This roundtable debates ideas and evidence in Diane Pfundstein Chamberlain’s recent book, Cheap Threats; Why the United States Struggles to Coerce Weak States. Pfundstein Chamberlain’s book considers the important puzzle described in the title, and in doing so puts forth a surprising new theory of coercive diplomacy. The reviewers praise some aspects of the book, but they raise concerns about both theory and evidence. Pfundstein Chamberlain responds comprehensively to the critiques. The debate is particularly interesting because the reviewers, though all experts in coercion and/or deterrence, approach the book from quite different angles.
H-Diplo | ISSF Roundtable, Volume X, No. 21 (2018)
Editors: Diane Labrosse and Seth Offenbach
Web/Production Editor: George Fujii
Dianne Pfundstein Chamberlain. Cheap Threats: Why The United States Struggles to Coerce Weak States. Washington, D.C.: Georgetown University Press, 2016. ISBN: (hardcover, $98.95); 9781626162815 (paperback, $32.95).
Published on 21 December 2018
© Copyright 2018 The Authors
This roundtable debates ideas and evidence in Diane Pfundstein Chamberlain’s recent book, Cheap Threats; Why the United States Struggles to Coerce Weak States. Pfundstein Chamberlain’s book considers the important puzzle described in the title, and in doing so puts forth a surprising new theory of coercive diplomacy. The reviewers praise some aspects of the book, but they raise concerns about both theory and evidence. Pfundstein Chamberlain responds comprehensively to the critiques. The debate is particularly interesting because the reviewers, though all experts in coercion and/or deterrence, approach the book from quite different angles.
Cheap Threats argues that U.S. coercive threats in the era of hegemony often have failed to achieve their ends precisely because the U.S. has been so successful in minimizing the costs of war, human as well as financial. The use of a volunteer army, for example, lessens the costs of war that come from public opposition. Pfundstein Chamberlain illustrates the argument with a fictional state that threatens to strike a target with cruise missiles if the target fails to dismantle its nuclear reactors (22-25). She argues that if the threat is cheap, for instance because cost-minimization makes escalation less dangerous, the target is likely to believe that the coercer will implement it (send the cruise missiles), but unlikely to comply with the coercer’s demand to dismantle the reactors, doubting the coercer’s ultimate will to proceed to war. In the book’s language, the coercer’s “cheap threat,” which in this book means a threat to take an action that is cheap, has immediate credibility because it is inexpensive to implement, but – and this is the surprising part of the theory — it lacks ultimate credibility and effectiveness for the same reason. Costly signals are more credible, she argues, because “only a highly motivated actor will be willing to pay the costs associated with such a signal” (23).
The book has significant policy implications. While the U.S. has tried to lower the costs of war, and, Pfundstein Chamberlain argues, has succeeded, she suggests that doing so has undermined the ultimate credibility of its threats. This conclusion need not mean that cost-minimization is the wrong decision, but it does suggest a re-evaluation of the strategy.
The reviewers laud some of the book’s conceptual discussions. Each of them notes the distinction between immediate credibility, the credibility of the threat to take the specified action, and ultimate credibility, the threat to fight a long and costly war, as a useful innovation to the field’s thinking about crisis behavior.
Both Crawford and Stigler focus their critiques on Pfundstein Chamberlain’s arguments about what makes threats costly or cheap and her application of these ideas to the crises she studies. Crawford concentrates his review on the book’s argument that bringing together a coalition to support coercive efforts makes military action less costly and threats less credible. He notes that the process of recruiting allies carries costs, while going it alone also signals a lack of support and of political ability. Stigler raises a number of concerns about the book’s discussion of costs in particular cases. For example, the book argues that the use of conscription and the relative lack of reliance on outside contractors made the U.S. threats in the Cuban Missile Crisis costly, but Stigler points out that the threat of nuclear war would have overwhelmed these smaller costs.
To me, the reviewers’ critiques of the theory are the most fundamental; if the use of force “were truly cheap and easy for the U.S.,” Sechser writes, “why would an adversary ever doubt America’s willingness to do it?” Stigler makes a similar point. As Sechser notes, the most obvious explanation for the ineffectiveness of U.S. threats in its period of hegemony lies in Pfundstein Chamberlain’s definition of effective coercion. As she states in this roundtable, she is not trying to argue that the U.S. attempted to communicate resolve to employ more significant force when it failed to do so, at least not always. For example, when President Barack Obama threatened Libya in 2011, he threatened to enforce a no-fly zone, but also stated that he did not plan to send ground troops (124), an honest indication that the U.S. was unwilling to engage in major war.
Had Obama made a different statement, would Libyan leader Muammar al-Qaddafi’s assessment of U.S. long-term resolve have been different? Literature on the effectiveness of what often is considered “cheap” diplomacy (diplomacy that carries no direct cost, though it can have costly repercussions) suggests that yes, the content of the language that leaders use is crucial. From this perspective, when the threat to take major action is credible, it tends to have greater impact on the target not because the action would be more costly for the coercer, but because it would be more costly for the target and the target wants to avoid it. Of course, if Obama had used other language in the Libyan crisis, this would not necessarily have changed the outcome. In any case, the hypothetical threat of U.S. ground troops in Libya would have been a bluff, and my own work suggests that leaders will hesitate to bluff, since doing so can have reputational repercussions that make communication more difficult in the near future.
The past two decades have seen substantial new work in the field of international relations on both coercive and deterrent diplomacy. As this work proceeds, scholars profitably could pay attention to why they are separating deterrence and coercion in both theory and empirics, a practice by no means unique to the work under discussion here. Most existing theories fail to examine the differences that truly exist, such as the fact that the status quo favors the target, or deterrer, rather than the challenger, or coercer. The field should pay more attention to the differences in order to assess their impact on disputes. If that impact is minimal, theories of coercion should apply to deterrence and vice versa.
Dianne Pfundstein Chamberlain, Ph.D., is an Associate Research Scholar with the Arnold A. Saltzman Institute of War and Peace Studies, Columbia University. She was formerly an Assistant Professor at the University of Massachusetts, Amherst, and a Postdoctoral Research Fellow in the International Security Program at the Belfer Center for Science and International Affairs, Harvard Kennedy School.
Anne Sartori is Research Scientist at the Sloan School of Management and Visitor in the Department of Political Science at MIT. Her research interests include interstate communication and nuclear strategy and stability. Her current project studies the impact of civil-military relations on international conflict, and her previous work includes the book Deterrence by Diplomacy (Princeton University Press, 2005).
Timothy W. Crawford (Ph.D., Columbia) is an Associate Professor of Political Science at Boston College. He is the author of Pivotal Deterrence: Third Party Statecraft and the Pursuit of Peace (Cornell University Press, 2003), which was the winner of the 2003 Edgar S. Furniss Book Award. He is also co-editor, with Alan J. Kuperman, of Gambling on Humanitarian Intervention: Moral Hazard, Rebellion, and Civil War (Routledge, 2006). His research interests include coercive diplomacy and alliance politics, and his current project examines the strategy and politics of dividing alliances. His latest work in this vein is “The Strategy of Coercive Isolation,” in Kelly M. Greenhill and Peter Krause, eds., Coercion: The Power to Hurt in International Politics (Oxford University Press, 2018).
Todd S. Sechser is the Pamela Feinour Edmonds and Franklin S. Edmonds, Jr. Discovery Professor of Politics at the University of Virginia. His research interests include coercion in international relations, the political dynamics of nuclear weapons, and the psychological effects of political violence. He is coauthor of Nuclear Weapons and Coercive Diplomacy (Cambridge University Press, 2017), and his research has been published in International Organization, the American Journal of Political Science, International Studies Quarterly, the Journal of Conflict Resolution, and other journals.
Andrew L. Stigler is an associate professor in the National Security Affairs Department of the United States Naval War College. His book, Governing the Military will be released in July 2018 by Routledge.
Dianne Pfundstein Chamberlain’s Cheap Threats offers a new way of understanding recent American forays into coercive diplomacy and adds to our tools for thinking theoretically about coercion in international politics. Its central claim is two-pronged: the things that make it relatively cheap for a unipolar power (i.e., the U.S.) to use force against weak states (1) encourage it to try to compel them more often, and (2) make it harder for it to do so effectively. Its scope, then, covers one of the enduring questions of coercion research: why do the efforts of the strong to coerce the weak often fail? The bulk of this review will focus on and critique one argument stemming from the book’s larger theoretical framework: that the unipole’s allies in these compellence attempts figure among those factors that make them fail. Before getting to that, further overview of the book’s theory and concepts is in order.
Pfundstein Chamberlain’s inventory of things that encourage the unipole to more often (and less effectively) try compellence starts with structure: the unipole’s extraordinary advantage in military power relative to all others in the system allows it to throw force against weak states with some impunity (27). Freed from the moderating effects of anticipated balancing, the unipole tries to compel weak states more often, and in pursuit of a wider range of goals, than it would in a world of great power peers. Along with these overarching incentives come specific characteristics of the United States. It has developed military capabilities and a range of cost-minimizing strategies that allow it to inflict violence on adversaries with relatively low losses to its own military personnel and machinery. That same technology and doctrine also allows the U.S. to reduce the harm it directly visits upon innocent civilians when it uses violence—which helps to moderate political backlash in the wider system and at home.
To these observations and deductions, the author harnesses the logic of costly signaling theory, which holds that threats that are costly (for the coercer) to implement will be more effective than those that are not, because the costly ones “convince the target that the coercer is highly motivated to defeat it” (24). This leads to the book’s counterintuitive argument noted at the outset. To whit—those advantages that make it cheaper for the U.S. to implement military threats ex post make it harder to convert threats into the desired political outcomes “before force is used,” because a threat that is not costly for the unipole to implement “will not signal that [it] is highly motivated” to persist if the immediate threat fails (24).
Undergirding this novel argument is another important contribution of the book—one that is conceptual. That is the distinction drawn between immediate and ultimate credibility. The former refers to the extent to which the target believes the coercer will “execute its immediate threat.” The latter refers to whether the target believes the coercer is willing to “apply additional force” and “escalate” if the target “resist[s] after the immediate threat has been executed” (21). The distinction must be made because, to paraphrase Pfundstein Chamberlain, it is not necessarily true that a state that is willing to initiate what it expects to be a cheap exercise in compellent force is also willing to persist in a protracted and costly military conflict (11). It matters because, as the author makes clear, it may be easier to cultivate immediate credibility than ultimate credibility, especially for a unipolar power. And high immediate credibility will often not be sufficient to compel effectively. If the target believes a threat is credible in the first way but not the second, then it may defy the threat wagering that after “execut[ing] limited military action” the coercer will “abandon [its] objectives rather than escalate” against stubborn resistance (22). While this distinction between the two kinds of credibility is especially relevant to the context of unipolar compellence versus weak states, the applicability of it—and the analytical traction it offers—extends to other kinds of coercion and influence techniques.
The evidence arrayed in the three post-Cold War, U.S.-as-unipolar-compeller cases (Libya 2011, Iraq 1991, Iraq 2003), provide good reasons to believe that the main thrust of the argument is correct. That is: structural and capability advantages, and “cost minimizing” strategies of the United States have opened up a gap between the immediate and ultimate credibility of its compellent threats against weak states; and this helps explain why they are ineffective. Yet there is something curious in the set of conditions that are said to sap the ultimate credibility of these compellent threats. That is, the compeller’s allies, who are relegated to a cost minimizing role. The remainder of this essay will focus on this point. While it is a narrow one, it takes us to some critical issues in the theory and practice of coercive diplomacy.
Comparison across the three post-Cold War cases reveals little about the impact of cost-minimizing allies on the credibility and effectiveness of the unipole’s compellence attempts. In the 2011 Libya case, many U.S. allies committed up front to supporting the United Nations (UN)-authorized intervention: according to Pfundstein Chamberlain, that made the U.S. use of force “less costly” (127). In the 1991 Iraq case, the UN backed U.S.-led intervention was also backed by a wider and even more potent coalition of European and regional partners, which again, made the use of force “less costly” (164). In the 2003 Iraq case, by contrast, the U.S. lacked a concurrent UN mandate, and had “a remarkably limited roster of allies” (193) backing it, which according to the author, made the use of force more (“moderately”) costly (195). Nevertheless, despite this cost variation imputed from the theory, the case results did not vary—compellence failed across the board. Thus, we cannot discern any differential effect of the varied levels of allied support. There is also a paucity of within-case evidence—and what little there is remains quite fuzzy—concerning how target leaders thought about the relationship between allied support of U.S. threats and their credibility. In short, we cannot tell from the process tracing whether the targets saw the ultimate credibility of U.S. threats bolstered or undermined by allied support. It is possible that in these cases indications of allied support enhanced ultimate credibility, but that the other and more significant cost-lowering factors identified in each case overdetermined failure.
We should reconsider, then, the theoretical framing of allied support for compellent threats. Not because it is unreasonable to posit that supportive allies help to defray some of the pecuniary and political expenses of executing military threats, Nor because the unipole’s expectations along these lines won’t encourage it to try to compel more frequently. Rather, we should reconsider the theoretical framing because cost-savings may not be the main reason for recruiting allies to support coercive diplomacy, and indeed, may not be the result of doing so. These are grounds to doubt that the leaders of weak states facing U.S. compellent threats key-in on the possible economizing U.S. allies represent as a negative indicator of the ultimate credibility of those threats.
Indeed, when it comes to allies, this focus on cost-minimizing misses two important things. First, it side-steps a significant thrust of alliance and coalition theory that views recruiting and working with allies as a costly business. From this perspective, assembling a coalition to back one’s threats should be a costly, not a cheap signal. Second, the cost-minimizing frame excludes other important strategic effects that allied support can have on the ultimate credibility of compellence. Having a coalition mobilized to back compellence can convey to targets that there is a political context in place that reinforces the ultimate credibility of threats. I will expand on both of these points below, but first we should look more closely at how Cheap Threats theoretically frames the unipole’s allies.
In Pfundstein Chamberlain’s theory, the unipole’s allies add little or nothing to the political and military punch behind its compellent threats. That is because “the unipole is, by definition, capable of executing military action around the world without assistance” (32). But there are drawbacks to “acting unilaterally”: it “is more expensive in financial terms and may be more likely to anger the public both at home and abroad” (32). So the unipole rounds up allies in order to “reduce the financial and political costs of executing” the threats.” In sum, “the logic of costly compellence asserts that the purpose of securing support from allies is not to enhance the overall effectiveness of US military action, but rather to redistribute the costs of an operation” (194).
Here Pfundstein Chamberlain channels one of two main ways of thinking about allies that have congealed in alliance politics and security studies. The “redistribute cost” motive aligns with the schema that sees allies as devices for gaining power or security cheaply. Thus, there is a nice affinity between her cost-minimizing-allies thesis and the concept of “substitutability” of arms and allies. In this schema we also find the concepts of “buck-passing” and “free-riding” within alliances. And we can find too the image of entrapment, in which a country manipulates its ally into expending blood and treasure to secure its parochial interests. The alternative schema, about which more will be said below, sees allies not as cost-savers but as costly means for gaining power and effectiveness against adversaries.
With the unipole’s allies defined as cost-saving devices, the author proceeds to argue that the targets of compellence will see the unipole’s allies in the same terms and thus as a sign of weak ultimate credibility. Two disturbing implications of this thesis come quickly to mind. The first is that all the diplomatic heavy-lifting that goes into forging a compellence coalition is actually counter-productive because, all else being equal, the unipole is in a weakened bargaining position when backed by many allies. The second is that going into a compellence crisis without allies is strong signal of resolve and thus ultimate credibility. A moment’s reflection suggests that while doing so may signal a certain determination, it will also reveal something about the unipole’s isolation if not political ineptitude, which cannot be good for ultimate credibility.
The alternative way to think about the relationship between allies and the costs of coercion highlights the costliness of allies. In this perspective, to work with allies and coalition partners means incurring “autonomy” and “policy coordination” costs. And recruiting supporters to back one’s coercive diplomacy means expending side payments and concessions in exchange for their backing. As Scott Wolford puts it: in “the strategic interaction between a would-be coalition builder and a potential partner [the latter] must be compensated for its cooperation in the costly endeavor of crisis bargaining and possible war…building coalitions entails a fundamental trade-off between increasing the chances of success and making political concessions to secure the cooperation of potential partners.”
In this schema, organizing allies to support the issuing and implementing of threats is not a formula for compellence-on-the-cheap, it is a costly way to increase the chances of successful coercion. This costliness, we should add, is particularly likely to hold for a unipole seeking to compel a weak state: in the unipolar systemic context, there will not be strong general (balancing) interest pushing would-be allies into the coercer’s camp. Specific—and costly—incentives must therefore be laid on the table to garner their cooperation. A unipole that pays those coordination and compensation costs in the process of issuing the threat reveals something about its seriousness. The political costs of building a coalition are sunk whether or not the unipole follows through on the initial threat, and whether it persists if resistance continues after the initial blow. Moreover, the unipole’s allies will add to the pressures on it to persevere, for their credibility will also be damaged if it walks back from its commitment after the fighting starts. An uncommitted compeller is therefore less likely to pay the costs of organizing a compellence coalition. Here is the key point: insofar as this is true—or target leaders generally think so—enlisting allied support for compellence may be a costly signal of ultimate credibility.
This alternative conception of allies and costs makes at least one thing clear: When a unipole builds a coalition to support compellence it reveals a willingness to pay costs in order—perhaps—to avoid others. Whether it is ultimately cheaper, on balance, for the unipole to act with allies or without them is an open question rather than a safe assumption. Given that, it is a long stretch to assume that targets will necessarily interpret the building of a compellence coalition as a cost-shirking move that exposes the unipole’s weak ultimate resolve.
Moreover, there are ways in which a strong backing coalition can augment the ultimate credibility of a threat, which is really what matters. These things can increase the chance of successfully applying force even if the unipole does not need allied military capabilities to do so. And targets are likely to pay attention to them when assessing ultimate credibility. When targets contemplate defying the unipole’s compellent threats, they will consider whether they can expect support from other powerful actors after the fighting has started. If such support were to materialize, it could help sap the unipole’s resolve. Forging a coalition that directly or indirectly sidelines potential supporters beforehand, then, may disabuse the target of such expectations. The presence of a compellence coalition, then, is thus not just an indication that the unipole wants to shed costs. It is a standing mechanism for effectuating and sustaining the target’s political isolation—both because the allies themselves will not be likely to assist the target, and because they too may use their influence to limit the extent to which others do so. In sum, an assembled compellence coalition can improve the downstream political effectiveness of applied force if the target does not comply and threats are implemented. Once the fighting starts, a target that is well isolated will be more vulnerable to total defeat than one that is not. Targets that understand this broader political context at the outset will factor it into their beliefs about the ultimate credibility of the unipole’s threats. A weak or non-existent compellence coalition will encourage them to hold out; a strong one will encourage them to fold early, before the use of force.
As mentioned earlier, comparisons across the three post-Cold War cases cannot adjudicate this question of whether it is better for the unipole’s ultimate credibility to have more (or less) allies backing its compellent threats. Despite variation on this, the outcome was the same—compellence failed. Nevertheless, honing in on the pair of Iraq cases (1991 and 2003) does provide additional ground for pushing the point further, and so push it we shall. We are encouraged to do this by an interesting distinction Pfundstein Chamberlain draws between the two cases. Though both threats failed, one may have been more likely to fail than the other. She compares the evidence surrounding “the 2003 threat with [that surrounding] the one that the United States issued in 1991” to infer a “relative prediction about the likely outcomes” (196). Her conclusion is that Iraqi President “Saddam [Hussein] was more likely to resist [in 2003] than he had been in 1991” (200). Why? “The United States’ 2003 threat was in many ways less costly than the one it issued and executed in 1991” (196). Thus, “the costly compellence theory suggests that Saddam was more likely to resist in 2003 than he was in 1991” (206). Another way of putting this conclusion is to say that Saddam’s refusal to comply in 1991 is more puzzling than his refusal to do so in 2003.
The general surmise may very well be correct (I think it is). But the problem here is that the threat variable capturing the cost-effect of allies across the 1991 and 2003 cases (see Table 5.2, p. 164, and Table 6.2, p. 195), points in the other direction. That is: as framed by Pfundstein Chamberlain’s theory, it indicates that allied support lowered costs more in the first case than in the second. In 1991, the U.S. mobilized a “large coalition” which, according to Pfundstein Chamberlain, made U.S. military action “less costly” (164). In 2003, the U.S. had “few” allies supporting its action, some of its “major allies” were among the no-shows, and it lacked UN authorization, which alienated still others—all of this made U.S. military action more (i.e., “moderately”) costly (195). Looking only at its cost-minimizing-allies hypothesis, then, Pfundstein Chamberlain’s theory would predict that Saddam was more likely to resist in 1991 than in 2003: the big 1991 coalition would signal to Saddam low U.S. resolve, and the small 2003 coalition would signal to Saddam high U.S. resolve.
But—recall—Pfundstein Chamberlain argues the opposite was true. The author suppresses the contradiction by arguing that the 2003 threat was expected to be cheap to execute for so many other reasons that “the relative lack of allies in early 2003 [made] little difference in terms of the anticipated costs of the operation” (197-98). That this was the judgement of U.S. policymakers seems likely. The author’s argument, however, also implies that Saddam’s overall calculus discounted the U.S.’s lack of allies in 2003 in the same way. Hence, “because the 2003 threat would be cheaper [overall] to execute…the Iraqi regime was more likely to doubt its ultimate credibility than the 1991 threat, and it would be less likely to concede before the use of force” (197).
The contradiction between the theory’s expectations about the higher-cost signal of “few” allies in 2003, and the claim that Saddam “was more likely to resist [in 2003] than he had been in 1991” (200), can be resolved in another, I think more plausible, way. That is to conceive of coalition building as a costly signal that enhances ultimate credibility, because allies and partners are costly to mobilize, and because building a coalition—especially a large one—indicates that the political ground is prepared to carry threats through to completion if the target resists. To put this more concretely: an important reason why Pfundstein Chamberlain may be right that Saddam was more likely to defy the 2003 threat is that it was backed by a “remarkable limited roster of allies,” due in part to Washington’s unwillingness to incur the high costs of assembling a coalition as it had in 1991. That unwillingness to pay coalition costs would have undermined ultimate credibility, and so too would the prospect that was raised by that narrow coalition—that U.S. follow-through could be blunted, not just by direct resistance, by the actions of others.
This reframing helps to bring out the puzzle embedded in the author’s conjecture that compliance was more likely in 1991 than in 2003. Saddam’s defiance in 1991 was truly anomalous. It is strange that he did not capitulate before he was attacked in 1991, when U.S. diplomacy with allies and in the UN had so thoroughly isolated him that his only supporters were the likes of Cuba, Yemen, Sudan, Jordan, and the Palestinian Liberation Organization. Critically, France and Russia ranged on the side of the U.S.-UN coalition. This, of course, was not the case in 2003. As Cheap Threats notes, shortly before the war Saddam had stated that “France and Russia would be able to prevent the U.S. from launching a war against Iraq. These two countries had strong economic ties to his country and it would be in their interests to block U.S. action” (204). Even if they could not stop the initial attack, that France and Russia stood opposed to it would have encouraged Saddam’s broader beliefs that the U.S. attack would be limited. If the U.S. did try for regime change, political opposition from France, Russia, Germany, and other opposed bystanders might inhibit it.
The narrow focus of this response has left a great deal of the theoretical ground covered by Cheap Threats (and much that is good about it) unremarked. Nevertheless, in keeping with that focus, I will conclude by trying to distill two points from the forgoing discussion, one theoretical and one practical.
First, alliance theories tell us that recruiting allies and building a coalition can have both cost-incurring and cost-minimizing effects. It may not be possible to say a priori whether, on net, the coalition builder will (in its judgement, or our own), come out in the red or the black. But it is safe to say that a theory about the calculations of those who build compellence coalitions that focuses only on the costs-reducing (i.e., the benefits) side of the equation, is missing some very important factors pertaining to their resolve and ultimate credibility. Related to this is the question of what the target thinks a compeller’s coalition signals about its ultimate credibility. Even if economizing is one of the benefits the compeller hopes to get from amassing allies, it would be unwise to assume that targets will seize upon this as an indicator of weak will, while ignoring the other ways that a coalition can be both costly to build and help the compeller to enforce its demands.
Second, though it is not central to Cheap Threat’s theoretical construct and overall thrust, the issue of allies and how they impact the credibility of the unipole’s attempts to compel weak states is magnified when we turn to policy relevance. In the conclusion of Cheap Threats, the author directly addresses the question: what can the U.S. do to make its compellent threats more effective (222-223)? One option implied by the costly compellence theory would be for the U.S. to act alone. To be clear, the author does not recommend that course of action. But if we follow the cost-minimizing-allies logic, unilateral action would (like a war tax) force Americans to bear the direct economic costs, not mention the wider political costs, of executing military action (32). And that should send targets a more credible signal of ultimate credibility.
This implied guidance—that Washington should not build effective allied support for its compellent threats, because targets will count it against U.S. credibility—will offend the political instincts of most policy makers. It also suggests that practitioners of political warfare have been getting it wrong for a long time, burning political capital to isolate targets and recruit allies in a self-defeating effort to buttress their coercive diplomacy. There is much that is insightful about Cheap Threats and leads to a better understanding of coercive diplomacy, but on this point, I think the intuitive theory of practice reflects a more promising formula for success than the counterintuitive one. Nevertheless, that intuitive theory offers no answer at all to the brute facts of cases covered in Cheap Threats: in 1991 (Iraq) and 2011(Libya) the unipole’s targets were deeply isolated and confronted by overwhelming U.S.-led coalitions, and yet compellence failed. In the costly compellence theory advanced by Pfundstein Chamberlain’s Cheap Threats, one does find such answers, and no serious attempt to understand why the weak so often defy the strong in coercive diplomacy can ignore them.
Coercion and threats are core pillars of U.S. foreign policy. In just the last year, the United States has turned to coercive bargaining in an effort to: (1) convince Iran to renegotiate the 2014 nuclear deal, (2) persuade North Korea to relinquish its nuclear weapons, (3) compel China to adjust its trade stance, and (4) obtain international support for all of these gambits. Notably, none of these attempts has yet succeeded. How can the United States become better at coercive diplomacy?
Academic research has grappled with the problem of coercive diplomacy for decades. Scholars have long observed that coercive diplomacy is difficult, and that coercive threats must meet a variety of conditions in order to succeed. Yet, in recent years scholars have noticed that powerful states, especially the United States, have a surprisingly poor record of success in coercive diplomacy. Coercive diplomacy is always difficult, but why does it seem to be more difficult for states with the greatest military advantage?
Scholars have offered a range of explanations. For example, Daniel Byman and Matthew Waxman claim that the United States is too averse to civilian casualties, allowing adversaries to hamper U.S. military operations by putting civilians in harm’s way. Todd Sechser argues that coercive targets fight to defend their reputations, hoping to deter future challenges from powerful states that cannot credibly commit to self-restraint. Michael Allen, Benjamin Fordham, and Phil Haun assert that coercive failures can be explained by demands that are too large, while others argue that U.S. practitioners have a poor understanding of coercion theory and are overconfident in U.S. military superiority.
Beyond these explanations, the literature offers a lengthy list of additional reasons that coercive threats sometimes fail, including asymmetric information about capabilities and resolve, loss-aversion, and even emotional affect.
Dianne Pfundstein Chamberlain hopes to add to this list with her valuable book Cheap Threats: Why the United States Struggles to Coerce Weak States. Pfundstein Chamberlain begins with the same observation that has vexed scholars for more than a decade: the end of the Cold War left the United States alone atop the global hierarchy, but this status bump did not translate into more coercive success. U.S. military operations against Afghanistan, Serbia, Libya, Iraq (twice), and Syria in the past two decades all originated, in part, from failed compellent threats. How can we understand this disconnect?
The puzzle is familiar, but Pfundstein Chamberlain offers a surprising explanation. Her argument starts with a reasonable premise: toward the end of the Cold War, the United States began to take steps to minimize the human, political, and financial costs of using military force. These include: adopting an all-volunteer force; relying more on military contractors; using advanced technology such as precision airpower and unmanned aerial systems; enlisting allies to share the burden of military operations; and paying for wars with deficit spending rather than taxes. Whether by design or by accident, these measures “render the use of force relatively cheap for the United States.” (47).
But there is a hidden downside to America’s shrinking costs of war. Costs, Pfundstein Chamberlain reminds us, can actually be useful in coercive bargaining. If issuing a compellent threat were costless, then anyone would be willing to do it, and the act of making a threat would convey nothing about one’s credibility. But if there is “some cost or risk” involved, as Thomas Schelling argued, then less-serious coercers will avoid making threats. Pretenders will be filtered out, allowing the truly resolved to differentiate themselves. Threats become meaningful signals of intent. In short, signals derive their meaning from the costs and risks one incurs by sending them. Costly signals are more credible than cheap ones.
The problem, Pfundstein Chamberlain argues, is that compellent threats no longer entail significant costs or risks for the United States. The danger that a compellent threat will fail and escalate to an armed conflict is less worrisome today, given that the United States can enlist unmanned drones, private contractors, or allies to do its dirty work. But by minimizing the costs of war, the United States has inadvertently rendered its threats meaningless as signals of resolve. The very fact that its threats are cheap makes them uninformative and therefore ineffective. Pfundstein Chamberlain writes: “it may be relatively easy for the unipole”—that is, the United States—“to employ cheap instruments, but that is precisely why cheap instruments do not signal that the unipole is highly motivated to defeat the target state and why the threat to employ them does not induce compliance” (36). By contrast, she argues, “costly threats are ultimately credible.” (25).
This is a book ostensibly about U.S. foreign policy, but the stakes are in fact considerably larger. Cheap Threats challenges us to rethink the dynamics of military power in international relations. It suggests that military superiority has countervailing effects: stronger states can more easily win wars, but may be less able to avoid them. The implications are jarring. If the cost-minimizing effects of military superiority impede a state’s ability to send credible signals of resolve during coercive diplomacy, powerful states may find themselves having to resort to force more often. Measures to minimize the costs of military force may prove self-defeating in the long run, lowering the price of military operations but increasing their frequency.
While the argument is compelling, the book’s evidence nevertheless raises some doubts. Pfundstein Chamberlain aims to demonstrate that U.S. compellent threats were more successful during the Cold War, before the United States adopted its low-cost model of warfare (76-78). If true, this would suggest that compellent threats are indeed more effective when force is costly to use, as Pfundstein Chamberlain asserts. Examining the widely-used International Crisis Behavior dataset, she argues that U.S. threats succeeded at a much higher rate during the Cold War (55%) than afterward (25%).
Yet this pattern rests on a sample of just 19 cases of U.S. compellence (eleven during the Cold War and eight afterward)—too small to draw any firm conclusions. Moreover, Pfundstein Pfundstein Chamberlain is sometimes too generous in her assessment of U.S. compellent threats during the “costly threat” period of the Cold War, leading her to overstate the degree to which U.S. threats succeeded during that era. For instance, Pfundstein Chamberlain argues that Secretary of State John Foster Dulles made a secret nuclear threat against China in May 1953, successfully compelling the Chinese government to accept U.S. terms for an armistice to end the Korean War (85). Yet there is actually no evidence that Dulles’s rumored ultimatum was ever transmitted to Chinese officials—and even if it was, it did not coerce China into making any concessions that it had not already agreed to.
A second example of a questionable compellence success involves the “Black September” conflict of 1970, during which the United States deployed naval forces to the Mediterranean. Pfundstein Chamberlain argues that this move coerced Syria into terminating its armored invasion of Jordan (86). But the purpose of this deployment was deterrence, not compellence: the Nixon administration sought to prevent Soviet interference in the event that Israel joined the war against Syria. If U.S. leaders also hoped that the move might intimidate the Syrians, it failed: Syria’s tanks were halted not by fear of American force, but by widespread mechanical breakdowns and ferocious Jordanian resistance.
In a third case, Pfundstein Chamberlain points to the 1972 Linebacker II bombings of Hanoi and Haiphong, which were intended to coerce North Vietnam into accepting U.S. terms for a cease-fire (87). The North Vietnamese did indeed sign the Paris Peace Accords shortly thereafter, promising not to annex or invade South Vietnam. But the agreement was virtually identical to one they had already accepted three months earlier, and Hanoi vitiated it altogether two years later when it invaded and conquered South Vietnam. Furthermore, it is difficult to see how this case constitutes a successful compellent threat, given that the United States had to execute it to achieve its objectives.
Reclassifying these three cases revises the U.S. success rate during the Cold War down to 27%, nearly identical to its post-Cold War success rate of 25%. In short, Pfundstein Chamberlain’s theory of “costly compellence” does not square well with the historical record: American threats were no more successful when U.S. costs for using force were high. To be clear, I do not mean to argue that these cases are unambiguous compellence failures—they are anything but unambiguous. But their very ambiguity suggests that we should be skeptical that there was ever a golden age when U.S. compellent threats were more credible. The grim reality is that compellence has always been difficult for the United States.
More generally, there is something odd about the notion that low costs for war undermine U.S. credibility. If using force were truly cheap and easy for the United States, then why would an adversary ever doubt America’s willingness to do it? To be sure, Pfundstein Chamberlain takes pains to emphasize that it is America’s long-term resolve that adversaries doubt: in a series of case studies, she shows that Iraqi President Saddam Hussein (1991 and 2003) and Libyan dictator Muammar el-Qaddafi (2011) believed that the United States might use limited force at first, but then “drop a few bombs and go home,” as she puts it (208). But this does not fully resolve the puzzle: why would adversaries believe the United States lacks staying power, if military force is genuinely cheap for it to use? What would prevent the United States from seeing its threats through to the end, if the costs of doing so are trivial?
Pfundstein Chamberlain’s case studies point to an answer, but one that is somewhat at odds with her theory. She argues that it is so inexpensive for the United States to use force that it cannot credibly signal its resolve, even when its resolve is high. But the book’s case studies demonstrate that the problem for the United States is not that military force is too cheap to use; rather, it is not cheap enough. It remains prohibitively costly to conduct the kinds of long-term ground operations that would allow the United States to impose its will on most stubborn adversaries. The reason coercion failed against Libya in 2011, for example, was not that the United States was unable to communicate its willingness to kill civilians, endure casualties, and occupy Libya in order to compel Qaddafi to accept U.S. demands. Rather, it was that the United States was never willing to do these things in the first place, and Qaddafi knew it. Indeed, President Barack Obama himself declared that “the United States is not going to deploy ground troops into Libya,” removing any doubt about the limits of U.S. resolve (124). America’s signals were hardly muddled; rather, they were too clear.
This suggests that U.S. doctrine has a rather different signaling effect than the one Pfundstein Chamberlain asserts. She argues that America’s cost-minimization measures “erode the signaling properties of its compellent threats,” making it harder for the United States to reveal when its resolve is high (47). The evidence, however, shows just the opposite: U.S. signals have actually become increasingly clear over time. The problem is that the signal is one that the United States would like to avoid sending. When the United States goes out of its way to shield the public from the costs of war, avoid inflicting collateral damage, and mitigate risks to its own soldiers, it may inadvertently advertise to its adversaries that it is sensitive to these costs and that its motivation to prevail is weak. The issue, then, is not that U.S. signals have become “uninformative” and reveal too little—it is that they reveal too much (35).
As Pfundstein Chamberlain rightly notes, even if these cost-minimization tactics have adverse effects on U.S. credibility, that is no reason to abandon them (225). Avoiding civilian casualties and protecting U.S. soldiers are important endeavors in their own right. Yet Cheap Threats makes a valuable contribution to our understanding of coercion, highlighting the complex nature of signaling and military power in international confrontations. Pfundstein Chamberlain has written an essential reminder that in coercive diplomacy, even seemingly straightforward policy choices can have unexpected and multifaceted consequences.
States in crises, even rich states, are often seduced by the allure of a cheap victory. The lengthy (and ultimately, for the United States, unsuccessful) conflict in Vietnam is perhaps the most potent example of the allure of a simple strategic paradigm that governed many U.S. decisions in Southeast Asia in the 1960s and early 1970s: surely the weaker state must give way to the might and will of the United States.
This is the focus of Cheap Threats: Why the United States Struggles to Coerce Weak States. Powerful as the United States is, it finds itself attracted to the prospect of bullying smaller states to get its way. Why, then, does the United States often fail in efforts to get weaker states to do its bidding?
For Pfundstein Chamberlain, the problem is that smaller states are not fooled into thinking the United States’ resolve is matched by its power. Pfundstein Chamberlain argues that the United States inadvertently sends signals of limited resolve by issuing cheap and easy-to-make threats. In addition, aspects of United States military posture as it has evolved since the end of World War II—the elimination of conscription, use of military contractors, and investments in high technology that seeks to win wars easily—also make it easier for the United States to issue cheap threats.
There are many aspects of this work that serve well to advance and deepen our understanding of these complex and risky coercive strategies. In particular, Pfundstein Chamberlain offers an excellent and nuanced discussion of resolve (8-11). She distinguishes between the will to initiate conflict, the will to suffer in conflict, and the will to kill the enemy. Some studies of coercion have over-focused on the question ‘Is the threat to attack credible?’ But Pfundstein Chamberlain convincingly exposes how this emphasis on the decision to attack is insufficiently nuanced. Sometimes states will launch an easy and cheap attack, and then decide they are disinterested in a longer campaign—and sometime their adversaries predict this lack of commitment. “It is not necessarily true that a state that is willing to initiate military action that it believes will be cheap is also willing to persist in a protracted and costly military conflict” (11).
One novel element of this study is Pfundstein Chamberlain’s distinction between immediate credibility and ultimate credibility. A threat that possesses immediate credibility is one for which the target believes the coercing state will apply some limited use of force. But the decision on whether to be coerced revolves around the ultimate credibility to follow through—that is, the coercing state’s willingness to take action, and bear costs, over the longer term to see the military effort through to success. This distinction makes sense. Many studies of coercion have given insufficient attention to the fact that coercive strategies involve a sequence of decision, not simply the initial decision of whether to strike and how.
Pfundstein Chamberlain’s focus on the United States’ coercive failures is relevant, since the United States (and, potentially, other powerful countries) is likely to attempt to coerce weaker states in the future. The difference between the Cold War period and the years following points toward a possible trend. According to Pfundstein Chamberlain, the United States issued a compellent threat in eleven of the forty-nine crises during the Cold War, or 22.4 percent. After the fall of the Berlin Wall, this rate rose considerably. Between 1989 and 2007, the United States was embroiled in fourteen crises and delivered compellent threats in eight of them, a rate of 57.1 percent. The datasets in Chapter 2 may prove particularly useful, since they seek to assess the stakes involved in crises, the threat used (or not used) and the ultimate outcome, among other factors.
At times, however, there are occasions to wonder if the indices that are employed to assess the context of compellent threats are somewhat overvalued and perhaps reflexively applied. For example, two of the four contextual variables that are assessed as making the U.S. threats during the Cuban Missile Crisis “more costly” are the use of military conscription in the U.S. military at the time, and the lack of heavy reliance on outside contractors. As Pfundstein Chamberlain notes, the Cuban Missile Crisis is unique among her case studies, in that it involved the risk of nuclear exchange—an event that would seem likely to have overwhelmed the other “cost” factors under consideration (103). It appears that Pfundstein Chamberlain is arguing that the U.S. military’s practice of conscription played a role in Soviet leaders’ estimations of the Kennedy administration’s credibility, by making the threat more costly and hence more credible. This seems a stretch, given the stakes.
A further criticism of the employment of conscription as a measure of the cost, and therefore the credibility, of the threat is that this aspect of a nation’s military posture cannot be selected at the time of the crisis. A president certainly has the latitude to determine if vague or specific threats with or without deadlines are issued. It makes sense for targets of coercive efforts to then consider the nature of the threat when gauging the credibility of a threat.
But when a crisis occurs, states threaten with the military they have at their disposal—conscript or volunteer. They also may bluff with that same military. To paraphrase Secretary of Defense Donald Rumsfeld’s famous quote, you coerce with the army you have, not the army you might want or wish to have. (I should mention Pfundstein Chamberlain does mention on page 30 the possibility of signaling resolve in a standoff by adopting conscription during the crisis, but this would be an unusual and lengthy crisis.) Military conscription would seem a less reliable indicator than Pfundstein Chamberlain appears to suggest. It may alter the ‘cost calculation’ of the threat slightly in some situations, but it is likely to be a minor factor in many instances, and almost certainly a non-factor in the Cuban Missile Crisis.
Pfundstein Chamberlain presents four variables in her contextual assessment of U.S. threats, but it is unlikely that she implies that each variable is weighted a quarter of the total. Nevertheless, one gets an impression that the credibility signaling impact of having a conscript army is somewhat overstated in parts of the book.
It might also be noted that an all-volunteer force could convey a more credible threat in ways other than cost. Dan Reiter and Allan C. Stam argue in part that democracies are more effective in war because they field more capable and often all-volunteer forces, which fight more reliably. Threats made by nations with more capable forces would enhance the impact of the threat.
There are other aspects of commitment that would seem appropriate to consider for some of the cases. Regarding the 1991 Gulf War, one of her case studies, Pfundstein Chamberlain assesses the case to be a win for costly compellence theory because the factors the theory gauges—employment of cost-avoiding military technology and tactics, attempts to spread the cost and burden among allies, and others—suggest that the target (Iraq’s government) could have reasonably questioned President George H. W. Bush’s commitment to liberate Kuwait.
But there were other costly commitments Bush made in the political realm that should have signaled credibility. When Bush was presented with a proposal from General Norman Schwarzkopf on 31 October 1990 to send an additional army corps to the theater to generate a robust option for a ground offensive, he agreed immediately, saying “If that’s what you need [for a ground option], we’ll do it.” The sheer size of the U.S. force deployment, approximately 700,000 troops, should have signaled determination and intent to any receptive audience.
For Bush, this was both an expensive move and a costly commitment. Politically, he could not then decide to pull out the ground force without suffering a humiliating setback. But Pfundstein Chamberlain’s construction of costly compellence does not include this in the range of commitment considerations. Surely this commitment—political, strategic, and financial—bears more consideration than whether the U.S. military is a force of conscripts or of volunteers.
Why, then, did coercion fail in 1990-1991? Poor Iraqi decision-making, inaccurate strategic assessment, and a sycophantic government are three reasons that come to mind. In both 1991 and 2003, Iraqi President Saddam Hussein did a bad job of reading what should have been clear signs of resolve from the United States. Costly commitments were made, and the Iraqi leader failed to grasp their significance. That conclusion may not lead to a satisfying theory of coercion. It may also not lead to good policy guidance, since it is hard to know in advance when another state will misread even a highly credible coercive threat. But that may well be a better explanation for the failure of potent coercive threats against Iraq in both 1991 and 2003.
The Iraq 2003 case is perhaps the most troubling of those in the book. In 1990, the initial troop deployment was to protect Saudi Arabia, and so there is room to debate when Iraqi leaders should have perceived a clear ground attack option for the coalition. But Iraq threatened no one in 2002 and 2003. The deployment of over 100,000 troops was a clear signal of serious intentions. That deployment was considerably smaller than the 1991 deployment, but Iraq’s military was considerably weaker in 2003, and the coalition succeeded overwhelmingly against the stronger Iraqi military in 1991.
There is much room to argue that Pfundstein Chamberlain understates the factors that lent credibility to the 2003 threat, even while she notes some of them. The United States had massed a combined force of 145,000 troops on Iraq’s border, as she notes, and proclaimed Iraq to be part of an axis of evil. This deployment was wholly at America’s initiative, and not a response to threats to Saudi Arabia or recent actions by Iraq, a fact that should have signaled hostile intent.
Furthermore, there were strong reasons to believe that the events of 9/11 had profoundly shifted senior American leaders’ perceptions of how their country needed to use military force to secure American interests, which would also make the threat more credible. Pfundstein Chamberlain notes some of these indications, such as President George W. Bush’s warning that he had “made up [his] mind that Saddam needs to go,” and numerous statements along similar lines (183-4, 186). The Americans demanded regime change, a maximalist goal that left no room for bargaining, and a goal that sharply reduced the likelihood of avoiding war.
Yet Pfundstein Chamberlain assesses all of the contextual threat variables to suggest a “less costly,” and therefore not credible, threat. All four contextual variables assessed in this case are either questionable or probably irrelevant. First, the United States was the unipolar power, so it was easy to generate the threat—but shouldn’t the overwhelming power the unipole represents be more of a factor in gauging threat? Second, the United States did not have a conscript army, and this reduced the political risks of a long-term campaign. But if the Americans executed the rapid plan they envisioned, there would be no political costs in any event.
Third, the United States made use of military contractors, making the threat less costly and, for Pfundstein Chamberlain, therefore less credible. But this could be a sign of a wealthy and a (somewhat more) efficient military—factors that speak to military power, and consequently threat credibility. Fourth, the United States planned to employ Revolution in Military Affairs (RMA) technology and doctrine, which would drive down U.S. costs, therefore again making the threat less credible. It must be said this is one of the oddest aspects of the costly compellence argument: the more devastating a nation’s military technology, and the more likely that technology is to minimize the user’s casualties, the less credible the threat to use that technology.
A flaw in Pfundstein Chamberlain’s assessment of the Iraq 2003 case is how she assesses the “Payment” element of the threat variables in Table 6.2 (195). She assesses payment for any attacks to be “less costly” in part because “oil revenues projected to offset costs.” This assessment of “less costly” then forms part of the foundation of Pfundstein Chamberlain’s assessment that the threat to invade was cheap, and therefore not credible. But the United States was partly depending on conquering and exploiting Iraq’s oil fields to make the attack less costly. The claim that the threat to invade was not credible, then, is partly based on an assumption that the U.S.-led coalition would invade and conquer.
Why did Saddam Hussein fail to perceive the threat’s credibility? His dysfunctional government misunderstood the highly credible threat he faced, and prioritized nonexistent regional threats over the massing American-led force on his border. The Iraqi Perspectives Project, which Pfundstein Chamberlain cites, conducted interviews with many senior Iraqi leaders after the successful invasion in 2003. This project offers any number of reasons to fault the Iraqi decision-making and assessment of coalition capabilities. Saddam Hussein felt that only Iran was “positioned” to threaten an occupation of Iraq. The Iraqi government focused more on regional threats than the U.S.-led coalition, even as invasion was imminent. When asked what he had expected to be the likely course of the coalition invasion, the Director of General Intelligence replied “We were more worried about Turkey and Iran.” It was a severely dysfunctional and ill-led regime.
To reiterate, Cheap Threats is a significant contribution to the literature on coercion. To this reader, it seems particularly noteworthy for the nuanced examination of resolve in coercive efforts, and its exploration of how costly signals may play a role in some cases. Coercive strategies are certain to figure into future crises, and Cheap Threats is likely to be an asset when those crises arise.
A major lesson of Pfundstein Chamberlain’s study is that these strategies often bring about unexpected risk and set one on a slippery slope of unexpected commitments. She quotes Thomas Schelling’s warning that “to engage in limited war is to start rocking the boat,” that is, to undertake an action which one cannot fully control (27).
It is privilege to have my book Cheap Threats reviewed in this forum, and by three reviewers who engaged with the core arguments and evidence in such a thorough manner. I want to thank the reviewers and the chair, Anne Sartori, for their time and consideration, and the editors at H-Diplo, particularly Seth Offenbach, for organizing this roundtable.
Cheap Threats asks: why do weak states resist threats issued by the United States? The book’s answer is straightforward but counterintuitive: weak states resist the United States’ threats when the threats are cheap to issue and to execute. The threat to use cheap force is not a convincing signal that the United States is sufficiently motivated to achieve its objectives if and when the threat and use of cheap force fail to modify the target state’s behavior. The United States has adopted both general conditions—the end of conscription and movement to the all-volunteer force, for example—and specific strategies—like the exclusive use of air-power to coerce target states—to render the use of force relatively inexpensive, particularly in the post-Cold War period.
The theory presented in Cheap Threats attempts to take something we know little about in real time—the target’s decision calculus—and use factors that we can observe—U.S. behavior and military strategy—to come up with an estimate for how the target will interpret the United States’ threat. I maintain that Cheap Threats provides a compelling theory that is more useful than its competitors for understanding the conditions under which compellent military threats succeed or fail, and that the imperfect dataset it presents is the best way to test this theory empirically. The progress of time and the inevitable presentation of future coercive campaigns will only provide more evidence for future tests.
Cheap Threats makes a key distinction between the immediate and ultimate credibility of compellent threats: a threat is immediately credible if the target believes that the coercer (the United States) will follow through on the threatened action, and ultimately credible if the target believes that the United States is willing to escalate to the level and duration of force necessary to extract a brute force victory from a stubborn target state after the use of coercion fails. A threat may have high immediate credibility and be ineffective in modifying the target state’s behavior. Cheap Threats argues that the United States’ threats fail not because they lack immediate credibility, but because targets doubt that the United States is willing to do more than drop a few bombs and then evacuate. Target states resist not because they doubt that the United States is willing to use force against them, but because they doubt that the United States has the stomach for a long and costly campaign.
Andrew L. Stigler highlights this important distinction between immediate and ultimate credibility in his fifth paragraph, but he does not consider this key distinction in his assessment of the role of advanced technology in the costly compellence theory I advance in the book. Stigler writes, “It must be said this is one of the oddest aspects of the costly compellence argument: the more devastating a nation’s military technology, and the more likely that technology is to minimize the user’s casualties, the less credible the threat to use that technology.” Cheap Threats does not suggest that the adoption of technology, doctrine, and strategy associated with the Revolution in Military Affairs (RMA) makes the threat to use the technology less credible. In fact, it makes the opposite argument: the adoption of these strategies makes the use of force easier for the United States, which in turn lends high immediate credibility to its threats. The use of these strategies does, however, make the United States’ threats less effective in modifying the target’s behavior.
Todd S. Sechser raises a similar objection when discussing the Libya case: “why would adversaries believe the United States lacks staying power, if military force is genuinely so cheap for it to use?” Again, the key argument in Cheap Threats is that the immediate credibility of threats to use cheap force is quite high: targets do not doubt that the United States will use cheap force against them. What is lacking on the U.S. side is the political will to use that force for a sufficiently long period of time, transition from airstrikes to conventional ground campaigns, or to escalate to the level of force necessary to gain a total monopoly on the use of violence within the target state (by, for example, drafting an additional 1.5 million young Americans to field a truly effective occupation force).
While some scholars believe that “staying power” is a necessary and sufficient element for achieving political (not battlefield) victory in twenty-first century conflicts, the record of recent U.S. military adventures suggests otherwise. If cheap force were effective in large doses, then Iraq, Afghanistan, Libya (and why stop there—Yemen, and perhaps Syria) would be stable and functioning societies with no need for continued U.S. military stabilization efforts. No doubt target states can appreciate this reality. Cheap Threats does not stress this point, but it is an important part of the explanation for why cheap force is both highly usable and highly ineffective as a coercive instrument.
Sechser also takes issue with the way Cheap Threats codes several cases from the Cold War period. When he reverses the coding of three cases I score as effective compellence, the difference in threat effectiveness that I observe between the Cold War and post-Cold War periods seems to disappear. Even if we accept Sechser’s (non-systematic) recoding of compellent threats and concede that compellence was similarly difficult across these two periods, that does not mean that the cause of compellent threat failure was the same. Sechser’s assessment attempts to impose a homogeneity on the phenomenon of threat failure that is unwarranted by the evidence. Sechser does not note, for example, my finding that the target’s response in the event of threat execution differed dramatically across these two periods: continued resistance was the outcome in only 25% of cases during the Cold War, but the target continued to resist in the face of actual U.S. force in 83% of cases in the post-Cold War period (70). Clearly, there is something interesting going on. It would also be difficult to argue that the United States is employing force in the same way in 2018 as it did in 1950. Even if we accept Sechser’s recoding, it is still possible that compellent threats fail in the post-Cold War period because the use of force has become relatively cheap and that they failed for some other reason in the early decades of the Cold War.
Furthermore, there is nothing inviolable about the particular line I draw at 1990, except to give us a rough point of comparison. As I note on pages76-78, we could choose to compare the pre- and post-1973 periods to assess the impact of cheap force on threat effectiveness, since conscription ended in the United States in 1973. When we draw the line here, the findings are actually more consistent with the outcomes predicted by the theory: successful compellence in 75% of cases 1945-1973 and only 18% of cases post-1973. With a theory that looks at long-term changes in the employment of American force, it will always be impossible to pinpoint a single date at which the impact of these changes will produce a change in outcomes, and thus it is productive to look at the trends at different points in time.
This discussion about differences in coding highlights what I believe is the single greatest challenge in working on interstate coercion: data availability and interpretation. This may be one of the reasons that there is a relatively small amount of academic work on this phenomenon relative to its prevalence in international politics. In some cases (including those on which Sechser and I disagree) it is difficult to determine whether and how a threat has been issued, let alone determine why a target may or may not have modified its behavior in response to a threat (or in response to something totally unrelated to the threat). In the absence of direct testimony from the leadership of a target state about why the state altered its behavior, we must make do with the behavioral cues that we can observe, and do our best to classify those behavioral cues in a consistent and meaningful way. The dataset in Cheap Threats is also relatively small, rendering the use of regression techniques unconvincing even where they are nominally possible. One advantage of a medium-N project like the one presented in Cheap Threats, however, is that it allows us to pull back the curtain on the coding of cases in a way that is not always possible with larger datasets, where the biases of the human coder are more hidden behind the veil of the spreadsheet or regression output table.
There is one area, however, in which I think Sechser and I actually do agree: that the United States is, more or less, signaling accurately. This is one area in which Cheap Threats could have been much more clear. In his penultimate paragraph, Sechser notes that, “the issue, then, is not that U.S. signals have become ‘uninformative’ and reveal too little—it is that they reveal too much.” He takes issue with the Libya case, arguing that, “The reason coercion failed against Libya in 2011, for example, was not that the United States was unable to communicate its willingness to kill civilians, endure casualties, and occupy Libya in order to compel [Muammar el-]Qaddafi to accept U.S. demands. Rather, it was that the United States was never willing to do these things in the first place, and Qaddafi knew it.” Exactly. The threat to use cheap force and the information communicated by President Obama (124) signaled that the United States had no intention of using American forces to depose Qaddafi if he continued to resist; recall that in the end, rebel forces on the ground aided by a drone captured and killed the disgraced Libyan leader (139). My argument is not that the United States was inaccurately signaling a low level of resolve. My argument is that the threat to use cheap force can never be a convincing signal of high resolve, and I agree with Sechser that in this (and perhaps most) cases, the cheap threat was accurately signaling low U.S. commitment.
Crawford’s very thorough and well-researched review focuses on a different element of the case analysis presented in Cheap Threats. Although he does not dispute the central argument that the United States’ compellent threats fail because they are cheap, he takes issue with one of the factors that I associate with the use of cheap force: the involvement of allies. He points out that in three cases (Iraq 1990 and 2003, and Libya), this variable is scored differently, but the outcome is the same (failed compellent threat). Stigler has a similar problem with the use of conscription as a variable that affects the outcome of a U.S. compellent threat. The variables that Cheap Threats identifies as factors contributing to the costliness of U.S. military action, and thus the effectiveness of U.S. compellent threats, were not intended to be taken as individually significant. The process tracing in the case studies is an attempt to evaluate a type of “index of threat costliness” based on the factors identified in the theoretical chapters. It was not my intention to assert that any one of these variables—the RMA model, the use of contractors—could be a reliable, independent predictor of threat effectiveness. Cheap Threats should have been clearer on this point.
Crawford’s key dispute, however, is with the characterization of allies as cost-reducing for the United States. He cites the alliance literature to argue that assembling allies is a politically costly undertaking, even when these efforts fail. Cheap Threats argues that the recruitment of allies reduces the United States’ costs for using force, thereby reducing the effectiveness of its compellent threats. Crawford argues that the choice to assemble a coalition of allies should in fact signal a high level of motivation to target states because doing so is costly, thus rendering U.S. threats more effective.
Cheap Threats is an argument about how the threat to use force can or cannot be a signal of high motivation to target states. The key logic is that actions and strategies that make the use of force less costly to the United States undermine the ultimate credibility of U.S. threats, and hence render them ineffective in modifying target behavior. Even if it may be politically costly to recruit allies—and I am not convinced that this is necessarily true for the United States, as I will discuss below—it could still be the case that the actual involvement of allies in a military mission reduces the United States’ costs for carrying out military action. In fact, this is the point of military alliances like NATO. A recurring theme in American foreign policy in the post-Cold War period (and particularly under the Trump administration) is that the United States’ European allies should do more to shoulder the burden for their own defense so that the United States can spend less. Furthermore, in cases where the United States acts with a “ready-made” alliance like NATO, it is not clear that recruiting allies is really that costly. As I discuss in Cheap Threats, the history of the 2011 Libya intervention suggests that the United States’ European allies led the push for intervention. The United States participated in a back-seat role, passing full control of the operation to these allies within twelve days. The evidence does not suggest that the United States paid high costs to act with its allies against Libya; the evidence suggests the exact opposite.
Consider a more recent example. In July 2018 when newspaper headlines bashed President Trump’s callous treatment of Theresa May during his first visit to the United Kingdom, no one used the opportunity to express excitement because the United States would no longer have the pay the costs associated with maintaining that alliance. The critiques of President Trump’s behavior toward international allies suggest that it is the loss of allies that makes U.S. action more difficult and more costly, consistent with the logic advanced in Cheap Threats. Even if there are some costs associated with the recruitment of allies—think of the whirlwind courtship of U.S. allies before the 1991 Gulf War that I discuss on pages 153-155—I would argue that the net effect of allied involvement is to lower costs for the United States, as was indeed the case for the first Gulf War. Crawford’s thorough theoretical analysis holds intuitive appeal, but I am still convinced that the evidence shows that the involvement of allies reduces the costs of military action, and thereby reduces the effectiveness of U.S. threats.
Finally, the reviewers are all somewhat skeptical about the book’s treatment of the Iraq wars. I concede that these are both extremely challenging cases. They also seem to be cases on which everyone has a personal favorite theory, but few people have actually read the available papers and transcripts on Iraqi President Saddam Hussein’s decision making. I argue in Cheap Threats that the United States’ threat was more likely to fail in 2003 than in 1991 because, on balance, the threatened force package would have been much less costly for the United States to execute in 2003 than in 1991. In fact, the 2003 invasion was explicitly designed to be as low-footprint and low-commitment as possible, a true test for Rumsfeld’s vision of cheap and easy high-technology warfare. That these plans were inexcusably short-sighted does not change the fact that this is what the United States was threatening in 2003, having previously executed several operations against the Iraq leader that stopped short of ousting him and his regime. In this space, I cannot review the bulk of the evidence on Saddam’s decision-making that I present in the book, so I invite readers to take a careful look at the discussion I present on the primary sources we have on Saddam’s regime.
The reviewers suggest that there are better explanations for why Saddam resisted in 2003—for example, the fact that he was more worried about regional rivals, a worry that I acknowledge on pages 201-202. This explanation is not inconsistent with Cheap Threats: it suggests that Saddam assessed the United States’ threat (a threat issued by the country with the world’s most powerful military) and judged that it was less of a danger to his regime than were his neighbors. Why would that be the case unless he doubted that the U.S. threat to use force against him was serious, i.e., unless he doubted its ultimate credibility?
The reviewers do not mention the evidence I cite from the Iraqi Perspectives Project that Saddam consistently downplayed the importance of U.S. technology, and that he viewed the reliance on this technology as an indication that the United States did not have the will to defeat him in a costly battle—key supporting evidence for the Cheap Threats theory (203). Is the cheapness of the threatened American action the only explanation for why he resisted? No. But much of the evidence is in fact consistent with the argument that the use of cheap force signals to target states that the United States lacks the will to defeat them in a long and protracted campaign.
The fact that we can find idiosyncratic explanations for a target’s behavior in a particular case does not indicate that a general theory is not useful for explaining the patterns of threat failure that we observe in the post-Cold War period. The test of a model is not whether every single variable and every single outcome line up with the theory’s prediction in every single case. As Kenneth Waltz noted, “The question, as ever with theories, is not whether the isolation of a realm is realistic, but whether it is useful. And usefulness is judged by the explanatory and predictive powers of the theory.” The reviewers did not mention the fact that as part of the process tracing in the case chapters, I also looked at a few prominent competing theories of threat effectiveness. None of these competing theories can provide a convincing explanation for the record of threat effectiveness we observe in these cases or for the patterns of threat failure that we observe in the dataset as a whole.
For example, one of the most commonly cited explanations for threat ineffectiveness is the failure to follow through on past threats—as in, ‘we must follow through on threats today or else tomorrow’s threats will fail.’ I have written about this “reputation theory” extensively elsewhere, but in Cheap Threats I demonstrate that this theory is entirely incompatible with reality. One of the more surprising findings of my research was that the United States does not bluff in the realm of compellence: in all of the cases from 1990 to 2007 where the United States issued an explicit compellent threat and the target resisted U.S. demands, the United States followed through on its threat. In other words, the United States never failed to execute a compellent threat in the face of target resistance, but threats against Iraq and Libya (and others) still failed. The other commonly accepted theories similarly struggle to account for patterns in the historical evidence. The fact that the three reviewers are all silent on this evidence does not diminish its importance.
The use of threats to persuade another actor to change his behavior has been and will continue to be a prominent feature of international politics, and one that deserves our careful scholarly attention. I am extremely grateful for the opportunity presented by these three serious reviewers and by the editors of H-Diplo to explore such an important topic in such detail.
 The use of “cheap threat” to mean a threat to take a cheap action differs from the common use in the literature. Usually, a “cheap signal” is one that is not costly to give, so that a state could make a cheap threat to take a costly action like an invasion (perhaps a leader stating privately an intent to invade) or a costly threat to take that same action (perhaps coupling the statement with military mobilization).
 Michael Joseph, “Explaining Peace in a Complex and Uncertain World: Multi-Dimensional Preferences, Great Power Rivalry, Diplomacy, and Peace,” Chapter 3, Dissertation Manuscript, George Washington University, 2018, https://scholarspace.library.gwu.edu/etd/fq977t99m; Anne E. Sartori, Deterrence by Diplomacy (Princeton: Princeton University Press, 2005); Robert F. Trager, Diplomacy: Communication and the Origins of International Order (Cambridge: Cambridge University Press, 2017).
 Joseph, “Explaining Peace in a Complex and Uncertain World,” Sartori, Deterrence by Diplomacy, and Trager, Diplomacy.
 For overviews of and recent contributions to coercion theory see: Kelly M. Greenhill and Peter Krause, eds., Coercion: The Power to Hurt in International Politics (New York: Oxford University Press, 2018); Peter Viggo Jakobsen, “Coercive Diplomacy,” in Contemproary Security Studies, ed. Alan Collins (Oxford: Oxford University Press, 2013), 235-55; Branislav Slantchev, Military Threats: The Costs of Coercion and the Price of Peace (New York: Cambridge University Press, 2011); Robert Art and Patrick Cronin, eds. The United States and Coercive Diplomacy (Washington, D.C.: USIP Press, 2003); Daniel Byman and Matthew Waxman, The Dynamics of Coercion (Cambridge: Cambridge University Press, 2002); Lawrence Freedman, Strategic Coercion: Concepts and Cases (Oxford: Oxford University Press, 1998).
 “Effectively” means targets comply with demands to alter their behavior “before the use of force” (21).
 Robert J. Art and Kelly M. Greenhill, “Coercion: An Analytical Overview,” in Greenhill and Krause, Coercion, 17-19.
 This flows from an illuminating prior discussion that excavates and parses different meanings of the concept of resolve found in the strategic literature (7-13). For another elaboration see: Dianne Pfundstein Chamberlain, “Doubling Down on Reputation: Defining State Resolve and Why it Matters.” War on the Rocks, 12 October 2016, https://warontherocks.com/2016/10/doubling-down-on-reputation-defining-state-resolve-and-why-it-matters/.
 The author argues that the effect of this large difference in allied support was minimized, however, because the Bush Administration decided to deficit spend rather than tax to pay for the action, and tap into Iraqi oil revenues to help defray war expenditures (194).
 That the U.S. was assured the backing of at least some allies in all three cases lends some support to the first prong of Cheap Threats’ main argument—that U.S. decisions to try to compel weak states are encouraged by perceptions that allies will help defray the costs of implementing threats.
 To be clear, there is evidence in some of the cases that U.S. policymakers saw allies as, at least partly, a way to lower the costs of implementing threats. See 137-138, 153-55.
 Although her theory ignores this aspect of allies, Pfundstein Chamberlain does not. In the account of the 1991 Iraq case, for example, she notes: “it was [not] easy for the United States to build and sustain the coalition opposing Iraq.” Nevertheless, “the purpose of assembling this coalition was…to minimize the United States’ costs” (162).
 As the author puts it: “Such allies may do little to enhance the unipole’s battlefield effectiveness, but they can lower the financial costs of executing military threats if they contribute troops, equipment, or money to the unipole’s campaign. Acting in conjunction with allies can also lower the political costs from military action if the unipole’s population views the approval of the allies as an indication of the action’s ‘correctness’” (32).
 I would add that the logic also asserts that targets will see it this way.
 As Benjamin Most and Randolph Siverson suggest, “alliances will often be chosen to allow budgetary savings”: Benjamin Most and Randolph Siverson, “Substituting Arms and Alliances, 1870-1914,” in New Directions in the Study of Foreign Policy, ed. Charles F. Herman, Charles W. Kegley, and James N. Rosenau (Boston: Unwin and Hyman, 1987), 131-160, esp. 135. Also see: Michael L. Altfeld, “The Decision to Ally: A Theory and a Test.” Western Political Quarterly 37:4 (December 1984): 523-544; Benjamin Most and Harvey Starr, “International Relations Theory, Foreign Policy Substitutability, and ‘Nice’ Laws.” World Politics 36:3 (April 1984): 383-406; James Morrow “Arms vs. Allies: Trade-Offs in the Search for Security.” International Organization 47:2 (Spring 1993): 207-233.
 Kenneth N. Waltz, Theory of International Politics (New York: Random House, 1979), 164-170; Barry Posen, The Sources of Military Doctrine (Ithaca: Cornell University Press, 1984), 63-64; John Mearsheimer, The Tragedy of Great Power Politics (New York: W.W. Norton, 2001), chapter 8.
 Glenn Snyder, Alliance Politics, 20, 44.
 Patricia Weitsman, Waging War: Alliances, Coalitions, and Institutions of Interstate Violence (Palo Alto: Stanford Univeristy Press, 2014; Patricia Weitsman, “With a Little Help from Our Friends? The Costs of Coalition Warfare.” Origins: Current Events in Historical Perspective 2:4 (January 2009): http://origins.osu.edu/article/little-help-our-friends-costs-coalition-warfare; Nora Bensahel, The Coalition Paradox: The Politics of Military Cooperation, Stanford University dissertation, 1999. Also see: Snyder, Alliance Politics, 44; James Morrow, “Alliances: Why Write Them Down,” Annual Review of Political Science 3 (2000), 70; Altfeld, “The Decision to Ally,” 526.
 “US coalition-building involves intense negotiations: coalition partners need to be bargained into participating in a specific US-led coalition. In addition, on many occasions, coalition participants require side-payments to motivate their coalition contribution”: Marina E. Henke, “The Politics of Diplomacy: How the United States Builds Multilateral Military Coalitions.” International Studies Quarterly 61:2 (June 2017): 410-422. Also see Jesse C. Johnson, “The Cost of Security: Foreign Policy Concessions and Military Alliances.” Journal of Peace Research 52:5 (2015): 665-679; Sarah Kreps, Coalitions of Convenience (New York: Oxford University Press, 2011); Randall Newnham, “Coalition of the Bribed and Bullied? U.S. Economic Linkage and the Iraq War Coalition.” International Studies Perspectives 9:2 (2008): 182-200.
 Scott Wolford, The Politics of Military Coalitions (New York: Cambridge University Press, 2015), 52-53.
 Thus, for a unipole that, in a given case, manifestly does not need its established allies to augment its military capability, opting to go it alone may be a low-cost signal, because the marginal benefits of adding allied capabilities are outweighed by the policy coordination and compensation costs.
 Timothy W. Crawford, “The Strategy of Coercive Isolation,” in Greenhill and Krause, Coercion, 228-250.
 Also see Kreps, Coalitions of Convenience, 115, 133-134.
 When a unipole’s key allies, or other major powers, withhold support, ultimate credibility is undermined because the target can infer that the bystanders could become tacit supporters after the initial application of force. The bystanders might try to use their influence to curtail the coercer’s commitment to force, or even provide back channel support to the target in later rounds.
 This refusal to back down in a situation of near-complete isolation may have contributed to subsequent western perceptions that Saddam was too irrational and reckless to be deterred. The author presents evidence suggesting that Saddam: “may have believed that US strength in 1991 was largely a function of the support it received from its allies” (171); and “identified the support that the United States received from its allies—that is, the support that minimized the United States’ costs for employing force against Iraq—as a critical component of American strength in 1991” (172). It is not clear whether and how this perceived “weakness” in the U.S. posture translated into a low-cost signal of resolve.
 As Pfundstein Chamberlain summarizes the evidence of Saddam’s mindset, the Iraqi leader “believed by early 2003 that an attack was likely but doubted that the United States would take the war to Baghdad” (205); “he resisted the United States’ demands not because he doubted that the United States would launch the threatened action, but because he doubted that the United States had the motivation to incur the costs he believed would be necessary to overthrow his regime” (206); he “thought he could withstand another limited US attack” (207).
 According to Pfundstein Chamberlain, “Both allies’ contributions and deficit spending shield the unipole’s public from the financial costs of using force and make it easier for the unipole to execute military action” (32). She later argues that one way to make U.S. compellent threats more effective would be to require [U.S.] administrations to “increase taxes or tap an outstanding surplus” when they seek “to execute military action against a small state…By making Americans pay the direct financial costs of fighting, policymakers could better convince the target of a U.S. compellent threat that the United States is highly motivated to defeat it” (222).
 For example, see Barry M. Blechman and Stephen S. Kaplan, Force without War: U.S. Armed Forces as a Political Instrument (Washington, D.C.: Brookings Institution, 1978); Alexander L. George and William E. Simons, The Limits of Coercive Diplomacy, 2nd ed. (Boulder: Westview, 1994); Peter Viggo Jakobsen, Western Use of Coercive Diplomacy After the Cold War: A Challenge for Theory and Practice (New York: St. Martin’s, 1998); Daniel Byman and Matthew C. Waxman, The Dynamics of Coercion: American Foreign Policy and the Limits of Military Might (New York: Cambridge University Press, 2001); Robert J. Art and Patrick M. Cronin, eds., The United States and Coercive Diplomacy (Washington, D.C.: U.S. Institute of Peace, 2003); Todd S. Sechser, “Militarized Compellent Threats, 1918–2001,” Conflict Management and Peace Science 28:4 (2011): 377-401; and Kelly M. Greenhill and Peter M. Krause, Coercion: The Power to Hurt in International Politics (New York: Oxford University Press, 2018).
 Daniel Byman and Matthew Waxman, “Defeating U.S. Coercion,” Survival 41:2 (1999): 107-120; Byman and Waxman, The Dynamics of Coercion.
 Todd S. Sechser, “Goliath’s Curse: Coercive Threats and Asymmetric Power,” International Organization 64:4 (2010): 627-660; Todd S. Sechser, “Reputations and Signaling in Coercive Bargaining,” Journal of Conflict Resolution 62:2 (2018): 318-345; Todd S. Sechser, “A Bargaining Theory of Coercion,” in Kelly M. Greenhill and Peter M. Krause, Coercion: The Power to Hurt in International Politics (New York: Oxford University Press, 2018).
 Michael A. Allen and Benjamin O. Fordham, “From Melos to Baghdad: Explaining Resistance to Militarized Challenges from More Powerful States,” International Studies Quarterly 55:4 (2011): 1025-1045; Phil Haun, Coercion, Survival, and War: Why Weak States Resist the United States (Stanford, Calif.: Stanford University Press, 2015); Peter Viggo Jakobsen, “Pushing the Limits of Military Coercion Theory,” International Studies Perspectives 12:2 (2011): 153-170; Rob De Wijk, The Art of Military Coercion: Why the West’s Military Superiority Scarcely Matters, 2nd ed. (Amsterdam: Amsterdam University Press, 2014).
 For example, James D. Fearon, “Rationalist Explanations for War,” International Organization 49:3 (1995): 379-414; Art and Cronin, The United States and Coercive Diplomacy.
 For example, James. W. Davis, Jr., Threats and Promises: The Pursuit of International Influence (Baltimore: Johns Hopkins University Press, 2000); Gary Schaub, “Deterrence, Compellence, and Prospect Theory,” Political Psychology 25:3 (2004): 389-411.
 Robin Markwica, Emotional Choices: How the Logic of Affect Shapes Coercive Diplomacy (New York: Oxford University Press, 2018).
 Schelling first highlighted the distinction between deterrent and compellent threats; see Thomas C. Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960), 195-199; and Thomas C. Schelling, Arms and Influence (New Haven: Yale University Press, 1966), 69-91.
 The best exegesis of the broader trend of cost-minimization, especially in modern democracies, is Jonathan D. Caverley, Democratic Militarism: Voting, Wealth, and War (New York: Cambridge University Press, 2014).
 A rich literature in economics and political science centers on this idea. The foundational work of costly signaling theory is often considered to be Michael Spence, “Job Market Signaling,” Quarterly Journal of Economics 87:3 (1973): 355-374.
 Schelling, Arms and Influence, 150. Other examples include James D. Fearon, “Signaling Versus the Balance of Power and Interests: An Empirical Test of a Crisis Bargaining Model,” Journal of Conflict Resolution 38:2 (1994): 236-269; James D. Morrow, “Alliances, Credibility, and Peacetime Costs,” Journal of Conflict Resolution 38:2 (1994): 270-297; James D. Fearon, “Signaling Foreign Policy Interests: Tying Hands Versus Sinking Costs,” Journal of Conflict Resolution 41:1 (1997): 68-90; Branislav L. Slantchev, “Military Coercion in Interstate Crises,” American Political Science Review 99:4 (2005): 533-547; David J. Lektzian and Christopher M. Sprecher, “Sanctions, Signals, and Militarized Conflict,” American Journal of Political Science 51:2 (2007): 415-431; Matthew Fuhrmann and Todd S. Sechser, “Signaling Alliance Commitments: Hand‐Tying and Sunk Costs in Extended Nuclear Deterrence,” American Journal of Political Science 58:4 (2014: 919-935; Abigail Post. “Flying to Fail: Costly Signals and Air Power in Crisis Bargaining,” Journal of Conflict Resolution (June 2018), https://doi.org/10.1177/0022002718777043.
 A χ 2 test of this correlation yields p = 0.198, falling short of conventional thresholds of statistical significance.
 Dulles made the threat to Indian Prime Minister Jawaharlal Nehru, apparently hoping that he would convey it to the Chinese. Yet Nehru denied that he ever passed it along. See Rosemary Foot, “Nuclear Coercion and the Ending of the Korean Conflict,” International Security 13:3 (1988), 104. One scholar calls Dulles’ supposed ultimatum a “fable.” See Edward Friedman, “Nuclear Blackmail and the End of the Korean War,” Modern China 1:1 (1975), 90. Pfundstein Chamberlain acknowledges that such skeptics exist but includes the case nonetheless.
 See Todd S. Sechser and Matthew Fuhrmann, Nuclear Weapons and Coercive Diplomacy (New York: Cambridge University Press, 2017), 174-180.
 See Richard A. Mobley, “U.S. Joint Military Contributions to Countering Syria’s 1970 Invasion of Jordan,” Joint Force Quarterly 55:4 (2009): 160-167.
 Indeed, Pape argues that the bombings “made no substantial difference in the terms of the agreement.” See Robert A. Pape, “Coercive Air Power in the Vietnam War,” International Security 15:2 (1990), 140.
 Bratton nicely summarizes the view that Linebacker II should not be considered a coercive success: “It is true that in the short term the Linebacker strikes disrupted the North Vietnamese offensive and guaranteed the independence of South Vietnam during the American withdrawal. However, what the North could not accomplish in 1973 it accomplished in 1975, when Saigon fell. It seems a stretch to list the Linebacker campaign with the surrender of Japan in 1945 or the liberation of Kuwait in 1991 as a successful example of coercion. To do so would lower the standards to endorse policies that at best only buy time or save face, rather than secure desired outcomes.” See Patrick C. Bratton, “When Is Coercion Successful? And Why Can’t We Agree on It?” Naval War College Review 58:3 (2005), 112.
 As Pfundstein Chamberlain indicates, the target’s response to a threat should be classified as “concede” only if “the target modified its behavior in accordance with the United States’ demands, stated or implied, after the threat was issued and before the threat was executed” (283; italics added). Indeed, Pape argues that the strikes were effective not because they foreshadowed future punishment, but because they blunted North Vietnam’s conventional strategy with brute force. See Pape, “Coercive Air Power in the Vietnam War.”
 Zegart, for example, argues that the low costs and high sustainability of drone warfare make U.S. coercive threats more credible rather than less. Amy Zegart, “Cheap Fights, Credible Threats: The Future of Armed Drones and Coercion,” Journal of Strategic Studies (February 2018), https://doi.org/10.1080/0402390.2018.1439747.
 We might more properly call these actions “indices” of U.S. resolve rather than signals, as they reflect a revealed preference rather than a claimed one. See Robert Jervis, The Logic of Images in International Relations (New York: Columbia University Press, 1970), especially chapter 2.
 Indeed, Byman and Waxman argue that U.S. cost-minimization efforts inadvertently give adversaries a playbook for how to outlast the United States. See Byman and Waxman, “Defeating U.S. Coercion.”
 Though she does suggest reinstating the draft and financing military operations with current revenues only (222-23).
 Dan Reiter and Allan C. Stam, Democracies at War (Princeton: Princeton University Press, 2002).
 Bob Woodward, The Commanders (New York: Simon & Schuster 1991), 319-320.
 Kevin M. Woods, with Michael R. Pease, Mark E. Stout, Williamson Murray,, and James G. Lacey, Iraqi Perspectives Projects: A View of Operation Iraqi Freedom from Iraq’s Senior Leadership (Norfolk, VA: United States Joint Forces Command Joint Center for Operational Analysis, 2006), esp. Chapter 2.
 Kenneth N. Waltz, Theory of International Politics (Boston: McGraw Hill, 1979), 8.
 See for example, Dianne Pfundstein Chamberlain, “It is Time to Drive a Stake into the Heart of the American Credibility Myth,” War on the Rocks, 27 September 2016, https://warontherocks.com/2016/09/it-is-time-to-drive-a-stake-into-the-heart-of-the-american-credibility-myth/.