In the 2000 U.S. presidential debates, George W. Bush criticized Vice President Al Gore for his role in the Clinton administration's foreign interventionism. Governor Bush pledged, instead, to have a "humble" foreign policy that would be "judicious in its use" of the American military. A president-elect Bush would differ from the Democratic administration by not "over-committing our military around the world," and by not engaging in a costly, "nation-building mission" that left the invaded country no better off than it was before (Bush 2000). The foreign policy positions of the two candidates were not surprising. They represented the two different ideologies of the parties during the Clinton administration: Democrats defending U.S. intervention in world affairs and Republicans arguing for less intervention (Schlesinger 1995).
From our contemporary standpoint, we know that the parties changed positions in the ensuing years. Just a few years later, the Democratic Party was criticizing Republicans for reckless war making, "over-committing our military around the world," a foolish "nation-building mission," and wasting American money and lives. This change was not only reflected in the discourse of party elites, but also in the attitudes of ordinary party identifiers. The American National Election Studies (ANES) regularly asks Americans whether they agree or disagree with the statement that "this country would be better off if we just stayed home and did not concern ourselves with problems in other parts of the world." In 1998, more Democrats than Republicans gave the interventionist response, but by 2002, significantly more Republicans than Democrats gave the interventionist response.
This evolution in Republican Party ideology, becoming relatively more interventionist under Bush, and the corresponding change in the Democratic Party, becoming relatively less interventionist, is just one of numerous instances in American political history of the two major parties changing their positions, rhetoric, and ideologies with regard to foreign policy. As students of American politics, how should we understand these developments? Are there any structural factors at work that can help explain when, why, and how party positions, rhetoric, and ideas change over time? Or, are these changes simply the product of historical contingency? This article seeks to answer those questions and finds that, in general, parties in long-term control of the presidency tend to become more interventionist on foreign policy while parties in opposition tend to become less interventionist.
Most scholars who have studied American party ideologies have focused on party ideas about "economic redistribution" or how active the federal government should be in managing the economy--what spatial modelers often call the "first dimension" of political ideology. This article instead focuses on party ideas about foreign policy, which have received less attention from students of American political parties and ideologies. (1) Those who have studied ideas about foreign policy have drawn a number of important conclusions, but none have tried to explain how party ideologies evolve.
For example, scholars of international relations have focused on the importance of elite ideology in determining American foreign policy, but have paid less attention to parties (Holsti 2006; Nau 2013). Early public opinion scholarship paid attention to parties, but argued that American attitudes toward foreign policy were largely incoherent: they lacked "intellectual structure and factual content" (Almond 1950, 56), (2) and they lacked ideological constraint (Campbell, Converse et al. 1960; Converse 1964). While more recent scholars have admitted that ideology matters in determining who gets elected and what foreign policies government officials pursue (Hurwitz and Peffley 1987; Gries 2014), this scholarship, in general, takes a snapshot view of parties, ideology, and foreign policy, rather than a moving picture view. (3) The scholarship that has examined party ideas about foreign policy over time has argued for stasis rather than dynamism (Dueck 2010). This article adds to this literature by focusing on the changes we observe in party ideologies with respect to foreign policy.
This article improves our understanding of American party ideology development in two ways. First, it emphasizes the analytical and conceptual insight that ideologies are endogenous structures subject to change over time by political actors. By reminding ourselves of this fact, we can better understand what ideologies are, what they do, and how they evolve. Second, once we understand that ideologies--including party ideologies--can, and do, change, this article points out that a polity-centered theory can help explain how and why party ideologies evolve. In particular, this article shows how party control of the presidency influences change in party theories of foreign intervention.
In making these two contributions, this article will proceed as follows. The next section explains how we should conceptualize party ideology development, and why this is important. The following section outlines a theory to explain variation in party ideologies over time and derives a hypothesis from this theory with regard to foreign policy. The rest of the sections test this hypothesis by examining nine observations that cover American history since 1900.
Conceptualizing Party Ideology Development
Ideologies are not static, philosophical structures that exist eternally, and immutably, in the realm of Platonic forms. Instead, ideologies are social and political constructs that are constantly subject to change. For example, famously, in the nineteenth century liberalism usually referred to laissez faire free market policies, but in the twentieth century liberalism came to represent government intervention in the economy. In the United States, Franklin Roosevelt (FDR) was crucial in changing the meaning of liberalism in this way (Rotunda 1986). To take another example, in terms of foreign policy, in the 1930s--40s "conservatism" referred to "America first" isolationism, while "liberalism" referred to hawkish internationalism. In the 1970s, "conservatism" had come to represent hawkish internationalism while liberalism had come to represent "bring America home" dovishness. These ideological evolutions are frequent and characteristic of American political development.
The evolution of ideologies is driven by political actors--whether politicians, political activists, political journalists, or ordinary citizens. As Converse explained, "the shaping of belief systems of any range into apparently logical wholes that are credible to large numbers of people is an act of creative synthesis" (1964, 211). Of course, once an ideology is created, it does not remain static: it is constantly undergoing transformations that 4 are also the product of further "creative synthesis." In this way, ideologies are endogenous political structures.
Problems with Our Current Understanding of American Political Parties
The dynamic character of ideology is important to remember because--without understanding this--we may be led to false inferences about the development of American political parties. For example, for most of the past century, America has had one major party identified with liberalism (the Democratic Party) and one major party identified with conservatism (the Republican Party). If we wrongly assume that the meaning and content of liberalism and conservatism are static, then we wrongly conclude that--as long as Democrats have been liberal and Republicans have been conservative--that the two major parties have not changed their ideas, rhetoric, or positions over time. When we realize that ideologies like liberalism, conservatism, and progressivism are constantly changing, then we can realize that a party is liable to dramatic change over time even when it identifies with the same ideological label.
To illustrate this point, I will briefly look at the problems we encounter when we attempt to use congressional roll-call scaling applications, like DW-NOMINATE, to measure party ideology development. Roll-call scaling applications posit that ideological positions on a liberal-conservative spectrum determine the roll-call voting behavior of a party's members of Congress (MCs), but those who use these applications typically do not explain how the meaning and content of their liberal-conservative spectrum changes over time. It is unclear what it means to say that a politician in one decade and a politician in another decade have the same DW-NOMINATE score and, therefore, the same ideological constraint. Without a detailed description of what conservative or liberal scores mean at each point in time, many scholars and journalists falsely assume that the meaning of these scores has remained static. In truth, the voting pattern of a liberal Democrat or conservative Republican in one period often represents the opposite issue positions of a liberal Democrat or conservative Republican in another time period.
It is true that some of the issues that liberals and conservatives debate change over time, but it is also true that the positions liberals and conservatives take on enduring issues change over time. For example, an extreme "conservative" DW-NOMINATE score in 1955 often described a Republican opposed to tax cuts and opposed to interventionist foreign policy. In contrast, an extreme "conservative" DW-NOMINATE score in 2005 often described a Republican in favor of tax cuts and an interventionist foreign policy. The same kinds of issue position reversals among liberals and conservatives can be found with regard to virtually every enduring issue in American politics. If the same number on an ideology index represents two ideologies that are the opposite of each other, depending on the time period...