1
The Senate was different, if not exceptional, from the start. But so were the presidency, the judiciary, and even the House of Representatives. If the Senate was exceptional it was not because of its elegant design and clarity of purpose. Emerging from the deliberations in Philadelphia, the Senate was rather like the committee’s version of a horse, which is to say a camel.1 The Senate was placed at the crossroads of the national system of separated institutions sharing power and at the intersection of state and national sovereignty. First, the Senate was half of a bicameral Congress endowed with legislative power. Second, it shared in the executive power of appointments and treaties, with a potentially decisive role in the composition of the Supreme Court. Finally, the politics of the Constitutional Convention added another special purpose—the Senate would be the institutional embodiment of federalism and state power in the national government. It was to be the smaller, wiser, and more detached legislative chamber, the institutional home of state power and state equality and the quasi-council to the executive on appointments and treaties. In this way the Senate became the crucible for resolving several of the thorniest problems of American constitutionalism. In particular, the Senate was exceptional for the manifest tension and outright contradictions between its roles as both a Senate of far-sighted and detached national statesmen and a Senate of delegates sent by self-interested state legislatures. That tension was at the heart of the Constitutional Convention, and from the time the Senate first achieved a quorum, on April 6, 1789, to today, it has never been resolved.
Its rich mixture of purposes and powers continues to make the Senate unique among the world’s political institutions. As the equal partner in one of the world’s most independent and robust legislatures, it is the most powerful upper house in the world. Whereas many countries have unicameral legislatures or upper houses or senates with significantly less power or even only symbolic importance, the US Senate combines a formidable array of institutional characteristics and powers that make it distinctive and autonomous. The higher age and citizenship requirements for members, as well as the longer and staggered terms, contribute, more or less, to differences in the character of the upper house compared to the lower. Similarly, state equality in the Senate contrasts sharply with the proportional representation of the House. The distinctions produced by elections are reinforced by the institutional elements. The House and Senate are granted nearly equal powers; in fact, the Senate has a few exclusive powers, including the approval of treaties and appointments, that give it a singular status among upper houses. And unlike many upper houses, the modern Senate has enjoyed at least equal (un)popularity and (dis)respect among the people.2
But it is this hybridity and complexity that have allowed the Senate over the years to conflate the various aspects of its constitutional role into a generalized idea that it has a special and different purpose that transcends the multiple but carefully defined roles given it by the Constitution. In particular, senators have drawn on and distorted the collection of attributes and responsibilities outlined above to justify what became the dominant feature of the Senate: its rules of procedure, which create and protect minority power, and the effect of those rules on Senate behavior. Modern senators have tended to portray minority power, as structured by Senate rules of procedure, as the sine qua non and raison d’être of their institution. This goes directly against what the Constitution actually mandated. The constitutional means to the end of better deliberation—first and foremost its smaller size, but also the age of senators and their length of citizenship, longer terms, and, originally, selection by state legislatures—have, over time, been displaced in favor of rules created by the Senate that empower minority obstruction.
This chapter focuses on key aspects of the Senate’s creation at the 1787 Constitutional Convention to show the Senate’s complex if not exceptional design: what the convention created, but also what it did not create. Each of the first three sections of the Constitution (Articles I on Congress, II on the executive, and III on the judiciary) divides into an opening section on composition and a second section on powers. That is, each article begins with how the branch is composed, particularly its mode of selection, qualifications for office, and duration of selection or appointment.3 The second part of each article is about powers, specifying the branch’s duties and authority. This chapter concentrates on two sides of the Senate’s composition. On one side were the elements of relative consensus, which were nearly everything about the Senate’s republican purpose and the characteristics suited to that, including age of senatorial candidates, length of citizenship, duration in office, indirect method of selection, and especially the small size of the Senate. On the other side was the element of nearly insurmountable conflict, which was the basis of representation: Would the Senate be proportioned by state population, like the proposed House of Representatives, or by state equality, like the existing Congress of the Confederation? The result of that blending of consensus and conflict was a hybrid and conflicted Senate.
Instead of providing a comprehensive account of the decisions made that summer in Philadelphia, this chapter highlights that all government institutions—the Senate, the House, the executive, and the judiciary—were equally novel creations; in some respects, the latter three were more innovative than the Senate. Also, it is important to recognize that the framers did not intend to enhance the power of a Senate minority; minority rights were not a central concern of those who drafted the Constitution, and equal representation in the Senate was not about minority rights. The framers’ decision to base Senate representation on state equality produced a compromised Senate with a conflicted mandate, however. Overall, the Senate was not the hub of Madison’s balancing act to “enable the government to control the governed; and in the next place oblige it to control itself.” It was the entire governmental system—composed of Congress, the executive, the court, and sovereign states—not the Senate per se that would prevent majority tyranny and allow for the safe and effective use of governmental power.
Consensus: A Republican Senate
Though they confronted a daunting task, the delegates who convened in Philadelphia in 1787 were not without some advantages. The formidable talents of many of the participants notwithstanding, many of the sharpest minds opposed to the project of fortifying the national government decided to boycott the meeting, partly in the expectation that it would collapse under its own weight. The convention was, as a result, populated by many of the most nationally minded among the political elite. This not only greatly enhanced the prospects for something more than a mere reformation of the Articles of Confederation, it broadened the level of agreement on several features of a good government. The delegates shared similar conclusions based on their republican beliefs and experiences in their state governments. One such conclusion was a widely shared belief in the desirability of a two-chamber legislature, with an upper house balanced against the more popular lower house.
At several points during the debates, the delegates expressed confidence about the self-evident appeal of bicameralism. Virginia’s George Mason argued that “the mind of the people of America, as elsewhere, was unsettled as to some points; but settled as to others.” He was sure it was well settled on two points, “1. in an attachment to republican government. 2. in an attachment to more than one branch in the Legislature.”4 Even the democratic soul of the convention from unicameral Pennsylvania, James Wilson, vigorously defended the need for a two-chamber legislature: “Is there no danger of a Legislative despotism? Theory and practice both proclaim it. If the Legislative authority be not restrained, there can be neither liberty nor stability; and it can only be restrained by dividing it within itself, into distinct and independent branches. In a single house there is no check, but the inadequate one, of the virtue and good sense of those who compose it.”5
The extent of this accord on bicameralism is evident from the opening actions of the convention. Madison and the Virginia delegation came to Philadelphia prepared with a detailed argument about the defects of the Articles of Confederation and, more important, a set of resolutions outlining the structure and powers of a new national government. Edmund Randolph “opened the main business” of the convention on May 29 by presenting both, and the fifteen Virginia resolutions became the basis of debate and deliberation on May 30. The first of these resolutions was a general call to action to correct and enlarge the Articles of Confederation. The second and third resolutions were about the structure of the new legislature. The second stated that representation in the national legislature “ought to be proportioned to the Quotas of contribution, or to the number of free inhabitants.” And the third resolved that “the National Legislature ought to consist of two branches.”6 In other words, the core of the entire proposal was a bicameral legislature based on some form of representation proportional to population.
The resolution for a two-house legislature was the opening item on May 31, and it engendered no debate or opposition. As Madison recorded in his notes, “The Third (3d) Resolution ‘that the national Legislature ought to consist of two branches’ was agreed to without debate or dissent, except that of Pennsylvania, given probably from complaisance to Doc. Franklin who was understood to be partial to a single House of legislation.”7 But Pennsylvania’s dissent on this issue disappeared thereafter. This effectively unanimous vote on bicameralism took place even after it was clear that the Virginia Plan would end the equal representation of states, which had been a foundational feature of the Confederation.
The proposal for proportional representation, discussed the day before, was more controversial and was ultimately postponed. Nevertheless, the delegates came close to agreeing that “the equality of suffrage established by the articles of Confederation ought not to prevail in the national Legislature, and that an equitable ratio of representation ought to be substituted.” However, when a delegate from Delaware reminded the convention of his state’s explicit instructions to protect state equality, even if it meant abandoning the proceedings, the convention agreed to a postponement.8 Yet Delaware did not object on the first vote on bicameralism the next day. Small states, it would seem, at first saw the need for bicameralism as more important than the potential threat to state representation.
The accord on bicameralism was based in no small part on the delegates’ widely shared assumptions about the distinct and special purposes of the upper house. As part of earlier work on the creation of the Senate, I found and categorized every remark or speech that made reference to a desired or ideal trait or purpose of a properly designed Senate.9 From the beginning of the proceedings to July 16 and the vote on the so-called Great Compromise, the decision to have the upper house composed of two senators from each state, nineteen delegates, most of the active participants during the period, offered at least some opinion about what they felt were the important characteristics of a Senate. Some only hinted at their preferences; others expounded repeatedly and at some length. A total of forty remarks or speeches contain one or more preferred traits or purposes of a Senate, producing a total of sixty-five individual invocations of one or another characteristic.10
Though the delegates voiced a number of opinions, the list is a harmoniously interrelated one and the level of agreement is evident. Aside from seven mentions of the importance of a second chamber as a check on legislative power, five characteristics dominate: small size (fourteen mentions), select appointment (eight), independence (eight), wisdom (eight), and stability (eight). These five account for 71 percent of the references to an ideal upper house. Moreover, they elicited no dissent beyond John Dickinson’s one claim that small size was not important. At fourteen references, small size is the most frequently invoked trait, and its central importance became apparent as the debate over legislative apportionment unfolded and dominated the convention. The remarks about size manifested concern for producing the quality of debate in which only a small group can engage. Madison would later provide this trenchant summary of the logic of numbers: “In all very numerous assemblies, of whatever characters composed, passion never fails to wrest the scepter from reason. Had every Athenian citizen been a Socrates, every Athenian assembly would still have been a mob.”11 Almost every delegate agreed with this judgment; they deemed small size the principal characteristic that would foster quality deliberation.
Small size would be complemented by a select appointment. As noted, these remarks were made prior to the final decisions about the composition of the Senate, but even so, it was clear that the upper chamber would not be directly elected. The Virginia Plan had the upper house selected by the lower chamber from nominations made by states. Complicated as it was by the later compromise in favor of having two senators per state, the eventual decision to have the Senate selected by state legislatures reflected the delegates’ consensus preference for an indirect and refined selection process. And regardless of the final form it took, the goal was to produce wiser, more experienced, and independent senators. It should be noted that the refined process of selection was as close as the delegates got to a proxy for wealth and property. A few delegates linked the Senate to representation and protection of property, but their views constituted a minority, even amid the secret deliberations. Even so, such remarks were usually not simply about representing the interests of property; instead, possession of considerable property would imbue officeholders with the wisdom and independence to form an effective Senate. Property, independence, education, and breadth of experience, in the belief of most of the delegates, came together as a package in society.
In turn, independence and stability would be enhanced by additional factors beyond the mode of selection, in particular the length of term. The Virginia Plan did not specify a length of term for either branch of the legislature. There was a blank to be filled in for the number of years for the first branch. The members of the other branch were “to hold their offices for a term sufficient to ensure their independency.”12 This implied only that the term of the second branch should be longer. In fact, this distinction between the two legislative chambers was confirmed and reinforced with every discussion and vote on the issue of terms. The initial votes gave the House a term of three years and the Senate a term of seven years. At other points, two delegates, Hamilton and Read, even proposed life terms for senators. While there was almost no support for life terms, there was no doubt that a significant difference between the two chambers would prevail. A few expressed concerns about the longest terms proposed, including nine, seven, and six years. Some resistance to much longer terms relative to the House came from small-state delegates who were holding on to the idea—which thus far had not been winning the day—that the upper house would represent states primarily. And some stemmed from anti-aristocratic if not democratic beliefs from delegates whose states had annual elections or who feared that senators, especially those from distant states, might get too independent and forget those who sent them. Nevertheless, the final decision on two years for the House and six-year staggered terms for the Senate evinced the delegates’ commitment to the kind of independence and stability that a longer term would produce.
An important note on staggered terms: they were designed and included to balance the dangers of the longer term by allowing partial renewal and change in the Senate every two years. Staggered terms, as we shall see in chapter 4, were not part of the Senate’s conservative design. They were not intended to make the Senate into a continuing body to prevent change. The convention never discussed such a concept.
Finally, the convention considered two other electoral qualifications relevant to such things as independence, wisdom, and stability. These were age and length of citizenship. Should there be a minimum age and period of citizenship to be eligible to serve in one or the other chamber of the national legislature? As with many such details, the Virginia Plan left these matters unspecified, leaving a blank space for the minimum age and not even mentioning citizenship. The age blank was filled in early on, well before the apportionment compromise was reached, when the consensus on the characteristics of an ideal Senate still drove the process. The convention decided initially to not have any age qualification for the first branch but specified thirty years for the Senate. The unspoken logic was clear: let the voters decide for the first branch, but a minimum age is prudent for the Senate. The convention amended this ten days later with a decision in favor of twenty-five years as a minimum age for members of the House.13
Decided much later, length of citizenship was more controversial than age, but not so much as a distinction between the legislative chambers. Citizenship qualifications did not appear until early August, with the draft presented by the Committee on Detail. Length of citizenship for both chambers then went through a bidding war between those delegates who sought the highest possible number of years and those who thought minimal requirements would do. Aside from general support for more stringent Senate requirements, the greater role of the Senate in foreign affairs via the power of treaties suggested the need for longer citizenship. In the end, with the decisions in favor of seven years for the House and nine for the Senate, rather stringent citizenship requirements adhered to both chambers.14
The small size, length of term, process of selection, and even age and citizenship would mitigate the passions and precipitation to which more numerous and popular assemblies are prone. A small number of senators, who would be, by and large, better men because they had achieved their positions through some refining mode of selection, and who were independent of any other branch of government, could best achieve the purposes or goals of a Senate. All these characteristics promoted superior deliberation, characterized by dispassion, wisdom, and a value for stability. This stability and wisdom would be primarily a check on the democratic excesses of the lower chamber (the comments on stability often implied a check on democracy). A properly constructed Senate would provide some institutional memory, a knowledge of and experience with various proposals and policies, and, as a result, would reduce the likelihood of constantly fluctuating laws and policies. The evidence from the Constitutional Convention and other founding documents such as The Federalist, particularly numbers 62 and 63, affirm that, rather than rules of procedure the Senate might later develop, it was the constitutionally mandated attributes of the Senate—the mode of selection, the smaller size, the length of term, and so forth—that were to structure and foster its quality as a deliberative body.
Conflict: Equal Representation and the Compromised Senate
Despite the delegates’ agreement on bicameralism and the best ways to ensure a high quality of deliberation in the proposed upper house of the new national legislature, the structure of legislative representation emerged as the most vexing issue of the Constitutional Convention. This became the main point of contention because the basic nature of the republic would be determined by the form or forms of representation in the national legislature. Discussions came close to collapsing in the face of this obstruction; once it was overcome, though other difficulties remained, the convention appeared certain to produce a final document. As James Madison lamented during the debates, “The great difficulty lies in the affair of Representation; and if this could be adjusted, all others would be surmountable.”15 At the center of this struggle was the composition of the Senate. Though this is common knowledge, the richness and complexity of how the Senate affected and was affected by the struggle over representation are lost or obscured in many accounts of the process.
While it would be stretching the truth to argue the representational compromise came into existence because of the Senate or, more precisely, because of the near consensus on the need for an upper house, the widespread agreement on the need for a Senate in a stronger national system helped to shape and produce the compromise, rather than the other way around. A republican Senate was crucial to the new system, regardless of any political compromises. The problem, however, was that the Senate would become essential as well to the resolution of the question of federalism, that is, the extent to which the new constitution would embrace or eliminate the role of states as sovereign actors. As a result, despite the agreement on a Senate in principle, and for republican purposes, the actual Senate that would emerge from the convention was the complicated and compromised product of the most important and bitter struggle that summer.
To portray the fight over representation as simply between large-state delegations for proportional representation and small-state delegations demanding state equality would be to strip the proceedings of the most interesting complexities in the mixture of interests and principles at work. In addition to these conflicting interests, two main relationships that became increasingly apparent over the course of the summer structured the conflict. The consensus on bicameralism and the general agreement that the government needed to be more national in character all but guaranteed that one of the legislative chambers would be proportional to population and directly elected. Yet nearly any form of proportional representation for the Senate would make it too large, and thereby vitiate its most important characteristics as a deliberative institution. This pushed many large-state delegates in the direction of equal representation in the proposed Senate, not just to placate small states but to keep that body small enough to fulfill its primary function.
As to the first relationship, it would take days of debate before the intractable relationship between the number of legislative chambers and the ratio of representation became apparent to the delegates as the crux of the proceedings. After the connection was made, bicameralism faced its only threat in the form of the so-called New Jersey Plan. Two weeks of debate that kept circling inevitably back to the question of representation culminated on June 15 in a 6–5 vote in favor of proportional representation in the Senate. Only a few days before this the delegates had agreed to proportional representation in the House by a 9–2 vote, demonstrating the broad support for both bicameralism and a different basis for representation in the lower house. But the 6–5 vote for the same in the Senate prompted the small states to rally and offer an alternative. William Patterson presented the New Jersey Plan on June 15. Up to this point, the opposition had no clear alternative. In part this was because there was broad agreement on the overall thrust of the Virginia Plan—the only point of significant dispute had been the issue of equal representation of states. The need for a more effective response to this bone of contention necessitated a more complete alternative. And until this point there had been no significant opposition to bicameralism. The New Jersey Plan, which called for a moderately strengthened Congress of the Confederation, retained the unicameral, equally apportioned legislature of the articles. This aspect of the design did not imply a philosophical preference for a single-house legislature. Patterson offered the plan in reaction to the convention’s initial and narrow support of proportional representation in both legislative chambers. The small states had voted, it should be remembered, for bicameralism at the beginning. After some debate this dramatically different alternative was rejected fairly easily on June 19 by a 7-3-1 vote.16 But it indicated that the small-state coalition, which had accepted and even embraced proportional representation in the House, was likely to hold out for equal representation in the Senate.17
As the deadlock over Senate representation developed, many delegates began to grow concerned about the size of the Senate. Simple mathematics made it clear that even a modified form of proportional representation in the Senate would produce a large body, and one that would grow rapidly with the addition of new states. This had not been a problem in the Virginia resolutions, which proposed that the membership of the second branch of the national legislature “ought to be elected by those of the first, out of a proper number of persons nominated by the individual Legislatures.” State legislatures would nominate candidates for the Senate and the House would select among them. Consequently, the second branch or Senate could be as small as desired because there was no geographic distribution. The House would select the best candidates regardless of residence. But many delegates rejected this idea for a refined selection of the Senate because this would make the Senate a creature of the House. Refined selection was still important, but not if it came at the expense of institutional independence.
After rejecting selection by the House, the convention agreed, in a 10–0 vote, to have state legislatures choose senators.18 This constituted a refined selection, but it also required some form of apportionment of senators among the states. Hence the second intractable problem: the adverse impact on the size of the Senate that would be produced by any form of proportionality. This surfaced several times during the debate, confirming delegates’ strong preference for a small Senate, significantly smaller than strict proportional representation would produce. Even very early on, Massachusetts delegate Rufus King drew the logical conclusion: the proposal to have the state legislatures select the second branch would make the Senate, from the outset, too numerous, unless “the idea of proportion among the States was to be disregarded.”19 Much later, as the convention edged toward the Great Compromise, North Carolina’s William Davie lamented the impact any form of proportional representation would have on Senate size: “It was impossible that so numerous a body could possess the activity and other qualities required in it.”20 King and Davie were not alone in such sentiments. Thrown back to the states for the primary role in selection and constricted by the nearly universal desire for a small Senate, the delegates were on the road toward the Great Compromise whether they realized it or not. As the problem of size and the role of state selection eroded support for proportional representation in the Senate, equal representation emerged as the only viable alternative. The politics of the Great Compromise were a complicated mixture of interest and principle. While perhaps not decisive, the widely shared sentiments of an ideal Senate—independent to be an effective check, small to promote deliberation, and selected (rather than elected) to enhance both deliberation and the check—pushed the convention toward the compromise. Efforts to elaborate purely interest-based explanations mistakenly overlook this evidence from the record. Among the factors that shaped the compromise and swayed the votes of a few individual delegates, these shared principles manifested their importance throughout the deliberations, even on the brink of compromise, and after.21
Following the Great Compromise, two other compositional issues with potentially profound consequences were solved rather easily, but with the effects of the apportionment compromise woven once again into the debate and decisions. First, states would be represented equally, but how many senators per state? Second, state equality implied that state representation in the Senate would be corporate, as in the Confederation Congress. If there were more than one senator per state, would they vote by state or per capita as individual senators?
These two profoundly important details were settled simultaneously on July 23 with relatively little debate and near consensus on the final votes. After considerable attention to unrelated articles in the Virginia resolutions, including the mode of ratification, Gouverneur Morris and Rufus King moved “that the representation in the second branch consist of ____ members from each state who shall vote per capita.” This was part of an overall counterattack to limit the damage done to the Senate by the equal representation compromise. By specifying that senators would vote as individuals, these large-state delegates were trying to ensure that the Senate would not be a replication of the Confederation Congress and to minimize the chances that senators would act simply as mouthpieces for state legislatures. This approach to voting also was closer to the ideal of a Senate composed of fewer and wiser men for better deliberation. Clearly, no one considered one per state a safe number, not only for purposes of deliberation but also for the purely pragmatic reasons of absences, illnesses, and deaths. Nathaniel Gorham, a supporter of per capita voting, called for two per state because a small number was best for “deciding on peace & war &c, which he expected would be vested in the 2d. Branch. The number of states will also increase.” Reflecting the delegates’ consistent preference for a small upper house, Mason agreed that more than two, with the addition of new states, would produce too large a Senate. Two per state was then agreed to unanimously. This was followed immediately by Luther Martin’s objection to per capita voting “as departing from the idea of the States being represented in the 2d. branch.” But the delegates quickly voted on the whole motion, including two per state voting per capita, and gave their approval by a 9–1 vote, with only Maryland in opposition.22 Even most small-state delegates, though committed to state equality, wanted to retain some of the Senate’s original purpose. Equality of votes might have been a political necessity, but not at the complete sacrifice of senatorial independence. Perhaps for some small-state delegates, this was their own compromise of interests and principles in the making of the Senate.23
From start to finish, the issue of Senate representation pervaded the deliberations of the delegates and affected some of their most important decisions. The nature and purpose of that body were vital issues to the end of the deliberations.24 For our purposes, the point is to understand how the ideal republican Senate and the compromised Senate of state equality came to be combined, whether harmoniously or not, in the same institution.
Minority Power or Rights
Senators, especially those serving after the nineteenth century, have with great frequency and conviction associated their institution with a special role: the protection of minority rights, or “the rights of the minority,” as senators often put it. To cite just two examples, former senator and Republican Majority Leader Howard Baker stated in 1993 that “the Senate is the only body in the federal government where . . . minority rights are fully and specifically protected. It was designed for that purpose by America’s Founders.”25 A couple of decades later, Democratic Majority Leader Harry Reid echoed Baker’s words: “There’s a reason for the filibuster. I understand it. It’s to protect the rights of the minority. The Senate was set up to protect the rights of the minority.”26 Such sweeping claims are devoid of historical evidence and made without pausing to define the key terms. What minorities and what rights? Minority interests are sometimes linked to states and federalism, but far more frequently minority rights are expressed in broad and unspecified terms. That is, the “minority” could be any individual or collective, whether defined by party or region or by some other interest. Likewise, “right” or “rights” are typically left hanging, as though self-evident. A review of thousands of pages of Senate debates from the nineteenth and twentieth centuries shows that the specific thing most commonly associated with minority rights is the right to be heard. This right to “voice” has often been characterized as “freedom of debate.” The actual history of the filibuster and supermajority cloture suggests that there is a bit more to it than that. As we shall see in later chapters, the rules of the Senate could encourage debate. They don’t. Despite what senators say, the rules have never been about allowing for more debate. They have been about power. The minority does not want to make its case; it wants to win. The hue and cry in the Senate about minority rights over the years has not been so much about voice as it has been about victory, that is, the power to obstruct, extract concessions, or veto outright what the majority would like to accomplish.
So what role did conceptions of “minority rights” play in the creation of the Constitution? Depending on which minorities and which rights were considered, what protections, if any, were afforded them? And was the Senate tasked with defending such rights in general? At the Constitutional Convention, minority rights as such were not a central concern. Convention delegates, in fact, worried more about forms of minority power or control. Even as the Senate was many different things to the authors of the Constitution, one thing it was not was a citadel for minority rights. In fact, strictly speaking, the framers never connected the Senate to any notion of minority power, let alone rights. In its primary and at times contradictory roles as both the federal institution constituted by state equality and comprising an older and wiser group of statesmen, the Senate represents a potential different majority (of states or simply senators). Or it represents a different quality of deliberation, owing to chamber size, the longer term, indirect election, and age-related experience. The records of the Constitutional Convention contain no discussion of the rights of the minority in general or any connection to the Senate specifically. In fact, they evince far more care for majority rule than they do concerns about minority rights. Insofar as equal representation in the Senate entailed an advantage for smaller-population states (and some delegates argued that population would not prove to be the enduring division among the states), then the Senate protected this very specific form of minority rights. But the debates made no mention of a more general notion of minority rights or the Senate’s role as their safeguard.
To begin, it bears repeating that the Senate of the Virginia Plan—Madison’s deliberative Senate, intended to balance the larger democratic House—would have been apportioned by population. That is, it would have reflected the majority of the population, not some minority, however construed. Madison and a majority of the delegates, for most of the convention, did not even want the Senate to represent states as states, let alone valorize the Senate as the guardian of states’ rights. When proportional representation for the Senate proved both unacceptable to smaller states and incompatible with a small Senate, equal representation of states became the barely acceptable alternative. Notice, however, that equal representation was really about representing or balancing two potentially different majorities: the majority of sovereign states in the Senate versus the majority of the national population in the House. The Great Compromise was so fraught that it is a fraud to extract a principle of minority rights from the highly contentious votes that nearly ended the proceedings. The basis of Senate representation that emerged from the convention was state equality: states, not any other potential minorities.
A search for the terms “minority” and “majority” in the entirety of the records of the convention reveals the following.27 The comments surrounding any use of the term “minority” are overwhelmingly against the ability of minorities to decide or prevent a decision by the majority. The only specific endorsements of minority decision-making power, not rights as such, concern states and their role in the national government. And even these are surprisingly few. As the political scientist Patrick Coby points out, “At the Convention there are three minority interests that receive the delegates’ sustained attention and solicitude. They are small states, all states, and southern states.”28 The fourth minority interest is property, but it draws less attention at the convention and does not end up with any direct protection in the Constitution.29
States, in this case, were not portrayed as exemplars of some universal notion that minorities of any variety should be protected. They were not stand-ins for some broader category. They were considered as specific political entities: discrete and preexisting political units that were given corporate and equal representation in the Senate only because a determined minority of the state delegations at the Convention were prepared to end the entire proceedings if this demand was not met.
Of the forty-three substantive comments surrounding any invocation of the term “minority,” only six refer exclusively to the Senate, as opposed to the government or the political process more generally. And five of those six are decidedly against the exercise of minority power by or within the Senate. The one exception came from Madison, who, while defending a long Senate term, noted that the Senate would or should represent the interests of larger property holders, who would be in the minority. It should be noted that this speech was more about the Senate as a source of wisdom and stability and it preceded the Great Compromise, after which Madison sought to take some power from the upper house structured by state equality. Property was, for the delegates who spoke on the subject, a proxy for independence and wisdom rather than the possession of a minority that needed protection.
Even the more general references to majority tyranny do not imply anything about the Senate. During the debates, Madison recorded himself as twice making versions of what would become his famous argument in Federalist no. 10. In familiar language, he warned that “In all cases where a majority are united by a common interest or passion, the rights of the minority are in danger.”30 But, as in no. 10, Madison found the cure for this danger in the size and diversity of the republic and the overall form of its institutions. His fellow Virginian, George Mason, also twice expressed concern about the tendency of the majority to oppress the minority, but for Mason this was about states and the power of the northern states, not some general principle. And he clearly ascribed the ability to defeat this defect of republics to the design of the whole system, not the virtues of the Senate. In fact, the Senate came in for some scorn from Mason because he felt it should not be given too much power and be able to entrench itself.31 Moreover, he also contradicted himself by worrying about a minority controlling the majority when, as he mistakenly forecast, the national population would shift in favor of the South and West.32
Instead, concerns about minority power, in general or in the Senate specifically, dominated. Delegates such as Wilson, Madison, King, Mason, Franklin, Morris, and General Pinckney argued against various forms of minority power or minority vetoes in areas such as trade, treaties, appointments, the creation of new states, and so forth.33 A particular example was the debate over the threshold for quorum for both chambers of Congress, with delegates expressing concerns about the delays and mischief that could attend supermajority quorum requirements.34 They rejected motions that would have required a higher quorum in the Senate for treaties and general business.35 They also rejected a motion to allow a single member of the Senate “to enter his dissent” on any vote (that is, enter the reasons for his vote in the journal).36 Moreover, the records are replete with concerns about a potential Senate aristocracy, from small- and large-state delegates alike. This also contradicts any idea that the Senate should be a forum for minority rights. Even Roger Sherman, one of the most important advocates for state equality, expressed opposition to excessive minority power, at least when it came to overriding presidential vetoes. He favored two-thirds rather than three-fourths—which was still in the draft Constitution on September 12—for overriding presidential vetoes, because “the States would not like to see so small a minority and the President, prevailing over the general voice. In making laws regard should be had to the sense of the people.”37
This might seem surprising given that “majority tyranny” and the fear of it are often portrayed as all but paramount among the concerns of at least some of the founders, particularly Madison. There is, however, a substantial difference between a fear of majority tyranny and a concern for minority rights writ large. Majority tyranny is about the durable or consistent domination of a minority by a cohesive majority. And, as implied by Madison in Federalist no. 10, for example, a majority faction is tyrannical only if it is united by a “common impulse of passion, or of interest, adverse to the rights of other citizens, or to the permanent and aggregate interests of the community.”38 That is, a majority decision by a government is not inherently tyrannical because it did not somehow listen to or compromise with the minority. The minority, just like the majority, can be a faction adverse to others’ rights or to the aggregate interests of the community. The difference is that the danger of minority factions is mitigated by majority rule, “which enables the majority to defeat its sinister views by regular vote.”39 In short, the majority has the right to decide on any particular issue, but the system is designed to mitigate or prevent domination over time. That is what the extended republic of sovereign states and the multiple decision-making institutions in the national government are meant to do.
When it comes to their institution, however, senators tend to equate minority rights with the following: that on any issue the minority must have some degree of influence over the process or outcome. That logic is in none of the decisions made at the convention.40 The default position of the Constitution is majority rule. To highlight that default position, the framers made specific exceptions to require supermajorities for five extraordinary actions: the override of a presidential veto, expulsion from Congress, the approval of treaties, conviction on impeachment, and the proposal of amendments to the Constitution.
At the Constitutional Convention, delegates used the term “tyranny” in reference to a number of things, including an executive that was too powerful or one that was not powerful enough. Concerns about majority tyranny specifically were few and far between and were expressed most directly by Madison previewing his Federalist no. 10 argument. As numerous were direct references to minority tyranny. North Carolinian Hugh Williamson was sure “it must be tyranny, if the smaller states can tax the greater, in order to ease themselves.”41 Or James Wilson of Pennsylvania, who thought “It is a part of the definition . . . of tyranny, that the smaller number governs the greater.”42 And finally, Roger Sherman, the leader of the small-state coalition, thought that “to require more than a majority to decide a question was always embarrassing as had been experienced in cases requiring the votes of nine States in Congress.”43
The Exceptional President, Supreme Court, and House of Representatives
As we have seen, the story of the Great Compromise, stripped of its essential details, can leave the impression that a second legislative chamber was introduced to save the day and the convention. This makes the Senate look like the bold innovation. Christopher Dodd, in his 2010 farewell speech to the Senate, made this mistake when he attributed the Senate to the actions of two Connecticut delegates, Roger Sherman and Oliver Ellsworth, “who proposed the idea of a bicameral national legislature.”44 This is simply false, as we have seen. The Connecticut delegates can be credited with advocating for taking what was already agreed to, a bicameral legislature, and using it as the basis for giving each side, so to speak, what it wanted, with proportional representation in one chamber and equal representation in the other. The representational compromise and what followed was something borrowed, something blue, mostly old, and a bit new, but with the necessity of a second chamber as all but a given. And however complicated the Senate became, it was a bit less innovative and a lot less elegant than the exceptionalism argument implies. What about the other branches? In short, the executive, judiciary, and the House of Representatives were in many respects as or more innovative than the Senate, and certainly more elegant in the clarity of their design and purpose.
We can begin with the fact that the United States under the Articles of Confederation had no executive or president. The closest the articles came to such a thing was a “Committee of the States,” composed of one member from each state, which convened only when Congress was in recess. A “president” was selected by Congress to “preside” over the Committee of the States, nothing more. States had governors, but the postrevolutionary constitutions created mostly weak executives, who were often picked by the state legislature or very much circumscribed by legislative power. Pennsylvania’s constitution, for example, had a twelve-member “executive council” instead of a governor. Georgia’s one-house legislature selected the governor, who served a one-year term and was all but under the control of a twelve-member executive council picked from the legislature. Indeed, whether powerful or not, such executive councils were included in most state constitutions. Two states, New York and Massachusetts, had created more autonomous executives. New York had an independently elected governor, and the governor was part of a “council to revise,” which included supreme court judges, that reviewed the constitutionality of legislation. Only Massachusetts had an independently elected governor with exclusive veto power subject to a two-thirds override. Overall, the first state constitutions remained some distance from an embrace of a strong executive.45
Despite fears of monarchical power, the presidency created by the Constitution was both an independent and a potentially powerful executive. Hamilton’s Federalist no. 70 laid out the “ingredients which constitute energy in the Executive”: unity, duration, an adequate provision for its support, and competent powers. To this end, the president was selected independently from the legislature for a longer term than the House and endowed with a veto that could be overcome only by mustering two-thirds in both the Senate and House. The convention rejected a plural executive and did not create anything like an executive council. This was a strong and unitary executive, designed for effective administration of the national government. Its autonomy from control by the legislature created a true separation of powers, something that had heretofore been an object of theorizing more than actual practice.
An independent judiciary was necessary to complete the distinctly republican separation of powers. This was a major leap forward. The colonies had had courts, but primarily those that handled ordinary legal disputes. As the historian Peter Charles Hoffer and colleagues write, “In some of the colonies, notably Connecticut, the legislatures performed the role of high courts.”46 The revolutionary state constitutions had then made strides toward a formal separation of powers and the recognition of judicial authority and independence. Nevertheless, as with the executive, many of these initial state constitutions did not have a fully independent or powerful high court. Pennsylvania’s constitution created a supreme court, but it was essentially just the peak of the pyramid of ordinary law courts. Constitutional questions were handled by a “Council of Censors” that met every seven years. South Carolina’s constitution mentioned various types of ordinary lower courts but specified that the lieutenant governor and privy council would serve as the court of chancery or equity. And during these years only a handful of cases suggested or asserted the authority of state courts to rule on the constitutionality of state acts or laws. At the national level, the Articles of Confederation said nothing about courts of any kind.
In sharp contrast, Article III of the Constitution announces that “the judicial Power of the United States, shall be vested in one supreme Court” and other inferior courts Congress is empowered to create. By dividing the appointment power between the president and the Senate and by giving federal judges and justices indefinite tenure “during good Behavior,” the Constitution goes nearly as far as possible in creating an independent court. Judicial power, moreover, “shall extend to all Cases, in Law and Equity, arising under this Constitution, the Laws of the United States,” and more. And while the concept and practice of judicial review—the court’s implicit authority to rule on constitutional questions—would take some time to evolve, this institutional architecture and broad constitutional mandate was a new and substantial addition to the practice of republican government, especially at the level of the nation-state.
The arguments about the innovative if not revolutionary nature of the presidency and judiciary are straightforward and perhaps unsurprising. Neither had existed in anything like the forms proposed in the Constitution. Nevertheless, reminding ourselves of the striking nature of these institutions, which were anything but givens or obvious, casts grave doubt on claims such as those cited in the introduction about the compromised and conflicted Senate as the “one touch of authentic genius” or “the most remarkable of all the inventions of modern politics.” And remarkable as each institution was, the true genius lay in the whole rather than in its parts.
But what about the House of Representatives? Surely, it is the least innovative among the four entities composing the branches of government. The House would seem to be the prosaic given of any parliament: a more or less popularly elected body with some or all of the authority to write laws. From a contemporary perspective, the House can seem like an unremarkable necessity rather than an extraordinary invention. I think that this is a significant distortion. In the most crucial constitutional respect, the House of Representatives was the true institutional innovation. Its relative simplicity and the relative lack of discussion about its creation are deceptive because they obscure the sea change the House represented. Madison and his allies lost a great deal in the battle for the ideal republican, and specifically more national, Senate. But that was in large measure because they won the central victory with the House. This victory centered on the creation of a bicameral legislature with at least one chamber selected directly by the people, not by or through their state governments. The House represents the creation of the crucial constituent connection, and thus power, for the national government.
While there were colonial and state assemblies that resembled the House in many respects, a popular legislature had not been part of the national government under the Articles of Confederation. They had created a unicameral Congress based on equal representation of states as states. States could send no fewer than two and as many as seven delegates. The number did not matter because states voted as states: one vote per state regardless of size. As we have seen, the New Jersey Plan with a unicameral legislature based on state equality was put before the convention only as a challenge to the insistence by many large-state delegations on proportional representation in both chambers of the legislature. This small-state alternative basically replicated the Confederation Congress. And it failed. The convention struggled with how representation would be structured in the Senate, but from the beginning, a strong majority of delegates favored a bicameral system with direct representation of citizens in at least one chamber.
This innovation, the direct representation of citizens in the national legislature, was the linchpin of “inventing the people,” in the words of historian Edmund Morgan. To create a genuinely national government, Morgan explains, “Madison was inventing a sovereign American people to overcome the sovereign states,” and in this creation the “composition of the lower house would thus be decisive” because everything else would be built from that foundation.47 George Mason made this clear early in the proceedings in Philadelphia. “Under the existing Confederacy,” he noted, “Congs. represent the States not the people of the States: their acts operate on the States not on the individuals. The case will be changed in the new plan of Govt. The people will be represented; they ought therefore to choose the Representatives.”48 The people would directly select their representatives, who would then make laws that acted directly on and for them, rather than on or through the states.
The resulting House of Representatives was radically democratic in design and potential, especially for the national legislature of a large country. Its British counterpart, the House of Commons, was based on a medieval form of geographic representation—places, towns, estates. American colonial and state legislatures were typically similar, with most distributing state assembly seats to towns and counties.49 Breaking from representation based on place per se, the House was apportioned by the population of each state based on an “actual Enumeration,” what we now know as the census. One can argue that states were preexisting “places,” and that point is well taken. But the key fact is that the Constitutional Convention did its best to divorce representation of citizens in the House from state control. States were all but unavoidable as the intermediary geographic unit for the distribution of House seats. Only states would be capable of administering the elections over the length and breadth of what was already a vast nation, with huge territories such as those just slated for formation into new states by the Northwest Ordinance, a law passed by the Congress of the Confederation in July 1787. The convention saved itself a world of trouble by giving the task of election management over to states. The jealousy of the states might not have tolerated a more nationalized system, but that was hardly an issue because the practical problems of a central government administering nationwide elections made such a system unimaginable. Nevertheless, the “people” within each state would directly pick their representatives to the national government. Moreover, the franchise would be as democratic as what each state allowed for in its “most numerous” branch of its legislature.
The new government was premised on a democratic assembly endowed with the primary powers of government. On this kind of geographic scale, this was a historic innovation. In 1787, the British House of Commons was marginally democratic and represented mostly the 50,000 square miles of England, with handfuls of seats in Scotland and Wales. Pennsylvania and New Jersey, two of the states still defined by their eighteenth-century boundaries, by themselves comprised about 55,000 square miles. And no other branch of the government was intended to be democratically selected. Even in the Virginia Plan, as we have seen, the upper chamber was to be proportioned by population, but selected by the lower house from nominations by the states. Almost none of the delegates imagined an unmediated democratic form of selection for the president, to say nothing of the judiciary.50 In fact, the antidemocratic mechanisms for selection of the other branches reinforce the point that the direct election of the House is certainly a more innovative feature of the new Constitution than the indirect selection of the Senate and the state equality attached to it.
It’s the System, Not the Senate
The Constitutional Convention created a multifaceted but also compromised and conflicted Senate. But no matter what the founders’ exact intentions were for the upper house of Congress, it was always part of a system. That system was composed of two layers. First, there was the extended republic of many states, which retained important aspects of sovereignty or self-government and which were characterized by a diversity of interests resulting from differences in geography, economy, and culture. Second, there was the national government itself, composed of multiple decision-making institutions of separated and shared powers. As we have seen, the Constitution created several exceptional institutions—the House, the executive, and the judiciary, along with the multifaceted Senate. Nevertheless, however important and innovative each part of the system was individually, they were designed, whether with complete success or not, to work together in a republican architecture of independent institutions. Each was to play a vital role in the project of “contriving the interior structure of the government, as that its several constituent parts may, by their mutual relations, be the means of keeping each other in their proper places.”51
Thus it was the Constitution as a whole that protected against majority tyranny, not any one part, such as the Senate. The Senate was part of this, but it was not the designated guardian of minority rights. Madison put it this way:
In all Govts. there is a power which is capable of oppressive exercise. In Monarchies and Aristocracies oppression proceeds from a want of sympathy & responsibility in the Govt. towards the people. In popular Governments the danger lies in an undue sympathy among individuals composing a majority, and a want of responsibility in the majority to the minority. The characteristic excellence of the political System of the U. S. arises from a distribution and organization of its powers, which at the same time that they secure the dependence of the Govt. on the will of the nation, provides better guards than are found in any other popular Govt. against interested combinations of a Majority against the rights of a Minority.52
The Senate was, and is, part of a system that in its totality was the true ingenuity. This system was innovative in the Constitution’s balancing act of the powers attached to each institution, in how each was organized, and in the relationship of each institution to a constituency that would support or resist its power.53
As we have seen, the most common characteristics ascribed to an ideal Senate by the framers were a limited number of members, a select form of appointment, and independence. A small number of carefully selected senators appointed for long terms would foster wisdom, stability, and effective deliberation. The quality of deliberation was likely to be better in wisdom and temperament than that of the House because of the elements of the Senate’s composition, namely, the much smaller membership, appointment by state legislatures, the longer term, and even the higher age and citizenship requirements of members. In all this, there was no discussion of the quantity of debate or even the right to debate. And there certainly was no discussion about rules of procedure at all, let alone as part of any bicameral differences with regard to deliberation.
In short, the exceptionalist mythology that senators and others regularly attach to that institution misses the truth on two fronts. First, it all but ignores the Constitution’s original intra-institutional components of the Senate’s deliberative difference and substitutes instead the Senate’s rules of procedure, which have been written and altered by senators from the Senate’s first meeting onward. Second, it reifies the Senate as the constitutional barrier against majority tyranny, when it is the inter-institutional system, the system as a whole, and not the Senate in particular, that protects against the abuse or unchecked aggregation of power.
The Senate was a product of multiple forces and motivations, and the product reflects this mixture. In the end, I would argue that knowing the exact balance of motivations of the framers is less important than understanding the consequences of the sometimes confused deliberations, sharp disagreements, and grudging compromises. Whatever the exact motivations behind all the decisions that produced the Senate, it was, as a result, a compromised institution, intended to embody two potentially incompatible purposes. Rufus King reminded the delegates of this dilemma on the last day of debate before the compromise vote: “According to the idea of securing the State Govts. there ought to be three distinct legislative branches. The 2d. was admitted to be necessary, and was actually meant to check the lst. branch, to give more wisdom, system, & stability to the Govt. and ought clearly as it was to operate on the people to be proportioned to them. For the third purpose of securing the States, there ought then to be a 3d. branch, representing the States as such and guarding by equal votes their rights and dignities.”54 Did the compromised Senate represent the states, or was it still the national institution and steward of governmental policy? Could the two functions be combined? Confusion and disagreement on the former question reigned throughout the rest of the convention. The latter question was never addressed and would have to be answered if and when the new constitution was put into practice.
On September 15, the busy final day of substantive debate on the proposed constitution, the last successful motion was about equality in the Senate.55 A few days before, the convention had added without debate one restriction on the procedure to amend the Constitution: “no amendment which may be made prior to the year 1808 shall in any manner affect” the provisions protecting the importation of slaves and the three-fifths formula on taxation. Roger Sherman noted the addition of this restriction and thought it “should be extended” such that “no State shall without its consent be affected in its internal police, or deprived of its equal suffrage in the Senate.” After dropping the internal police provision, the convention approved the second half of Sherman’s restriction without debate or dissent, “being dictated by the circulating murmurs of the small States.”56 The implication of this wording was that instead of undergoing the general and already arduous process for amending the Constitution, any proposed change to equal representation in the Senate would require the approval of every state. Thus was entered into Article V the only indefinite and seemingly insurmountable limitation on the power to amend the Constitution. The Great Compromise, the creation and effects of which had pervaded the deliberations of the Constitutional Convention from start to finish, was not only great, it was apparently permanent as well. And it is to this fundamental feature of the Senate and the Constitution that we now turn.