With the establishment of the national party convention, the process used to select the delegates to the national convention became of paramount importance. State and local party conventions selected the national convention delegates, but those conventions were often conducted in deeply undemocratic ways, excluding many party voters or using parliamentary rules such as winner-take-all and/or the unit rule to marginalize political minorities in the state. Nevertheless, the Democratic Party expressly endorsed the use of winner-take-all and unit-rule voting in the nomination process, which allowed party bosses to control the composition (and therefore candidate preference) of their state delegation. The Republicans were initially more hostile to boss control, forbidding the unit rule, but they, too, ultimately endorsed winner-take-all delegate selections in 1916. Moreover, both parties routinely seated delegates from states in which the convention process had been run in an undemocratic fashion. Thus, for most of the nineteenth and twentieth centuries, the party convention process was run by a small coterie of party bosses, who ultimately chose the party’s nominee.

The Framers at the Constitutional Convention were initially uncertain regarding how the President should be elected. For most of the convention, they favored appointment by Congress, but fears that the President would become too dependent on Congress persuaded them to entrust his selection to an independent body of electors – the Electoral College. In the Framers’ expectation, however, the Electoral College would just identify and nominate potential candidates; the House of Representatives would do the actual electing, selecting the winner from among the top candidates nominated by the College. Moreover, the Framers envisioned that the choice of President would be made on meritocratic, not partisan, grounds. After the Washington administration, however, national political parties emerged, transforming the Presidential election into a partisan contest. In this newly partisan environment, each party assumed the responsibility for selecting its nominee, and they initially did so through informal discussions among prominent party leaders.

Following the end of Reconstruction, southern states began adopting legal restrictions to prevent African-Americans from voting. Although the U.S. Supreme Court struck down state laws that expressly banned African-Americans from voting, the Court allowed states to use other, ostensibly race-neutral means to disenfranchise African-Americans, such as the poll tax and onerous registration requirements. Following the Second World War, the two national parties struggled with how to address the fact that their southern parties were excluding African-Americans from the nomination process and sending all-white delegations to the national convention. The Democratic Party regularly seated all-white delegations from its southern wing. Only in 1964 did it warn its southern parties that they could no longer exclude African-Americans from party affairs. Meanwhile, in the Republican Party, “lily-white” party organizations gradually took over the southern Republican parties and similarly excluded or marginalized African-Americans in party affairs. As a result, even into the 1960s, African-Americans in the south were regularly excluded from the presidential nomination process in both parties.

The McGovern-Fraser Commission had banned ex officio delegates so as to ensure that the nomination process was controlled by ordinary voters. The result was a convention in 1972 in which there were relatively few party leaders or officeholders. That convention also then selected a nominee who badly lost the general election, which party leaders blamed on the absence of officeholders and party leaders. To prevent the nomination of an unelectable candidate in the future, the party decided to reserve a small number delegate spots for party leaders and officeholders (PLEOs), but the PLEOs were pledged to support the candidates favored by party voters in their state’s primary election or caucus. Following another general election loss in 1980, the Democrats created a category of unpledged delegates who could support any candidate they wished – the so-called “superdelegates.” Although the superdelegates never played a decisive role in the selection of the nominee, fears that they would do so mounted over time, and in 2018, the Democratic Party stripped the superdelegates of the right to vote in the first round of balloting, effectively ending their influence in the process.

Early in the nineteenth century, the political parties formalized their nomination process, entrusting the choice of nominee to their congressional caucuses. Soon though, concerns about representation and corruption began to undermine the legitimacy of the process. In 1824, in an effort to displace the congressional caucus process, state Democratic-Republican legislative caucuses in five different states nominated five different candidates for the party. The result was disastrous, with the five candidates splitting the Electoral College vote and triggering the contingent election process. In 1832, keenly aware of the need for a centralized process that would select one nominee for the party, the Democrats settled upon holding a national convention of delegates from the various states to select their nominee. The creation of the national party convention, however, did not democratize the process as much as reformers hoped. As the first national party conventions revealed, most of the delegates were state and local officeholders and prominent citizens. Ordinary voters still remained largely excluded from the process.

Political scientists largely agree that, today, the modern presidential nomination process favors the rise of ideologically extreme candidates who contribute to the ideological polarization that the country is experiencing. Political scientists, however, disagree about the direction in which reform should move. Most political scientists believe that the process has become too democratic and that the cure for the current ideological polarization is to return the nomination process to the control of party leaders. This prescription for reform, however, ignores the fact that, when party leaders did control the process in the nineteenth and twentieth centuries, they too often chose ideological extremists or populist demagogues. Rather, as the Conclusion elucidates, the problem with the current process is that it is insufficiently democratic: the rules governing the process exclude too many ideologically moderate voters, thereby encouraging the selection of more ideologically extreme candidates. The Conclusion closes with several suggestions for how the nomination process could be opened more fully in the future so as to remedy this ideological polarization.

Although the McGovern-Fraser reforms opened the nomination process to ordinary voters in both parties, the democratization of the process proved to be incomplete. State laws and party rules often excluded many voters, such as independent voters in states that used a closed primary restricted only to party members. Restrictive party reenrollment rules also disabled many voters from changing parties in the weeks or months leading up to the presidential primary election, thereby preventing them from voting for the candidate they supported. Moreover, voters in some states found that they possessed less influence in the process than voters in other states. Voters in smaller states were allotted more national convention delegates (and therefore influence in the process) than their populations warranted, and, by the end of the twentieth century, it was apparent that voters in states that held their primary or caucuses early in the calendar had more influence than voters in later states.

As the 1968 convention had commanded, the Democratic Party empaneled a rules reform commission to examine the party’s nomination process, most notably the rules governing how national convention delegates were selected. Following a series of public meetings, the McGovern-Fraser Commission adopted eighteen guidelines that would fundamentally reshape the presidential nomination process in the party. To comply with the new guidelines, many states adopted the presidential primary, and, unlike early primary elections, the new primaries had to allow voters to choose which presidential candidate they supported (either by registering their preference directly or by choosing delegates whose candidate preference was expressly listed on the ballot). The new primaries also became binding, ensuring that the winner of a state’s primary election actually received the support of the delegates from that state. The end result was a nomination process that empowered ordinary voters for the first time in American history, and, because the new primaries were codified in state law, the McGovern-Fraser reforms also had the effect of democratizing the Republicans’ nomination process as well.

Towards the end of the nineteenth century, frustration with party boss control of the nomination process continued to grow. At the beginning of the twentieth century, at the behest of Progressives, several states adopted the presidential primary election, allowing voters to directly participate in the nomination process for the first time. Both national parties accepted the validity of the primary election process, but the early primaries did little to empower ordinary voters. In most primary states, voters did not directly vote for the presidential candidate but rather for the individual delegates to the national convention, whose candidate preference was often unknown and not disclosed on the ballot. As a result, uncommitted and favorite-son delegations, both of which were typically stand-ins for the state and local party bosses, often won the primary elections. Moreover, even when a candidate won a state’s primary, the national convention often allowed delegates from that state to vote for a different candidate; the primary result was not viewed as binding. Thus, despite the initial promise of primary elections, ordinary voters remained on the periphery of the nomination choice.

The 1968 Democratic National Convention illuminated all of the flaws in the presidential nomination process to that time: The ineffectiveness of presidential primary elections; the failure of party bosses to follow democratic norms in state conventions that selected national convention delegates; and the continued exclusion of African-Americans from the southern parties’ nomination process. Hubert Humphrey won the party’s nomination, but the manner in which he did so left many Democrats convinced that their nomination process was fundamentally flawed. Before it closed, the 1968 convention demanded that the party’s national committee create a commission to examine the party’s nomination process, with an eye to opening it up to ordinary voters in the future. In so doing, the 1968 Democratic National Convention set the stage for a reform movement that would fundamentally transform the presidential nomination process in both parties.

The Electoral College can misfire, electing a president who loses the popular vote (as in 2000 and 2016), but the presidential nomination process can misfire as well, producing a nominee (as in 1952 or, more recently, 2016) who is less popular among party voters than other candidates in the party. The cause of this is the byzantine web of state laws and party rules governing the process. This book explores those rules, enabling us to make sense of the process and understand how presidential candidates have been selected throughout American history. Surprisingly, for much of American history, the major party’s nominees were chosen by party leaders, not ordinary voters, and even today, the process is far less democratic than many imagine. Not every voter is able to participate in the process, and not every vote is weighted equally. This book examines the evolution of the rules governing the nomination process and how those rules contribute to the increasing ideological polarization of our politics today.