In Leaders at War, Elizabeth Saunders examines the use of military force by states to intervene in other nations’ domestic affairs. Why, she asks, do some military interventions explicitly seek to transform the societies and institutions of the states they target while others do not? And more basically, “why do great powers like the United States undertake overt intervention in some conflicts or crises but not in others?” (2) As Saunders rightly notes, it’s not enough to study interventions that occurred; we should also examine those that might have occurred but did not.
H-Diplo | ISSF Roundtable, Volume III, No. 8 (2012)
Thomas Maddux and Diane Labrosse, H-Diplo/ISSF Editors
George Fujii, H-Diplo/ISSF Web and Production Editor
Commissioned for H-Diplo/ISSF by Thomas Maddux
Elizabeth N. Saunders. Leaders at War: How Presidents Shape Military Interventions. Ithaca: Cornell University Press, 2011. ISBN: 978-0-8014-4922-2 (hardcover, $35.00).
Published by H-Diplo/ISSF on 15 February 2012
Introduction by Fredrik Logevall, Cornell University. 2
Review by Paul K. MacDonald, Wellesley College. 5
Review by Rose McDermott, Brown University. 9
Review by Dan Reiter, Emory University. 13
Review by Thomas A. Schwartz, Vanderbilt University. 16
Response by Elizabeth N. Saunders, George Washington University. 19
Introduction by Fredrik Logevall, Cornell University
In Leaders at War, Elizabeth Saunders examines the use of military force by states to intervene in other nations’ domestic affairs. Why, she asks, do some military interventions explicitly seek to transform the societies and institutions of the states they target while others do not? And more basically, “why do great powers like the United States undertake overt intervention in some conflicts or crises but not in others?” (2) As Saunders rightly notes, it’s not enough to study interventions that occurred; we should also examine those that might have occurred but did not.
In addressing when and how states choose military action Saunders attaches particular importance to the role of individual leaders—in this case, U.S. presidents Dwight D. Eisenhower, John F. Kennedy, and Lyndon B. Johnson. To many H-Diplo readers it will seem axiomatic that presidents matter enormously in questions of war and peace, but in the International Relations field in political science, Saunders tells us, this is an “often-overlooked factor.” Many IR theorists see individual leaders as too idiosyncratic to examine analytically, while others believe that all decision-makers respond to foreign or domestic challenges in similar ways. Saunders, by contrast, maintains that leaders “vary systematically in how they perceive threats, and that these different threat perceptions help explain when and how states intervene.” (2)
The reviewers, Paul K. MacDonald, Rose McDermott, Dan Reiter, and Thomas A. Schwartz, find much to like in the book, praising Saunders for the empirical depth of her research, for the quality of her prose, and for the fair-mindedness with which she presents her arguments. Writes Schwartz: “Her credibility is greatly enhanced by the fact that she is not selective in her sampling of historical evidence, and repeatedly references material that may not fit her argument.”
Yet the reviewers also express concerns about Saunders’s central claim: that the crucial variable that distinguishes leaders is the degree to which they believe that the internal nature of other states is the ultimate source of threats. Separating leaders into two ideal-types—those who see threats as emerging solely or mostly from the foreign policies of other states and those who diagnose threats as arising from the domestic institutions of states—she argues that leaders in the latter category are more likely to use military intervention to try to transform the institutions of the state in question. MacDonald credits Saunders for conceding the limitations of her approach and for noting that internally focused leaders will not necessarily intervene more frequently than their externally focused counterparts. But he nevertheless finds that she underplays the degree to which “multiple and overlapping sets of beliefs shape how leaders respond”—beliefs, in the case of these three presidents, about the nature of the Soviet threat, about U.S. capabilities, and about the power of the United States to actually reform other societies. McDermott, too, wonders about other potential motivations for intervention, and she also wishes that Saunders had done more to engage prior scholarship on the psychology of leadership and the nature and function of biases.
Reiter and MacDonald applaud Saunders for giving close attention to the three presidents’ views on foreign policy before they came to the White House, the better to explain their beliefs about the origins of threats and the necessity of transformative change. Yet for Reiter this analysis of the pre-presidential years ultimately proves to be of limited utility in explaining why the presidents arrive at different assessments regarding what military interventions would accomplish. He further asks how important it is to find out whether a resort to intervention seeks transformation or not, given that such a determination tells us little about how large and destructive the military action will be.
Schwartz, meanwhile, as the lone historian in the group, stresses the importance of paying due attention to the international and domestic political context and to periodization. The situation within a given country, and the assessment of key external actors, he argues, can matter more than a U.S. president’s beliefs in determining the nature of a military intervention—and indeed whether the intervention happens at all. On Vietnam (which gets close and sustained attention in Saunders’s empirical chapters), both Schwartz and MacDonald point out that the choices that LBJ faced in early and mid-1965 were not the same as those that JFK faced two years before. The situation on the ground in South Vietnam had changed, as had the political context within the United States.
For political scientists, certainly, if not so much for historians, the question that looms large at the end of this stimulating and intelligent study is, as McDemott says, whether its author’s findings are generalizable and can help us better understand future decisions about the use of military force, not least by American presidents. The reviewers express skepticism in various ways. But all agree that Saunders succeeds marvelously in showing how much individual leaders matter when crunch time comes. A good thing to remember as another U.S. presidential campaign gets under way.
Elizabeth N. Saunders is an Assistant Professor of Political Science and International Affairs at George Washington University. She received her Ph.D. from Yale University. Her research and teaching interests focus on international security and U.S. foreign policy. Her work has appeared in International Studies Quarterly, International Security, the American Journal of Political Science, and International Studies Review. She is currently working on two projects, one on the consequences of the “electoral disconnection” in U.S. foreign policy, and the other on the origins and scope of power projection capabilities.
Fredrik Logevall is the John S. Knight Professor of International Studies and Professor of History at Cornell University, where he also serves as director of the Mario Einaudi Center for International Studies. He received his Ph.D. in 1993 from Yale University and previously taught at UC Santa Barbara where he co-founded the Center for Cold War Studies. His new book, Twilight War: The Fall of an Empire and the Making of America’s Vietnam, is forthcoming in August from Random House.
Paul K. MacDonald is an Assistant Professor in Political Science at Wellesley College. He has held research positions at the Williams College, American Academy of Arts and Sciences, the Belfer Center for Science and International Affairs, the Olin Institute for Strategic Studies, and the Center for International Security and Cooperation. He has published articles in International Security, Review of International Studies, Security Studies, Daedalus, and the American Political Science Review.
Rose McDermott is a Professor of Political Science at Brown University. McDermott received her Ph.D. (Political Science) and M.A. (Experimental Social Psychology) from Stanford. McDermott is the author of three books, a co-editor of two additional volumes, and author of over eighty academic articles across a wide variety of disciplines encompassing topics such as experimentation, identity, emotion, intelligence, decision making, and the biological bases of political behavior. She is the incoming President of the International Society of Political Psychology.
Dan Reiter is chair of the political science department at Emory University. He received his Ph.D. in political science from the University of Michigan in 1994. He is the author or coauthor of dozens of articles as well as three books: Crucible of Beliefs: Learning, Alliances, and World Wars (Cornell, 1996); Democracies at War with Allan C. Stam (Princeton, 2002), and How Wars End_(Princeton, 2009). One of his current projects is the role of secrecy in American entry into war.
Thomas A. Schwartz is a Professor of History at Vanderbilt University. He earned his Ph.D. from Harvard University studying under Professor Ernest R. May and Professor Charles S. Maier. He is the author of America’s Germany: John J. McCloy and the Federal Republic of Germany (1991), Lyndon Johnson and Europe: In the Shadow of Vietnam (2003), co-editor of with Matthias Schultz of The Strained Alliance: U.S.-European Relations from Nixon to Carter. Cambridge: Cambridge University Press, (2010). He is currently finishing a biography of former Secretary of State Henry Kissinger entitled Henry Kissinger and the Dilemmas of American Power.
Review by Paul K. MacDonald, Wellesley College
In her new book Leaders at War, Elizabeth Saunders asks a question of critical importance to American foreign policy: Why do Presidents intervene in some international crises and not others? In addition, why do some Presidents adopt strategies designed to remake the domestic structure of a targeted state while others strive simply to change some aspect of a state’s foreign policy?
Security studies scholars have traditionally argued that the decision to intervene, as well as the magnitude of intervention, will be shaped by the strategic context confronting a would-be intervener. Minor crises will require modest means, while major crises may compel leaders to commit substantial resources. To the extent that scholars have complicated this rather simplistic picture, they have tended to focus on domestic political or bureaucratic pressures that conspire to drive states away from carefully calibrated strategies. Casualty-averse publics, self-interested bureaucracies, and myopic groups of like-minded decisionmakers are among the various factors blamed for excessive or incompetent overseas adventurism.
Saunders does not dismiss these previous theories, but argues instead that the perceptions leaders have of the sources of foreign threats can play a decisive role in deciding when and how states intervene. Saunders separates leaders into two ideal-types: “externally focused” leaders who view threats solely in terms of the foreign policies adopted by other states, and “internally focused” leaders who see a close connection between these foreign policies and the domestic political structure of these states (30-31). Saunders maintains that the beliefs a leader holds about the source of foreign threats colors the way they approach the question of intervention. In particular, internally focused leaders are more likely to embrace “transformative strategies” designed remake the political institutions of states targeted in an intervention (42-43). Externally focused leaders, in contrast, are willing to adopt more modest strategies designed simply to compel a target to remove a corrupt leader or end an objectionable policy.
Saunders’ analytical framework provides a simple and straightforward theory about how leaders’ beliefs can shape interventions, and she admits the limitations of her approach. Her theory does not claim, for example, that internally focused leaders will intervene more frequently than their externally focused counterparts. Indeed, internally focused leaders may be more hesitant to intervene, because they believe successful interventions require a significant commitment in time and resources. Nor does she completely dismiss the importance of strategic context; external constraints still shape leader’s decisions. She is also careful to note that her theory does not predict when interventions will be successful, although she does claim that leaders may try to apply their preferred strategy in circumstances where they are ill suited.
The supreme virtue of Saunders’ book lies in its empirical chapters. Saunders tests her theory through a careful comparison of the intervention choices of three Cold War Presidents – Dwight D. Eisenhower, John F. Kennedy, and Lyndon B. Johnson. She cleverly decides to code these leader’s beliefs about the origins of threats by examining their views about foreign affairs before they became President. This allows her to differentiate between the contemporary pressures these leaders faced as Presidents from their long-standing, and deeply held, beliefs about the nature of Cold War threats.
The story Saunders tells about these three men is both compelling and illuminating, based on the type of careful and comprehensive primary document research that more political scientists should emulate. Kennedy, in particular, stands out as the individual who viewed the Cold War most in internally focused terms. Whereas Eisenhower was content to ignore the brutality and corruption of Third World Allies as long as they remained anti-communist, Kennedy fretted about how authoritarianism and poverty could render these states susceptible to communist subversion and takeover. Saunders persuasively argues that these concerns colored each President’s stance towards intervention. Eisenhower proved content to show the flag in places like Lebanon to shore up American credibility, while Kennedy pressed forward with ambitious programs to reform and remake the Third World, most notably in South Vietnam.
While I came away impressed by the theoretical sophistication and empirical depth of Saunders’s research, some outstanding questions remain. The first concerns the extent to which Saunders sees leaders’ beliefs about external threats as reasonable frames through which to interpret an inherently ambiguous world or as deep-seated biases that can distort their perceptions. In her theoretical chapter, Saunders notes that dynamics that result from “psychological bias” lie “beyond the scope of the theory” (38), and in her empirical chapters, she seems reluctant to pass judgment on Eisenhower’s ignorance of the internal situation in Lebanon or Kennedy’s relentless faith in modernization theory. At times, one even gets the sense that Saunders sees both internally focused and externally focused views as equally valid perceptions of the world, and that the preference for one or the other depends not on evidence but faith.
Many of Saunders’s specific predictions, however, rely on some implicit conception of cognitive pathology. She claims, for examples, that when faced with multiple opportunities to intervene, leaders will “channel” their interventions to cases that most fit their preconditions (42). She likewise argues that when facing domestic pressures to intervene, leaders may push for their preferred strategies, even when these are ill suited to the particular environment (40). For each of these predictions, one has to assume that leaders ignore evidence that contradicts their preconceived beliefs and that they fail to learn from their mistakes once the costs of an ill-advised intervention mount. One can understand why Saunders is reluctant to issue a final verdict on the question of whether threats are in fact a product of the external policies of states or of their internal conditions. But she needs to clarify how she can see leaders as rational calculators of costs and benefits within their particular belief systems (e.g. p. 37), yet appear unable to make cost and benefit calculations about the reasonableness of the beliefs themselves.
A second question that is worth further consideration is whether we can reduce the decisions of the leaders Saunders discusses solely to their beliefs about foreign threats. Indeed, throughout her case studies, one gets the sense that multiple and overlapping sets of beliefs shape how leaders respond. One set of beliefs, for example, has to do with the nature of the Soviet threat and the extent to which leaders view external aggression or internal subversion as the primary strategy the USSR will use to undermine the free world. Another set of beliefs concerns the United States’ capacity to mobilize resources to meet threats, and whether an aggressive Cold War posture might bankrupt the American economy or corrupt the American political process. A third set of beliefs – and perhaps the most decisive – concerns the extent to which a leader believes societies in the Third World can or should be reformed.
These various beliefs are undoubtedly intertwined, perhaps even reinforcing. But Saunders tends to ascribe the beliefs of leaders to their narrow views about Third World threats, while ignoring the way in which broader beliefs about the Cold War, American institutions, and modernization and development more generally might have been motivating the leaders she examines. Indeed, Eisenhower’s reluctance to “dictate to nations” (60) or Johnson’s hesitancy to reform nations “in our own image” (167) seem driven less by threat assessment than by beliefs about the extent to which Arabs and Asians could be transformed into Americans in the first place.
A third question Saunders could address is the relative weight one should place on leaders’ beliefs in comparison to other factors such as the external context in which a crisis takes place. My own sense reading Saunders detailed case studies is that leaders’ beliefs are much more important in shaping the form an intervention will take than in explaining variation in intervention choices, where factors such as the magnitude of the crisis and the feasibility of intervention seem paramount. Eisenhower’s decision to intervene in Lebanon but not Iraq, for example, seems to have as much to do with differences in the duration and intensity of the two crises, as well as the military feasibility of showing the flag in Beirut rather than Baghdad. Feasibility questions likewise seemed critical in Kennedy’s decision to intervene in Vietnam rather than Laos. Similarly, Johnson’s decision to forgo intervention in Panama is not surprising given the modest nature of the crisis in comparison to that which gripped the Dominican Republic.
Even the choice of intervention strategies seems, at times, to be driven by changes in context rather than leaders’ beliefs. Consider the case of American intervention in Vietnam, which plays a reoccurring role in Saunders’s empirical chapters. To be sure, Kennedy emphasized the interplay between domestic weakness in South Vietnam and the escalating guerilla war, and shaped the American advising mission accordingly. But as Saunders admits, the situation facing Johnson in late 1964 and early 1965 was dramatically different than that of his predecessor. As Dale Andrade describes, the North Vietnamese Party Central Committee took the momentous decision in September 1964 “to mobilize…the entire armed forces to concentrate all our capabilities to bring about a massive change in the direction and pace of expansion of our main force army on the battlefield.” As a result, by spring 1965, an additional seven regiments of PAVN main force units had moved south, where communist forces were already engaged in large multi-battalion battles with their increasingly feeble South Vietnamese adversaries. Johnson’s decision to escalate the war with American conventional forces, therefore, was not simply a reflection of deep-seated beliefs, but also a reaction to negative trends on the battlefield.
Saunders’s book is an impressive achievement – one that departs from “great men” theories of international politics and provides a clear and compelling theoretical framework for understanding the intervention choices different leaders make. Yet it remains unclear whether leaders’ beliefs shade interventions in subtle ways in permissive international environments, or whether they can induce leaders to ignore external constraints altogether and embrace disastrous interventions they would be wise to avoid.
Review by Rose McDermott, Brown University
Elizabeth Saunders’ book Leaders at War: How Presidents Shape Military Interventions offers an extremely well-written, clearly presented, carefully researched examination of the way in which leaders’ threat perceptions affect both their propensity to engage in military intervention and also shapes the type and extent of intervention in which they engage. Using extensive archival data, she examines Dwight Eisenhower, John F. Kennedy and Lyndon B. Johnson’s decisions to intervene as well as cases of non-intervention, noting the importance of the dogs that did not bark in the night to influence our understanding of the entire domain of choice.
The importance of the book, of course, lies in the central role played by individual leaders in the analysis, a position rarely taken by scholars of international relations outside those who, as she puts it, draw on psychological theory. In this regard, Saunders’ work takes seriously the admonition of Daniel Byman and Kenneth Pollack to bring the leader back into the core of theorizing about important international outcomes. Saunders does an admirable and self-conscious job of both controlling for those factors, as well as considering the alternative explanations for military interventions which tend to be privileged by most scholars of international relations, including structural/material considerations and domestic competition.
As one prone to engage those very psychological models which tend to be excluded by most international relations scholars, and as one who also has worked extensively in the archives of American presidents, I found a couple of the theoretical aspects of the book surprising. First, Saunders develops what she describes as:
a simple typology of leaders. At least in terms of threat perception, leaders do not come in infinite varieties bur rather can be usefully categories into one of two ideal types. Leaders’ causal beliefs about the origin of threats systematically influence decisions to intervene by alternative the cost-benefit calculus of intervention itself and by shaping the tools available to states when they undertake military intervention (212).”
It remains unclear, however, how this kind of research into the American presidency relates to other influential and simple categorizations of leaders, such as that put forward by James David Barber (1972), which classified leaders according to their positive or negative approach and their passive or active style of leadership, or the less well known but directly relevant categories put forward by Lloyd Etheredge (1978) based on whether they were introverted or extroverted and whether they were high or low dominance. Saunders does not really engage prior typologies of leaders to explain how her model either disputes or builds on such models.
Indeed, given that Saunders puts forward a model that largely depends on the beliefs of leaders, she provides surprisingly little explication for the psychological basis of such cognitions. It seems odd that in a book designed to celebrate the critical role of individual leadership in crucial decision making, almost all previous work examining the psychology of leadership is set aside from the outset. Some of these choices seem particularly strange given the value of that prior work. To pick simply the most glaring omission, Blema Steinberg’s magisterial contribution examining the role of narcissism in driving decisions to intervene in Vietnam on the part of Dwight Eisenhower, Lyndon Johnson and Richard Nixon seems a particularly unfortunate oversight. In justifying this choice, Saunders writes:
There is also a rich individual-level tradition that draws on psychological theories. These theories highlight factors that may produce error or bias in the way individuals father information or make decisions, such as misperception or the analogies and schemas that help decision making make sense of a complex world….I do not focus on error or bias in threat perception of information processing, or differences in the way leaders carry out policies, but rather concentrate on how their substantive beliefs shape perceptions of threat.” (29)
It seems counterproductive to dismiss the very arguments whose main thrust work only to support the central contention of this work, namely that leaders matter. Saunders may not want to engage literature which treats inferential biases as a universal characteristic of human nature, wherein all brains are assumed to be wired to respond in an identical fashion. However, not even the most extreme evolutionist would ignore the role of individual variance, particularly as substantiated in embodied manifestations. Indeed, important work on the nature and function of biases does not demand a clear separation between universal mechanisms and individual variance. For example, Philip Tetlock’s notable and important work does not assume that all biases affect all people, or affect them in a similar manner; indeed, his careful work documents the expression of reliable individual differences. If Saunders wants to make the case that biases matter in ways that can be systematically described, and thus serve predictive utility, she need not abandon all work which outlines the idiosyncratic role which individual variance can exert on decisions where leaders have some leeway for choice. The value of any explanation which privileges richness over rigor, as all individual level analyses do almost by definition, is that they highlight subtle aspects of causality, such as beliefs, that might otherwise be lost in a more blunt approach.
I was quite confused when I first read this, thinking that Saunders was trying to have it both ways, invoking psychological notions of beliefs as causal explanatory variables without having to address or include any previous substantive work in this area, including Robert Jervis’ critical work on the nature and function of beliefs, where he built on Smith, Bruner and White’s earlier work to distinguish the ways in which beliefs designed to test reality differ in important ways from those which serve social and psychological purposes. But over the course of the rest of the book, I came to realize that Saunders has a very delimited definition of the nature of beliefs, one which comes closer to what others might called an unmotivated or cold bias. But that again begs the question as to why she seems unwilling to engage the previous literature on bias in a more substantive way. It is not that her presentation is wrong or inaccurate, rather simply more limited than it need be in the audiences it can speak to or engage. It will be hard to win over readers who advocate realist beliefs (bias intended) for explaining the reasons for military intervention to a position which locates causal force in the individual. But similarly it will be hard to win over a crowd predisposed to believe in the role of individuals by dismissing and excluding previous work which directly address this phenomena from the outset.
However, once the limits of her consideration are accepted, and the role of individual leaders in shaping the nature and extent of military intervention is explored, the nature of her causal mechanism within that dynamic becomes salient. Specifically, Saunders locates individual variance in leadership in leader beliefs about threat perception. While this certainly constitutes a reasonable conjecture, Saunders does little to explain why she chooses this factor from among the myriad potential choices of leader variance to explain the source of divergence in outcome. What is it about threat perception, as opposed to drive for power, or desire for affection, or financial incentive or anything else that would expect this factor to drive decision about military intervention more than any other? The explanation for the critical selection of this important independent variable is never fully explicated. While a great deal of time and attention is given to explaining and justifying the case selection, and these selections seem satisfying, I was never clear as to why threat became the central focus of leader attention in driving decisions regarding military intervention.
By eliminating the likelihood that motivated beliefs provide causal leverage, perceived threat thus becomes the result and not the cause of policy preference. While an independent role for causal beliefs constitutes a strong point of the book, it does raise additional questions regarding the source of such beliefs. What if beliefs result from preferences rather than driving them? As a result, I was left wondering about the ontological origins of the beliefs Saunders explicates. While Saunders does an admirably meticulous job delineating the pre-presidential beliefs of each leader she exposits, she delves much less into the origins of such beliefs, and I was curious as to their foundation. Do these beliefs reflect dispositional differences in risk taking propensity, sensitivity to reward or propensity for fear, which may be present from birth and embodied in some kind of biological or genetic difference? Do such beliefs instead reflect differences in educational background and/or parenting styles? Or are they founded in some aspect of early political or personal experience, where the person learned some critical lessons about what does and does not work when trying to persuade others of the normative worth of one’s central values? I found myself wishing that Saunders had been more willing to speculate about the psychological foundations for the beliefs she finds so pivotal to explaining the aspects of military intervention she seeks to illuminate.
Saunders finishes her book outlining the ways in which her study contributes to scholarship, policy and contemporary debates and all her conclusions seem reasonable. Whether Saunders’ conclusions will generalize to improve our understanding of future decisions regarding military intervention will depend, in part, on observers’ abilities to ascertain leaders’ beliefs prior to action. But her fundamental point, and central contribution clearly deserve acclaim: Leaders clearly matter in definitive if ineffable ways in shaping all kinds of policy, including those most important ones related to the use of military force.
Review by Dan Reiter, Emory University
In Leaders at War: How Presidents Shape Military Interventions, Elizabeth N. Saunders proposes that the personal beliefs of American presidents shape American military interventions, specifically determining whether or not interventions seek to transform society. She proposes that structural/materialist explanations of whether an intervention seeks to be transformative are insufficient, and that beliefs need to be accounted for to make more precise predictions. She tests her theory mainly on military intervention decisions made by Presidents Dwight Eisenhower, John F. Kennedy, and Lyndon Johnson, with brief discussions of intervention decisions made by other presidents, including Theodore Roosevelt, Woodrow Wilson, and George W. Bush. She finds support for her central proposition that the beliefs of presidents determine whether or not interventions aimed to transform.
This book is squarely within a very exciting trend within contemporary international relations scholarship, that focuses on the behavior of individual leaders. In some ways, the focus on beliefs connects Saunders’ theory to older international relations scholarship, including work on operational codes, social psychological explanations of leader behavior, and learning theories focusing on leaders’ past experiences. Her work runs parallel to other leader-centered scholarship that uses alternative approaches, including formal approaches to leader behavior, scholarship focusing on variation in a leader’s domestic political institutions, psychological scholarship examining the effect of personality, emerging work on neuroscience and genetics, and others.
In this short essay, I explore Saunders’ independent variable, beliefs, and her dependent variable, whether or not an intervention is transformative. She posits that presidents may or may not believe in the necessity of attempting transformative change. Presidents who believe that threat comes from external factors are more likely to slough off the need for transformative change, and instead settle for non-transformative change. Conversely, presidents who believe that the internal characteristics of states determine levels of threat may accept the necessity of pursuing transformative change. Certainly, leaders’ beliefs are an important causal force in international relations. Structure is often too indeterminate a variable, and can be insufficient in making adequately precise predictions about leader behavior. Belief-based theories can help provide more precise predictions, helping us understand, for example, whether a leader will attempt to confront or appease an emerging threat.
Saunders’ findings push us to think about an important follow up question: where do beliefs come from? Saunders’ three main cases beg this question, as these presidents arrive at different beliefs about the necessity of transformation despite sharing similar backgrounds. Eisenhower, Kennedy, and Johnson all lived through and were deeply affected by America’s experience in World War II. That war culminated with a major American commitment to transform its defeated foes, Germany and Japan, a commitment that emerged from a belief that internal characteristics, such as Japanese hypernationalism and German Nazism (as well as Italian fascism), were taproot causes of Axis aggression. Further, the transformation of the former Axis nations into peaceful democracies was recognized by all observers as successful. If beliefs emerge from experience, then we might expect that all three would have similar beliefs. Yet Saunders posits that the three do not have common beliefs, as Eisenhower and Johnson prefer seeking non-transformative change but Kennedy prefers seeking transformative change. Political party does not predict this variance, as Kennedy and Johnson were Democrats and Eisenhower was a Republication. Military experience does not predict this variance, as all three served in the war, though with varying degrees of exposure to enemy fire. In short, the sources of the divergences of these three individuals’ beliefs remain a mystery.
Saunders’ dependent variable is more original than it might seem at first glance. Some past studies have explored whether or not a leader supports intervention. Others have explored the scale of intervention, and relatedly whether an intervener will pursue foreign-imposed regime change. However, Saunders is not interested in the scale of intervention or the goal of overthrowing a regime, but rather in attempts to transform a target’s social, political, and economic institutions. For her, transformative change is not necessarily comparable to major intervention or seeking regime change of the target. Transformative change can be small scale, such as Kennedy’s plans for intervention in Vietnam. Non-transformative change can be massive in scale, such as Johnson’s plans for intervention in Vietnam, and can seek to depose the targeted leader, as in the case of the 2003 invasion of Iraq.
The dependent variable calls for closer consideration. Specifically, we might explore exactly why it is important if an intervention seeks transformation or not. As noted, non-transformative interventions can be utterly massive, as in Vietnam from 1965 on and the 2003 Iraq War, and transformative interventions can be minor, as in Kennedy’s vision of intervention. Some might argue that only transformative interventions have a chance of successfully democratizing a target country, but studies indicate that attempts to impose democracy externally have a poor record of success. If we fear that transformative interventions may be destabilizing, the empirical evidence indicates that the key aspect of intervention determining how destabilizing it might be is whether or not it deposes the target leader, not whether or not it attempts intervention.
A last point. Upon finishing this book, the reader is left wondering about how the theory might fare in two tremendously important American interventions, World War I and World War II. A common interpretation is that the postwar peace following World War I collapsed because intervention was not transformative, as the Allies did not seek to root out the sources of German militarism and install robust democratic institutions. Conversely, the postwar peace following World War II was stable because the Allies successfully revamped German and Japanese societies.
These two cases immediately push us to ask how we should view the decision of whether or not to seek transformation, and relatedly the beliefs of Presidents Wilson and Harry Truman (and, perhaps, Franklin Roosevelt, as Roosevelt’s groundwork affected Truman’s postwar policies). Truman is probably easily judged as believing in the necessity of transforming Germany and Japan. Wilson in April 1917 publicly stated a belief in transformation, declaring that German autocracy and militarism were the taproot causes of the war and that the war was needed to make the world safe for democracy, but his private views of Germany were of course much more complex. Ultimately, the Allies adopted a non-transformative policy towards Germany in the Versailles Treaty. A complete case study could lay out whether the outcome of non-transformative intervention correlated with Wilson’s beliefs, or whether Wilson really desired transformative intervention, but was unable to accomplish it because of Allied and/or domestic political opposition.
The possibility of transforming threatening nations has resurfaced in the twenty-first century, especially in the wake of the 2001 Afghanistan War and the 2003 Iraq War. Political scientists and historians have begun to examine different dimensions of this kind of foreign policy initiative, in particular exploring the conditions under which such efforts have worked or would work. The academy also needs more scholarship building on Saunders’ book, helping us to understand the conditions under which presidents (or other leaders) might seek to transform other nations.
Review by Thomas A. Schwartz, Vanderbilt University
As an unreconstructed historian of American foreign relations, I have taken to approaching books by political scientists on the topic with great caution. Having to wade through insufferable jargon and/or complex mathematical equations about decision making leaves me frustrated. I begin to worry that I can no longer communicate with my disciplinary counterparts about shared interests and important questions. So for me this book was quite an intellectual treat. Not only is it exceptionally well-written and clearly structured, but it shows an extremely sophisticated use of recent historical work on both the Cold War and the three presidents – Dwight Eisenhower, John Kennedy, and Lyndon Johnson – used for its case studies. Saunders demonstrates an awareness and understanding of the historical debates that exist about these men, and uses both published primary sources as well as the most current historical scholarship. Her credibility is greatly enhanced by the fact that she is not selective in her sampling of historical evidence, and repeatedly references material that may not fit her argument. Her conclusion, that individual leaders and their ideas matter, is one that will find considerable support among those of us who study American Presidents and their foreign policies.
I wanted to begin this way because my criticism should be seen in light of how strongly I admire the book. Saunders argues that there are two approaches which American leaders apply to security threats and military interventions. One approach is to treat the threat as emanating from the internal make-up of the state, and to use military intervention to transform the institutions and society of that state. The other approach is to regard the threat as coming from that state’s foreign and security policies, and to limit any intervention to dealing with those issues alone, avoiding any “transformative” goals. Saunders makes a strong case that the ideas that Presidents bring into office about threats and the relationship of those threats to the internal order of states shape their approach to military intervention. Eisenhower and Johnson tended to focus on external threats and were not interested in the internal order of the countries where they intervened. Kennedy believed that threats arose from the domestic institutions of a society, and that these required transformation in order to eliminate that threat. Although all three of her case studies stem from the Cold War era, it is not hard to see the contemporary foreign policy significance of the Saunders argument. In the current discussion about dealing with states such as Iran, Syria, and North Korea, and with the continuing American engagements in Afghanistan and Iraq, whether and what type of military intervention would be in the U.S. national interest is an extraordinarily timely subject.
My reservations about the argument in this book stem from a different assessment of the relative importance of certain factors in shaping military interventions, as well as how significant the difference between leaders on this issue really is, especially compared with other issues. For example, Saunders argues that the Kennedy Administration’s approach to the Diem government in South Vietnam reflects the transformative view of interventions. To some extent she is certainly correct. The American pressure on Diem to broaden the political base of his government in order to win the war certainly reflects this. The distaste with which his repression of the Buddhists was viewed also underlines this aspect of Washington’s perspective. And Saunders is correct to point out that Lyndon Johnson as Vice President opposed the coup against Diem. But as she herself notes, Kennedy himself was very ambivalent about the coup, and a number of those close to him, including Robert Kennedy, opposed it. And even though Johnson’s subsequent decision to increase the number of U.S. troops can be seen as reflecting his emphasis on the external threat compared with internal reform in Saigon, it was also the case that the overthrow of Diem altered the conditions on the ground in the South, and led North Vietnam, as Mark Lawrence put it, “to strengthen the Southern insurgency in hopes of scoring quick battlefield victories that would bring the NLF to power before the United States could intervene more fully.” The situation within the country – relative chaos or stability – or the changing calculations of other external actors – in this case, North Vietnam – can have a greater effect on the nature of the intervention than a leader’s beliefs. (One could probably argue this as well for Iraq, where George W. Bush might have preferred a simple decapitation of the Saddam Hussein regime but ultimately found himself inextricably entangled in a nation-building exercise.)
Saunders herself acknowledges that her three case studies were all shaped by the environment of the Cold War, which at its heart was the conflict with the Soviet Union, a regime which Americans judged to be a threat, both because of its external behavior and its internal nature. The Cold War only came to an end when the Soviet Union was transformed, although the significance of the American role in that transformation remains a contested issue among historians. Whether or not the United States could actually live or “co-exist” with the Soviets was a central question during the Presidencies of Eisenhower, Kennedy, and Johnson. Eisenhower’s early years were marked by the rhetoric of rollback and its implementation, albeit by covert means, in Iran and Guatemala. Kennedy seemed to be continuing this with the Bay of Pigs invasion, but he embraced an easing of relations after the Cuban Missile Crisis. Although Saunders does show some similarities between Eisenhower and Johnson on issues of intervention, LBJ actually had more in common with Kennedy in his willingness to continue to promote détente with the Soviets, especially in the policies he carried out in Western Europe.
This leads me to the question of how significant ideas about the nature of threats were as compared to the domestic politics of foreign policy. Eisenhower’s New Look policy limited the possibility for conventional military interventions, but as the Soviet Union seemed to be advancing in nuclear weapons, this created room for a critique of “massive retaliation” as the strategy to defend the West. As it became clear that nuclear weapons deterred both the two superpowers, the competition for influence seemed to require new methods and ideas about intervention. In this environment, it really is not surprising that Kennedy and the Democratic Party latched on to the critique of American diplomacy embodied in the book, The Ugly American, with its suggestion that the United States, because of its cultural ignorance and failure to understand communist subversive methods, including guerilla war, was losing the struggle for Asia. (Kennedy was one of “a group of distinguished citizens” who sent the book to every member of the United States Senate.) Certainly this influenced the rhetoric of his inaugural address, in which he pledged to poorer countries “our best efforts to help them help themselves, for whatever period is required — not because the Communists may be doing it, not because we seek their votes, but because it is right.” What I’m suggesting is that Kennedy’s interest in a “transformative” agenda was in part a domestic political tool to distinguish his more activist approach to the Cold War from that of the Eisenhower Administration. Once in office, Kennedy’s reformist zeal waned. By 1963 he was becoming increasingly willing to accept military regimes in Latin America and was absolutely determined to prevent another communist government in the region. The difference between his and Johnson’s approach was something his liberal supporters like Arthur Schlesinger proclaimed, but I remain skeptical about its real significance.
Overall I think Elizabeth Saunders has identified an interesting and important difference in the way our leaders perceive threats. But watching the various changes President Barack Obama has gone thorough in defining American foreign policy, from an initial policy of engagement with Iran and Syria, to a current policy of sanctions toward Iran and calls for regime change in Syria, not to mention “leading from behind” in Libya, reminds us that Presidents are forced to adapt to a changing international environment as well as the vagaries of domestic politics. They may bring certain ideas into office with them, but the world may not cooperate.
Response by Elizabeth N. Saunders, George Washington University
I am grateful to the editors of H-Diplo for organizing this roundtable on Leaders at War. I thank the reviewers for their thoughtful and engaging comments, which have prompted me to think about the book’s argument in new ways. I will organize my comments around what I see as the main issues raised by the roundtable participants, concentrating especially on those that cut across several of the reviews.
First, the reviewers raise questions about the book’s central explanatory factor: leaders’ beliefs about the origins of threats. One important issue that arises in the reviews by Rose McDermott and Dan Reiter is the origins of these beliefs. I wrestled with this question as I began my research, but I concluded that explaining the origins of beliefs would represent a very different enterprise from the one I focus on here. The origins of beliefs are an important story in their own right, but the story is likely to be complex and multifaceted. One could imagine that causal beliefs about the origin of threats stem from prior military experience, education, or other aspects of a leader’s background. While it is possible that beliefs systematically track individual traits, at least in the cases I have examined, there are multiple independent paths to beliefs. I view these multiple paths as a strength of the theory, in that they show that threat perceptions are an independent and important source of leaders’ actions and are not simply a proxy for an underlying cause I have missed. I chose to bracket the origins of beliefs since my goal was to establish that these beliefs have crucial implications for how presidents approach the world, and act within it. I believe that demonstrating the importance of these beliefs is a valuable contribution, particularly since many scholars continue to either deny a role for individuals or despair of making systematic arguments about them. But I welcome further research on the origins of beliefs, and indeed other scholars are investigating whether traits like military experience or education systematically affect leaders’ behavior in international conflicts.
Of course, whether I am focusing on the appropriate set of beliefs is another matter. McDermott asks another important question about my main argument: why focus on threat perceptions, rather than some other aspect of beliefs? There are several reasons why the focus on threats makes sense in the context of explaining military interventions. First, many scholars have suggested that states intervene to protect national interests, and thus implicitly, when they perceive threats to those interests. Even when the “threat” appears remote in terms of its impact on the United States, as in the case of the Vietnam War, the reasons for intervention (though perhaps misguided) are usually connected to a perceived threat to national security, such as the danger of an ally’s collapse or the potential damage to U.S. credibility. While intervention in humanitarian contexts is often far removed from traditional conceptions of the “national interest,” even here we often hear security-based justifications relating to instability in certain parts of the world (e.g., the Balkans) or ‘spillover’ effects. Threat perception is thus intimately related to military intervention in a variety of international settings, from U.S. interventions in the Caribbean in the early twentieth century (which dealt, for example, with the impact of local instability on debt collection and the potential for European interference in the Western Hemisphere), to the Cold War, and the post-Cold War (whether one is focused on the rise of new transnational threats, threats from within other states such as terrorist havens, or external behavior like WMD proliferation).
Furthermore, although the categories I identify do not overlap perfectly with ‘realists’ and ‘liberals’ for reasons I discuss in the book, we have strong theoretical reasons to believe that there is an important distinction between those who see threats as arising from internal behavior (or theorists of the “second image,” as Kenneth Waltz put it) and those who do not. As the book discusses, this distinction informed understandings of threat going all the way back to the founding era in the United States, as seen in the debate over the posture the United States should take in the European war in the 1790s. Thus as realists often suggest, threats are important for understanding when and how states intervene—yet as Robert Jervis has emphasized, what constitutes a threat can be highly subjective.
McDermott also asks why the book does not directly engage psychological theories to a greater extent, tapping a larger issue about the nature of the beliefs on which I focus. She is right that I sought to establish the theoretical and empirical importance of what might be thought of as “cold” or unmotivated beliefs in shaping conceptions of threat and by extension, intervention choices. I made this choice not because I saw motivated or “hot” cognition as irrelevant to intervention decisions, but because I felt it was important to focus initially on unmotivated causal beliefs for two reasons. First, as suggested above, I saw the distinction between internally and externally focused threat perceptions as an important cleavage in beliefs. Additionally, since many international relations scholars deny the importance of individuals’ beliefs (obviously excepting those like McDermott who are experts on how cognitive factors affect world politics), it is important to establish that these beliefs are coherent, distinguishable, and relevant, before considering how motivated bias might also operate. But as I write in the book, I see important connections between the “cold” beliefs I identify and sources of motivated bias. For example, given a leader’s causal beliefs about the nature of threats, we can then ask whether an especially vigorous attempt to make policy investments based on those beliefs leads to motivated bias in perceiving whether those investments have been successful (perhaps leading to wishful thinking that the military has adapted to the leader’s investments) or even a perception that a given intervention will be easy or low-cost. I certainly see research that links my framework to psychological approaches focusing on motivated bias as a fruitful avenue for future work.
A final issue concerning leaders’ threat perceptions emerges from Paul MacDonald’s review, which highlights an interesting paradox in the argument: how can leaders be considered rational calculators of costs and benefits when they ignore evidence that contradicts their beliefs? Thus, why not consider the “reasonableness” of beliefs? While I had wrestled with variants of this question, I had not considered it in precisely the terms MacDonald uses and find his question compelling. I start from the assumption that leaders weigh the costs and benefits of intervening. Some inputs into the cost-benefit calculation will be shaped by “brute facts” like terrain, the distance of the potential target state from the intervening state, or whether the intervener is simultaneously involved in other conflicts. But other aspects of the cost-benefit calculation will be shaped by beliefs themselves—notably, how the leader values the benefits of transformation, and perceives the costs of intervening (since beliefs lead to policy investments, which affect preparedness). Admittedly this blends rational and biased perception, but leaders do not have biased perceptions of all aspects of the cost-benefit calculation—only those specified by the theory. Overall, I see beliefs as a thumb on the scale, rather than a determinative factor. Leaders do not ignore major elements of the situation, but also filter key inputs into the decision that are especially likely to be shaped by beliefs—namely, benefits and the costs associated with preparedness—through the prism of their threat perceptions. As discussed above, however, these unmotivated beliefs may in turn be associated with motivated bias in perceiving follow-on aspects of the intervention, such as how it is faring, with the result that leaders exhibit biased information processing once the intervention is underway. Such mechanisms may explain why leaders appear to be slow in learning from mistakes and updating their beliefs.
A second major issue raised by the roundtable concerns what the book is trying to explain. Reiter asks why I focus on transforming institutions, rather than whether or not the intervention removes the leader of the target state. One reason is that regardless of the overall scale of the intervention, transformation (in the sense of the intervener directly undertaking institutional change in the target) may require different capabilities than a nontransformative operation, even one that involves removing the leader. Another reason concerns the scope of my argument. While Reiter rightly highlights research on the long-term effects of regime change, my argument does not aim to explain the effects of transformation on the target state, but rather the intended policy choice in the intervening state. To be sure, the effects on the target state are very important, and scholars have highlighted the gap between stated goals of democratization and actual outcomes, which often fall far short of democracy. But explaining policy choice in the intervening state is an important, if not analytically prior, question that can have crucial consequences for the ultimate success of the intervention itself. The intervener’s level of preparedness—which I argue is shaped by leaders’ beliefs and policy investments—is important for intervention outcomes. In cases such as Iraq, where the initial goal was a ‘decapitation’ of the regime rather than thorough transformation, understanding the intervener’s initial intentions—as Leaders at War aims to do—is important for explaining why there was such a gap between ends and means when the strategy later shifted in a more transformative direction. The United States prepared for a quick, nontransformative intervention that would replace the top layer of leadership in Iraq but not involve nation-building. Once nation-building became the U.S. goal, it took several years for strategy and capabilities to catch up.
Finally, the roundtable participants ask whether I have duly accounted for the effects of other factors, especially in the context of the Cold War and the Kennedy and Johnson presidencies. Both MacDonald and Tom Schwartz bring up the change from Kennedy to Johnson and the issue of whether we should attribute changes in U.S. policy to the shift in administrations or to evolving external circumstances, particularly changes on the ground in Vietnam. This problem is challenging, but it puts the theory through an especially demanding test. Even as circumstances changed, we can see evidence of leaders’ beliefs. The evidence from Johnson’s vice presidency, when he disagreed with Kennedy’s approach, is particularly illuminating. While Schwartz rightly points out that external conditions can overwhelm a leaders’ initial choices—such as the North Vietnamese effort to score quick victories in the South before the United States could intervene more fully, or George W. Bush finding himself embroiled in a nation-building intervention when he hoped to avoid just such an enterprise—leaders’ beliefs and initial policy choices can leave a very important mark on the conduct of the war, with important consequences for the intervening state and its target alike. As noted above, in Iraq, the attempt to prosecute the war as a ‘decapitation’ affected the course of the war for several years. The gap between ends and means was a key element of the way the war unfolded, even if ultimately the United States moved to a more transformative policy.
Schwartz also argues that “Kennedy’s interest in a ‘transformative’ agenda was in part a domestic political tool to distinguish his more activist approach to the Cold War from that of the Eisenhower Administration.” This argument is intriguing, but Kennedy’s own statements and writings seem to suggest otherwise. Kennedy showed internally focused beliefs as early as 1946, and his internally focused view of the Third World became clear in his 1951 travels in the Middle East and Asia—all before Eisenhower officially got into the 1952 race. Furthermore, while I am highly sympathetic to the argument that presidents seek to distinguish themselves from their predecessors, it seems difficult to imagine that calling for more U.S. involvement in the Third World would be a highly beneficial domestic political move in the wake of the Korean War. More plausible is the idea that liberal elites would look kindly on a transformative agenda. But again, given American entanglements in the years prior to Kennedy’s run for president, advocating a transformative agenda does not seem like a necessary component of national political success.
The larger issue raised by Schwartz and MacDonald is how much we can attribute policy choices to leaders’ beliefs rather than other factors like the international environment or domestic politics. This question is a central concern of the book and one of the reasons why studying leaders is so difficult, despite the intuitive appeal of leader-based arguments. My solution was to look at the Cold War, when the international context was relatively fixed and threats were generally thought to be well-understood, and to restrict my analysis only to the United States, so that domestic institutions remained constant. But even within the context of the United States during the Cold War, there were undoubtedly shifting international and domestic political currents. No study of real world politics can eliminate this problem completely, but the book aims to show that we can trace the impact of leaders in systematic ways even as we acknowledge these shifts.
© Copyright 2011-2015 The Authors.
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 United States License.
 See Dale Andrade, “Westmoreland was right: learning the wrong lessons from the Vietnam War,” Small Wars & Insurgencies, 19, no. 2 (2008) 152-154.
 Daniel L. Byman and Kenneth M. Pollack, “Let Us Now Praise Great Men: Bringing the Statesman Back In,” International Security 25 (4) (2001): 107-146.
 James David Barber, The Presidential Character: Predicting Performance in the White House (Englewood Cliffs, NJ: Prentice-Hall, 1972); Daniel L. Byman and Kenneth M. Pollack, “Let Us Now Praise Great Men: Bringing the Statesman Back In”, International Security 25 (4) (2001): 107-146; Lloyd Etheredge, A World of Men (Cambridge, MA: MIT Press, 1978).
 Blema Steinberg, Shame and Humiliation: Presidential Decision Making on Vietnam (New York: Cambridge University Press, 1998).
 Philip Tetlock, Expert Judgment. How good is it? How can we know? (Princeton, NJ: Princeton University Press, 2005).
 Robert Jervis, “Understanding Beliefs”, Political Psychology 27 (5): (2006): 641-663.
 See, for example, Nathan Leites, A Study of Bolshevism (Glencoe, IL: Free Press, 1953); Deborah Welch Larson, Origins of Containment: A Psychological Explanation (Princeton: Princeton University Press, 1985); Yuen Foon Khong, Analogies at War: Korea, Munich, Dien Bien Phu, and the Vietnam Decisions of 1965 (Princeton: Princeton University Press 1992); Dan Reiter, Crucible of Beliefs: Learning, Alliances, and World Wars (Ithaca: Cornell University Press, 1996).
 H. E. Goemans, War and Punishment: The Causes of War Termination and the First World War (Princeton: Princeton University Press, 1995); Bruce Bueno de Mesquita, Alastair Smith, Randolph M. Siverson, and James D. Morrow, The Logic of Political Survival (Cambridge: MIT Press, 2003); Giacomo Chiozza and H. E. Goemans, Leaders and International Conflict (Cambridge: Cambridge University Press, forthcoming 2011); Scott Wolford, “The Turnover Trap: New Leaders, Reputation, and International Conflict,” American Journal of Political Science 51 (October 2007): 772-788; Maryann Gallagher, “Who Ups the Ante? Personality Traits and Risky Foreign Policy,” Ph.D. diss., Emory University, 2010; Stephen Peter Rosen, War and Human Nature (Princeton: Princeton University Press, 2005).
 Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976); Judith Goldstein and Robert O. Keohane, eds., Ideas and Foreign Policy: Beliefs, Institutions, and Political Change (Ithaca: Cornell University Press, 1993).
 William Easterly, Shanker Satyanath, and Daniel Berger, Superpower Interventions and Their Consequences for Democracy: An Empirical Inquiry (Washington: Brookings, 2008); Bruce Bueno de Mesquita and George W. Downs, “Intervention and Democracy,” International Organization 60 (Summer 2006): 627-649.
 Goran Peic and Dan Reiter, “Foreign Imposed Regime Change, State Power, and Civil War Onset, 1920-2004,” British Journal of Political Science 41 (July 2011): 453-475.
 Mark Atwood Lawrence, The Vietnam War: A Concise International History (New York: Oxford University Press, 2008), p. 81.
 Thomas A. Schwartz, Lyndon Johnson and Europe: Ion the Shadow of Vietnam (Cambridge: Harvard University Press, 2003).
 William J. Lederer and Eugene Burdick, The Ugly American (New York: Norton, 1958).
 John F. Kennedy, “Inagural Address” (delivered 20 January 1961), American Rhetoric, transcribed by Michael E. Eidenmuller, http://www.americanrhetoric.com/speeches/jfkinaugural.htm (accessed 14 February 2012).
 See, for example, Michael C. Horowitz and Allan C. Stam, “How Prior Military Experience Influences the Future Militarized Behavior of Leaders,” Unpublished Manuscript, University of Pennsylvania, August 2011.
 See the discussion and notes on p. 28 of Leaders at War.
 See Kenneth N. Waltz, Man, the State, and War: A Theoretical Analysis (New York: Columbia University Press, 1959), chap. 4.
 Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976), 19-20.
 For a recent debate on this issue, see the forum surrounding Alexander Downes’ argument about the effectiveness of regime change. See Downes, “Regime Change Doesn’t Work,” Boston Review 36, no. 5 (September/October 2011), 16-22.