Last Wednesday, Feb. 28, a bipartisan coalition of senators introduced a proposed new joint resolution seeking to compel the Trump administration to end U.S. support for the Saudi-led campaign in Yemen. This resolution has already kicked off a new round of what is, in our view, a much-needed debate about the merits of U.S. involvement in the Yemen conflict, the humanitarian cost of which is increasing daily. Yet even before the text of the joint resolution was released, it drew criticism. In particular, Senate Majority Leader Mitch McConnell circulated a letter from the acting general counsel for the Department of Defense William Castle preemptively asserting that the joint resolution’s “fundamental premise is flawed” and that it “would not achieve its apparent purpose of restricting U.S. support to the [Saudi]-led campaign” even if enacted. In turn, Castle’s letter has raised new concerns that the Trump administration has embraced a broad view of presidential power, one that will allow it to proceed with its activities in Yemen even if Congress enacts legislation opposed to it.
Much of this debate, however, has overlooked the unique procedural context in which the joint resolution is being put forward, which is designed to ensure it can be made the subject of a vote even over possible opposition by others in the Senate. That said, the same procedure—which is tied to the 1973 War Powers Resolution (WPR)—limits the legal effect that the joint resolution can have. This context is important for understanding the Castle letter, which makes clear the executive branch’s view that the joint resolution will have no legal effect on current U.S. activities in Yemen if enacted but on narrower legal grounds than some fear. Further, while the joint resolution’s significance is likely to be primarily political, not legal, as a result, this context also points toward other avenues for advocacy that critics of U.S. involvement in Yemen may wish to pursue.
The Procedural Hook
The joint resolution on Yemen is designed to take advantage of special expedited procedures that Congress adopted in order to adapt the WPR (codified at 50 U.S.C. §§ 1541-48) to the Supreme Court’s 1983 opinion in INS v. Chadha. Section 5(c) of the WPR allows Congress to direct the removal of U.S. armed forces “engaged in hostilities” overseas without a declaration of war or statutory authorization through a “concurrent resolution,” a type of instrument that only requires approval in the House and Senate and does not go to the president for his signature or veto. This approach—known as a “legislative veto”—was used periodically by Congress as a way to try to increase its influence over the implementation of policy by the executive branch. In Chadha, however, the Supreme Court found a similar legislative veto in a different statutory context to be unconstitutional, raising serious doubts about section 5(c)’s own constitutionality.
Instead of amending section 5(c), Congress chose to respond to Chadha by enacting parallel procedures that would allow it to similarly direct withdrawals through joint resolutions and bills, which do not raise the same constitutional concerns post-Chadha as they are subject to presidential presentment and veto. These supplemental procedures were ultimately codified at 50 U.S.C. § 1546a. Specifically, they require Congress to consider any joint resolution or bill with the same objective as a concurrent resolution under section 5(c) of the WPR—namely, to require the removal of U.S. armed forces “engaged in hostilities” overseas without a declaration of war or statutory authorization—in line with the expedited procedures available for the review of arms sales, as set forth in section 601(b) of the International Security Assistance and Arms Export Control Act of 1976.
These expedited procedures, like similar ones created periodically since the early 1970s, represent a choice by members of Congress to increase their collective influence over executive branch actions by requiring some legislators give up part of their individual power within the chamber. They ease the process of bringing a joint resolution (or bill) to the floor, which can lessen the influence of committee members who might oppose it. If the committee to which the joint resolution is referred has not reported it out within 10 calendar days (subject to certain adjustments for adjournment periods), then any senator supporting the resolution may move to discharge it from the committee through a vote of the entire Senate. The motion to discharge is privileged and debate on it is limited to one hour, making it easier for a senator to force at least one vote related to the resolution. When combined with a privileged motion to begin debate on the resolution itself and an overall 10-hour cap on debate, this prevents a filibuster of the joint resolution or bill and reduces the ability of individual senators to obstruct it. Section 1546a also allows for such joint resolutions or bills to be amended and, in the event that they are vetoed by the president, sets out a condensed time-frame for considering a possible override in the Senate, conditions not usually present for other measures pursued through these procedures.
Notably, section 5(c) is not the only provision of the WPR that dictates when U.S. armed forces must be withdrawn. Section 4(a) requires the president to inform Congress in writing within 48 hours if U.S. armed forces are “introduced . . . into hostilities or into situations where imminent involvement in hostilities is clearly indicated by the circumstances[,]” among other situations, while section 4(c) requires a periodic report updating Congress on the status of such hostilities every six months. Section 5(b) in turn requires the president to withdraw U.S. armed forces from any such situations within 90 days if Congress has not enacted a declaration of war or statutory authorization, unless Congress is physically unable to meet due to an armed attack. Each of these provisions also plays a role in understanding the debate surrounding the joint resolution, as described below.
The Castle Letter in Context
This context is important for understanding both the proposed joint resolution itself and the Castle letter’s critique of it. The joint resolution explicitly states that it is being enacted pursuant to the section 1546a procedures. To ensure that it qualifies, it specifically directs President Trump “to remove United States Armed Forces from hostilities in or affecting the Republic of Yemen, except [where] engaged in operations directed at al Qaeda or associated forces,” within 30 days. This “remove . . . from hostilities” language directly parrots section 1546a, which itself parrots section 5(c) of the WPR. In this sense, the joint resolution is directing a specific set of actions defined in terms of the WPR.
This is why the Castle letter asserts that the joint resolution “would not achieve its apparent purpose of restricting U.S. support to the [Saudi]-led campaign” even if enacted. The joint resolution directs the president to remove U.S. forces “from hostilities.” But according to Castle the executive branch does not believe that U.S. support for the Saudi-led coalition constitutes involvement in “hostilities” for WPR purposes in the first place. The Castle letter describes U.S. involvement as being limited to the following, closely tracking similar descriptions by the Obama administration:
Since 2015, the United States has provided limited support to [Saudi]-led coalition military operations against Houthi and Saleh-aligned forces in Yemen. With the exception of a defensive strike in October 2016, U.S. forces are not taking direct military action in this Saudi-led effort in Yemen. Instead, the United States provides the [Saudi]-led coalition defense articles and services, including air-to-air refueling; certain intelligence support; and military advice, including advice regarding compliance with the law of armed conflict and best practices for reducing the risk of civilian casualties.
In this context, Castle asserts that the executive branch has long interpreted “hostilities” as used in the context of the WPR to mean “a situation in which units of U.S. armed forces are actively engaged in exchanges of fire with opposing units of hostile forces[,]” a threshold that U.S. support for the Saudi-led campaign—except in relation to the October 2016 incident described above—hasn’t crossed.
This position isn’t a new one; indeed, it’s reflected in the prior actions of both the Obama and Trump administrations. Except for the October 2016 incident mentioned above, neither administration has filed 48 hour reports related to U.S. support for the Saudi-led campaign under section 4(a) of the WPR as would be required if it believed current circumstances constituted “hostilities”. Nor were these activities mentioned in the periodic reports required under section 4(c) of the WPR—at least until the most recent December 2017 report, which included a note that “United States forces, in a non-combat role, have also continued to provide logistics and other support to regional forces combatting the Houthi insurgency in Yemen.” While a confusing departure from prior practice, it’s not clear that this reflects a change in view on what constitutes “hostilities”. In contrast, both the Obama and Trump administrations have routinely included direct action against al-Qaeda in the Arabian Peninsula (AQAP) and the Islamic State in Iraq and Syria (ISIS)—which the Castle letter acknowledges as constituting hostilities for WPR purposes—in their section 4(c) periodic reports.
If the executive branch were to accept that its support for the Saudi-led campaign in Yemen constituted hostilities, then it would conceding that it is already in violation of the WPR, both for failing to report its involvement in earlier reports and for allowing that support to continue past 90 days from when a report was first required. And if the Trump administration were to instead argue that it is not required to comply with these section 4 and section 5 requirements—for example, if it viewed the WPR as unconstitutional, as several presidential administrations have suggested in the past—then it would almost certainly consider itself not bound by the proposed joint resolution either. For its part, the Obama administration has stated that it sought to comply with the WPR’s reporting requirements, whatever constitutional reservations it may have had. And the Trump administration has showed no clear signs of moving away from this practice as of yet.
This does not mean that the executive branch’s interpretation of what constitutes “hostilities” is objectively correct. There is much reasonable space in which to criticize that interpretation and other relevant actors might find alternative interpretations more persuasive. The authors of the joint resolution themselves contend that section 8(c) of the WPR expands the scope of “hostilities” to include the types of support being provided to Yemen, an argument some have found persuasive. (For an excellent discussion of the merits of this position, see this recent analysis by Professor Oona Hathaway and Aaron Haviland.) Yet the fact that the executive branch has maintained its current interpretation across presidential administrations—and relied on that interpretation in pursuing actions that, in its view, put it in compliance with the WPR—means that it is unlikely to change its view anytime soon. Further, because the executive branch controls the deployment and activities of U.S. military forces—as well as the enforcement of federal law—its view of how to interpret relevant statutory requirements is likely to remain the operational one. And while federal courts could in theory step in if they disagreed and the issue presented itself in litigation, they have generally proven reluctant to do so.
The bottom line is that Castle is likely correct, at least in effect: even if enacted, the joint resolution will not end U.S. support for the Saudi-led coalition, because the executive branch views that support as entirely consistent with what the joint resolution requires. The silver lining, however, is that Castle’s letter is thus not the bold assertion of presidential authority that many fear it to be. He is not saying that Congress could never enact legislation that would require the executive branch to end support for the U.S.-led campaign in Yemen. Instead, he is simply asserting that the specific joint resolution in question—which is designed to fit through the narrow gates of the section 1546a procedures—does not require as much.
That Footnote, Though . . .
There is, however, one part of the Castle letter that arguably belies this narrow reading. Following the assertion that “President Obama directed . . . military and intelligence support [to the Saudi-led campaign] pursuant to his authority under Article II of the Constitution as Commander in Chief and Chief Executive and his authority to conduct foreign relations[,]” Castle inserts footnote 3, which states:
Because the President has directed U.S. troops to support the [Saudi-]led operations pursuant to his authority under Article II, and because the limited operation does not implicated [sic] Congress’s constitutional authority to Declare War, the draft resolution would raise serious constitutional concerns to the extent it seeks to override the President’s determination as Commander in Chief.
While far from a model of clarity, this footnote could be read as drawing a sharp binary between Congress’s constitutional authority under Article I’s declare war clause and the president’s Article II authority as commander in chief and chief executive. Such a formulation would seem to put any use of military force not within the purview of the former necessarily within the exclusive domain of the latter, making any legislative interference with such activities constitutionally suspect.
This would be an aggressive view of the president’s exclusive authority to direct military operations, as well as a substantial departure from the executive branch’s prior positions. Generally speaking, under Justice Robert Jackson’s widely-cited tripartite framework from his 1952 Youngstown Sheet & Tube Company v. Sawyer concurrence, the president may act in certain circumstances where congress is silent and the relevant division of constitutional powers between Congress and the president overlap or are unclear. But presidential authority is at its “lowest ebb” where he or she acts contrary to legislation, requiring that the president have some Article II power through which he or she can assert exclusive authority (thereby rendering any contrary legislation unconstitutional). The executive branch has generally accepted that overseas uses of military force fall into the former category of shared constitutional authority, even as it has argued that structural considerations and a long history of congressional delegation and authorization give the president wide discretion in pursuing the use of military force “at least insofar as Congress has not specifically restricted it” and so long as the duration and intensity of the resulting conflict falls short of those levels requiring Congress’s approval under the declare war clause. This is a far cry from arguing that the scope of the president’s Article II powers precludes Congress from enacting any limiting legislation whatsoever. Indeed, even where it has asserted exclusive Article II authority not subject to legislative limitations, the executive branch has tied it to the existence of a state of emergency or imminent threat to U.S. national security, not asserted it categorically over all uses of military force not subject to the declare war clause.
Fortunately, there are reasons to think that Castle might not have intended for us to read this footnote quite so broadly. Most notably, the notation for footnote 3 directly follows citations to two authorities that contradict this reading. The Supreme Court’s 1850 declaration in Fleming v. Page that, “[a]s commander-in-chief, [the president] is authorized to direct the movements of the naval and military forces placed by law at his command” is in the context of a congressional declaration of war; further, the rest of the opinion focuses on the limits on the president’s ability to use military force to pursue certain objectives overseas—in this case, the annexation of territory—absent congressional authorization. And the 1941 opinion of the Attorney General that Castle cites as support for the conclusion that the president’s authority “extend[s] to the dispatch of armed forces outside of the United States, either on missions of goodwill or rescue, or for the purpose of protecting American lives or property or American interests”—written by the same Robert Jackson who would later articulate the Youngstown framework—is predicated in part on both affirmative congressional support for such activities and the fact that “no statute . . . seeks to negative this authority in the President.”
Both opinions do, however, arguably support the view that the president as commander-in-chief has the sole authority to direct how military personnel are deployed and used within any broader framework of authorization, which may be the narrower point that Castle is attempting to make. Castle’s constitutional concerns may focus specifically on efforts to direct how the president chooses to use U.S. troops to provide “military and intelligence support”, especially if he or she is doing so in furtherance of his or her substantial authority over U.S. foreign relations. Further, such concerns would not necessarily extend to other U.S. conduct in support of the Saudi-led campaign. (Indeed, this would explain why the Castle letter goes on to identify various statutory authorizations pursuant to which it is supporting the Saudi-led campaign.) Of course, the exclusive scope of the president’s constitutional authority over foreign relations—particularly in the face of contrary legislation—is itself the subject of debate, and the extent to which it extends to conduct that may make the United States a party to an armed conflict or even violate international law is, in our view, questionable. Yet this framing would at least more squarely align footnote 3 with prior executive branch opinions and avoid a wholesale abandonment of the Youngstown framework.
Regardless, footnote 3 of the Castle letter is ambiguous and important enough to warrant clarification. Hopefully the Trump administration will take steps to provide this clarity, perhaps as part of its forthcoming report on the legal framework for national security operations. And if it does not, Congress should push the issue.
So What’s the Goal Here?
If the Castle letter is correct that the joint resolution is unlikely to have any direct impact on U.S. policy, then why are its sponsors pursuing it? The answer is most likely political. The section 1546a procedures being used make it easier for the joint resolution’s sponsors to force a vote—most likely on whether or not to discharge it from committee, not the joint resolution itself—beginning next week. Because Senate Majority Leader McConnell wields significant scheduling power in the Senate, the joint resolution’s sponsors would have difficulty getting a vote over his objections—which he clearly has—without these procedures. Forcing the vote will in turn require individual Senators to adopt a stance on the Trump administration’s continued involvement in Yemen, the outcome of which could prove politically damaging either for them individually if they support continued involvement in the increasingly unpopular Saudi-led campaign or for the Trump administration if they do not.
The sponsors may choose to go ahead and force the vote to impose political costs on the supporters of U.S. involvement in Yemen. Indeed, Sen. Chris Murphy (D-CT), one of the joint resolution’s co-sponsors, made use of similar procedures in relation to arms sales to Saudi Arabia in the past. Alternatively, they could attempt to leverage the threat of forcing a vote under section 1546a procedures in order to gain concessions from the Senate leadership on related items. This would resemble the path that the sponsors of a similar concurrent resolution in the House of Representatives—tellingly pursued under section 5(c) of the WPR despite post-Chadha constitutional concerns—followed last fall. They ultimately negotiated for a vote on, and won the eventual (overwhelming and bipartisan) passage of, a separate non-binding House resolution that raised various concerns with the conflict in Yemen and confirmed that Congress had not authorized U.S. involvement. It’s unclear, however, if Senate leadership would be receptive to a similar move without significant pressure from fellow Republicans.
And what if supporters of the joint resolution supporters still want to push for a legislative bar on U.S. participation in Yemen? Nothing prevents Congress from pursuing legislation that specifically and unequivocally prohibits the executive branch from providing any support to the Saudi-led coalition. Such legislation would no doubt raise constitutional concerns from executive branch lawyers, but it would at a minimum force the executive branch to make clear its views on the limits of the president’s exclusive Article II authority over the use of military force and may set up a situation where the courts would be more willing to intervene if the executive branch proceeds regardless. That said, such legislation could not move using the section 1546a procedures, making it easier for Senate leaders to block consideration. Hence, if they wish to pursue this route, opponents of U.S. involvement in Yemen will need to bring pressure on congressional leaders to allow a vote. Or their allies in Congress will need to wait for an opportune moment—such as the next defense authorization act or round of relevant appropriations—when they can incorporate their demand into a broader legislative package that will be more difficult to obstruct.
Whichever of these paths its supporters decide to take next, the joint resolution currently under consideration is unlikely to put an end to the controversy surrounding U.S. involvement in Yemen. But it may prove to be an important step towards eventually winning the political debate.