Most critics of the Electoral College would replace this relic of eighteenth-century constitutional thinking with a popular vote that would take place throughout a single national constituency that, for convenience’s sake, we can call the United States of America. In his new book, Edward B. Foley stakes a different criticism: our working Electoral College was not the one that the Constitutional Convention cobbled together at the tag end of its debates, but the one that was fashioned with the adoption of the Twelfth Amendment in 1803–4. The well-known result of that amendment was to allow presidential electors to “designate” separate votes for president and vice president, thereby solving the problem of the inadvertent tie between Thomas Jefferson and his ostensible running mate Aaron Burr that threw the 1800 election into the lame-duck House of Representatives. But an equally important consequence, Foley proposes, lay elsewhere: in the establishment of a Jeffersonian consensus that victors in presidential elections should acquire a “federal conception of a compound majority-of-majorities” (p. 37), meaning a majority of electors formed with a coalition of states wherein the winning candidate gained majority support.

It is this “electoral college of 1803” principle that has largely prevailed ever since, with a handful of exceptions led by James K. Polk’s election in 1844, which occurred only because in New York the antislavery candidate James Birney siphoned enough votes from Henry Clay to allow Polk to carry the state. But the spark igniting Foley’s book comes from three recent elections (1992, 2000, and 2016) when third-party candidates (John Anderson, Ross Perot, and the combo of Gary Johnson and Jill Stein, respectively) allowed the winning candidates to carry a number of states essential to their victory by mere pluralities. Where many commentators are driven to political madness when there is a disparity between national popular and electoral vote winners, Foley’s commitment to a state-based system of presidential elections leads him to a different conclusion. It is the danger of third-party candidacies preventing majoritarian decisions that drives Foley to analytical distraction. The solution to this ostensible dilemma does not lie through an Article V amendment to the Constitution, Foley concludes, but rather by having the states individually adopt electoral laws that would produce popular majorities. The most obvious of these would involve ranked-choice voting, redistributing the second-plus-n choices of voters supporting losing candidates until a majority is produced. But Foley ingeniously identifies several other methods that turn the majoritarian trick, including (to take one example) something like the French presidential runoff.

Foley is a distinguished scholar of American electoral law, and his expertise is evident in his careful review of problematic presidential elections and the alternative results to which rules for producing state-based majorities could have led. Yet, although one can be impressed with his imaginative parsing of electoral contingencies and procedural reforms, some elements of his argument remain subject to qualms, reservations, and outright objections.

Foley’s argument rests on a fresh reading of the Twelfth Amendment that is politically plausible but constitutionally problematic. He forces us to recognize that the debates over the amendment were not limited to providing a solution for the collective action glitch that threw the 1800 election into the House of Representatives. The congressional debates of 1803 included extensive discussions of the need for the electoral votes of individual states to reflect the will of their majorities. The issue of whether electors should be chosen by districts was also discussed, another reform that continued to be agitated in the 1810s and 1820s (as Alexander Keyssar demonstrates in his new book, Why Do We Still Have the Electoral College?). Foley’s bold move in elaborating this larger rationale for electoral reform is thus a significant achievement.

But the Twelfth Amendment did not pursue these further measures. Whatever tacit consensus existed in 1803 was not constitutionally entrenched. In practice, most presidential elections have satisfied the Jeffersonian norm because the two-party system is naturally conducive to that result. It is precisely because that outcome occurs so regularly that we have forgotten the historical origins of the Jeffersonian norm that Foley has recovered. Yet absent explicit constitutional language that would implement this norm, state legislatures were left free to manipulate electoral rules as they wished. As Foley concedes, after 1828 “plurality winner-take-all elections became the overwhelmingly dominant method that states used to appoint their presidential electors” (p. 50). This became the true electoral equilibrium that has operated ever since, the nominal Jeffersonian consensus notwithstanding.

Plurality winners will only occur when there are third-party candidates. The existence and potential persistence of these candidates are the true source of Foley’s concern. Rather than ask why we should maintain a system in which a victorious candidate can gain three million fewer votes nationally than his rival but win on the basis of narrow victories in three states, Foley prefers using ranked-choice voting, the one mechanism he likes best to avoid the Jacksonian-era resort to winner-take-all plurality victories. Yet it is one thing to argue that ranked-choice voting would work well for, say, reducing a city council pool of 15 candidates to a final cluster of 5 or 7 winners; it is another to ask why it should apply to the sole office in which the entire “executive power” of the United States is vested in a single person. Why should one privilege the second- or third- or nth-choices of voters who consciously want to make symbolic electoral statements in favor of noncompetitive candidates, rather than an effectual choice between the actual contenders? Perhaps Ralph Nader’s Florida voters might have preferred Al Gore to George Bush in 2000, or perhaps they would have just abstained from voting. The one thing we know is that Nader-style nudniks consciously decided not to vote for the obvious contenders in a battleground state that everyone knew was closely contested. Their initial choice should be respected as the free exercise of the suffrage by informed citizens, but it was also a decision to avoid a vote that would have affected the actual outcome of the election.

Finally, one other silent supposition of Foley’s normative scheme deserves consideration: his default commitment to the federal character of presidential elections. That was also part of the putative Jeffersonian consensus, and it remains the default condition that we seemingly cannot escape. Yet why, on principled democratic grounds, should one prefer a state-based presidential election system in which a federal “compound majority-of-majorities” could be gained without a majority of the national popular vote? Why should the modern norm of one person, one vote, not prevail in presidential elections today? A commitment to a federal scheme of presidential voting assumes that each state has some coherent interest that a majority of its voters are qualified to determine. But Americans do not determine their presidential preferences by asking which candidate will best serve the interest of their state. They do so instead on the basis of all those individual preferences that shape their individual political affiliations. The states are only the arbitrary geographic divisions that determine how these individual preferences are distributed across the nation. Optimal electoral reform should make the national expression of those accumulated preferences its proper goal through a popular vote in that single national constituency, the United States of America.