Sarah Kreps has made a superb contribution to the burgeoning academic literature on the causes of military intervention. This literature reflects the enormity of the task social scientists face in comprehending world affairs.
This enormity stems from the many choices social scientists have to make in order to investigate reality. Unlike physicists, they do not start with a single universe inhabited by objects that obey physical laws and have no independent ideas or social institutions of their own. They confront a universe in which objects have minds of their own, behave interactively through various social networks and institutions, and may or may not act in line with material forces. On top of all that, social scientists inhabit the universe they examine. They are studying themselves. They belong to specific countries and favor certain ideas and institutions. Should they focus on the United States or on other countries? Should they privilege the causal power of ideas, power or institutions? Should they look at the world from an individual, country, or systemic level of analysis?
H-Diplo | ISSF Roundtable, Volume IV, No. 7 (2012)
Sarah E. Kreps. Coalitions of Convenience: United States Military Interventions After the Cold War. New York and Oxford: Oxford University Press, 2011. ISBN: 9780199753802 (hardback, $27.95).
H-Diplo/ISSF Editors: Thomas Maddux and Diane Labrosse
H-Diplo/ISSF Web and Production Editor: George Fujii
Commissioned for H-Diplo by Thomas Maddux
Published by H-Diplo/ISSF on 16 November 2012
PDF- https://issforum.org/ISSF/PDF/ISSF-Roundtable-4-7.pdf
Contents
-
Introduction by Henry R. Nau, Elliott School of International Affairs, The George Washington University 2
-
Review by Stephen G. Brooks, Dartmouth College. 7
-
Review by Patricia L. Sullivan, University of North Carolina at Chapel Hill 13
-
Review by Alexander Thompson, Ohio State University. 19
-
Review by Dominic Tierney, Swarthmore College. 23
-
Author’s Response by Sarah E. Kreps, Cornell University. 28
Introduction by Henry R. Nau, Elliott School of International Affairs, The George Washington University
Kreps’s Study Illustrates Choices and Consequences of Social Science Research
Sarah Kreps has made a superb contribution to the burgeoning academic literature on the causes of military intervention. This literature reflects the enormity of the task social scientists face in comprehending world affairs.
This enormity stems from the many choices social scientists have to make in order to investigate reality. Unlike physicists, they do not start with a single universe inhabited by objects that obey physical laws and have no independent ideas or social institutions of their own. They confront a universe in which objects have minds of their own, behave interactively through various social networks and institutions, and may or may not act in line with material forces. On top of all that, social scientists inhabit the universe they examine. They are studying themselves. They belong to specific countries and favor certain ideas and institutions. Should they focus on the United States or on other countries? Should they privilege the causal power of ideas, power or institutions? Should they look at the world from an individual, country, or systemic level of analysis?
Kreps’ study illustrates these choices. It focuses on U.S. rather than Chinese or Russian policy toward military intervention. It is interested in the form of intervention, unilateral vs. multilateral, not whether the intervention is undertaken in the first place or succeeds. It privileges an instrumental or rationalist approach that emphasizes material causes – threat and operational costs – of the form of intervention, not a constructivist approach that emphasizes ideas[1] and norms[2], or an institutional approach that emphasizes rules.[3] And it adopts a decision-making level of analysis, not a systemic[4], interactive[5], or comparative domestic one.[6] Finally, it selects only four cases from many more that exist and controls for other possible explanations by holding variables constant or rejecting alternative hypotheses.
Thus Kreps holds structural variables constant by examining interventions only after the end of the Cold War. Throughout this period the United States dominated, and therefore structural variables, which were constant, cannot explain variations in the form of intervention. By focusing on the United States, she also holds other variables constant such as democracy and culture, dealing only with domestic politics such as Congress and public opinion. And, she considers additional hypotheses such as multilateral norms, regional power dynamics, and briefly individual leaders’ preferences, learning theory, and the willingness of other powers to contribute to multilateralism.
Kreps makes all of these choices knowingly and explicitly. The study, as one reviewer suggests, is a model of social science analysis. But her choices have consequences, and these consequences account for most of the criticisms of her work. From the reviews that follow, let me give a few examples.
Stephen Brooks raises a first dilemma, the study’s definition of multilateralism. Kreps agrees “with normative arguments that states would rather have more legitimacy than less when they use force abroad” (5). But then she defines legitimacy exclusively in terms of multilateralism: “policy coordination among multiple actors that have influence over decision-making.” More participants with more influence confer more legitimacy. The United Nations becomes the gold standard. This definition follows from her rationalist rather than constructivist perspective; legitimacy is procedural, not substantive. Multilateral cooperation equates with legitimacy whatever the individual or collective ideology of the countries participating.
But what if legitimacy is a function of ideology – say democracy –rather than participation and influence? State leaders have to be accountable in free and fair elections before they can claim that their interventions are legitimate. If NATO authorization becomes more legitimate than U.N. authorization, then the Bosnian, Kosovo and Afghanistan interventions rank higher than the Gulf War, Haiti and Somalia. Moreover, maybe multilateralism itself is valued because it is a product of democracy, not because it is merely instrumental, as Kreps argues. Recall that Kreps holds structural factors constant. But one of these structural factors, which she does not highlight, is the dominance of democracy in the post-Cold War period. Not only American power but also American values suffused this period. So maybe multilateralism was the rule because it was ‘appropriate,’ which is a constructivist argument. Imagine if the Soviet Union had won the Cold War rather than the United States. Is it likely then that states would have associated legitimacy with multilateralism? It is doubtful. Legitimacy might have had a different substantive content, that of Marxist-Leninist orthodoxy rather than democratic pluralism.
Patricia Sullivan raises a second criticism of Kreps’ choices. There are no examples in the study of participatory multilateralism – interventions with no formal authorization of an international organization but high levels of participating countries. Why? Kreps attributes this to the fact that after the Cold War there was no longer a Soviet veto and the United States could more easily obtain authorization from international organizations. But that draws on a constant structural variable, which she has excluded. And it doesn’t apply to NATO authorization, which should have become more difficult to obtain because the Soviet Union was no longer the looming enemy inducing alliance cooperation. In addition, as Sullivan points out, two other structural variables might explain the absence of participatory multilateralism just as well – different threat perceptions and capabilities of other countries that do not align with those of the United States. Kreps’ decision-making level of analysis omits such interactive (relative perceptions) and structural (relative capabilities) level variables.
Dominic Tierney identifies still a third dilemma. Not only are variables omitted in social science analysis; they also interact and sometimes cause one another. Tierney questions whether the two variables Kreps highlights are really independent of one another. Kreps acknowledges that the directness of threat or “time horizon tends to dominate” resource requirements (34). As Tierney points out, if the threat is great but the costs to deal with it are low, why is it a threat in the first place? Or, conversely, if the operational commitments of intervening are enormous, as proved to be the case in Iraq and Afghanistan, decision-makers may persuade themselves that the threat is less direct. The United States is withdrawing militarily from both Iraq and Afghanistan and convincing itself, at the same time, that persisting sectarian conflict in Iraq and Taliban participation in a future Afghan government will not recreate the same threats that occasioned the interventions in the first place.
Alexander Thompson raises the most fundamental question for social scientists. How do they measure things like perceptions of threats and costs? Since scholars are studying events about which they have preferences, how do they code various behaviors? Do they take the word of decision-makers at face value? Leaders who intend to intervene may hype threat – President Harry Truman in 1947-48 when he made decisions to intervene militarily in western Europe, or Bill Clinton in 1998 when he bombed Iraq, or George W. Bush in 2003 when he invaded Iraq. And if leaders intend to intervene, they may also underestimate costs – to wit, Iraq and Afghanistan. If the analyst substitutes his or her own judgment for these assessments, are these judgments necessarily objective?
Thus, it may not be self-evident when threats are high and costs low, or vice versa. Thompson notes that the task in the Persian Gulf War in 1991 was more modest – to expel Iraq from Kuwait without changing regimes – than the task in the Iraq War in 2003 – to overthrow a government and change regimes. Yet Kreps codes the former as high operational costs and the latter as low operational costs. Here another theoretical framework, which Kreps considers only briefly, may be at work – a learning model of intervention: the low cost of the successful operation in Afghanistan in 2001 dictated the same approach to the more ambitious operation in Iraq in 2003, with the less successful consequences that followed.
Social science scholars do the best they can, given the cards that reality deals them. They have to make multiple choices that narrow the value of the conclusions they reach. It would help if every study started with a careful abstract of these choices and consequences. Then scholars might claim less for their results, and other scholars might be encouraged to duplicate results. Unlike physicists, social scientists seldom bother to replicate studies. Kreps’ study, with its careful choices and modest claims, would be a great place to start.
Participants:
Sarah E. Kreps is an Assistant Professor of Government at Cornell University. She has undergraduate and graduate degrees from Harvard and Georgetown respectively and has held fellowships at the Belfer Center for Science and International Affairs (Harvard University) and the Miller Center (University of Virginia). She is the author of Coalitions of Convenience: United States Interventions after the Cold War (Oxford University Press, 2011) as well as numerous articles on topics of international security, which she has published in such journals as the Journal of Conflict Resolution, Security Studies, Foreign Policy Analysis, Political Science Quarterly, and the Journal of Strategic Studies.
Henry R. Nau is Professor of Political Science and International Affairs at the Elliott School of International Affairs, The George Washington University. He holds Ph.D. and M. A. degrees from The Johns Hopkins University School of Advanced International Studies and a B.S. degree from the Massachusetts Institute of Technology. His published books and essays include, among others, Worldviews of Aspiring Powers: Domestic Foreign Policy debates in China, India, Iran, Japan and Russia, edited with Deepa M. Ollapally (Oxford University Press, 2012); Perspectives on International Relations: Power, Institutions, and Ideas (CQ Press, 3rd ed., 2011); At Home Abroad: Identity and Power in American Foreign Policy (Cornell University Press, 2002); The Myth of America’s Decline: Leading the World Economy into the 1990s (Oxford University Press, 1990); “U.S. Grand Strategies,” in David Coates, ed. Oxford Companion to American Politics (Oxford University Press, 2012); “The Jigsaw Puzzle and the Chess Board: The Making and Unmaking of Foreign Policy in the Age of Obama.” Commentary (May 2012); and “Ideas have consequences: The Cold War and today,” International Politics, (July 2011). He has just completed a book manuscript on “Conservative Internationalism: Integrating Force and Diplomacy under Jefferson, Polk, Truman and Reagan.”
Stephen G. Brooks is an Associate Professor of Government at Dartmouth, and has previously held fellowships at Harvard and Princeton. He is the author of Producing Security: Multinational Corporations, Globalization, and the Changing Calculus of Conflict (Princeton University Press, 2005) and World out of Balance: International Relations and the Challenge of American Primacy (Princeton University Press, 2008). He has published articles in journals such as International Security, International Organization, Journal of Conflict Resolution, Perspectives on Politics, Security Studies, and Foreign Affairs. He received his Ph.D. in Political Science with Distinction from Yale University in 2001, where his dissertation received the American Political Science Association’s Helen Dwight Reid Award for the best doctoral dissertation in international relations, law, and politics.
Patricia L. Sullivan is an Assistant Professor in the Department of Public Policy and Curriculum in Peace, War, and Defense at the University of North Carolina-Chapel Hill. She completed her Ph.D. in political science at the University of California, Davis. She is the author of a recently released book on the utility and limitations of military force as a policy instrument, Who Wins? Predicting Strategic Success and Failure in Armed Conflict (Oxford University Press, 2012). She has published research on international conflict, military aid, foreign military intervention, and public support for military operations in journals including the Journal of Conflict Resolution, the Journal of Politics, the Journal of Peace Research, and Foreign Policy Analysis.
Alexander Thompson received his Ph.D. from the University of Chicago and is an Associate Professor of Political Science and a Faculty Associate of the Mershon Center at Ohio State University. His research and teaching are in the area of international relations, with a particular focus on the politics of international organizations and the design of international agreements. He is the author of Channels of Power: The UN Security Council and U.S. Statecraft in Iraq (Cornell University Press, 2009) and numerous journal articles. His current projects address the dynamics of treaty ratification and renegotiation, the evolution of the global climate change regime, the choice between bilateralism and multilateralism in institutional design, and the role of secrecy in international organizations.
Dominic Tierney is Associate Professor of political science at Swarthmore College. He received his Ph.D. in international politics from Oxford University, and is a senior fellow at the Foreign Policy Research Institute, and a correspondent at theAtlantic.com. He is the author of Failing to Win: Perceptions of Victory and Defeat in International Politics (Harvard University Press, 2006), with Dominic Johnson, FDR and the Spanish Civil War: Neutrality and Commitment in the Struggle that Divided America (Duke University Press, 2007), and How We Fight: Crusades, Quagmires, and the American Way of War (Little, Brown and Co., 2010).
Review by Stephen G. Brooks, Dartmouth College
This is an excellent book. The core argument that Kreps advances is that the U.S. is an instrumental legitimacy seeker; this is what causes it to prefer, but not have an ironclad preference for, multilateralism in its military interventions. This is an important argument that various scholars have pointed toward but have not adequately developed or defended. Moreover, Kreps provides a compelling account of when the U.S. is likely to deviate from its preference for multilateralism: when it has short time horizons and/or when the intervention is expected to be resource-intensive.
As with any book, there are elements of this book that I find less compelling. Yet rather than being critical for its own sake, I will instead be forward thinking; below I delineate four important questions that emerge from Kreps’ analysis that I think need to be further addressed in future research.
One issue I wondered about in light of Kreps’ analysis concerns the extent to which the U.S. will feel pressure to seek endorsements from the UN in future years. Kreps maintains that the changing distribution of power is likely to mean that the U.S. will undertake more and more ‘forum shopping’ in future years (11). She notes further that “the United States is generally willing to operate without UN authorization but will forum shop for alternative IOs [international organizations] such as NATO or resort to an ad hoc coalition, with each forum conferring less and less legitimacy, but also less and less constraint” (157). Here, the question that springs to mind is how much less legitimacy NATO (or some other coalition or institution) has as compared to the UN, and with which actors? Do all states, and all populations in those states, highly value the UN, or just some? In an ongoing coauthored research project on U.S. public attitudes I am working on, our initial findings suggest that there is not, in fact, a legitimacy premium attached to a UN endorsement of a military action.[7] If NATO (or some other coalition or institution) has either the same or only slightly less legitimacy than the UN among the states/populations that the U.S. cares most about, then it would seem that the U.S. will be less prone to seek UN endorsements in the future given that it has clearly moved beyond the “golden era” of relatively easy to obtain UN cooperation that existed in the immediate aftermath of the Cold War’s end. If, on the other hand, there is an extremely large legitimacy premium attached to the UN endorsement among the states/populations that the U.S. cares most about, then we would expect the U.S. to often seek UN approval even if it is very cumbersome to work through and/or the chances of receiving an endorsement are not especially high.
In turn, Kreps notes that other countries such as France and Canada have much to contribute militarily to U.S.-led operations and may be prone to tie their military contributions to receiving an endorsement from the UN; if so, she posits that this will create heavy pressure on the U.S. to work through these institutions (158-159). Yet if many potential military contributors of this kind are heavily cutting their already-constrained defense budgets (as they now seem to be doing) and thereby have even less that they can militarily provide to U.S.-led operations in the future – and, moreover, if they are progressively falling further behind the U.S. in military technology, thus increasing the inter-operability problem – then will the UN have the same pull for the United States even if these states link their contributions to a UN endorsement?
The second general issue I wish to raise concerns financial contributions from allies. Kreps notes that states that offer financial resources for an operation (such as Japan’s contribution to the 1991 Gulf War) do so “in exchange for a say over decisions” (18). This may be true sometimes, but I find it doubtful that it is always, or even predominantly, the case. Many of the countries that have made, or may someday make, financial contributions to U.S.-led military operations are countries that depend on it for their security. Such states may well be compelled by the United States to make financial contributions to help fund conflict operations or may act without U.S. direct pressure out of a desire to stay on the good side of their security patron. Either way, the financial contribution may be made without the state in question being able to and/or expecting to gain any real say over the nature of military operations. For example, did Japan really have any substantial control over how the Gulf war was conducted?
Why is this issue potentially important? In the view of Kreps, there is no free lunch when it comes to financial contributions – they are a quid pro quo, in which money is exchanged for some control over decisions about the conduct of the conflict (18). If that is the case, then the U.S. will always need to follow a cost/benefit calculation as to whether to accept such financial contributions and there will be no ‘cost free’ choice: in order for the operation to be less expensive, the U.S. will have to bear various costs that make the operation less efficient than if it was planning it by itself. Alternatively, if it is sometimes, or often, the case that the U.S. can garner financial contributions without having to give up any real autonomy over how the war is fought, then this would be an ideal situation for the U.S. In light of the fact that many U.S. allies effectively cannot help on the battlefield (i.e. Japan) or, as noted above, are likely to have an ever lower ability to contribute on the battlefield in the future due to defense cuts (i.e. many U.S. allies in Western Europe), the prospect of the U.S. receiving financial contributions may well be the main way that many states can, in fact, contribute to these military operations in future years. And if such financial contributions are a free lunch for the United States, then it may well regard them as a worthy substitute for troop contributions. In terms of making a prediction about how often the U.S. will seek such contributions – and in terms of reaching an assessment about how valuable it would be for the U.S. to actually receive them – we really do need to know much more about how often they have a free lunch quality. Kreps has usefully put this issue on the table, but more research on this question needs to be undertaken.
The third issue I wish to discuss concerns the determinants of time horizons. Kreps maintains that “uncertainty about the future will create short time horizons and temptations to scoop the short-term gains of unilateralism” (7). She notes further that “the challenge is in explaining a priori the conditions under which a state will feel more immediate security threats and experience shorter time horizons” (29). Her answer is that it “depends on the security environment. A relatively safe, predictable security environment” causes the time horizon to be long, whereas “an environment with more security hazards creates shorter time horizons” (30). It may well be that time horizons shift in response to objective changes in the security environment; if so, the job of analysts who wish to understand and make predictions about time horizons would certainly be easier. But my sense is that it is likely more complicated than this.
Different policy makers may well understand the same security environment in different ways, and thus have different time horizons, because they perceive it differently. One reason perceptions may differ is that different individuals may have different underlying theories about how the world works – and/or different underlying psychological foundations – which cause them to see the same security environment in different ways. In an earlier article, I noted that some realist IR scholars – such as John Mearsheimer and Kenneth Waltz – always view the world in extremely pessimistic terms and presume that statesmen think the same way.[8] I noted that Waltz and Mearsheimer both see states as being conditioned by the mere possibility of conflict; in this view, policy makers always adopt a worst-case perspective. In contrast, the vast majority of other realist scholars – such as Charles Glaser and Stephen Walt – see policy makers as being conditioned by the probability of conflict. I noted further that these divergent assumptions regarding whether states are conditioned by the probability of conflict as compared to the possibility of conflict lead directly to an additional set of assumptions regarding the discount rate: viewing the world in worst case/possibilistic terms reflects the view that actors heavily discount the future, while viewing the world in probabilistic terms reflects the view that actors will often make intertemporal trade-offs. It is certainly possible that some statesmen may well best be captured by the worst case/possibilistic conception (and such individuals will heavily discount the future) while others may be may well best be captured by the probablistic conception (and such individuals will not heavily discount the future). The larger issue I am pointing to in this paragraph is that the discount rate may vary from one policy maker to another in ways that cannot be explained simply by looking at the objective nature of the security environment. More research is needed on this general topic.
The fourth and final issue I want to raise concerns how we define unilateralism. At the beginning of her book, Kreps identifies several analysts who forward a “structural” perspective which posits that “unilateralism will be tempting to the most powerful states” (4). My understanding of this perspective is that it does not see unipolarity as directly leading to unilateralism; rather, it sees unipolarity as reducing some (but not all) of the costs of going it alone in security affairs and thereby making it a more tractable option than it would otherwise be. Kreps notes that if unipolarity makes unilateralism tempting, it is surprising that we do not see more of it than we actually do; she argues that very few cases of unilateralism actually exist. Although she does a much better job than all other analysts to date of operationalizing unilateralism and multilateralism – her book draws upon her earlier exemplary 2008 article on this topic[9] — I wonder if she has set too forgiving a bar for coding the U.S. as engaging multilateralism by having such a narrow definition of unilateralism. Specifically, she defines unilateralism as an operation that (a) is not authorized by any multilateral organization and (b) is conducted by just one state. This definition has many virtues (many analysts may well find it is the best one we could possibly use) and Kreps does an amazing job of operationalizing it. But there are certainly some ways that the U.S. could be seen as acting in a unilateral manner that are not considered within her definition of the concept. In this regard, there are at least three other potential definitions of unilateralism that should also conceivably be factored in – none of which, as will be seen, are as easy to operationalize as the one Kreps employs (here I draw upon the discussion in Brooks 2012, which was motivated in part as a reaction to Kreps’ 2008 article).[10]
First, some scholars define unilateralism as any action that does not receive the backing of the United Nations.[11] Kreps argues persuasively that it sets too high a standard to say that any action should be considered unilateral if it fails to receive UN authorization; if this criterion were adopted, then the 1999 Kosovo War would be considered unilateral even though this military action was approved by NATO and was prosecuted collectively by its members. In contrast, Kreps maintains that a military action should be considered procedurally multilateral if any international institution votes to approve it; it is unilateral only if it completely fails to receive any institutional backing. While it does seem too restrictive to say that an action is unilateral unless it receives UN approval, it is perhaps too forgiving to say that any regional institution whatsoever can bless an action for it to be considered multilateral. Not all regional institutions are created equal: it is much harder and more significant to gain institutional approval from NATO – an organization with 28 members with a combined population of 882 million people – than it is to gain approval from, say, the Organization of Eastern Caribbean States, a seven member organization of current or past British dependencies whose combined population is 550,000 people – that is, slightly smaller than Oklahoma City. The point is simply that many scholars – and perhaps also many policy makers and much of the public – may well regard actions that are approved by most kinds of regional institutions as being unilateral, whereas Kreps would code them as procedurally multilateral.
Acting without consulting other states is a second definition of unilateralism that should also potentially be factored in.[12] In this regard, the Bush administration was generally viewed as acting unilaterally during its first term in significant part because it was seen as not consulting very much with its allies. Kreps recognizes the potential significance of this dynamic, but does not include it within her conception of unilateralism: she notes that a key difference between the George H.W. Bush and Bill Clinton administrations and that of George W. Bush “was not goals, but statecraft: the ability to listen, appreciate other arguments, and ultimately persuade states” (167). It should be noted that the literature does not specify how much consultation a state must undertake to avoid being regarded as pursuing a unilateral foreign policy. The literature also does not specify what specific form consultation must take for an action to avoid being regarded as unilateral. Does it suffice to merely give other states a ‘heads up’ that an action will be taken? Or do other states have to be given the opportunity to provide feedback? If they do need to give feedback, then is it sufficient if other states provide information about where they stand without actually being in a position to have any actual influence on the hegemon’s decision? Or must feedback be sought with a simultaneous willingness on the part to shift positions in response? Manfred Bertele and Holger Mey argue in favor of the highest standard of consultation: for an action to be multilateral, “consultation should occur after interests have been defined (an act that inevitably remains unilateral) but before a strategy has been conceptualized—before a decision to act has been made.”[13]
A third definition of unilateralism that should also potentially be factored in is unwillingness by a state to change its preferred policy course in response to the actions or interests of others. In this respect, Charles Krauthammer argues:
“Unilateralism does not mean seeking to act alone. One acts in concert with others if possible. Unilateralism simply means that one does not allow oneself to be hostage to others. No unilateralist would, say, reject Security Council support for an attack on Iraq. The nontrivial question that separates unilateralism from multilateralism–and that tests the “pragmatic realists”–is this: What do you do if, at the end of the day, the Security Council refuses to back you? Do you allow yourself to be dictated to on issues of vital national–and international–security?” [14]
Using this rubric, unilateralism is obviously easy to identify in those situations when a state acts by itself against the expressed wishes of all other countries. Conversely, an action is clearly not unilateral if all countries agree that it should be undertaken. Yet it is almost unheard of in international politics that either all or no states agree with a given action, and the literature is unclear regarding the fact that much agreement must exist before an action should no longer considered unilateral.
What is the bottom line of this discussion of these alternative definitions of unilateralism? If one or more of these additional definitions were factored into our understanding of the concept, then there would be more instances in which we would code the U.S. as engaging in unilateral behavior than if we use the narrower definition that Kreps forwards. Recognizing these alternative conceptions of unilateralism does potentially help to square the circle between Kreps’ contention that the U.S. almost never engages in unilateral behavior with the view held by many analysts – as well as many foreign publics and foreign leaders – that the U.S. does frequently engage in unilateralism. In the end, Kreps has done a great service with her conceptual work on the definitions of unilateralism (and multilateralism) and in showing how they can be operationalized. But I think that more work is nevertheless needed in this area.
In closing, let me note that the most useful books do not simply raise some specific questions that they provide answers to, but also illuminate additional issues that require future investigation. That is certainly the case with this book. It successfully defends its core argument that the U.S. is an instrumental legitimacy seeker, and in so doing it raises a series of additional interesting research questions – four of which I have highlighted here – that scholars should now further investigate.
Review by Patricia L. Sullivan, University of North Carolina at Chapel Hill
In Coalitions of Convenience, Sarah Kreps builds an elegant, intuitive argument about the conditions under which powerful states use force multilaterally and the factors that create incentives to go it alone. The book is a model of conscientious scholarship. The author constructs her theory thoroughly and deliberately and generates concrete, falsifiable hypotheses. She pays a great deal of attention to operationalizing concepts, which allows for reasonably objective measures of her explanatory and dependent variables. And she thoroughly tests her argument against leading alternative explanations in a series of four case studies that draw on a broad range of sources—including interviews and government documents. Kreps also pays careful attention to case selection and makes a transparent case for the interventions she has chosen to evaluate. The book is well-written and the case studies in particular are engaging and convincingly argued.
Coalitions of Convenience argues that even the most powerful states often want to use military force in concert with other states because doing so (1) increases the perceived legitimacy of military intervention, and (2) getting help from allies allows the state to conserve resources. Attaining the authorization of an international organization (IO) prior to using force and incorporating many diverse states in carrying out an intervention increases the perceived legitimacy of using force because these things “are viewed as an independent, neutral assessment of the lead state’s ambitions” (149). Powerful states want to increase the perceived legitimacy of their military interventions because less legitimate uses of force risk provoking balancing behavior or sanctions from other states. Acting unilaterally can prompt other powerful states to take countermeasures that make an intervention more costly or less likely to succeed. Acting within established norms and institutions, on the other hand, preserves good will, strengthens an international order that favors status quo powers like the United States, and allows a state to conserve limited resources.
Given these incentives for multilateral action, why doesn’t the United States always use force multilaterally; why go it alone? Kreps maintains that while intervening with a coalition has benefits, it also limits freedom of action. The process of securing IO authorization and the participation of allies takes time and makes it difficult to act quickly. Coordinating with allies, all with their own divergent interests, is slow and often requires compromise on goals, timing, strategies, side payments, and command structure.
So when will the U.S. choose multilateral action over the unilateral use of force? Coalitions of Convenience argues that which path a powerful state chooses is a function of two factors: 1) the state’s time horizon (how immediate and direct the security threat is) and 2) the operational commitment (how resource-intensive the mission is expected to be). The more immediate and direct the threat, the shorter a state’s time horizons. When a state faces an immediate threat to its homeland, short-term security concerns dominate and a state is likely to forgo multilateral channels in order to act swiftly and preserve maximum control over operations. When time horizons are longer, lead states have an incentive to pursue formal authorization from an international organization before taking action. Even when time horizons are long, however, Kreps argues that strong states may not seek the full participation of other states in their military operations if the state believes it can achieve its objectives quickly and cheaply. Because intervening with other states constrains the lead intervener’s freedom of action and can compromise its efficiency, when the United States believes a mission can be accomplished with limited resources it will seek formal authorization but restrict the actual involvement of foreign forces to keep coordination costs to a minimum. Only when intervening in a region with powerful rival states that could act to counter U.S. interests (what Kreps calls a “contested zone”) (8), or against a target with a large military, will the U.S. seek both IO authorization and the full participation of other states.
Kreps is careful to define multilateralism in a rigorous way. She considers two separate dimensions of multilateralism: (1) institutional (authorization from international organizations) and (2) participatory (the material participation and decision-making influence of other states). The participatory dimension is more fully developed with a consideration of how many states contributed to the coalition, the proportion of troops and resources committed by those states, whether regional allies were involved, and the relative material power of the participants. I am sure that other scholars will adopt this useful construct for thinking about multilateral military action.
Kreps tests her hypotheses against the predictions of four alternative explanations through a careful comparison of U.S. interventions in the first Gulf War (1990-1991), Haiti (1994-1996), Afghanistan (2001- ongoing), and Iraq (2003-2011). Each of the first three campaigns is disaggregated into two phases. Figure 2.1 on page 20 and Table 2.1 on page 22 categorize each of these interventions, as well as six additional post-Cold War U.S. military interventions, along the institutional and participatory dimensions to produce four types of cooperation strategy: unilateralism/bilateralism/minilateralism, formal multilateralism, participatory multilateralism, and full multilateralism.
According to the author, Gulf War I began as bilateral intervention in which the United States deployed thousands of troops to the region before attaining UN authorization and conducted its initial deterrence operations with only Saudi support. After UN Resolution 678 authorized the use of force to evict Iraq from Kuwait on 29 November 1990, the intervention became a case of full multilateralism with IO authorization and a large coalition of states with tangible influence over decision-making. The intervention in Haiti in 1994 is a case in which a long time horizon and relatively low operational commitment would have predicted formal multilateralism (IO authorization without much real contribution from other states). However, after the initial invasion was conducted almost entirely by American troops, the United States pulled together a robust multilateral coalition for post-conflict stability operations because it was concerned about the cost of an open-ended commitment to ensuring stability in the country. In Afghanistan, there were many countries willing to help, but the U.S. initially intervened to overthrow the Taliban regime in coordination with a non-state rebel movement, the Northern Alliance, with only minimal support from Central Asian republics in the form of access and basing rights. After the regime was ousted, the U.S. welcomed the assistance of NATO allies for reconstruction efforts in areas of the country that had been stabilized. Finally, Kreps considers the 2003 Iraq War, which she characterizes as “one of the most unilateral interventions in the post-Cold War period” (48), an outcome she attributes to U.S. leaders’ (mistaken) beliefs that time horizons were short and the operational commitment would be low.
As always, reality is more complicated than our theories would suggest. The question is not whether this theory captures all of the factors that determine whether a powerful state will seek a coalition or act unilaterally, but how useful this theory is in explaining variation in intervention behavior. Could another theory provide a better explanation and superior predictive power for an equal or broader range of cases? Overall, I think Kreps does a commendable job of creating a useful framework for understanding the United States’ motivation for seeking IO authorization and the participation of other states in its military interventions. In particular, I am convinced that leaders’ beliefs about the severity and immediacy of a threat have a strong influence on decisions about whether to seek authorization from international organizations and the participation of other states before using force abroad. The book makes a strong case that long time horizons and the desire to share the costs of open-ended security and reconstruction missions provide a better explanation for U.S. multilateralism than post-Cold War norms, the preferences of individual leaders, regional power dynamics, or domestic politics.
I am puzzled, however, by a significant contradiction between the predictions made by the theoretical model and U.S. behavior in the decades since the end of the Cold War; while Kreps’ model predicts a “participatory multilateralism” outcome when time horizons are short and the operational commitment is expected to be high (35-36), this outcome is never observed (20 Fig 2.1). Short time horizons and high operational commitments would seem to be logically connected. Direct, immediate threats to security should result in short time horizons and be more likely than less direct threats to require robust, resource-intensive military responses. It is more difficult to see how a longer-term political, humanitarian, or economic threat would warrant a major military response. Moreover, by the criteria established in the book, there are a number of cases that could be classified as short time horizon and high operational commitment. Kreps describes the first Gulf War as a case in which “a surprise military attack on an ally created short time horizons” but in which it was “clear from the outset…that despite its power, the United States would be unable to carry out the intervention alone” (152). I would argue that both Afghanistan and the 2003 intervention in Iraq would also fit the short time horizon, high operational commitment characterization. U.S. decision-makers grossly underestimated the operational commitment in both of these cases, but they are both ambitious, regime-change operations against militarily formidable foes (at least relative to the Haitian armed forces and Somali warlords) in undeniably contested zones.
The book’s explanation for the absence of participatory multilateralism outcomes is that, since the end of the Cold War, UN authorization has been relatively easy to attain given U.S. power preponderance and the termination of the United States’ rivalry with the USSR. But this explanation is unsatisfying for two reasons. First, it is entirely outside the factors that Kreps argues explain variation in intervention behavior: time horizons and operational commitments. Second, and more importantly, the cases in which participatory multilateralism would be predicted by the theory—the first phases of Gulf War I and Afghanistan, as well as the 2003 invasion of Iraq—are coded as unilateral/bilateral/minilateral outcomes rather than full multilateralism as this explanation would predict.
I believe the absence of participatory multilateralism has important implications for Kreps’ theory and our understanding of multilateralism. In particular, it highlights the importance of three factors to which the book gives inadequate attention: the importance of the supply-side piece of a complete explanation for the variance in multilateralism we observe; the key role of the unique military capabilities and war-fighting preferences of the United States; and the limited explanatory power of operational commitment as defined.
The absence of cases in which the United States intervenes with the broad participation of allies but without IO authorization (i.e., participatory multilateralism) suggests that Kreps too easily dismisses the supply side of the equation. Although she is correct that the U.S. can unilaterally opt out of operating multilaterally, the U.S. cannot unilaterally decide to intervene with allies. All the cases of full and formal multilateralism in Figure 2.1 required the U.S. to pursue a multilateral strategy, but also required other states that were willing to authorize the use of force. In a statement that seems incongruous with the core of her theory, Kreps observes that the United States acts multilaterally if it “can produce UN authorization and collect allies easily” (10). In her discussions of the 1991 Gulf War and 1994 intervention in Haiti, she also notes that UN authorization was a precondition for the participation of a number of traditional U.S. allies. These remarks suggest that one reason we do not see participatory multilateralism is that sometimes the U.S. cannot gain the authorization of an international organization quickly or easily because other powerful states have serious reservations about U.S. intentions. In these cases, the U.S. is also unlikely to find many partners willing to make a major contribution to its military mission. When other states’ threat perceptions do not align with those of the United States, the U.S. can find itself without international support regardless of its preferences. A lack of international support for U.S. objectives vis-à-vis the Saddam Hussein regime seems to be at least a piece of a complete explanation for the unilateral U.S. intervention in Iraq in 2003.
The book’s discussion of the four case studies also partially obscures an important detail: High-intensity, offensive, ground combat operations are carried out essentially unilaterally by the United States. In contrast, all ‘post-combat’ reconstruction and stabilization operations are multilateral to the extent that the U.S. can get other countries to pitch in. One conclusion this suggests is that a major impediment to fully realizing the participatory dimension of multilateralism may actually be military officers who are reluctant to participate in truly joint operations or cede any decision-making control on the battlefield. Moreover, one of the primary reasons for their reticence has become much more severe since the first Gulf War. There is a significant and growing technology gap between the United States and potential coalition partners that creates substantial operational challenges. It is difficult for U.S. ground forces to operate jointly with other militaries, especially when, as is increasingly the case with even traditional U.S. allies like France, Australia, and the U.K., these militaries are less technologically advanced. Kreps discusses the “accommodation strategies” the U.S. military has developed to enable it to work with less capable allies (154). But despite statements by policymakers and military leaders about the value of operating multilaterally and institutionalizing coalition operations, the U.S. military’s accommodation strategies often boil down to relegating other states to operating at the periphery in order to keep them from interfering with the efficiency and effectiveness of U.S. combat operations. Increasingly, the U.S. essentially tells its partners to stay out of its way while American troops do the war-fighting and then invites them to in to help clean up. There are a number of obvious drawbacks to this arrangement. One is that as the U.S. develops an ever more technologically sophisticated fighting force, it locks itself into a position in which it is the only state with the capacity to carry out many missions and loses out on the resource-conserving benefits of burden sharing. Another is that the U.S. risks alienating its traditional allies as they become more and more reluctant to play the reconstruction role in the wake of American offensive operations.
One final possibility that the absence of participatory multilateralism suggests is that operational commitment plays a more nuanced role than Kreps’ theoretical argument implies or her hypotheses predict. In the theory chapters, Kreps discusses operational commitment in terms of real financial and ground troop commitments. She defines operational commitment as the “level of resources directed toward an intervention” (31) and the discussion on pages 33 and 34 suggests that what determines the extent to which the U.S. seeks the participation of other states is the level of resources the mission will require in some absolute sense. Operational commitment will be high when the U.S. intervenes in “contested zones” and countries with a large number of military-age men. In the intervention cases, however, variation in U.S. desire for multilateral participation does not track well with variation in the target country’s military strength, or whether the intervention is to take place in a contested zone. Operations commitment is a poor explanation for seeking contributions from allies in places like Somali and Haiti and eschewing it in Iraq in 2003. In Haiti the operational commitment was not high in any absolute sense, or relative to the operational commitment that could reasonably be anticipated before the U.S. intervened to overthrow regimes in Afghanistan or Iraq. Instead, the operational commitment of the stability phase was perceived to be high only relative to the low salience of the issues at stake for the U.S. Kreps argues that the level of resources that a military mission will require creates incentives to seek the participation of other states for burden-sharing. She states that there are some circumstances in which American military power is insufficient. But it would be more appropriate to say there are times when the human, material, and financial costs of a mission are greater than the costs the U.S. is willing to pay given the issues at stake. In actuality, what leaders anticipate the intervention will cost (which may be wildly off the mark) relative to what they perceive to be the salience of the interests at stake (which could also have little relation to the objective threat posed by a target) appears to determine how much incentive the U.S. has to trade-off control of the mission for assistance from other states. We don’t see the cases of participatory multilateralism hypothesis 1b predicts when time horizons are short and operational commitments are high because direct, immediate threats make the U.S. willing to bear much higher costs.
None of my critiques in any way diminish the value of this book. Sarah Kreps has produced a thought-provoking work on an important question that has not received much attention from political scientists. It represents a major advance in our understanding of multilateralism, providing strong evidence that the “inputs to cooperation strategy relate to the intervention itself…not to a particular distribution of power, political party, or individual” (150). Kreps’ typology of cooperation strategies and the clarity of her argument will undoubtedly provide the foundation for additional research in this area.
Review by Alexander Thompson, Ohio State University
The question of why states sometimes pursue military interventions unilaterally and other times opt for a multilateral approach is important and timely, with broad implications for policy and theory debates. Given its means, the United States, in particular, is always tempted with the option of going-it-alone, thereby avoiding the potentially costly constraints and delays that accompany coalition building and bids for international organization approval. And yet, especially since the end of the Cold War, the United States has often acted with substantial coalitions and channeled its policies through multilateral institutions. We saw this most recently with the intervention in Libya, which was deliberately conducted with the support of the UN and regional organizations. Unilateral interventions are now the exceptions—albeit dramatic ones when they occur.
In her clearly argued and engaging book, Sarah Kreps outlines a framework for understanding the choice between unilateralism and multilateralism and offers an empirical assessment in the context of U.S. post-Cold War interventions. The book is part diplomatic history and part foreign policy analysis, and it will appeal to those scholarly communities, but its main contribution and intended audience lie in the field of international relations (IR). For IR scholars, Kreps’ puzzle is critical. The long tradition of Realism in IR explains state behavior as a function of power and predicts that a dominant state—such as the United States in the wake of the Cold War—should prefer the flexibility and practical advantages that come with acting alone. In contrast, more recent work on how norms shape state decision-making predicts that multilateral approaches should dominate because they are more legitimate and socially acceptable. Without denying that both power and norms matter, Kreps makes a strong case that neither perspective can explain the substantial variation in U.S. policies or the motivation behind them. We require a more nuanced explanation.
For Kreps, states act out of “cunning and consequence” (5). The book thus falls in the tradition of rationalist work that attempts to explain the advantages of institutional commitments and multilateralism, while simultaneously recognizing the tradeoffs that make unilateralism appealing in some cases.[15] The basic argument is that states with short time horizons, a product of more immediate and serious security threats, choose unilateralism, while states that face less direct and long-term threats have the luxury of paying the up-front costs of forging a multilateral response in return for its substantial benefits. These benefits include burden sharing with coalition partners and the ability to signal benign leadership by “acting legitimately within the established international order rather than contravening it” (25). Other things being equal, even the most powerful interveners prefer to capture the political advantages of multilateralism; they only turn to unilateralism in cases of exceptional urgency or threat. This helps to explain why multilateralism seems to be the default and modal approach to intervention in contemporary international affairs.
A second layer of the argument addresses the form that multilateralism takes. Kreps conceptualizes multilateralism along two dimensions, institutional authorization and the degree to which other states participate, offering a useful contribution to sometimes-woolly discussions of multilateralism among academics and in the press. She proposes that the higher the “operational commitment”—the resources and dedication required for the intervention—the greater the incentive for a fully multilateral approach, one that combines authorization from an international organization with meaningful participation by coalition partners. In sum, the combination of short time horizons and low operational commitment is most conducive to unilateralism, while long time horizons and high operational commitment produce the most multilateral policies.
The empirical portion of the book covers four U.S. interventions: Iraq (1991), Haiti (1994), Afghanistan (2001) and Iraq (2003). The case studies are commendable in many ways. They are well organized, rely on various sources of evidence (mostly secondary but some primary), and effectively demonstrate that multilateralism is an explicit policy choice that takes on various forms across and within cases. Moreover, Kreps explicitly considers alternative explanations based on the possible influence of three factors on the intervening state: international norms, domestic politics, and power dynamics in the region of the intervention. While these alternative explanations generally fail to explain the observed outcomes, Kreps readily concedes when the evidence points in their direction. For example, she outlines the importance of regional dynamics in the Gulf War case, especially the need for the United States to use multilateralism in order to reassure and co-opt the Soviet, Saudi and Turkish governments (69-71). And in the aftermath of the 9/11 attacks, the U.S. public and Congress provided a conducive domestic political atmosphere for acting unilaterally in Iraq (138-44). In these cases, however, the alternatives point mostly in the same direction as Kreps’ own hypotheses. Indeed, they do not challenge her argument so much as they demonstrate that some of the outcomes are over-determined.
The careful attention to the dependent variable and to alternative arguments is laudable but has the effect of highlighting a weaker aspect of the book’s research design: Kreps could be more precise and thorough when it comes to outlining the observable implications of the main hypotheses. While Kreps has a nice methodological discussion of how to distinguish her argument from the alternatives (44-46) and devotes a substantial portion of each empirical chapter to an assessment along these lines, she offers less guidance on the type of evidence that is most probative with respect to her key independent variables, the urgency of the security threat and the expected operational commitment required by the intervention. One result is that the case studies leave room for debate when it comes to showing these factors at work.
While I do not disagree fundamentally with her empirical claims regarding the values of the independent variables, in places the evidence offered seems a bit superficial, especially when it relies heavily on claims made by the decision-makers themselves. For example, Kreps concludes that Iraq was viewed as a serious and urgent threat by the George W. Bush administration in the lead-up to the 2003 war, offering numerous examples of officials making claims along these lines and attributing their impatience to a post-9/11 mindset (115). She also discounts accusations that these officials intentionally exaggerated Iraq’s weapons of mass destruction programs and its links to terrorism (125). This raises important questions. Advocates of a war will naturally claim that the threat is serious and urgent in order to rally support. Such evidence must be treated with skepticism (genuine concern must be discerned from hyperbole) and should be complemented by an assessment of the objective threat posed by Iraq to the United States at the time. While I certainly agree with Kreps that the 9/11 attacks triggered a new sense of vulnerability in Washington and changed its perspective on Iraq, most independent analysts did not view Iraq as a grave threat and most governments saw little need for war and favored more time for UN inspections. Far from addressing an urgent threat to U.S. national security, the war was viewed even by some insiders as one of choice rather than necessity.[16]
The measurement of the operational commitment variable poses similar challenges, also illustrated by the Iraq War case. The push for a small force and the prediction of a swift and inexpensive war in Iraq turned out to be wildly off the mark. Kreps correctly notes that this is less relevant as evidence than what decision-makers believed going in (133), but of course there were many military planners and State Department officials who accurately anticipated the post-invasion requirements and argued that a large force and a long-term reconstruction plan would indeed be necessary. The same could be said of the Afghanistan case. Kreps argues that the choice to intervene without a coalition was driven by a focus on the relatively limited, short-term goals of toppling the Taliban and eliminating al Qaeda (103-4), but we should consider the possibility that the tail was wagging the dog: the insistence on conducting the invasion without a coalition and with a very light force made it difficult to pursue a more ambitious mission that would maximize post-invasion stability and the chances of Osama bin Laden’s capture.[17]
This concern is not limited to cases of unilateralism. In the Gulf War, Kreps argues that the George H.W. Bush administration sought coalition partners because their added capabilities were required: “The operational commitment would be enormous, and the motivation to share it with allies considerable” (59). An alternative interpretation of the case is that Bush Sr.’s foreign policy team decided for a variety of political reasons that they would operate multilaterally and then characterized the participation of so many coalition countries in terms of the needs of the mission. After all, the task at hand in 1990-1991 (reversing Saddam’s invasion of Kuwait) was more modest than that of the 2003 war (total defeat, regime change and occupation), and yet Kreps codes the latter case as one of low operational commitment and the former as one of high operational commitment. This is not necessarily contradictory insofar as the perceived operational needs (a subjective matter) are what matters most analytically for the book. But even Bush and his national security advisor Brent Scowcroft later complained that the coalition was useful politically but a burden on the battlefield.[18]
My goal is not to challenge Kreps on her conclusions or on the historical record but rather to raise questions and make a broader methodological point. With both the time horizon and operational commitment variable, the desired policy outcome is likely to determine the public rhetoric and even the mindset of decision-makers. There is certainly evidence in the 2003 Iraq case that members of the Bush administration, in order to justify a relatively unilateral intervention, shaped the intelligence and actively stifled more realistic assessments of the threat and mission requirements.[19] In these situations, it becomes difficult to objectively measure Kreps’ key explanatory variables and to disentangle them from the political motivations underlying the intervention policy. While there is no easy solution to this problem, the book might have offered more guidance on how to sift through the evidence to reliably establish the perceived time horizon and anticipated operational commitment and to link them causally to the outcome.
Coalitions of Convenience will have a lasting impact on how we think about multilateralism and U.S. intervention policies. The nuanced conceptualization of multilateralism is a useful contribution in itself and the arguments about time horizons and operational commitment will contribute to lively debates on how states navigate between unilateralism and multilateralism. The normative and policy implications of the book are clear: When the leaders of the United States or other powerful countries are tempted by the short-term advantages of unilateral intervention, they should think twice. Along these lines, Kreps recommends in her concluding chapter that governments should improve their capacity to make accurate assessments in order to avoid unilateral policies that are later regretted (63-4). The larger implication is that the benefits of multilateralism—especially, I would argue, of international organization approval—are broad and consequential. Not only does a cooperative approach improve the chances of success for ambitious interventions, it has the enduring benefit of allowing powerful states to pursue their interests without suffering the sort of backlash that will eventually erode their influence.[20] Multilateralism is prudent and effective, not merely legitimate.
Review by Dominic Tierney, Swarthmore College
At the heart of Coalitions of Convenience is a puzzle: why would a colossus like the United States use force multilaterally? And why does the nature of this multilateralism vary significantly across post-Cold War cases? After all, the United States sometimes acts with both UN blessing and a broad coalition, whereas at other times it uses force with UN support but only a handful of partners, and, in cases like Iraq, essentially goes it alone.
In many respects, the book is a model for Ph.D. students to follow in producing a tightly crafted argument. It’s well structured and sensitive to methodological problems. The case studies are based on considerable research including extensive interviews. The book is particularly strong in teasing out the contested concept of multilateralism. The writing is crisp and effective.
One of the key predictions of Kreps’ theory is that direct threats produce more unilateralism. Multilateralism offers benefits in terms of legitimacy and burden sharing, but it is also “more time-consuming, less reliable, and more limiting than operating alone” (6). Faced with a military threat to the homeland or a key ally, powerful states like the United States will move unilaterally to head off the danger, rather than take time to cobble together a coalition.
At first glance, this thesis runs counter to the dominant realist theory of alliances. Stephen Walt and other authors have argued that external threats produce multilateralism because states form counter-balancing alliances (realists do not put as much weight on the international-organization side of multilateralism).[21] External threats encourage states to overcome their differences (for example, of regime type) and align against the common danger. In the face of a rising Germany, Republican France and Tsarist Russia, for example, allied in the 1890s. Similarly, the threat posed by Germany, Japan, and the Soviet Union led the United States to abandon its long tradition of unilateralism and form the grand alliance in World War II, as well as the peacetime NATO alliance in the Cold War. In these cases, Washington did not find that multilateralism was ‘more time-consuming, less reliable, and more limiting than operating alone.’ Instead, fighting alone was viewed as more time consuming, less reliable, and more limiting.
Indeed, Kreps recognizes that coercive behavior may induce a balancing response, including the formation of rival multilateral alliances. Napoleon’s aggression produced a coalition that “overcame earlier coalition failures and marshaled enough manpower and unity to bring down Napoleon” (25).
But this begs a question: if other states tend to band together in the face of a threat, why does Washington respond to a threat by abandoning its allies and going it alone?
One means of reconciling these arguments is that a direct threat causes more balancing behavior than no direct threat. But the response may still be more unilateral than substantively multilateral.
Another part of the answer is that Kreps uses the language of threat in a slightly different way than does Walt. For many realists, threat is a function of capabilities, aggressive intent, proximity, and offensive capacity. The greater the degree of threat, the more likely we are to see alliance formation or multilateralism. Kreps, however, distinguishes between two key variables. The first is “time horizons” or the “directness of threat,” including: (1) threats against the homeland, (2) immediate versus distant threats, and (3) military versus political or economic threats. Directness of threat refers not just to a certain type (immediate military threats against the homeland) but also to the scale of the threat. Kreps equates directness of threat with “rapidly changing conditions” and “an environment with more security hazards” and “immediate security threats” (29-30). Direct threats like the invasion of Kuwait in 1990 or 9/11 create an urgency to act quickly and promote unilateralism.
The second variable is “operational commitment” or whether success can be achieved quickly and at low cost. A high operational commitment, or uncertain odds of success, encourages the United States to burden-share and act multilaterally.
Kreps could explain more clearly why a state’s tendency to form multilateral coalitions in the face of threat can be trumped by other dynamics. After all, Walt’s “threat” and Kreps’s “direct threat” overlap to a considerable extent: a proximate state with extensive capabilities, which was exhibiting aggressive intent, would also tend to represent a direct threat. But Kreps’s point is that not all threats are alike. Those that are direct but seem easy to resolve will induce unilateralism rather than alliance formation or balancing.
An important issue that needs further consideration is enemy capabilities. For realists, higher enemy capabilities mean a greater threat and enhanced odds of alliance formation and balancing. In Kreps’s argument, rising enemy capabilities seem to produce two contrary effects: heightening direct threats to the United States and therefore the unilateral temptation, while also increasing the required operational commitment to win, thereby encouraging multilateralism. If the enemy becomes more capable, do the odds of unilateralism increase or decrease?
Kreps seems to be aware of this issue because at one point she argues that the directness of threat (or time horizon) trumps operational commitment (34). In other words, when the United States faces an immediate military threat it will act unilaterally, even if there are substantial costs involved. After 9/11, for example, the United States invaded Afghanistan unilaterally, despite the high degree of commitment required.
But the reverse may be true: operational commitment can trump directness of threat. If the cost of tackling an adversary grows steep enough, the United States (or any actor) will have no choice but to act in concert with allies.
Indeed, if one examined the effect of heightening direct threats on the odds of unilateralism, we might discover a bell curve. As direct threats rise from a low threat level scenario like ‘Haiti 1994’ to a medium threat level scenario like ‘9/11,’ the United States becomes more inclined to act unilaterally. But as the direct threat heightens still further, to a ‘World War II’ scenario, Washington is once again inclined to act multilaterally.
At the extreme, operational commitment trumps other factors. If the United States engaged in a war against China, with a very high direct threat level, Washington would quickly discover its love of allies. The high direct threat level would be correlated with an extreme degree of operational commitment.
Now let me turn to some other aspects of the book. Is it true that the benefits of multilateralism are primarily long term while the gains from unilateralism are primarily short term? Here, the distinction is drawn too sharply. Multilateralism and unilateralism provide a complex series of short-term and long-term costs and benefits that defy easy categorization. Yes, multilateralism can provide long-term gains in terms of legitimacy and bolstering the international system. But it also offers very immediate pay-offs in burden sharing, the use of foreign bases, and so on.
Similarly, unilateralism can produce short-term benefits in efficiently delivering military success. But unilateralism also produces long-term benefits, for example, by swiftly removing a threat to the international system. Whether the beneficial ends (removing a threat) outweigh the costly means (unilateralism) in terms of long-term system stability is an open question.
Kreps argues that states will not choose to act multilaterally if doing so will undermine the effectiveness of the mission (154). This is generally true but an exception may exist for post-conflict stabilization. A U.S.-led operation may be more effective at stabilization or nation-building than a broad based multilateral coalition—yet Washington still prefers to hand this kind of mission over to partners.
Consider the U.S. intervention in Somalia in 1992-1993. Kreps categorizes the case as “formal multilateralism” (22). But there were really two stages to the mission. The first phase, from December 1992 to May 1993, was a U.S.-dominated ‘formal multilateral’ venture called UNITAF, with 28,000 American troops. But in May 1993, UNITAF handed over to a ‘full multilateralism’ operation called UNOSOM II, with a much-diminished U.S. role (around 3,000 troops).
UNITAF was a well-organized intervention that succeeded at stabilizing southern Somalia and delivering aid, and may have saved 100,000 Somali lives. By contrast, UNOSOM II’s motley collection of troops from Botswana, Bangladesh, Pakistan, Italy, and elsewhere were no match for U.S. forces, and arrived without even basic equipment. No single country provided leadership; instead, each member of the coalition pursued its own agenda. The mission eventually unraveled in the wake of the October 1993 ‘Black Hawk Down’ battle in Mogadishu, when eighteen Americans died.
Arguably, the U.S. favors multilateral nation-building less in the service of maximizing the effectiveness of the mission, and more because Washington wants to rid itself of an unsavory duty. The United States traditionally favors regime change missions over nation-building—preferring to cook the dinner rather than wash the dishes. In Somalia, the large number of coalition members in UNOSOM II did not represent a force multiplier. Instead, it was America’s ticket to reduce its role and then exit precipitously. The Clinton administration gained political cover for the U.S. withdrawal from Somalia by falsely blaming the United Nations for the ‘Black Hawk Down’ battle—even though Washington had actually masterminded the mission in Mogadishu.
If the sweet spot for American unilateralism is a direct threat and a low operational commitment, the puzzle is why Americans would feel threatened by an adversary they believe to be easy to defeat. As Kreps suggests in the conclusion, actors are prone to overconfidence about the outcome of war and therefore downplay the need for allies, for example, with Iraq.
Here, one might wonder why actors are overconfident about using force but not overconfident about the potential contribution of allies. In fact, the psychological literature backs up Kreps by suggesting that there is a systematic bias at work. Leaders tend to exhibit positive illusions about their own country’s capabilities but simultaneously display a negative bias when considering the behavior of other states, including potential allies.[22]
Kreps’ case studies are well written and clear. The reader may naturally wonder if the theory applies more widely. The argument that, “Powerful states act multilaterally as often as possible because multilateralism is a power-conserving strategy” (151) does not hold before World War II, when the United States almost invariably acted unilaterally (even in World War I it was an ‘associated power’ rather than a formal ally of Britain and France). This begs the question: what changed so dramatically?
Kreps does not consider Cold War cases, in part because unilateralism was less a choice than the product of an inevitable veto by one of the two superpowers in the UN Security Council. Still, it is not clear from Kreps’s argument why the United States pursued a more multilateral strategy in Korea than in Vietnam, including constructing a much broader international coalition on the ground. The theory would explain this result in terms of the shorter time horizons or lower operational commitments in Vietnam. But the time horizon was shorter in Korea than in Vietnam. North Korean troops looked set to overrun South Korea in the summer of 1950. Washington, however, took the trouble of building a wide-ranging alliance that stretched from the British Labour government to South Korea’s Syngman Rhee. And operational commitments turned out to be much greater in Vietnam, where the U.S. had 500,000 troops deployed by 1968.
More recently, in 2011, the U.S. pursued a ‘full multilateral’ strategy in Libya. Kreps’s theory would explain this in terms of long time horizons and high commitments. But Washington saw this as a low commitment operation and therefore we would predict ‘formal multilateralism’ or a U.S.-dominated mission with UN blessing. Here, historical learning is surely important, as Washington sought to avoid a U.S.-dominated operation similar to that of Iraq.
Overall, this is a very successful and thought-provoking book that deserves a wide readership.
Author’s Response by Sarah E. Kreps, Cornell University
Let me first start by sincerely thanking the distinguished set of roundtable participants for their careful reading of my book, Coalitions of Convenience. Their insights are invaluable both to me and undoubtedly will be for those who work on issues of international organizations, the use of force, and international relations theory.
By way of addressing the comments, I would like to say a few words about the book’s origins. After the Iraq War, there was considerable debate among academics not just about the decision to intervene, but also how the U.S. intervened, that is, without UN authorization and flanked only by a ‘coalition of the willing.’ The latter debate approached the level of hand wringing among international relations scholars, the overwhelming consensus being that unilateralism would come at high costs to the U.S. Stephen Brooks and William Wohlforth challenged this conventional wisdom in a 2005 Perspectives on Politics piece, noting that, “to anyone familiar with this normally squabbling scholarly community, such widespread agreement is noteworthy.”[23] That scholars, irrespective of their theoretical proclivities, would be unified in their concerns about a policy of unilateralism struck me as unusual and possibly incongruent with their theoretical views.
It turned out that unilateralism was at odds with what constructivists and institutionalists expected to see even of powerful states such as the U.S. but consistent with many variants of structural realism, which would have expected unipolarity to “select for” unilateralism.[24] The problem is that none of these existing accounts seemed to explain variation: why would we see unilateralism in cases like Iraq but multilateralism in the 1991 Gulf War? Are patterns of cooperation better explained by individual leaders that have proclivities towards multilateralism or unilateralism? Is there something about the particular intervention or security environment that encourages unilateralism in some cases but not others? What did we even mean by multi and unilateralism?
While a puzzle for many theories of international relations, cooperation on the part of a powerful state seemed quite consistent with a classical realist interpretation of state behavior. As I note in the book, classical realists took factors such as legitimacy and morality seriously, although mostly for instrumental reasons. A state that had become powerful within a certain world order could better perpetuate its power by working within those norms than by contravening them. Not that these norms would override power considerations, but they could operate side by side and could be mutually reinforcing. Based on this assumption, the theoretical framework assumed that multilateralism was the default cooperation strategy for a powerful state but that short time horizons and large burdens would create incentives for unilateralism.
Although the reviewers seem to be persuaded by the overarching theoretical framework of my book, they do raise a number of points that push on both theoretical and empirical points. I appreciate the opportunity to address these comments, focusing in particular on those that more than one reviewer raises.
The Dependent Variable
One of the most important questions has to do with the way I define the dependent variable: multilateralism and unilateralism. My goal in establishing an operationalizable definition was to go beyond the ‘eye of the beholder’ definition that was made manifest with the Iraq war: the Bush Administration and its supporters asserted that the ‘coalition of the willing’ that intervened would be multilateral but to critics it appeared to be a U.S. unilateral intervention masquerading (not even very well) as multilateralism. Drawing on theories of institutional design, I crafted a definition that I believe captured the logic of true multilateralism: that the parties involved have some voice in decision-making (i.e., it’s not just about numbers). The two-level approach—institutional authorization and active state participation—means that true unilateralism is rare. Indeed, there are few Panama-style interventions, where the U.S. has no international organization (IO) authorization, no allies, and indicates among principal decision makers that it will notify other states after the intervention.
Brooks notes, however, that there are less restrictive definitions of unilateralism that would produce a more unilateral record than if we used my more restrictive definition. First, not having UN authorization could be construed as unilateral. Considering this more easily operationalized definition would, however, only change the coding of one intervention: Kosovo. Thus, the analysis is not drastically different with this coding.
The second and third definitions have to do with the degree to which the U.S. consults other states and is willing to change its strategy if it encounters opposition from others. These two latter definitions are certainly squishier and, as Brooks concedes, difficult to operationalize. What degree of consultation is required for it to count as multilateral? Unilateralism defined as ‘acting against the wishes of other countries’ has similar problems. Those latter two, which turn on the degree of consultation and approval of other countries, bring us back to the fundamental question, which is how to operationalize consultation and voice opportunities in a way that goes beyond the ‘know it when you see it’ test. Nonetheless, I acknowledge that my efforts to advance the conceptualization of multilateralism and unilateralism were not the last word on the subject and hope that others can move this forward.
The Independent Variables
A second main point in the comments had to do with the way I conceptualized the two main variables, time horizon and operational commitment. Perhaps more than any other issue, this is the one I ruminated over the most as I wrote the book. Both of these variables are basically based on threat, and threat invariably is difficult to operationalize because a security environment is interpreted in part, as Dominic Tierney notes, through one’s risk propensities, world views, and experiences. In other words, there’s no objective threat ‘out there’ so the question for an argument based on threat is how we discern a leader’s actual perception of threat as opposed to his politically-motivated rhetoric, which would tend to inflate the threat. This criticism, as Alexander Thompson points out, is especially germane in the case of Iraq, which is the subject of the most lively debates about what we knew and when we knew it. That there remains scholarly debate about this question is evident even among the reviewers; Patricia Sullivan, for example, codes Iraq as a short time horizon (high threat), and Thompson observes that Iraq was not “a grave threat.”
A similar question arises with the operational commitment variable—whose assessment of the burden do we consider? Debates in the run-up to the Iraq War illustrated the divisions within high levels of government about the burden in terms of troops and financial resources—Army Chief of Staff General Eric Shinseki famously estimated that several hundred thousand troops would be needed and Treasury Secretary Lawrence Lindsey estimated that the war would cost $100-$200 billion. As Thompson observes, the State Department fully anticipated the post-war reconstruction work that would be required. The key ‘deciders,’ however, gave short shrift to those views and gave more weight to the estimates from the Pentagon. Sullivan is correct to say that reality is more complicated than our theories would suggest; the complicated reality of this story would include a theory of bureaucratic politics, illustrated by the effect of the Pentagon’s heavy hitters such as Secretary of Defense Donald Rumsfeld and Under Secretary of Defense for Policy Doug Feith who trumped more cautious voices in the State Department. As Tierney rightly observes, this is even more than a story of bureaucratic politics, it is a question of why negative biases often prevail. Or put a slightly different way, it’s a story about “why hawks win”[25] and tend to be more influential in foreign policy debates.
These questions about how to operationalize—or as Thompson formulates it, what is “the type of evidence that is most probative”?—the two independent variables give me a valuable opportunity to clarify these ideas. Leaders certainly have incentives to misrepresent both domestic and international audiences, but it would be surprising for classified documents to misrepresent what leaders thought. And in the case of the Iraq War, it is clear, for example, that the military had planned for being out of Iraq by 2006;[26] the story about whose views were taken into account in terms of generating these predictions is the complicated one outlined above, but the upshot is that the U.S. government concluded that it would win the war easily and get out. With the erroneous expectation that the U.S. could intervene quickly, the U.S. was more willing to intervene with a coalition in which it contributed the lion’s share of the troops and financial resources. An appropriate counterfactual question is what cooperation strategy the US would have pursued had it known that the war would cost $800 billion and whether it would have waited for more allies before intervening. One of the senior members of the State Department whom I interviewed made it clear that the response would have been an unqualified yes.
Offering dispositive evidence regarding the threat variable is a trickier proposition because we never have complete information about another state’s capabilities; we therefore have to make inferences based on incomplete evidence, whether that of satellite images, human intelligence sources, etc. The task is even more difficult for authoritarian states and those for which we have few good human intelligence sources, which was the case for Iraq after it evicted UN inspectors at the end of the 1990s. The intelligence community then typically generates National Intelligence Estimates (NIE) that would characterize the threat. These estimates are far from flawless of course. Robert Jervis notes in his study of intelligence failures associated with Iraq’s WMD programs that “it is clear that Iraq was a case of collection failure in that the evidence collected was scattered, ambiguous, and often misleading.”[27] In spite of these shortcomings, the NIE remains the best guess as to another state’s capabilities and declassified NIEs would presumably lack the political motivations to misrepresent as would rhetoric or NIEs that are unclassified.
Beyond the question of operationalizing the two main variables comes the question of enemy capabilities that Tierney raises. I appreciate the opportunity to say more about this as it relates to the operational commitment variable. Certainly, one thing that makes the unipolar world different from other distributions of power is that one state has a preponderance of power relative to others. As a result, there are far more match-ups that look like the U.S.-Haiti or U.S.-Somalia than the U.S.-Soviet Union. Nonetheless, conflict with states that could put up a fair fight does arise—the U.S. thought the 1991 Gulf War was one of those cases—and these calculations about capabilities are subsumed under the operational commitment variable. How much an effort the intervention will require is a function of how formidable the adversary is. Tierney suggests that this could produce a bell curve-like response with both the least and most formidable adversaries prompting a multilateral response. While this is an interesting proposition, it seems to conflate the two main variables. For example, if Iran attacked the Fifth Fleet in the Gulf, my argument would expect the U.S. to act unilaterally at the outset and then try to collect allies after its initial response. Its time horizon would be short but the operational commitment would be high, producing a unilateral initial response followed by multilateral response, perhaps like the Gulf War (although maybe not with UN authorization given the likelihood of a Russian veto). Thus, rather than a bell curve, the more likely response would be a two-stage hybrid strategy in which the US would employ one strategy at the outset—since the U.S. is arguably the only country that has global expeditionary forces, which de facto means that most initial responses would be unilateral or bilateral alongside the host country—followed by a more robust multilateral response.
Also picking up on the operational commitment variable, Thompson raises the question of whether the causal relationship is actually inverted, with the strategy determining the notion of commitment rather than the other way around, which is what my argument expects. According to this argument, the U.S. would determine its cooperation strategy and then reverse engineer the case for why a particular military strategy is in order. This seems problematic because we still need a theory of cooperation preferences. The obvious candidate, that certain leaders have cooperation propensities, doesn’t hold up to scrutiny. For example, although George H.W. Bush intervened multilaterally in Iraq, the Panama intervention was as unilateral an intervention as there has been. And Bill Clinton’s unilateral strikes against Iraq in 1993 and Afghanistan and Sudan in 1998 were the subject of international censure. Thus, it does not appear as though leaders have consistent preferences for a particular cooperation strategy. Another candidate explanation, which is that leaders would just generally intervene with UN authorization and a coalition, whether because of norms or for burden-sharing, does not explain the number of instances in which the U.S. intervened unilaterally.
As evidence that the cooperation preference determines the military strategy rather than the other way around, Thompson points to Brent Scowcroft’s comment that the “welcome troop contributions themselves became a challenge to manage.” In other words, the difficulty of integrating these allied military forces suggests that military imperative could not have been the motivation. However, I read this comment more as a ‘necessary evil’ remark, much in the way that Generals Montgomery and Dwight Eisenhower regarded each other in World War II. Historian Stephen Ambrose raises the “nagging question, Why did Eisenhower put up with Montgomery? He had no choice. He had to cooperate with the difficult and exasperating British general.”[28] I interpret Scowcroft’s remarks in the same light, which is that the coalition produced challenges but was thought to be unavoidable given what the U.S. expected of an Iraqi military seasoned from nearly a decade of fighting in the 1980s.
To this point, the anticipated strength of the opponent influences the cooperation strategy as much or more than the operational goals. I agree that the 2003 war certainly had more ambitious objectives, but against what the U.S. thought was a much less fearsome foe. Similarly, the U.S. crafted its unilateral approach in Afghanistan based on the expectation of a relatively effortless war; as I note in the book, the US had an embarrassment of riches in terms of available allies, but its military strategists believed that these allies would be extraneous in light of the “Afghan Model,”[29] a strategy designed around a small military footprint and reliance on local allies. Having said all this, I agree with Thompson that the book could have done more to show how to sift through the evidence and link the independent variables with the cooperation outcome. Doing so would have helped to guide the interpretation of historical events, a point raised in several comments.
Lastly, under the heading of the independent variables, Tierney notes as an historical aside that Korea had a shorter time horizon but was multilateral compared to Vietnam. As I explain below, I did not include these cases in my book because of the conscious effort to hold constant a particular distribution of power, unipolarity. Nevertheless, I see these cases as being generally consistent with the argument, in part because my understanding of the degree of multilateralism in Korea is different from the more robust coalition effort that Tierney suggests. The historian William Stueck notes that the U.S .and South Korea assumed the vast majority of the burden in part because “the United States had been clumsy in soliciting assistance during the first two months of the war, when a psychological moment existed to secure foreign contributions.”[30] Stueck also quotes General MacArthur’s testimony that fighting without the allies the U.S. did have “would have no material effect upon the tactical situation.”[31]
To be sure, the U.S. had requested a British contingent in the early weeks,[32] but other efforts to secure contributions were anemic. Only once stalemate set in and the U.S. began to scrutinize the thin contributions of allies did the Department of Defense redouble its efforts for more allied contributions. Thus, the Korean War was largely a U.S.-South Korean effort, which is entirely consistent with the argument, since short time horizons (in this case the prospect of overrunning the Korean peninsula) would not be expected to produce a robust multilateral response. Also consistent with the argument is the latter day effort to secure allies, once the U.S. realized the magnitude of the burden in Korea.
Implications of the Theory
A couple of the reviews raise the question of implications, both historical and for the future. Both deal with issues of scope conditions of the theory. As I noted in the book, many theories of international relations say little to nothing about what happens when power is concentrated in the hands of one state (rather, most theories speak to conditions of bipolarity or multilaterality).[33] My explicit intention was to focus on conditions of unipolarity, as I note in chapter two of the book, since existing theories have little to say about this distribution of power. The shortage of theoretical insights on unipolarity and the enduring nature of unipolarity[34] made this choice of scope conditions appropriate.
The discussion of scope conditions also addresses several points raised by the reviewers. First, it speaks to Tierney’s question of whether this theory really challenges realism. As he notes, Stephen Walt’s theory of balancing assumes that states band together to challenge threats. However, this makes the assumption that one state needs to aggregate the capabilities of other states. Part of the premise of the book is that in a unipolar world, the U.S. has a preponderance of power, meaning that it does not necessarily band together with states to confront threats. The most extreme form of this was the Haiti intervention; in 1994, the year the U.S. intervened, the U.S. defense budget of $288 billion was twenty times the entire gross domestic product of Haiti. But similar points can be made regarding Afghanistan and Iraq. The point is that balancing and alliances work differently for a state that holds a large concentration of power. To answer one of Tierney’s queries, this is why a direct threat might actually provoke unilateralism rather than multilateralism: in many cases the U.S. does not need to aggregate power the way Austria, Britain, and other members of the coalition did in the Napoleonic Wars. Even when other countries share the threat, they too have incentives to buck-pass to the U.S. and indeed it is not surprising that a number of U.S. allies have reduced their defense budgets (as Brooks notes), i.e., they know the U.S. can undertake most of these operations on its own.
It also addresses Sullivan’s question of why there are no instances of participatory multilateralism in the universe of cases. A brief discussion on page 21—which would have been conveyed more powerfully with a graph—indicates that vetoes plummeted after the Cold War. The Permanent Five essentially became the Permanent One, in which a powerful U.S., eviscerated Russia, and still-rising China created a set of dynamics in which the possibility of a UN veto was virtually zero. With enough time and side payments—as the 1991 Gulf War of cajoling allies and rotating members of the Security Council suggests—the U.S. was generally in a position to win enough support for its foreign policy initiatives. The dynamic was different in the Cold War, when explicit and implicit vetoes meant that most interventions consisted of cobbled together coalitions, often feeble ones like that in the 1965 Dominican Republic intervention, which consisted of 23,400 US forces, and about 1500 token Brazilians, Hondurans, Paraguayans, Nicaraguans, Costa Ricans, Salvadorans, and no UN authorization.
In his comments, Brooks raises the question about how the argument travels over time. In this context he raises the question of whether there is in fact a legitimacy premium on acquiring UN authorization and therefore whether in its act of “forum shopping,” the U.S. would jettison the pretense of only acting with UN authorization. After all, if the findings of his recent research are correct then the public is not enamored of UN authorization and U.S. leaders would incur few costs of intervening without UN authorization. My own work drew somewhat different conclusions—that the U.S. public prefers UN authorization but does not necessarily want to be constrained from acting unless they have it—but in some ways there are two more important considerations.
The first is whether leaders act as if UN authorization matters, and there’s evidence to suggest that they do. Even President George W. Bush and his legal team went to great lengths to frame their Iraq intervention in a way that was consistent with UN resolutions.[35] Why, more recently, did the U.S. and its allies insist on UN authorization to intervene in Libya? As a corollary, it is clear that a main deterrent of intervening in Syria is that the Russians vetoed every attempt for more stringent measures against the Assad regime. The U.S. is reluctant to intervene without UN authorization. Second, insofar as American leaders want the participation of allies, then the UN imprimatur is crucial, since foreign domestic audiences seem preoccupied with it, if not U.S. domestic ones. As was the case in Iraq, President Bush wanted to intervene with the UK and Prime Minister Tony Blair made it clear—or acted as if—his own domestic audience saw UN authorization as a sine qua non. In this sense, the public’s actual views on the question are less relevant than the beliefs leaders have about their publics, which is that they prefer the UN’s legitimation function.
With these points in mind, I would still expect the U.S. to have a preference for multilateralism although the recent experience with Syria points to how that preference could erode. Regardless of how much the conflict degenerates, Russia has opposed international intervention. Of course it is not clear how much countries such as the U.S., UK, and France were interested in intervening, but generating any consensus in the UN was a non-starter. Libya was unique in that its leader, Moammar Ghadaffi, was universally despised and had no major allies. Almost every other country is likely to have an ally that would make P5 consensus in the UN Security Council difficult to achieve, more so with a U.S. in relative decline and Russia and China with a stronger voice. I suspect that a NATO intervention would have considerable appeal for the U.S. if it hoped to intervene but could not generate UN authorization. In any case, the high water mark for UN authorization with regards to Syria has likely passed and notwithstanding the unique situation with Libya, I would expect the U.S. to intervene more often flanked by a ‘coalition of the willing’ or like-minded NATO. Brooks wonders if the material incentives to intervene with a coalition have declined as other states have fallen behind in technology over the last decade, but this has corresponded with—and is likely offset by—a decreased appetite for intervention after Iraq and Afghanistan.
I would like to conclude by reiterating my gratitude to the reviewers for their thoughtful comments and to Thomas Maddux for organizing the roundtable. I hope that this discussion can guide a fruitful continuation of related work.
© Copyright 2012-2015 The Authors.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License.
Notes
[1] See, for example, Elizabeth Saunders, Leaders at War: How Presidents Shape Military Interventions (Ithaca: Cornell University Press, 2011); Mark L. Haas, The Ideological Origins of Great Power Politics, 1789-1989 (Ithaca, NY: Cornell University Press, 2005).
[2] See, for example, Martha Finnemore, The Purposes of Intervention: Changing Beliefs about the Use of Force (Ithaca, NY: Cornell University Press, 2003).
[3] See, for example, G. John Ikenberry, Liberal Leviathan: The Origins, Crisis, and Transformation of the American World (Princeton: Princeton University Press, 2011).
[4] See, for example, Stephen M. Walt, The Origins of Alliance (Ithaca, NY: Cornell University Press, 1990).
[5] See, for example, Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976)
[6] See, for example, Jack Snyder, The Myth of Empire: Domestic Politics and International Ambition (Ithaca NY: Cornell University Press, 1993).
[7] Deborah Brooks and Stephen Brooks, “Do Security Institutions Matter to the American People?” Working Paper.
[8] Stephen Brooks, “Dueling Realisms,” International Organization, 51:3 (19997): 445-477.
[9] Sarah Kreps, “Multilateral military interventions: Theory and practice”, Political Science Quarterly 123(2008): 573-603.
[10] Stephen Brooks, “Can We Identify a Benevolent Hegemon?” Cambridge Review of International Affairs, 25:1 (2012): 27-38.
[11] Martha Finnemore, The purpose of intervention: Changing beliefs about the use of force (Ithaca: Cornell University Press, 2004).
[12] Manfred Bertele and Holger Mey, “Unilateralism in theory and practice”, Comparative Strategy, 17: 2, (1998): 197-207; John Dumbrell, “Unilateralism and ‘America First’? President George W. Bush’s foreign policy”, Political Quarterly, 73 (2002): 279-287.
[13] Bertele and May, 199.
[14] Charles Krauthammer, “The unipolar moment revisited”, National Interest, 70 (2002-2003): 17.
[15] Terrence Chapman, Securing Approval: Domestic Politics and Multilateral Authorization for War, (Chicago: University of Chicago Press, 2011); Alexander Thompson, Channels of Power: The UN Security Council and U.S. Statecraft in Iraq, (Ithaca, NY: Cornell University Press, 2009); David A. Lake, Entangling Relations: American Foreign Policy in Its Century, (Princeton, NJ: Princeton University Press, 1999).
[16] Richard N. Haass, War of Necessity, War of Choice: A Memoir of Two Iraq Wars, (New York: Simon & Schuster, 2009).
[17] Secretary of Defense Donald Rumsfeld, in particular, was insistent on using a small force. On the impact of insufficient forces in Afghanistan, see Terry H. Anderson, Bush’s Wars, (New York: Oxford University Press, 2011), pp. 85-88.
[18] George Bush and Brent Scowcroft, A World Transformed, (New York: Alfred A. Knopf, 1998), p. 342. See also Thompson, Channels of Power, pp. 51-3.
[19] See Thomas E. Ricks, Fiasco: The American Military Adventure in Iraq (New York: Penguin, 2006); Chaim Kaufmann, “Threat Inflation and the Failure of the Marketplace of Ideas: The Selling of the Iraq War,” International Security 29 (Summer 2004), 5-48; and Scott McClellan, What Happened: Inside the Bush White House and Washington’s Culture of Deception, (New York: PublicAffairs, 2008).
[20] Stephen M. Walt, Taming American Power: The Global Response to U.S. Primacy (New York: W.W. Norton, 2005); Thompson, Channels of Power; G. John Ikenberry, After Victory: Institutions, Strategic Restraint, and the Rebuilding of Order after Major Wars, (Princeton, NJ: Princeton University Press, 2001).
[21] Stephen M. Walt, The Origins of Alliances (Ithaca: Cornell University Press, 1987)
[22] Dominic Johnson and Dominic Tierney, “Bad World: The Negativity Bias in International Relations,” paper presented at the annual meeting of the International Studies Association, San Diego, March 2012.
[23] Stephen Brooks and William Wohlforth, “International Relations Theory and the Case against Unilateralism,” Perspectives on Politics, Vol 3, No. 3 (2005), 509-525; 510.
[24] John Ikenberry, “Is American Multilateralism in Decline?” Perspectives on Politics, Vol 1, No.3 (Sept 2003), 533-550; 533.
[25] Daniel Kahneman and Jonathan Renshon, “Why Hawks Win,” Foreign Policy, Jan/Feb 2007.
[26] See declassified Top Secret Polo Step planning slides generated by the Defense Department, available at http://www.gwu.edu/~nsarchiv/NSAEBB/NSAEBB214/index.htm
[27] Robert Jervis, “Reports, Politics, and Intelligence Failures: The Case of Iraq,” The Journal of Strategic Studies, Vol. 29, No. 1 (Feb 2006), 3-52; 13.
[28] Stephen Ambrose, The Victors: Eisenhower and His Boys the Men of World War II (New York, NY: Simon and Schuster, 1998), 147.
[29] Stephen Biddle, “Allies, Airpower, and Modern Warfare: The Afghan Model in Afghanistan and Iraq.” International Security 30, no. 3 (Winter 2005/06): 161-176.
[30] William Stueck, The Korean War, (Princeton, NJ: Princeton University Press, 1995), 195.
[31] Stueck, The Korean War, 194.
[32] M.L. Dockrill, “The Foreign Office, Anglo-American Relations and the Korean War, June 1950-June 1951,” International Affairs, Vol. 62, No. 3 (Summer 1986), 459-476.
[33] The 2009 special issue of World Politics dealt with whether and how international relations theories operate differently under unipolarity.
[34] Nuno Monteiro, “Unrest Assured: Why Unipolarity is Not Peaceful,” International Security, Vol 36, No. 3 (Winter 2011/12), 9-40; Michael Beckley, “China’s Century? Why America’s Edge Will Endure,” International Security Vol. 36, No. 3 (Winter 2011/12), 41-78.
[35] John Yoo, “International Law and the War in Iraq,” American Journal of International Law 97 (2003), 563.