The Yale Law Journal


The Insidious War Powers Status Quo

08 Mar 2024

abstract. This Essay highlights two features of modern war powers that hide from public view decisions that take the country to war: the executive branch’s exploitation of interpretive ambiguity to defend unilateral presidential authority, and its dispersal of the power to use force to the outer limbs of the bureaucracy.


American presidents have long found the means to use force abroad unilaterally, bringing the country to war or its brink on their own prerogative. They have typically justified their actions through expansive interpretations of extant statutory or constitutional authority. This executive-branch maximalism has come under attack from various angles—from originalists who see the modern balance of power as an upending of the constitutional order, from institutionalists who hope to see Congress play a more responsible role in world affairs, and from policymakers who would prefer the government prioritize its soft power toolbox. Those who support presidential unilateralism, by contrast, defend it as either constitutionally or pragmatically sound, and view congressional attempts to rein in the President’s use of force as either legally or logistically ill-advised. Professor Matthew C. Waxman’s essay, with which this piece is in dialogue, falls somewhere between these poles, but it ultimately views the modern status quo as less dangerous than characterized by its critics. My own view is that it is worse.

Efforts to tilt the power balance away from the executive branch and back toward Congress have focused largely on substantive fixes to two categories of statutes: The first are the specific delegations of authority aimed at particular conflicts, like the 2001 and 2002 Authorizations for Use of Military Force (AUMFs), war authorities Congress bestowed on the President to fight al Qaeda and the Taliban after 9/11 and to intervene in Iraq, respectively. The second is broader legislation aimed at resetting the balance of war powers as a whole, most notably the 1973 War Powers Resolution (WPR), which Congress passed in the wake of Vietnam to confront perceived abuses of presidential wartime powers and recapture a role for Congress in decisions about the use of force.1 Reformers have sought repeal of the 2002 AUMF for Iraq and repeal or amendment of the 2001 AUMF, on which presidents have relied in the years since 9/11 to use force against groups in Somalia, Syria, Iraq and elsewhere that in some cases did not even exist at the time the statute was passed.2 Others have directed their efforts at forward-looking reform, either to proposals to include sunsets and other measures in future AUMFs, or to reform the WPR itself to clarify its scope and extend its constraints on the President.3

There is significant merit in these reform efforts and I myself fall within this camp,4 but my most significant concerns about the modern war powers status quo center on problems that I locate outside of legislative text. These concerns lie in the heart of the executive branch itself—in its aggressive control over the interpretation of war powers authorities and in the sub-delegation and dispersal throughout the executive branch of these sprawling powers to use force.

Substantive statutory solutions like those proposed to repeal AUMFs or reform the WPR may be necessary but insufficient. The problems internal to the executive branch are deep, multilayered, and perhaps in some cases insurmountable. The executive branch is an unwieldy goliath and the process of decision-making within it often seems entirely opaque. Meanwhile, congressional reticence to tackle executive unilateralism stems largely from a mix of political apathy regarding the use of force, hesitancy by members of Congress to shoulder responsibility for matters where they see little personal electoral upside, and the American love affair with military uniforms. Even were Congress to enact one-time fixes now to any extant statute, this alone would not resolve the political pressures (or lack thereof) that keep members from more regularly engaging. Reining in the Department of Defense (DOD) in particular—bureaucratically or financially—does not tend to be a political winner. (Even former President Trump, who made no secret of his fever dreams about the “deep state,” exacerbated DOD’s bureaucratic independence by sub-delegating more power to lower-level officials and military officers.5) And if we cannot change the political realities for Congress, we can still work with where they are willing to apply pressure, and we may need to refocus efforts on executive branch interpretation of its authorities and intra-executive bureaucratic decision-making.

Professor Waxman argues that current war powers reform debates overemphasize war initiation and insufficiently acknowledge the roles Congress plays in managing the course of war. I agree that there are many roles Congress can (and does) play in managing war efforts, and that much more emphasis can be placed on Congress’s capacity to manage war holistically. But unlike Professor Waxman, I view the lead-up to war and its initiation, including the smaller scale uses of force the Executive might not label as “war” and even decisions about troop movement, as the most critical junctures for influencing the path of U.S. war powers decisions and precisely where reform efforts must focus. Once U.S. troops are committed and in harm’s way, it becomes infeasible for Congress to decline to support them.

As a result, my concerns with the focus on reforms to the WPR and the post-9/11 AUMFs lie not in their overemphasis on war initiation but rather in their fixation on a statutory solution to a mix of political and bureaucratic problems that, in my view, have primarily come to require an internal executive-branch resolution. As I discuss in Part I, lawyers within the executive branch—specifically the executive branch’s own arbiter of presidential power within the Department of Justice, the Office of Legal Counsel (OLC)—will read any statutory fixes Congress enacts through their longstanding lenses of statutory and constitutional interpretation. Those lenses have developed through decades of executive practice and government memoranda that rationalize an enormous and amorphous unilateral power for the President to use force. No war powers reform effort can succeed without a concomitant undertaking to understand and address executive-branch interpretation of statutory and constitutional war powers.

Furthermore, as I will address in Part II, this unilateral power to use force does not reside only within the President’s hands but is in fact dispersed formally and informally to lower-level officials throughout the executive branch. The combination of these factors both means that the executive branch as a whole wields significant and fluid power and that decisions to deploy that power are at times made not by the President and senior advisors but instead in the outer limbs of the executive branch. For those war powers reformists who favor congressional control over the President’s use of force for reasons of accountability, transparency, or opportunity for debate, there are arguments that these factors can be recreated to some degree through internal intra-executive dynamics. But that requires high-level and multi-agency engagement. Thus the fact that presidential war power is sometimes deployed without even the President’s control should be especially concerning. Unlike Professor Waxman, I lean toward thinking that this unilateralism is more dangerous than not. Addressing these hazards requires all hands on deck, and I propose here solutions for reform that task government critics, members of Congress, and executive-branch officials themselves with projects to reform the status quo.

I. the problem of interpretation: constraint as power

The practical exercise of war powers in the United States today cannot be ascertained by merely reading the Constitution or the texts of the WPR and legislative authorizations to use military force. Those texts suggest a dominant role for Congress in decisions about bringing the country to war, and a secondary role for the President who may use force only when the nation is actually under attack or Congress has otherwise authorized him to do so. But this description bears little resemblance to reality. To understand the President’s practical power to use force today, one must understand how the vast machinery of the executive branch interprets war powers authorities. Over the last several decades alone, the executive branch (across ideologically distinct administrations) has embraced an aggressive reading of the President’s authorities and a narrow reading of any constraints imposed by the Constitution, statutes, and international law. The result is that presidents across the ideological spectrum have embraced a presidential prerogative to use force abroad with little to no deference to their colleagues in the legislative branch.

Much of this is well-trodden territory. Some scholars have focused in particular on certain sweeping claims to virtually uncabined presidential power, such as those made in the early post-9/11 era seeking to justify everything from indefinite detention to torture to presidential power to launch a ground invasion of another country without congressional assent.6 These are extreme examples and have been met with such impassioned pushback that there is reason to consider them part of an executive-branch anticanon rather than precedent.

My concerns lie in the bread and butter of more pedestrian government lawyering, which has over time justified significant presidential unilateralism with a lot less drama.7 For any statutory resolution, Congress must grapple with the most slippery of these interpretive moves: over time, executive-branch lawyers manage to interpret constraints on the President—even constraints imposed by their own predecessors—as levers for expanding executive power. They do so through a combination of techniques, including embracing a permissive rule as the only limitation on the President’s power, while ignoring other more constraining checks. For their part, the courts and Congress have often acquiesced in this approach or failed to weigh in at all. Three examples illustrate this phenomenon, spanning constraints imposed by the Constitution, Congress, international law, and the Department of Justice itself.

A. Reading Authority into Constitutional Constraints

The Constitution, of course, makes the President the Commander in Chief and gives to Congress the power to “declare war” along with other war-regulating authorities.8 That much is clear, but after that point, much is up for grabs. It is well established that the President has some amount of constitutional authority to use force unilaterally. The Framers expected that the President could use force to repel a sudden invasion on the nation—which I will refer to as the “repel attacks” authority.9 The Supreme Court itself accepted in the Prize Cases that the President could prosecute a war that was not of their own making in the interest of defending the nation, and in fact that it was their responsibility to do so.10

There is a clear logic to lodging the authority to “repel sudden attacks” with the Commander in Chief despite leaving the decision to declare war with Congress. At the time of the framing, convening Congress was quite an ordeal. (Some might say it remains so today.) And the Framers would not have wanted to leave the President without recourse to defend the state against an existential threat while waiting for members of Congress to saddle up their horses. But over time, through executive practice, expanded legal justification, and at least perceived congressional acquiescence, this unilateral authority to repel attacks became untethered from that original compromise.11

Today, the executive branch claims an authority to act unilaterally under two different legal theories.12 The first, linked most closely to Republican administrations, can be classified as the “self-defense” theory. While in theory self-defense derives most clearly from the “repel attacks” authority, the self-defense justifications that the executive branch has espoused over the last several decades bear little resemblance to that early concept.13 The executive branch has extended that theory to justify a range of actions far beyond the immediate defense of the homeland, especially in the aftermath of 9/11, when OLC issued opinions endorsing unilateral presidential power to engage in “anticipatory self defense” in Iraq and to use force “preemptively against terrorist organizations or the states that harbor or support them, whether or not they can be linked to the specific terrorist incidents of September 11.”14 The implication of the first: the President could on his own authority order a full scale ground invasion of a sovereign nation that had not attacked us. The implication of the second: the President could use force anywhere in the world against any actor or state or take “whatever actions he deems appropriate to pre-empt or respond to terrorist threats.”15 These are aggressive readings of self-defense, no doubt, but they are familiar moves—taking a small grant of power and expanding upon it until it bears little resemblance to the original delegation. We see the executive branch make analogous moves regularly with statutes in the war powers space. As one example, the executive branch’s current interpretation of its 2001 statutory authority to use force extends to groups such as ISIS that did not exist until many years after that statute was enacted.

The more nuanced of the government’s theories legitimating unilateral action is what I will refer to as the “not war” or “not a war in the constitutional sense” theory. Under this second theory, the President is justified in using force unilaterally so long as it does not cross some constitutional line where Congress’s “declare war” power starts. In other words, the Constitution’s designation of Congress as the branch of government with the responsibility to “declare war” is read as an implied grant of power to the executive branch to do, at the very least, anything short of war. That is the first constraint-as-power move. Having embraced this argument, the executive must then determine what counts as “war” under the Constitution.

B. The Nature, Scope, and Duration Test

The “not war” theory embodied different parameters in its early incarnations within the government until it took on more definitive shape. It crops up in internal government memos from as early as 1965,16 in which we can spot the precursors of a test for determining when the “not war” theory applied. That test was finally articulated in a 1994 opinion by Walter Dellinger, then the head of OLC, the principal arbiter of presidential power questions for the executive branch.17 In that opinion, Dellinger first laid out what is now the deeply rooted “nature, scope, and duration” test for determining whether U.S. actions rise to “‘war’ in the constitutional sense.”18 At that time, Dellinger was writing against a backdrop of internal historical embrace by executive-branch actors of the President’s power to act unilaterally as long as he stayed below some “war” threshold. At what Professor Waxman calls the “high water mark” of executive unilateralism,19 the “not war” theory had been used to justify President Truman’s unilateral commitment of troops to support the United Nations’ (U.N.) campaign in the Korean War under the theory that a U.N. “police action” was not a “war” per se.20 By contrast, while the Dellinger memo did justify the unilateral use of force in a particular context—in that case to remove the military government which came to power in the 1991 coup in Haiti and reinstall the democratically elected leadership—it was also careful in crafting both requirements for and limits on the use of that test. In particular, the Dellinger test was notable for placing a clear constitutional constraint on the President’s unilateral power to use force, a recognition that the executive branch had only nodded at internally in the preceding era.21

Dellinger’s 1994 opinion represented a significant step for the executive branch: first in publicly recognizing a constitutional constraint on the President’s power to use force unilaterally, and second, in establishing a test that continues in use today for determining where that constitutional line is drawn.22 That test incorporated specific constraints: the “nature, scope, and duration of the anticipated deployment” would be carefully examined to ensure that it fell below the threshold of “‘war’ in the constitutional sense.”23 The opinion noted other factors that featured in its determination that unilateral action was justified, including that the deployment took place “with the full consent of the legitimate government of the country involved,” that it “did not involve the risk of major or prolonged hostilities or serious casualties to either the United States or Haiti,” and that there would be a limited risk of resistance. The opinion also took care to note the lack of “extreme use of force, as for example preparatory bombardment.”24

Over time, those factors have fallen by the wayside in subsequent discussion of and reliance on this test, and as a result the constraints built into the test have been eroded.25 Subsequent OLCs have deployed the 1994 “nature, scope, and duration” test to justify everything from the targeted killing of a foreign state’s general abroad,26 to the extended bombing of Libya in support of a multilateral mission that ultimately facilitated the overthrowing of a government,27 to air strikes in Syria in the interests of humanitarian intervention.28 A feature named as a particular constraining factor in one opinion—such as consent from the recognized government in the Haiti opinion, or the existence of U.N. Security Council authorization in the Libya opinion—later goes unmentioned in a similar future context where the government has not so consented or the Security Council has not authorized force, as seen in the 2018 opinion justifying air strikes against Syria.29

There are various means by which this dilution of constraints occurs over time. One is a problem inherent in any factors test. As discussed above, the 1994 Haiti opinion included several critical factors that were relevant to a determination that the President could use force without Congress, without stating that any one factor was necessary or sufficient. The next time OLC needed to make a determination, it may not have seemed a significant stretch to find authority when several if not all of the factors were present. But when this process is repeated and factors are often dropped and rarely added, the test is watered down over time.

I have written previously about a related cause of constraint-erosion: the norm against written redlines in these OLC memoranda.30 Legal advice denying the President authority is rarely memorialized—in part because if it staves off action, then the advice is no longer needed—and when OLC does enshrine its guidance in writing, it tends to restrict its response to only the question before it.31 The latter practice is familiar to lawyers as a tenant of judicial decision-making, and thus can be understood as connected to the cultural belief within OLC that it is the quasi-judiciary for the executive branch.32 But unlike OLC lawyers, judges do memorialize their “nos.” And the above two factors together mean that the OLC law of the executive contains few written redlines. I have described how this results in a slow accretion of executive power as follows:

Considering how lawyers address novel questions, by comparing new scenarios to old precedents, this lack of written redlines means that over time new lawyers have only those opinions authorizing action on which to build and draw analogies. One can readily envision how small incremental extensions of precedents in one direction can build over time to a significant expansion of power.33

Today the nature, scope, and duration test is deployed by government lawyers as a broad source of authority to act up until a bar that always seems to be just a little higher than the operation they seek to justify. The constraints that Dellinger and even his predecessors considered part of the test have been progressively eroded one by one such that they no longer serve as a meaningful constraint on the President’s unilateral use of force. Instead, today the nature, scope, and duration test is deployed primarily as a theory for justifying unilateral action.

C. Reading Authority into Statutory Constraints

On top of this broad understanding of the President’s constitutional authority, the executive branch has also interpreted its delegated statutory authority aggressively, either to assert additional layers of independent authority or as “belt and suspenders” to bolster its constitutional claims. Like the executive branch’s aggressive reading of its inherent self-defense authorities, its expansive interpretation of congressional authorizations to use military force over the last two decades has been a regular subject of critique.34 Any congressional attempts to dial back that power grab will need to contend with the executive branch’s interpretative moves that read significant authority into any text that can be claimed as ambiguous. But more insidious again is the executive’s ability to find power between the lines of statutory constraint.

For example, Congress enacted the WPR in 1973 against a political backdrop in which it sought to reclaim power in the wake of the Watergate scandal and subsequent impeachment of President Nixon, the Vietnam War, and rampant concerns that the President had been dishonest with Congress in seeking to expand and continue the war.35 The WPR was Congress’s attempt to recover its primacy over war powers, and it both delineates the “only” circumstances under which the President holds unilateral power—namely in the face of “a national emergency created by an attack upon the United States, its territories or possessions, or its armed forces”—and seeks to regulate those limited uses with additional constraints.36 It does not provide any additional power for the President to act unilaterally beyond his independent constitutional authorities. In fact, the WPR expressly states that it does not provide any independent authority to the President to introduce U.S. forces into hostilities.37 It could not be clearer in this regard. And yet a string of OLC opinions analyzing presidential war powers over several decades regularly dismisses Congress’s description of those limited uses—deriding it as a mere “policy statement”—and instead reads unilateral authority into the constraints imposed on the President.

According to OLC, Congress in the WPR “implicitly recognized th[e] presidential authority” to act unilaterally not “only” in the face of “a national emergency created by attack upon the United States, its territories or possessions, or its armed forces,” as Congress stated in that statute, but also in accordance with the ways the President has acted previously.38 This implicit recognition can be found, according to OLC, in the regulations the WPR imposes on the President’s use of force. These include the requirement that, “in the absence of a declaration of war, the President must report to Congress within forty-eight hours of taking certain actions, including introduction of U.S. forces ‘into hostilities’” or the imminent likelihood thereof.39 Another regulation the WPR imposes is colloquially known in war powers circles as “the clock”—the requirement that, absent congressional blessing, when the President uses force unilaterally, he “generally must terminate such use of force within sixty days (or ninety days for military necessity).”40

Within the text of the WPR, each of these requirements is a constraint on those already narrow authorities the WPR recognizes the President has constitutional authority to deploy unilaterally—namely, “only” when the country is under attack or Congress has otherwise authorized force. But OLC reads each of these requirements instead as acknowledgements of the President’s power to act broadly, subject more or less to these constraints alone. It is as if the initial “only” language laying out the extent of that power did not exist. In other words, the reporting requirement and the clock are interpreted as implicit legislative grants of or acquiescence in the President’s power to use force unilaterally at least up to sixty days.41 Congress’s attempt in the WPR to reset the balance of powers and to rein in the President’s unilateral use of force is thus understood by the President’s lawyers as acceptance of the President’s power to continue using force unilaterally well beyond the Act’s express narrow exceptions.

That itself might have been a reasonable interpretive move—regulation of action as recognition of authority to act—were it not for the fact that the Act specifically lays out those circumstances under which it understands the President as having the power to act unilaterally.42 And if OLC’s disregard for that statement as either “incomplete” or nonbinding is correct, then we might nevertheless consider the WPR’s later statement that nothing in the statute should be read to empower the President beyond his existing authorities as suggesting the WPR is not intended to be read as doing precisely that.43 Given that Congress both explicitly laid out those specific circumstances it viewed as falling within the President’s unilateral authority and reiterated that the statute should not be read as empowering the President, there is little argument in favor of reading its regulation of the President’s constitutional authority as condoning power beyond it.

D. Reading Authority into International Law Constraints

Finally, as I have explored in prior work, the executive branch at times invokes limits contained in international law as a means of enabling the President on the domestic front.44 The use of international law as an interpretive tool in understanding both constitutional and legislative text has a long pedigree going back to the Framers, despite apparent sentiment among recent Supreme Court nominees that they must denounce it in order to be confirmed.45 When considering what the Framers and ratifiers meant by “declare war” in the Constitution, for example, it is eminently reasonable to consider what they understood war to mean. And these were men who regularly supported their positions with references to leading international jurists of their time like Emmerich de Vattel.46

One might reasonably apply the understanding that international law informs the content of relevant legal terms of art to the modern day. For example, one might assume that when Congress authorizes the President to use force, it intends that the President do so against the backdrop of laws of war that the United States has long recognized (not to mention drafted and promoted) as regulating the conduct of hostilities. This is consistent with the Supreme Court’s longstanding Charming Betsy canon of statutory construction, under which statutes should be read not to conflict with international law whenever possible.47 Thus, for example, if Congress authorizes the President to use force, we do not take that authorization to include the power to raze a civilian city to the ground. If Congress authorizes the President to detain enemy soldiers, we might presume the intention is that they be afforded appropriate treatment under the Geneva Conventions and other applicable international law rules. In other words, the international law governing war may—and I would argue does—inform the outer bounds of the authority Congress intends to delegate to the President and regulate his use of that authority.

But this is not the only way the executive branch and others have looked to international law to inform the President’s authority. At times, and in a move similar to that described above with the war powers clock, executive-branch lawyers, judges, and others have looked to international law, which regulates the state as a whole, as the sole outer limit on the President’s authority, thus running roughshod over domestic separation-of-powers concerns or other domestic law constraints. As I discuss below, this may take the form of reading implicit additional authority into a domestic grant of power, or swapping out the more permissive rules of international law for a more restrictive domestic law regime, or even gesturing at a vague international law concept to provide a smokescreen for an amorphous power grab.

For example, in addressing the President’s detention authority in the conflict with al Qaeda, both the executive branch and courts have looked to international law to inform the President’s statutory authority under the 2001 AUMF. That statute explicitly authorizes the President to use force against al Qaeda, but it is silent on the question of detention.48 Thus, in the first 9/11-related habeas case to reach the Supreme Court, there existed no explicit statutory authority granting the President authority to detain in that conflict.49 Instead, the Court looked to the only statute on the books, the 2001 AUMF, and read into the text authorizing force the separate power to detain.50 In order to do so, the Hamdi plurality looked to prohibitions under international law—specifically the rule stating that detainees must be released at the end of hostilities.51 That rule provides an outer limit on states’ detention of soldiers in armed conflict. The Court took that outer limit on the state and read it into the domestic statute an additional grant of power to the President, holding that the AUMF gave the President authority to hold detainees up until the end of the conflict, the point at which international law would require their release.52

Similarly, the executive branch has turned to limits found in the international laws of war to entirely replace otherwise-applicable domestic law rules. In a 2010 memorandum justifying the targeted killing abroad of a U.S. citizen accused of plotting terrorist attacks against the United States, government lawyers argued that certain constitutional and statutory protections and prohibitions could be satisfied if—and only if—the individual was a lawful target as a matter of international law.53 In each of these circumstances, the government was not arguing that no law applied. Rather, it sought freedom to act up until the limits of what international law permitted. And while international law does impose constraints on a state’s wartime actions, these are inherently permissive rules. Unlike normal U.S. statutory or constitutional law, the international laws of war permit detention without trial. They permit killing that would otherwise be murder. Simply swapping out domestic constraints for international law in such cases—or interpreting domestic authority to expand to the limits of international law—often means relying on an extremely permissive legal regime as the only constraint on actions that would otherwise be heavily circumscribed.

And in the context of interpreting the breadth of groups covered by the 2001 AUMF, the executive branch has gestured at a concept of “co-belligerency” that it purports to draw from international law.54 Under the Executive’s theory, Congress surely intended the 2001 statute as authorization for the President to use force against not just those groups that planned the attacks of 9/11, but also any groups that joined them. Pushed to develop a more concrete theory, the government explained to courts and others that a concept drawn from the international law governing war between states should be imported into a conflict with a nonstate actor, and then should be used to inform—and in fact to expand—the President’s authority vis-à-vis Congress as a matter of the domestic separation of powers. If this were not sufficiently tenuous, the international law theory itself was obsolete. But the gesture at constraints drawn from a body of international law too obscure for the courts to examine closely was sufficient to get the courts off the government’s back, and the expansive interpretation stands to date.55

International law plays a critical role, of course, in regulating states at war, and this is particularly so in the interaction between states. But international law does not generally address the way states set up their legal systems internally, and it certainly does not speak to the proper allocation of power between the President and Congress. It may well be a useful interpretive tool for shedding light on the intentions of the Framers or the expected outer limits of a delegation of wartime power, but it is not a stand-alone replacement for domestic legal authorities and constraints. International law is not intended as a sole constraint on a leader’s domestic powers or as a means of allocating internal authorities, and thus it is inherently not up to those tasks. Furthermore, in the United States in particular, the executive branch exercises virtually total control over international law interpretation, especially in the wartime space, and it has been known to push the boundaries beyond what other states accept as black-letter law. The courts, Congress, and others charged with checking the executive branch frequently lack the expertise or the will necessary to understand let alone fully engage with the Executive’s interpretation. These features together make it that much more dangerous for the Executive to deploy international law as the only constraint on its action or as determinative of its power vis-à-vis Congress.56

Rarely do the interpretive moves I discuss in this Section represent bald power grabs alone. Assertions of power are typically interwoven with considerations of constraint. And it is in part for this reason that many of these moves fly beneath the radar. Even the lawyers pointing to constraints as authority may not realize the extent to which they are deploying the limits contained in statutes, the Constitution, or international law as affirmative power to act up to those limits. The executive branch’s use of constraint as power is a nuanced tool and one that develops and evolves as lawyers—often in perfectly good faith and under pressing circumstances—try to noodle out whether the President can take one more incremental step beyond what they have done before.

II. the dispersal of executive war powers within the executive branch

The risks of presidential unilateralism may appear problematic enough when the concern is that one person, the elected President, is taking upon himself the right to decide whether to bring the country to war. But while presidential “unilateralism” of this sort cuts out the collective decision-making body that is Congress, it is in fact less of a solitary endeavor than the word “unilateral” suggests. Most decisions made by “the President” go through an extensive bureaucratic interagency process before ever reaching his desk. Typically decisions to use force made at the presidential level involve a decision-making process run by the National Security Council with input from not only the Pentagon but the intelligence community, the Department of Justice, and the State Department, which brings foreign policy expertise, regional familiarity, and diplomatic considerations and reflection on potential consequences. Feeding into this policy process is an analogous interagency legal process known as the “Lawyer’s Group,” comprised of lawyers from these same agencies. Some have suggested that this internal process is a second-best replica of the traditional checks and balances of the courts and Congress.57

The problem of executive-branch unilateralism extends far beyond this paradigm of presidential decision-making. It is not always the elected and very public President of the United States and their senior advisers making these decisions with the benefits of vetting by expert agencies. Instead, certain key authorities to use force have been delegated to or claimed by lower-level decision makers dispersed throughout the national security bureaucracy. Those decisions can engender very little process, if any, throughout the executive-branch bureaucracy and often entirely shut out the relevant decision makers and critical experts from other key agencies who each bring important information and interests to bear—officials who would be involved were this decision actually happening at the level of the President. This dispersed decision-making is where the purest form of unilateralism lies, and it means that at times, a decision to act that could ultimately draw the country into war rests in the hands of a lower-level soldier in the field.

Debates over the allocation of war powers tend to conjure a binary choice between Congress and the President and thus miss this problem of intraexecutive power dispersal entirely. The question is typically posed as, “Should the President decide whether to bring the United States to war, or should Congress?” But executive-branch exercises of power are not all equal, and do not all involve the President issuing a directive or even making a single decision. Unilateralism by the executive branch includes not just presidential decisions but also decisions and actions taken by much lower-level executive-branch actors. From cyber operations to more traditional uses of force, from formal sub-delegations to organic power grabs, lower-level actors throughout the executive branch hold the power to take action that may bring the country to war without the President weighing in at all.

The dispersal of executive power down throughout the executive branch takes many forms. As I have discussed in prior work, executive-branch power is often “sub-delegated” or “sub-allocated”58 to lower-level officials within the executive branch.59 “These sub-allocations may be ephemeral or longstanding, formal or informal, intentionally created or organic, written or unwritten, and they may derive from presidents and other political actors both present and past.”60

Sub-allocations of power exist throughout the executive branch and are a necessary component of complex governing within the modern administrative state. It is self-evident that it is simply impossible for presidents or even their closest advisers to play a role in every decision that happens throughout the executive branch. It should not come as a surprise then that war powers decisions are also often dispersed. Yet discussions surrounding the allocation of war powers rarely grapple with the pragmatic reality that the Executive’s deployment of such powers is not always wielded by the President’s own hands.61 The risks of a lower-level unelected official bringing the country to war are not negligible. In fact, we are and have been at war initiated under just such circumstances.62 If a primary concern about presidential unilateralism is the lack of collective decision-making by elected officials, the threat of going to war on the sole decision of a commander in the field or an operator at Cybercom compounds these hazards exponentially.

Scholarship about executive-branch sub-delegation and sub-allocation is stymied by the fact that many sub-delegations in the national security space are classified.63 I draw my examples from entirely public unclassified sources, which necessarily constrains the discussion. This lack of transparency and concomitant difficulty engaging in open analysis is one of many challenges and hazards posed by this dispersal of the power to make war. I will focus here on two different forms of sub-allocations: formal and organic. Formal sub-allocations may (though need not) arise via explicit, written sub-delegations of authority. These are often signed off on by the President himself, or one of his predecessors, after deliberation at the highest levels of the administration. The Trump Administration engendered significant debate when it loosened the Obama-era requirement of a direct presidential decision for lethal actions taken outside active warzones on the basis that DOD should be allowed to act more quickly and without centralized review.64 Similar consternation about the lack of presidential accountability and interagency deliberation occurred when President Trump delegated the power to set force management levels in Iraq, Syria, and Afghanistan to the Secretary of Defense.65

In addition to these formal, top-down delegations of power, sub-allocations also occur when lower-level actors acquire power organically. One such example is the development of informal customary authorities that DOD and military officials have claimed for themselves, like the evolving doctrine of unit self-defense66 and the designation of partner forces as eligible for “collective” self-defense. The basic concept of unit self-defense holds that “a commander, or an individual soldier, sailor or airman [has] the automatic authority to defend his or her unit, or him or herself, in certain well defined circumstances.”67 The U.S. military’s standing rules of engagement, issued by the Joint Chiefs, assert that “[u]nit commanders always retain the inherent right and obligation to exercise unit self-defense in response to a hostile act or demonstrated hostile intent.”68 As I have elaborated elsewhere, such a broad automatic delegation of power provides “a significant mechanism for escalation of conflict, whether intentional or not.”69

The risk of escalation comes into sharp relief when one considers the fraught circumstances into which the United States has deployed and stationed (and continues to station) its troops for extended periods. The initial international law basis for deploying troops to a particular location abroad may be based in consent or national self-defense, and the domestic legal basis might be constitutional or statutory. But once there, U.S. troops stationed in, say, Iraq or Syria may (and have) become targets for a variety of actors against whom the President had not previously decided to use force.70 Under the concept of unit self-defense, U.S. troops may respond immediately “to a hostile act or [even] demonstrated hostile intent,” without necessarily running the decision back up the chain at home.71 This is not to say that the President and high-level decision makers in Washington are never involved in such decisions. For example, recent strikes against Iran-aligned groups in Syria and Iraq were reportedly ordered by the President in response to an attack in Jordan that killed three U.S. soldiers.72 The incredibly fraught array of policy considerations, expertise, and risk assessments that ideally feed into such a decision to use force—as well as how to do so and where—highlights just how significant it is that these decisions are ever made at the ground level. And this latest round of attacks is not happening in a void; rather this conflict is escalating against a backdrop of years of back and forth that has included decisions made by commanders in the field in the name of unit self-defense.73

The DOD concept of “collective unit self-defense” or “ancillary self-defense” attenuates the decision to use force further still from Washington decision makers. Under collective unit self-defense, U.S. forces may extend their authority to engage in unit self-defense to defense of “partner” forces—including both state and nonstate actors.74 The implications are enormous. There is no publicly available mechanism for designating a partner force as eligible for such U.S. protection, nor clarity on who within the executive branch would make such a determination in the moment.75 The implication is that an individual or commander stationed abroad as part of a particular mission could make a game-time decision to use force in defense of some other entity (be it a state force or some undefined nonstate actor) for purposes bearing little or no connection to the underlying U.S. mission, without any input from the President, the White House, other agencies, or even the civilian leadership at DOD, let alone Congress.

Some of these assertions of power are fairly recent. Collective unit self-defense is a relatively novel concept. Professor Oona A. Hathaway notes its use in 2017, “when the United States announced military operations in defense of the Syrian Democratic Forces (SDF), a coalition of ethnic militia and anti-government groups operating in North and East Syria with which the United States had partnered since 2015.”76 Its provenance likely dates back further, at least to the 2016 airstrikes against al Shabab in Somalia, in which U.S. forces defended the strikes as “unit self-defense—of both U.S. and [African Union Mission in Somalia] forces.”77 The al Shabab case provides a cautionary tale: within a few short months of taking these strikes in unit or collective unit self-defense, the United States had classified al Shabab as an “associated force” in the conflict with al Qaeda, thus bringing it fully within the ambit of war with the United States.78 In other words, the executive branch asserted the domestic legal authority to continue using force against al Shabab indefinitely, with no need for a specific authorization from Congress—from that point forward, it would read that authority into the 2001 AUMF.

Notably, while centralization in the President is seen as a more efficient means of action when compared with congressional involvement in war powers decisions, that very centralization of decision-making is viewed as a roadblock to action by those within the executive branch who favor speed. Presidential control is viewed as bringing with it interagency process and the need to grapple with differences of views among differing stakeholders, as well as the resulting delays that such process entails—in short many of the features often associated with congressional involvement. Thus, the dispersal of decision-making power away from the brain stem of the White House and out to the fingertips of the executive branch is internally viewed as a means of removing roadblocks to efficient action. Some illuminating examples are the Trump-era decisions to loosen presidential control over decisions about direct action.79 These procedural changes were hailed by those who felt that the prior rules unnecessarily tied the hands of commanders in the field and criticized by those who wanted to see greater accountability and diverse input.80

If we were to lay out actual war powers decision makers along a line chart and place the features of transparency, accountability, and consideration of diverse views on the left side of the chart and efficiency and speed on the right, we might place congressional control on the far left. But true Presidential control does not sit on the far-right side of our chart; rather on the chart of actual decision makers, Presidential control is in the middle. The far-right side of this spectrum holds the decision makers in the field as exercising the most efficiency and speed, and the least of the qualities like transparency, accountability, and wide airing of views. We can trace a historical path that has moved from congressional control on the far left, toward presidential control, and now may well be passing that center and heading toward the right end of the spectrum.

At the core of most war powers debates is a single question: Who is best positioned to make decisions about bringing the country into war? Those who favor the President when compared with Congress often prefer the alacrity and precision with which they imagine he can act. Those who favor Congress prefer deliberation and transparency and may also be hoping that additional time will impose some discipline on what should be a decision of last resort. But there is an underlying assumption in debates that focus on the President-Congress binary that the decision maker under either scheme would ultimately be an elected representative of the people. The existence of formal and organic sub-allocations of war powers indicates that this assumption is misplaced. And once the complex reality of internal executive-branch decision-making is laid bare, the factors favoring efficiency and operational flexibility suddenly shift from centralized power in the President to dispersed power down through the branches of the executive branch. And the reverse is true as well—the features of accountability, transparency, and diversity of perspectives feeding into a deliberative process suddenly favor presidential control when compared to that dispersed power.

III. can we change the war powers status quo?

Efforts to change the war powers status quo must contend not only with a recalcitrant Congress whose war powers muscles have long since atrophied. They must also grapple with the reality of aggressive interpretation and decision-making dispersal that is taking place virtually unchecked within the executive branch.

The solutions to problems of executive interpretation require a multifaceted approach with tasks for critics, for members of Congress, and for executive-branch actors themselves. Scholars and critical government watchers seeking to sway executive interpretation from the outside or inside when serving in an administration must be attuned to and persistent in calling out interpretive overreach. They must insist that government lawyers show their work, so to speak, but that alone is insufficient. They must also demand clarity on the outer limits of the authority that is being asserted. And they must remain vigilant in recognizing the gradual accretion of power through what might appear to be marginal steps, and not hesitate to push back on ratcheting up claims to power even when they may not appear extreme in any one instantiation.

Members of Congress and those focused on congressional reform must both undertake reforms to the current statutory framework in which they have delegated amorphous, indefinite power to the President, and engage with how executive-branch lawyers review congressional statutes in practice postenactment. They must find ways to continuously reassert Congress’s interpretive understanding via hearings, litigation, and resolutions, and yes, also statutory fixes that explicitly grapple with and reject specific executive interpretations. A one-and-done approach where Congress passes one piece of legislation and then ignores how it is interpreted for years to come entirely cedes the floor to executive-branch lawyers going forward.

The most significant challenges may be for executive-branch officials themselves to address, though they could be prompted along by political pressure from each of the above groups. If the executive branch continues to claim the mantle of decision-making over war, based on an often-plausible sense that it is the more responsible branch, then it is going to need to take on the features that are missing when Congress goes silent. Some of the most significant of these features include transparency, deliberation, and accountability.

The executive branch should—and outside actors should compel them to—publish memoranda not only defending their uses of force but also including explicit red lines explaining where proposed action would have been unlawful. This will better compel future executive-branch officials to constrain their actions within those redlines and permit both internal actors and outside government watchers to see whether they have in fact done so. For questions involving the jus ad bellum and other matters of international law, including where an international law theory is deployed to interpret domestic authorities, government lawyers with international law expertise—and in particular the Office of the Legal Adviser at the State Department — should have a decisive role and either sign off on any international legal theories in the Department of Justice’s memoranda or, better still, draft and publish their own memoranda on those matters.

These features of transparency and deliberation and especially accountability are also critical defects of the system of executive sub-delegations. Like the problem of interpretation, the dispersal of power inside the executive branch will require a multipronged approach to address. Here, Congress may find common cause with presidents who wish to impose greater top-down “presidential administration” over war powers decision-making.81 There is no world in which the President of the United States makes all decisions for the executive branch. But there are unlimited options for structuring the decision-making process inside the government, and for raising or lowering the involvement of high-level actors and relevant expertise. In prior work, I have examined tools Congress has to manage executive-branch decision-making in the foreign relations and national security space by creating and tweaking process requirements that empower or disempower different actors inside the executive.82 For formal sub-delegations of power, Congress might demand information through hearings, letters, and other means, regarding who within the executive branch has been authorized to make decisions that risk bringing the country to war. These should explicitly include operations and activities that the executive branch might interpret as falling below use of force, but which could nevertheless pose risks of escalation. Some of the examples they uncover may and should prompt reform efforts in themselves, geared either toward insisting that such decisions be made at a higher level within the executive branch, or that they involve engagement with Congress.

Dialing back the informal and organic dispersal of power is more fraught. DOD’s assertion of broad authority to engage in unit self-defense to protect troops in ways that could easily escalate into full-blown war will not be politically palatable for Congress to easily rein in. “Collective” or “ancillary” unit self-defense is much more attenuated and the easier of the two to tackle. At a minimum, Congress should insist on understanding the Executive’s decision-making process for designating groups that are eligible for collective self-defense, request an up-to-date list of any groups so designated as already required by statute, and most significantly, it should insist on presidential control over that process. This last would require a statutory fix from current reporting requirements that go so far as to acknowledge that these decisions may be made in the field by the “armed forces” and do not require direction or control from above. Congress could also simply prohibit the President’s unilateral use of U.S. troops for collective unit self-defense.83 This would not mean that the United States could never come to a partner force’s aid; it would simply require that the President come back to Congress for authorization before doing so, just as he would have to do if he were contemplating coming to the defense of a foreign nation, like Ukraine, that was the victim of aggression.

Actual unit self-defense—the ability of a commander or individual soldier to use force “in response to a hostile act or demonstrated hostile intent”—is naturally charged. Congress and the executive branch could both do a better job of defining and publishing the parameters and constraints on such action. But surely U.S. troops we send abroad need to be able to defend themselves. And it is no less true that certain contexts may involve discretion where the use of force could lead to unnecessary escalation. This is a foreseeable problem that should be considered and addressed before placing troops in such a situation. For this and other reasons, I agree with Professor Waxman that we need more congressional engagement on the leadup to war. In this case, I would specifically propose heavy engagement on where troops are stationed and when they are sent abroad, and then regular and periodic reassessments of the merits and risks of their continued placement. The WPR already requires reporting to Congress when combat-equipped troops are introduced into a new country or their numbers substantially enlarged,84 and this is one area where we should consider statutory fixes that would heighten Congress’s engagement in that decision and periodic review. As part of that engagement, Congress should interrogate the purpose of the troops being stationed abroad and the cost-benefit analysis to doing so, including factoring in the risks of starting or escalating a war, risks that may and do change over time and must be regularly recalculated. U.S. troops stationed abroad pose a constant risk of bringing us into conflict—not only because they may get attacked and prompt political backlash, but also because they may immediately strike back, thus making unilateral decisions about escalating a conflict at the micro level before the President has had a chance to consider the matter, let alone Congress or the American public.

As for executive-branch officials, the features of transparency, deliberation, and accountability counsel that, at a minimum, decisions that could take the country to war must come to the President. Unlike decisions made by an operator or a soldier in the field, decision-making at the presidential level, or even at the level of other key officials in Washington, typically prompts high-level interagency deliberation. That interagency consultation feeds into the President’s ultimate decision, or in cases of agreement sometimes stands in for it, and should be a regular process for decisions about the use of force. As discussed above, these processes are typically run by the National Security Council and will include the policy expertise of the national-security and foreign-policy agencies, as well as legal guidance from the “Lawyer’s Group”—the mix of high-level lawyers at the relevant national security agencies who meet weekly and whose legal advice feeds into the policy process. This process not only facilitates the airing of concerns and considerations of a range of experts and critical players; it also permits the executive branch to keep track of its reasoning and create a record that at least theoretically could be shared. It is this last piece that is honored more often in the breach, but greater transparency about the executive branch’s justifications for waging war is essential and the Lawyer’s Group should find a means of publishing its legal conclusions. Finally, the accountability prong militates unequivocally in favor of the decision to use force lying in the hands of, if not Congress, then at the very least an elected President.


The focus of the debate over war powers is, naturally, the allocation of the power to wage war and Congress’s ceding of that power historically to the President. In this Essay, I have sought to raise attention to two particularly insidious problems with the unilateral presidential use of force: the executive branch’s exploitation of interpretive ambiguity and its dispersal of the power to wage war to the outer limbs of the bureaucracy. The combination of these two features means that the vast majority of U.S. uses of force tend to fly under the radar for most of the American public. My primary concern is not that Congress as an institution somehow deserves a bigger hand on the wheel; in fact, they often seem to shun such a role. Rather it is that decisions about leading the country to war should be happening at the highest and most transparent levels of government, where decision makers can consider not only the exigencies of a particular exchange of fire but also world events and diplomacy and long-term risk, in a manner that is responsive and accountable to the people. Ultimately, determining the proper allocation of war powers is so imperative not because of the significance of constitutional intent or the balance between executive and legislative powers but because these are questions about bringing the country to war—literally life and death, the gravest choices a leader can make.

The decision to wage war should also be one that leaders would prefer to avoid. And in fact, the use of force is merely one narrow, particularly blunt instrument for managing conflict. The United States has other tools and might have significantly more and better tools if we had a responsible Congress invested in responsible foreign policy more broadly. We know where Congress has been on the questions of whether to go to war and whether to end it: largely absent or acquiescent. And perhaps that ship has sailed, never to return. Moreover, I take on board Professor Waxman’s excellent point that greater congressional involvement could in some cases escalate rather than de-escalate a conflict, in part by spreading the political risk. But it may not be too late, especially if the American public is itself tiring of endless war, for Congress to focus on investing in the tools necessary to prevent conflicts from boiling over. Where is Congress on soft power?

The Russia-Ukraine conflict has placed a glaring spotlight on a reality that foreign policy experts have long understood: not everything can be solved by dropping bombs. For decades we have focused on our hammers, and everything has looked like a nail. Military funding dwarfs the diplomatic arm of the state by orders of magnitude. At the same time, the Senate regularly holds up State Department confirmations—including, in recent years, of the Ambassador to China.85 We have now found ourselves facing a conflict that policymakers across the political spectrum understand we cannot fix by jumping in guns blazing. As a result, we are suddenly seeing the very people who spent years trying to defund and dismantle U.S. soft power and international institutions wondering, where are these international institutions and foreign policy tools and will they save us? This is precisely the strategic work of a State Department Legal Adviser, and yet that position is held up too; the Senate has failed to confirm one since 2018. Congress should do a better job of engaging with decisions about when the country goes to war, yes. It should also do a better job of helping the President keep the wars at bay.

Professor of Law, Cardozo Law School; Senior Fellow, Reiss Center on Law and Security at NYU School of Law. I recently served as the Counselor on International Law at the U.S. Department of State, but the views expressed here are my own and do not necessarily reflect the views of the U.S. Government. I am indebted to Tess Bridgeman, Ashley Deeks, Stephen Pomper, Jessica Thibodeau, and Matthew Waxman for excellent conversations and comments on drafts, to the Yale Law Journal editors for incisive feedback, and to Jake Dow for superb research assistance.