Contrary to the numbers games of today's majoritarians, America's federal republic reflects not a trace of national, numerical democracy.
Niccolò Machiavelli was one of the principal architects of American constitutionalism. A heterodox thinker in his day, Machiavelli observed in his Discourses on Livy (1531): “Those who blame the quarrels of the Senate and the people of Rome condemn that which was the very origin of liberty.” In other words, the institutionalized conflict between the optimates and populares was essential to preserving the Roman Republic. Machiavelli was the first significant figure in the Western tradition to acknowledge the importance of conflict in sustaining a body politic. And his understanding of the relationship between conflict and political order would eventually influence the creation of the American Senate in 1787.
Yet most people today are unfamiliar with Machiavelli’s influence on the Senate. For example, consider the recurring debate over the institution’s undemocratic nature. Both critics and fans of the Senate assume it was created to serve the interests of the states qua states. Jonah Goldberg summed up this point recently asserting, “the Senate was created to represent the interests of the sovereign states.” The only issue of contention in the debate is whether one thinks this is a good or bad thing.
Yet the conventional wisdom is mistaken, and impedes our ability to assess of how the Senate at present is, or is not, working.
The Options the Founders Considered
The U.S. Senate exists for one overriding reason: to check the popularly elected U.S. House of Representatives. Throughout the summer of 1787, James Madison and his fellow delegates to the Federal Convention highlight, again and again, the Machiavellian observation that institutionalized conflict was essential to the preservation of the republic. Trying to inject an updated understanding of Machiavelli’s dictum into the heart of the new federal government, they created a Senate whose institutional features—size, membership-selection process, nature of representation, length of term of office, compensation—are properly understood only in relation to the body’s House-checking role.
If the Senate were going to check the House, its members could not be drawn from the same source. That meant senators could not be popularly elected. If they were, they would end up being subject to the same interests and passions as their House colleagues, gravely impairing their ability to be a corrective. According to Madison, “In the states where the Senates were chosen in the same manner as the other branches . . . the institution was found to be no check whatever against the instabilities of the other branches.”
Much less would the Senate be able to check the House if its members were dependent on the House for their seats, which was initially a feature of the Virginia Plan. Almost all of the delegates to the Convention opposed having House members appoint senators; the latter might feel beholden to the former. Connecticut’s Roger Sherman reflected their general sentiment, stating that he “was of opinion that if the Senate was to be appointed by the first branch and out of that Body that it would make them too dependent, and thereby destroy the end for which the Senate ought to be appointed.”
Once the delegates ruled out popular election and House appointment as being incompatible with the Senate’s checking role, having each state legislature choose that state’s U.S. senators was the only option they had left. Admittedly, both Madison and fellow Virginian Edmund Randolph originally opposed state selection, favoring instead the selection of senators by the House. Their reason was that state lawmakers were even closer to the people than House members would be. Consequently, allowing state lawmakers to select U.S. senators would readily carry the popular interests and passions into the Senate whereas selection by the U.S. representatives might do so to a lesser degree. The two men opposed popular election of senators as a recipe for bringing this ill-effect into the Senate quickest of all. Each option had its downside for Madison and Randolph. That they, in the end, set aside their qualms about state selection shows how paramount was the Senate’s checking role in their minds.
The delegates’ desire to have the Senate check the House also dictated that it have a smaller membership than the House. They believed that there was something intrinsic to large legislative assemblies, separate and apart from the way their members were selected, that led them to conduct their business in similar ways. In Federalist 62, Madison observed,
The necessity of a Senate is not less indicated by the propensity of all single and numerous assemblies to yield to the impulse of sudden and violent passions, and to be seduced by factious leaders into intemperate and pernicious resolutions . . . a body which is to correct this infirmity ought itself to be free from it, and consequently ought to be less numerous.
Similarly, Pierce Butler suggested that the Senate should be small enough “as to be exempt from the passionate proceedings to which numerous assemblies are liable.” Butler’s fellow South Carolinian, Charles Pinckney, was even more specific, arguing that “If the small states should be allowed one senator only, the number will be too great, there will be 80 at least.”
Why They Thought Smaller Was Better
North Carolina’s William Richardson Davie summed up the problem such views created for those delegates who supported the allocation of Senate seats on a proportional basis:
Allowing the legislators to choose the Senate, and establishing a proportional representation in it, seemed to be impracticable. There will according to this rule be ninety members in the outset, and the number will increase as new states are added. It was impossible that so numerous a body could possess the activity and other qualities required in it.
As Davie surmised, the widespread opposition to a large Senate among the delegates precluded a Senate in which seats were allocated proportionally by population. Assuming each state was guaranteed at least one member, the total number of senators proportionally allocated based on population quickly surpassed the size with which the delegates were comfortable.
Madison hinted on June 30, 1787 that he could live with equal representation in the Senate. However, he argued that the delegates should take steps to make senators independent of the states if the convention adopted equal representation. In other words, Madison did not want senators to feel pressure to represent the states inside the federal government. The indirect mechanism by which popular interests and passions would enter the Senate could then lead senators to refrain from checking the popularly elected House. Madison predicted that failing to make the Senate independent of the states would turn the institution into “another edition of Congress” (that is, the Confederation Congress whose members were selected by, and dependent on, the state legislature that selected them).
Independence, the Highest Senatorial Value
Significantly, the small-state delegates agreed with Madison. The Convention’s adoption of equal representation is what flies in the face of the assumptions of Goldberg et al today: Equal representation was intended to make senators independent of the states that selected them. The delegates approved longer terms for senators than what they had already approved and compensation out of the federal treasury. They also opposed making senators subject to instruction and recall by the state legislatures that selected them.
What is important is not that the Senate served as an institutional representative of the several states in the federal government as a result of state selection and equal representation. Instead, it’s that a Senate so conceived would interact with, and counteract any unwise decisions by, the House. This, in turn, would help to control legislative majorities and frustrate presidential action. While the delegates did indeed disagree about how best to ensure that the Senate could check the House, all had this checking function in mind. It was crucial to establishing a new form of republican government that would be stable but flexible.
Taken separately, the House and Senate are imperfect institutions. But this is precisely the lesson that Madison and his fellow delegates took from Machiavelli. Taken together, the House and Senate are necessary ingredients for institutionalized conflict. They make possible a properly constrained Congress rooted in the consent of the governed.