Skip to main content

You are here

The Founding Fathers:
A Conserving Caucus in Action

Winter 2014 - Vol. 56, No. 1


This essay appears in the Winter 2014 issue of Modern Age. To subscribe now, go here.


Several fine and highly readable accounts of the Philadelphia Constitutional Convention have appeared over the decades. What characterizes most of these works—as well as, we should add, the countless articles dealing with the Convention—is their focus not only on the delegates, their backgrounds and views, but also on the conflicts that arose between them during their deliberations. Perhaps most prominent among these readings is John Roche’s seminal essay “The Founding Fathers: A Reform Caucus in Action,” which casts delegates to the Convention as politicians rather than as theorists and emphasizes their employment of compromise to overcome conflict.1 The result, Roche writes, is that “the careful observer of the day-to-day work of the Convention finds no over-arching principles.” Such principles as appear to exist, he suggests, were retrospective justifications applied to the products of compromise. This focus is quite understandable, if for no other reason than that it provides a context for a better understanding of the dynamics of the Convention, how differences of interest and theoretical persuasion were reconciled, and why the Constitution took the form it did. Such is the case, to take the most dramatic example, with the well-known account of the “large state, small state controversy” that almost resulted in the breakdown of the Convention.

Yet this portrait of the Philadelphia Convention as a “reform caucus” overlooks the considerable extent to which the delegates were constrained by, and therefore only relatively modestly modified, long-established political forms in use in the American colonies and states for decades. This is to say, they showed an inclination to follow John Dickinson’s admonition, “Experience must be our only guide. Reason may mislead us.”2 In this sense, the Convention was less a reform caucus than a conserving one. Indeed, while it is undeniably true that the delegates were skilled at the art of compromise, the Convention was also characterized by a remarkable degree of consensus on fundamental matters of governance. This consensus reached not merely the republican underpinnings of the regime but also several particulars as to its form, such as bicameralism and the separation of powers.

Roche correctly notes that the framers were not abstract theorists, but they were theoretically sophisticated. The general absence of abstract theory in their debates may largely reflect the fact that few fundamental theoretical issues were in dispute; on the contrary, such conflicts as did exist pertained largely to the best practical means of realizing ideals on which the overwhelming proportion of delegates agreed. The sheer magnitude of the task before the delegates—that of uniting independent and largely sovereign states under one government—would probably have been impossible lacking a consensus on the basic principles of governance. Moreover, given the speed with which the Convention completed its work, this consensus clearly had to embrace a number of concrete matters of governance, principally those relating to the procedures and structures of government necessary for the realization of the basic principles.

Equally important, a major source of this consensus, we believe, is to be found in the American political tradition, starting with the principles and practices of government stretching back to the earliest colonial times. Prior forms provided the basis of broad consensus and set practical boundaries to the options available to the framers. Consequently, the most important areas of consensus did not need to be “arrived” at or achieved by compromise to begin with; they were supplied by experience. To put this otherwise, if we place the Philadelphia Convention and its handiwork into a broader historical perspective, we encounter again a trait in the American political tradition that was evident in the period leading up to the Revolution, that which Friedrich von Gentz identified as a “lack of abstract theory.”3 On Gentz’s showing, the American Revolution, quite unlike the French, could even be considered reactionary, since the colonists sought a return to the conditions that prevailed during the colonial period of “salutary neglect.”

We do not mean to suggest that the framers were copycats. Clearly this was not the case. Nor could it be, given the need to address the delicate issues surrounding federalism and what the role of the states in the structure and processes of the new government should be, matters about which the political tradition was largely silent. Moreover, they looked upon their political institutions and tradition with a critical eye, acknowledging the failures and weaknesses of state constitutions. In this sense, the prior experiences indicated what avenues should not be traveled, or they pointed to potential dangers in some of the institutional or procedural arrangements set before the Convention. Yet, as even this “negative” role reveals, the political tradition served to establish crucial parameters in the debates and deliberations of the Convention; that is, it not only provided the common grounds and shared experiences for a meaningful exchange of views; it also limited the range of potential alternative arrangements the framers would consider.

This consensus reached a remarkable array of issues that covered virtually the full range of questions political theorists might ask about a new regime, including its form—republican—and such institutional arrangements as separation of powers. The Virginia Plan, the template on which the Convention’s deliberations were based, supplies a compelling example. Virtually every feature of it that received serious consideration by the delegates can be clearly traced to prior forms. The plan of a lower house electing an upper house—famously attributed to James Madison’s study of David Hume4—in fact appears in the colonies as early as the Fundamental Orders of Connecticut and in the postrevolutionary constitutions of Georgia and South Carolina, as well as in Charles Pinckney’s draft of a national plan of government.5 Its commitments to bicameralism and establishment of three separate branches of government, as will be seen below, mirror nearly the entirety of state constitutions. The Council of Revision, comprising representatives from the executive and judicial branches, was based on a similar institution in New York. Only the national negative on state laws could be called innovative, yet this measure pertained largely to federalism and was, in any event, less a source of controversy at the Convention than a simple nonstarter that never stood a serious chance.

Similarly—and in stark contrast to the many historical accounts that emphasize the possibility that abuses of popular rule might drive the nation to monarchy—the delegates evince an almost unanimous consensus that the new government would not only be republican but would also tilt toward the populist end of the republican spectrum. Once the decisive question of whether the new regime would operate on individuals was settled, no one questioned that its center of gravity would be a lower house elected by the people for relatively brief terms. Indeed, given the paucity of evidence to the contrary, it is difficult to see how the idea that republicanism was in danger ever took such a strong hold on the historiography of the founding. The Convention never considered any regime other than republican; and, indeed, so prevalent was the consensus in favor of the republican form that the rare comments raising concerns about that prospect are notable chiefly for the almost bashful tone of apology that attended them. Dickinson, for example, spoke admiringly on June 2 of the British system, but hastened to add that a “limited Monarchy however was out of the question.” Hamilton, for his part, was sufficiently self-conscious about his British-style proposal on June 18 that he had to beg “Gentlemen of different opinions [to] bear with him” on the subject.

To be sure, disagreement over various aspects of these principles and how best they could be secured was substantial, though nowhere near as intense as that over a range of issues surrounding federalism (for example, what the role of the states should be in the central government, the extent of national power vis-à-vis the states)—the one issue, again, on which the tradition provided the least guidance. Nevertheless, the delegates—all of whom at least acknowledged the need for a stronger national government—recognized the necessity of tackling these issues. In relatively short order, consensus on these matters produced an understanding that the Articles of Confederation had to be abandoned, that the Convention would have to ignore that part of their instructions officially authorizing the Convention “for the sole and express purpose of revising the Articles of Confederation.” Thus, in several key areas of the Convention’s work, we can see broad consensus bounded by experience:

Abandonment of Articles and Scope of National Powers. The introduction of the Virginia (or Randolph) Plan on May 29, just the third working day of the Convention, was a clear indication that the Articles of Confederation would be scrapped. This plan abandoned the defining characteristics of the confederate model: its key institution—the “prime mover,” so to speak, within the system—was a popularly elected lower chamber of a bicameral legislature, with its representation proportioned among the states according to population, and its laws operating upon individuals rather than the states. To be sure, there were some who believed that the Convention, in deliberating on the Virginia Plan, was going beyond its mandate from Congress or the instructions of their state legislatures. On May 30, for instance, George Read of Delaware points out that “deputies from Delaware were restrained by the commissions from assenting to any change of the rule of suffrage” from that of equality of state representation. The major assault on the Virginia Plan comes on June 9 from William Paterson, who goes beyond charging that the Convention was acting beyond its legal authority in abandoning equal state representation to maintaining that such abandonment would not be endorsed by the people.

Paterson’s remarks, as well as his introduction on June 15 of the Small State or New Jersey Plan as an alternative to the Virginia Plan, however, must be placed in context. In many ways, his plan conformed to the Convention’s official mandate simply to alter the Articles, although, taken as a whole, these alterations were substantial. For instance, it did go a long way in meeting the objectives of those who wanted a stronger union, such as that envisioned in the Virginia Plan, by granting the central government taxing powers and control over interstate and foreign commerce, and by providing as well that its laws would be the “supreme law of the respective States.” This scope of federal powers extended to most of the same areas in which the Virginia Plan proposed to empower the national government. It did not, to be sure, include farther-reaching ideas such as Madison’s national negative on state laws. But the important point to observe is that the divisions between the two pertained not to the scope of federal powers but rather to the institutional forms necessary to ensure that the states did not encroach on the national regime—a possibility all understood to be a potential problem, as indicated by Paterson’s proposal to empower the national government to enforce the collection of taxes within the states.

The differences in institutional forms, to be sure, were no small things, but we can also see that the authority accorded to the national government by both the Virginia and New Jersey Plans reflects a remarkable degree of consensus within the Convention on matters that would, in the ratification debates, prove highly controversial. It would not be an exaggeration to say the plans agreed on all the important questions of authority. While other features of the New Jersey Plan—for example, those relating to the judiciary and plural executives—were not likely to gain much support in the Convention, the most notable and controversial provision was for a unicameral legislature with equality of state representation; that is, the retention of the Congress established by the Articles with even more extensive responsibilities. That Paterson would advance such a proposal in light of what had already transpired is, we believe, best understood as taking an extreme position to establish a “bargaining” stance from which a compromise could be secured that would embrace proportional representation in one chamber and equality of state representation in the second. Moreover, Paterson’s plan did not reflect any theoretical hostility to bicameralism: if the Congress was going to continue to serve as essentially a diplomatic body, the theoretical need for a second chamber dissipates. In any event, only two days—largely consisting of withering critiques by Wilson and Madison—are devoted to a discussion of Paterson’s plan. The abandonment of the Articles is rendered official on June 19 by a vote of 7–3, with the Maryland delegation divided, to abandon further consideration of the Paterson Plan and return to a consideration of the Virginia Plan as “altered, amended, and agreed to.” The issue of representation in the second chamber, however, remained and would prove troublesome.

The Separation of Powers. In Federalist No. 47, Madison writes of separation of powers that “no political truth is . . . of greater intrinsic value, or . . . stamped with the authority of more enlightened patrons of liberty.”6 A survey of the state constitutions adopted after the Declaration of Independence would seem to bear out Madison’s appraisal. Of the six states that prefaced their constitutions with a statement or declaration of rights, four expressly called for the separation of the three branches of government. The “Declaration of Rights” for the Massachusetts Constitution of 1780 contains the most elaborate statement regarding this separation and its purpose: “. . . the legislative department shall never exercise the executive and judicial powers, or either of them; the executive shall never exercise the legislative and judicial powers, or either of them; the judicial shall never exercise the legislative and executive powers, or either of them, to the end that it [the Massachusetts government] may be a government of laws and not of men.” The first article in the body of the Georgia Constitution of 1777 is more succinct: “The legislative, executive, and judiciary departments shall be separate and distinct, so that neither exercise the powers properly belonging to the other.”

While other state constitutions were not as explicit as Georgia’s, their basic frames of government clearly reveal a concerted effort to adhere to this principle. It bears emphasis that these rhetorical declarations of devotion to the separation of powers appear in the revolutionary constitutions as an explicit theoretical justification for institutions that had long evolved—a self-conscious invocation of the authority of Montesquieu. So prevalent was Montesquieu’s authority on this matter—he has been rated as the most often cited theoretical figure on the American shores during the era7—that any departure from the principle of separation of powers would have required clear justification. It is the absence of such commentary, not the presence of it, that proves the extent of consensus on this principle.

Thus, it is not surprising that the Virginia Plan embraced the separation of powers, even though Madison himself would come to regard its protections for the principle as insufficient. That the Paterson Plan, which sought to salvage the Articles, also incorporated separation of powers might, at first blush, seem unusual. Yet the mere fact that this plan provided for an energetic government that could enforce its will upon the states, quite unlike the state of affairs under the Articles, necessitated a division of powers in order to have any chance of success within the Convention.

Following on the analysis of Montesquieu, Federalist No. 47 makes it clear that the purpose of separation of powers is to prevent arbitrary rule. The decisive point to notice is that the Convention was undivided on the importance of that principle. The fault lines that emerged centered not on the theoretical foundations of separation of powers but rather on different opinions as to the institutional threats to it and the proper institutional responses: some delegates wanted to strengthen the executive and judiciary against an encroaching legislature; others, to weaken the presidency lest its occupant become tyrannical; and still more favored a strict and inviolable separation between all three branches. But the separation of powers was not an accidental by-product of extraneous institutional debates.

The degree of consensus surrounding the separation of powers is evident in the fact that certain important lines pertaining to it were drawn early in the proceedings. The Virginia Plan called for a Council of Revision that would empower “the Executive and a convenient number of the National Judiciary” to veto measures passed by the legislature. Both Madison and James Wilson were persistent and forceful proponents of uniting the judiciary with the executive, seeking adoption of this provision on June 6 and July 21 without success. What can rightly be called a third effort, on August 15, took the form of allowing either “the Executive” or the “Supreme Judiciary Departments” to exercise a veto: if both vetoed, “3/4 of each House” would be required to override; if only one, “2/3 of each House.” This proposal was soundly rejected by an 8–3 vote.

Why were Madison and Wilson so persistent in seeking this union of the two departments? The reason was to retain rather than to jettison the separation of powers. The principal justification for their efforts is to be found in Madison’s remarks of July 17 to the effect that at the state level the legislatures have aggrandized virtually all powers and that a meaningful separation of powers therefore does not obtain. He notes as well that “the Executives of the States are in general little more than Cyphers; the legislatures omnipotent.” On July 21 he repeats his concern in even more alarming and specific terms: “Experience in all the States had evinced a powerful tendency in the Legislature to absorb all power into its vortex. This was the real source of danger to the American Constitutions; & suggested the necessity of giving every defensive authority to the other departments that was consistent with republican principles.” On August 15, Wilson points out that, while tyranny was most commonly associated with a “formidable” “Executive,” “where the Executive is not formidable” tyranny should appropriately be linked with the legislature.

Thus, their primary purpose was to protect the two weaker departments, the executive and judiciary, from legislative aggrandizement—again, to preserve the separation of powers in practice. As Wilson put this at an earlier point in the debates, on July 21, “the joint weight of the two departments was necessary to balance the single weight of the Legislature.” Equally important, however, such a union would render the exercise of the veto power in the face of legislative aggrandizement more likely. The executive standing alone might not have the confidence, firmness, or courage, deficiencies overcome by uniting with the judiciary. Yet Madison, speaking on the same day, doubted that even this union of executive and judicial branches would do the job; “the Legislature,” he believed, “would still be an overmatch for them.”

The debate over the Council of Revision assumes importance because its opponents, while not opposed to the ends sought by its proponents, hold to a vision of separated powers and what its implementation required that eventually wins out and finds expression in the Constitution. On June 6 Dickinson succinctly set forth the overriding objection, namely, that the “junction of the Judiciary” with the executive branch “involved an improper mixture of powers.” Earlier, on June 4, Elbridge Gerry, apparently the most vocal opponent of this provision, observed that judging “the policy of public measures” was “foreign” to the judicial function. Following upon this, Rufus King maintained that “the Judges ought to be able to expound the law as it should come before them, free from the bias of having participated in its formation.” Caleb Strong, on July 21, makes essentially the same point: “The Judges in exercising the function of expositors might be influenced by the part they had taken . . . in framing the laws.” Luther Martin is even more emphatic in rejecting any role for the judges in the legislative process on largely different grounds: “A knowledge of Mankind, and of Legislative affairs cannot be presumed to belong in a higher degree to the Judges than to the Legislature.” Not unrelated to Martin’s objection is Pinckney’s observation regarding Madison’s final proposal of August 15 that “the interference of the Judges in the Legislative business . . . will involve them in parties, and give a previous tincture to their opinions.”

Looking at this dispute from a wider perspective, the major difference between those who favored this combination of executive and judicial powers and those who opposed it revolved around their respective estimates of the capacity of the legislature to overwhelm the executive and judicial branches. All, however, were united in a theoretical commitment to separation of powers, which in fact motivated these other disputes. Greatly influenced, no doubt, by what they witnessed taking place in the states after independence, those who sought to fortify the executive called for checks on the legislature that would be extremely difficult to overcome. Indeed, some suggested that the only sure remedy against legislative tyranny resided in giving the executive an absolute veto.8 Those opposed to these measures simply did not view the legislative threat in such an apocalyptic light.

At the same time, a wider perspective also reveals fundamental grounds of agreement that in the end served to facilitate compromise on one of the most critical issues confronting the Convention. Put otherwise, even though the proposal to unite the executive and judicial branches to fend off legislative encroachments was rejected, there was still a consensus that the executive needed a check on the legislature in order to prevent tyranny from that quarter. As Nathaniel Gorham observed during the debate on July 21, the last time the Council of Revision was proposed, “All agree that a check on the legislature is necessary.” And in fact, from an early stage in the deliberations (June 4), a solid majority backed an executive veto power whose override would require a two-thirds vote in both legislative chambers—the same requirement for an override of a governor’s veto contained in the New York and Massachusetts constitutions.

The disputes surrounding these issues were, to be sure, substantial, but they should not obscure the degree to which the delegates were, so to speak, deliberating in the same political universe. That is, they understood the need for the separation of powers; they did agree that the legislature would have to be checked; they did appreciate the reasons underlying their differences; and they were mindful of how their decisions would impinge upon other basic values, not the least of these being republicanism. In short, they shared the same basic objectives. Their eventual resolution of these differences came down to a matter of judgment: Would, for example, the independence of the judiciary suffer from even a partial union with the executive? Would a three-fourths override provision render the president too powerful? Would a two-thirds provision be sufficient to prevent legislative encroachment?

The issue of who should elect the president perhaps better illustrates the fundamental agreement on principle that pervaded the deliberations. The Virginia Plan provided for the election of the executive by the combined chambers of the legislature. Early in the deliberations key issues arise that will later be debated at some length: Who should elect the executive—the legislature or the people—and what should be his term of office? On June 13 the Committee of the Whole reports on the Randolph Resolutions affirming the Convention’s prior decision that the executive be elected by the national legislature for a term of seven years, adding the proviso that he “be ineligible a second time.”9 On July 17 the delegates vote unanimously for election by the national legislature but also, on a 6–4 vote, eliminate the proviso barring reeligibility. On this issue the delegates apparently accepted Gouverneur Morris’s July 17 view that “ineligibility . . . tended to destroy the great motive to good behavior, the hope of being rewarded by a re-appointment. It was saying to him, make hay while the sun shines.”

The ramifications of the decision to allow for reeligibility were almost immediately apparent. Jacob Broome indicated that with reeligibility he was now for a shorter terms of office, preferring a longer term if “he had remained ineligible.” But James McClurg, in moving “to strike out the 7 years, and insert ‘during good behavior,’ ” raised the issue that hit to the heart of the separation-of-powers principle. By eliminating the ineligibility clause, he argued, the executive “was put into a situation that would keep him dependent forever on the Legislature; and he conceived the independence of the Executive to be equally essential with that of the Legislature.” McClurg’s motion was supported by Gouverneur Morris, who at the same time indicated that “he was indifferent how the Executive should be chosen, provided he held his place by this tenure.”10

Madison, in more extensive remarks, reiterated the point that “the Executive could not be independent of the Legislature, if dependent on the pleasure of that branch for a reappointment.” Invoking the authority of Montesquieu, he pointed to the consequences of a union of these two branches: “tyrannical laws may be made that they may be executed in a tyrannical manner.” While it should be noted that even though Madison—unlike Wilson, with whom he shared almost identical views on the need for separation of powers—never endorsed life tenure for the executive, he nevertheless felt that McClurg’s motion deserved “a fair hearing & discussion, until a less objectionable expedient” could be found to prevent legislative tyranny.

George Mason regarded “during good behavior” to be only “a softer name for an Executive for life,” and he raised the specter of a “hereditary Monarchy”; an “easy step,” in his view, if the “good behavior” provision were adopted. This charge drew denials of any such intention from McClurg, Madison, and Gouverneur Morris. Perhaps the most incisive of these was Madison’s argument that the preservation of republican government “required some expedient for the purpose” of restraining the legislature, but “in devising it” that “the genuine principles of that form . . . be kept in view.”

On July 19, presumably in light of the difficulties attendant upon reeligibility, a majority voted in favor of a motion by Oliver Ellsworth for election by electors “appointed by the Legislators of the States” using a formula that would take state population into account in allotting electoral votes. But by July 24 a majority reversed this decision and returned to election by the national legislature, which prompted Gerry and Martin to move “to re-instate the ineligibility of the Executive a 2d time.” Gerry, acknowledging the need for an executive independent of the legislature, went on to contend that “the longer the duration of his office the more will his dependence [on the legislature] be diminished,” and he suggested terms of “10, 15, or even 20, years and [that he] be ineligible afterwards.”

Clearly there were those such as Gouverneur Morris for whom reeligibility was critically important and who, therefore, could not accept any provision along the lines suggested by Gerry. King, for instance, stated that reeligibility was “too great an advantage for the small effect it will have on his independence.” It is with this in mind that Wilson, recognizing the difficulties that “spring from the mode of election,” set forth a proposal designed to overcome them and allow for reeligibility: “the Executive be elected for 6 years, by a small number, not more than 15 of the Natl. Legislature, to be drawn from it, not by ballot, but by lot and who should retire immediately and make the election without separating.” While Gouverneur Morris believed the plan worthy of consideration, Wilson’s plan failed to gain any traction.

On July 26, after two days of debate and deliberation, Mason surveyed all the proposals suggested for the election of the executive—different modes of popular election, “election . . . by Electors chosen by the people,” election by state legislatures or by state governors, and Wilson’s lottery scheme—and concluded “that an election by the Natl. Legislature as originally proposed . . . was the best.” On his motion, the original provision for a seven-year term with no reeligibility is adopted by a vote of 6–3 with two states divided. This decision, as we know, did not settle the matter, nor given its deficiencies should we have expected it to. If nothing else, the debates and deliberation to this juncture, particularly in pointing up the difficulties associated with the election of the executive by the legislature, are sufficient to explain why the eventual resolution of this issue, election by the Electoral College, was gratifying to most of the delegates; namely, it ensured executive independence from the national legislature, along with a relatively short term and reeligibility. Added to this, of course, was the composite nature of the process that embodied both the federal principle, by providing a role for the states, and the national principle, by apportioning delegates largely on the basis of state population.

Again, the give and take over this issue has to be understood in a wider context. The experiences at the state level, coupled with the basic strictures of the separation of powers and republicanism principles, provided the grounds upon which the deliberations proceeded. Put otherwise, these principles, perhaps imperfectly provided for in the state constitutions, were uppermost in the minds of the delegates. The debates centered not on the principle itself but over what was required to ensure its realization in practice. In the last analysis, this was a matter of judgment. As Morris put it on July 24 amid the debates over executive-legislative relations, “It is most difficult of all to rightly balance the Executive. Make him too weak: The Legislature will usurp his powers: Make him too strong. He will usurp on the Legislature.”

Bicameralism. On June 20, in the midst of wide-ranging remarks in the Convention, Mason noted that while much had been said about the “unsettled mind of the people of America,” “he was sure” there were “two points” upon which “it was well settled,” namely, “an attachment to Republican Government” and “an attachment to more than one branch of the legislature.” He went on to observe that the constitutions of the states “accord so generally in both these circumstances, that they seem to have been preconcerted.” “This,” he believed, “must either have been a miracle, or have resulted from the genius of the people.” Attributing the attachment to bicameralism to the genius of the people comes close to the mark, since there is good reason to believe—contrary to the supposition that Americans sought to “copy” the British model—it naturally evolved over time.

What is widely regarded to be the first American constitution, the Fundamental Orders of Connecticut (1639), reveals the earliest stage of this evolution, with its provision for the election of magistrates and governor by the General Assembly. These magistrates, who are also members of the General Assembly, assumed executive responsibilities with the governor. Consequently they can be seen as standing somewhere between the legislative and executive departments. An amendment to the Fundamental Orders in 1645 enhances considerably the magistrates’ legislative dimension by providing that they vote as a unit within the General Assembly, a majority of their number being necessary for the passage of legislation.11 Finally, in keeping with the logic of this change, in 1698 the General Assembly is divided into two houses—an “Upper House” and a popularly elected “Lower House”—with the provision that “no act shall be passed . . . but by the consent of both houses.”12

The Connecticut evolution is not unique. In 1644 a Massachusetts ordinance, after noting “inconveniences” associated with “magistrates & deputies sitting together” in one assembly, ordained that “the magistrates may sit & act business by themselves” and that their concurrence is necessary for the enactment of laws.13 Likewise, the Rhode Island charter was amended in 1666 so that “the deputyes may sitt apart from the magistrates as a House by themselves; and consequently the magistrates to sitt as a House by themselves; and that of these two houses may consist the law makeing power.”14 The move to bicameral legislatures was not confined to New England: by 1643, the Governor’s Council, which had been part of the original Virginia Grand Assembly, a unitary body established in 1619, became a separate second legislative chamber, the other being the House of Burgesses.15 In fact, Virginia’s bicameral arrangement, wherein an appointed governor’s council constituted a second legislative chamber, is that which eventually emerges in all the royal and proprietary colonial governments save Pennsylvania and Georgia.

Without going into the details of the constitutions adopted by the states, both before and after the formal separation from Great Britain, certain trends should be noted. The most notable, of course, is that the governor’s councils—a feature of the royal colonies—are eliminated, though their functions are assumed by new constitutional arrangements. These arrangements varied somewhat. The New Jersey Constitution of 1776 provided that an upper chamber, the “Legislative Council,” should be elected, but provided as well that the governor, elected by both Houses, should be “President of the Council,” thereby still linking the council to the executive. Delaware’s constitution indicates the path followed by most states: the governor’s council was replaced by an elected body, the “Council,” which constituted the upper chamber. In addition, there was provision for a “privy council” whose duties were primarily executive in character. The Maryland constitution likewise provided for the election of the upper house and a “privy council” with largely executive duties. Following Virginia’s example, the North Carolina constitution provides for election to both legislative chambers and establishes a “Council of State” whose main function is to “advise the Governor in the execution of his office.”

Quite aside from the more or less natural evolution, there was widespread agreement for bicameral legislatures on theoretical grounds. John Adams’s widely circulated “Thoughts on Government” (1776), which offered principles that should guide states in the drafting of new constitutions in the wake of their independence from Great Britain, was a highly influential work that defended the necessity of bicameralism to maintain liberty. Here Adams, a proponent of “mixed” government, indicts unicameral legislatures on various grounds, the first being the “passion, flight of enthusiasm . . . or prejudice . . . productive of hasty results and absurd judgments,” inherent faults that require the steady hand of a second chamber either to stymie or correct. Like considerations lead Wilson, the most democratically inclined convention delegate, to insist in his 1791 Lectures on Law that a second chamber was absolutely necessary in light of the “passions” and “prejudices” of a single assembly that would, if left unchecked, eventually produce “despotism, injustice, and cruelty.”16 In short, the need for a second chamber was widely recognized across the political spectrum, a state of affairs that could not help but be reflected in the Convention.

The composition and character of the second or “upper” chambers in both the colonial and the period between the Declaration and the Constitution provided a consensus, to some degree tacit, on certain critical issues surrounding the form and function of the second chamber. To begin with, in keeping with both the morality and practice of the colonial and state governments, as well as with the more theoretical injunctions, they wanted an upper chamber that would be more sedate and deliberative than the lower. Above all, it was assumed that a second chamber, like the councils in the colonial and state governments, should closely examine the legislative output of the lower chamber with an eye to determining whether it was suitable and, if so, how it might be improved. To realize this, in turn, it was understood that the second chamber should have fewer members than the lower in order to facilitate the necessary deliberation.

There were still other critical features of a second chamber upon which there was consensus. That the terms for its members should be longer than for those of the lower chambers was a widely held position that gained strength even over the course of the deliberations. On June 12, Randolph proposes a seven-year term for the upper chamber in order to ensure “firmness & independence”; a proposal immediately supported by Madison on grounds that such a term would provide a critically needed “stability” in order to check the excesses of the lower chamber. Later, on June 26, again expressing his concern that the second chambers in the states have been no match for the lower chamber, he again maintains that “considerable duration ought to be given” the upper chamber, even terms of nine years. On the same day, Wilson observes that in addition to concerns over “anarchy & tyranny,” the Senate “will probably” have responsibilities relating to foreign relations and “treaties” that would render lengthy terms desirable in order to promote “respectability in the eyes of foreign Nations.”17

In this Wilson is implying that the Senate would share certain functions with the executive, not unlike the councils and upper chambers of the colonies and states. And, not unexpectedly, this turns out to be the case. Underlying all the formal constitutional provisions relating to term, size, and functions was a recognition, largely implicit, but articulated by Madison on June 26, that a properly constituted second chamber might well come to be respected by the people for its “wisdom & virtue,” thereby posing an effective and republican check on the ill-conceived measures of the lower chamber. Setting the minimum age for eligibility for the Senate at thirty, compared with twenty-five for the House, thereby seeking to provide for a more mature membership, may be understood as one small step in this direction. So, too, was their insistence on some form of filtered election, another staple of both the colonial and state governments.

This consensus is underscored rather than impaired by Paterson’s proposal for a unicameral legislature. Paterson’s proposal was made before consensus emerged that the new regime would operate on individuals, thus removing the theoretical motivation for bicameralism. Paterson’s legislature, like that of the Articles, was—to borrow John Adams’s metaphor—less a legislative body than a diplomatic assembly. The key fact to observe is that once the Convention agreed that it would instead function as a Congress with authority over individuals rather than states, all parties—including those still pushing for the states to retain some sovereignty—apparently agreed that a bicameral legislature was needed.

In sum, while the principled differences over who or what the Senate should represent have understandably been a focus of attention among students of the Convention, what is often overlooked are the areas of consensus or substantial agreement that rendered the delegates’ task manageable and the fact that this consensus arose substantially from experience. Had they begun their deliberations on a second chamber de novo, without the benefit of the prior political tradition, it is doubtful they would have reached any resolution.

* * *

The foregoing has attempted to establish that the delegates to the Philadelphia Constitutional Convention of 1787 began their work with a broad-based consensus on core theoretical matters and, further, that this consensus was guided and bounded by the long-standing practice of the states. The importance of this consensus for understanding the American political tradition is myriad. Most clearly, it underscores the fact that American political institutions are the products of a steady evolution reaching back to the earliest colonial institutions, not a sudden burst of innovation in Philadelphia, and still less of theoretical contemplation detached from experience. However, and equally important, the fact that the framers relied on experience does not diminish the theoretical sophistication or coherence of their work. This is true not merely because their gradualism and caution itself arose from a theoretical suspicion of detached reason. One of the most important features of the American political tradition is that it is characterized not by ad hoc institutional design—as Roche suggests—but rather by theory arising from practice. Put otherwise, Gentz’s distinction between the American and French Revolutions was that the former was not the product of “abstract” theory. It is far from the case, however, that a lack of abstract theory indicates an absence of theoretical grounding altogether. On the contrary, the American tradition exhibits a commitment to theory moored to practice.

Roche, arguing otherwise, describes Madison and Hamilton as “inspired propagandists with a genius for retrospective symmetry” for having, in The Federalist, draped the product of a “reform caucus” in theoretical garb after the fact.18 But this overlooks the extent to which the theoretical consensus at the Convention arose from a continuous practice of introspection nearly two centuries in the making before the delegates arrived in Philadelphia. Such institutions as bicameralism and separation of powers evolved from gradual and cautious speculation in the states that was anchored to practice and bounded by awareness of the limits of abstract reason. The framers’ silence on many theoretical matters reflects the extent to which this consensus existed, and their decision to embrace it, in turn, reflects both their own theoretical commitment to gradualism and their inheritance of a long and coherent theoretical tradition they chose to retain. That makes the Convention less a “reform” caucus than a conserving one.

Finally, the foregoing considerations also cast serious doubt on Madison’s much-vaunted role as “Father of the Constitution.” This encomium derives in considerable measure from his presumptive authorship of the Virginia Plan, which, as we have seen, served as the basis of the Convention’s substantive deliberations. Yet the theoretical arguments Madison adduced within the Convention and in The Federalist, especially with respect to separation of powers, seem largely incompatible with core features of the Virginia Plan. The case is complicated by the fact that only circumstantial evidence connects Madison to the plan. He does outline some of its core features in correspondence with Washington as well as Edmund Randolph in advance of the Convention, but these letters do not specify the modes of election—the lower house electing the upper and each of them electing the president—that seem incompatible with his position on separation of powers.19 The primary elements these letters do specify are either compatible with long tradition, such as bicameralism and representation, or—as in the case of the scope of federal powers—proved largely uncontroversial because their necessity was generally seen as self-evident. As we have seen, the only theoretically novel device these letters propose is Madison’s much criticized proposal for a national negative on state laws. Not only was this device repeatedly defeated, it would not be too much to say it never received serious consideration—a fact that concerned Madison enough that he predicted the Constitution would fail because of it. This suggests that Madison may have been accurate, not merely modest, when he famously described the Constitution as “the product of many minds.” The overwhelming influence of tradition on constitutional forms shows it was the product of many decades as well. ♦


George W. Carey (1933–2013), a longtime contributor to Modern Age, was one of the foremost authorities on the political theory of the American Founding. His voluminous writings included Basic Symbols of the American Political Tradition (with Willmoore Kendall), The Federalist: Design for a Constitutional Republic, and In Defense of the Constitution. He taught at Georgetown University for more than fifty years.

Greg Weiner is assistant professor of political science at Assumption College in Worcester, Massachusetts, and the author of Madison’s Metronome: The Constitution, Majority Rule, and the Tempo of American Politics.

1.     John P. Roche, “The Founding Fathers: A Reform Caucus in Action,” American Political Science Review 55, no. 4 (December 1961): 799–816.

2.     Elliot’s Debates, 424 (August 13). All references to the Convention are to Elliot’s Debates, to the date indicated.

3.     Friedrich von Gentz, The Origin and Principles of the American Revolution compared with the Origin and Principles of the French Revolution, trans. John Quincy Adams (Indianapolis: Liberty Fund, 2010).

4.     See, inter alia, Douglass Adair, “That Politics May Be Reduced to a Science” in Fame and the Founding Fathers (Indianapolis: Liberty Fund, 1998).

5. Pinckney’s plan calls for the House to elect a Senate “either from among themselves or the People at large.” All references to colonial documents and to postrevolutionary constitutions are to unless otherwise indicated.

6.     Alexander Hamilton, James Madison, and John Jay, The Federalist (Gideon edition), eds. George W. Carey and James McClellan (Indianapolis: Liberty Fund, 2000), 249.

7.     Donald S. Lutz, “The Relative Influence of European Writers on Late Eighteenth-Century American Political Thought,” American Political Science Review 78, no. 1 (March 1984): 189–97.

8.     Among those who at one time or another favored an absolute veto were Hamilton, Wilson, and Gouverneur Morris. Mason seems to have expressed the general sentiment of the Convention in his remarks on June 4. He concluded by contending that an absolute veto would be tantamount to ceding “all the rights of the people to a single Magistrate.”

9.     Ibid., 113.

10.     Ibid.

11.     Donald S. Lutz, ed., Colonial Origins of the American Constitution (Indianapolis: Liberty Fund, 1998), 239–40.

12.     Ibid., 253.

13.     Ibid., 90–91.

14.     Ibid., 208.

15.     See The assembly constituted in 1619 is today commonly referred to as the House of Burgesses and the first representative body in America. It is the first representative body, but the House of Burgesses did not come into existence until 1643 as the “lower” house of the Virginia legislature.


17.     On June 25 the idea of staggered elections—i.e., only a portion of the Senate seats being contested at any given national election—was apparently well received. This, in turn, meant that eventually the debate over length of terms came down to six or nine years, with one-third of the Senate elected either every two years or three years. This arrangement helped to satisfy to some degree both those who wanted stability and those who were concerned about accountability.

18.     Roche, “The Founding Fathers,” 804.

19.     Madison to Washington, April 16, 1787,; Madison to Randolph, April 8, 1787,