Skip to navigation – Site map

Document 14

S. Toner on Gelphi, Feaver and Reifler’s Paying the Human Costs of War

Full text

1Christopher Gelphi, Peter D. Feaver and Jason Reifler. Paying the Human Costs of War: American Public Opinion & Casualties in Military Conflicts. New Haven: Princeton University Press, 2009. 280 pp. ISBN 978-0691139081

2In one of their less astute observations the authors of Paying the Human Costs of War suggest ‘that the United States does not go to war often’ (67). While war and public opinion is a notoriously difficult subject to study, few countries offer as great an opportunity to study the topic as the United States. Despite their dismay at the apparent paucity of foreign military engagements to examine, Christopher Gelphi, Peter Feaver and Jason Reifler have made an interesting and sometimes controversial contribution to the literature on military casualties and public opinion. In examining the way in which the US public structures its attitudes toward war and military casualties, they found that it was the interactive effect of prospective judgements about the likelihood of a mission succeeding and retrospective judgements about the initial rightness of the war that led Americans to continue supporting a mission in the face of mounting casualties (20). Of these two, however, expectations of success is the most important factor affecting casualty tolerance.

3The book grew out of the Duke University project ‘Wielding American Power’ and seeks to probe the extent of the US public’s casualty tolerance. Perhaps not surprisingly, given the Duke project’s title, the authors are interested in uncovering the factors that lead to continued public support for some wars despite substantial casualties. Rather than viewing public opinion as a welcome limitation on risky American military adventures, they often treat it as something that can be shaped and manipulated. Gelphi and Feaver presented the findings of their opinion polls to the Bush administration in 2004 and although they also consulted with the Kerry campaign team, Feaver’s media contributions leave little doubt as to his political inclinations.1 Feaver subsequently went to work at the National Security Council from 2005- 2008 and it certainly seems that his research fed into the presidential rhetoric of the time, particularly the desire to frame the war in Iraq as heading toward victory.2

4In his landmark works on war and US public opinion John Mueller of the Ohio State University came to the conclusion that there existed a ‘simple association: as casualties mount, support decreases’.3 Mueller is Gelphi, Feaver and Reifler’s intellectual bête noire. For these authors ‘the image of the American public as a paper tiger - a mirage of strength that collapses in the face of casualties- is as incorrect as it is popular’ (2). They argue that US policymakers are not as constrained by public opinion as the conventional wisdom would hold.

5The authors argue that the public is capable of structuring its attitude to foreign engagements in a coherent manner and engages in a rational cost-benefit analysis when lending its support to a mission. Gelphi et al. reject Mueller’s ‘inexorable decline’ argument by examining support for American wars at times of both battlefield success and battlefield defeat. They claim that US citizens have a much greater tolerance for casualties at times of demonstrable success and they offer some convincing examples from wars past and present.

6Building on the literature, the authors suggest that while the mission type, elite consensus and multilateral support increase casualty tolerance, none of these has so great an impact as expectations of success (122-123). What the authors never quite pin down is the degree to which these factors feed into perceptions of success, something that they treat as an independent rather than the sum total of several factors.

7The strength of the work lies in the authors’ ability to carry out surveys during a major war, from September 2003 to November 2004. Most studies in the area of war and public opinion rely exclusively on posing hypothetical questions to survey respondents whereas Gelphi et al. conducted their surveys in response to real world events. There is also no doubt about the quality of the statistical analysis to which the authors submit their data, although the statistics often make for some impenetrable reading. They also examine their data at the individual level, in contrast to most studies in this area that examine aggregate data. As such they are able unpack some of the causes of support and offer some insights on the demographics and individual motivations of casualty tolerance.

8Nonetheless, the work is found wanting in several critical areas. Firstly, ‘expectations of success’ is an abstract concept at best. The authors offer a section on what success in Iraq means to the US public and they highlight that the public is responsive to the way in which the administration frames victory. One might suggest that most Americans rely on the administration for definitions of success, however accurate those definitions may be. Gelphi et al. are convinced that presidential rhetoric can shape public perceptions of success but their own work shows that this only causes short term spikes after which support declines once it becomes obvious that success on the ground is not sustained.

9The primary weakness of the material on the Iraq war, the bulk of the study, is that the surveys were carried out between September 2003 and November 2004 prior to the most deadly phase of the insurgency. No reason is offered for this even though the book did not go to press until the fall of 2008. Conveniently enough for the authors, the problems raised by conducting surveys during only the first year of the war can be ignored because not once is the length of a conflict considered as a factor which affects the robustness of support.

10The authors also undermine their own position by highlighting, after 200 pages in which they have repeatedly argued that the public has a reasonable and reasoned grasp on issues of foreign policy, that only 48 percent of respondents to one of their 2004 surveys could identify Donald Rumsfeld as Secretary of Defense. Other figures are alarming, such as the 16 percent of respondents to one survey who suggested that they would tolerate up to 50,000 US deaths in Iraq. Such figures make a mockery of the ‘pretty prudent public’ argument.

11The authors ignore the huge changes that have occurred in the way that America wages war in the past 20 years. It is difficult to quantify whether the US public as a whole has made sense of the shift from ‘revolution in military affairs’-style conflicts like Kosovo in 1999 to counterinsurgency operations in Iraq and Afghanistan. What has been the psychological impact of the switch from the ‘virtual wars’ of the Clinton years to the population-centric ‘COIN’ approach of the ‘War on Terror’ and the attendant increase in casualties? It would have been interesting if the authors had examined this.

12Perhaps most depressingly of all is the absence of any consideration of civilian casualties. The authors suggest that this is an important area for future research but one would have thought that the horrific internecine violence unleashed in Iraq by the US invasion would have had so considerable an impact on support for the war that it would have to be included in any survey.

Simon Toner, University College Dublin

Top of page


1 See for example Feaver’s contributions to Foreign Policy’s ‘Shadow Government’ blog,
2 ‘Bush's Speech on Iraq War Echoes Voice of an Analyst’, New York Times, December 4, 2005.
3 John Mueller, ‘The Iraq Syndrome’, Foreign Affairs, November/December 2005
Top of page


Electronic reference

« S. Toner on Gelphi, Feaver and Reifler’s Paying the Human Costs of War », European journal of American studies [Online], Reviews 2009, document 14, Online since 20 September 2009, connection on 27 March 2017. URL :

Top of page


Creative Commons License

Creative Commons License

This text is under a Creative Commons license : Attribution-Noncommercial 2.5 Generic

Top of page