The Convention Handbook

Before you readA Note on the Status of This Handbook

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 does two things at once: it lays out a complete, workable procedure for a convention limited to a single subject, and it flags — plainly — where that procedure rests on contested ground.

Where the law 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.

Part IFoundations

§1 Purpose and Scope

This handbook governs a single-subject Article V convention whose only authorized purpose is to propose an amendment on the single subject the sponsoring campaign has chosen and stated in its applications. It is written for all real and theoretical stakeholders in the process: congress members, state legislators and election officials, convention delegates and their staff, host-state administrators, government HR offices, the press, and the coalition running the effort.

It is deliberately topic-disciplined. A convention limited to a single subject is only defensible if every actor — from the drafter of a state application to the parliamentarian on the floor — treats the single-subject limitation as the spine of the whole enterprise. Most of the failure modes discussed below trace back to a breach of that discipline.

Mutable takes no position on which amendment any coalition should pursue. This handbook is about procedure — how a convention is applied for, called, staffed, apportioned, run, recorded, and ratified — and every recommendation in it is designed to hold regardless of subject. It is written to be adopted, cited, and demanded by anyone: the coalitions pursuing an amendment today, the state legislators who will write the enabling laws, and the organizations that oppose holding a convention at all but would rather it be run this way than any other.

§2 The Constitutional Text

Four thresholds are fixed by the text and are not negotiable: 34 states must apply to trigger the call; Congress “shall call” (a duty, not a discretion); a convention only proposes (it cannot enact); and 38 states must ratify. Everything else — how delegates are chosen and apportioned, how the convention votes, how long it sits, who pays — is left open by the text and must be supplied by practice, by Congress’s call, by the states, or by the convention’s own rules. This handbook weighs the evidence and attempts to provide workable solutions to all these missing questions and more.

§3 The Single-Subject Limitation

The convention is limited to the single subject stated in the applications, and every mechanism in this handbook exists to hold it there. That limit is declared in the applications (§7), carried into the congressional call (§11), written into each state’s delegate instruction (§21), and enforced on the floor by a germaneness rule any delegate may invoke (§42). Whether a convention can be limited as a matter of law is the deepest contested question in Article V, and this handbook does not pretend otherwise — it states the authority on both sides:

No court has ever resolved this. In Coleman v. Miller (1939) the Supreme Court called amendment-process questions “political” and left them to Congress — which cuts against judicial enforcement of a convention’s limits — though later cases such as Baker v. Carr (1962), which held legislative-apportionment challenges justiciable rather than "political," suggest the modern Court might engage. Mutable therefore does not rely on courts to police the limit; it relies on the procedural locks and, above all, on ratification.

This limited-convention framework is not a partisan artifact. Two serious Senate efforts converged on it from opposite ends of the spectrum, two decades apart: Ervin, a Democrat, in 1971 and 1973, and Hatch, a Republican, across the 1980s. Both treated subject-limited conventions as valid, both aggregated applications only by the same subject within a seven-year window, and both assigned Congress the role of policing scope. Reporting Hatch's bill in 1984, the Senate Judiciary Committee asserted Congress's power not merely to set but to enforce the limits stated in the state applications — an institutional endorsement of limitability standing alongside the ABA and DOJ reports.

§4 Register of Open Legal Questions

These questions have no authoritative answer. The handbook adopts a working position on each (stated in the relevant Part) but treats none as closed. Coalition messaging should concede them openly rather than be caught flat-footed.

1. May a convention be limited to one subject, and is the limit judicially enforceable? (Authority is split — see §3.)

2. May Congress aggregate non-identical applications, and who counts them?

3. Do applications expire, and may a state rescind before the 34th application?

4. Who selects delegates — legislatures, the people, or Congress — and may they be recalled or instructed?

5. How is voting apportioned: one state, one vote, or by population?

6. What vote threshold proposes an amendment (simple majority vs. supermajority of states or delegates)?

7. May Congress fund, time-limit, or set procedures for the convention, or only fix place, time, and subject?

8. Which mode of ratification applies, and may Congress choose conventions over legislatures?

§5 Guiding Principles

Six principles run through every procedure in this handbook. Where a procedural choice is close, the option that better serves these principles wins.

1. Single-subject discipline. Nothing outside the single subject is proposed, debated, or voted — enforced procedurally, not just rhetorically.

2. Transparent by default, protected where negotiation requires it. Every application, count, floor session, and roll-call vote is public and contemporaneously recorded. Committee negotiation may be closed only under the narrow, logged, time-limited procedure of the standing rules.

3. Democratic legitimacy. Delegates derive authority from the voters wherever feasible; the process should read as the people’s work, not an insiders’ deal.

4. Bipartisanship by design. Rules are written so that neither party gains procedural advantage; a subject-limited amendment that looks partisan cannot reach 38 states.

5. Anti-runaway safeguards, layered rather than sequential. The germaneness rule and the ratification threshold bind every convention this handbook describes and do not depend on any state’s legislative choices. Matching applications and state faithfulness statutes are recommended, additive layers on top of that floor — valuable, but not preconditions a convention must clear before it can safely proceed.

6. Honest uncertainty. Contested points are labeled; the process never claims settled authority it does not have.

Part IIThe Application Phase (States → Congress)

§6 Standardize the Wording

Every applying state adopts the same or substantially similar model application wording. Prior Article V drives have foundered in part on inconsistently worded resolutions across states. Non-identical applications hand Congress an excuse not to aggregate, and they muddy the “same subject” question that determines whether 34 even counts. The remedy is a single model application adopted in materially identical form in every state, with an explicit aggregation clause that ties each application to the others on the shared subject and disclaims aggregation with any other subject.

§7 What the Application Must Do

1. Apply under Article V for a convention “for proposing amendments,” limited to the coalition’s single chosen subject (Article V does not let an application dictate exact amendment wording, so the application defines the subject, not the final text).

2. Define the subject affirmatively — as a power the people and the states should hold — rather than as an attack on a single court decision, which ages badly and narrows coalition appeal.

3. Include a single-subject limitation clause stating the convention may consider no other subject and instructing the state’s delegates accordingly.

4. Include a calibrated aggregation clause: count with other applications on the same subject; do not count toward a plenary or any other-subject convention.

5. Direct transmission to the President of the Senate, the Speaker and Clerk of the House, the state’s congressional delegation, and the presiding officers of every other state legislature (the 1787-era practice of inviting cooperation).

6. State a continuing-application term and the conditions for rescission, so the state controls its own count.

§8 The Path to 34 — Managing Application Aggrigation

Because aggregation issues involving arguably disarate subject matters have led to contested counting, a coalition’s sequencing should never let an application on one subject read as joinable to an application on another. Two operational practices follow: (a) each new application recites the specific sister applications it joins, by state and year, building a clean, self-documenting family; and (b) the sponsoring coalition maintains a public, continuously updated ledger of the count, with the text of every application linked, so the coalition — not opponents — owns the authoritative tally. Any coalition, on any subject, is welcome to adopt this same discipline; the target here is not a rival movement but a specific bad practice — an application drafted loosely enough to be aggregated with a subject it never sought.

§9 Transparency in the Count

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 — requiring the Archivist to separate, organize, and publicly post all applications and rescissions, and requiring Congress to maintain a public website of the count.

Part IIIThe Congressional Call

§10 Congress’s Role: a Ministerial Duty

The better-supported reading — Hamilton in Federalist No. 85, the earliest congressional practice, and most modern scholarship — is that once 34 valid same-subject applications exist, Congress’s duty to call is ministerial, mandatory, and non-discretionary. Most scholars consider that duty judicially enforceable in principle, though whether a court would compel a recalcitrant Congress is less settled — Coleman v. Miller (1939) treated amendment-process questions as "political" — so in practice the real levers are the clean public count and political pressure. Congress may not deliberate over whether to call; it may only specify the convention’s time, place, and subject. The President has no formal role in the call, and the calling resolution appears not to require presentment or signature.

§11 What Congress May — and May Not — Do

Honest caveat: whether Congress may also impose a duration limit or baseline procedural rules is genuinely disputed. Some 1970s–1990s Senate bills (the Ervin and later proposals) assumed Congress could. Mutable’s working position is that Congress may set place, time, subject, and funding, and may recommend but not impose internal procedure — leaving rules to the convention under §34. This is not a contradiction: §11 says Congress may recommend but not impose the convention’s internal procedure, and the duration limit in §12 is not Congress imposing anything — it is the convention binding itself, fully consistent with a rule under which only the convention controls its own internal procedure.

§12 Convention Timeframe

Mutable recommends Congress (or the call) allot one month for the convention, renewable only on a showing that an extension would produce an outcome rather than prolong a deadlock. A short, focused term suits a single-subject convention, discourages mission creep, and mirrors the brisk pace of the 1933 ratifying conventions (many concluded in a single day) and the four-day National Women’s Conference. The extension standard should be written into the convention’s own rules (Part IX), with the convention itself — not Congress — judging whether the standard is met, to preserve the convention’s independence.

§13 The Mode-of-Ratification Decision

Congress should specify ratification by state conventions. Article V gives Congress the choice between ratification by state legislatures and by state conventions, and the choice can be strategically decisive, especially for an amendment that would restrict the interests of incumbents themselves, and is treated at length in Part XII. In brief, this handbook recommends Congress choose ratifying conventions, for the same reason Congress chose them for the Twenty-First Amendment (repeal of prohibition): to route around the very officeholders who benefit from the status quo. The recommendation belongs in the convention’s transmittal letter (Part XI) so it reaches Congress with the proposal.

Part IVThe Host State

§14 Why a State Should Want to Host

Hosting is a prize. The host state gains the civic prestige of seating the first Article V convention in American history; substantial economic activity (the 1977 National Women’s Conference in Houston drew ~2,000 delegates and 15,000–20,000 observers and blanketed national television); a generational tourism and earned-media moment; and a permanent place in the historical record. A state legislature can even volunteer in its application or by separate resolution, expediting the location decision. This is not in tension with the application’s single-subject discipline: matching language governs the substance of what the convention would consider, so every state’s application has to agree on that. Hosting is pure logistics — it says nothing about the amendment — so a state volunteering to host needs no matching language from any other state.

§15 What the Host State Is Responsible For

Mutable’s cost-allocation principle (Part VII) is simple: the convention is national work, so the federal government pays the people; the host state provides the place and its operation. Concretely, the host state should furnish:

1. Venue. A large university lecture hall, auditorium, or amphitheater with fixed seating for the full delegation count plus committee rooms, a press center, and a public gallery.

2. Security. State and local law enforcement for the venue, delegates, and the public, coordinated with federal authorities as warranted by the event’s national profile.

3. Technology. Livestream and broadcast infrastructure, microphones and a recorded floor-audio system, an electronic or roll-call voting mechanism, reliable connectivity, and the public-record capture described in Part X.

4. Staffing. Floor staff, clerks, stenographers/transcription, IT support, ushers, and ADA accommodation personnel.

5. Daily meals. Lunch and refreshments for delegates and credentialed staff during session days.

6. Local logistics. Wayfinding, shuttle/parking, an information desk, and a liaison office working with government HR (Part VIII).

§16 Venue Requirements

§17 Selecting the Host State

To keep the choice transparent and non-political, Mutable recommends a published request-for-proposals process: interested states submit bids against the Part 15–16 criteria; bids and scoring are public; and the decision-maker (Congress in the call, or a neutral body it designates) selects on the published criteria. A neutral, criteria-driven selection forecloses the appearance that the host was chosen to tilt the outcome.

Part VDelegates: Selection, Apportionment, and Instruction

§18 How Delegates Are Selected

Delegates are elected by the voters of each state. Three methods are on the table and the choice is genuinely contested, but only popular election produces a convention the country will recognize as its own work rather than an arrangement among insiders.

Unsettled: whether a state legislature may bind or recall popularly elected delegates is unresolved. Alabama’s 1933 advisory opinion suggested binding instructions are permissible; Maine’s suggested at-large selection could violate geographic-representation norms. Mutable’s model leans on instruction-by-statute plus the ratification backstop rather than assuming recall power exists.

§19 Delegate Count

Each state sends one delegate for every seat it holds in the House of Representatives, plus two at large — the House-plus-two formula, yielding a convention of 535. Article V is silent on delegation size, and the historical record shows the number has always been fixed by the calling authority, never by any built-in rule. The 1787 Convention seated wildly uneven delegations — from two delegates to eight — and the 1933 ratifying conventions varied more sharply still, because each state’s own enabling act set its delegate count. State constitutional conventions, the deepest body of American practice, almost always sized their delegations to the state legislature, roughly one convention delegate per legislative seat, so the convention mirrored an apportionment the people already recognized. The lesson is not a magic number but a principle: tie the count to an existing, legitimate measure of representation rather than inventing one.

Mutable adopts the measure a chamber of Congress actually endorsed. The Ervin bill — passed by the Senate in 1971 and again in 1973 — fixed each state’s delegation at its total congressional representation: one delegate per House seat, plus two at large for its senators. That is the House-plus-two formula, and it yields a convention of 535 delegates — exactly the size of Congress itself. Mutable keeps it for two reasons: it is the only delegate-count formula a house of Congress has ever passed, which gives it unmatched procedural pedigree, and its modest two-delegate floor is a small, honest nod to the federal character of the enterprise that does not surrender the individual vote.

In candor, the most authoritative neutral study went further. The ABA’s 1974 committee judged even House-plus-two "seriously questionable," because the two at-large delegates over-represent the smallest states by fifty percent or more — and Alaska, in 1974 terms, by nearly 375 percent — and it preferred apportionment by House seats alone. A state, or the convention itself, that wanted the purest one-person, one-vote structure would simply drop the two at-large delegates. Either way, the decisive point is settled elsewhere: however the delegates are counted, on the floor they vote individually, one delegate one vote (§37). The count controls only how much population deviation is built in; it does not smuggle state-bloc voting back in.

Two honest limits are worth naming here. First, House-apportioned delegate seats are elected within each state’s existing congressional district lines, so a gerrymandered map carries into delegate elections exactly as it carries into House elections — this handbook does not invent a new districting standard to fix that, any more than it takes a position on any other subject outside its own; the district lines a state uses are that state’s own choice, for better or worse. Second, on outside money reaching delegates directly: this is addressed, not ignored — candidate financial-disclosure and conflicts statements (§20), mandatory disclosure at filing (§22), and the gift ban and lobbying registration rules that apply throughout the convention itself (§47) are the mechanisms built to reach it.

§20 Delegate Eligibility and Conflicts of Interest

A convention’s own integrity rules must be unimpeachable, and are most exposed when the convention’s subject touches the interests of the people seated at it. Each state’s enabling law requires candidates and delegates to: be qualified electors of the state; file a public financial-disclosure and conflicts statement; disclose campaign funding for any delegate election; and abide by the convention’s gift, lobbying, and recusal rules (Part X). This handbook does not recommend a blanket bar on all sitting officeholders, since that raises its own ballot-access and equal-protection questions. It does recommend a narrower, specific exclusion, well-precedented and worth adopting: Hatch's convention-procedures bill barred sitting members of Congress and other federal officeholders from serving as delegates — a sensible rule, given that the convention exists precisely to act on a subject Congress will not. Beyond that line, candidate guidance favors delegates without a direct financial or professional stake in whatever the amendment would change — consistent with the 1933 “no political axe to grind” rationale.

§21 Instruction and Faithfulness

Each state binds its own delegates to the subject its application named, and does so before a single delegate is elected. How a state binds them is the state’s own prerogative, and three approaches are legitimate. What is not legitimate is instructing a delegate on substance — telling them which answer to reach inside the subject — for the reasons given below.

Option one: a scope-only faithfulness statute. This is the baseline this handbook recommends, and the one most states will find easiest to pass. The statute (a) limits the delegate’s authority to the single subject stated in the application and the call; (b) voids any vote cast outside it; (c) provides for orderly replacement of a delegate who breaches — by the next-highest vote-getter, in the elected model — without a special election; and (d) requires a public oath to the instruction before the delegate is seated. It fixes the boundary and nothing else.

Option two: binding to a citizens’ assembly. A state may go further and tie its delegates to the deliberated judgment of a randomly selected, demographically representative assembly of its own citizens, rather than to a subject-matter boundary alone. This is a more demanding and more expensive mechanism, and it binds delegates to the people rather than to the legislature — which is precisely what distinguishes it from the commissioner model rejected below. Appendix D sets out the design, the working precedents, and the honest costs.

Option three: no additional state instruction. A state may also elect delegates under its enabling law and add no separate binding statute at all, relying on the convention’s own germaneness rule and the ratification threshold to hold the line. This is a weaker position for that state, and this handbook recommends against it — but it is a real choice, not a failure, and a state that makes it does not break the process for anyone else. The next paragraph explains why.

No coalition should wait on faithfulness legislation, and no critic should treat its absence as fatal. The two mechanisms that do the heaviest anti-runaway work — the germaneness rule enforced on the convention floor (§42) and the thirty-eight-state ratification threshold (Part XII) — operate no matter what any individual state has or has not enacted. Germaneness binds the body itself, not one delegation; ratification binds the outcome. A faithfulness statute is an additional, state-specific lock bolted on top of that floor by a state that wants one. It is worth having, worth passing, and worth recommending — but it is not a precondition, and a convention in which only some states have adopted one is not thereby unsafe. Rappaport, examining these same statutes, reached a compatible conclusion by a different route: he judged legal controls of this kind potentially useful but uncertain enough that he would not rest the process on them, resting it instead on structural and political controls.

The approach is already on the books. Several states — among them Georgia, Florida, Indiana, Arkansas, and Utah — have enacted Article V convention-delegate statutes, distinct from though modeled in part on the better-known faithless-elector laws that govern presidential electors. Each binds a delegate to the same three things: the state’s application, the congressional call, and the state’s instructions. A delegate may not vote to consider or approve an “unauthorized” amendment, meaning anything outside that subject; a vote cast in violation is nullified; the delegate is subject to replacement; and in several states the act is a crime. These statutes are the most developed expression of the front-line, in-convention anti-runaway mechanism available today. A model statutory clause is in the companion Model Legal Language document.

The legal footing is stronger than it once was, though not settled. In the 2020 “faithless elector” cases, Chiafalo v. Washington and Colorado v. Baca, the Supreme Court held unanimously that states may bind presidential electors to their pledge and remove those who defy it. Whether that reasoning extends to Article V delegates — who, like electors, are constitutionally created officers without a clear textual guarantee of independence — is a serious argument that no court has yet tested in this context, and this handbook does not treat it as decided. It is not a new argument either: writing a decade before Chiafalo, Rappaport made the same argument. A state relying on it should cite it for what it is — strong support, not settled law.

Mutable does not support “legislature instructions”. The mentioned existing statutes should be borrowed from with open eyes, because of where they come from and whom they were written for. The "faithful delegate" or "delegate limitation" template is not a set of independent state innovations; it is coordinated model legislation produced by the convention movement — most prominently the Convention of States project's "Commissioner Selection Resolution" and "Criminal Penalty Bill," drafted and vetted by figures including Robert Natelson and Michael Farris, and promoted alongside the balanced-budget and term-limits convention drives. More important than the provenance is the model embedded in the text: these statutes assume an appointed, legislature-controlled delegate — a "commissioner" whom the legislature selects, instructs, and recalls at will. That is the opposite of the popularly elected delegate this handbook prefers (§18). Mutable therefore takes the mechanism from these laws — bind the delegate to the subject, void the out-of-scope vote, penalize the breach, seat an alternate — but not the selection and recall framework that surrounds it. Adapted to an elected delegation, the instruction flows from the voters and the application rather than from a legislative majority that can rewrite it mid-stream, and replacement is tied to objective triggers rather than discretionary legislative recall (§23). The faithfulness machinery is sound; the appointed-commissioner scaffolding it currently sits inside is not Mutable's model.

§22 “So You Want to Be a Delegate?”

Public-facing guidance for prospective delegates should make the role, the commitment, and the transparency expectations clear before anyone files:

1. Publish a bio and position statement. Election officials collect a standardized biography and a written position statement from every candidate and publish them in a single public guide, in print and online, well before the election.

2. Understand the time commitment. Up to one month at the convention site, plus travel and preparation. Treat it like a deployment.

3. Know the single-subject pledge. Candidates affirm in writing that they will confine themselves to the subject stated in the application and honor their state’s instruction. The commitment itself is uniform — every state’s candidates make this pledge; the statute that enacts and enforces it is each state’s own law, and its exact text will vary even though what it commits a candidate to does not.

4. Disclose. Candidates file the financial-disclosure and conflicts statement (§20) at the time of filing; it is published with the bio.

5. Provide for your own absence. Ultimately each delegate is responsible for arranging their work and home affairs — see the employer-accommodation best practice in Part VIII.

§23 Vacancies, Alternates, and Recall

Each state’s enabling law should name alternates (next-highest vote-getters, in the elected model) to fill vacancies without a special election, and should specify the narrow grounds for replacing a delegate — resignation, incapacity, or a faithfulness breach — with a transparent, on-the-record process. Mid-stream partisan recall (e.g., a legislature flipping control and swapping its slate) is exactly the destabilizing scenario scholars warn about; Mutable’s model disfavors discretionary recall and ties removal to objective triggers only. The commitment to objective-only removal is what this handbook recommends uniformly; each state enacts it through its own enabling law, and the exact replacement procedure will vary state by state within that shared commitment.

Part VIDelegate Elections

§24 The Role of State Election Officials

Delegates are popularly elected — the recommendation this handbook makes throughout and does not treat as optional — so the state’s chief election officer (typically the Secretary of State) runs the delegate election on the existing election machinery. Their core duties:

1. Set and publish the election calendar within the window fixed by the state’s enabling law (the 1933 model: gubernatorial call → election in a set number of days → seated convention).

2. Administer candidate filing (petition signatures and/or a modest fee), verify eligibility, and publish the certified candidate list.

3. Collect, standardize, and publish every candidate’s bio and position statement in an official voter guide.

4. Design a ballot that lets candidates declare as pledged (for / against / unpledged) on the proposed subject, on the 21st-Amendment model.

5. Conduct the election, canvass results, and certify the delegation and alternates.

6. Administer the public delegate oath to the state’s instruction before certification is final — in every state, since the oath is what makes the instruction enforceable and public.

§25 Ballot Design and the Pledged-Delegate Model

Candidates declare their position on the ballot, so the delegate election itself functions as a public vote on the question. The 1933 conventions worked exactly this way: delegates publicly pledged how they would vote, which let voters treat the election as a near-referendum on the amendment. This handbook adopts the same transparency — candidates declaring for, against, or unpledged, printed on the ballot — with one essential adjustment, because 1933 was a ratifying convention and this is a proposing one.

§26 Candidate Filing and Transparency

Filing requirements should be low enough to be broadly accessible (the project’s legitimacy depends on ordinary citizens being able to run) but structured enough to ensure seriousness — e.g., a nominating petition plus a nominal fee, sized to screen for seriousness rather than to test a candidate’s finances, with a hardship waiver available, on the 21st-Amendment-state pattern. Each filing is published as it comes in, and the complete, final candidate list — every bio, position statement, and disclosure, compiled together — is published no later than a fixed date before the election, so voters can compare the full field at once rather than piecing it together as it trickles in. No material about a candidate that election officials rely on should be non-public.

§27 Election Timeline

The delegate-election schedule follows from the convention date, which means it is mostly Congress’s to set in the call. Two schedules are possible, and the trade-off is speed against cost. The 1933 pattern gives a concrete benchmark: Washington’s enabling statute fixed the delegate election at not less than one month nor more than six weeks before the convention itself, and Utah’s governor called for an election on October 10, held it November 7 — four weeks later — and seated the convention December 5, four weeks after that. A dedicated election needs comparable lead time: figure four to eight weeks from the call to the election, and another four to six weeks from the election to a seated convention, once filing, campaign, and canvass periods are built in.

Where speed is essential or no regular election is near, a standalone special election on the compressed schedule below is ideal — the path the 1933 repeal conventions used:

Part VIIMoney: Compensation, Costs, and Funding

§28 Who Pays for What

Mutable’s allocation rests on a principle: a convention proposing an amendment to the United States Constitution is national work, not state business, so the people doing it are paid by the federal government, while the host state — which reaps the prestige and the economic benefit — supplies and operates the venue.

What It Would Cost, in Round Numbers

A one-month convention on this pathway — 535 delegates, federal pay and per diem, a host state for venue and security — runs on the order of $75 million all-in in current dollars. One line dominates everything: security. Treated as a National Special Security Event it is roughly $50 million on its own (range $35–75M); delegate salary (one month of House pay, about $7.8M), per diem (about $2.6M), venue, staffing, and the official record together come to under $20M combined.

Presented here are estimated planning figures; a Convention Cost Model supplement, with adjustable assumptions, accompanies this handbook.

§29 The Federal-Funding Precedent

Federal funding of a national convention is not hypothetical. In 1977, Congress appropriated $5 million (roughly $27–28 million in today’s dollars) to fund both the state delegate-electing conventions and the national women’s conference, including scholarships to widen participation. That precedent supports a federal appropriation covering delegate compensation, per diem, convention administration, and the public record — and supports funding measures (like participation scholarships) that make the delegate pool look like the country.

§30 Per Diem and Travel Mechanics

Per diem should follow the published GSA rate for the host locality so it is objective and non-negotiable. Delegates book lodging and file per-diem and reimbursement paperwork through a government HR liaison (Part VIII), not ad hoc, so that every dollar is documented and auditable — consistent with the transparency principle. Travel funded by the delegate’s state should likewise run through a documented state process.

Part VIIIGovernment HR and Administrative Support

§31 The Role of Government HR

A dedicated government HR function — staffed federally for convention-wide matters and mirrored by a host-state liaison office — should handle the administrative load so delegates can focus on deliberation. Its remit:

1. Onboarding and credentialing of every delegate, alternate, and staffer.

2. Per-diem enrollment and payment processing against the GSA rate.

3. Hotel booking and lodging coordination for delegations.

4. Salary administration for the convention term (federal payroll).

5. Travel-reimbursement processing (coordinated with each state’s travel funding).

6. Disclosure and conflicts filings — intake, publication, and recordkeeping.

7. ADA and accessibility accommodations for delegates and the public.

8. A single help desk so no delegate has to navigate multiple bureaucracies.

§32 Employer Accommodation: the Reservist Standard

Mutable’s best-practice standard is that a delegate’s regular employer should treat convention service like a reservist’s deployment: the employer makes every reasonable effort to accommodate the absence, holds the position, and does not penalize the delegate for serving. Mutable should advocate for state enabling laws to codify a job-protection provision on the reservist model.

§33 Credentialing and Onboarding

On arrival, each delegate completes a single onboarding pass: identity and certification check against the state’s certified delegation; oath confirmation; credential and secure-access issuance; payroll and per-diem enrollment; disclosure verification; and a briefing packet (the rules, the schedule, the single-subject charge, and the ethics rules). Credentials should visibly distinguish delegates, alternates, staff, press, and public.

Part IXThe Convention Proceedings & Apportionment

A convention needs working rules the moment it convenes — quorum, officers, a voting method, a threshold to propose, a way to enforce its own subject limit, transparency practice, and a term. None of that is optional in the sense that a convention could simply have no rule on these points and still function. What is a matter of choice is which specific rule to adopt on each one, and the sections that follow state Mutable’s recommendation for each, argued with its reasoning and its strongest counterargument, not presented as a neutral default. A complete model rules package embodying every recommendation in this Part is in the companion Model Legal Language document, Part B, ready for a Committee on Standing Rules to propose on day one (§36) — but the convention retains full authority under §34 to adopt different rules than the ones recommended here.

§34 Rulemaking Authority

Consistent with the dominant scholarly view, the convention adopts and may amend its own rules; neither Congress nor the states prescribe them (Congress may, at most, recommend). The convention may alter even its voting rule — but never to escape the single-subject limitation, which derives from the applications and the call, not from the rules.

§35 Convening and Roll Call

The convention is called to order by a temporary presiding officer named in the congressional call (§11), who immediately conducts a roll call of the states. The roll call at convening is by state — the traditional “is [State] present and ready to be seated” call that establishes which delegations have arrived, not a headcount of every individual delegate. Individual delegate presence is tracked separately once business begins, for the quorum-of-delegates rule in §36. Until the convention adopts its own rules, it operates under general parliamentary law — the common practice that governs any deliberative assembly before it writes its own rules, and under which even a germaneness objection to an out-of-scope motion will lie (§42). It does not default to one state, one vote — that would contradict the one-person, one-vote apportionment this handbook recommends (§37). The only prior national precedent points the other way: at the 1977 National Women’s Conference, the closest modern analogue, delegates voted individually, adopting each plank of the national plan by a majority of the delegates present, not by state bloc — this is the precedent for how a national convention votes once assembled. A separate question, taken up in the table below, is how many delegates each state gets in the first place; that apportionment question has real state-level precedent through Reynolds v. Sims, but no confirmed national-convention precedent tying delegate count to population the way the House-plus-two formula does.

§36 Day One: the Committee on Standing Rules

A Committee on Standing Rules presents a complete proposed rules package on the first day; the sponsoring coalition and host state should see that a draft package is prepared in advance so no time is lost, but nothing in it binds the convention. (Making the package a congressional mandate would clash with the convention’s authority over its own rules — §11, §34 — so it is prepared as a service to the convention, not imposed on it.) Debate follows, and the convention adopts a final structure by a two-thirds vote — remember, the convention ultimately decides its own rules.

The package presented should, at minimum, fix:

1. Officers (president, secretary, parliamentarian) and the manner of their election.

2. The voting rule on the floor (one person, one vote), and the threshold specifically to propose an amendment — the convention’s one binding, consequential output. Ordinary floor business, adopting rules, referring a matter to committee, procedural motions, runs on simple majority, the same way Congress itself operates.

3. Quorum (a majority of seated delegates) and the manner of casting and recording an individual delegate's vote.

4. The committee structure (Rules, Credentials, Drafting, Style) and referral procedure.

5. The single-subject germaneness rule and the point-of-order process to enforce it (§41).

6. Debate, amendment, and reconsideration procedures.

7. The transparency rules, including the narrow closed-session procedure for committee negotiation: open sessions, livestream, the journal, and recorded votes by default.

8. The duration rule and the deadlock-vs-progress standard for any extension (§39).

9. The form of the final product and the transmittal procedure (Part XI).

§37 Apportionment and Voting Weight

On the floor, every delegate casts one vote individually. The convention does not vote in state blocs. Two decisions are easily run together here and must be kept apart: the apportionment question — how many delegates each state sends — and the voting-unit question — whether those delegates vote as individuals or whether each state casts a single bloc vote. Section 19 settles the first (a House-plus-two delegation, popularly elected); this section settles the second. This is not a contradiction, though it is worth being explicit about the two different levels at work. §34 already confirms the convention has full authority over its own rules — Mutable is not claiming the legal power to bind that choice. What follows is a recommendation for what the convention should choose once it gets there, backed by the ABA’s conclusion, the Reynolds line of reasoning, and the weight of state-convention practice — the same way this handbook argues for a position everywhere else, rather than a claim that the convention lacks the power to decide otherwise.

The tradeoff, stated honestly. One-state-one-vote has a genuine pedigree: it is how the 1787 Convention voted. But that is precisely the model the ABA calls archaic, suited to a confederation of sovereigns we no longer are; and in our own time it is promoted chiefly by the organized convention movement — ALEC, the Convention of States project, and Robert Natelson — whose coalition it happens to advantage. The serious argument for it is prudential: small states might be likelier to apply, and to reach thirty-four, if promised an equal floor vote. That deserves weight, but it does not prevail. Small states already hold a controlling check at the other end: ratification is state-equal, and thirteen of the least-populous states can block any amendment however the convention voted. There is no need to malapportion the proposing body to protect an interest the ratification stage already protects — and doing so would forfeit the democratic legitimacy that a one-person, one-vote convention carries into the country's judgment of its work.

One structural point is not contested at all: the convention meets and votes as a single, unicameral body, not split into two chambers. Every American constitutional convention on record has operated this way — a structural corollary of one-person, one-vote and majority rule, and a norm strong enough that recent scholarship of the state-convention record calls it a universal expectation. Mutable's model rules package (Part B) reflects this without further debate.

§38 Officers

The convention promptly elects a permanent president, plus a secretary. Both are elected from among the seated delegates and keep their vote — presiding over or recording a body does not strip a delegate of their vote, the same way a House Speaker still votes though rarely does except to break a tie or on a major question. The parliamentarian is different, and deliberately so: a professional, non-delegate appointee with no vote at all. The president presides, rules on order (subject to appeal), and signs the transmittal. The secretary keeps the journal and certifies the record. The parliamentarian advises but does not decide — a deliberate check against a presiding officer steering outcomes through procedure.

§39 The Convention as a Deliberative Body: Hearings and Testimony

A proposing convention is a deliberative body, not a rubber stamp (§21): within its single subject it reaches its own judgment on the text. To do that well it must be able to inform itself — and the public — by hearing testimony. Nothing in Article V forbids this, and every comparable deliberative body does it; but because the convention writes its own rules, this is stated as practice the convention’s standing rules should adopt, not a procedure imposed from outside.

A workable model the rules can provide for:

1. A substantive committee — or the committee of the whole — may invite testimony. Written submissions and scheduled oral testimony from constitutional scholars, election administrators, practitioners in the convention’s subject, state officials, and reform and watchdog organizations, should be invited to participate.

2. Opposing and dissenting voices are invited deliberately. A balanced, bipartisan witness list is part of the legitimacy of the result: the convention should hear the strongest case against a provision it is weighing, not only the case for it.

3. All testimony is public and on the record. Taken in open session, livestreamed and transcribed like every other proceeding, with written submissions published. No private briefing of delegates by interested parties — the gift, lobbying, and recusal rules (Part X) govern that.

4. Structured and time-limited. Hearings run on a published schedule, with set times for testimony and for delegate questioning, on the congressional-hearing pattern, so the convention’s one-month term is not consumed by them.

5. Distinct from public comment. Hearings are the convention pulling information in — it invites specific witnesses to help it draft. Public comment (Part X) runs the other direction: anyone may submit input on their own initiative, whether or not the convention asked. Both are public and both feed the same record, but one is solicited and one is not, and neither substitutes for the other.

6. Distinct from closed negotiation sessions. Hearings are testimony-taking, and they are always public without exception. A committee's closed session is a different thing entirely — delegates negotiating language among themselves, permitted only under a narrow, recorded-vote procedure (§36), and never producing anything but a public report. Hearings inform; closed sessions negotiate; neither is ever a private vote. Congress’s own practice draws this same line: House Rule XI requires the vote to close a committee session to happen by recorded vote in open session, and any measure the committee later approves, including votes held in markup, becomes part of the public committee report. Negotiation may be private; the vote and its outcome are not.

§40 Duration and Extension

The convention sits for the one-month term recommended in §12. An extension requires an affirmative finding by the convention itself — by the same two-thirds used to adopt the rules — that more time would yield an outcome rather than entrench a deadlock. A convention unable to make that finding should adjourn and report honestly that it reached no proposal; a single-subject convention that cannot agree should not be kept alive indefinitely.

§41 The Proposal Threshold

The convention proposes an amendment only on a two-thirds vote of the delegates, parallel to the threshold Article V sets for Congress. The threshold is formally unsettled — the default rule for a deliberative body would be a simple majority, though Article V requires two-thirds when Congressproposes. A supermajority proposal: (a) signals broad, bipartisan agreement, which any subject-limited amendment needs to reach 38 states; and (b) blunts the “narrow-majority runaway” critique. The trade-off — a higher bar to produce anything — is acceptable given the stakes.

§42 Enforcing the Single Subject on the Floor

The germaneness rule is the convention’s internal anti-runaway lock. Any delegate may raise a point of order that a motion, amendment, or proposal exceeds the subject of the call. The president rules; the ruling is appealable to the floor. A proposal ruled non-germane is out of order and cannot be voted. Two of these locks are structural and need no state’s cooperation to operate: this germaneness rule, enforced by any delegate’s point of order, and the 38-state ratification threshold, which is the final word regardless of anything else. Delegate instruction and state faithfulness statutes add further, recommended protection on top — real and worth having, but not a precondition for trusting the process elsewhere, as §21 explains. Together, this is the structural answer to the runaway-convention fear: not a promise that every state will legislate, but a design where the subject cannot leave its lane even if none of them do.

This lock is not hypothetical; the U.S. House has enforced a germaneness rule — Rule XVI, clause 7 — since 1789, barring any floor amendment on a subject different from the matter under consideration, and by precedent that principle applies even before a body adopts its own rules, under general parliamentary law. At Pennsylvania's 1967–68 convention — limited by its enabling act to four subjects — delegates tried to introduce proposals outside that mandate, including one to direct state money to non-public schools. The convention president ruled the strays out of order, the body confined itself to its subjects, and the voters ratified every question it ultimately produced. Clear historical proof that a presiding officer's germaneness ruling, backed by the floor and the ratification check, actually holds a convention to its mandate.

§43 The Record

The convention keeps a complete public record: a journal of motions and recorded votes by delegate; a verbatim transcript; and a preserved livestream archive. The record is the raw material of legitimacy and of any later judicial review, and it is the antidote to 1787-style secrecy. Every floor session, every roll-call vote, and the convention's product are part of the record without exception; the narrow committee-negotiation exception is itself logged — the fact, duration, and stated ground of every closed session are permanent entries in the journal, so even what is not transcribed is never hidden.

Part XMedia and Public Transparency

§44 Transparency Is the Default

The convention meets in public. That is the single greatest contrast with the 1787 convention, and the default is near-absolute rather than absolute. The 1787 delegates deliberated in total secrecy for four months, with no public sessions and no record released in most delegates' lifetimes; that blanket secrecy, more than any single decision it produced, is what fuels modern runaway fears. Mutable rejects that model without swinging to its opposite extreme. Every floor session is open, every roll-call vote is recorded, and the convention's product is public; the only exception is the narrow, logged, time-limited allowance for committee negotiation (§36) — a more open standard than Congress itself operates under, and one that recent research on deliberative bodies suggests actually improves, rather than undermines, the odds of an honest, negotiated outcome. 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.

§45 Media Accreditation and Accommodation

1. Open, neutral accreditation. A simple, published process open to outlets of every size and ideology, plus independent and digital press; criteria are content-neutral and capacity-based, never viewpoint-based.

2. A real press center. Filing desks, power, connectivity, a clean broadcast feed, and a briefing room — furnished by the host state (Part IV).

3. A free program feed. The host state’s broadcast feed is offered free to any outlet and to the public, so coverage is not gated by network resources.

4. Regular briefings and a spokesperson. Scheduled on-record briefings from a convention spokesperson, plus access to committee chairs and the president, so reporting is accurate rather than speculative.

5. Equal access. No exclusive deals; pooled coverage only where space requires, with the pool feed shared to all.

§46 Public Participation and Observation

Beyond the press, the public itself should be able to watch — in the gallery and online. The host 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. The 1977 model of open state conventions feeding a national plan is the template for making the public a participant, not a spectator.

§47 Lobbying, Gifts, and Ethics During the Convention

A convention cannot let outside money influence its own drafting. Ethics rules should require: public registration and disclosure of anyone lobbying delegates; a strict gift ban; recusal on disclosed conflicts; and full transparency of any organized advocacy on the floor. These rules should be in the standing-rules package and enforced on the record. The integrity of the process is itself part of the message.

Part XIPresenting the Results

§48 The President’s Transmittal Letter

When the convention proposes an amendment (or adjourns without one), the president drafts a transmittal letter to Congress accompanying the result — the convention’s formal hand-off, on the model of the 1977 conference’s plan-of-action report to the President and Congress. A model letter is in the companion document, Part D. It should contain:

1. Background. How the convention was triggered (the application family and count), called, and conducted, with the single-subject charge stated.

2. The proposed amendment text, exactly as adopted, with the vote tally.

3. A record of the deliberation: the committees, the alternatives considered, and the rationale, so Congress and the states understand the proposal.

4. The convention’s recommendation on the mode of ratification — see §49 — with reasons.

5. Certification by the president and secretary, and a reference to the complete public record (journal, transcript, archive).

§49 The Convention’s Recommendation on Ratification

Although Congress chooses the mode of ratification, the convention should make a reasoned recommendation in its transmittal — and Mutable’s recommendation is ratification by state conventions, for the reasons developed in Part XII. Putting the recommendation in the transmittal ensures the case reaches Congress at the decisive moment, framed by the body that did the work.

§50 Certification and Transmission

The certified proposal and transmittal go to the President of the Senate and the Speaker of the House. The convention’s job ends with proposal and transmittal; it has no power to enact.

Part XIIRatification

§51 The Two Modes, and Why It Matters

An amendment becomes law only when 38 states ratify, by legislatures or by conventions, as Congress directs. The choice is strategically central whenever an amendment would restrict the very people asked to ratify it, because in that case the current system’s primary beneficiaries are the very legislators who would otherwise vote on ratification.

Ratification's significance goes beyond strategy, though — it is also the last word, doctrinally. This is not insurance against a runaway convention: the four independent locks described in §42 are what keep the convention on its subject, and nothing here relaxes them. What ratification protects is different — the finality of a decision the people have already knowingly made. Once matching applications, delegate instructions, faithfulness laws, and floor germaneness have all operated as designed and a text goes to 38 states for a genuine up-or-down vote, the ratification itself is the constitutional judgment — not a court's after-the-fact opinion about a technical scope question nobody but a lawyer would recognize as a defect. State courts have treated it that way for a century: once the people ratify, they decline to unwind the result over exactly this kind of claim. The last time any court even entertained the argument was 1975, in Rhode Island, and it was dismissed as moot. The four locks are what prevent a runaway. This doctrine is what prevents a legitimately-approved amendment from being relitigated out of existence by a lawyer years later.

§52 Mutable’s Recommendation: Ratifying Conventions

§53 How a State Ratifying Convention Works

There is no uniform federal template — in 1933 the 38 state conventions followed a variety of procedures — so each state legislates its own enabling act. The recurring 1933 pattern, which Mutable promotes as a model, ran:

1. Congress, in proposing the amendment, specifies ratification by convention and may set a ratification deadline (seven years, in 1933).

2. Each state enacts an enabling act fixing the timetable and the delegate-election rules.

3. The governor calls an election of convention delegates within a set window.

4. Delegates are elected — typically having declared for, against, or unpledged — so the election functions as a public vote on the amendment.

5. The convention meets (often briefly — in 1933, sometimes a single day), and ratifies or rejects the amendment as written, with no power to alter it.

6. The result is certified to the federal government; ratification by the 38th state completes the amendment.

Note: a ratifying convention, like a legislature, may only accept or reject — it cannot change a word. Mutable has prepared model enabling-act language for ratifying states as a follow-on to this handbook, mirroring the 1933 state convention laws.

§54 Ratification Timeline and Margin

The sponsoring coalition should plan the ratification campaign around any congressional deadline and around the arithmetic that 13 states can block ratification — Justice Scalia’s observation that a relatively small population, concentrated in the smallest states, can defeat an amendment. That math cuts both ways: it is why the proposal must be genuinely bipartisan and affirmatively framed from the very first state application, and why the convention should propose by supermajority (§41). Ratification strategy is downstream of every legitimacy choice made earlier in this handbook.

Appendix APrecedents at a Glance

Appendix BMaster Timeline Checklist

Appendix CSources and Key Authorities

Selected authorities, grouped by weight. Neutral and institutional sources carry the most weight; reform-advocacy and convention-movement sources are cited for their arguments or their procedural models, not as disinterested authority.

Tier 1 — Neutral and institutional — neutral or governmental; the firmest ground

• 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 (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) — exclusion of sitting members of Congress as delegates. congress.gov/bill/93rd-congress/senate-bill/1272

• Coleman v. Miller, 307 U.S. 433 (1939); Hawke v. Smith, 253 U.S. 221 (1920); Dyer v. Blair, 390 F. Supp. 1291 (1975); Reynolds v. Sims, 377 U.S. 533 (1964)

• Chiafalo v. Washington, 591 U.S. 578 (2020), and its companion case Colorado Department of State v. Baca — the “faithless elector” cases holding states may bind and remove presidential electors who break their pledge; the extension of this logic to Article V delegates is a serious argument, not a decided one supremecourt.gov/opinions/19pdf/19-465_i425.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

• David E. Pozen & Thomas P. Schmidt, The Puzzles and Possibilities of Article V, 121 Colum. L. Rev. 2317 (2021) — the leading treatment of Article V’s textual and doctrinal indeterminacy columbialawreview.org/content/the-puzzles-and-possibilities-of-article-v

• Russell L. Caplan, Constitutional Brinksmanship: Amending the Constitution by National Convention (1988) — the first systematic legal study of the convention clause archive.org/details/constitutionalbr00capl

• 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 (§5) researchonline.lse.ac.uk/id/eprint/225/1/open-door8.pdf

• Jill Lepore, The Philosophy of Amendment, 112 Calif. L. Rev. 2249 (2024) — the Jorde Symposium lecture prompting Pozen’s response; a skeptical-but-engaged account of America’s amendment drought lawcat.berkeley.edu/record/1311641

Tier 2 — Reform-skeptic — independent of the movement; argue limits may fail

• Center on Budget & Policy Priorities (Leachman & Super), States Likely Could Not Control a Constitutional Convention (2017) cbpp.org/research/states-likely-could-not-control-constitutional-convention-on-balanced-budget-amendment-or

• Walter Dellinger, The Recurring Question of the “Limited” Constitutional Convention, 88 Yale L.J. 1623 (1979) https://scholarship.law.duke.edu/faculty_scholarship/316/

• Michael Stokes Paulsen, A General Theory of Article V, 103 Yale L.J. 677 (1993) https://openyls.law.yale.edu/server/api/core/bitstreams/2168002e-9d76-49ac-bc4b-71ebc5613d6e/content

• Walter Olson (Cato Institute), commentary on the open questions of Article V https://www.cato.org/commentary/article-v-constitutional-convention-wrong-idea-wrong-time

• Michael B. Rappaport, Reforming Article V: The Problems Created by the National Convention Amendment Method and How to Fix Them, 96 Va. L. Rev. 1509 (2010) — argues the convention method is structurally defective and should be replaced by state drafting; also canvasses the limited/unlimited debate and the legality of state controls on delegates ssrn.com/abstract=1748854

• Common Cause coalition statement (with the League of Women Voters and others) https://www.commoncause.org/resources/coalition-statement-opposing-an-article-v-convention/

• Scholars Strategy Network, delegate-selection & representation problems scholars.org/contribution/delegate-selection-representation-problems-and-difficulties-article-v-convention

Tier 3 — Convention-movement — advocacy; cited for procedural models, not politics

• R. Natelson / Independence Institute “Article V Information Center” (one author) — “Counting to Two Thirds” (Federalist Society Review); “Founding-Era Conventions,” 65 Fla. L. Rev. 615 (2013) fedsoc.org/fedsoc-review/counting-to-two-thirds-how-close-are-we-to-a-convention

• Michael Rappaport, The Constitutionality of a Limited Convention, 81 Const. Comment. 53 (2012) https://scholarship.law.umn.edu/cgi/viewcontent.cgi?article=1793&context=concomm

• ALEC model application — corporate-funded advocacy; cited only as a drafting template for the aggregation clause, not as authority https://alec.org/publication/article-v-a-handbook-for-state-lawmakers/

• Convention of States / COSAction materials conventionofstates.com

Appendix DOptional State-Level Deliberative Mechanisms

Everything in this appendix is optional. Nothing in Parts I through XII of this handbook depends on any state adopting it, and a coalition that adopts none of it loses no protection this handbook otherwise provides. What follows is a menu for states that want to bind their delegates more deeply than a subject-matter faithfulness statute alone — tying a delegate’s vote not just to the right topic, but to the deliberated judgment of a representative sample of their own state’s people.

The Underlying Method: Deliberative Polling

The mechanism draws on “Deliberative Polling,” a methodology invented by Stanford political scientist James Fishkin in 1988 and applied more than 150 times worldwide. A random, demographically representative sample of a population — typically several hundred people — receives balanced briefing materials, deliberates in small and large groups, questions competing experts, and is measured for opinion change before and after. Fishkin ran the largest American test of the method himself: the 1996 National Issues Convention, a $3.5–4 million, roughly 500-person exercise broadcast on PBS ahead of that year’s presidential primaries.

Does This Work for Constitutional Change Specifically?

The clearest test case is Ireland. Its Convention on the Constitution (2012–14, two-thirds randomly selected citizens, one-third elected officials) recommended the change that became the 2015 marriage-equality referendum; its successor Citizens’ Assembly (2016–17, 99 randomly selected citizens, five sessions, 25 expert witnesses) recommended the change that became the 2018 abortion referendum. Since 2012, assembly recommendations have led to three ratified constitutional amendments and one defeat.

One honest complication is worth stating plainly: at least one academic analysis found the final public referendum vote on abortion was actually more favorable than the assembly’s own internal vote — which cuts against any claim that the assembly produced uniquely “better” judgment rather than simply anticipating where public opinion was already heading. The mechanism is a genuine, tested tool. It is not a guarantee of a particular outcome, and this handbook does not present it as one.

Mongolia offers the closest analogue to formally wiring deliberation into an amendment process: a 2017 law requires a deliberative poll before its parliament may consider any constitutional amendment, and the country has used the process for two successful amendments since, most recently adding proportional representation to its electoral system. Iceland has also used deliberative polling to inform constitutional amendments.

Lawrence Lessig’s Three-Pillar Strategy, Adapted for Article V

Harvard law professor Lawrence Lessig has proposed adapting this tool specifically to Article V convention delegates, in a three-part strategy.

Pillar 1 — Delegate binding. Discussed at §21: the 2020 “faithless elector” cases (Chiafalo v. Washington and Colorado v. Baca) establish that states may bind and remove presidential electors who break their pledge. Lessig’s strategic addition is a specific design choice: if a state binds its delegates to the judgment of its own people — through a citizen assembly, rather than through the legislature’s own preference — the legislature itself should have no power to override that binding after the fact. Binding to the people, not to the legislature, is what keeps this mechanism democratic rather than simply another version of legislative control.

Pillar 2 — Citizen assemblies. A state convenes a large, randomly selected, demographically representative assembly of its own citizens. Participation is mandatory, jury-style, and genuinely compensated — travel, government per diem, lost wages, job protection, and a meaningful bonus, not token pay. Participants receive briefing materials on the amendment topics under consideration, deliberate in small and large groups over an extended period, and vote. A state statute then provides that no delegate may vote to support a proposed amendment whose topic at least 60 percent of the assembly did not support.

One nuance is worth being precise about, since it changes how large a departure this actually is from this handbook’s core model: the binding rule, as Lessig states it, is keyed to the topic under consideration, not to specific draft language. Read this way, an assembly blesses a direction; it does not dictate the words. That leaves meaningful room for the delegate to negotiate actual text at the convention — which is what makes this compatible with, rather than a replacement for, the “bound as to subject, not as to substance” principle in §21.

Pillar 3 — Preemptive rejection resolutions. A state legislature passes a resolution in advance, stating it will reject any amendment arriving from a convention that lacks genuine democratic safeguards — regardless of that convention’s subject or who is running it. Because ratification requires 38 states, a single legislative chamber in each of just 13 states can block any amendment; a resolution like this uses that leverage proactively rather than waiting to exercise it after a convention has already finished its work. This is not a mechanism for designing a state’s own delegation — it is a strategic tool a state can use to press any convention, including one it does not control, toward the safeguards this handbook recommends. States building a Ratification Lock statute, of the kind discussed in Part XII, are already most of the way to this tool.

How This Fits the Core Model

None of this appendix changes the core recommendation in Parts I through XII: popularly elected delegates, apportioned House-plus-two, voting individually, bound to their state’s chosen subject by a faithfulness statute. A citizen assembly does not replace that statute; where a state adopts one, it operates as an additional, more demanding layer on top of it. And nothing here should be read as an endorsement of every claim made by sources that also discuss these tools — for example, one site presenting Lessig’s strategy also asserts that one-state-one-vote is the “dominant view” on convention voting mechanics; §37 shows that claim is contested, not settled, and this handbook’s own position is unchanged by anything in this appendix.

Honest Cautions

Two more, beyond the ones already stated. First, Chiafalo’s extension from presidential electors to Article V convention delegates is a serious legal argument, not a decided one; no court has tested it in this context, and a state relying on it should say so plainly rather than treat it as settled law. Second, a genuinely representative, adequately compensated citizen assembly is expensive and administratively demanding — real per diem, real job protection, and real staff to run the process are not incidental costs, and a state should budget for them honestly rather than assume the mechanism is free because Ireland’s was comparatively modest in scale.

Sources for This Appendix

• Lawrence Lessig, Making an Article V Convention Safe for Democracy, N.Y. Rev. Books (Apr. 17, 2023) nybooks.com/online/2023/04/17/making-a-constitutional-convention-safe-for-democracy-lessig

• James S. Fishkin, Can Deliberation Cure the Ills of Democracy? (Oxford Univ. Press, 2025) global.oup.com/academic/product/can-deliberation-cure-the-ills-of-democracy-9780198944416

• Chiafalo v. Washington, 591 U.S. 578 (2020), and Colorado Dept. of State v. Baca (see Appendix C) supremecourt.gov/opinions/19pdf/19-465_i425.pdf

• Ireland’s Citizens’ Assembly and Convention on the Constitution — overview of process and outcomes electoral-reform.org.uk/the-irish-abortion-referendum-how-a-citizens-assembly-helped-to-break-years-of-political-deadlock

• James S. Fishkin, Deliberation Can Save Democracy (2025) — discusses Mongolia’s 2017 law and its constitutional amendments persuasion.community/p/deliberation-can-save-democracy

End of handbook. Companion documents:

Mutable — Model Legal Language for a Limited Article V Convention (state application, standing rules, delegate-instruction statute, transmittal letter, illustrative amendment language).

Mutable — Model State Ratifying-Convention Enabling Act

Mutable — Model Convention Cost Spreadsheet