As a presidential candidate, Donald Trump vowed to pursue more aggressive detention and interrogation policies for terrorism suspects. He supported sending detainees to Guantánamo Bay (“We’re gonna load it up with some bad dudes, believe me”) and torturing suspects (“Don’t tell me it doesn’t work—torture works”). Yet for all of Trump’s bluster, his administration’s actions on those matters have proved decidedly modest. For example, an executive order signed by Trump in January generally preserved the status quo for the 41 foreign citizens still in U.S. military custody at Guantánamo, none of whom were sent there by Trump. For better or worse, the Trump administration has shown little interest in pushing the boundaries of who may be held, and under what conditions, in conjunction with the ongoing armed conflict between the United States and al Qaeda and its affiliates.

Yet despite the administration’s reticence, those limits are being tested by a lawsuit that has flown largely under the public radar. Known as Doe v. Mattis, it involves an unnamed American citizen who allegedly fought alongside the Islamic State (also known as ISIS) in Syria and who has been detained by the U.S. military in Iraq since September 14, 2017. It raises questions that go to the core of U.S. counterterrorism policy: Has Congress authorized the government to use military force against ISIS? Even if it has, can such force be used against American citizens? And the case has posed an even more fundamental question: For how long can the U.S. government manage to detain one of its own citizens without giving a legal rationale for doing so by impeding the courts’ ability to entertain such a lawsuit? If the tortuous path the case has taken so far is any indication, the answer is not reassuring.


One week after the 9/11 attacks, President George W. Bush signed into law the Authorization for the Use of Military Force, the statute that, to this day, provides the principal source of the U.S. government’s domestic legal authority to use military force against terrorists. Although the AUMF is often characterized as declaring war on terrorism, Congress was, in fact, far more nuanced. It empowered the president to use military force only against “those nations, organizations, or persons he determines planned, authorized, committed, or aided the [September 11 attacks], or harbored such organizations or persons.” Thus, the AUMF raised, but did not answer, two enormous legal questions: Which nations, organizations, or persons would fall within its aegis? And, less obviously, could the U.S. government use military force against American citizens? (The Non-Detention Act, a 1971 statute enacted partly in response to World War II–era internment camps, requires specific authorization from Congress before the government can detain its own citizens.)

In 2004, the Supreme Court took a half step toward answering the latter question when it concluded in Hamdi v. Rumsfeld that the AUMF did authorize the military detention of a U.S. citizen who had been captured during active combat in Afghanistan while fighting alongside the Taliban against U.S. forces. But Hamdi is a notoriously narrow ruling. As Justice Stephen Breyer (who cast the key vote in favor of the government in that case) wrote in 2014, Hamdi said nothing about detention of people (Americans or foreigners) captured outside Afghanistan, or whether, even assuming their initial detention was lawful, there were any legal constraints on how long the government could hold them.

Hamdi also did not address what has become the dominant question surrounding the AUMF: just how far beyond al Qaeda and the Taliban its authority stretches. Both the Bush and Obama administrations argued that Congress had authorized military operations against any “associated force” of al Qaeda, a reading that, by the middle of 2016, meant that the AUMF had been used to authorize 37 distinct military campaigns in 14 different countries. But that interpretation has never been blessed by the Supreme Court, and even if it is a fair reading of the 2001 statute, it doesn’t encompass ISIS, a group that broke from al Qaeda and thus can’t be said to have “entered the fight alongside” it, part of the official definition of “associated force” used by the Obama administration.

Instead, the United States has based its campaign against ISIS on the notion that the organization is a “derivative group” of al Qaeda and therefore within the ambit of the AUMF, much as a splinter group that broke off from Nazi Germany at the end of World War II would have been covered by the 1941 U.S. declaration of war against Germany. It’s not an implausible argument, but it’s also not self-evident (indeed, Congress separately declared war against each of Germany’s co-belligerents during World War II). That’s why, for the better part of the last four years, a bipartisan array of politicians, policymakers, and commentators (including President Barack Obama himself) have called for a new AUMF that would specifically identify those groups and individuals against which the United States may use military force and which would respond to the ways the antiterrorist campaign has changed since 2001. But despite widespread rhetorical support, such proposals have stalled in Congress, both because the devil is in the details and because there has been no pressing legal imperative for such a clarification. Thanks to Doe v. Mattis, that may soon change.


At the heart of the case is John Doe, a dual citizen of the United States and Saudi Arabia who was allegedly fighting on behalf of ISIS in Syria in September 2017 when he turned himself in to the Syrian Democratic Forces, a U.S. ally. The SDF promptly handed Doe to the U.S. military, which transported him to Iraq, where he has since been detained at an undisclosed location as an enemy combatant.

On October 5, the American Civil Liberties Union filed a habeas petition on Doe’s behalf. Although the ACLU had no relationship with him (and had no idea who he was), it argued that, because the government refused to identify Doe (and, therefore, allow lawyers to contact family members who could authorize such a suit), someone had to be allowed to ascertain whether Doe wanted to challenge the legality of his detention. The government objected, arguing that the ACLU was trying to bootstrap its way into court, and that even if Doe had a right to judicial review at some point he didn’t have it yet, because the government was still deciding what to do with him. On December 23, exactly 100 days after Doe was transferred to U.S. custody, D.C. federal district judge Tanya Chutkan agreed with the ACLU and ordered the government to allow ACLU lawyers access to Doe. Two weeks later, the ACLU reported back that Doe did indeed want to challenge his detention and that he wanted the ACLU to represent him.

With that procedural underbrush cleared, Chutkan may now finally be poised to decide the key question: whether the 2001 AUMF allows the use of military force against ISIS. Moreover, even if she rules that it does, Doe’s status as a U.S. citizen will complicate the matter, because it’s possible that, thanks to the Non-Detention Act, a clearer statement of Congress’ intent is required for the government to detain an American. Thus Doe v. Mattis could prompt Congress to finally revisit the AUMF and reassess the entire legal framework for U.S. counterterrorism policy.

For months on end, the U.S. government has detained one of its own citizens without having had to provide any legal rationale for doing so.

Perhaps wary of such a decision’s potential impact, the Trump administration has apparently been hard at work trying to moot the case by arranging to transfer Doe to foreign custody—possibly, given his dual citizenship, to Saudi Arabia. On January 23, Chutkan ruled that the government had to provide Doe and the court with 72 hours’ notice before transferring him, in case there were legal grounds on which Doe could object. The government is appealing that ruling (the appeal will be argued on April 5), but unless it is overturned, it seems likely that the district court will have to settle the merits of Doe’s detention, one way or the other.


In one sense, then, the true significance of Doe v. Mattis remains to be seen. It is still an open question what, if anything, the court will say about the AUMF’s applicability to ISIS and those fighting on ISIS’ behalf and whether, before the case gets that far, it settles the circumstances in which U.S. citizens suspected of terrorism can be transferred to foreign custody.

Yet Doe has already set an important and disturbing precedent. For months on end, the U.S. government has detained one of its own citizens without having had to provide any legal rationale for doing so, let alone one that could win in court. In ordinary criminal cases, the Supreme Court has long held that an individual must be presented before a neutral magistrate within 48 hours of his or her arrest. Two days may not be long enough for citizens captured on foreign battlefields, but Doe will shortly enter his sixth month of military detention without even a preliminary ruling from a federal judge as to its legality.

It’s easy to be unsympathetic to the plight of an American who left home to fight for ISIS in Syria. But without the judicial review that the government has spent the last six months impeding, what’s to say that that’s what really happened? And what’s to stop the government from pushing the boundaries even further next time? No matter what happens in John Doe’s case, the government’s ability to stall the legal process may be its most significant—and troubling—legacy.

You are reading a free article.

Subscribe to Foreign Affairs to get unlimited access.

  • Paywall-free reading of new articles and a century of archives
  • Unlock access to iOS/Android apps to save editions for offline reading
  • Six issues a year in print, online, and audio editions
Subscribe Now