![]()
p.15
PART I
The Origins of Primary Elections
In the introduction to this book, we considered some aspects of the history of primary elections. As noted there, one of the most interesting aspects of this history is its relative obscurity. Many Americans have some understanding of the development of the current presidential nominating system, but there have been few detailed treatments of the development of the direct primary, of the establishment of primary elections for offices other than the presidency.
This lack of knowledge is reflected in the relative paucity of historical work on primary elections between the late 1920s and the early 2000s. Although there were some regional histories of elections published during this time, there were no book-length treatments of primary elections written between the publication of Charles Merriam and Louise Overacker’s Primary Elections in 1928 and Alan Ware’s The American Direct Primary in 2002. As Ware pointed out in his book, primaries were widely studied in the 1910s and 1920s, but in any generation, historians will bring their own biases and perspectives to the study of historical events. This means that there is much room today for a revisiting of questions regarding why primaries developed and why they spread.
The first section of this book provides examples of some of the best contemporary scholarship on the development of primaries. In each case, the scholars here bring to their analyses distinctly modern ideas about the motivations of politicians.
In Chapter 1 of this handbook, Alan Ware discusses difficulties in defining what a primary is. His chapter was chosen to lead off this book, in part because of his pioneering work in describing the origins of primaries, and in part because Ware, who has spent most of his teaching career in the United Kingdom, brings a different perspective to the study of primary elections than do many Americans. Ware’s intent in this chapter is not only to elaborate upon debates among early twentieth-century Americans to determine what a primary election should be, but also to question whether the “primaries” that have been adopted in many Western democracies over the past twenty years – and which are the subject of Part V of this book – are really analogous to the American direct primary.
Chapter 2 is by John Reynolds, the author of The Demise of the American Convention System, 1880–1911, published in 2006. As the title of that book suggests, Reynolds has studied the period immediately before the adoption of the direct primary, and his chapter shows that, for Reynolds, the adoption of the direct primary was not necessarily as radical a change to American politics as some have alleged. The seeds of the direct primary lay in part in changes in how candidates campaigned for office and in part in the increasing difficulty political parties had in placating different factions. Both Reynolds and Ware emphasize the strategic significance of adopting primaries; they were not as much a “good government” reform, as some Progressives would have had it, as they were a means for parties and party supporters to resolve disputes.
p.16
Another perspective on the adoption of primaries is offered by Jamie Carson and Ryan Williamson in Chapter 3. Carson’s approach to the direct primary has been less that of a pure historian and more that of a political scientist studying the development of Congress as a political institution. In his 2013 book Ambition, Competition, and Electoral Reform: The Politics of Congressional Elections Across Time (co-written with Jason Roberts), Carson explored the effects of changes in election laws in the nineteenth and twentieth centuries on the development of the incumbency advantage and of changes in general election competition. In the chapter here, Carson and Williamson describe some of the successes and failures of the direct primary. Contrary to the goals of Progressives, the primary did not make elections more competitive, but it did reduce the power of political parties and arguably made candidate selection more democratic.
Finally, it is worth noting that much of what we know, or at least expect, about primaries stems from assumptions about candidate and party behavior in multi-stage elections. That is, the winnowing process for which primaries are employed in the U.S. and elsewhere has a firm grounding in economic theories about how democratic choices are made. Formal models of candidate selection have been used for several decades now to explain what should happen in primaries, and many of these formal claims were rather informally stated even in the early years in which primaries were used. Chapter 4 presents a summary of many of the formal claims about the logic of primary elections. Gilles Serra develops a spatial model of the candidate selection process, and then explores the conditions under which political polarization might result from primary competition. He concludes that if voters rationally seek to help their party to win elections, polarization is unlikely. Serra’s theory provides a means of using some of the early claims about primary elections, addressed in the earlier chapters of this section, to inform the link between primaries and polarization that will be addressed in the empirical chapters of Parts II and III of this handbook.
![]()
p.17
1
WHAT IS, AND WHAT IS NOT, A PRIMARY ELECTION?
Alan Ware
Like many political concepts, the idea of a “primary election” has changed over time. As with both “liberal” and “conservative,” for instance, the transformation has been so great that a time traveler from the nineteenth century might struggle to comprehend its usage in some contexts today. Equally important, it has now become more open-ended in its application than in the earlier period, when its meaning was quite specific. The aim of this chapter is to explain how and why different meanings of the term “primary” or “primary election” developed over time, both within the United States and, much later, in other democracies. Unlike some American inventions, such as basketball, the original American model was subject to further modification once it was “imported” by other countries. However, and arguably more important, many changes had occurred within the United States itself before the use of primaries elsewhere. For that reason, the initial discussion here is exclusively about the U.S.
In the contemporary United States to speak of a primary election could be to refer to one of three rather different institutional arrangements:
1 For the historian of nineteenth-century America, it could be the first stage in a multi-stage process for selecting candidates in one of the two major parties. The term “primary election” was not used universally during that century, however, and in some regions of the U.S. primary elections were called either primaries (in the Mid-West and most of the West) or caucuses (in New England and a few western states). Yet they were “primary elections” in Pennsylvania and the South, as well as in the statute books of most states except those in New England (Dallinger 1897, n53). This concept of a primary is discussed in the first section of the chapter, along with the pressures that led to a radically different system of nominating candidates emerging form the subject of the first two sections.
2 From the early twentieth century onwards this meaning was largely abandoned, and primaries were now any formal election in the process of candidate selection by a party. Usually, speaking of a primary was to refer to a direct primary – that is, one in which the winner of the primary automatically became the party’s candidate. This was because direct primaries had become by far the most common method for nominating candidates. However, in some states, and especially for presidential nominations, a “primary” retained its original meaning of being the first stage in a nomination process. These and related matters form the subject of the third, fourth, and fifth sections.
p.18
3 From the second decade of the last century yet another meaning came into use when political parties in several states, mainly in the West, were barred from nominating candidates for some public offices. In the absence of direct party nominations to the ballot, a “primary” came now to refer to no more than the first election in a double-ballot electoral system. The primary is the election preceding a run-off election between the two candidates receiving the most votes in that initial election. This is discussed in the penultimate section of the chapter.
Until the last decades of the twentieth century primary elections were understood as an essentially American phenomenon. When, from that period onwards, parties in various other countries started to introduce what they sometimes called primary elections the term was reserved exclusively for direct primaries – that is, procedures for nominating candidates involving an election, the result of which determined who would be party’s nominee for some office. When confronted by it, which is rarely, Europeans and others usually find the third meaning of “primary election” (above) incomprehensible since there are other ways of describing the procedures identified. As for the original American meaning (1), non-Americans would normally find this usage confusing too. However, exporting the idea of a primary election overseas has also been accompanied by some extension beyond the core American notion that the purpose of such elections was to produce nominees for a general election. It is now starting to be applied to the selection of individuals to positions of leadership in a political party, a role that is both broader and more significant than their being candidates for public offices. These issues are examined in the final section.
Primaries in the Nineteenth Century
A simplified account of the emergence of primary elections would be that they developed from three sources during the 1830s. First, there was the much older New England tradition of direct democracy involving men taking decisions for their communities in town meetings. Second, during the 1820s there was a major decline of long-established social deference in much of America, deference that had survived the Revolution but which would persist only in a few places later, including Rhode Island (Silbey 1991). Finally, there was the impact of the Jacksonians whose conception of democracy was that its practice was made possible by people engaging with the activities of political parties. The parties became rejuvenated between the mid-1820s and the late 1830s with a large mass base of participation. In this regime, nominating candidates was a process that began, and in the case of local offices ended, in the lowest level of governmental structures. Anyone eligible to participate in a party nomination, with eligibility usually being defined broadly for white adult males but with others usually excluded, was encouraged to do so at the first stage of candidate selection. (To use the term increasingly popularized by political scientists today, the parties’ “selectorates” were large.1) For higher-level offices, participation would entail them being involved in electing delegates to subsequent stages in the nomination process.
As noted earlier, there were regional differences in what that first stage was called, but functionally they were identical. Although caucuses in New England had originally lived up to their name, and involved prior discussion and not mere voting for candidates, this had to be abandoned later. In the decades after the Civil War debating ...
