Paper presented at the Carnegie Council Fellows' Conference 2004. To access the full paper click on the "download" button at the bottom of the page.
During the early years of the Cold War, numerous scholars and statesmen readily accepted the claim that democracies, particularly the United States, could not initiate preventive wars against their adversaries. As the United States grappled with a dramatically changing threat environment created by the USSR’s development of its own nuclear weapons capability, this claim reflected a strong normative rejection of preventive war as a strategic option to arrest this power shift. Not only did President Truman declare that preventive wars were “weapons of dictators, not of free democratic countries like the United States,” NSC-68 of 1950 also asserted, “It goes without saying that the idea of ‘preventive’ war – in the sense of a military attack not provoked by a military attack upon us or our allies…would be repugnant to many Americans.” In 2003, however, the United States did indeed launch a preventive war against an adversary state. Moreover, the 2003 war with Iraq was authorized by a joint resolution from Congress passed by large majorities, and enjoyed consistent support from the American public, as reflected in national surveys.
The purpose of this research is to examine if there is in fact a general democratic anti-preventive war norm, to what extent this norm might be present among American citizens (at both the congressional and public levels), and to determine if there are any patterns in the distribution of this norm within the population. Evidence for this research was generated from several sources: 1) a content analysis of the transcripts of congressional debate over war with Iraq in October 2002; 2) public opinion polls prior to the initiation of war; and 3) a national opinion survey on preventive war conducted in November 2003 by the author with support from the Carnegie Council’s Fellows Program.
The main finding is that there is no generalizable democratic anti-preventive war norm in the United States. In the abstract, most Americans do agree that the U.S. should comply with a general principle against initiating preventive wars. A poll before the war showed a strong preference for deterrence (66%) over preventive war (25%), and in the survey conducted in November 2003, 76% agreed that the U.S. should not attack states that do not pose an imminent threat (18% disagreed). However, in the context of specific threat scenarios, support for the logic of preventive war climbs dramatically, to a point that discredits the claim that democratic citizens and political leaders consider preventive war normatively problematic. This does not mean that an anti-preventive war norm is completely absent. Evidence shows that a fair portion of the public and the Congress does indeed consider preventive war illegitimate as a way to address the threat of weapons of mass destruction proliferation, even in the post-September 11th threat environment. Of great significance, data from numerous sources confirm that support for or opposition to preventive war divides starkly based on political ideology and party identity.
The best evidence of normative resistance to preventive war comes from an analysis of the congressional debate over war with Iraq in the fall of 2002, which shows that a sizable number of senators (22%) and representatives (23%) explicitly reject the preventive war doctrine for normative reasons and refused to approve of war with Iraq in the absence of an imminent threat. In fact, among opponents of war this was the major reason cited. However, the evidence also shows that with few exceptions this normative view is confined to political liberals. The widespread support for preventive war in Congress was overwhelmingly driven by an explicit link that most supporters drew between Iraq and the threat of future terrorist attacks with weapons of mass destruction (71% of supporters in the House, 68% of supporters in the Senate). Even as most supporters of war acknowledged that this threat was amorphous, hard to prove and lacked immediacy, the risk was still considered compelling enough to justify preventive war in the absence of direct provocation.
Much the same can be said about the attitudes of the general public. As noted above, in the abstract most Americans do not embrace the logic of preventive war as a way to prevent power shifts to adversary states. Yet the level of support for or opposition to preventive war is contingent on the context of a specific case. When chemical, biological or nuclear weapons are specified as the military capability being developed by another country, and when the identity of the suspect country is specified (Iran in this survey), there is a strong upward shift in the number of respondents who would support the use of force under preventive war conditions (even if that country is not expected to directly threaten the U.S. with those weapons). This does not mean that American voters move en masse toward enthusiastic acceptance of using force under these conditions. Instead, American voters split on these questions: 50% support the use of force when chemical, biological, and nuclear weapons are specified, while 43% oppose this option; 47% support the use of force against Iran if it is suspected of developing nuclear weapons, while 43% oppose it. Most importantly, the best predictors of whether an individual will support or oppose the use of force under preventive war conditions are political ideology and party identification. Across every single question in the survey from November 2003, on the preventive war scenarios described above, on support for the use of force against Iraq and other suspect countries even after the failure to find evidence of Iraqi WMD, and in the level of trust in the Bush administration if the U.S. does not find Iraqi WMD, conservatives and Republicans by very large margins support the use of force and are willing to follow the Bush administration’s lead in these cases. In contrast, liberals and Democrats by margins just as large oppose preventive war and are unwilling to accept the Bush administration lead in these types of cases. Voters that characterized themselves as political moderates or independents fell between the extremes of support and opposition reflected in the attitudes of conservatives and liberals, and tended to split relatively evenly on most of the survey questions.