This thesis asks when and why Democrats and Republicans respond differently to American military interventions abroad. Using quantitative and case-study methodology for conflicts in the post-Vietnam era, it concludes that partisans in the mass public respond to conflicts in ways that comport with their basic worldviews and deeply held values.
THIS THESIS IS not about the war in Iraq in which the United States of America presently finds itself engaged. But it most certainly is inspired by that war. Pundits and scholars will debate for decades the consequences of President George W. Bush’s fateful decision to topple Saddam Hussein’s government in the spring of 2003. One such consequence, though, is so readily apparent that it has already served as fodder for dozens of articles in the popular press and for a handful of academic works. It is that members of the Republican Party are profoundly more supportive of the war than are members of the Democratic Party. Among all the cleavages in American society—race, religion, class, region, gender, education—none exerts anywhere near as strong an influence on an individual’s likelihood to favor the American presence in Iraq as party affiliation does.
Partisanship in American politics is not new, but the intensity with which it now manifests itself on a basic question of war and peace is. “No military conflict in modern times has divided Americans on partisan lines more than the war in Iraq, scholars and pollsters say—not even Vietnam,” wrote Robin Toner and Jim Rutenberg in the New York Times.1 Gary Jacobson has observed that by early 2006, the gap between Republican and Democratic support for the war had widened to an astonishing 60 percentage points, more than twice that observed for any other war since the rise of public-opinion polling.2 In its 2005 political typology of the American public, the Pew Research Center found that while economic and social issues produced divisions cutting across—rather than along—party lines, “foreign affairs assertiveness now almost completely distinguishes Republican-oriented voters from Democratic-oriented voters; this was a relatively minor factor in the past.”3
Though the causes of this unprecedented level of partisan polarization remain rather hazy, it is generally considered to be an unhealthy development.4 Some have speculated that partisanship is now at such an extreme level that the foreign-policy positions espoused by parties and candidates must be more a function of domestic political considerations than of any substantive vision. “We have reached a situation that, were the partisan composition of the White House and Congress reversed, spokesmen for both parties would change positions on a dime,” said presidential scholar Al Felzenberg, arguing that the partisan divide can be attributed to opportunistic elites who engineer mass polarization to serve their own political ends.5 That accusation, if true, strikes at the deepest theoretical justifications for multiparty democratic government. Is the Republican–Democratic divide over the war in Iraq a function of partisanship for its own sake, or are more profound causes at work?
The Puzzle, the Context, and the Argument
As interesting a question as this is, it is too difficult at this stage to pin down with any precision the causes of the vast partisan gap in support for the war in Iraq, whose history is still being written. The preceding decades, however, can offer guidance on the dynamics of partisan support for war. In the years between the end of the Vietnam War and the beginning of the current Bush presidency, the U.S. used military force in numerous instances around the globe, under a vast array of different circumstances. Some interventions were undertaken by a Republican president, others by a Democrat; some lasted for years, others for a single day; some tried to topple Communist dictatorships, while others strove to keep fractious societies from erupting into sectarian warfare.
As a result of these varied circumstances, some of the interventions resulted in fierce partisan strife, though none on the scale of the current conflict in Iraq. The study of these cases has prompted legions of scholars to ridicule the maxim that “politics stops at the water’s edge.” (There may be no more widely mocked aphorism in all of political science.) But in today’s partisan climate, it is easy to forget that other interventions, as recently as a decade ago, produced almost no divide between members of the two parties. The dance of consensus and dissensus has been an ongoing one since it first emerged in the wake of the Vietnam War.6
This is the puzzle I seek to explain: what factors account for the variation in both the magnitude and direction of the partisan gap observed in public support for the use of force during the last two decades of the twentieth century? I hope to bring evidence to bear on the question of when and why Republicans and Democrats respond differently to military intervention. Not only are these important matters in their own right, but they can help make sense of the public’s highly partisan reaction to the conflict in Iraq. They also point to the areas that should be the focus of attention for future research into the intersection of partisanship and public opinion in wartime.
To a first approximation, there are two competing conceptions of American public opinion on foreign policy. One, which dates to the early and middle decades of the twentieth century but has roots even deeper in the Western intellectual tradition, takes a dim view of the public’s capacity for responding intelligently to matters of state. From the end of the Second World War until the 1970s, the bulk of the American foreign-policy establishment subscribed to its basic tenets—that the public was ill-informed about foreign policy, held poorly defined and fickle views on the most pressing issues of the day, and responded impetuously to developments abroad. The second school, a revisionist camp, is more sanguine about the nature of public opinion on foreign policy. Drawing heavily on public-opinion polls from the past thirty years, it holds that the mass public is capable of responding to foreign crises intelligently, using information heuristics and cues from trusted elites in government and the media to arrive at policy judgments that reflect real, coherent preferences. It first gained adherents inside the academic community, but grew in influence as the Vietnam War demonstrated the public’s capacity for holding enduring, passionate views about international affairs.
My argument falls squarely in the second camp. The title of the thesis, a play on that of an influential work within the revisionist school of thought, sums up my basic claim.7 I contend that Republicans and Democrats in the mass public are “pretty prudent” in the sense that they respond to the use of force in predictable and logical ways. They tend to take positions that are shaped to a substantial degree by their deeply held, fixed attitudes about the desirability of American hegemony and the acceptability of violence as a method of solving international problems. Their responses to any given military intervention are functions both of the nature of the intervention itself—its principal objective, the ideology of the adversary, the presence or absence of key allies—and of cues from the president and members of Congress.
Scholars have demonstrated the ways in which factors such as these affect public opinion as a whole. No one, to my knowledge, has undertaken a systematic study of the post-Vietnam period looking separately at Republicans and Democrats. The popular attention to partisan polarization in the case of Iraq is not mirrored by any scholarship on the period leading up to it. In analyzing interventions between 1980 and 1999, I conclude that Republicans and Democrats have substantive, fundamental value disagreements as to when military force should be used. Republicans, relative to Democrats, are more likely to favor interventions directed toward regime change, but less likely to favor the use of American troops as peacekeepers. They place less emphasis on gaining multilateral support for intervention, and are more inclined to support military action in an abstract sense. Elites within each party shape the behavior of followers, but actual on-the-ground circumstances exert a significant and independent influence on the responses of members of both parties.
These findings help explain why the gap between members of the two parties is far greater in some cases—for instance, during interventions that occur under Republican presidents, and during those of longer duration—than in others. They enable us to predict the circumstances under which Republicans will be relatively more or less inclined than their Democratic counterparts to support American military operations. While my results cannot directly explain the current partisan divide on the Iraq conflict, they imply that that divide is probably neither a coincidence nor a product of mindless partisanship. Rather, given the ways in which Republicans and Democrats have responded to other interventions in the recent past, it can reasonably be viewed as a logical result of the nature and circumstances of American actions. As a prolonged intervention undertaken by a Republican president, aimed at regime change, and garnering substantial opposition both internationally and from Democratic elites, the current war in Iraq bears many of the hallmarks of a conflict that could be expected to produce a substantial gap in public opinion in the direction of greater relative Republican support.
Methodology and Overview
My argument relies primarily on evidence from public-opinion polls, both to describe the general foreign-policy beliefs of the mass public and to measure levels of support for specific military interventions included in my analysis. A word is in order about the suitability of such polls—can they really be used to paint an accurate picture of public opinion on the use of force?
Polls are, to be sure, blunt instruments. They are highly susceptible to a litany of problems that can seriously impair their capacity for providing accurate information.8 These include biases in sampling and selection, low response rates, and uncertainty in interpretation. There are two objections in particular that can be raised to the use of polling data in the context of this thesis. The first is the problem of nonattitudes.9 It is not clear whether members of the mass public can really be expected to have well-defined views on the wisdom of using force in far-off countries about which they know very little. Second is the problem of question wording. It is well documented that small differences in wording between surveys can produce large swings in answers to questions.10 This is especially troubling here, because it is easy to imagine instances in which Republicans and Democrats might react differently to subtle wording cues in a polling question.
The problem of nonattitudes is discussed more thoroughly in the second chapter. While public knowledge about several of the interventions included in my analysis was indeed very low, there is strong evidence that the public made efficient use of its limited knowledge in responding to poll questions. More significantly, the model I construct can itself be viewed as a test of the nonattitudes problem. If, in fact, public responses to polls lack substance and reflect nothing more than random variation, we should not observe any relationship between public opinion and the circumstances of interventions. If we do detect such a relationship, it can be considered prima facie evidence that the public’s views are more concrete than the nonattitudes hypothesis suggests. I address the problem of question wording, to the greatest extent possible, by omitting questions with potentially biasing or misleading cues.
