The United States’ Senate has its origin in the Philadelphia Convention of 1787. Initially intended as a meeting to revise and strengthen the Articles of Confederation, the Convention instead established the Constitution which serves as the basis for today’s American government – including the characteristics and composition of the Senate.

Would there be a Senate – and if so, how would it represent the states? As he waited for the Convention to formally begin, James Madison, Delegate from Virginia, sketched out a proposal for organizing the new American government. Known now as the “Virginia Plan,” the draft proposed – among other elements – a bicameral legislature featuring a popularly elected “lower” chamber and an “upper” chamber elected by the lower. While several other proposals were introduced, the Virginia Plan – introduced on May 29 – would serve as the basis for the Convention’s deliberation on organizing the legislature. The notion of a two-chambered legislature, similar to many of the States’ yet dissimilar to the legislature under the Articles of Confederation, was largely accepted by the delegates.

The Virginia Plan proposed that both chambers’ number of members would be decided proportional to state populations – an arrangement that would favor the larger states. In mid-June, a caucus of smaller states, fearful of the power this could offer large states, created a proposal in response – the “New Jersey Plan.” It preserved the Article of Confederation’s one-vote-per-state representation under a single legislative body. While this plan was rejected, it offered the smaller states a point around which to rally.

Having hit deadlock on upper chamber apportionment, the delegates referred the problem to a committee to reach compromise. There, a proposal by Richard Sherman, Delegate from Connecticut, was taken up. It called for representation in the lower chamber to be based on population, while states would be equally represented in the upper chamber. Known variously as the “Connecticut Compromise,” the “Sherman Compromise,” or the “Great Compromise of 1787,” the idea was supported by Benjamin Franklin, who – to satisfy the larger states – further proposed that matters concerning money must originate in the lower chamber. Despite continuing debate and disagreement, the Convention narrowly adopted the plan on July 16.

The issue of apportionment was settled, but the question of how many Senators would represent each state remained outstanding. Few delegates considered a single Senator per state as enough – recognizing that an absence would leave a state without representation, and that more Senators would increase the knowledge and competency of the body. Conversely, too large a Senate would undermine its distinct purpose and membership. The Convention debated the merits of two versus three Senators per state.

Gouverneur Morris of Pennsylvania – who believed that three Senators per state was necessary to form acceptable quorum – and Rufus King of Massachusetts arrived at a proposal to force the question: that the “representation in the second branch consist of [blank] members from each State, who shall vote per capita.” The Convention voted favorably on the proposal on July 23rd. Only Pennsylvania voted in favor of three Senators when it came time to decide, and the rest, save for Maryland, voted for two. With that, the Convention decided that the Senate would consist of two Senators per state, each voting independently instead of as a bloc.

Other features of the Senate also emerged out of the Philadelphia Convention – such as term length and Senate “classes.” Madison’s Virginia Plan, as originally proposed, did not specify lengths of upper chamber terms; rather, on June 13, the Convention, having consulted the composition of the States’ senates, reported an amended version which designated seven-year Senate terms. This was met with criticism from some of the delegates – Alexander Hamilton sought lifelong terms, for example, while Madison advocated for nine-year terms. The issue soon became linked with the notion of a Senate “class” system – a staggered rotation of election for the Senate.

On June 25, Nathaniel Gorham, Delegate from Massachusetts, suggested a four-year term with a fourth of the Senate body being elected each year. Edmund Randolph, of Virginia, supported the idea of staggered rotation; biennial elections, the delegates hoped, would bring stability to the Senate and protect the Senate from a rapid turnover in ideas. Arguing for a lengthy seven-year term, Randolph, with Madison, cited Maryland’s state senate – with its five-year term, longer than others in the union – as an example of a successful body with a several-year Senatorial term. Hugh Williamson, of North Carolina, countered that a six-year term would be more easily divisible into equal election cycles. On the 26th, Gorham brought the question of a six-year term, with a third of members going out every second year. The Convention considered, but rejected, a nine-year term, and then passed the six-year, three-class Senate by a vote of 7 to 4.

Similarly, the Virginia Plan left the age requirement for Senate membership up to the Convention. On June 25, three days after designating twenty-five as the minimum age requirement for Representatives, the delegates unanimously approved an age of thirty for Senators. Later, Madison, in Federalist No. 62, would argue that the Convention felt that Senators needed a “greater extent of information and stability of character” – which would come with age – than members of the House.

While the original Virginia Plan also made no mention of a citizenship requirement for Senate members, the reported draft included a four-year citizenship requirement. Delegates debated between “total exclusion of adopted citizens” and “hasty admission of them,” fearing both foreign influence and control of the Senate and closing the institution to naturalized citizens. The Convention struck a compromise between these opposing camps by settling on a nine-year requirement. Pennsylvania’s Gouverneur Morris moved to replace this clause with a fourteen-year minimum, while  James Wilson of Pennsylvania wished to reduce it to 6 years, but the Convention voted both down – along with thirteen and ten year minimums – before agreeing on a nine-year requirement on an 8 to 3 vote.

Who would be the Senate’s presiding officer? An early draft of the Constitution, presented on August 6, allowed the Senate to choose its own President and designated that individual as the executive’s successor. Yet, as clamor for a separation of powers between branches increased, the Convention opted to create the Electoral College. While doing so, it proposed that the newly created executive position of “Vice President” would serve as President of the Senate – an arrangement similar to New York’s constitution, which had the lieutenant governor be president of its Senate and have a decisive vote on ties.

George Mason, Delegate from Pennsylvania, and Elbridge Gerry, of Massachusetts, spoke against this proposal, believing it to be in conflict with the principle of keeping the executive and legislature separate. Yet Roger Sherman, in defending it, noted that if “the Vice-President were not to be President of the Senate, he would be without employment… and some member by being made President must be deprived of his vote.”  This seemed to sway most delegates, who passed the proposal on a vote of 8 to 2.

Finally, the Convention took up the question of nominations. Some wanted the executive to have sole power over appointment, while others – following the precedent established in the Articles of Confederation and most State constitutions – wanted the legislature to have the responsibility. The delegates, in a compromise, first decided to grant the president the power to appoint executive branch officers, while the Senate would appoint the judiciary. Yet this was not enough to satisfy delegates who favored a strong executive, nor those who feared an overly strong executive. Instead, on September 4, the Convention was presented an amended appointment process, modeled off Massachusetts constitution – which had divided responsibilities between the governor, who made nominations, and the legislative council, which confirmed them. On September 7, this model was unanimously approved for the new federal government, with the Senate holding the power to confirm nominations.

After finishing deliberation on other matters related to other branches and making final modifications to style, the Convention engrossed the new Constitution on September 15 and submitted it for signing on September 17 – where 39 of the 55 delegates present signed it. From there, it was released to the public to begin the ratification process. Nearly a year and a half later, on April 6, 1789, the first Senate, modeled upon the decisions made at the Convention, achieved its quorum and elected its officers. On March 4, 1789, the first Session of the first Senate began.