Article I, Section 8 gives Congress, not the president, the power “To declare War.” The Framers wrote it that way on purpose — Madison changed the word from “make” to “declare” specifically so the executive would keep only the power to repel a sudden attack, not start a war on its own. This audit grades six features of that design against how war actually gets made in 2026: the clause itself, the 1973 law meant to enforce it, the authorizations that have outlived their targets by two decades, the money that funds it all, the courts that won’t referee it, and Congress’s own recent attempts to reclaim the power on paper. Same method as every institutional entry in this series — one design feature at a time, scored 0–100, every measurement cited. Reporting on documents, not legal advice.
The pattern
The Declare War Clause itself has never been directly overridden — no court has ever held that a president can unilaterally declare war, and Congress’s formal authority to do so is undisturbed. What has happened instead is that the clause has been made almost irrelevant by everything downstream of it. Congress hasn’t formally declared war since 1942. In its place: two open-ended authorizations from 2001 and 2002, stretched by four different administrations to justify force in roughly twenty countries their original targets never touched; a 1973 law written to force a 60-day clock on unauthorized war that no president has ever complied with on its own terms; courts that have never once ruled on the merits of a war-powers dispute between Congress and the president; and, as of this year, a formal administration position that an entire category of lethal military action simply falls outside the 1973 law’s reach altogether.
The scored card
The usual warning: every factual figure below is a measurement, cited to a court ruling, a statute, a CRS report, or a documented vote count. The final grade is not. It is the journal’s editorial judgment of how well each design feature actually functions in 2026. We publish the reasoning so you can argue with the math.
Scale: 85+ functioning as designed, no real erosion · 65–84 solid, real but limited strain · 45–64 mixed — meaningful erosion alongside real function · 25–44 substantially bypassed in practice · under 25 functionally hollowed out.
The Declare War Clause itself
Art. I, §8, cl. 11 — 11 declarations, 5 wars, last in 1942The Framers gave the war-making power to the legislature on purpose. At the Constitutional Convention, Madison and Elbridge Gerry moved to change the war clause's verb from "make" to "declare" specifically so the president would retain only the power to repel a sudden attack, not to start a war unilaterally — sword and purse deliberately split between two branches. That formal authority has never been taken from Congress by a court or a statute. Per the Congressional Research Service, Congress has formally declared war eleven times across five conflicts — the War of 1812, the Mexican-American War, the Spanish-American War, and World War I and II — with the last declarations issued in 1941 and 1942. No formal declaration since. The clause's exclusivity is intact; it has simply gone eight decades without being used for its stated purpose, as the wars fought since have run through other legal vehicles entirely.
The War Powers Resolution's compliance record
50 U.S.C. §§1541–1548 (1973)Passed over Nixon's veto in 1973, the law requires a president to report to Congress within 48 hours of introducing forces into hostilities, starting a 60-day clock after which forces must withdraw absent a declaration of war or specific statutory authorization. In its more than fifty years on the books, no president of either party has ever conceded that a use of force triggered that 60-day withdrawal requirement and actually pulled troops out because of it. Every administration since Reagan has either sidestepped the reporting and consultation requirements or contested the law's reach — a narrower, more specific dispute than "the whole law is unconstitutional," centered mainly on the legislative-veto mechanism the Supreme Court's reasoning in INS v. Chadha, 462 U.S. 919 (1983), effectively undercut. The clock has run more than fifty years without once forcing a withdrawal.
The 2001 and 2002 authorizations
Pub. L. 107-40; Pub. L. 107-243The 2001 AUMF authorized force against those who planned or aided the September 11 attacks and those who harbored them — al-Qaeda and the Taliban, by name. The 2002 AUMF authorized force against Saddam Hussein's Iraq. Per the Congressional Research Service and Brown University's Costs of War project, the 2001 authorization alone has since been cited to justify military operations in roughly twenty countries far beyond that original target, including Somalia, Yemen, Syria, Niger — where the 2017 Tongo Tongo ambush killed four American soldiers under its authority — and the Philippines. The stretching runs through an "associated forces" doctrine that appears nowhere in the statute's text, applied by four administrations of both parties to groups that in several cases did not exist in 2001. Two authorizations written for two specific targets have functioned, in practice, as a standing license.
The power of the purse as a war check
Case-Church Amendment (1973); Boland Amendments (1982–86)Congress's clearest successful use of the purse to end a war came in 1973, when a series of appropriations riders barred any further funding for combat in or over Vietnam, Laos, and Cambodia after August 15 of that year — Nixon did not contest Congress's authority to do it. But the tool cuts both ways: an earlier 1973 attempt to bar funding for the Cambodia bombing campaign specifically, the Eagleton Amendment, was vetoed and the House failed to override it — the same two-thirds problem that dogs the War Powers Resolution itself. And the Boland Amendments barring CIA and Pentagon funds for the Nicaraguan Contras were evaded outright through non-appropriated, off-books financing in the Iran-Contra affair. Per the Congressional Research Service, the deeper structural problem is that cutting off funds is far more usable prospectively, to block a new operation, than retroactively, against troops already deployed — where it is treated politically as abandoning forces in the field.
Judicial review
Campbell v. Clinton, 203 F.3d 19 (D.C. Cir. 2000)In 1999, 31 members of Congress sued over Clinton's Kosovo air campaign, conducted without a declaration of war or War Powers Resolution-compliant authorization. Both the district court and the D.C. Circuit dismissed the case — not by ruling the campaign lawful, but primarily on legislator-standing grounds, the same bar the Supreme Court applied in Raines v. Byrd, 521 U.S. 811 (1997). The Supreme Court denied certiorari. That pattern has held since: courts have consistently avoided ruling on the merits of war-powers disputes between Congress and the president, reaching for standing and ripeness doctrines rather than deciding who is right. More than two decades and several more undeclared conflicts later, no federal court has ever issued a merits ruling settling a war-powers dispute between the two elected branches.
Congress's own reassertion attempts
Yemen (2019), Iran (2020, 2025–26), Venezuela (2025)In April 2019, Congress passed a War Powers Resolution to end U.S. support for the Saudi-led coalition in Yemen — the first time in history a war-powers resolution cleared both chambers. Trump vetoed it; the Senate override attempt fell short of two-thirds. The same pattern repeated in May 2020 after the Soleimani strike: both chambers passed a resolution, Trump vetoed it on May 6, and the Senate's override vote failed 49-44 the next day. In November 2025, after the administration disclosed dozens of lethal strikes on alleged drug-trafficking vessels near Venezuela without advance notice to Congress, a Senate resolution to end the strikes failed 51-49. And on July 23, 2026, after a second round of U.S.-Iran hostilities broke out following a collapsed ceasefire, the House passed a war-powers resolution 214-208 — with four Republicans joining every Democrat — while the Senate rejected a companion measure the same day, 47-49. In more than fifty years under this law, a war-powers resolution has passed both chambers of Congress exactly twice, and been vetoed both times. It has never once survived an override vote.
The lowest grade in the Capitol Audit so far. The clause itself stands undisturbed at the top of the card — no one has ever successfully argued a president can declare war alone. Everything built to make that power functional in practice sits at the bottom: a 1973 law with a fifty-year unbroken record of never once forcing a troop withdrawal on its own terms, two decades-old authorizations doing duty in nations their authors never named, courts that decline to referee the dispute at all, and a Congress that has managed, twice in over five decades, to pass a war-powers resolution through both chambers only to watch it die on a veto override both times. The purse power is the one meaningful exception — it ended Vietnam once — and even that came with a documented instance, Iran-Contra, of the executive simply routing around it.
The last word
The 2025 Venezuela strikes mark a new low-water mark worth stating plainly: this is the first time an administration has told Congress, as a formal legal position rather than a contested implication, that an entire class of lethal military action — strikes without American ground troops in harm’s way — falls outside the War Powers Resolution’s reach altogether. That is not a dispute over how the law applies. It is a claim that it does not apply. Congress’s response was a floor vote it lost 51-49, on a resolution that would not have bound a president who disagreed with it anyway, since none has yet. The design the Framers built gave Congress the sword’s authorization and the purse’s leverage specifically so no single person could start a war alone. Eighty-four years after the last declaration, the authorization runs on statutes stretched past their targets, the leverage runs on funding no one wants to be blamed for cutting, and the check either branch could still, in theory, provide — the courts — has for over two decades declined to referee the fight at all.
Ubi non est lex, non est transgressio — where there is no law, there is no transgression. The law here is real, and Congress wrote a second law specifically to enforce it. Neither has stopped a war yet.
Sources & methodology:
- U.S. Const. art. I, § 8, cl. 11 (Declare War Clause); Madison’s Convention notes on changing “make” to “declare” — Constitution Annotated
- Formal declarations of war, 11 across 5 conflicts, last in 1941–42: CRS RL31133, “Declarations of War and Authorizations for the Use of Military Force”
- War Powers Resolution of 1973 (50 U.S.C. §§1541–1548), passed over Nixon veto: CRS IF13134
- INS v. Chadha, 462 U.S. 919 (1983): legislative-veto reasoning affecting WPR §5(c)
- 2001 AUMF (Pub. L. 107-40) and 2002 AUMF (Pub. L. 107-243) application beyond original targets, ~20 countries: CRS R43983; Costs of War Project, Brown University — “The 2001 AUMF”
- Case-Church Amendment (1973) ending funding for Indochina combat operations; Eagleton Amendment veto and failed override: contemporaneous congressional record
- Boland Amendments (1982–86) and Iran-Contra circumvention: congressional record; DOJ Iran-Contra Independent Counsel final report
- Congressional authority and practical limits of funding cutoffs against deployed forces: CRS R41989, “Congressional Authority to Limit Military Operations”
- Campbell v. Clinton, 52 F. Supp. 2d 34 (D.D.C. 1999), aff’d 203 F.3d 19 (D.C. Cir. 2000), cert. denied 531 U.S. 815 (2000)
- Raines v. Byrd, 521 U.S. 811 (1997): legislator standing doctrine
- Yemen War Powers Resolution, S.J.Res.7 (116th Congress), passed both chambers, vetoed April 16, 2019, override failed: Congress.gov; Arms Control Association
- Iran War Powers Resolution, S.J.Res.68 (116th Congress), passed both chambers, vetoed May 6, 2020, Senate override failed 49-44 on May 7, 2020: White House veto message; Senate.gov vetoes record
- Venezuela vessel strikes beginning September 2025 and administration’s WPR-inapplicability position: Just Security; PBS
- Senate Venezuela War Powers Resolution vote, 51-49, November 2025: Courthouse News; CBS News
- June 2025 U.S. strikes on Iranian nuclear sites and July 2026 collapse of ceasefire: contemporaneous reporting
- House and Senate Iran War Powers votes, July 23, 2026 (House 214-208; Senate 47-49): CNBC; The Hill; NPR
- The grades are the journal’s editorial judgment; every underlying factual claim above is independently sourced as cited.