The Article V Convention Handbook
A map of how a convention could work, what keeps it safe, and what we're still figuring out.
These snippets are from an early draft of a more comprehensive, off-line version. Wording, evidence, and recommendations are bound to change with continued input from experts. As always, we also invite people who will argue with it.
Finalized handbook will be published in print and for download.
What this handbook is
This is a plain-language guide to the one path the Constitution gives ordinary citizens to amend it without Congress: the Article V convention. No Article V convention has ever been held. The Constitution provides the mechanism in a single sentence, and almost every operational detail below is a matter of recommendation, scholarly debate, or analogy rather than settled law. This handbook seeks to do two things at once: lay out a complete, workable procedure for a convention limited to campaign-finance reform, and flag — plainly — where that procedure rests on contested ground.
We are not writing from scratch. The convention method has been studied carefully before, and we lean on that record: the American Bar Association's 1974 Special Constitutional Convention Study Committee report, the Department of Justice's 1987 Office of Legal Policy analysis of limited conventions, and the procedures Senators Sam Ervin and Orrin Hatch seperately proposed throughout the 1970s and 1980s to govern a convention if one were ever called. Each concluded, in its own way, that a convention limited to a single subject is both possible and governable. Where the law or scholarship is genuinely unsettled, the text says so. Where Mutable takes a position, it is labeled Mutable’s recommendation and paired with the strongest competing view, so that a coalition partner, a skeptical legislator, or a court can see the reasoning and the alternative. The goal is a process that would survive scrutiny as a triumph of democratic principle and transparency — not a process that pretends more certainty than exists.
1787 Federal Convention — the only national constitutional convention and the cautionary “runaway” example.
Twenty-First Amendment ratifying conventions (1933) — the only use of the convention method for ratification; the model for elected, pledged delegates who “have no political axe to grind.”
National Women’s Conference (1977) — a federally funded ($5M), state-elected, fully televised national convention; the closest modern model for funding, delegate elections, media, and a plan-of-action report to Congress.
Various State Constitutional Conventions — the states have held conventions roughly 250 times throughout the nation's history. Several noted conventions were subject limited with the majority popularly electing delegates.
Guiding principles
Six principles run through every procedure in this handbook. Where a procedural choice is unsettled, the option that better serves these principles wins.
- 1Single-subject discipline. Nothing outside campaign finance is proposed, debated, or voted — enforced procedurally, not just rhetorically.
- 2Transparency. Every application, count, floor session, and roll-call vote is public and contemporaneously recorded. Committee negotiation may be closed only under a narrow procedure of the standing rules.
- 3Democratic legitimacy. Delegates derive authority from the voters wherever feasible; the process should read as the people's work, not an insiders' deal.
- 4Bipartisanship by design. Rules are written so that neither party gains procedural advantage; a campaign-finance amendment that looks partisan cannot reach 38 states.
- 5Anti-runaway safeguards. Multiple independent locks (matching applications, instructions, faithfulness laws, germaneness rules, ratification) so no single failure produces a runaway.
- 6Honest uncertainty. Contested points are labeled; the process never claims settled authority it does not have.
The two doors of Article V
Article V offers two ways to propose amendments to the U.S. Constitution. The first, used for all twenty-seven amendments so far, runs through Congress: two-thirds of both houses vote to propose. The second has never been used to completion: two-thirds of the states — thirty-four legislatures — can compel a convention to propose amendments, with no permission from Congress required. Both proposing pathways still require final ratification from the states.
The second door is not theoretical. States have filed hundreds of convention applications over the past two centuries, on dozens of subjects, from direct election of senators to a balanced budget. None has yet crossed the thirty-four-state line for a single, agreed subject — but the mechanism is live, and it has come closer than most people realize. Mutable takes this door for one simple reason: the first depends on the very people the money system put in office, and the second does not.
The Congress, whenever two thirds of both Houses shall deem it necessary, shall propose Amendments to this Constitution, or, on the Application of the Legislatures of two thirds of the several States, shall call a Convention for proposing Amendments, which, in either Case, shall be valid to all Intents and Purposes, as Part of this Constitution, when ratified by the Legislatures of three fourths of the several States, or by Conventions in three fourths thereof, as the one or the other Mode of Ratification may be proposed by the Congress; Provided that no Amendment which may be made prior to the Year One thousand eight hundred and eight shall in any Manner affect the first and fourth Clauses in the Ninth Section of the first Article; and that no State, without its Consent, shall be deprived of its equal Suffrage in the Senate.
The single-subject limitation
A convention begins when thirty-four state legislatures pass "applications" calling for one. These are ordinary legislative resolutions, transmitted to Congress, and they do more than count toward a threshold — their wording can define and confine what the convention is permitted to consider.
Our approach is a single uniform application — the same operative text adopted state to state — narrowly limited to the subject of money in politics. Uniform wording matters for a practical reason explored in the next section: the more clearly thirty-four applications describe the same subject in the same terms, the harder it is for anyone to argue they don't count together. We'd rather remove that argument before it starts than win it later.
We're not starting from nothing. Vermont passed a campaign-finance convention application, in 2014, as did Rhode Island in 2016 — both remain active today. Their language drifts a bit, but we can build on their groundwork.
Visit the Library for our draft state application
The deepest objection is that a limited convention has never been tested. That is true only at the federal level. At the state level, subject-limited conventions are a routine, repeatedly successful American practice. Tennessee and Rhode Island have both held five, each was confined by the legislature’s authorizing act to enumerated subjects, each one staying within its assigned subjects, and each proposal sent to the voters for the final say. Limited conventions have also happended in Pennsylvania and Virginia, plus several in the South throughout the late eighteen, early nineteen hundreds.
The model Mutable proposes — a call that fixes the subject and elects delegates — mirroring the models several states have followed for years. It is also the practice the ABA’s 1974 study relied on in concluding a convention may be limited. And it is affirmed by the Ervin bills passed in the Senate.
The aggregation solution
Whether applications on different subjects, or filed decades apart, can be added together to reach thirty-four is genuinely contested. Two questions sit underneath it: must applications be on the same subject to be aggregated, and do old applications ever expire? Some efforts — the long-running balanced-budget push — rely on counting applications filed across many years, and some of those states have since voted to rescind them, which raises the further question of whether a rescission actually removes an application from the count.
We reject loose aggregation entirely: only applications calling for the same, single, clearly worded subject should count toward a convention. Unfortunatley, as it stands, there is no official referee — theoretically, Congress itself tallies the applications, however there exists no "live" count. Additionally, no court or other party is empowered to second-guess it, which means the counting fight is ultimately political rather than procedural. That is precisely why uniform wording and active management of old applications matter so much.
Visit the Library for our draft state application
Congress has never formally tabulated applications, and the National Archives treats them as congressional records it has no authority to compile. That vacuum is a transparency risk.
Mutable backs the reform — already proposed in Congress and endorsed in model policy — requiring the Archivist to separate, organize, and publicly post all applications and rescissions, and requiring Congress to maintain a public website of the count. Whoever maintains the authoritative, public count shapes the debate over whether the threshold is met.
The call, and Congress's role
Once thirty-four valid applications exist, the Constitution says Congress "shall call" a convention. But the details of that duty have never been tested. In practice Congress would likely decide how the count was certified, set the time and place, and provide funding and logisticl support. None of that is spelled out in the text.
The closest thing to a roadmap is the Federal Constitutional Convention Procedures Act that Senator Sam Ervin pressed in the early 1970s. It proposed exactly this missing machinery: how applications are counted, a window within which they must be contemporaneous, how delegates are apportioned, and how a convention transmits its work. The Senate passed a version of it more than once; the House never did, so it never became law. That leaves us with a sketch of how Congress could run a convention, and no binding rule that it must run it that way — and a real possibility that a reluctant Congress drags its feet or sets hostile terms.
Set a prompt call. Mutable proposes Congress call the convention within a fixed, short window after the 34th application is certified — enough lead time for states to run delegate elections, but no foot-dragging.
Specify logistics and affirm the subject. Place, time, subject. The call should restate the single subject in the states’ own terms, and Congress should publish its application tabulation and reasoning.
Provide administrative and financial support. Mutable takes the position that amending the U.S. Constitution is federal work. Delegates should be compensated. The federal government should provide HR and administrative staffing needs. Visit the Library for our Modeling the Cost suppliment.
Who goes: choosing delegates
The Constitution is silent on how states pick their convention delegates, which makes it one of the most consequential open questions. The choice is between legislatures appointing delegates and voters electing them. We favor popular election: a delegation accountable to citizens, not to the politicians who might prefer to control it.
On the question of apportionment — how many delegates each state sends — our draft proposal is a "House-plus-two" basis: each state sends delegates equal to its number of congressional districts plus two, mirroring the size of its congressional delegation, with delegates elected by district plus two at large.
Mutable recommends that every applying state pair its application with a delegate faithfulness statute that (a) limits the delegate's authority to the single campaign-finance subject; (b) voids any vote cast outside it; (c) provides for orderly replacement of a delegate who breaches; and (d) requires a public oath to the instruction before the delegate is seated.
Instruction is easily misread as a substantive mandate. It is not, and the distinction is the spine of a legitimate convention. A delegate is bound as to subject, not as to substance. The faithfulness laws fix the jurisdiction — delegates may work only on campaign finance and may not stray. But within that subject, no one hands the delegates the answer. Not the legislature, not a sponsoring organization, not the voters who elected them. The delegates deliberate over the text, weigh alternatives, hear testimony, and reach their own judgment on what the amendment should say.
The rules of the floor
A convention writes its own rules when it convenes — the 1787 convention did exactly that, behind closed doors. That fact is both the root of the "runaway" fear and the place to defuse it. The same power that lets a convention set its rules can be used to bind it: rules agreed in advance can hold it inside its lane.
The draft rule-set we're assembling has a few load-bearing parts: a hard single-subject limit tied to the applications that called the convention; a high supermajority required to propose anything; open and recorded proceedings rather than the secrecy of 1787; a defined quorum; and a credentials process that seats only delegates properly instructed by their states. The unresolved question is how much of this can be locked in before anyone walks in the room, versus adopted by the delegates once they do — and that, in turn, depends on the contested questions about Congress's role as discussed in Section 05.
A Committee on Standing Rules presents a proposed rules package on the first day. Debate follows, and the convention adopts a final structure by a two-thirds vote. There is no precedent requiring a supermajority to adopt convention rules; in fact, deliberative bodies ordinarily adopt their rules by simple majority. Mutable nonetheless recommends the two-thirds bar, not because precedent compels it, but because clearing a supermajority on day one, before any substantive question is reached, signals broad buy-in and gives the whole proceeding an unimpeachable legitimacy. Visit the Library for our Model Convention Standing Rules suppliment.
Media and public transparency
Delegates at the 1787 convention deliberated in total secrecy with a pledge to withhold the record for 50 years. The idea of replicating that blanket secrecy helps fuel modern runaway fears. Mutable rejects that model, with carve outs. Every floor session is open and every roll-call vote is recorded. Media should be provided access via a press center and interview opportunities. The 1977 National Women’s Conference — covered nationally on television, with gavel-to-gavel public-television coverage — shows a national convention can be conducted almost entirely in public without losing seriousness.
However, compromise and consensus sometimes require private debate. Congress itself operates under this standard. Indeed, research on deliberative bodies suggests some privtae time actually improves, rather than undermines, the odds of an honest, negotiated outcome.
Beyond the press, the public itself should be able to watch — in the gallery and online. We should also provide structured, transparent channels for public input that do not disrupt the floor: published written-comment intake, and — where the convention chooses — designated public-testimony windows in committee.
Ratification: the backstop
This is the safety net beneath everything else. Nothing a convention proposes becomes part of the Constitution until three-quarters of the states — thirty-eight — ratify it. That means thirteen states, acting alone, can stop anything. It is an enormous, deliberately high wall, and it works no matter what happens at the convention itself.
A few details matter. Article V stipulates that Congress chooses the mode of ratification — by state legislatures, or by special state conventions; the latter has been used only once, for the repeal of prohibition. By modern custom Congress sets a window, usually seven years, within which ratification must be completed. And the Supreme Court held in Coleman v. Miller (1939) that key questions about the ratification process are "political questions" for Congress rather than the courts — which cuts both ways, but means the wall is policed by the states and Congress, not litigated away.
In 1933, Congress chose ratifying conventions for the Twenty-First Amendment (repeal of prohibition) precisely to route around state legislators who were beholden to the temperance lobby — leaving “gun-shy legislators … off the hook” and putting the question to specially elected delegates who had “no political axe to grind.”
Campaign-finance reform presents the same structural problem in mirror image: the officeholders who would ratify are the ones who benefit most from the status quo. Ratifying conventions move the decision to the people’s elected delegates and approximate a one-state, one-vote national referendum on the amendment — the most legitimate and most winnable path for this subject.
The runaway fear, answered
The honest objection: open the Constitution, and won't a convention rewrite everything? Our answer is multi-tiered. The first level is the call — single-subject applications, instructed and electable delegates, pre-agreed rules — are designed to keep a convention narrow. Next, the rules and the record — rules requiring supermajorities, mirroring Congress, and pronounced transparency — keep the convention honest. At the last level stands the ratification wall: even if every front-end limit somehow failed, thirteen states can refuse to ratify, and the proposal dies. History backs the wall: Congress itself sent the country genuinely dangerous amendments that simply died for lack of ratification, like the Corwin Amendment of 1861, which would have shielded slavery forever.
We owe skeptics the strongest version of their own case, which is 1787 itself: the Philadelphia Convention was called only to revise the Articles of Confederation and instead wrote an entirely new Constitution. That is the original "runaway." But the circumstanes of the day required something new, and notice what still held even then — the new document had no force until it was ratified, and it nearly wasn't. The backstop worked in the one case people fear most. Visit the Library for our History of Article V examination.
What's still missing
This is a v0.1, and the gaps are exposed rather than hidden. What it still lacks: deep legal citations under every claim; a full review by constitutional counsel; a state-by-state map of the convention applications already on the books, and which are live; finished model documents to sit alongside the prose; and worked-out answers to many remaining open questions.
› Notes & Sources
U.S. Const. art. V, constitution.congress.gov/browse/essay/artV-3-3/ALDE_00013051/
CRS, The Article V Convention: Contemporary Issues for Congress (R42589); and Historical Perspectives (R42592) , congress.gov/crs_external_products/R/PDF/R42589.15.pdf
American Bar Assn., Amendment of the Constitution by the Convention Method Under Article V (1974) — establishment study concluding a convention may be limited, partly on democratic-legitimacy grounds, archive.org/details/ABA_Report_of_the_Special_Constitutional_Convention_Study_Committee
U.S. Dept. of Justice, Office of Legal Policy, Limited Constitutional Conventions Under Article V (1987) — executive-branch legal opinion (Meese-era originalist lean) , ojp.gov/ncjrs/virtual-library/abstracts/report-attorney-general-limited-constitutional-conventions-under
John Dinan, Explaining the Prevalence of State Constitutional Conventions in the 19th and 20th Centuries, 34 J. Pol. Hist. 297 (2022) — scholarly survey of the state convention record, including limited conventions (documenting limited single-subject conventions on voting and registration in Virginia 1945 and Rhode Island 1944 & 1958) , doi.org/10.1017/S0898030622000136
Tennessee constitutional convention record, 1953–1977 (Tenn. State Library & Archives; Tennessee Encyclopedia) — five limited conventions, each fixed to enumerated subjects and voter-ratified, tennesseeencyclopedia.net/entries/government/
Rhode Island constitutional convention record, 1944–1986 (R.I. Dept. of State Archives; R.I. ConCon Clearinghouse) — a series of limited conventions, each fixed to enumerated subjects and voter-ratified; the one unlimited full-revision convention (1964–68) was rejected at the polls, rhodeisland.concon.info/?page_id=158
Pennsylvania Constitutional Convention of 1967–68 (50 Constitutions, Univ. of Wisconsin Law School; Pennsylvania Bar Association) — limited by enabling act to four subjects, with out-of-scope proposals ruled out of order from the chair; all five ballot questions ratified, 50constitutions.org/pa/timeline
Federal Constitutional Convention Procedures Act, S. 1272 (Ervin) — passed the Senate 1971 & 1973; seven-year same-subject aggregation. Constitutional Convention Implementation Act (Hatch) — S. 1710 (1979), S. 119 (1984, reported from the Judiciary Committee, S. Rept. 98-594), and S. 214 (1991) — same-subject seven-year aggregation, House-plus-two delegate apportionment, exclusion of sitting members of Congress as delegates; the 1984 committee report asserted Congress's power to enforce the subject limits, congress.gov/bill/93rd-congress/senate-bill/1272
Coleman v. Miller, 307 U.S. 433 (1939), https://tile.loc.gov/storage-services/service/ll/usrep/usrep307/usrep307433/usrep307433.pdf
Hawke v. Smith, 253 U.S. 221 (1920), https://tile.loc.gov/storage-services/service/ll/usrep/usrep253/usrep253231/usrep253231.pdf
Dyer v. Blair, 390 F. Supp. 1291 (1975), https://law.justia.com/cases/federal/district-courts/FSupp/390/1291/1966458/
Reynolds v. Sims, 377 U.S. 533 (1964), https://tile.loc.gov/storage-services/service/ll/usrep/usrep377/usrep377533/usrep377533.pdf
National Constitution Center, Report: Article V Constitutional Conventions, constitutioncenter.org/news-debate/special-projects/article-v-amending-the-constitution/report-article-v-constitutional-conventions
Constitution Annotated, Ratification by Conventions (21st Amendment), constitution.congress.gov/browse/essay/artV-4-3/ALDE_00013057/
1977 National Women’s Conference (Pub. L. 94–167); State ratifying conventions (procedures overview), en.wikipedia.org/wiki/1977_National_Women%27s_Conference
List of state Article V applications (overview & current count), en.wikipedia.org/wiki/List_of_state_applications_for_an_Article_V_Convention
David E. Pozen, The Common Law of Constitutional Conventions, 112 Calif. L. Rev. 2213 (2024), — argues the deep state-convention record supplies a workable middle path between originalism and ad hoc invention for a federal convention, doi.org/10.15779/Z38RX93F7T
Jonathan L. Marshfield, American Democracy and the State Constitutional Convention, 92 Fordham L. Rev. 2555 (2024), — empirical survey of the state-convention record underlying the delegate-selection, one-person-one-vote, and deliberative-body norms, ir.lawnet.fordham.edu/flr/vol92/iss6/10
Dennis F. Thompson, Democratic Secrecy: The Dilemma of Accountability, 114 Pol. Sci. Q. 181 (1999), — the foundational theory of when secrecy is compatible with democratic accountability, onlinelibrary.wiley.com/doi/10.2307/2657736
David Stasavage, Open-Door or Closed-Door? Transparency in Domestic and International Bargaining, 58 Int’l Org. 667 (2004), — empirical study, including the 1787 Convention and early Congress, finding open bargaining increases posturing and the risk of breakdown; supports Mutable’s closed-session rule, researchonline.lse.ac.uk/id/eprint/225/1/open-door8.pdf
This document is unfinished. So is the movement.
When it's done, the handbook will be published for distribution — in print and as a free download. Until then it's built in the open, and there's room at the founding for the people who'll help complete both.
Get involved