to abstain from hearing Al-Nashiri’s habeas petition. In other words, did the district court commit legal error in extending the abstention principles established in Schlesinger v. Councilman, 420 U.S. 738, 95 S.Ct. 1300, 43 L.Ed.2d 591 (1975), which dealt with courts-martial, to Al-Nashiri’s pretrial challenge to the subject matter jurisdiction of a military commission?3 Concluding that the district court did not err as a matter of law, we then ask whether its ultimate decision to abstain based on any circumstances unique to Al-Nashiri’s case was appropriate. Because we conclude that it was, we affirm the district court.
A
The district court did not err, as a matter of law, in extending the principles announced in Councilman to Al-Nashiri’s case.
i
Federal courts generally “have a strict duty to exercise the jurisdiction that is conferred upon them by Congress.” Quackenbush v. Allstate Ins. Co., 517 U.S. 706, 716, 116 S.Ct. 1712, 135 L.Ed.2d 1 (1996). This duty “is not, however, absolute.” Id. In the context of criminal prosecutions, federal courts routinely decline to adjudicate petitions that seek collateral relief to prevent a pending prosecution. See, e.g., Henry v. Henkel, 235 U.S. 219, 228-30, 35 S.Ct. 54, 59 L.Ed. 203 (1914) (petition seeking habeas relief); JMM Corp. v. District of Columbia, 378 F.3d 1117, 1120 (D.C. Cir. 2004) (petition seeking injunc-tive and declaratory relief). This practice stems in part from a “basic doctrine of equity jurisprudence,” which provides that courts should not exercise their equitable discretion to enjoin criminal proceedings, as long as the defendant has an adequate legal remedy in the form of trial and direct appeal. Jarkesy v. SEC, 803 F.3d 9, 26 (D.C. Cir. 2015); see also Deaver v. Seymour, 822 F.2d 66, 68-69 (D.C. Cir. 1987). Thus, where the issue the petitioner challenges can be litigated in pretrial motions and raised as a defense at trial, federal courts typically require the petitioner to navigate that process instead of skirting it. See Jarkesy, 803 F.3d at 26.
In Councilman, the Supreme Court extended this basic doctrine to a new context: courts-martial. The case involved a court-martial convened to try an Army officer for selling and possessing marijua*119na. At the time, Supreme Court precedent required that an alleged offense be “service connected” to be constitutionally triable by court-martial. See O’Callahan v. Parker, 395 U.S. 258, 272-73, 89 S.Ct. 1683, 23 L.Ed.2d 291 (1969) (establishing “service connection” rule), overruled by Solorio v. United States, 483 U.S. 435, 107 S.Ct. 2924, 97 L.Ed.2d 364 (1987). Councilman filed suit in district court to enjoin the court-martial from proceeding, arguing that the military lacked jurisdiction to try him because his alleged offense was not connected to his service in the army. See Councilman, 420 U.S. at 741, 95 S.Ct. 1300. The district court granted the injunction, and the Tenth Circuit affirmed. Id. at 739-40, 95 S.Ct. 1300. But the Supreme Court reversed, holding that “when a serviceman charged with crimes by military authorities can show no harm other than that attendant to resolution of his case in the military court system, the federal district courts must refrain from intervention, by way of injunction or' otherwise.” Id. at 758, 95 S.Ct. 1300.
The Court grounded its decision in the corresponding abstention doctrine for state criminal prosecutions announced four years earlier in Younger v. Harris, 401 U.S. 37, 91 S.Ct. 746, 27 L.Ed.2d 669 (1971). Abstention in favor of ongoing state criminal proceedings in Younger was based on two considerations: the traditional rule that courts of equity should not enjoin criminal prosecutions where an adequate remedy at law exists, see id. at 43-44, 91 S.Ct. 746, and interests of “comity,” perhaps better described in that case as “federalism,” id. at 44-45, 91 S.Ct. 746. Interference in ongoing state proceedings would disrupt the careful balance between state and federal power. See id.
The Councilman Court acknowledged that the “peculiar demands of federalism” were not applicable to courts-martial, but it explained that “factors equally compelling” justified its decision to allow courts-martial to run their course without interference by the federal courts. 420 U.S. at 757, 95 S.Ct. 1300. As the Supreme Court explained in Hamdan v. Rumsfeld, 548 U.S. 557, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006), Councilman relied on two “comity” factors other than federalism, focusing on the military interests advanced by allowing courts-martial to proceed uninterrupted and on the adequacy of the court-martial system in protecting service members’ rights:
First, military discipline and, therefore, the efficient operation of the Armed Forces are best served if the military justice system acts without regular interference from civilian courts. Second, federal courts should respect the balance that Congress struck between military preparedness and fairness to individual service members when it created “an integrated system of military courts and review procedures, a critical element of which is the Court of Military Appeals consisting of civilian judges completely removed from all military influence or persuasion....”
Id. at 586, 126 S.Ct. 2749 (quoting Councilman, 420 U.S. at 758, 95 S.Ct. 1300) (internal citations omitted). As the Court later explained, “abstention in the face of ongoing court-martial proceedings is justified by our expectation that the military court system established by Congress — with its substantial procedural protections and provision for appellate review by independent civilian judges — ‘will vindicate servicemen’s constitutional rights.’ ” Id. (quoting Councilman, 420 U.S. at 758, 95 S.Ct. 1300).
In Hamdan, the Supreme Court considered whether to extend the principles set out in Councilman to abstain from adjudicating a Guantanamo detainee’s challenge *120to his trial before a military commission. To reiterate, the commission set to try Hamdan was convened by the President without specific congressional authorization. In that context, the Court declined to abstain, concluding that neither of Councilman’s comity considerations was present. As to the first, the Court said simply that Hamdan was “not a member of our Nation’s Armed Forces, so concerns about military discipline do not apply.” Id. at 587, 126 S.Ct. 2749. And as to the second, the Court explained that the military commission trying Hamdan was “not part of the integrated system of military courts, complete with independent review panels, that Congress has established.” Id. Unlike Councilman, the Court emphasized, Ham-dan had no right to appeal a conviction to a review body that was “structurally] insulated] from military influence.” Id. Rather, any conviction would be reviewed only by Executive Branch officials: first a panel of three military members selected by the Secretary of Defense, then the Secretary himself, and finally the President. Id. And because these review bodies lacked the structural independence of the Court of Appeals for the Armed Forces, whose civilian judges review court-martial convictions, they bore “insufficient conceptual similarity to state courts to warrant invocation of abstention principles.” Id. at 588, 126 S.Ct. 2749. The Court further explained that the government had not identified any other “important countervailing interest” that justified abstaining. Id. at 589, 126 S.Ct. 2749 (quoting Quackenbush, 517 U.S. at 716, 116 S.Ct. 1712).
The Hamdan Court instead determined that Ex parte Quirin, 317 U.S. 1, 63 S.Ct. 1, 87 L.Ed. 3 (1942), was the most relevant precedent. In Quirin, rather than decline to intervene in ongoing proceedings of a military commission, the Court convened a special Term to hear the case and expedited its review, explaining that the issues were of great public importance. See id. at 19, 63 S.Ct. 1. The Hamdan Court closed its discussion by noting: “While we certainly do not foreclose the possibility that abstention may be appropriate in some cases seeking review of ongoing military commission proceedings (such as military commissions convened on the battlefield), the foregoing discussion makes clear that, under our precedent, abstention is not justified here.” 548 U.S. at 590, 126 S.Ct. 2749.
ii
Much has changed since Hamdan. Within four months of the Supreme Court’s opinion — and in direct response to it— Congress passed the MCA, which established enhanced procedural protections and rigorous review mechanisms for military commissions. The committee report accompanying the House version of the MCA indicated that the legislation was an effort to respond to Hamdan, in which “[t]he Court [ ] suggested that the President could ask the United States Congress to authorize commission rules that diverge from the UCMJ, provided that they were consistent with the Constitution and other laws.” H.R. Rep. No. 109-664, pt. 1, at 4-5 (2006). And when signing the 2006 MCA, President Bush explained that the Supreme Court had ruled that the military commissions he had established after September 11 “needed to be explicitly authorized by the United States Congress.” See Statement by President George W. Bush upon Signing S. 3930, 2006 U.S.C.C.A.N. S61 (Oct. 17, 2006). The President explained that he “asked Congress for that authority, and they [] provided it” by passing the MCA. Id.
Al-Nashiri and amici urge that despite the significant changes enacted in the MCA, abstention remains as inappropriate here as it was in Hamdan. They argue that Al-Nashiri, like Hamdan, is not a *121member of the Armed Forces, and commissions are fundamentally different from courts-martial. By contrast, the government contends that the MCA established the rigorous system of review found lacking in Hamdan and that the district court was warranted in allowing the military commission to proceed. It insists that while Councilman does not directly control, it is the closest analogue in our jurisprudence, because comity considerations “equally compelling” as those in Councilman, 420 U.S. at 757, 95 S.Ct. 1300, point in favor of abstention here.
To determine whether “equally compelling” factors exist here, we must identify the precise role played by Councilman’s two comity considerations. Evaluating those considerations, we conclude that to abstain we must be assured of both the adequacy of the alternative system in protecting the rights of defendants and the importance of the interests served by allowing that system to proceed uninterrupted by federal courts. The comity consider- ' ations in Councilman established both of these elements. With respect to adequacy, the Court did not evaluate the on-the-ground performance of courts-martial in protecting service members’ rights. Instead, it “assumed” the sufficiency of the structure Congress created, with its substantial procedural protections and provision for appellate review by judges insulated from military influence. Id. at 758, 95 S.Ct. 1300; see also Hamdan, 548 U.S. at 586, 126 S.Ct. 2749 (characterizing Coun-oilman’s reasoning as such).4 And as for importance, the Court explained that abstention would serve a vital interest by permitting the military to discipline soldiers without immediate interference by federal courts. Councilman, 420 U.S. at 757, 95 S.Ct. 1300.
The Court’s emphasis on these two considerations made sense in light of its abstention jurisprudence, developed in the context of state-court proceedings. That precedent made clear that abstention was appropriate only (1) where the petitioner would have an adequate remedy in the alternative forum, see Kugler v. Helfant, 421 U.S. 117, 124, 95 S.Ct. 1524, 44 L.Ed.2d 15 (1975) (“The policy of equitable restraint [in favor of state criminal proceedings] is founded on the premise that ordinarily a pending state prosecution provides the accused a fair and sufficient opportunity for vindication of federal constitutional rights.”); Younger, 401 U.S. at 45, 91 S.Ct. 746 (“The accused should first set up and rely upon his defense in the state courts ... unless it plainly appears that this course would not afford adequate protection.” (quoting Fenner v. Boykin, 271 U.S. 240, 243-44, 46 S.Ct. 492, 70 L.Ed. 927 (1926))), and (2) where abstention would “clearly serve an important countervailing interest,” Allegheny Cty. v. Frank Mashuda Co., 360 U.S. 185, 189, 79 S.Ct. 1060, 3 L.Ed.2d 1163 (1959), such as reducing friction between federal and state governments, see Younger, 401 U.S. *122at 44, 91 S.Ct. 746 (emphasizing the need to “respect [ ] state functions” by avoiding pretrial intervention in state criminal prosecutions). The Councilman Court simply applied these central considerations to the context of courts-martial.
Taking our cue from Councilman, then, we ask two questions to determine whether any sufficiently “compelling” factors justified the district court’s decision to abstain. First, we consider whether the system enacted to adjudicate Al-Nashiri’s guilt will adequately protect his rights. And second, we examine whether an “important countervailing interest” justifies the decision to avoid the district court adjudicating a pretrial challenge to the subject matter jurisdiction of a military commission created under the MCA.
iii
To answer the first question, we are convinced that the MCA’s review structure is adequate because it is virtually identical to the review system for courts-martial approved by the Court in Councilman. In the MCA, Congress established an “integrated” scheme dictating how enemy belligerents are to be tried and obtain appellate review, Councilman, 420 U.S. at 758, 95 S.Ct. 1300, and two Presidents sanctioned this approach — President Bush in 2006, when the MCA was first enacted, and President Obama in 2009, when it was revised. Pursuant to that structure, Al-Nashiri faces a trial with a military judge presiding and a “jury” that, in capital cases, generally consists of twelve military officers known as “members” of the military commission. 10 U.S.C. §§ 948m, 949m(c). If he is convicted, the convening authority — -the Defense Department official who initially referred the case to trial — may review the guilty finding and set it aside, or reduce it to a finding of guilty of a lesser-included offense. Id. § 950b. The convening authority must review a sentence to approve, disapprove, commute, or suspend it in whole or in part. Id. A final guilty finding, as modified by the convening authority, will then be reviewed by the CMCR unless the defendant properly waives this right of review. Id. §§ 950f, 950c. The CMCR is composed of both military and civilian judges and has the power to review factual and legal questions alike. Id. § 950f. The defendant may appeal the CMCR’s decision to our court, and we are empowered to review all questions of law, including the sufficiency of the evidence. Id. § 950g. Finally, our ruling can be challenged via petition for writ of certiorari in the Supreme Court. Id. § 950g(e).
These review structures “closely (and intentionally) mirror[] the current structure for ... review of courts-martial.” Stephen I. Vladeck, Exceptional Courts and the Structure of American Military Justice, in Guantanamo and Beyond 163, 175 (Fionnuala Ni Aolain & Oren Gross eds., 2013). Not only does the composition of the commission itself closely mirror that of a court-martial — both have twelve members in capital cases and a presiding military judge — but the structure of appellate review is virtually identical across the two systems. The “scope of the CMCR’s post-conviction review is a word-for-word copy” of the portion of the UCMJ that sets out the authority of each service’s Court of Criminal Appeals, the military body that reviews court-martial convictions. Id. Compare 10 U.S.C. § 950f, with id. § 866. Similarly, the authority given to this court to review the CMCR’s decision is as broad as the authority that the UCMJ gives the Court of Appeals for the Armed Forces, the tribunal that Councilman approved as sufficiently “removed from [] military influence or persuasion,” 420 U.S. at 758, 95 S.Ct. 1300 (citing Noyd v. Bond, 395 U.S. 683, 694-95, 89 S.Ct. 1876, 23 L.Ed.2d 631 *123(1969)). Compare 10 U.S.C. § 950g(d), with id. § 866(c).
The similarity of the two systems’ review mechanisms strongly suggests that, if the review procedure for courts-martial is considered adequate to protect defendants’ rights, the same should be true of the review procedure for military commissions. Indeed, in one sense the review structure for military commissions is more insulated from military influence than is the structure for courts-martial. The judges on our court, unlike those on the Court of Appeals for the Armed Forces, enjoy Article Ill’s guarantees of life tenure and salary protection, further assuring that our review is not swayed by political pressures. See Hamdan, 548 U.S. at 675-76, 126 S.Ct. 2749 (Scalia, J., dissenting).
We do not overlook the fact that although the review structures are virtually identical, the evidentiary and procedural rules in a military-commission trial differ in some regards from those in courts-martial. Even so, Al-Nashiri’s trial before a military commission will include a number of significant procedural and evidentiary safeguards. Among other things, he will have the right to be represented by counsel, 10 U.S.C. § 949c, be presumed innocent, id. § 949Í, obtain and offer exculpatory evidence, id. § 949j, call witnesses on his behalf, id. and challenge for cause any of the members of the military commission and the military judge, id. § 949f. In fact, Al-Nashiri does not argue before us that any evidentiary or procedural defects will prevent the military commission and various appellate bodies from fully adjudicating his defense that his conduct occurred outside the context of hostilities. Cf. JMM Corp., 378 F.3d at 1127 (“For Younger abstention to be appropriate in the face of pending state proceedings, the federal plaintiff must ‘have a full and fair opportunity to litigate’ its constitutional claims in those proceedings.” (quoting Ohio Civil Rights Comm’n, 477 U.S. at 627, 106 S.Ct. 2718)). We therefore conclude that, at least where a defendant identifies no such defect, the MCA’s “integrated system of military courts and review procedures,” Councilman, 420 U.S. at 758, 95 S.Ct. 1300, is sufficiently adequate to point in favor of abstention.
Al-Nashiri argues against this conclusion by identifying various features of military commissions that, in his view, suggest that they are deficient as compared to the court-martial system. According to Al-Nashiri, the commissions established by the MCA lack the established track record that courts-martial" had at the time of Councilman. He also points to two instances in which our court overturned military-commission judgments on appeal. But Al-Nashiri does not argue that these features render military commissions unlawful or will prevent him from presenting a full defense. Instead, by pointing to these alleged shortcomings, Al-Nashiri asks us to do what the Supreme Court notably did not do in Councilman: determine whether pretrial intervention is warranted by examining the on-the-ground performance of the system that Congress and the Executive have established. See 420 U.S. at 758, 95 S.Ct. 1300 (“[implicit in the congressional scheme embodied in the [UCMJ] is the view that the military court system generally is adequate to and responsibly will perform its assigned task. We think this congressional judgment must be respected and that it must be assumed that the military court system will vindicate servicemen’s constitutional rights.” (emphases added)). In the absence of any claim that the shortcomings to which Al-Nashiri points render the congressional scheme unlawful or will prevent Al-Nashiri from fully defending himself, the district court did not err in deeming that scheme adequate.
*124iv
We next ask whether an “important countervailing interest” permits a federal court to decline to adjudicate a defendant’s pretrial claim that a military commission lacks subject matter jurisdiction to try his offense. It does. By providing for direct Article III review of Al-Nashiri’s jurisdictional challenge on appeal from any conviction in the military system, Congress and the President implicitly instructed that judicial review should not take place before that system has completed its work. And where this judgment was made out of concern for national security needs — an arena in which the political branches receive wide deference — we must follow their directive. We turn now to examining the vital interest we identify: the need for federal courts to avoid exercising their equitable powers in a manner that would unduly impinge on the prerogatives of the political branches in the sensitive realm of national security.5 Comity demands restraint in such circumstances, just as it requires federal courts to avoid interfering with the functions of states and the military. See, .e.g., Wash. Research Project, Inc. v. Dep’t of Health, Educ. & Welfare, 504 F.2d 238, 253 (D.C. Cir. 1974) (“Considerations of inter-branch comity impel us to withhold coercive orders that are not demonstrably necessary.” (emphasis added)).
Congress — with the approval of two Presidents — exercised its legitimate prerogatives when it decided, in response to Hamdan, that the ordinary federal court process was not suitable for trying certain enemy belligerents. Therefore, Congress crafted a separate-scheme under which they would be tried and potentially convicted. Longstanding historical practice supports trying such enemy belligerents by military commission, see, e.g., Quirin, 317 U.S. at 28-29, 63 S.Ct. 2, and the scheme Congress crafted in the MCA contains substantial additional protections as compared to the commissions used in past conflicts. One key difference, as we have explained, is that the MCA allows defendants an appeal as of right to our court. Article III courts therefore play a far more robust role in overseeing the actions of modern military commissions than they did in the past. See, e.g., Johnson v. Eisentrager, 339 U.S. 763, 787, 70 S.Ct. 936, 94 L.Ed. 1255 (1950) (“Correction of [military commissions’] errors of decision is not for the courts but for the military authorities which are alone authorized to review their decisions.” (quoting In re Yamashita, 327 U.S. 1, 8, 66. S.Ct. 340, 90 L.Ed. 499 (1946))). They also play a much larger part than they do in the review structure for courts-martial, which provides no appeal as of right to an Article III court.
Crucially, while the scheme Congress created in the MCA incorporates Article III review, it also delays it until a specific point. Before an Article III appellate court may step in, a defendant must first be tried and convicted in the military system, *125the convening authority must have approved the conviction, and the defendant must appeal the conviction to the CMCR or affirmatively waive his right to do so. Ordinarily, when Congress instructs that adjudication of certain types of cases should begin in specialized, non-Article III tribunals and end with review in an Article III court, we suppose that Congress intended for litigants to proceed exclusively through that scheme. See City of Rochester v. Bond, 603 F.2d 927, 931 (D.C. Cir. 1979). In other words, by providing for Article III involvement at a particular point, Congress “implicitly” signals that Article III courts should get involved no sooner. Jark-esy, 803 F.3d at 15. Litigants may not ordinarily seek to prevent the proper operation of the congressional scheme by pursuing equitable relief in district court.
We are particularly confident that Congress did not intend to allow a defendant to halt the workings of a military commission by challenging in federal court an issue that could just as easily be considered by the commission and reviewed by a federal appellate court: the commission’s own subject matter jurisdiction. The structure of the MCA makes this clear. For starters, the MCA explicitly empowers military commissions to make findings sufficient to determine their own jurisdiction, see 10 U.S.C. § 948d, and permits a presiding military judge to “hear[] and determine motions raising defenses or objections which are capable of determination without trial of the issues” bearing on guilt or innocence, id. § 949d. These provisions suggest “[b]y implication” that jurisdictional challenges are not ordinarily to be raised pretrial in district court. Deaver, 822 F.2d at 70; cf. id. at 69-70 (explaining that the existence of a procedure allowing defendants to move to dismiss an indictment pretrial suggests that defendants may not mount a collateral equitable challenge to the indictment on the same ground).
Moreover, a military judge’s order denying a motion to dismiss charges on jurisdictional grounds cannot be appealed to us until after final judgment. See Khadr v. United States, 529 F.3d 1112, 1114-15 (D.C. Cir. 2008). Our court has “exclusive jurisdiction to determine the validity of a final judgment rendered by a military commission,” as approved by the convening authority, once “all other appeals under this chapter have been waived or exhausted.” 10 U.S.C. § 950g(a)-(b). An order denying a motion to dismiss charges is not a “final judgment” under 10 U.S.C. § 950g(a), not least because it has not been approved by the convening authority. See Khadr, 529 F.3d at 1115-16. District courts would “undermine the final judgment rule” laid out by Congress were they routinely to entertain motions for equitable relief of the sort Al-Nashiri seeks, “with [their] attendant rights of appeal.” Deaver, 822 F.2d at 71.
Heeding the political branches’ instruction as to the timing of Article III review qualifies as an “important countervailing interest” warranting abstention, at least where that instruction is based on those branches’ assessment of national security needs. In the realm of national security, the expertise of the political branches is at its apogee. See Hamdi v. Rumsfeld, 542 U.S. 507, 531, 124 S.Ct. 2633, 159 L.Ed.2d 578 (2004) (plurality opinion) (“Without doubt, our Constitution recognizes that core strategic matters of warmaking belong in the hands of those who are best positioned and most politically accountable for making them.”); Al-Bihani v. Obama, 590 F.3d 866, 875 (D.C. Cir. 2010) (noting “the wide deference the judiciary is obliged to give to the democratic branches with regard to questions concerning national security”); Hamad v. Gates, 732 *126F.3d 990, 1006 (9th Cir. 2013) (“Congress’s decisions with respect to [Guantanamo] detainees are at the core of Congresses authority with respect to ‘the conduct of foreign relations [and] the war power.’ ” (quoting Mathews v. Diaz, 426 U.S. 67, 81 n.17, 96 S.Ct. 1883, 48 L.Ed.2d 478 (1976))). Acting on the guidance set out in Hamdan, the President sought authority for the military-commission trials that “he believe[d] necessary,” 548 U.S. at 636, 126 S.Ct. 2749 (Breyer, J., concurring), and Congress gave it to him, deciding in the process that Article III courts should not step in before the military system has issued a final decision. The district court did not err by declining to disturb this joint determination.
Al-Nashiri and amici raise several counterarguments, asserting that the interests supporting abstention in the military-commission context are less significant than those in the court-martial context. Al-Nashiri contends initially that Councilman does not apply because he is not a service member; and, as the dissent likewise points out, concerns of military discipline are therefore inapplicable. True enough. But nothing in the Supreme Court’s case law requires the interests justifying the district court’s decision to be identical to those in Councilman-, it is enough that they are “equally compelling.” Councilman, 420 U.S. at 757, 95 S.Ct. 1300; see also Hamdan, 548 U.S. at 589, 126 S.Ct. 2749. To require identical interests would be to suggest that abstention principles developed in the context of criminal proceedings in one forum can never be extended to another. But this cannot be correct. Indeed, Councilman itself was an' outgrowth of Younger abstention, which dealt with ongoing criminal proceedings in state courts and had nothing to do with military discipline.
To be sure, the Court in Hamdan did not consider interests other than military discipline in determining that it would hear the habeas petition before it. It noted simply that Hamdan was not a member of the Armed Forces, and that concerns of military discipline therefore did not apply. But the Court did not hold that abstention is appropriate only where concerns of military discipline are present. To the contrary, it left open the possibility that some other “important countervailing interest” might justify abstention in a future case. Hamdan, 548 U.S. at 589, 126 S.Ct. 2749 (quoting Quackenbush, 517 U.S. at 716, 116 S.Ct. 1712). The Court had no occasion in Hamdan to consider whether the vital interest we have identified here would point in favor of abstention, because Congress had not specifically authorized Hamdan’s military commission — much less incorporated Article III courts into the applicable review scheme. Indeed, the Supreme Court in Hamdan expressly declined to consider whether Congress’s provision of “limited” Article III review in the Detainee Treatment Act of 2005 pointed in favor of abstention, because Hamdan had no right to such review under that Act. Id. at 588 n.19, 126 S.Ct. 2749. That Hamdan did not consider interests other than military discipline, therefore, does not preclude us from doing so.
Al-Nashiri and amici further assert that abstention applies only to court systems that are wholly separate from the federal judicial establishment. They note that decisions of courts-martial and state courts are not directly reviewed by federal courts; moreover, these alternative judicial systems have a long history of operating undisturbed by federal intervention. Therefore, they argue, while the Court in Councilman was concerned with Article III courts intruding where they as a whole had no place, no similar concern is at play here, where Congress built Article III *127courts into the review mechanism. Comity interests are not implicated by such a structure, according to Al-Nashiri and am-ici.
Our role in reviewing military-commission convictions does, of course, distinguish the MCA’s review structure from that of state courts and courts-martial. But this distinction points away from pretrial intervention rather than toward it. For starters, while courts often invoke the term “comity” to refer to respect for separate judicial systems such as state courts, the term is more capacious than that. As we have explained, we have invoked inter-branch comity to avoid exercising our equitable discretion to interfere with the prerogatives of coordinate branches of government. Comity can also justify a district court’s discretionary decision to “transfer, stay, or dismiss a case that is duplicative of a case filed in another federal [district] court,” even though both courts are part of the same judicial system. Federal-Comity Doctrine, Black’s Law DictionaRY (10th ed. 2014); see, e.g., Pacesetter Sys., Inc. v. Medtronic, Inc., 678 F.2d 93, 94-95 (9th Cir. 1982).
Moreover, the eventual involvement of an Article III appellate court lessens the need for immediate intervention because an Article III court can remedy any errors on appeal. Indeed, before cases like Younger and Councilman, the traditional rule that equity should not interfere with a criminal prosecution generally applied only to cases in which a defendant had an adequate non-equitable remedy in a federal court. See Trainor v. Hernandez, 431 U.S. 434, 441, 97 S.Ct. 1911, 52 L.Ed.2d 486 (1977) (explaining that “the existence of an adequate remedy at law barring equitable relief normally would be determined by inquiring into the remedies available in the federal rather than in the state courts,” but Younger “broadened” the inquiry “to focus on the remedies available in the pending state proceeding”). If the availability of legal remedies in Article III courts has historically barred criminal defendants from receiving pretrial equitable relief, we do not see why in this case the availability of such remedies would counsel in favor of permitting pretrial relief.
Al-Nashiri and the dissent also contend that the military possesses no special expertise in addressing questions related to the laws of war. Thus, both argue, while part- of the reason for abstaining in Councilman was to defer to the military’s expertise in handling criminal matters connected to military service, no similar interest exists here. We are not convinced. For one thing, Councilman set out a rule that applies broadly — even to those claims that implicate military expertise to a lesser degree. See Solorio, 483 U.S. at 436-37, 107 S.Ct. 2924 (holding that courts-martial may try service members even for crimes unrelated to their military service). For another, Councilman suggested that expertise can be built over time; thus, the relative novelty of the military commissions need not necessarily count against them. See 420 U.S. at 758, 95 S.Ct. 1300 (noting that the civilian judges who reviewed court-martial convictions “would gain over time thorough familiarity with military problems”). And finally, Councilman cited military expertise as just one of several practical benefits of abstention. In addition to serving the needs of the military, avoiding pretrial intervention also eliminates “duplicative proceedings,” potentially “obviate[s] the need for judicial intervention,” and “inform[s] and narrow[s]” eventual Article III review. Id. at 756-58, 95 S.Ct. 1300. These advantages apply in full force here.
As in Councilman, then, an important countervailing interest supported the dis*128trict court’s decision to abstain from hearing Al-Nashiri’s petition.6
B
Having determined that the district court applied the proper legal standard when it decided that it could abstain in favor of ongoing military-commission proceedings, we next examine whether .its ultimate decision to abstain was appropriate, in light of any features unique to Al-Nashi-ri’s case. Al-Nashiri advances three arguments for why abstention was inappropriate here; none has merit.
i
The Supreme Court has instructed that federal courts can intervene in ongoing criminal proceedings in a few narrow and limited circumstances. In particular, a federal court may intervene where a plaintiff shows that “extraordinary circumstances” both present the threat of “great and immediate” injury and render the alternative tribunal “incapable of fairly and fully adjudicating the federal issues before it.” Kugler v. Helfant, 421 U.S. 117, 123-24, 95 S.Ct. 1524, 44 L.Ed.2d 15 (1975) (quoting Younger, 401 U.S. at 45, 53, 91 5.Ct. 746); see also Huffman v. Pursue, Ltd., 420 U.S. 592, 601, 95 S.Ct. 1200, 43 L.Ed.2d 482 (1975) (“[A] movant must show not merely the ‘irreparable injury’ which is a normal prerequisite for an injunction, but also must show that the injury would be ‘great and immediate.’ ” (quoting Younger, 401 U.S. at 46, 91 S.Ct. 746)). Al-Nashiri contends that this exception to abstention obligated the district court to intervene in his case because his proceeding before the military commission will cause him irreparable psychological harm and will require him to divulge his defense in advance of a possible retrial in federal district court. These harms, he asserts, amount to the sort of “great and immediate” irreparable injury that the Court has recognized could support a federal court’s decision not to abstain in his particular case. Councilman, 420 U.S. at 756, 95 S.Ct. 1300 (quoting Fenner v. Boykin, 271 U.S. 240, 243, 46 S.Ct. 492, 70 L.Ed. 927 (1926)).
Al-Nashiri’s argument is foreclosed by the Supreme Court’s definition of what constitutes “great, immediate, and irreparable” injury justifying a federal court’s intervention in ongoing criminal proceedings. Moore v. Sims, 442 U.S. 415, 433, 99 S.Ct. 2371, 60 L.Ed.2d 994 (1979); see also Councilman, 420 U.S. at 756, 95 S.Ct. 1300. As the Court explained in Councilman, “certain types of injury, in particular, the cost, anxiety, and inconvenience of having to defend against a single criminal prosecution, [cannot] by themselves be considered ‘irreparable’ in the special legal sense of that term.” 420 U.S. at 755, 95 S.Ct. 1300 (quoting Younger, 401 U.S. at 46, 91 S.Ct. 746). Instead, abstention is appropriate where a plaintiff “can show no harm other than that attendant to resolution of his case in the military court system,” even though those harms are “often of serious proportions.” Id. at 754, 758, 95 S.Ct. 1300; see also McLucas v. DeCham-*129plain, 421 U.S. 21, 33, 95 S.Ct. 1365, 43 L.Ed.2d 699 (1975) (holding that avoiding the possibility of erroneous incarceration throughout a court-martial proceeding does not qualify as “irreparable injury” for purposes of abstention). Put simply, Al-Nashiri’s alleged harms are “attendant to resolution of his case in the military court system” and, as a result, do not render abstention inappropriate here. Councilman, 420 U.S. at 758, 95 S.Ct. 1300.
Moreover, even setting this clear proscription aside, the dissent’s argument that Al-Nashiri’s case could qualify for the “extraordinary circumstances” exception is unavailing. Focusing on the word “extraordinary,” the dissent makes a sympathetic ease that Al-Nashiri’s harms are different in both kind and magnitude from those that he would experience in a federal court or from the harms experienced by the average criminal defendant. But that alone does not bring those harms under the limited and narrow meaning of the exception. Although the dissent may be correct that Councilman itself had “no occasion to attempt to define those circumstances” that might be sufficiently extraordinary to warrant abstention, 420 U.S. at 761, 95 S.Ct. 1300, several subsequent cases have clarified the scope of this exception. See Ku-gler, 421 U.S. at 124, 95 S.Ct. 1524; Trainor v. Hernandez, 431 U.S. 434, 441-42, 442 n.7, 97 S.Ct. 1911, 52 L.Ed.2d 486 (1977); Moore, 442 U.S. at 433, 99 S.Ct. 2371. For a plaintiff to come within the exception, he must show both that he will suffer a “great and immediate” harm absent federal-court intervention and that the alternative tribunal is “incapable of fairly and fully adjudicating the federal issues before it.” Kugler, 421 U.S. at 123-24, 95 S.Ct. 1524. Al-Nashiri’s allegations regarding his treatment during his detention, while deeply troubling, do not provide any reason to fear that he will not be given a fair hearing in the military commission. See id. at 124, 95 S.Ct. 1524. Instead, Al-Nashiri’s allegations are about his particular vulnerabilities to a trial by a military commission at Guantanamo Bay. Because they say nothing about the competence of the military commission itself, those harms do not meet the requirements of the “extraordinary circumstances” exception.
The dissent responds that we need not feel bound by this precedent because Al-Nashiri’s case is different. The cases defining the “extraordinary circumstances” exception arose in the context of Younger abstention, not abstention in favor of courts-martial or military commissions, and therefore, the dissent contends, the definition of extraordinary circumstances articulated in the Younger cases does not apply in the military context.7 But Councilman is not as far removed from Younger as the dissent suggests. As we explained above, Councilman’s, abstention discussion is based on the same principles underlying Younger. See Councilman, 420 U.S. at 757, 95 S.Ct. 1300 (determining that Younger principles “apply in equal measure to the *130balance governing the propriety of equitable intervention in pending court-martial proceedings”). Accordingly, other circuits have concluded that Councilman is simply an application of the Younger doctrine to the courts-martial context. McCune v. Frank, 521 F.2d 1152, 1157 (2d Cir. 1975) (“Younger is not limited to criminal proceedings.” (citing Councilman)); Bowman v. Wilson, 672 F.2d 1145, 1156-59 (3d Cir. 1982); Lawrence v. McCarthy, 344 F.3d 467, 470 (5th Cir. 2003) (“The Supreme Court has since applied Wmwger-abstention in various other contexts, including that of Schlesinger v. Councilman....”); Hennis v. Hemlick, 666 F.3d 270, 274 n.5 (4th Cir. 2012) (“[T]he Supreme Court extended Younger abstention to restrict federal court intervention into on-going court-martial proceedings”). In following the lead of Younger and Councilman here, we heed the Court’s guidance that the exceptions it has crafted to abstention in favor of an ongoing criminal proceeding are narrow. See Huffman, 420 U.S. at 602, 95 S.Ct. 1200 (describing the “traditional narrow exceptions” to abstention doctrine). What the dissent proposes would redefine the scope of the “extraordinary circumstances” exception and create a novel free-floating exception for psychological harms. Such an approach belies the Court’s past treatment of the exceptions to abstention, and, as a result, we will not expand the “extraordinary circumstances” exception to include psychological harms that do not implicate the fairness of the military-commission proceedings.
Before moving on to Al-Nashiri’s other arguments, we again emphasize that Al-Nashiri’s sole claim in this appeal relates to whether the district court erred in declining to hear his challenge to the military commission’s subject matter jurisdiction. Al-Nashiri does not argue that Congress exceeded its constitutional authority in creating the military-commission system under the MCA or in defining “alien unprivileged enemy belligerent” in a manner that includes him. Nor, to repeat, does he contend that any procedures of the system Congress created in the MCA are unconstitutional or will prevent him from fully litigating his jurisdictional defense. He also makes no claim that delaying habeas review in his case amounts to an unlawful suspension of the writ. This is perhaps because the Supreme Court has explained in the court-martial context that “a deferment of resort to the writ until other corrective procedures are shown to be futile” is “in no sense a suspension of the writ of habeas corpus.” Gusik v. Schilder, 340 U.S. 128, 132, 71 S.Ct. 149, 95 L.Ed. 146 (1950).8 Indeed, Al-Nashiri does not dis*131pute that the MCA provides substantial “other corrective procedures,” including the right to appeal a conviction to our court. Finally, to .the extent that Al-Nashi-ri’s arguments regarding psychological harm challenge his treatment while in custody, nothing in our opinion forecloses him from challenging those conditions by filing a habeas petition in district court.
ii
Al-Nashiri next argues that post-trial Article III review will come too late'to vindicate his constitutional and statutory “right not to be tried” by a military commission that lacks subject matter jurisdiction over his offenses. See Councilman, 420 U.S. at 759, 95 S.Ct. 1300. The district court’s decision to abstain violated this right not to be tried, he contends. And because this right will be lost at the moment his trial begins, he argues that appellate review in our court cannot vindicate it.
In support, he points to the text of the 2009 MCA, which provides that an offense “is triable by military commission under this chapter only if the offense is committed in the context of and associated with hostilities.” 10 U.S.C. § 950p(c) (emphasis added). He asserts that the use of the word “triable” instead of “punishable” or “liable” suggests that Congress conferred a right not to be tried by a military commission at all, rather than merely a right not to be subject to a binding judgment by a commission. We understand his constitutional claim to assert something similar: the military commission has jurisdiction under Article I to try only war crimes, which by definition must have a nexus to hostilities. Whether Al-Nashiri locates his alleged right not to be tried in the MCA or the Constitution, the crux of this “right” is that Al-Nashiri is entitled to an initial determination in an Article III court of whether his military commission has jurisdiction over his offense. We disagree.
Some statutory and constitutional provisions indeed provide express guarantees that trial will not occur. In such cases, trial itself creates an injury that cannot be remedied on appeal. But only a handful of such guarantees have been recognized. The key question, then, is whether there is any express statutory or constitutional language that gives Al-Nashiri a right not to be tried, instead of simply a right not to be subject to a binding judgment, should his alleged crimes have taken place outside the context of hostilities. As the Supreme Court has explained it:
There is a crucial distinction between a right not to be tried and a right whose remedy requires the dismissal of charges. A right not to be tried ... rests upon an explicit statutory or constitutional guarantee that trial will not occur- — as in the Double Jeopardy Clause (“nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb”), or the Speech or Debate Clause (“[F]or any Speech or Debate in either House, [the Senators and Representatives] shall not be questioned in any other Place”).
Midland Asphalt Corp. v. United States, 489 U.S. 794, 801, 109 S.Ct. 1494, 103 L.Ed.2d 879 (1989) (internal citations and quotation marks omitted). The statutory language to which Al-Nashiri points might appear at first blush to create such an explicit guarantee: it describes when an offense is “triable” by military commission. 10 U.S.C. § 950p(c) (“An offense specified in [the MCA] is triable by military commission ... only if the offense is committed in the context of and associated with hostilities.”). But our case law demonstrates that the mere use of terms like “triable” does not transform a right not to *132be subject to a binding judgment into a right not to be tried.
Particularly instructive is our opinion in Khadr. There, we held that an erroneous jurisdictional ruling against a defendant in the military-commission system can be adequately remedied on appeal from final judgment, despite statutory language in the MCA that might suggest a defendant was not triable by military commission. Khadr, 529 F.3d at 1117-18. The presiding military judge in Khadr had determined that under the 2006 MCA, neither he nor the military commission’s members had the power to find the defendant an “unlawful” enemy combatant, as required for the military to have jurisdiction over the defendant. Id. at 1114. The military judge therefore dismissed the charges for lack of jurisdiction. On appeal, the CMCR held that the military judge could make the necessary jurisdictional finding and remanded accordingly. Id. at 1115. The defendant petitioned for interlocutory review of the CMCR’s decision.
This court rejected the defendant’s petition, explaining that the CMCR’s “procedural decision, as well as any subsequent jurisdictional decision, will be reviewable if necessary following a final judgment.” Id. at 1118 (emphasis added). We explained that “the denial of a claim of lack of jurisdiction is not an immediately ap-pealable collateral order” as the jurisdictional provisions at issue created a “right not to be subject to a binding judgment,” not a right to be free from trial altogether. Id. (quoting Van Cauwenberghe v. Biard, 486 U.S. 517, 527, 108 S.Ct. 1945, 100 L.Ed.2d 517 (1988)). And the right not to be subject to a binding judgment “may be effectively vindicated following final judgment.” Id. (quoting Van Cauwenberghe, 486 U.S. at 527, 108 S.Ct. 1945). Notably, Khadr dealt with language that could be read to suggest the existence of an express “right not to be tried.” See id. at 1114 (explaining that under the 2006 MCA, a military commission had “jurisdiction to try any offense made punishable by this chapter or the law of war when committed by an alien unlawful enemy combatant” (quoting former 10 U.S.C. § 948d(a)) (emphasis added)).
Our conclusion holds even if the military commission lacks subject matter jurisdiction not simply under the MCA, but instead under the Constitution. This much is apparent from Councilman. There, Councilman argued that his alleged offense was not constitutionally triable by court-martial because it was not “service connected.” Councilman, 420 U.S. at 741-42, 95 S.Ct. 1300; see also Solorio, 483 U.S. at 440-41, 107 S.Ct. 2924 (explaining, in overruling the “service connection” rule, that the rule was a “constitutional principle” interpreting Congress’s power under Article I); O’Callahan, 395 U.S. at 272-73, 89 S.Ct. 1683 (justifying the “service connection” rule by reference to Article I and the limits set. out by the Fifth and Sixth Amendments). And when the Supreme Court established the “service connection” rule, it spoke in terms of trial and not punishment. See O’Callahan, 395 U.S. at 274, 89 S.Ct. 1683 (holding, in establishing the “service connection” rule, that “since petitioner’s crimes were not service connected, he could not be tried by court-martial but rather was entitled to trial by the civilian courts” (emphases added)).- But the Court concluded in Councilman that any “service connection” deficiency could be adequately remedied after trial. 420 U.S. at 754, 95 S.Ct. 1300.
Al-Nashiri nevertheless gleans the existence of a constitutional “right not to be tried” from two cases in which the Supreme Court enjoined pending military trials: Reid v. Covert, 354 U.S. 1, 77 S.Ct. 1222, 1 L.Ed.2d 1148 (1957), and Hamdan. *133In both Reid and Hamdan, the Supreme Court heard pretrial habeas petitions .and found that the tribunal at issue — a court-martial in Reid, a military commission in Hamdan — lacked the authority to proceed. But we cannot infer from the mere fact of intervention before trial that a constitutional “right not to be tried” exists, much less one that extends to Al-Nashiri. And Al-Nashiri points to no pronouncement in Reid or Hamdan stating that a defendant has a right to have an Article III court determine in the first instance whether the military system has jurisdiction to try his offenses.
Instead, taking Hamdan first, Al-Nashi-ri observes that according to a plurality of the Justices, “deficiencies in the time and place allegations” against Hamdan signaled that the “offense [alleged] is not triable by law-of-war military commission.” Pet’r’s Br. 45 (quoting Hamdan, 548 U.S. at 600, 126 S.Ct. 2749 (plurality opinion)). Al-Nashiri apparently quotes this language to suggest that the reason the Court intervened pretrial was to vindicate a right not to be tried for offenses that were not “triable” by military commission. But the Court never said so. Instead, at other points in the opinion, a majority of the Court explained that it chose to intervene before the military commission issued a judgment because (1) no comity considerations justified abstaining under Councilman; (2) Hamdan “ha[d] no automatic right” to judicial review of the commission’s “final decision”; and (3) there was a strong reason to believe unlawful procedures would actually be used in Hamdan’s trial, because they were “described with particularity” in a presidential order and “implementation of some of them ha[d] already occurred.” Hamdan, 548 U.S. at 616, 126 S.Ct. 2749. In other words, the Court intervened because Article III appellate review was not available and no compelling considerations counseled in favor of awaiting the military commission’s judgment.
Al-Nashiri is correct, however, that Reid and similar cases suggest abstention is inappropriate where individuals raise “substantial arguments denying the right of the military to try them at all,” arid “the legal challenge turns on the status of the persons as to whom the military asserted its power” — that is, where “there is a substantial question whether a military tribunal has personal jurisdiction over the defendant.” Hamdan, 548 U.S. at 585 n.16, 126 S.Ct. 2749 (citing United States ex rel. Toth v. Quarles, 350 U.S. 11, 76 S.Ct. 1, 100 L.Ed. 8 (1955) (internal quotation marks omitted)). The precise contours of this “status” exception are unclear, but the Supreme Court has offered two examples of challenges that may come within its scope. First, where the military attempts to court-martial a defendant who is “undisputed[ly]” a civilian, the Court has intervened to prevent trial. New v. Cohen, 129 F.3d 639, 644 (D.C. Cir. 1997); see also Councilman, 420 U.S. at 759, 95 S.Ct. 1300 (citing Toth, Reid, and McElroy v. United States ex rel. Guagliardo, 361 U.S. 281, 80 S.Ct. 305, 4 L.Ed.2d 282 (1960)). In these cases, the “issue presented concerned not only the military court’s jurisdiction, but also whether under Art. I Congress could allow the military to interfere with the liberty of civilians even for the limited purpose of forcing them to answer to the military justice system.” Councilman, 420 U.S. at 759, 95 S.Ct. 1300. Requiring civilian defendants to first proceed through the military system would be “especially unfair” because of the “disruption caused to [their] civilian lives” and the accompanying “deprivation of liberty.” Id. (quoting Noyd, 395 U.S. at 696 n.8, 89 S.Ct. 1876). And second, the Hamdan Court suggested, in dicta, that the status exception might apply to Hamdan’s challenge, which alleged *134that his military commission was not “regularly constituted” under the Geneva Conventions. An irregularly constituted court is “ultra vires” and therefore necessarily lacks personal jurisdiction over any defendant, the Court reasoned. Hamdan, 548 U.S. at 589 n.20, 126 S.Ct. 2749.
Whatever the precise scope of this exception to abstention, it does not require that Al-Nashiri’s jurisdictional challenge first be heard by an Article III court. We do not understand Al-Nashiri to challenge his status as an alien unprivileged enemy belligerent who is subject to detention and to trial by military commission for certain types of conduct. Instead, he argues that the nature of his alleged offenses is such that the military lacks the authority to try them. His claim is therefore similar to that presented in Councilman, where the defendant did not challenge his status as a service member, but instead argued that the military could not try his offenses because they were not connected to his service in the Army. See 420 U.S. at 759-60, 95 S.Ct. 1300. Like the Supreme Court in Councilman, then, we conclude that this type of claim does not fit within an exception to abstention. Nor does Al-Nashiri argue that the commissions created by the 2009 MCA generally lack jurisdiction over defendants because they are so procedurally deficient that they are wholly ultra vires. The district court therefore did not err in abstaining from deciding Al-Nashi-ri’s pretrial challenge to the commission’s subject matter jurisdiction.
We recognize that our court’s opinion in Hamdan spoke of the status exception in broad terms. See Hamdan v. Rumsfeld, 415 F.3d 33 (D.C. Cir. 2005), rev’d on other grounds, 548 U.S. 557, 126 S.Ct. 2749, 165 L.Ed.2d 723 (2006). We suggested that the “theory” behind this exception “is that setting aside the judgment after trial and conviction insufficiently redresses the defendant’s right not to be tried by a tribunal that has no jurisdiction.” Id. at 36. But the Supreme Court’s subsequent opinion in Hamdan clarified that this exception to abstention applies to cases in which “the legal challenge turns on the status of the persons as to whom the military asserted its power.” 548 U.S. at 585 n.16, 126 S.Ct. 2749 (emphasis added) (internal quotation marks omitted). Thus, despite the broad wording of our statement in Hamdan, we cannot conclude that the status exception covers all non-trivial jurisdictional challenges that a military-commission defendant might raise. Indeed, such a reading would conflict with Councilman, which allowed a court-martial to go forward even though the defendant contested the tribunal’s jurisdiction to try the offense with which he was charged.
iii
Al-Nashiri also contends that intervention is required because his military-commission proceedings have been unreasonably delayed. He points to, the provision of the MCA that eliminates the UCMJ’s speedy trial guarantee, see 10 U.S.C. § 948b(d)(A), and notes that the government’s interlocutory 'appeals before the CMCR — and, as a result, his trial before the military commission — were stayed for nearly a year pending the confirmation of military judges to the CMCR. Al-Nashiri estimated in his briefing that trial will not commence until 2018 at the earliest. The government did not challenge this estimate at oral argument. Now that the CMCR’s stay has been lifted, the government has informed us that military-commission proceedings will resume in September 2016. Al-Nashiri’s counsel further estimated in rebuttal at oral argument that appellate review in this court will not occur until 2024. He provided no information, however, to explain why so much time would pass between trial and appeal.
*135We need not decide whether an unreasonable delay in military-commission proceedings could come within an exception to abstention. Cf. Nissan Motor Corp. in USA v. Harding, 739 F.2d 1005, 1011 (5th Cir. 1984) (explaining that “excessive delay causing significant impairment of constitutional rights” can counsel against abstaining in favor of an ongoing state proceeding). Although the stay before the CMCR delayed the processing of the government’s interlocutory appeals — and therefore Al-Nashiri’s trial — for nearly a year, Al-Nashiri never opposed this postponement. Indeed, when the government asked the CMCR to lift this stay after the confirmation of two military judges to that tribunal in April 2016, Al-Nashiri moved to continue the stay. We decline to label unreasonable or excessive a delay that Al-Nashiri has not contested. Cf. Sirva Relocation, LLC v. Richie, 794 F.3d 185, 196 (1st Cir. 2015) (noting that claims that a state proceeding is inadequate due to adjudicative delay are “undermine[d]” by a plaintiffs “failure to pursue potentially available state judicial remedies”). Nor has Al-Nashiri explained why the delay caused by the government’s interlocutory appeals was unreasonable or excessive. In fact, it was Al-Nashiri himself who argued that, in accordance with the Rules for Military Commissions and “basic equity,” the military-commission proceedings should be stayed while the government pursued its interlocutory appeals. Order, United States v. Al-Nashiri, AE340J (Apr. 10, 2015).
To be clear, we are troubled by the estimate of Al-Nashiri’s counsel that appellate review in this court might not occur until 2024. But counsel offered this prediction for the first time during rebuttal at oral argument, providing no information on the cause of this anticipated lag between trial and appeal to our court, and no opportunity for the government to respond. We are therefore not prepared at this juncture to forecast that any such delay will occur or be excessive as a matter of law. Should an unreasonable delay materialize, Al-Nashiri may pursue available remedies at that time.
Relatedly, Al-Nashiri suggests that where it is “plain” that the law of war does not apply, a district court should not abstain from adjudicating a military-commission defendant’s pretrial challenge, because requiring the defendant to first proceed through the military system “would serve no purpose other than delay.” Reply Br. 25 (quoting Strate v. A-1 Contractors, 520 U.S. 438, 459 n.14, 117 S.Ct. 1404, 137 L.Ed.2d 661 (1997)). But, as we explain below in rejecting Al-Nashiri’s mandamus petition, there is nothing “plainly]” erroneous about applying the law of war here. As a result, we take no stance on whether pretrial intervention would be appropriate — or, indeed, required — in such a case. Rather, we simply hold that in this case, the district court was not required, as a matter of law, to intervene.
Moreover, because the district court did not err in abstaining, we reject Al-Nashi-ri’s arguments that the court was obligated to rule on the merits of his petition for preliminary injunctive relief and that it abused its discretion by issuing a stay that mooted the request for injunctive relief. Abstention permits a court to decline to reach the merits of a petitioner’s claim. “It would be illogical for a federal court to preliminarily enjoin a [parallel] court proceeding when it [will] abstain from reviewing [that] proceeding altogether.” Phelps v. Hamilton, 122 F.3d 885, 891 (10th Cir. 1997). Accordingly, we affirm the district court’s treatment of Al-Nashiri’s request for injunctive relief.
Ill
We turn finally to Al-Nashiri’s mandamus petition. As we emphasized in *136rejecting his prior mandamus petition, mandamus is a “drastic remedy” that is appropriate only if three conditions are met. In re Al-Nashiri, 791 F.3d 71, 78 (D.C. Cir. 2015). First, the party seeking mandamus must have “no other adequate means to attain the relief he desires.” Id. (quoting Cheney v. U.S. Dist. Court for D.C., 542 U.S. 367, 380, 124 S.Ct. 2576, 159 L.Ed.2d 459 (2004)). Second, he must show that “his right to issuance of the writ is clear and indisputable.” Id. (quoting Cheney, 542 U.S. at 381, 124 S.Ct. 2576). And even if the first two conditions are satisfied, the court must believe “the' writ is appropriate under the circumstances.” Id. We deny Al-Nashiri’s petition because he has not met the high bar of showing a “clear and indisputable” right to issuance of the writ.
According to Al-Nashiri, it is “clear and indisputable” that his conduct did not take place in the context of hostilities, and therefore that he is entitled to mandamus relief. He contends that hostilities exist only when the political, branches say so in a “contemporaneous public act”; the existence of hostilities cannot be determined after the fact. And in his view, no contemporaneous public act established that hostilities existed either before September 11, 2001, or in Yemen, where his alleged offenses took place.
In fact, Al-Nashiri asserts, public acts at the time of his offenses suggested that America was at peace. He points to the President’s public statement, in response to the Cole bombing, that the nation was not at war. And while the President reported to Congress under the War Powers Resolution that he had introduced forces “equipped for combat” into Yemen after the Cole attack, he did not report that he had introduced forces “into hostilities.” Compare 50 U.S.C. § 1543(a)(1) (requiring the President to provide a written report to Congress if he introduces troops “into hostilities”), with id. § 1543(a)(3) (same if he introduces troops “in numbers which substantially enlarge United States Armed Forces equipped for combat already located in a foreign nation”). Further, the Federal Bureau of Investigation led the investigation of the Cole bombing, treating it as a crime scene rather than a combat zone. In Al-Nashiri’s view, these facts suggest that' the President did not believe “hostilities” existed around the time of the Cole bombing.
The government responds that the existence of hostilities is established by looking not merely to the contemporaneous acts of the political branches, but to a totality of the circumstances, including al Qaeda’s conduct. Implicit in this argument is the notion that the existence of hostilities can be assessed after the fact, at trial. Applying this totality-of-the-circumstanees standard, the government argues that the Cole attack was part of al Qaeda’s larger strategy to wage war against the United States, which culminated in the attacks of September 11. It notes that al Qaeda publicly declared jihad against the United States in 1996 and attacked the U.S. embassies in Kenya and Tanzania in 1998, and that after these bombings, the President ordered, missile strikes on al Qaeda training camps in Afghanistan and a chemical weapons facility in Sudan, and invoked the right to self-defense undér the United Nations Charter. The government also points to the MCA, which authorizes military commission jurisdiction for conduct occurring “before, on, or after” September 11, 2001. See 10 U.S.C. § 948d. To the government, this language suggests that Congress believed hostilities existed before September 11, even if no public act was taken until the passage of the AUMF on September-14, 2001.
*137The disagreement between the parties thus boils down to two central questions: Should the existence of hostilities be determined based on the totality of the circumstances, or only on the understanding of the political branches? And may it be based on a retrospective analysis, or only on what decisionmakers believed at the time of the events? Al-Nashiri and amici believe the judgments of the political branches at the time are what matters; the government takes a broader view.
Whatever the answers to these questions, they are not clear and indisputable, as the Supreme Court’s opinions in Ham-dan make clear. There, a four-justice plurality suggested that the conflict against al Qaeda began only after September 11, 2001, and the enactment of the AUMF. Hamdan v. Rumsfeld, 548 U.S. 557, 598-600 & n.31, 126 S.Ct. 2749, 165 L.Ed.2d 723 (plurality opinion) (questioning the legality of a charge encompassing acts from 1996 until 2001, since “the offense alleged must have been committed both in a theater of war and during, not before, the relevant conflict,” id. at 600, 126 S.Ct. 2749). The plurality may therefore have believed that some kind of contemporaneous public act of the political branches is needed to establish hostilities, although it did not expressly say so.
By contrast, in a dissent for three members of the Court, Justice Thomas argued that the judiciary cannot “second-guess” the Executive Branch’s view' expressed in its charging documents that an accused acted within the context of an armed conflict. Id. at 684, 126 S.Ct. 2749 (Thomas, J., dissenting). He further contended that the Executive’s “determination that the present conflict dates at least to 1996 is supported by overwhelming evidence.” Id. at 687, 126 S.Ct. 2749. In support, Justice Thomas cited much of the same evidence that the government relies upon here, including the 1996 declaration of jihad against the United States and the 1998 embassy bombings. See id. at 687-88, 126 S.Ct. 2749. The dissenting opinion therefore implies that a contemporaneous public act is not needed: al Qaeda’s actions, rather than only those of our political branches, could be considered in determining when hostilities began. Id. at 685, 687-88, 126 S.Ct. 2749. Justice Thomas’s argument that the Executive could determine when hostilities began in its charging documents is also inconsistent with the view that a contemporaneous act is needed.9
The debate in Hamdan indicates that whether hostilities against al Qaeda existed at the time of Al-Nashiri’s alleged offenses, and whether Al-Nashiri’s conduct in Yemen took place in the context of those hostilities, are open questions. And open questions are “the antithesis of the ‘clear and indisputable’ right needed for mandamus relief.” Al-Nashiri, 791 F.3d at 86.
The authority Al-Nashiri cites does not clear up this uncertainty. He points to cases emphasizing that the determination of when hostilities end is left to the political branches. See Ludecke v. Watkins, 335 U.S. 160, 170, 68 S.Ct. 1429, 92 L.Ed. 1881 (1948); Al-Bihani v. Obama, 590 F.3d 866, 874 (D.C. Cir. 2010). These cases do not, however, clearly establish that this political determination must be made in the *138form of a “public act” such as a proclamation or report to Congress. Nor do these cases speak directly to when hostilities begin. Al-Nashiri also relies on The Protector, 79 U.S. (12 Wall.) 700, 20 L.Ed. 463 (1871), which explained that it was “necessary ... to refer to some public act of the political departments of the government to fix the dates” of the Civil War. Id. at 702, 106 S.Ct. 2718; see also Masterson v. Howard, 85 U.S. (18 Wall.) 99, 105, 21 L.Ed. 764 (1873) (citing The Protector). But The Protector spoke only of the Civil War, 79 U.S. (12 Wall.) at 700; it did not purport to lay down a rule to govern future conflicts. As the Supreme Court later held, the terms “at war” and “at peace” may change meanings across contexts. Lee v. Madigan, 358 U.S. 228, 231, 79 S.Ct. 276, 3 L.Ed.2d 260 (1959). The Protectors reliance on a “public act” is therefore not clearly and indisputably applicable here. As a result, it cannot be grounds for mandamus relief.
Because Al-Nashiri cannot show that his conduct clearly and indisputably took place outside the context of hostilities, we deny his petition for mandamus relief.
IV
We deny Al-Nashiri’s petition for a writ of mandamus and affirm the district court’s denial of his motion for a preliminary injunction.
. We need not weigh in on whether the district court had subject matter jurisdiction to adjudicate Al-Nashiri's motion for preliminary injunctive relief. Although the government suggests in its briefing before us that Al-Nashiri’s claim does not sound in habeas — a claim that calls into question the district court's statutory jurisdiction, see 28 U.S.C. § 2241 (e)(2) — we affirm the denial of that motion for reasons we explain below. Because the motion was properly denied on threshold grounds, we need not consider the district court's subject matter jurisdiction any further. See Sinochem Int'l Co. v. Malay. Int’l Shipping Corp., 549 U.S. 422, 431, 127 S.Ct. 1184, 167 L.Ed.2d 15 (2007) (''[A] federal court has leeway ‘to choose among threshold grounds for denying audience to a case on the merits.’ ” (quoting Ruhrgas AG v. Marathon Oil Co., 526 U.S. 574, 58.5, 119 S.Ct. 1563, 143 L.Ed.2d 760 (1999))).
. Finality principles would normally prevent us from reviewing a decision to stay a case. But when the denial of a preliminary injunction — which is a reviewable final judgment, see 28 U.S.C. § 1292(a)(1) — is based on the decision to stay a case, we can review the propriety of the stay. See Privitera v. Cal. Bd. of Med. Quality Assurance, 926 F.2d 890, 892-93 (9th Cir. 1991). To treat a stay as unre-viewable under such circumstances "would mean that the denial of the preliminary injunction would be effectively unappealable because a reversal on that issue would have no effect.” Id. at 892.
. As an initial matter, we note that Al-Nashiri and the government disagree about the role that the hostilities requirement plays in the MCA. Al-Nashiri argues that the existence of hostilities is a.legal question that does not hinge on the facts proved at trial. For its part, the government contends that the hostilities requirement is a “necessary element of the offense with which he has been charged” that the government must prove at trial. We assume Al-Nashiri is correct that the hostilities requirement is a legal question going to the commission’s subject matter jurisdiction. Even so, as we will explain, the district court did not err in permitting the military commission to resolve the question in the first instance.
. Although the Court in Councilman assumed that the alternative judicial system at issue would adequately protect defendants' rights, we doubt that it would have reached the same result if the plaintiff had identified flaws in that system that would prevent him from fully litigating his defenses. Indeed, case law indicates that abstention is appropriate only where a plaintiff has "a full and fair opportunity to litigate” his claims in the alternative forum. JMM Corp. v. District of Columbia, 378 F.3d 1117, 1127 (D.C. Cir. 2004) (quoting Ohio Civil Rights Comm’n v. Dayton Christian Sch., Inc., 477 U.S. 619, 627, 106 S.Ct. 2718, 91 L.Ed.2d 512 (1986)); see also Browder v. City of Albuquerque, 787 F.3d 1076, 1084 (10th Cir. 2015) (Gorsuch, J., concurring) (explaining that abstention is inappropriate where state processes will not remedy the plaintiff's injury because they are inadequate either on their face or in practice).
. Habeas corpus "is, at its core, an equitable remedy,” Schlup v. Delo, 513 U.S. 298, 319, 115 S.Ct. 851, 130 L.Ed.2d 808 (1995), as is the injunctive and declaratory relief that Al-Nashiri's habeas petition requests, see Samu-els v. Mackell, 401 U.S. 66, 72, 91 S.Ct. 764, 27 L.Ed.2d 688 (1971). Thus, like the Court in Councilman, the district court here faced the question whether to exercise its equitable jurisdiction to intervene in a pending criminal prosecution. We assume that the form of relief Al-Nashiri seeks — a writ of habeas corpus — does not affect our analysis of the interests justifying abstention, and Al-Nashiri does not argue otherwise. Cf. In re Justices of the Superior Court Dep’t of the Mass. Trial Court, 218 F.3d 11, 17-18 (1st Cir. 2000) (collecting cases for the principle that “the federal courts have routinely rejected petitions for pretrial habeas relief” on Younger grounds, even though Younger dealt with a motion for in-junctive relief).
. By holding that an important countervailing interest justified the decision to abstain in this case, we do not suggest that a district court may always abstain from exercising its equitable jurisdiction simply because it perceives that some important' interest would be advanced by staying its hand. As the Supreme Court has made clear, abstention is appropriate outside the criminal context only in certain enumerated circumstances. See Sprint Commc'ns, Inc. v. Jacobs, — U.S.-, 134 S.Ct. 584, 593, 187 L.Ed.2d 505 (2013) (holding that Younger abstention does not extend to state civil proceedings merely because they implicate “important state interests” and provide an "adequate opportunity to raise [federal] challenges”).
. According to the dissent, Councilmans exception to abstention for "personal jurisdiction” challenges shows that we may consider other factors that the Supreme Court has not yet identified. But it is not clear that Councilman’s "personal jurisdiction” exception is unique to courts-martial, as the dissent suggests. Councilman grounded that exception in a right not to be tried, see 420 U.S. at 759, 95 S.Ct. 1300, which courts have recognized in other contexts as an “extraordinary circumstance” under Younger. See Gilliam v. Foster, 75 F.3d 881, 904 (4th Cir. 1996) (en banc) (holding that Younger abstention did not apply where plaintiff alleged potential Double Jeopardy Clause violations because “a portion of the constitutional protection [the Clause] affords would be irreparably lost if Petitioners were forced to endure the second trial before seeking to vindicate their constitutional rights at the federal level”).
. We take no stance on whether abstention could amount to a suspension of the writ, as this issue is not properly before us. But we observe that federal courts routinely decline to allow claims that can be raised in pretrial motions and addressed on direct appeal to instead be raised via pretrial habeas petition, whether trial is set to take place in federal court, state court, or a court-martial. See, e.g., Henry v. Henkel, 235 U.S. 219, 229, 35 S.Ct. 54, 59 L.Ed. 203 (1914) (federal prosecution) ("[T]he hearing on habeas corpus is not in the nature of a writ of error nor is it intended as a substitute for the functions of the trial court.... [A defendant] cannot, in either case, anticipate the regular course of proceeding by alleging a want of jurisdiction and demanding a ruling thereon in habeas corpus proceedings.”); In re Justices of the Superior Court Dep't of the Mass. Trial Court, 218 F.3d 11, 17-19 (1st Cir. 2000) (state prosecution) (“[T]he federal courts have routinely rejected petitions for pretrial habeas relief raising any variety of claims and issues.... Defendants are not entitled to consideration of their federal habeas claims until a time when federal jurisdiction will not seriously disrupt state judicial processes.” (internal quotation marks omitted)); Dooley v. Ploger, 491 F.2d 608, 610 (4th Cir. 1974) (court-martial prosecution) ("Before seeking [habeas] relief from a district court, [a defendant] must first exhaust his military remedies[.]”).
. The Hamdan dissent's suggestion that courts cannot question the Executive’s charging documents also puts to rest AI-Nashiri's argument that the military judge acted in a clearly unlawful manner when it denied Al-Nashiri’s motion to dismiss by, in part, deferring to the Executive Branch’s determination that Al-Nashiri’s conduct occurred in the context of hostilities. "Even if we ultimately agreed with [A]l-Nashiri on the merits," the military judge’s decision was not clearly and indisputably erroneous. Al-Nashiri, 791 F.3d at 86.