Can Congress Constitutionally Restrict the President’s Troop Withdrawals?

Ashley Deeks
Wednesday, February 6, 2019, 11:15 AM

As others have discussed on Lawfare, Congress recently has begun to feel its oats when it comes to U.S. foreign policy.

Secretary of the Defense Jim Mattis at Minot Air Force Base, N.D., Sept. 13, 2017 (Source: U.S. Air Force/J.T. Armstrong)

Published by The Lawfare Institute
in Cooperation With

As others have discussed on Lawfare, Congress recently has begun to feel its oats when it comes to U.S. foreign policy. In the wake of general dissatisfaction with President Trump’s decision to pull troops out of Syria, the Senate rebuked him by declaring that the Islamic State’s presence and activities in both Syria and Afghanistan continue to pose a national security threat. Though the measure warned against U.S. troop withdrawals, it did not mandate that the president cease withdrawing U.S. forces.

On the House side, though, Democratic Rep. Tom Malinowski and Republican Rep. Van Taylor introduced a bill that would, if enacted, impose such obligations. The “Responsible Withdrawal From Syria Act” would prohibit the use of fiscal 2019 funds to draw down U.S. forces from Syria below 1,500 troops, unless the secretary of defense, secretary of state, and director of national intelligence submit a report to Congress addressing a range of policy and threat-related issues. Malinowski also introduced a bill that would prevent the president from reducing the number of U.S. troops in South Korea below 22,000.

This raises an important question: To what extent can Congress constitutionally require the president to keep troops overseas? Even though the president has now suggested that he might either retain some troops in Syria or shift troops to Iraq (which would preserve their ability to respond to developments in Syria), and even though he seems to have changed his mind about withdrawing troops from South Korea, these questions may well arise again in short order.

Historical Precedent

Past practice is a good place to look in trying to evaluate Congress’s authority to prevent troop withdrawals. (For an extremely comprehensive account of congressional efforts to regulate executive military operations in a way that conflicts with executive preferences, see the two Harvard Law Review articles by David Barron and Marty Lederman. I draw most of the examples below from those pieces.) It turns out that there are not many cases in which Congress has mandated that the president affirmatively continue to engage in military activities in the face of a presidential interest in seeking to unwind those activities. There are some related cases and a few direct precedents, however.

Congress has, with some frequency, imposed temporal limits on the deployment of troops. For instance, Congress imposed an 18-month limit on the deployment of U.S. troops as part of a peacekeeping force in Lebanon in 1983, and in 1995 Congress’s Defense Appropriations Act prohibited the use of funds to maintain U.S. forces in Somalia after a certain date, except for personnel protection. Malinowski and Taylor’s bill presents the reverse situation, however—they are seeking to maintain military forces in Syria, not limit the presence of those forces—so these cases are imperfect precedents.

Congress also has occasionally required the president to employ certain types of military actors in certain ways. In an early case, Congress in 1791 gave George Washington an additional regiment but indicated that those troops were to be used to provide frontier protection. In 1909, President Teddy Roosevelt issued an executive order restricting the Marines to on-shore bases only, precluding them from manning naval vessels. Congress objected to this and enacted a statute requiring the president to deploy at least eight Marines for every 100 enlisted navy men on battleships and cruisers. This is close to forcing the president to deploy armed forces, but not precisely the same thing. The statute required the president to allocate some spots to Marines on certain ships, if and when he decided he wanted those ships manned. It did not actually require him to deploy those ships, even though he may have needed to do so as a practical matter.

In 1949, Congress mandated that President Harry Truman engage in certain defense spending. Truman had asked Congress to fund 48 Air Force groups, but the House insisted that he create 58. Truman signed the final bill, but, citing his commander-in-chief powers, ordered the secretary of defense to impound the extra funding for those additional 10 groups. The president thus resisted Congress’s attempt to impose additional personnel and structures on him.

More recently, President George H.W. Bush objected to a statute that would have restricted his authority to relocate defense personnel from an air base in Spain. He issued a signing statement noting his understanding that this provision did “not constrain my authority to deploy military personnel as necessary to fulfill my constitutional responsibilities as President and Commander in Chief,” —objecting as a constitutional matter to a statute that would have limited his ability to move armed forces from a U.S. air base.

Then there is Congress’s decision to require the president to maintain the detention facilities at Guantanamo Bay. These provisions (which first appeared in the 2012 National Defense Authorization Act and have carried across the years in slightly different forms to the current NDAA) make it virtually impossible for the president to release certain individuals held at Guantanamo, even though the executive has concluded that they no longer pose a threat. In signing the 2019 NDAA, the president noted, “Sections 1033 and 1035 purport to restrict transfers of detainees held at the United States Naval Station, Guantánamo Bay. I fully intend to keep open that detention facility and to use it, as necessary or appropriate, for detention operations. . . . I reiterate the longstanding position of the executive branch that, under certain circumstances, restrictions on the President’s authority to transfer detainees violates constitutional separation-of-powers principles, including the President’s constitutional authority as Commander in Chief.” The current administration followed the Obama administration in deeming this limitation unconstitutional, at least in certain circumstances. Just as the “Responsible Withdrawal From Syria Act” proposes, these provisions effectively would require the president to continue to engage in a certain type of military activity that he wishes to cease.

There is one example that is nearly identical to the South Korea bill’s requirements. Section 1264 of the 2019 NDAA prohibits the Defense Department from using funds to reduce the number of active-duty members of the U.S. armed forces in South Korea below 22,000. According to Section 1264, the provision can be waived if the secretary of defense certifies that the reduction is in the U.S. national security interest and will not significantly undermine the security of U.S. allies in the region, and that the United States has appropriately consulted with U.S. allies, including South Korea and Japan, about such a reduction. Here, too, the Trump administration included an objection in the signing statement, reiterating “the longstanding understanding of the executive branch that these types of provisions encompass only actions for which such advance certification or notification is feasible and consistent with the President’s exclusive constitutional authorities as Commander in Chief and as the sole representative of the Nation in foreign affairs.”

In short, there are a limited number of examples in which Congress has forced (or tried to force) the executive to affirmatively engage in military activity that it seeks to avoid. And at least in the modern cases, the executive has resisted, either in practice or in a formal statement. The dearth of precedent here is not surprising. Barron and Lederman unearthed a salient quote from University of Illinois Professor Clarence Berdahl, who stated that the idea that Congress would ever declare war over the president’s veto “cannot be imagined” because the “successful prosecution of a war would be impossible without the hearty cooperation of that department of the government which has in its sphere the actual direction and management of the war.” And yet the Guantanamo provisions, the 2019 NDAA provision related to South Korea and the proposed Syria and South Korea withdrawal restrictions head partway down that path, asking the president to affirmatively undertake military actions to which he objects.


The line between constitutional and unconstitutional statutes regulating the president’s war powers is a notoriously contentious one, though those contentions rarely are resolved in court. On the one side are those, such as Barron and Lederman, who think that Congress constitutionally can engage in a wide range of legislation to regulate military activity and that the executive has, for large parts of U.S. history, complied with those laws. Scholars in this camp might conclude that there is little difference between imposing troop limits on the president and requiring him to keep a certain number of forces in the field, or between telling the president that he may not use troops for a certain purpose and that he must use troops for a certain purpose. In all of those cases, they would argue, Congress is ordering the president to undertake a military action that he opposes—but is not unduly infringing on the president’s tactical commander-in-chief power.

On the other side are those who believe that the president has nearly preclusive commander-in-chief powers. Most modern presidential administrations have taken an assertive approach to war powers, claiming that the president possesses a broad set of exclusive powers. This began with the Truman administration in 1951, which argued that “since the direction of the armed forces is the basic characteristic of the office of the Commander in Chief, the Congress cannot constitutionally impose limitations on it.” Modern administrations have staked out similarly aggressive claims to that authority, even though they sometimes have chosen not to test the statutory limits that Congress imposed. This school of thought might be particularly troubled by the tension that these types of statutes might create for U.S. troops, who are being forced to stay in the field—and possibly to continue to fight an armed conflict—when their leader, the president, does not support their mission.

These two groups presumably come out differently on whether and to what extent Congress can limit the president’s ability to withdraw troops from overseas. Both sides might agree that there are some outer limits on congressional authority, such as a statute purporting to require the president to transfer 500,000 forces away from an ongoing exercise of self-defense in State X and to station them in State Y instead. Yet, as Scott Anderson noted last year, even the 2019 NDAA signing statement did not expressly note that the provision requiring the executive to maintain 22,000 troops in South Korea was facially unconstitutional. Instead, the signing statement focused on the imposition of the certification or notice requirement, and noted that the Trump administration would read those requirements to encompass only those actions for which such certification or notice is feasible and consistent with the president’s exclusive commander-in-chief authorities. Malinowski’s South Korea bill contains more expansive certification requirements than the 2019 NDAA, and so, if enacted into law, the executive might conclude that more of that bill’s requirements would fall within the president’s exclusive authorities. Likewise, the Malinowski/Taylor bill contains a long list of questions for the executive to address in a report before it could withdraw troops below the 1,500 troop level. But the executive might also conclude that some core of those statutes could be constitutional.


For now, President Trump may have obviated (or postponed) the conversation about U.S. troop withdrawals from Syria and South Korea. Further, it is unclear that these withdrawal bills would survive a presidential veto. Nevertheless, as Congress explores its powers to object to President Trump’s military decision-making, it is worth keeping in mind the historical precedents requiring that the president affirmatively undertake certain military activities over his objection.

The issue of whether and how to regulate troop withdrawals by statute may also re-emerge as a result of the upcoming U.S.-North Korea summit, if the executive considers offering U.S. troop withdrawals from South Korea as part of a future agreement. According to the 2019 NDAA’s Joint Explanatory Statement, the conferees believe that the “significant removal of United States military forces from the Korean Peninsula is a non-negotiable item” in negotiations with North Korea over its denuclearization. If such an agreement emerges from U.S.-North Korea talks, Congress’s ability to enact this principle into law and enforce it may be put to the test.

Ashley Deeks is the Class of 1948 Professor of Scholarly Research in Law at the University of Virginia Law School and a Faculty Senior Fellow at the Miller Center. She serves on the State Department’s Advisory Committee on International Law. In 2021-22 she worked as the Deputy Legal Advisor at the National Security Council. She graduated from the University of Chicago Law School and clerked on the Third Circuit.

Subscribe to Lawfare