Gaps in the Record: Teaching with the Constitutional Convention

How do we know what happened in the Pennsylvania State House over the summer of 1787 when delegates convened there to write the federal Constitution?

On its face, answering this question doesn’t seem to pose any insurmountable obstacles. Afterall, the meeting was one of great historical import; the men who gathered in it were elite, literate, and even used to the idea of preserving records. Surely, it should be easier to answer than similar questions about what happened in any myriad of past events of less apparent significance or involving actors who did not or could not record their thoughts or actions. 

Figure 1: Details of the Constitution’s framing were not known until years after the conclusion of the convention and signing of the document. Howard Chandler Christy, Scene at the Signing of the Constitution of the United States, 1940. Howard Chandler Christy, Public domain, via Wikimedia Commons.

Yet, providing an account of what transpired in that series of famous meetings is much harder than one with a passing interest might suspect. The question of what took place behind closed doors over the summer of 1787 started to interest me as I was writing my book, Democracy in Darkness: Secrecy and Transparency in the Age of Revolutions. I worked with the Max Farrand edited collection of convention records throughout my graduate career and frequently turned to them to mine for evidence of what delegates thought (or at least expressed) about various topics. Among those was the issue of secrecy in government, including at the convention itself.

As I read through these records looking for what deputies reportedly said or wrote down about secrets and their place in politics, I began to reflect on how the secrecy surrounding the deliberations affected the very records I was consulting. The fact that the delegates met in secret rendered the process of forming the Constitution fundamentally opaque. The effect, as I argue in my book, was to help vest the finished product with an air of unanimous approbation and transcendent authority.

Too often, historians turn to records of constituent or legislative deliberations from the eighteenth century as a source to draw quotations from early American political figures, without questioning their accuracy. Even in meetings that were open to the public, the technological limitations of the era made recording speech anywhere near verbatim difficult—and this is on top of considerations like space constraints, printing costs, intended audience, or the distortions of memory. When a meeting was held behind closed doors, like the Constitutional Convention, there is the additional possibility of distortion due to the intervening time between when notes were taken and when they were revisited, edited, and published. Teaching with these types of sources highlights their limitations and, for me, forced a more critical engagement with them in my own scholarship. 

Figure 2: Later depictions imagine the mood or tone in the room over the summer of 1787, but it can be hard to nail down with the limited sources we have. Franklin at the National Convention/Lossing-Barritt (Philadelphia: n.p., ca. 1840-1890). Photography. Retrieved from the Library of Congress.

When I embarked on teaching a historical methods course for undergraduate history majors and minors, it struck me that using the records of the Constitutional Convention as a case study might be a useful way to get students thinking about the construction of the archive, biases of sources, and the challenge of conjecture in writing history. To set up a unit on working with primary documents to write a research paper, I devised a lesson plan to encompass a single 75-minute class meeting.

As preparation, I assigned the students a podcast interview with Mary Sarah Bilder on Ben Franklin’s World in which she discusses the findings of her book on the story of James Madison’s notes from the convention: Madison’s Hand. In addition to giving background on the convention and legislative note-taking in the eighteenth-century, Bilder goes into how she uncovered evidence about the way Madison wrote and then extensively revised his notes. The podcast provides a good entry point into the class session, especially to start thinking about what Madison’s notes—the most complete we have from that summer—can and cannot tell us about the Constitution and its formation.

To begin the exercise, I pass out a packet to each student containing all the notes included in the Farrand volume for a single day of the Convention: May 29, 1787. I selected this date because it is early on and evidently contained discussion of meeting logistics in addition to being the day the Virginia Plan was introduced. Furthermore, the different degrees of detail and points of emphasis in each set of notes immediately highlights the challenge of producing a definitive account of what transpired in that session. 

Figures 3a-b: The official journal from May 29, 1787, provides a limited account of what happened in the meeting that day, focusing on the establishment of procedural rules. Journal pages for May 29 in Max Farrand, The Records of the Federal Convention of 1787 (New Haven: Yale University Press, 1911), 1:15–16. Retrieved from the Library of Congress.

To start, I ask the students to read through the official journal (just two pages) for that date. Once they have done this, we pause and I ask them what strikes them about it. What impression do you get of the meeting that day and what was discussed? Did it seem like a long day? Could you get a sense of the feeling in the room? Could you look at this and tie any views to any specific people; would you feel comfortable quoting anyone from this source?

Students immediately note the brevity of the notes and the lack of any kind of speeches recorded—most conclude that they would definitely not be able to quote anyone from this record. The day’s meeting appears to have been short and mostly focused on establishing procedures. Many note an apparent consensus prevailing on the proposed rules and the introduction of “sundry propositions” by Mr. [Edmund] Randolph and a “draught of a foederal government” by Mr. Charles Pinckney, both apparently to be discussed later (Farrand 16). In terms of the feeling in the room, some students generally suggest it seemed efficient or matter of fact, while one or two usually conclude that there is little they could say about the atmosphere based on this source alone. What’s a historian to do in such a situation? First, look for additional sources.

In this instance, I have some at the ready for them. I give students a further ten minutes to read through the next set of notes: those belonging to Madison. These are much more extensive than the official journal record (seven pages); they include one vote tally on a particular question, they attribute names to ideas raised, and—perhaps most significantly—they present segments of fairly detailed speech from particular deputies. Most notable among these is the recorded speech Edmund Randolph gave to introduce the Virginia Plan, which Madison calls the “main business” of the day (Farrand 18). In fact, the bulk of his notes for this date are made up of a record of what Randolph purportedly said and the contents of the plan he laid out. Madison gives just two brief sentences at the end to note that “Mr. Charles Pinkney laid before the house the draught of a federal Government which he had prepared to be agreed upon between the free and independent states of America” (Farrand 23). 

Figure 4: Madison heavily edited his notes after the conclusion of the convention, as Mary Sarah Bilder has shown in her book, Madison’s Hand. James Madison, John Payne’s Copy of James Madison’s Original Notes on Debates in the Federal Convention of 1787. Manuscript/Mixed Material. Retrieved from the Library of Congress.

When we debrief after their consideration of Madison’s notes, most students immediately identify how much more extensive they were than the official journal. Many come away with a different impression of the meeting that day; it was not devoted mainly to rules, but more to discussing the Virginia Plan. This record conveys more of a sense of gravity, even urgency, to the meeting that is not present in the official journal. Typically, one or two students remember Bilder’s discussion of how Madison revised his notes later and pick up on a difference in how Randolph’s plan is described as “the establishment of a national government” (Farrand 16) in the journal compared to “revising the foederal system” (Farrand 18) in Madison’s records. While some hesitate to say they would quote Randolph directly from these notes, most conclude that they would feel comfortable attributing particular ideas or points to him.

At this point, we pause and I have students read a page from Bilder’s book (pp. 180–81) in which she describes Madison asking Randolph in 1789 to recreate his speech from May 29, 1787, introducing the Virginia Plan. She recounts how Randolph handed over his notes, but refused to “dilate” the speech, as he put it (Bilder 181). Randolph apparently “found it impossible to retrace the subject” and noted that he would “mingle inadvertently much of what I have heard since, without being able to separate it from what occurred then” (Bilder 181). Nonetheless, as Bilder details, Madison proceeded to reconstruct the speech based on what Randolph had furnished and his own memory and existing rough notes.

Figure 5: James Madison’s notes are the most extensive we have from the convention. James Madison: Fourth President of the United States (New York: Nathaniel Currier, 1841). Courtesy, American Antiquarian Society.

I then ask the students if they would revise their approach to Madison’s notes based on knowing this detail about how he composed them. While some suggest this information is not too surprising and that they would continue to trust the notes about the same as they had upon first inspection, others’ confidence is more shaken. Usually someone says something along the lines of: the notes might not be totally made up, but to cite them as direct evidence of what was said seems tricky. I ask them what they would do if they were still intent on writing an account of that day’s proceedings. As good historians in training, everyone generally suggests seeking out more sources—specifically, wanting to cross-check with additional notes available from the date.

Luckily for them, I have curated those as well. I give them another ten minutes or so to flip through the last three sets of notes included in the Farrand volume for May 29: the notes of Robert Yates (about one page), James McHenry (about three pages), and William Paterson (about one page). All three sets of notes include little, if anything, of procedural discussions in that session; Yates and McHenry note Randolph’s focus on the defects of the Confederation, while Paterson merely lays out the points he proposes as part of the Virginia Plan; neither McHenry or Paterson note that Charles Pinckney introduced a plan at all.

Students usually identify some small points of convergence across several of the sets of notes. For one, Yates writes that Randolph gave a “long and elaborate speech,” (Farrand 23) which seems to confirm the sense that this was the focus of the day’s session—an interpretation further backed by McHenry’s first line that “Governor Randolph opened the business of the convention” (Farrand 24). According to Yates, Randolph “candidly confessed that they were not intended for a federal government—he meant a strong consolidated union, in which the idea of states should be nearly annihilated” (Farrand 24). This is backed by Paterson’s record, which has Randolph saying: “We ought to be one Nation” (Farrand 27). 

Figure 6: Edmund Randolph introduced the so-called Virginia Plan on May 29 in the convention, though the details of what he said are difficult to pin down. Constantino Brumidi, Edmund Randolph, Head-and-Shoulders Portrait (Washington, D.C.: Detroit Publishing Co., 1904). Photograph. Retrieved from the Library of Congress.

McHenry’s notes go into much more detail on Randolph’s criticisms of the confederation—its “imbecility,” as he puts it at one point (Farrand 25)—and also introduces the idea that the plan was presented as a response to defects in the state governments. A brief footnote at the end of McHenry’s records also advises that: “In all essential particulars McHenry’s copy of the Virginia Plan is identical with that of Madison. It is accordingly omitted here” (Farrand 27). The editor’s interjection seems to reinforce the accuracy of Madison’s plan report.

At this point, we take stock as a class of what we feel confident saying about what happened that day in the convention by considering the following questions:

Would you be able to/feel confident quoting anyone from the meeting? What about describing what they said or conveyed?

What is the effect of Madison including procedural notes where the others don’t? How does this create a sense of greater trust in his notes as more comprehensive or official?

Would you feel confident describing the length or mood of the meeting based on these notes?

What other sources would you want to consult in order to flesh out your description of what happened at the convention on this date? (This question provides a good opportunity to discuss how to weigh legislative notes with official journals and private correspondence, diaries, and/or memoirs. It also leads to a fruitful consideration of the lack of press reports due to the secrecy imposed on the convention and how that poses further limitations for historians.)

To conclude the exercise, I give the students a few final minutes to read from a secondary source describing this day in the convention: pages 66-67 of Carol Berkin’s A Brilliant Solution. What I like about using this passage is that Berkin’s book is clearly written to engage an interested audience beyond merely academics. As a result, she attempts to convey intangibles in her narrative—a prospect that provokes good debate among the students. How does she know that “Randolph struck a perfect note of humility and sincerity” (Berkin 66)? Was Madison “no doubt aware that all eyes were upon him” as the actual author of the Virginia Plan (66)? Is it accurate to say that after Randolph spoke of the confederation’s deficiencies “the specter of ruin and humiliation menaced the East Room” (Berkin 67)?

Figure 7: Delegates met behind closed doors in this room over the summer of 1787. Pennsylvania Assembly Room, Independence Hall. Reading Tom from Reading, UK, CC BY 2.0, via Wikimedia Commons.

This can open a thoughtful conversation not only about how historians work and write, but also about how the Constitution is interpreted and the difference between history and law. The exercise generates questions about the role of conjecture in historical writing, particularly how historians attempt to account for gaps in the sources and build narratives to capture things like the mood of a room when the evidence doesn’t explicitly lay it out. We talk about whether a historian is responsible for indicating when something is conjecture in writing, how they can do this, and the extent to which they must walk the reader through their process.

Getting students to closely consider the available notes from a single day of the Constitutional Convention proves to be an engaging way to talk about the challenges of using primary sources to build a narrative of the past. More specifically, it can also give rise to questioning how the Constitution gets interpreted legally, particularly through the framework of originalism. Most students walk away convinced that it’s a lot harder than it might seem to know what the framers were doing or thinking over the summer of 1787. If it’s the only takeaway from the class, it’s a worthwhile one.

Further Reading

Max Farrand, ed., The Records of the Federal Convention of 1787, 3 vol. (New Haven: Yale University Press, 1911).

Katlyn Marie Carter, Democracy in Darkness: Secrecy and Transparency in the Age of Revolutions (New Haven: Yale University Press, 2023).

Mary Sarah Bilder, Madison’s Hand: Revising the Constitutional Convention (Cambridge: Harvard University Press, 2021).

Carol Berkin, A Brilliant Solution: Inventing the American Constitution (New York: Harper Collins, 2002).


This article originally appeared in February 2024.

Katlyn Marie Carter is an assistant professor of History at the University of Notre Dame. She is the author of Democracy in Darkness: Secrecy and Transparency in the Age of Revolutions (New Haven: Yale University Press, 2023).