Who Picked the Committees at the Constitutional Convention?

Critical Thinking

September 13, 2018
by David O. Stewart Also by this Author


Journal of the American Revolution is the leading source of knowledge about the American Revolution and Founding Era. We feature smart, groundbreaking research and well-written narratives from expert writers. Our work has been featured by the New York Times, TIME magazine, History Channel, Discovery Channel, Smithsonian, Mental Floss, NPR, and more. Journal of the American Revolution also produces annual hardcover volumes, a branded book series, and the podcast, Dispatches

Through four months in the summer of 1787, passionate arguments over political principles filled the Pennsylvania State House while hard-nosed political horse-trading buzzed in the taverns and drawing rooms of Philadelphia. Fifty-five American politicians were writing a new charter of government for the United States, the Constitution. They produced the longest-surviving constitutional republic in human history, and that summer has drawn the rapt attention of historians and legal scholars ever since.

One part of that constitutional story, however, has proved elusive.  Although the debates among the delegates sometimes were electrifying and often were important, many of the most difficult questions were resolved by specially-appointed committees of a few delegates.[1]

Because key constitutional provisions were produced by committees, not on the convention floor, we might learn a good deal about the dynamics of the convention if we knew how the committee members were chosen. That, in turn might explain why certain delegates were included on a committee and why others were omitted, and that could illuminate much about the attitudes of the delegates toward each other, and toward the challenging issues before them.

But we don’t know how the delegates chose the members of each committee.  Neither the convention nor any delegate left a record of the procedure they followed. The most likely answer, it turns out, has been hiding in plain sight all along.[2]

The Committees
Two committees of five delegates handled the most daunting work of the summer: the Committee of Detail produced the first draft of the founding document; the Committee of Style produced the final draft. Despite their innocuous-sounding titles, those committees completed the essential work of synthesizing the elements of the founding document into a coherent whole.

In addition, eight “committees of eleven”—consisting of one delegate from each state then represented at the convention—resolved the thorniest questions of the summer, the ones that threatened to blow the convention up:

  • How to balance representation in the legislative branch between small and large states?
  • How to control the explosive risks of slavery, reassuring slaveholding states while not totally embracing the nefarious institution?
  • How to structure a presidency that had energy, but not monarchical power?

Many of the delegates who played key roles on the committees have largely been ignored by history: John Rutledge of South Carolina, Gouverneur Morris and James Wilson of Pennsylvania, and Edmund Randolph of Virginia. Rutledge chaired three critical committees, including the Committee of Detail, where Wilson and Randolph made important contributions. For the Committee of Style, Morris single-handedly wrote the final draft of the Constitution.

Through what process did the delegates choose these men to undertake the most delicate and important work of that summer?

Several answers to that question are possible: That the delegates chose members of committees by voting in their state delegations, with one vote for each state; or that each delegation chose the person who represented that state on a given committee; or that all the delegates voted together as individuals; or that they voted by different methods depending on the nature of the committee being selected.

When the delegates voted on substantive matters at the Convention, they voted only as state delegations, with each state casting a single vote.  That was always the practice in the Continental Congress, beginning in 1774. If a state had an even number of delegates who deadlocked on a motion, the state cast no vote on that motion. Voting by states gave extra influence to small states like Delaware, which were dwarfed in population and land mass by their larger siblings, Pennsylvania and Virginia. But per-state voting respected the principle of state sovereignty. So one possibility is that the delegates applied this familiar practice—one-state, one-vote—when choosing committee members.

Yet per-state voting would seem an odd practice to follow for the “committees of eleven,” which included one delegate from each state. Why should Georgia delegates vote on which Massachusetts delegate, or which New York delegate, or which Maryland delegate, would serve on a committee?  Would the Georgians even be able to make an informed judgment on which delegate had time or interest or background to be an effective member of that committee? Or which delegate best reflected the views of the other members of the state delegation? A far more plausible approach would be for each delegation to choose its own representative to those committees of eleven.

What about the two pivotal committees with only five members? It would be impossible for each state to choose one of its delegates for those committees, since no more than five states could have a member on each such committee. Consequently, the balloting for those committees was likely to have been on some other basis. Did the delegates vote in those instances as individual delegates, or as state delegations?

These questions have confounded decades of constitutional study, and the answers are important. The method of choosing committee members can shed important light on the mindset and intentions of the delegates at key points through the summer.

For example, the first “committee of eleven” addressed the issue that brought small states to the brink of abandoning the Convention:  whether each state should have an equal vote in the national legislature, or whether legislative representation should be apportioned on the basis of state population or wealth.

For that first committee of eleven, the members representing the large states were, surprisingly, delegates who had not been pressing for proportional representation, even though proportional representation was in the interest of those states.  Thus, two strong proponents of proportional representation—James Wilson of Pennsylvania and James Madison of Virginia—were passed over in favor of Ben Franklin and George Mason, respectively. Franklin and Mason had been far less engaged on the question, and might be expected to have been far more willing to compromise on it.[3] How did that happen?

If each state delegation chose its own committee member, then choosing Franklin and Mason were signals from the two largest states that they were open to compromise, no matter how passionate Wilson and Madison had been during floor debates.  Alternatively, if the full convention chose the committee members by per-state votes, then the small states tilted the committee’s membership in their favor by taking advantage of their greater numbers.  Or, if the committee members were chosen by all delegates voting as individuals, then perhaps the small states enjoyed an edge because their delegations often were just as large as those of the large states, while there were more small states than large ones.

Which is true?  Herewith the evidence.

The False Leads
Because the Convention’s proceedings were closed to the public, the first place to look for an answer is in the Convention’s own records.  The only official record is the journal maintained by William Jackson, secretary to the Convention. His work has sometimes been described as sloppy and sometimes has been defended, yet it remains the official source.[4] He did not try to create a transcript of the proceedings, but recorded only the motions and resolutions that were proposed and how they were decided.

A second source, sometimes accorded nearly official authority, is the record compiled by James Madison of Virginia, who set out to record the delegates’ debates through the long sessions. Of lesser authority are the notes of Robert Yates of New York, who abandoned the convention after six weeks, and occasional notes taken by delegates including James McHenry of Maryland and Alexander Hamilton of New York.[5]

On its first day in session, the Convention appointed a three-member committee to develop the body’s procedural rules. That committee, according to Jackson’s journal, was chosen “by ballot.” The Convention also selected Jackson as secretary, which Jackson recorded as also occurring by “ballot;” Madison’s notes reflect that five states supported Jackson while two voted for another candidate. Madison, like Jackson, used the term “ballot” to describe the selection process.[6]

In its report, that three-person rules committee addressed the selection of committees by the convention. Committee members shall be, the report states, “appointed by ballot,” with “the members who have the greatest number of ballots, although not a majority of the votes present, be the Committee.”[7]

The rules specified that each state would cast one vote on substantive issues, but that rule sheds no light on the question whether, when choosing committees, the “ballots” or “votes” were to be cast by states or by individual members. The rules are silent on that point.[8]  Remember that silence. It’s important.

Further textual study encounters more silence. When recording the choice of the first “committee of eleven”—one from each state—Jackson’s journal records the voting “by ballot;” Madison did also.[9]  Yet, as noted above, having each state cast a vote for another state’s committee member seems offensive to each state’s sovereignty, and also cumbersome.

Moreover, never does any of the sources record the votes cast for individual committee members. Thus, we do not know if delegate X was chosen for committee Y by the votes of six states, or by the votes of twenty-one individual delegates; if such a record were available, it might reveal an interesting regional or philosophical pattern or alliance among the delegates. The records, however, disclose no such information.

The picture grows more vexing with the selection in late July of the five members of the Committee of Detail. Jackson’s journal describes the Convention having “produced to ballot” for that panel, while Madison called it “a ballot for a committee.”[10] As noted above, as a practical matter a five-member committee could not be chosen by having each state delegation choose a member, so all of the delegates voted—either by states or as individuals—for those committee members. Yet both Jackson and Madison used the term “ballot” to describe the selection process for that smaller committee; they used the same term for the committees of eleven.

In short, the available texts from the convention provide no definitive answer. “Ballot” describes all of the occasions on which the delegates chose committee members, but that term does not reveal the nature of the voting contest. Voting by “ballot” could mean voting by states, by individual delegates, or sometimes by one method and sometimes by another.[11]

The Power of Precedent
With no clarity in the available texts, the remaining source is the practice of the American Congress since 1774, when the Continental Congress first convened. Many of the delegates were lawyers trained in the English common-law traditions, so they understood and accepted the logic of following prior precedent in legal matters.

In every Congress for thirteen years, as every delegate in Philadelphia knew, each state cast a single vote; the practice was incorporated in the Articles of Confederation, which took effect in 1781. The Articles also authorized Congress to appoint a “committee of the states” which would tend to business while Congress was in recess.  Each state could have one member on a committee of the states.  Congress also could create “such other committees . . . as may be necessary.”[12]

Based on its practice since 1774, Congress in 1781 adopted two rules for exercising the power under the Articles to appoint “such other committees.” For important issues, Rule 19 authorized the use of a “grand committee” with a member from each state. Rule 19 specifically stated that “each state shall nominate its member” of a grand committee. According to a leading study of congressional procedure, the Continental Congress always followed this practice in the years before the rule took effect.[13]

But Congress also had made ample use of its residual power to create smaller committees, relying on them to address discrete problems and then to disband, often in a matter of a few days or a week. The proliferation of three-member and five-member committees was astonishing.  In 1783 alone, Congress appointed 498 committees, or nearly four new committees in every three-day period; over eight years, Congress utilized an average of 280 committees each year.[14]

Congress’s Rule 20 governed the choice of those smaller committees, stating that “the states shall ballot for small committees,” adding that if the committee could not be completed with individuals receiving a majority vote (that is, the votes of seven states), the “house shall by a vote or votes determine the committee.”[15]  Although the “states” might cast an initial ballot for the members of small committees, it was mathematically impossible for more than one delegate to command a majority of state votes, which would leave at least two seats to be filled by the vote of the “house” (or four vacant seats for the five-man committees). Voting by individual ballot—by the “house”—surely was the predominant method for filling hundreds of committee seats every year.

Congress did not ordinarily record the voting on selection of committee members, but it did in February 1777, when it established a three-person committee to examine desertions from the army. Twenty-seven members of Congress were present that day; twenty-seven ballots were cast for members of the committee. The three delegates commanding the highest number of votes (nine votes, five votes, and five votes, respectively) were named to the committee.[16]

Indeed, in 1783 Congress amended Rule 20 to specify the procedure by which individual delegates voted for members of small committees.  The ballots were to be completed by the delegates “from their seats,” with one member from each delegation placing the ballots in a box that would be delivered to the president of Congress.[17]

This precedent—which was the contemporary reality well-known to the Convention delegates—provides a sensible template for projecting how the Constitutional Convention selected its committee members. The Convention’s Committees of Eleven may be likened to the “grand committees” of Congress, with each delegation selecting the member who would represent that state on the committee. That procedure preserved the prerogative of each state to determine which delegate would represent it, at a time when state prerogatives were jealously defended.

For smaller committees like the Committee of Detail and the Committee of Style, the delegates almost surely cast ballots as individual members of the whole house, as the Continental Congress did when choosing members of its small committees. Thus, a delegate from a small state might be chosen—as William Samuel Johnson of Connecticut was voted onto the Committee of Style, and Oliver Ellsworth from the same state on the Committee of Detail—solely because his talents were thought to match the task facing the committee, while other prominent delegations provided no delegates to that panel.

Only if the delegates voted as individuals within “the house” could they have selected Alexander Hamilton for the Committee of Style.  The New York delegation could not choose Hamilton for that committee because New York was not even officially present at the Convention at that time. New York’s legislature had specified that at least two of its delegates had to be present in order to cast the state’s ballot at the Convention, but Hamilton was the only New Yorker at the Pennsylvania State House after early July. With no state delegation to choose him, Hamilton could have been chosen only by the vote of the individual delegates.[18]

These conclusions have the considerable virtue of matching congressional precedent and practice before 1787, which dictated many of the procedures the Convention followed, including per-state voting and closing its proceedings to the public. These conclusions have the further virtue of making sense. That they are not grounded in an authoritative text of the Convention is a frustration, but not one that should obscure the best available explanation of the committee selection procedure.

[This article results from the enlightened prodding and ideas of William Treanor, John Vile, Mary Sarah Bilder, John Mikhail, and Richard Beeman, for which the author is most grateful.]


[1]E.g.,John R. Vile, “The Critical Role of Committees at the U.S. Constitutional Convention of 1787,” American Journal of Legal History, 48:147 (2006).

[2]In my book on James Madison, Madison’s Gift: Five Partnerships that Built America (New York: Simon & Schuster, 2015), 337, I complained that despite considerable effort, I had not been able to figure out how the delegates chose committee members, and reported a consultation with three other scholars of the Convention (Richard Beeman, John Vile, and Mary Sarah Bilder) who expressed similar frustration.  I have since continued to pick at that scab and herewith present the results of that work.

[3]Vile, “The Critical Role of Committees,” 159.

[4]See Richard Beeman, Plain, Honest Men: The Making of the American Constitution (New York: Random House, 2009), 70 (calling Jackson “an exceptionally poor choice” as Secretary). Mary Sarah Bilder, “How Bad Were the Official Records of the Federal Convention?” George Washington Law Review 80, no.6 (2012): 1674–1679.

[5]As explored in detail by Mary Sarah Bilder in Madison’s Hand: Revising the Constitutional Convention (Cambridge: Harvard University Press, 2015), Madison somewhat obsessively revised – or even enhanced – his notes for years after the convention.  That singular practice does not, however, affect the question addressed in this article.

[6]Max Farrand, ed., The Records of the Federal Convention of 1787 (New Haven: Yale University Press, 1911), 1: 2, 1: 4 (May 25, 1787).

[7]Ibid., 1: 9 (May 28, 1787).  Ties, the report added, would be resolved in favor of the member “first on the list in the order of taking down the ballots.”  It is not clear how a member would come to be “first on the list.” That may have referred to something as simple as alphabetical order; it may have referred to the custom of recording votes of states from north to south, from New Hampshire to Georgia.  E.g., Ibid., 1: 32 n.5.

[8]Ibid., 1: 8 (May 28, 1787).

[9]Ibid., 1: 509, 1: 516 (July 2, 1787).

[10]Ibid., 2: 97, 2: 106 (July 24, 1787).

[11]One scholar has suggested that the delegates voted as individuals on the selection of all committee members.  Jack Rakove, Original Meanings: Politics and Ideas in the Making of the Constitution (New York: Knopf, 1996), 379n38. That assertion is supported, however, only by a citation to Madison’s notes on the selection of the first “committee of eleven.”  Madison’s entry on that occasion states only that the committee members were selected “by ballot,” a term that both he and Jackson used to describe the selection of Jackson as Convention Secretary on May 25, when each state delegation cast one vote.  Accordingly, use of the term “ballot” sheds no light on the question addressed in this article.

[12]Articles of Confederation, Art. V, cl. 4; Art. IX, cl. 5.

[13]Journals of the Continental Congress, 20: 479 (May 4, 1781); Calvin C. Jillson and Rick K. Wilson, Congressional Dynamics: Structure Coordination, and Choice in the First American Congress, 1774-1789 (Stanford: Stanford University Press, 1994), 98.

[14]Ibid., 96-97.

[15]Journals of the Continental Congress, 20: 480 (May 4, 1781) (emphasis supplied).

[16]Journals of the Continental Congress, 9: 109 n.1 (February 12, 1777); Jillson & Wilson, Congressional Dynamics, 66.

[17]Journals of the Continental Congress, 24: 344 (May 15, 1783).

[18]The delegates voting for Hamilton plainly overlooked the further question of whether he was even a member of the Convention if he was not part of a state delegation then present.  George Washington, ever punctilious, noted that problem when he recorded in his diary for September 17, 1787, that the Constitution had commanded the “unanimous assent of 11 states and Colo. Hamilton’s from New York (the only delegate from thence in Convention).”


  • This is really interesting. Folks don’t often appreciate the the intricacies of committee assignments and rules. I worked on two committees in the House of Representatives and interned at a non-legislative committee in the Senate back in the dark ages. Rules and assignments play an outsized role that won’t show up in a written record and can be extraordinarily difficult to sleuth out in determining outcomes. Ironically, I’ve begun to conclude that it’s those grey areas that enable work to get done and decisions to be made. There were very good reasons to swear delegates to secrecy and close the shutters in the 18th century!

  • Many thanks for this most illuminating article. My ‘man’ Wilson was far more involved in the nitty-gritty of everyday practical politics than I thought.

Leave a Reply

Your email address will not be published. Required fields are marked *