Early Presidential Elections: The Questionable Use of Electors to Correct Voter Imbalances

An important issue that the Congressional delegates faced when drafting the Constitution was how to create an equitable balance in voting rights between the larger states (Massachusetts, Pennsylvania, Virginia) and the smaller ones (Delaware, Georgia, New Hampshire). Although the delegates were sworn to secrecy throughout their debates (May through September 1787), once the debates were over the delegates were at liberty to freely express their views and many did so. On November 29, 1787, Luther Martin, a Congressional delegate from an intermediate size state, Maryland, delivered the following remarks to the Maryland House of Representatives.

When I join’d the Convention I found that Mr. Randolph [Governor of Virginia] had laid before that Body certain propositions for their consideration . . . one of which was that [as few as] seven States might proceed to Business, and therefor four States composing a Majority of seven, might eventually give the Law to the whole Union . . . It must be remembered that in forming the Confederacy the State of Virginia proposed, and obstinately contended (tho unsupported by any other) for representation [in voting] according to Numbers [that is, population size] . . . These Views in the larger States, did not escape the observation of the lesser [states] and meetings in private were formed to counteract them[1]

Among the delegates’ major concerns over this matter was how to elect a chief executive officer for the country as a whole and, at the same time, provide an appropriate means to cope with the vast population differences that existed among the several states. To address this matter, on July 19, 1787, Oliver Ellsworth suggested that the states should not be allotted votes based on population size alone but on a ratio system devised to reflect a proportional representation of each state’s population. To begin the discussion, Ellsworth moved that the president should “be chosen by Electors appointed for that purpose by the Legislatures of the States . . . One person from each State whose numbers shall not exceed 100,000, Two from each of the others, whose numbers . . . shall not exceed 300,000, and Three from each of the rest.”[2] On July 20 James Madison underscored the significance of Ellsworth’s motion when he observed,

that this would make in time all or nearly all the States equal. Since there were few that would not in time contain the number of inhabitants entitling them to 3 Electors; that this ratio ought either to be made temporary, or so varied as that it would adjust itself to the growing population of the States.[3]

While a vote on the first part of Ellsworth’s motion (appointed by the Legislatures) was approved, a vote on the second part (the nature of the ratio) was postponed. As the debate continued, a further concern also arose “from the disposition in the people to prefer a Citizen of their own State, and the disadvantage this would throw on the smaller States.”[4] To address this added concern, on July 25 Hugh Williamson suggested,

that each man [elector] should vote for 3 candidates. One of these he observed would be probably of his own State, the other 2, of some other States; and as probably of a small as a large one.[5]

In agreement with Williamson, but to simplify his suggestion, Gouverneur Morris then proposed that “each man should vote for two persons [instead of three] one of whom at least should not be of his own State.”[6] According to Madison the only objection to Morris’s double-voting procedure would be that the elector,

after having given his vote for his favorite fellow Citizen, would throw away his second on some obscure Citizen of another State, in order to ensure the object of his first choice. But it could hardly be supposed that the Citizens of many States would be so sanguine of having their favorite elected, as not to give their second vote with sincerity to the next object of their choice.[7]

While neither Williamson’s suggestion nor Morris’s proposal were ever put to a vote, it was Morris’s wording, with two clarifications, that eventually found its way into the following material in Article II Section 1 of the Constitution which was sent to the House in final draft form by the Committee of Revision on September 12, 1787:

The electors shall meet in their respective states, and vote by ballot for two persons, of whom one, at least, shall not be an inhabitant of the same state with themselves . . . The person having the greatest number of votes shall be the President . . . In every case, after the choice of the President by the representatives, the person having the greatest number of votes of the electors shall be the Vice-President.[8]

The first clarification involved the phrase, “The electors shall meet in their respective states, and vote by ballot.” These words were added to avoid the possibility of collusion if all the electors were to meet at the same time and in the same location when they nominated their candidates and cast their votes. The second clarification involved the term vice-president, which was added merely to provide a home for one of the two votes, since it was not initially clear what the duties should be for the person holding this office.[9] The vice-president’s duties were eventually defined during the final stages of ratification.[10]

Despite the second clarification, without also specifying which vote should apply to which office, it remained unclear how the elector’s votes should be tabulated. The only information in Article II Section 1 that addressed this matter stated that the final nominations along with their respective numbers of votes, would be recorded, sealed, and transmitted to the president of the Senate for final tabulation.[11] Unfortunately, this wording did not address the issue at hand. To rectify this situation, it was decided that the candidate with the greatest number of overall votes, based on one of the two allotted votes, would become president. The candidate with the next highest number, based on the other of the two allotted votes, would then become vice-president. When ties occurred, the House of Representatives would determine the outcome.

Implications

Because of the clause that permitted the electors to “vote by ballot for two persons, of whom one, at least, shall not be an inhabitant of the same state with themselves,” the electors, in fact, were now totally free to cast one of their two votes for their state’s favorite son, which is precisely what happened, and which, in turn, led to a series of results diametrically opposite to what the delegates initially had hoped to achieve. In essence, the endorsement of this clause had permitted the large states to dominate the outcome of many of the early presidential elections.

In 1797, for example, Massachusetts, using one of its allotted votes, gave all sixteen of its electoral votes to John Adams, the state’s favorite son, which enabled him to win the presidency in 1797. The state then spilt the second of its allotted votes and gave a further sixteen electoral votes to candidates from Connecticut (Samuel Johnson two votes, Ellsworth one vote) and South Carolina (Thomas Pinckney thirteen votes). Similarly Virginia, using one of its allotted votes, gave all or nearly all of its electoral votes to its favorite sons: Washington in 1789 and 1793, and Thomas Jefferson in 1797 and 1801, then spread the same number of electoral votes during each of these years across several of the other states. Although Jefferson failed to win the presidency in 1797, he did win the vice-presidency that year but only with Virginia’s help.[12]

The one exception to this procedure was Pennsylvania. Presumably, owing to the poor health and subsequent death of its favorite son, Benjamin Franklin in 1790, and the apparent lack of any other favorite sons who chose to run for office, Pennsylvania opted to support the favorite sons of Virginia in 1789, 1793, and 1797. Only in 1801 did its electors choose to split their votes equally between Jefferson from Virginia and Aaron Burr, the favorite son from New York.

Although this highly questionable procedure was finally brought to an end in 1801, the end occurred not because of this procedure, but because of a tie vote in the election that year when Jefferson and Burr each received an overall total of seventy-three electoral votes that required thirty-six ballots in the House to resolve. As a result, Article II Section 1, came under severe attack over what has since been labeled “one small defect” in its construction, namely, its failure to designate a specific office for each of the two votes.[13] To correct this defect, the Twelfth Amendment, approved on June 15, 1804, mandated the assignment of one of the votes to the office of president and the other to the office of vice-president with the names of both to appear on the ballot in accordance with the following wording:

The Electors shall meet in their respective states and vote by ballot for President and Vice-President, one of whom, at least, shall not be an inhabitant of the same state with themselves; they shall name in their ballots the person voted for as President, and in distinct ballots the person voted for as Vice-President

Despite this correction, a careful reading of the amendment shows that it contained words identical to those originally used by Morris, which meant that his procedure remained totally intact. If the electors did not allocate any votes to a candidate from their own state who was running for the office of vice-president, it was still possible for the state’s electors to allocate all their votes to their state’s favorite son who was vying for the office of president, which again is exactly what happened. Between 1805 and 1821 Virginia allocated all its electoral votes to Jefferson, Madison, and James Monroe, while Massachusetts, in 1817, cast all its votes for Rufus King, a resident of Massachusetts. Pennsylvania, as before, opted to allocate all its votes in the same manner as Virginia. Each of these states, also as before, then distributed their same allotment of votes to candidates from a number of other states who were vying for the office of vice-president.[14]

Discussion

In spite of the lengthy debates that had taken place among the delegates with the aim of equalizing voting across the states, with Morris’s provision in place the favorite sons of the larger states continued to play an important, if not a dominant role, in the choice of president for the United States in all the elections from 1789 through 1821. In view of this outcome, it is worth asking if a procedure other than Morris’s could have led to a different result. On June 2, 1787, James Wilson introduced the following resolution.

That the states be divided into districts—and that the persons, qualified to vote in each District, elect Members for their respective Districts to be electors of the Executive Magistracy. That the electors of the Executive Magistracy meet and they or any of them shall elect by ballot, but not out of their own Body, a Person in whom the Executive authority of the national government shall be vested.[15]

The major difference between Morris’s procedure and Wilson’s was that the latter called for a single vote by the electors whereas the former required a double vote. Of equal importance, although neither allowed the states to vote for their own favorite sons, Wilson’s explicitly eliminated this likelihood, whereas Morris’s did not. Unlike Morris’s procedure, which was never put to a vote, Wilson’s was defeated by a vote of seven to two.[16]Nevertheless, it is instructive to consider what would have occurred if Wilson’s had been approved.

By reporting the total electoral count under each procedure for the first four presidential elections, the following table permits a direct comparison in outcome between the two. Although the total count under Morris’ procedure remained as before, under Wilson’s approach the new count was obtained by subtracting the number of votes allocated by each state to the state’s own candidate. For example, under Wilson’s approach in 1793 the count allocated to Virginia for Washington was reduced by twenty-one, for Adams from Massachusetts, the count was reduced by sixteen, and for George Clinton from New York, the count was reduced by the twelve.[17]

While no changes resulted in the first two elections under either procedure because of the very large number of electoral votes that Washington had received in both, such was not the case for the next two elections. In 1797 under Wilson’s approach, Pinckney from South Carolina, instead of Jefferson from Virginia would have become vice-president, and in 1801 Burr from New York instead of Jefferson would have become president without the need for the House of Representatives to intervene. While neither candidate was from a small state, using Wilson’s approach both were elevated in their rank order positions and therefore both provided alternatives to the large state nominations.

Conclusion

Given that Wilson’s approach was more in keeping with the delegates’ desire to minimize the importance of the larger states than Morris’s, it is reasonable to ask why Morris’s procedure was endorsed whereas Wilson’s was not. The most promising explanation that comes to mind stems from the remarks attributed to Luther Martin, as quoted above, as well as from Jefferson’s summary of events that were unfolding at the time: “The larger colonies had threatened they would not confederate at all if their weight in Congress should not be equal to the numbers of people they added to the confederacy.”[18] Because both of these remarks strongly suggest that the larger states were clearly in favor of maintaining their positions of authority in governing the country, it would not be unreasonable to assume that, by supporting Morris’s approach and not Wilson’s, such action by the larger states would have been to their advantage.

In line with this explanation, recall that two of the large states, Pennsylvania and Virginia, when making use of Morris’s approach, assigned all or nearly all of their electoral votes to Virginia’s candidates. Needless to say, their combined action not only skewed the overall results in favor of these states but also completely dwarfed the impact of the small states. By way of example, consider once again the election of 1797. Whereas Virginia allotted twenty of its twenty-one electoral votes to Jefferson, and Pennsylvania also gave fourteen of its fifteen votes to him, this total of thirty-four votes meant that even if the five small states (Delaware, Georgia, Kentucky, Tennessee, and Vermont) had also acted in concert, they would only have been able to accumulate a total electoral vote count of eighteen, well below the number needed to overcome the enormous lead of these two large states in their choice of Jefferson.[19] In view of the delegates’ level of education, range of experiences and overall sophistication, coupled with the elementary mathematics involved, it is difficult to believe that this future scenario would not have been realized by the delegates when they opted to favor Morris’s approach over Wilson’s.

[1]Records of the Federal Convention, 3 (November 29, 1787), Appendix A, CXLVIb, 151.

[2]Records of the Federal Convention, 2 (July 19, 1787), 50.

[3]Ibid., July 20, 1787, 63.

[4]Ibid., July 25, 1787, 111.

[5]Ibid., 113.

[6]Ibid.

[7]Ibid., 114.

[8]Elliot’s Debates, 1 (September 12, 1787), 302.

[9]Ray Raphael, Mr. President(New York, NY: Alfred A. Knopf, 2012), 118.

[10]Records of the Federal Convention, 2 (September 4, 1787), 498.

[11]Elliot’s Debates, 283.

[12]Svend Petersen, A Statistical History of the American Presidential Elections(New York, NY: Frederick Ungar Publishing Co., 1968), 11-13.

[13]Raphael, Mr. President, 227. For a full account of the origin of the controversy surrounding Article II Section 1 see Tadahisa Kuroda, The Origins of the Twelfth Amendment(Westport, CT: Greenwood Press, 1994).

[14]Petersen, A Statistical History, 14-16.

[15]Records of the Federal Convention, 1 (June 2, 1787), 77; Elliot’s Debates, 1 (June 2, 1787), 156.

[16]Records of the Federal Convention, 77.

[17]Petersen, A Statistical History, 3.

[18]Eli Merritt, Disunion Among Ourselves(Columbia, MO: University of Missouri Press, 2023), 147.

[19]Petersen, A Statistical History, 3-12.