Concept
Sharing War Power
War, defense, and foreign affairs powers
Sharing War Power
The Constitution gives the United States the complete power to make war, and it spreads that power across two branches. Article I, Section 8 lets Congress declare war, raise and support armies, provide and maintain a navy, and spend for the common defense. Article II, Section 2 makes the President commander in chief. The Necessary and Proper Clause and the nation's inherent authority over foreign affairs complete the grant.1 A grant to the nation does not say which branch exercises it when the nation must decide whether to fight, and the text answers that question only in part.
The gaps come from two features of the text. The Constitution is relatively inarticulate about the foreign affairs powers of the national government, so doubt arises about where specific authority resides. Some war and foreign affairs powers that it does spell out are shared jointly by Congress and the President, so each branch can claim a given decision as its own.2 Some of the resulting disputes have never reached a court, and others have been litigated without a definitive resolution.3 An answer must therefore be assembled from the constitutional text, from the statutes Congress has passed, and from the practice of the two branches.
The four subsections below build that answer in stages. The first covers the declaration power and the exception for sudden attack, the firmest points in the text. The next two cover the instruments Congress uses to consent to military action without a formal declaration, and to restrain it once it begins: force authorizations and the War Powers Resolution. The last covers control of funding, Congress's most practical check on a war.
Declaring War and Sudden Attack
Article I, Section 8, clause 11 gives Congress the power “To declare War,” and Article II, Section 2, clause 1 makes the President “Commander in Chief of the Army and Navy of the United States.”4 Together the clauses set two rules. The decision to initiate war or hostilities belongs to Congress, and the President may act in an emergency to defend the nation without first obtaining congressional approval.5 Each rule has its own reason, and the hard questions in this field arise at the line between them.
The first rule answers the question of who should be able to commit the nation. The Founders gave the declaration power to Congress deliberately, remembering how easily the English Crown could lead the nation into war without the consent of Parliament.6 Ely argued that the stake is the judgment that “no single individual should be able to take the nation into war and thereby risk the lives of all of us, especially our young people.” A Congress that has not unequivocally authorized a war at the outset, he added, is “a good deal more likely later to undercut the effort.”7
The drafting history explains the second rule. An early draft of Article I, Section 8 would have given Congress the power to “make” war. Madison's notes on the Convention show that the verb was changed to “declare” so as to “leav[e] to the Executive the power to repel sudden attacks.”8 The change turns on what each verb covers. A power to make war would reach everything in a war, including the first moment of resistance to an invasion, while a power to declare war reaches the decision to begin one. An invasion does not wait while two houses assemble, debate, and vote, so a rule that held back every defensive use of force until Congress acted would leave the nation defenseless in the one situation where speed decides the outcome.
In Prize Cases, the Court applied that understanding. The decision makes clear that the President can respond to attacks on the United States without awaiting congressional authorization, because the President “does not initiate the war, but is bound to accept the challenge without waiting for any special legislative authority.”9 The verb initiate carries the holding. The nation had already been attacked, so the President's response answered someone else's war and began none of his own. The Court has given little guidance about presidential use of force under other circumstances.10
Two flaws in the President's position correspond to the two rules. First, the attack, even if sudden and unexpected, struck a French island and not the United States, its territories, or its armed forces, so the deployment falls outside the emergency exception that the Founders contemplated and the Court recognized in The Prize Cases.11 Second, the treaty cannot carry the weight placed on it. Clause 11 assigns the declaration power to Congress, meaning the House and the Senate, and a treaty confirmed by the Senate alone cannot take the place of a congressional declaration of war.12 A declaration needs two chambers and a treaty needs one, so treating the treaty as a substitute would drop the House from the decision to go to war. Defense treaties often recognize the point. The Southeast Asia Collective Defense Treaty, for example, committed the United States to respond to an attack on a signatory “in accordance with its constitutional processes.”13
Between a response to an attack on the nation and a war the President starts alone lies a middle ground that the text leaves open. Even if the President holds an inherent power to defend the nation, does that power include offensive strikes? Can defense be construed so broadly that it reaches foreign military actions taken to “defend” the nation's allies?14 The drafting history speaks of repelling sudden attacks, and each question asks how far that phrase stretches.
Practice has moved toward the President. With increasing frequency, Presidents have sent the armed forces into hostilities on their own initiative, even where there was no pretext of defending the United States against sudden attack, and a 1967 Senate Foreign Relations Committee study found that “[t]he use of the armed forces against sovereign nations without authorization by Congress became common practice in the 20th century.”15 Congress has often done little to push back, and courts have dismissed the vast majority of constitutional challenges as nonjusticiable under principles of standing, ripeness, or the political-question doctrine.16
Commentators divide over whether courts should police the line. Ely advocated a strong judicial role to rein in the executive and so induce Congress to discharge its constitutional responsibilities. Sofaer replied that efforts to have courts pass on the legality of executive decisions to involve the nation in hostilities would generally be futile and harmful, because judicial involvement tends to create hard-and-fast rules in intensely political areas of overlapping authority. He added that a President does not act irresponsibly merely because Congress has not specifically authorized an action, so long as the President acts in the absence of legislative direction to the contrary.17 Because courts have mostly declined to decide, the political branches work the dispute out themselves, through force-authorization statutes, the War Powers Resolution, and the control of funding.
Authorizations for Use of Force
Congress now usually consents to military action by statute and not by formal declaration. Particularly since World War II, it has adopted “force authorization” statutes that authorize the President to engage in specified military activities without explicitly declaring a “war” with another nation.21 The best-known modern example followed the September 11 attacks. Section 2(a) of the 2001 resolution gave the President authority “to use all necessary and appropriate force against those nations, organizations, or persons he determines planned, authorized, committed, or aided the terrorist attacks that occurred on September 11, 2001, or harbored such organizations or persons, in order to prevent any future acts of international terrorism against the United States by such nations, organizations or persons.”22
A statute can serve in place of a declaration because the declaration power exists to make sure that the decision to fight comes from the legislature and not from one person. A force authorization is enacted by that same legislature, so the allocation the Founders chose is honored whatever label the document bears. Because the statute authorizes specified military activities, Congress can also shape its consent to the circumstances.
Hamdi v. Rumsfeld shows an authorization operating as a source of power. A U.S. citizen captured as an enemy combatant was held in custody in the United States, and the Court held that the September 11 resolution was sufficient authority to hold him, though he was entitled to a procedural due process hearing to determine his status.23 The first half of the holding treats the authorization as the statutory basis for military action. The second half shows that an authorization supplies power and leaves in place the constitutional limits on how that power may be used against a person.
How far consent can fall short of a formal authorization is less settled. Two questions illustrate the gap: whether the Gulf of Tonkin Resolution was sufficient to constitute a declaration of war for Vietnam, and whether repeated congressional approval of funding could suffice without a resolution approving the war itself. The Supreme Court has never clearly answered either one, and because it treats such foreign policy disputes as political questions, answers are unlikely to come from the judiciary.24 Courts generally dismiss constitutional challenges to presidential uses of force as nonjusticiable, so the working law of authorization has been built mostly by the two political branches.25
Breadth is the risk. If a resolution leaves every choice to the President, the decision to commit the nation has in substance returned to the executive, and the legislative check that justified treating a statute as consent survives only on paper. The 2001 resolution shows how open-ended these instruments can be, since its reach turns on whom the President determines to have planned, authorized, committed, or aided the attacks or to have harbored those who did. Modern Presidents have also assigned a sweeping meaning to force-authorization statutes, some of them quite broadly worded, in claiming power to commence and conduct military actions.26 Section 2(b) states that the resolution constitutes specific statutory authorization as the War Powers Resolution defines it, and whether it is the sort of open-ended resolution that statute envisioned remains a fair question.27
One way to press the breadth objection asks whether the authorization contains articulated standards that confine the President's choices. That is an argument for separating a genuine congressional decision from a handoff of the decision, and no court has adopted it as a rule for war authorizations. The 2001 resolution at least gives something to measure against, since it identifies its targets by conduct tied to the September 11 attacks and states a purpose, preventing future acts of international terrorism, although the President determines who meets the description.
Whether to fight, against whom, for what end, and for how long all rest on the President's own determination, so the argument that Congress has handed over its power and has not exercised it is strong. A resolution that named the adversary, described the purpose, and fixed a period would weaken the argument, because Congress itself would have made the choices that matter.
An essay should keep two questions apart. Whether Congress has consented at all is answered by looking for a statute, either a declaration or a force authorization, and Hamdi treated the 2001 resolution as sufficient authority for detention. Whether the consent is so open-ended that it fails to be Congress's own decision is a delegation question, argued by asking whether the authorization contains articulated standards.
War Powers Resolution
Congress enacted the War Powers Resolution of 1973 over President Nixon's veto.30 It is a federal statute and not just a congressional resolution, and it is unusual: a structural, quasi-constitutional law that focuses on processes and relationships and not on substantive policy.31 It fills a gap, because the Court has given little guidance about presidential power to use armed force outside a response to attack.32 Its stated purpose is to fulfill the intent of the framers and ensure that the collective judgment of both Congress and the President will apply to the introduction of United States Armed Forces into hostilities, or into situations where imminent involvement in hostilities is clearly indicated by the circumstances, and to the continued use of those forces.33
The statute states Congress's view that the President's constitutional powers as Commander in Chief to introduce forces into such hostilities “are exercised only pursuant to (1) a declaration of war, (2) specific statutory authorization, or (3) a national emergency created by attack upon the United States, its territories or possessions, or its armed forces.”34 The first two bases are the two forms of congressional consent, the second being the force-authorization statute that Congress has used more often than formal declarations since World War II.35 The third basis tracks the sudden-attack exception. It lists attacks on the United States, its territories or possessions, and its armed forces and says nothing about attacks on allies, the very question the previous subsection left open.36
Congress chose a process over a flat prohibition because of the position a deployment creates. Once the President introduces troops into a conflict, Congress must choose between funding the intervention and cutting off funding to an army in the field, which can be a politically suicidal act, and the Resolution is intended to avoid that predicament.37 It therefore attaches a clock to the introduction of forces and makes withdrawal the default when the clock runs out. The statute in effect countenances a temporary period of unilateral presidential engagement, subject to two restraints: Congress alone can order a stand-down by concurrent resolution at any time, and any continuation past the prescribed window requires affirmative legislative approval.38
The French-island deployment from the previous subsection shows the clock running. Suppose the President sent the forces on March 1. A report was then required by March 3, and the sixty days ran out on May 2. If unavoidable military necessity required more time for a safe withdrawal, thirty more days carried the deadline to June 1, unless Congress meanwhile authorized the action or extended the period.45 The clock started whether or not the President gave notice, because the statute runs the period from the date a report is submitted or required to be submitted.46 If the period began only on filing, a President could keep it from ever starting by withholding the report.
The first episode turns on what the forces are doing, since the supply and training exception looks to the character of the mission, and the detachment's entry into foreign territory does not by itself require a report. The response in the second episode rests on the third basis, a national emergency created by an attack on the armed forces, so neither a declaration nor a specific statute is needed to authorize it. A report is still due, because the statute requires one whenever forces are introduced into hostilities in the absence of a declaration of war.47 Finding a national emergency therefore resolves only the authority question, and the consultation, report, and sixty-day steps must still be run.
President Nixon vetoed the Resolution on two constitutional grounds, and the two have fared differently. His first objection, which looked to Congress's acquiescence in past practice, was dubious, because the Supreme Court has noted that separation of powers violations cannot be waived even if the branch whose authority was encroached upon acquiesced in the intrusion. Even if Congress had over the years permitted the President to usurp the power to declare war, that would not allow the executive now to claim the power as its own. A longstanding practice may still help a court define the acceptable boundaries between the branches.48
His second objection, which concerned the legislative veto, was well taken after the Court struck down the legislative veto in Immigration and Naturalization Service v. Chadha.49 A concurrent resolution is passed by both houses but is not presented to the President, so it escapes the veto.50 The Chadha opinions list the Resolution's provision for removal of forces by concurrent resolution among the legislative-veto statutes.51 The weakness is confined to the one mechanism that lets Congress bind the President without his participation. The other provisions rest on a statute that Congress enacted over the President's veto, so the presentment objection has less force against them.
No court has definitively resolved whether the statute's requirements are constitutional or effective.52 Challenges to presidential action in federal court have usually been rebuffed under the doctrines of standing, ripeness, or political question.53 Presidents have claimed wide-ranging power to commence and conduct military actions without an explicit congressional sign-off, relying on broad readings of statutes and on the Commander-in-Chief Clause, and Congress has often done little to push back.54 Presidents sometimes comply with the notice duty while asserting constitutional authority of their own. In 2017, President Trump notified Congress within 48 hours of launching a strike and said that he “acted in the vital national security and foreign policy interests of the United States, pursuant to [his] constitutional authority to conduct foreign relations and as Commander in Chief and Chief Executive.”55 An answer that calls the statute void because Presidents contest it overshoots, and so does one that predicts a court will order troops home under the sixty-day rule.
When Congress does push back, it must act through both houses. In December 2018, the Senate voted 56 to 41 to check the President's authority to direct U.S. military assistance to Saudi Arabia for its war in Yemen, and the measure died when the House declined to take up a parallel bill. The Senate also pushed back in the opposite direction against a proposed troop withdrawal from Syria, which raises the reverse question whether Congress can force a President to keep U.S. forces engaged abroad.56 Because courts decline to enter these disputes, the branches themselves settle where the power to start a war lies. Justice Jackson's warning in Youngstown Sheet & Tube Co. v. Sawyer, 343 U.S. 579, 654 (1952), describes the stakes: power “belongs in the hands of Congress, but only Congress itself can prevent power from slipping through its fingers.”57 Congress's most practical means of keeping that power is its control of money, which the next subsection takes up.
Funding as a Check
The text puts the money in Congress's hands. Article I, Section 8 gives Congress the power to spend for the common defense in clause 1 and the power to raise and support armies in clause 12.59 Clause 12 adds a condition that appears on no other kind of appropriation: no appropriation to raise and support armies may run for a longer term than two years.60 The Founders paired that limit with the assignment of the declaration power to Congress as a further safeguard against an executive who might engage in war without the consent of the people.61
The limit works because an appropriation is legal permission to spend, and the permission lapses with its term. After two years the army stays funded only if Congress votes to fund it again. Two results follow. The limit ensures that the President has only a limited “war chest,” and it forces each new Congress to reassess the wisdom of continuing hostilities that a prior legislature may have approved.62 The second result carries more weight, since a war begun with broad support continues only as long as successive Congresses keep choosing to pay for it.
Students often misread the two-year limit as a deadline for the war or the army. The clause limits the term of each appropriation and says nothing about how long a conflict may last, so a Congress that wants to keep the Army in being simply renews the funding, one appropriation at a time.
The seven-year term cannot stand, because clause 12 allows no army appropriation to run longer than two years. The two-year statute satisfies the clause, and the Army can then be funded for decades under successive appropriations. The statutes differ in only one respect, the term, and the term is the one thing clause 12 regulates. The size of the appropriation and the popularity of the war play no part in the analysis.
The two-year clause is a structural limit, and Congress's wider power over the purse supplies the working tools. Justice Jackson observed in his concurrence in Youngstown that Congress alone controls the raising of revenues and their appropriation and may determine in what manner and by what means they shall be spent for military and naval procurement.63 A Congress determined to protect its Article I authority can therefore act at two points. It may refuse to appropriate new funds to support a presidential war, as it did toward the end of the Vietnam War, and it may bar the further expenditure of funds that have already been appropriated.64 The tools answer different problems. Refusing new money stops a commitment from growing, while the bar on spending reaches money already appropriated that the executive could otherwise keep spending on the operation after Congress had turned against it.
The funding check is practical because it does not depend on a court. Federal courts have usually rebuffed challenges to a President's unilateral introduction of forces into hostilities under the doctrines of standing, ripeness, or political question, and Congress holds money directly.65 Its weakness is political cost. Once the President has put troops in the field, Congress must choose between funding the intervention and cutting off an army already engaged, and a vote for the second can be politically suicidal.66 That predicament is why Congress turned to the procedural limits of the War Powers Resolution, which makes withdrawal the default once the sixty-day period runs out. The purse is therefore the strongest check in the abstract and the hardest to use while a deployment is under way, and the Vietnam episode fits that pattern because Congress acted toward the end of the war.
Funding limits are statutes, and a statute restrains a President only if the President obeys it. Should a President flout these restrictions, a resolute Congress could invoke its ultimate power of impeachment.67 Impeachment is the backstop because it is the one remedy that needs neither the executive's cooperation nor a court's willingness to reach the dispute. Justice Jackson's warning in Youngstown, noted in the previous subsection, that only Congress itself can keep its power from slipping away, describes every tool here, because each works only if Congress is willing to use it.