H-Diplo | ISSF Roundtable XII-6
Oriana Skylar Mastro. The Costs of Conversation: Obstacles to Peace Talks in Wartime. Ithaca: Cornell University Press, 2019. ISBN: 9781501732201 (hardcover, $39.95).
21 January 2021 | https://issforum.org/to/ir12-6
Editor: Diane Labrosse | Commissioning Editor: Avery Goldstein | Production Editor: George Fujii
- Introduction by Avery Goldstein, University of Pennsylvania.. 2
- Review by James D. Fearon, Stanford University.. 6
- Review by Todd H. Hall, University of Oxford.. 11
- Review by Xiaoyu Pu, University of Nevada, Reno.. 13
- Review by Kristopher W. Ramsay, Princeton University.. 16
- Review by Robert F. Trager, University of California, Los Angeles. 20
In The Costs of Conversation: Obstacles to Peace Talks in Wartime, Oriana Skyler Mastro tackles an undertreated and undertheorized topic in security studies—the decision to begin negotiations with an adversary during war. The result is an elegantly presented and persuasively argued theory that she applies to carefully considered case studies. For scholars, Mastro’s book will be a valued addition to the literature on security studies as well as theoretically informed work on diplomatic history. For policymakers, the book provides a straightforward logic to help untangle, clarify, and potentially correct mistaken intuition and conventional wisdom about the factors that shape an adversary’s willingness to enter negotiations during military conflict.
Although one might fairly argue that the Mastro’s work does not supplant what are several of the currently dominant approaches to understanding wartime negotiations, it clearly and convincingly complements such work by providing a theoretical argument to answer a prior question that falls outside the range of explanation for competing approaches—when and why negotiations begin. Existing literature about negotiations with an adversary, especially a growing corpus of work that draws on bargaining theory to illuminate the importance of uncertainty about information, commitment problems, and domestic audience costs that regimes face, has provided important insights about the steps leading up to military conflict, and about the conditions that may be conducive to war termination. Mastro focuses on the strategic calculus that determines if and when a belligerent will be open to talking to its adversary. The key consideration, according to Mastro, is the need to demonstrate strength before entering talks. Otherwise, the adversary will be tempted to interpret a willingness to negotiate as a sign of weakness—of capabilities or resolve—that it could exploit.
To test her arguments, Mastro re-examines three major international conflicts in Asia during the Cold War. Although these cases (the Korean War, the Sino-Indian War, and the American phase of the war in Vietnam) have each been thoroughly covered by journalists, historians, and political scientists, the author brings to light newly available evidence, including declassified documents, archival materials, and interviews, that both enrich the empirical record and help clarify the mechanism she sets forth in her theory about the costs of conversation.
The five reviewers in this roundtable are unanimous in their admiration for the project, even as they offer suggestions for clarification about some of its claims, propose alternate explanations for the patterns Mastro studies, and indicate the ways in which her work can provide a foundation for others to build on the questions she asks and the answers she offers. Our five contributors bring to bear the perfect mix of expertise to explore this book’s themes. James Fearon and Kristopher Ramsay have been important participants in elaborating work that applies bargaining theory to an understanding of conflict. Robert Trager’s work has focused much attention on the interconnected issues of beliefs, perceptions, and diplomatic communication. And Todd Hall and Xiaoyu Pu, are leaders among a new cohort of China scholars whose work reflects a strong grasp of contemporary international relations theory with an impressive depth of regional knowledge.
Fearon indicates that Mastro illuminates an important implication of the distinctiveness of the bargaining that occurs during warfighting. He briefly notes the links between other scholarship that examines conflict as a bargaining problem and Mastro’s central argument about the remarkable reluctance of adversaries to adjust their stance in response to changing facts and information that is revealed as fighting proceeds. In thinking about this puzzle, Fearon draws special attention to the significance of the initial war aims of the adversaries, especially in cases where the adversaries have extreme or absolute war aims. Such war aims exacerbate the familiar problems introduced by the dubious credibility of bargaining offers absent a third-party to enforce commitments. Fearon would like to see more from Mastro about how these considerations together contribute to the fear that escalating demands will follow from an offer to talk because it could be interpreted as a sign of weakness. He provides illuminating historical examples that clarify the way in which his suggestions could help build on the contribution of Mastro’s book.
Hall underscores the elegance of Mastro’s argument and the impressiveness of the sources she taps to provide evidence that demonstrate its logic in practice. Hall usefully suggests that Mastro’s focus on the costs of conversation may shortchange the role that strategic benefits could play in driving a decision to enter peace talks. And he notes a puzzle prompted by his reading of Mastro’s book as to why actors who are interested in negotiations fail to recognize how their posture and actions shape its adversary’s decision about whether to join talks. Although acknowledging the limits to making broad claims based on even a detailed examination of a small number of historically specific cases, Hall emphasizes that the book offers novel insights about decision-making in these important conflicts and recognizes their usefulness as a plausibility probe for the logic Mastro spins out.
Ramsay notes the gap in the otherwise rich literature on bargaining and conflict that Mastro’s work fills- the logic and timing of a state’s decision to “open the door to a diplomatic solution” to an ongoing war. Ramsay also directs attention to Mastro’s distinction between signaling resiliency and the more traditional focus on signaling resolve in the bargaining literature. Rather than focusing solely on the issue of limits to the generalizability of her argument, Ramsay lauds the stringent criteria that Mastro establishes for deciding whether the empirical evidence in her cases demonstrates the significance of the concern she emphasizes—that the adversary might view an offer to talk as a sign of weakness that it could exploit. Ramsay, however, is also left with a puzzle: if openness to negotiations could be interpreted as a sign of weakness, the problem would go away if all states simply had a standing policy of openness to peace talks in war—an idea Mastro also raises in the book. Ramsay, then, is calling for further research into how states get locked into the “closed diplomatic postures” that Mastro identifies as complicating the decision to open talks with the adversary during war. In this, he adds to Fearon’s suggestion about the possible relevance of initial war aims by suggesting that the war strategies that states choose may have implications for the diplomatic postures that are feasible as the fighting proceeds.
Pu shares his colleagues’ favorable view of the book’s theoretical innovations, empirical evidence, and policy implications. He echoes Ramsay’s observation about the possibilities for states to embrace peacetime policies that help avoid the “costs of conversation.” In this respect, like Ramsay, Pu endorses Mastro’s recommendation that states may want to establish as standard policy a willingness to negotiate during conflicts. Pu also sees merit in Mastro’s policy advice for resolving the potentially dangerous crises and conflicts in which the United States and China may find themselves in the coming years. In addition, he echoes Mastro’s warning about the tendency for strong countries to rely on coercion to bring adversaries to the bargaining table, an approach that, as Mastro’s work explains, may actually make an adversary fearful of signaling weakness and therefore make it less likely to agree to peace talks. Pu raises an additional intriguing question inspired by the book—whether the interest in signaling might spill over into peace talks themselves and provide states with incentives to engage in strategic deception.
Trager commends Mastro’s “compelling account of why and when states come to the negotiating table during wartime.” He sees the case studies as providing convincing evidence in support of the book’s central claims about the inhibitions induced by the fear that agreeing to talk will be interpreted as a sign of weakness. Yet Trager questions the scope of this argument beyond the cases that are treated in the book, as illustrated in the historical examples he cites. Trager indicates that increased pressure has in fact sometimes convinced an adversary which is under pressure and losing the conflict to open talks, an outcome that is at odds with Mastro’s expectations. In some cases, he argues, a weaker belligerent’s decision to talk under pressure may be part of its strategy to buy time. If so, it may seek a peace settlement knowing that it plans to resume fighting in order to alter its terms at a later date when it has increased its military power. Trager also offers two alternative explanations for the failure to initiate peace talks during conflict. In the first, Trager shares Fearon’s view that initial war aims and motivations may shape the willingness of the parties to open talks once fighting is underway. The second alternative suggests an approach to thinking about the role of third parties that is different from Mastro’s argument. She argues that this role is limited to powerful patrons who can exert decisive pressure on clients to open talks. Trager instead points to the potential significance of third parties who are important allies of one of the states, rather than powerful patrons. In such cases, the interest in managing an alliance may constrain decisions about opening peace talks or continuing to fight the war.
In her thoughtful response to the reviewers, Oriana Mastro clarifies her motivation for writing the book and briefly draws out its central arguments. She then addresses the questions and concerns the reviewers raise. First, Mastro elaborates on the reasons why states may fear that open postures signal weakness, including two sources that are distinct from the strategic calculations she emphasized in explaining the closed postures adopted by belligerents. Second, Mastro engages with the argument that benefits to be gained by entering negotiations, and not just the costs, might determine states’ choices. While acknowledging the logic of this view, she notes that she did not find historical evidence that anticipated benefits from negotiated settlements led to the initiation of talks and suggests reasons for this empirical observation. Third, Mastro addresses the reviewers’ point about the war aims of belligerents affecting the prospects for opening negotiations. She explains why her argument is relevant even in cases where states have absolute war aims, as well as in the much more common cases where, public rhetoric notwithstanding, they have limited aims. Finally, Mastro agrees with the reviewers that her work leaves some of their questions unanswered. She shares their hope that the book will prompt others to address remaining theoretical puzzles, further enrich the empirical record, and thereby “strengthen our understanding of how diplomatic strategies and military strategies are connected and impact one another.”
My brief introduction does not fully capture the range and subtlety of the many issues the reviewers include in their essays as they reflect on Mastro’s work and that she includes in her response. Their stimulating exchange of ideas serves to demonstrate why the Costs of Conversation is a major contribution to the literature about war and diplomacy, one that is noteworthy for the innovativeness of its argument and the excellence of its empirical research.
Oriana Skylar Mastro is a Center Fellow at the Freeman Spogli Institute for International Studies at Stanford University where her research focuses on Chinese military and security policy, Asia-Pacific security issues, war termination, and coercive diplomacy. She is also Foreign and Defense Policy Fellow at the American Enterprise Institute and an officer in the United States Air Force Reserve for which she works as a strategic planner at INDOPACOM. For her contributions to U.S. strategy in Asia, she won the Individual Reservist of the Year Award in 2016. She has published widely, including in Foreign Affairs, International Security, International Studies Review, Journal of Strategic Studies, The Washington Quarterly, The National Interest, Survival, and Asian Security. Her book, The Costs of Conversation: Obstacles to Peace Talks in Wartime (Cornell University Press, 2019), won the 2020 American Political Science International Security Section Best Book Award (Pre-Tenure Faculty Member). She holds a B.A. in East Asian Studies from Stanford University and an M.A. and Ph.D. in Politics from Princeton University.
Avery Goldstein is the David M. Knott Professor of Global Politics and International Relations in the Political Science Department, Inaugural Director of the Center for the Study of Contemporary China, and Associate Director of the Christopher H. Browne Center for International Politics at the University of Pennsylvania. His research focuses on international relations, security studies, and Chinese politics.
James D. Fearon is Geballe Professor in the School of Humanities and Sciences at Stanford University, and Senior Fellow at Stanford’s Freeman-Spogli Institute for International Studies. He is the author of a number of articles on explanations for interstate and civil war.
Todd H. Hall is Professor of International Relations and the current director of the China Centre at the University of Oxford.
Xiaoyu Pu is an associate professor of political science at the University of Nevada, Reno. He is a Public Intellectuals Program fellow with the National Committee on United States-China Relations and a non-resident senior fellow with the Inter-American Dialogue in Washington, D.C. He is the author of Rebranding China: Contested Status Signaling in the Changing Global Order (Stanford University Press, 2019).
Kristopher Ramsay is Professor of Politics at Princeton University. He is the co-director of the Program for Analytical and Quantitative Political Science at Princeton and the associate editor for international relations at the Quarterly Journal of Political Science. He has published numerous articles on diplomatic communication, information processing, and war.
Robert F. Trager is Associate Professor of Political Science at the University of California, Los Angeles. He has also taught at Yale and Oxford Universities, held an Olin Fellowship at Harvard University, and worked in investment banking in New York. His research focuses on the strategic implications of artificial intelligence developments; how states form beliefs about the intentions of other states, and in particular on the role of diplomatic communication; the moral and emotional bases of foreign policy preferences; and a variety of other topics. He is currently completing his second book, which examines the enormous effects of women’s suffrage on the international system.
Review by James D. Fearon, Stanford University
Based on four carefully done case studies, Oriana Skylar Mastro’s excellent book proposes to explain why, in general, state leaders are so reluctant to participate in peace talks after a war has begun. She argues that they fear that willingness to talk will be interpreted as evidence of military weakness or low resolve, thus encouraging the other side to be more militarily aggressive. Short of outright defeat, leaders propose or agree to talks only when they believe that the other side will not draw a particularly adverse inference or when the other side lacks the military capability to ‘ratchet up’ its efforts. This is most likely, she argues, when they are clearly militarily dominant or believe that they have fought successfully enough that the other side will not draw the wrong conclusion from their willingness to talk. In short, leaders in war often reject talks in order to maintain or further a reputation for military capability or high resolve.
Four cases cannot establish a broad empirical regularity. For what it is worth, however, I believe that Mastro is right that refusal to engage in talks during interstate and civil wars is common. I also think that this empirical pattern, which I refer to elsewhere as “fighting rather than bargaining,” is difficult to explain with standard theories of bargaining, but is also important for understanding why wars occur. Mastro’s evidence and analysis suggest that reputational considerations can—often?—explain why states simply fight rather than bargain in the normal sense of exchanging and adjusting serious settlement offers.
My comparative advantage is not the specifics of the diplomatic history of the Korean, Sino-Indian, or Vietnam wars. I am going to focus on Mastro’s theoretical arguments, her engagement with alternative hypotheses, and whether these can make sense of broader empirical patterns that she suggests her case studies illustrate. I will argue that there are two other mechanisms that often explain why states fight rather than bargain. At the same time, I will argue that Mastro’s evidence as well as other empirical considerations suggest that what she calls the “strategic costs of conversation” have been neglected relative to their importance (14). This mechanism has been neglected not only as an explanation for when talks occur but also for why armed conflict bargaining looks so different from bargaining in other areas.
Mastro situates her study in part in the context of a literature in Political Science that followed on R. Harrison Wagner’s “call to treat war as a bargaining process, not as a result of bargaining breakdown” (22). As she is well aware, there is now a long tradition that conceives of armed conflict as a type of bargaining, a view famously associated with Thomas Schelling.
If war is a type of bargaining, however, it is a pretty strange type. Paradigmatically, in ‘bargaining’ the parties go and back forth, exchanging offers that are serious in the sense that they expect there is a non-zero chance the other side will accept. Sometimes a side may delay in order to signal toughness, but in the expectation that there is a chance the other will come to accept what is on the table.
By contrast, in bargaining in both interstate and civil war contexts, the parties often put out extreme demands—initial war aims—that everyone knows will not be accepted in full unless one side is completely defeated. These extreme aims are often surprisingly invariant with respect to developments on the battlefield, and parties go long periods without exchanging and revising serious offers at all. The latter is the main stylized fact that Mastro’s book seeks to explain.
The observation raises a very good question. Given the high costs and risks of continued fighting, why not make and adjust serious offers fairly rapidly, as happens in, say, buyer-seller bargaining over the price of rug, car, or house? If war is a bargaining process, why is it so inefficient? Why does it appear that the parties must often fight without talking for an extended period before they can agree to sit down to exchange serious offers?
Mastro’s argument about the strategic costs of conversation is closely related to what in bargaining theory is called the ‘ratchet effect.’ This is the idea that agreeing to a demand tells the other side that you are willing to be pushed at least this far, so that if one is not protected by a contract, the other side may renege (or take back their offer) and try to push you even farther. Third-party enforced contracts are not normally feasible in interstate or civil wars, which may help to explain why ratchet-effect concerns are so prominent in this context. Better to keep fighting to maintain or develop a reputation for being a “tough type” (strong military capabilities or high resolve) than to undermine the adversary’s belief in this possibility, encouraging them to be recalcitrant and aggressive in bargaining.
Mastro’s argument is slightly different in that she stresses the downside of agreeing to talk as a “ratcheting up of the [military] force” (14) used by the other side, rather than a change in bargaining behavior in negotiations. I thought there was support for this interesting idea in the case study evidence, although often leader statements and comments could be interpreted as referring either to bargaining behavior or subsequent military attacks, or both. Since greater belief in one’s military prospects tends to imply greater demands in negotiations, the two concerns effectively go together.
In her case studies, Mastro evaluates the strategic costs of conversation mechanism against four alternative theoretical perspectives, described as “the traditional bargaining model of war” (25), ideational factors, and domestic and international costs of conversation. She also poses as a “scope condition” that “The costly conversations thesis is designed primarily to explain diplomatic posture in limited interstate conflicts,” as opposed to wars in which one or both sides “harbor absolute aims” (8). The reason is that if a state “is unwilling to compromise and demands complete concession with respect to the issue at hand (absolute aims)” (19), then it may see negotiations as pointless and a potential diversion.
Taken literally, the latter is a problematic restriction. Unwillingness to compromise—or at least unwillingness to talk about possible compromises—is the thing the costly conversations thesis is supposed to explain. If harboring “absolute [war] aims” is another possible explanation for lack of interest in negotiations, then this is should be an alternative hypothesis, not a scope condition.
How do we know that, when it comes to the importance of reputational concerns, the three wars studied here are representative of the broader universe of wars, or even “limited interstate conflicts”? Could “absolute aims” be another common explanation for refusal to talk and, in turn, war duration and destructiveness? Some of Mastro’s evidence points in this direction. She explains closed diplomatic postures for parts of the Korean War for both the U.S. and China by reference to absolute aims, thus a substantial part of one of the three wars selected for the case studies. She notes in passing several other interstate wars where one side is said to have had absolute aims, and argues that the strong reluctance of governments and rebel groups in civil wars to begin negotiations may result from their having absolute aims (20-21, 132-33).
As noted, “absolute aims” are defined as seeking “complete concession with respect to the issue at hand.” This is not “relatively rare in modern international relations” (20) but instead is arguably what state leaders hope for and publicly state as war aims at the start of most wars.
Indeed, the idea that wars begin with parties expressing incompatible, often extreme war aims that are then gradually tempered based on learning from the battlefield is the essence of the alternative hypothesis that Mastro associates with “the traditional bargaining model of war” (25-26). On page 26 she draws from this model specific implications concerning the timing and circumstances of becoming open to negotiations, such as that “the losing side would be the first to offer talks” but also that states shift to open postures at the same time, or not at all. These implications are later found wanting in the case studies.
I read the implications of this “traditional model” somewhat differently. I would say that both Geoffrey Blainey’s original account and some of the formal versions imply that, first, if the parties believe that completely incompatible military expectations are behind their completely incompatible “absolute” war aims, then they might forgo talks simply because they see no point. They believe that they have to teach the other side a lesson on the battlefield first. In Mastro’s cases, plausible examples include Chairman Mao Zedong’s views and closed diplomatic posture up to February 1951 in the Korean War, and the North Vietnamese position before the Tet Offensive. My impression is that many others can be adduced.
Second, it is not the case in the “traditional model” that war ends when one side quits (as opposed to a negotiated settlement); that full convergence in beliefs is required for talks to begin; or that sides would necessarily switch to open diplomatic postures at the same time. Instead, as learning from battles occurs, the “traditional model” implies that one or both sides might become interested in reaching out to explore if the other has changed its aims and demands, or to convey that it has reduced has its own.
While this certainly does occur sometimes, it seems that even following highly informative battlefield losses, surprisingly often state leaders are not willing to talk. I agree with Mastro that ratchet-effect reputational concerns may be an empirically common explanation for this distinctive feature of armed conflict bargaining (or at least, what I think is a distinctive feature).
In sum, at the start of some wars, “absolute aims” based on strongly incompatible military expectations may provide an important alternative explanation for lack of interest in negotiations. But as battlefield evidence accumulates, continuing lack of interest or active refusals—which appear quite common—may be better explained by the reputational costs-of-conversation mechanism.
There is a different explanation for “absolute [war] aims” that in my view provides a third empirically common explanation for lack of interest in bargaining while fighting. As Mastro observes, “states may pursue absolute aims to eliminate the commitment problems that may have led to the outbreak of war in the first place” (20). Quite a few wars—most civil wars and plausibly a good number of interstate wars—are wars of regime change in the sense that one side’s leaders have no interest in a negotiated settlement because they do not believe that the other side’s leadership can be trusted to implement any deal that they would prefer to their military option. So if fighting seems promising enough, the (more objectively) absolute aim of removing the hostile regime by force can make sense. Examples include the George W. Bush administration’s disinterest in virtually anything short of deposing Iraqi President Saddam Hussein, and the Allies’ early agreement on unconditional surrender as their war aim against Nazi Germany.
This argument is well-developed in excellent books by Alex Weisiger and Dan Reiter. Weisiger, for example, captures the main idea in a quote from a September 1941 (!) report in which United States’ General George Marshall and Admiral Harold Stark say that “An inconclusive peace between Germany and her present active military enemies would be likely to give Germany an opportunity to reorganize continental Europe and to replenish her strength.” It may be that essentially the same commitment problem helps to explain the puzzle of World War I’s duration (and the infrequency of serious negotiations), as opposed or in addition to mutual optimism or a costly conversations account. After Germany controlled Belgium, any peace that allowed consolidation of control could lead to more coercion by a stronger opponent, whereas Germany could reasonably hope to keep at least this much by continued fighting.
In Weisiger’s analysis, whether a state’s war aims are ‘limited’ or ‘unlimited’ (that is, seeking regime change) is endogenous, determined by beliefs about the adversary’s nature (preferences, or type) and thus ability to commit to a negotiated settlement. This is a useful way to conceive of “absolute [war] aims”—not as “seeking complete concessions with respect to the issue at hand” but rather as the military objective of nothing less than disarming the opponent. This is plausibly another important explanation for lack of interest or refusal to negotiate in interstate and civil wars, one that happens not to be present in Mastro’s four case studies.
One last comment on policy implications of the costly conversations thesis: Mastro notes that her findings “open up additional opportunities for third-party mediation” (138). The idea is that third parties may be able to reduce the reputational costs of agreeing to talk by allowing the combatants to attribute their willingness to third-party pressure, or to say ‘we are talking because my enemy told the mediator that they would talk.’ I think this is quite right. Indeed, further evidence of the importance of the costly-conversations mechanism can be seen in the common practice of back-channels and unofficial communications through individuals with unclear negotiating authority.
But once a state’s leaders agree to direct talks, the ratchet-effect problem can still apply to the making of serious offers (here, offers that reduce demands from stated war aims). In his work on negotiations during interstate wars over the last 200 years, Eric Min finds that there is a sharp break before and after 1945. Combatants held direct talks much more often and started talking earlier in interstate wars after 1945. But post-45 direct talks have been far less productive in the sense that they are less predictive of war termination, compared to the earlier period. Min shows further that powerful third parties have exerted much more pressure on combatants to meet for talks after 1945. So it may be that third-party mediation and pressure can help get parties to the table, but the several reasons for reluctance to talk discussed above apply almost as much for reluctance to make serious offers—that is, to bargain rather than just fight.
All in all, Oriana Mastro’s book is an important contribution to what in my view has been a productive and exciting back-and-forth between relatively empirical and relatively theoretical work on interstate war that has taken place over the last 25 years or so. Detailed case studies based on—and, in Mastro’s book, adding original archival research to—diplomatic history have been central to this agenda and its progress. The above considerations are offered less as critique than as suggestions about how Mastro’s evidence and arguments based on these four cases fit into the larger empirical picture.
Review by Todd H. Hall, University of Oxford
In The Costs of Conversation: Obstacles to Peace Talks in Wartime, Oriana Skylar Mastro begins with a simple but novel question: what determines when a state which is party to an armed conflict will agree to engage in talks with its adversary? To be clear, this is not to be confused with agreeing to end a conflict, or even agreeing to a ceasefire. As Mastro points out, talks can drag on for years without the conflict lessening. Nor is it to be confused with ‘yes, but…’ offers to start dialogue that are contingent on the other side fulfilling certain conditions first. The focus of Mastro’s study is simply the willingness to sit down and engage in direct discussions with no strings attached.
Certainly, multiple reasons come to mind as to why a state might choose to seek dialogue. On the one hand, more traditional approaches would tell us that talk is cheap. And if it is clear which way the winds are blowing, would not the losing party want to begin extending feelers for a way out? A state may also be attempting to appease a more powerful sponsor. But then again, one can also think of reasons for why a state would not agree to talk. Maybe the domestic costs would be too high—sitting down at a table with those who are killing your soldiers (and, in some cases, also civilians) might not necessarily be a move well received at home. And there are also issues to do with honour, pride, or ideology that may stand in the way.
Mastro argues, however, that the primary answer is none of the above. Instead, what matters is whether or not states believe that agreeing to talks will send the wrong signal to their adversaries—one of weakness to be precise—and that those adversaries will seek to take advantage of this. In other words, the problem is that states frequently believe that if they agree to talk, this may be read as evidence that their capability or will to continue fighting is flagging. If their adversaries appear to have the resources and drive to escalate if they sense weakness, assenting to direct talks may have quite deleterious consequences. The flip side of this is that states will only open themselves up to dialogue when they have clearly established that they are not acting out of weakness and not under duress, or are confident that their adversary in not in a position to ramp up the conflict in response.
The argument is elegant and simple, and Mastro demonstrates how this logic unfolded in the context of several major Cold War conflicts in Asia: the Korean War, the Sino-Indian War, and the Vietnam War. In each case, she makes impressive use of primary and secondary sources, and in some instances also interviews, to compare the explanatory purchase of her hypotheses vis-à-vis competing explanations derived from the literature. The arguments are persuasive. Mastro does an excellent job of making a convincing case for why the state actors involved decided to express a willingness (or not) for talks when they did.
One could make the argument that the limited number of cases she is engaging—and the fact that they are occurring during a specific historical period involving a small number of actors—hampers her ability to assert with confidence that her hypotheses will hold more broadly. Granted. But even still, for an initial trial of the plausibility of her claims, the book fairs extremely well. In the process, it also sheds important light upon the decision-making logics at work within some quite significant historical conflicts. So even if the generalizability of Mastro’s claims still remains to be more robustly demonstrated, the contribution to our understanding of these historical cases is hard to dispute.
Possibly a more pointed critique would be that Mastro, in focusing on the costs, fails to sufficiently put these in conversation with the benefits. This is not to say Mastro does not recognise the potential benefits of diplomatic engagement; she notes its use for gaining intelligence, seeking a respite from fighting, playing to divisions in the adversary’s camp, or as a means to gain legitimacy and seek favour on the international stage (24). And yet, the decision to talk is treated less as the function of a cost-benefit analysis and more as something that happens when certain external conditions are met (position of strength, inability of the adversary to capitalise on perceived weakness, etc.) The account Mastro offers is a strategic one that is focused primarily on the variables that increase or decrease the perceived risk of agreeing to talks. Unexamined are the variables that might increase or decrease the relative strategic benefits from talking, or even just appearing willing to talk. But could there not be circumstances in which such perceived benefits increase to the point that state actors become willing to run the risk of appearing weak? For instance, is it not possible that offering talks in the face of escalating aggression could—in some situations—supply the chance to cast oneself as the reasonable, innocent victim, increasing the likelihood of external intervention? What about the nature and magnitude of what the other side appears to be offering should talks begin? The stance that “strategic cost dominates decision making” (29) forecloses the possibility of considering in theoretical depth what factors might alter the perceived benefits of talking. And yet, in all situations where the perceived risk of engaging in talks is not zero, the magnitude of perceived benefits should logically play a role.
It is also worth noting a puzzle that emerges from Mastro’s findings. What is striking is that various state actors which were on the receiving end of “the silent treatment” (13, footnote 7) from their adversaries were baffled by the behaviour or engaged efforts to force them to the table that ultimately were counterproductive. In one particularly memorable quote, Mastro cites Premier Zhou Enlai expressing exasperation with his Indian counterparts: “They wouldn’t talk with us! What can I do?” (64). However, if this logic is so evident and widespread, why is it simultaneously so difficult for policymakers who are on the receiving end to grasp? This puzzle possibly opens larger questions about the limits of strategic empathy, a topic which is beyond the purview of the book, but is nevertheless an interesting one.
Nothing should take away from the fact that Mastro has authored an impressive work of theoretical and empirical scholarship. Moreover, it is extraordinarily accessible and readable. I unreservedly recommend it as reading for those who are interested in diplomacy and war, from entry-level students to more specialist scholars and practitioners alike. It is a true piece of academic craftsmanship.
Review by Xiaoyu Pu, University of Nevada, Reno
While war might start in many different ways, it is often ended through peace talks. But talks are not easy in wartime. After a fight breaks out, when do the warring states decide to talk to their enemies? Why do the warring states sometimes refuse to talk? Oriana Skylar Mastro has published an excellent book to address these questions. The Costs of Conversation has the making of an important book, including innovative theoretical ideas, solid empirical evidence, and important policy implications.
Mastro suggests that there are two types of diplomatic posture in wartime: an open diplomatic posture or a closed diplomatic posture. What factors would impact leaders’ decisions to maintain or change their diplomatic posture? Mastro argues that the willingness to talk might give the enemy an impression of weakness and the enemy might take advantage of such a situation. Considering these strategic costs of peace talks, leaders would look to two factors when making decisions: the likelihood the enemy will see openness to talk as a sign of weakness and how the enemy may change its strategy in response to such an interpretation. Leaders are only willing to engage diplomatically with the enemy if these costs are low
Mastro examines the costly conversations thesis through four cases: Chinese decisions in the Korea War, Chinese decisions in Sino-India War, Indian diplomatic posture in the Sino-India War, and North Vietnamese decisions in the Vietnams War. In the Korean War case, China had a closed diplomatic posture in the early stage of the war, and it shifted to an open diplomatic posture later on. In the Sino-India war, China was open to peace talks throughout the conflict. In contrast, India maintained a closed diplomatic posture in the war. During the Vietnam War, North Vietnam’s diplomatic posture shifted from a closed posture to an open posture. In each of these cases, Mastro demonstrates that the costly conversations thesis best explains the timing and nature of these countries’ approach to peace talks.
The book makes significant contributions to international relations. First, Mastro demonstrates that talk in international politics is both more critical and more complicated than we conventionally assume. While scholars often emphasize that talk is cheap, Mastro’s book joins a number of new studies that rediscover the crucial role of talk in international politics. Furthermore, Mastro clarifies that the reasons why peace talks could be potentially more difficult than we conventionally assume. Policymakers worry that the willingness to talk will be interpreted as a sign of weakness, and the enemy might intensify and prolong the fight in response to such an interpretation. Policymakers must overcome the problems before they are confident to engage in peace talks. These obstacles to peace talks explain why some wars are prolonged, and peace talks are often delayed.
Second, Mastro sheds new light on several conflicts in Asia. These case studies are carefully designed, and the arguments are supported with rich evidence. Taking a rationalist approach, Mastro makes sense of some crucial conflicts in the Cold War era. For instance, why was a weak and fragile China more willing to fight rather than talk in the early stage of the Korean War? Why was China under the same leadership open to talks during the Sino-India war? Moving beyond the usual focus of depositional factors (such as strategic culture or personality), Mastro provides powerful interpretations of these historical cases with an innovative framework. The empirical evidence is impressive, and her studies serve as excellent examples of qualitative analyses. Mastro employs a “three-tiered collection strategy to avoid systematic bias” (32), including internal documents related to critical decision-makers, diplomatic documents in China, U.S., and India, and original interviews with decision-makers in Vietnam.
Finally, the book has crucial policy implications for international relations and U.S.-China relations in particular. In recent years, there are growing concerns that the U.S. and China might be engaging in a new Cold War or even a hot war. Given the importance of Sino-American relations, a mere possibility of war should justify all the reasons to plan for war termination in advance. Mastro suggests that the U.S. and China should not only strengthen its crisis management mechanism, but they should also consider establishing openness to wartime talks as official standard policy. By doing so, U.S. and Chinese leaders will not have to worry that agreeing to talk might be seen as a sign of weakness if war occurs. Furthermore, Mastro also proposes that powerful countries should reevaluate their tendency to rely on coercion to get an adversary to talk. Far from driving the opponent to the negotiating table, coercive tactics are more likely to harden the position of the opponent. Mastro’s policy proposals are insightful and thought-provoking.
While highlighting the book’s key contributions, I would like to raise some questions inspired by Mastro’s book. First of all, while Mastro provides a rational framework on the costly conversations thesis, a further question is how deception and misperception might play a role in peace talks. The costly conversations thesis assumes that leaders worry their willingness to talk might signal weakness. This might leave room for strategic deception. In international relations, both psychological and rationalist approaches have emphasized the tendency for states to misrepresent their capabilities or resolve. Robert Jervis suggests that “deception involves minor and relatively cheap (although often not easy) changes in behaviors to project a desired image.” According to James Fearon, states have rational incentives to conceal their capabilities or resolve. Mastro probably acknowledges the role of manipulation when she advocates that countries should work to destroy the linkage between talks and weakness (139). While states can delink talks and weakness for a benign purpose, they could also potentially shape their image into a deceptive direction. In other words, while Mastro frames peace talks mainly as a straightforward policy, leaders could also project a deceptive image whey they pretend to talk (or not talk). Furthermore, while leaders have room for deception when sending signals, misperceptions often occur when these signals are received. While Mastro might recognize the possibility of deception and misperception, her framework ultimately emphasizes a rational framework of cost-benefit analysis. Some follow-up research could investigate when and how leaders might behave differently from the rational framework of costly conversations thesis.
Second, the costly conversations thesis is mostly based on a widespread assumption of impression management in disputes: an intention to talk might give others an impression of weakness. But it is unclear why such an impression is always accurate in social life as well as in international politics. While I am persuaded that policymakers do have such kind of concerns in the historical cases described by Mastro, I am not sure how accurately this might reflect a tendency in a broader context. Mastro acknowledges that under extreme circumstances, talk might not be costly. For instance, when a nation is significantly stronger than its enemy, the leaders might not have to worry about the potential costs of peace talks (23). If so, should leaders always have to worry about talking might automatically signal weakness? Further research through multiple method design can provide more empirical and experimental evidence to examine the thesis.
Third, do some of the alternative approaches offer competing explanations or complementary explanations? In comparison to the costly conversations thesis, Mastro highlights four alternative explanations: the traditional bargaining model, ideational factors, domestic costs, and international costs. In each empirical chapter, she compares the costly conversation thesis with these alternative explanations. I am not sure if all of these alternative explanations are competing explanations. For instance, political leaders typically face both domestic audience and the international audience. When making decisions, the domestic audience is often as influential as the international audience. If the concern over projecting a weak image is related to both domestic and international audience, it might be useful to unpack the process of domestic and international interactions. Comparing with the cost conversations thesis, domestic politics sometimes might be a competing explanation, but other times it might be a complementary explanation.
Finally, how do multiple audiences shape peace talks in wartime? While Mastro briefly mentions that peace talks can happen privately, publicly, or secretly (135), her empirical cases primarily focus on direct peace talks during the Cold War era. As leaders conduct diplomacy in a more globalized environment, a promising approach is to examine peace talks in wartime that happen in different information context. The arts of impression management on backstage, frontstage, and off-stage as proposed by Erving Goffman might help illustrate some interesting dynamics in diplomacy.
Above all, The Costs of Conversation is an exemplary work that combines innovative theorizing with first-rate empirical research. The book will remain an essential reading for scholars, students, and policymakers for many years to come.
Review by Kristopher W. Ramsay, Princeton University
Oriana Skylar Mastro’s The Cost of Conversation: Obstacles to Peace Talks in War Time is an important work that opens the door to understanding how diplomatic strategy is related to the military strategy of war fighting. It is a must-read for anyone interested in the process of war and how political, military, and strategic objects interact to shape the path of conflict and the road to peace. While there is a rich literature on crisis bargaining and diplomacy and a nascent understanding of the two tracks of bargaining and fighting, we understand very little about why and under what conditions states at war are willing to open the door to a diplomatic solution.
Mastro argues that the decision to follow an open diplomatic posture is deeply strategic and intertwined with the military strategy and political objectives at any given point of a war. Her analysis shows that rational goal-oriented leaders calibrate the costs and benefits of opening direct diplomatic exchanges when they are deciding whether to offer talks, or when they are responding to the offer of their adversaries. This cost-benefit analysis is shaped by two dominant factors: the strategic cost of adverse inference and the ability of the enemy to respond to that inference with actions that are detrimental to the state's objectives.
When it comes to choosing an open diplomatic posture, the adverse inference of weakness looms large in a state's decision calculus. The state fears that the enemy may perceive the move to an open diplomatic posture “as a concession made because of weakened resolve, degraded military capabilities, or reduced war aims” (12). As a result, even the willingness to talk about a peace settlement may encourage an adversary to prolong, intensify, or escalate the conflict in hopes of achieving a decisive military victory.
The task of the state is then to use a closed diplomatic posture, one where the leaders are unwilling to entertain any direct discussion of a settlement, to persuade their enemy that they are not weak, but resilient. Here resiliency is differentiated from the traditional concept of resolve. Where the latter implies that the state is willing to bear the costs of war in order to get its desired outcome, a resilient state can both absorb and deflect the cost at a given level of violence. As Mastro writes, the goal is to “create a shared recognition that while one state may not be able to win with overwhelming force, it also cannot be coerced through increased pain and pressure” (15). Once resiliency is understood, the fear associated with an adverse inference of weakness is eliminated.
The case is then made for the costly conversation argument through an examination of four cases: the Chinese diplomatic posture in the Korean War, the Chinese diplomatic posture in the Sino-Indian War, the Indian diplomatic posture in the Sino-Indian war, and the North Vietnamese diplomatic posture in the Vietnam War. These cases were selected from thirty-eight major interstate wars. With the goal of making the cases as comparable as possible, the particular wars were selected from a common twenty-year window and all occurred in Asia during the Cold War. The process tracing and case analysis draw on primary sources, semi-structured interviews where feasible, and secondary works of history. The analysis also focuses primarily on within-state variation in conditions and behavior in order to better control confounding factors, though the between-state variation also provides evidence for the costly conversation logic.
As a final methodological comment, the analysis takes a very specific, and I would say stringent, approach to evaluating the working of the costly conversation mechanism. Clearly there are many factors and forces at play in each of these, and probably all, cases of war. Many things weigh on the decision-makers in times of war, especially when it comes to how to pursue a possible end to the conflict. There are individual preferences, domestic political pressures, and international pressures and commitments, to name a few. Here Mastro sets the high bar in arguing that adverse inference of weakness must be a prominent and documented point of discussion among elites for it to be considered an important cause. This is a very stringent criteria, meaning that the concern for adverse inference not only be present, but that we can find documentary evidence that, in the minds of the decision-makers, this strategic cost was a significant factor shaping their choice of diplomatic posture. The fact that some clear evidence of this cost is evident in all the cases speaks to the power of the concern and the importance of the idea that has been identified in this book.
The book’s main weakness is that one is left wondering why a closed diplomatic posture is ever used. For the open posture, or switching to an open posture, to have meaning two things must be true. First, states sometime—maybe often or always—start wars with a closed posture. Second, everyone expects that non-resilient states are inherently attracted to an open posture, otherwise there would be no reason to make an adverse inference from seeing a state switch from one strategy to the other. That is, the signal implied by the switching of strategies only makes sense if countries already believe that open postures imply weakness. But there is no reason why this has to be the case. In a world where all countries always pursued a strategy of having an open diplomatic posture during all wars the problem would go away.
So why are there ever closed postures? The answer, with respect to these cases, seems to be that a closed posture is the result of absolute war aims, as when the avowed goal of the state is total victory over the adversary—or when the initial goal of one side is not really open to compromise. In such a circumstance an open posture is irrelevant and, even if it only implies the smallest of costs, a pointless strategy. Why would states bother talking to enemies they have every intention of completely defeating, or when the goal is producing a fait accompli?
To start, consider the Chinese strategy when China entered the Korean War in October in 1950. When it did so, as Masto shows, Chairman Mao Zedong had one goal in mind: removing UN and American forces from the peninsula. Such an absolute goal did not leave much room for compromise and thus made an open diplomatic strategy pointless. In fact, it was not until Mao determined that he must update his military objective that an open diplomatic strategy made sense. But the conditions of costly conversation then arise. When is it best to open, and what are the costs and implications of changing the diplomatic strategy both on potential settlements, and maybe more importantly, the war effort? Mastro gives us a convincing and detailed explanation of that process.
The Sino-Indian case is a little harder to match exactly with the theoretical argument, but that is largely because it is unclear how India got trapped in a closed diplomatic stance at the beginning of the war. Beginning with China’s open diplomatic stance makes much sense of the incentives and the interconnectedness of diplomatic and military strategies. The Chinese were at the same time making military moves on the border and open to reaching a peaceful settlement that finally demarked the boundary between Chinese and Indian territory. They were working from a position of strength but, from the beginning, Premier Zhou Enlai was open to a diplomatic solution. This makes a lot of sense given the many items on China’s domestic and international agenda in 1962.
In the case of India during the Sino-Indian War, the narrative makes more sense if one starts with the assumption that the first action in this war was India’s Forward Strategy. At first, Prime Minister Jawaharlal Nehru believed that the local balance of power favored India in the border region and that India could grab the territory and generate a fait accompli without provoking a Chinese response. Obviously that failed, but at that point the closed posture was a fact on the ground. This seems unfortunate because Nehru explicitly had a policy of being always open to talks, which, if he had followed through on during the Forward Policy, would have left no room for adverse inference. This is an example where, if Nehru had maintained his reputation for always being open to talks, the adverse inference would have been defused. There is no inference to be made if the underlying condition of resilience does not affect the diplomatic strategy. But once India had gone down the road of a closed posture, there was no way out and the logic of costly conversation drove decision-making around diplomatic strategy.
North Vietnam is a very insightful case, but also not a perfect match to the story. On the one hand, the adverse inference effect is clearly present. The evidence shows a deep concern that an open diplomatic strategy would lead the U.S. to believe that its bombing strategy was softening Hanoi’s resolve and, most importantly, that such an action might increase the amount of resources the U.S. was willing to allocate to the war effort (109). It is not clear if the North Vietnamese ever had any intention of pursuing a peace deal, even when they opened the channel for direct diplomatic communication after the Tet Offensive. Like China in the early phases of its entry in the Korean War, the North Vietnamese leadership’s objectives were not really compatible with compromise, and even their internal documents note that opening direct talks would not mean lessening the war effort (108). In fact, they hoped that the initial phase of opening might give them an opportunity to regroup and also move more freely in the absence of the bombing campaign.
In this way, the Vietnam case emphasizes more the consequences of adverse inference on the opponent’s level of effort when the country with the closed strategy knows that it is not ready to reach an agreement at that time. Again, we find here that military goals and strategies that are not compatible with compromise make opening difficult or even irrelevant.
What these examples show us is that Mastro’s argument that diplomacy and war are not alternative ways of settling a dispute but rather are connected by the overall political strategy is, in an important way, even stronger than she claims. In defining the modest scope conditions for the work, focusing on the decision to start costly conversations about peace, The Cost of Conversation does not make the even stronger connection that some war strategies are so deeply connected to diplomatic strategies that they can only be chosen together. This has important implications not only for the path of war, but also how we should think about strategies during war initiation and even crisis bargaining.
Mastro has written a compelling and interesting book that brings to the forefront an important question of wartime politics: under what conditions can we expect belligerents to open the door to peace talks? The answer is not obvious, and the historical record shows us that the timing of talks varies in ways we did not understand. Here we learn that an important factor, may be the most important factor, that determines the decision to transition from a closed diplomatic posture to open direct talks is managing the adverse inferences the enemy may make about one’s capacity or willingness to fight if one agree to open a channel of communication. In the end, the narratives constructed from primary sources, semi-structured interviews, and the secondary sources paint a compelling picture of the strategic costs of conversation. This is a significant contribution to our understanding of how diplomatic and military strategies co-evolve to shape the trajectory of war.
Review by Robert F. Trager, University of California, Los Angeles
Oriana Skylar Mastro has written a compelling account of why and when states come to the negotiating table during wartime. Offering to do so poses the danger of appearing weak, she writes; states are willing to talk when factors such as battlefield events mitigate this risk. While journalistic accounts and some scholarship tend to treat talks between adversaries as an unalloyed benefit to all, Mastro explains why this is not the case. Her work offers four carefully researched, convincing case studies supported by interviews of key decision-makers.
In war and outside of it, states draw inferences from the mere fact of talks. Outside of war, the movements of leaders and envoys are carefully tracked for this reason. Indeed, merely broaching a topic can be informative. In the middle of the nineteenth century, for instance, when the Tsar broached the topic of dividing the Ottoman Empire with the British Envoy George Hamilton Seymour, many concluded that the Tsar intended to bring about the Empire's dissolution.
When nations are at war, as Mastro explains, willingness to talk carries even more weight. In fact, it serves as an important signal of resolve in itself. Talks are often the place to work out the details of negotiated solutions, and those details – what would a simultaneous withdrawal from the border look like, when would it take place, and the like – are necessary to an agreement. Since concessions cannot be made without coordination of this sort, unwillingness to talk is a strong signal of resolve, particularly so during a war when delay is inherently costly.
The analysis does not rule out other routes to negotiations, however, including one in which Mastro expresses doubt: increasing pressure from one side may force another to negotiate. This did not work for the United States in Vietnam, but we do not have a strong reason to believe that it never does. Indeed, many wars seem to end when battlefield victories convince one side to make political concessions to the other. In this regard, consider the decision of Russia to negotiate the peace agreement that ended the Crimean War. Russia was determined to revise the settlement on the day that it signed it, but understood that an end to the fighting would allow it to reconstitute its forces, improving its position. As the Tsar said at the time, “we'll wait for our time to come.” Thus, it was the military victories of British and French combined with pressure from Austria that forced Russia to negotiate a settlement it did not like and was determined to overturn. Mastro's analysis also does not rule out other reasons states refrain from talking. In some of the cases the book discusses, for instance, states use talks to play for time. Mastro recognizes the importance of this, and in these cases, the state being ‘played’ understands this danger, but is willing to pursue talks in any case. In other cases, however, states refuse to be drawn in to negotiations for fear that they may be a ruse to gain operational advantage.
There are other alternative hypotheses. One is that the factors that make the war happen in the first place also determine states' willingness to go to the negotiating table. In this sense, it is difficult to separate out the study of what makes wars stop from the study of what makes them start. This point can be appreciated in the U.S. refusal to negotiate directly with Iraq after the start of the air campaign but before the ground invasion in 1991. The U.S. goal, by that point, was to reduce the capabilities of the Iraq military and thereby change the balance of power in the region. This was the reason the war continued. Iraq, by contrast, was willing to negotiate. Similarly, before the First World War, Austria-Hungary had decided that its greatest fear was a negotiated solution. It is no wonder therefore that it did not want to talk in the early days of the war. This was a case where Austria-Hungary did not fear looking weak - it feared that Serbia might not be sufficiently resolved to fight and thus that the war, which Austria considered necessary, would be avoided. Thus, an alternative explanation for the reticence of powers to talk in the early days of wars is that whatever made them decide to fight in the first place also makes them unwilling to talk for the time being.
The analysis discounts the role of pressure from international third parties as a significant factor in deciding when to negotiate, but the extent to which this is warranted may still be unclear. Why, for instance, did Hitler's Germany declare war on the United States after the Japanese attack on Pearl Harbor? It did so because of the fear of what it would mean for the alliance with Japan if Germany did not and the belief that Germany would eventually end up at war with the U.S. whatever Germany did. The importance of the international audience - Japan - in forcing Germany's declaration can be clearly seen. Japan was not a “powerful patron,” as Mastro says is required for such considerations to carry weight (29), but only an important ally. If Germany's interest in maintaining its alliance with Japan was important on the day Germany declared war, it must have been important on the day after too. It would therefore have been a factor influencing German willingness to discuss terms of settlement before and after the declaration. Thus, in Mastro's account, the role of some factors may be reduced more than is warranted.
Overall, the book is excellent, and even exciting to read. It poses questions provocatively. It describes the thinking of decision-makers at crucial junctures authoritatively. The dynamics the book describes are real, and, as Mastro argues, they often constitute the most important factors explaining state decisions to negotiate in wartime.
 As reflected in our reviewers’ citations, a rich and varied literature linking bargaining and conflict has emerged over the past five decades. It includes Thomas Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960); Arms and Influence (New Haven: Yale University Press, 1966); Fred C. Iklé, Every War Must End (New York: Columbia University Press, 1971); Geoffrey Blainey, The Causes of War (New York: The Free Press, 1973); Paul R. Pillar, Negotiating Peace: War Termination as a Bargaining Process (Princeton: Princeton University Press, 1983); James D. Fearon. “Rationalist Explanations for War,” International Organization 49:3 (Summer 1995): 379-414, DOI: https://doi.org/10.1017/S0020818300033324; R. Harrison Wagner, “Bargaining and War,” American Journal of Political Science 44:3 (July 2000): 469-84; Hein Goemans, War and Punishment: The Causes of War Termination and the First World War (Princeton: Princeton University Press, 2000); Branislav Slantchev, “The Principle of Convergence in Wartime Negotiations,” American Political Science Review 97:1 (2003): 621-632; Robert Powell, “Bargaining and Learning While Fighting,” American Journal of Political Science 48:2 (2004): 344-361; Suzanne Werner, “Bargaining and Fighting: The Impact of Regime Type on War Onset, Duration, and Outcomes,” American Journal of Political Science 48:2 (March 2004): 296-313, DOI: https://doi.org/10.1111/j.0092-5853.2004.00071.x; Dan Reiter,, How Wars End (Princeton: Princeton University Press, 2009); Kristopher W. Ramsay. “Cheap Talk Diplomacy, Voluntary Negotiations, and Variable Bargaining Power.” International Studies Quarterly 55:4 (December 2011): 1003-1023, DOI: https://doi.org/10.1111/j.1468-2478.2011.00687.x; Jean-Pierre P Langlois and Catherine C. Langlois, “Does the Principle of Convergence Really Hold? War, Uncertainty and the Failure of Bargaining,” British Journal of Political Science 42:3 (July 2012): 511-36, DOI: http://www.jstor.org/stable/23274136; Jessica Chen Weiss, “Authoritarian Signaling, Mass Audiences, and Nationalist Protest in China,” International Organization 67:1 (2013): 1-35; Roseanne W. McManus. “Fighting Words: The Effectiveness of Statements of Resolve in International Conflict,” Journal of Peace Research 51:6 (2014): 726–740; Eric Min, “Talking While Fighting: Understanding the Role of Wartime Negotiation,” International Organization 74:3 (Summer 2020): 610-632.
 In addition to Fearon (1995) and Ramsay (2011), see James D. Fearon, “Bargaining, Enforcement, and International Cooperation,” International Organization 52:2 (Spring 1998): 269-305; Kristopher W. Ramsay, “Settling It on the Field: Battlefield Events and War Termination,” Journal of Conflict Resolution 52:6 (December 2008): 850-879, https://doi.org/10.1177/0022002708324593.
 Robert F. Trager, Diplomacy: Communication and the Origins of International Order (Cambridge: Cambridge University Press, 2017).
 Todd H. Hall, Emotional Diplomacy: Official Emotion on the International Stage (Ithaca: Cornell University Press, 2015); Xiaoyu Pu, Rebranding China: Contested Status Signaling in the Changing Global Order, Studies in Asian Security (Stanford: Stanford University Press, 2019).
 James D. Fearon, “Fighting rather than Bargaining,” unpublished paper presented at the 2007 Annual Meetings of the American Political Science Association. The current version is Fearon and Xin Jin, “Armed Conflict Bargaining.” Both are available at http://fearonresearch.stanford.edu/53-2/.
 In my view “reputational costs of conversation” would have been a more exact label, since “strategic” has many possible referents in this context.
 R. Harrison Wagner, “Bargaining and War,” American Journal of Political Science 44:3 (July 2000): 469-84.
 Thomas Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960), chapter 3; Arms and Influence (New Haven: Yale University Press, 1966), chapters 1 and 4. In “Bargaining and War,” Wagner makes a fascinating and I think compelling case that Clausewitz’s analysis of war “as simply a continuation of political intercourse” expressed an understanding of what he termed “real war” as a bargaining process.
 Dan Reiter finds this in his study of 22 “key decision times” in six major wars in his book How Wars End (Princeton: Princeton University Press, 2009); see also Fred C. Iklé, Every War Must End (New York: Columbia University Press, 1971). Mastro provides examples in several of her cases.
 A stylized fact is an empirical regularity that is probably true but that lacks extensive documentation. For instance, Mastro says that “a review of interstate wars in the Correlates of War [COW] dataset” leads her to conclude that “In over half of the wars since World War II, at least one belligerent refused to talk, publicly or privately, throughout the entire conflict” (2, and footnote 5 on page 143). Work by Eric Min, “Talking While Fighting: Understanding the Role of Wartime Negotiation,” International Organization 74:3 (Summer 2020): 610-32, has recently provided more extensive documentation concerning when parties to interstate war conduct “direct or mediated” negotiations. His results roughly support Mastro’s assessment. He finds that for the 93 COW interstate wars between 1816 and 2007, talks were present for only 17% of war days, and in about one third of wars, either no talks were held at all or occurred for the first time just before the war ended. Note also that this coding notes only whether there are talks between “officially appointed representatives” of each country, not whether they are exchanging serious offers.
 This is the main argument in Fearon, “Fighting rather than Bargaining” and Fearon and Jin, “Armed Conflict Bargaining.” The intra-war bargaining model presented there drops the standard assumption that if a state agrees to a demand, it is implemented and the war ends. Instead, the other state can take advantage by pressing new demands, which makes for the ratchet-effect problem when combined with private information about military prospects. On the ratchet effect, see Oliver Hart and Jean Tirole, “Contract Renegotiation and Coasian Dynamics,” Review of Economic Studies 55:4 (October 1988): 509-40. Interestingly, in economic contexts like buyer-seller bargaining, the ratchet effect tends to lead to immediate agreement rather than long delay (as in the armed conflict case), because there is no equivalent of the option to just fight in hopes of winning or learning the other side’s type from battlefield performance. Examples of models of intra-war bargaining that assume commitment to accepted offers – models of what Mastro calls “the traditional bargaining model of war” – include Robert Powell, “Bargaining and Learning While Fighting,” American Journal of Political Science 48:2 (2004): 344-361; and Branislav Slantchev, “The Principle of Convergence in Wartime Negotiations,” American Political Science Review 97:1 (2003): 621-32.
 World War II, the US war in Afghanistan, the Gulf War, Kosovo, and the 2003 Iraq War (20-21, page 133, footnote 17).
 Geoffrey Blainey, The Causes of War (New York: The Free Press, 1973).
 See for examples Hein Goemans, War and Punishment: The Causes of War Termination and the First World War (Princeton: Princeton University Press, 2000); Dan Reiter, How Wars End (Princeton: Princeton University Press, 2009).
 Reiter, How Wars End, found that often state leaders did not change their aims or demands at all, or even increased them, after major battlefield setbacks. Iklé said that “After the initial battles, a great deal more is known about the enemy’s relative strength than before the war,” but “a government rarely reverses itself after the first campaigns.” Iklé, Every War Must End, 16. See also discussion in the Conclusion of Fearon and Jin, “Armed Conflict Bargaining.”
 Alex Weisiger, The Logics of War: Explanations for Limited and Unlimited Conflicts (Ithaca: Cornell University Press, 2013), 124; and Reiter, How Wars End. James D. Fearon, “Why Do Some Civil Wars Last So Much Longer than Others?,” Journal of Peace Research 41:3 (May 2003): 275-302, develops a closely related commitment problem model that yields as implications that civil wars will tend to be wars of regime change with little serious bargaining, and will tend to have longer duration than interstate wars. On the Iraq war, see Andrew Coe, “Containing Rogues: A Theory of Asymmetric Arming,” Journal of Politics 80:4 (2018): 1197-1210, who develops a commitment-problem account of Bush administration war aims (among other features of this war).
 Min, “Talking while Fighting.”
 Precisely, she looks at competing explanations rooted in work on bargaining theories of conflict and domestic audience costs, such as that by James Fearon; theories of ideological beliefs, as showcased in Andrew Kennedy’s work; and theories of small state behaviour, such as are found in David Vital’s writings. See, respectively, James D. Fearon, "Rationalist Explanations for War," International Organization 49:3 (1995): 379-414; Andrew Kennedy. The International Ambitions of Mao and Nehru: National Efficacy Beliefs and the Making of Foreign Policy (Cambridge: Cambridge University Press, 2011); David Vital, The Inequality of States: A Study of the Small Power in International Relations (Oxford: Clarendon Press, 1967).
 Examples include Dustin Tingley and Barbara F. Walter, “Can Cheap Talk Deter? An Experimental Analysis,” Journal of Conflict Resolution 55:6 (2011): 996-1020; Roseanne W. McManus, “Fighting words: The Effectiveness of Statements of Resolve in International Conflict,” Journal of Peace Research 51:6 (2014): 726–740.
 For instance, see Avery Goldstein, “First Things First: The Pressing Danger of Crisis Instability in US-China Relations,” International Security 37:4 (2013): 49-89. Christopher Layne, “Preventing the China-U.S. Cold War from Turning Hot,” The Chinese Journal of International Politics 13:3 (2020): 343–385.
 Robert Jervis, The Logic of Images in International Relations (New York: Columbia University Press, 1989), 11.
 James Fearon, “Rationalist Explanations for War,” International organization 49:3 (1995): 379-414.
 Robert Jervis, Perception and Misperception in International Politics (Princeton: Princeton University Press, 1976).
 Fearon, “Domestic Political Audiences and the Escalation of International Disputes,” American Political Science Review 88:3 (1994): 577-592; Jessica Chen Weiss, “Authoritarian Signaling, Mass Audiences, and Nationalist Protest in China,” International Organization 67:1 (2013): 1-35
 See Erving Goffman, The Presentation of Self in Everyday Life (New York: Anchor Books, 1959); recent examples in IR include Austin Carson, “Facing Off and Saving Face: Covert Intervention and Escalation Management in the Korean War," International Organization 70:1 (2016): 103-131; McManus and Keren Yarhi-Milo, “The Logic of ‘offstage’ Signaling: Domestic Politics, Regime Type, and Major Power-protégé Relations," International Organization 71:4 (2017): 701-733.
 James D. Fearon, “Rationalist Explanations for War,” International Organization 49:3 (Summer 1995): 379-414, DOI: https://doi.org/10.1017/S0020818300033324; Kenneth A. Schultz, Democracy and Coercive Diplomacy (Cambridge: Cambridge University Press, 2001); Anne E. Sartori, “The Might of the Pen: A Reputational Theory of Communication in International Disputes,” International Organization 56:1 (Winter 2002): 121-49, DOI: http://www.jstor.org/stable/3078672; Alexandra Guisinger and Alastair Smith. “Honest Threats: The Interaction of Reputation and Political Institutions in International Crises,” Journal of Conflict Resolution 46:2 (April 2002): 175–200, DOI: https://doi.org/10.1177/0022002702046002001; Shuhei Kurizaki, “Efficient Secrecy: Public versus Private Threats in Crisis Diplomacy,” The American Political Science Review 101:3 (August 2007): 543-558, DOI: https://doi.org/10.2307/27644465; Kristopher W. Ramsay, “Cheap Talk Diplomacy, Voluntary Negotiations, and Variable Bargaining Power,” International Studies Quarterly 55:4 (December 2011): 1003-1023. DOI: https://doi.org/10.1111/j.1468-2478.2011.00687.x; Keren Yarhi-Milo. “Tying Hands behind Closed Doors: the Logic and Practice of Secret Reassurance,” Security Studies 22:3 (August 2013): 405-435, DOI: https://doi.org/10.1080/09636412.2013.816126; Yarhi-Milo, Knowing the Adversary: Leaders, Intelligence, and Assessment of Intentions in International Relations (Princeton: Princeton University Press, 2014); Robert F. Trager, Diplomacy: Communication and the Origins of International Order (Cambridge:Cambridge University Press, 2017).
 Paul R. Pillar, Negotiating Peace: War Termination as a Bargaining Process (Princeton: Princeton University Press, 1983); Alastair Smith, “Fighting Battles, Winning Wars,” Journal of Conflict Resolution 42:3 (June 1998): 301–320, DOI: https://doi.org/10.1177/0022002798042003005; Branislav L. Slantchev, “The Principle of Convergence in Wartime Negotiations,” The American Political Science Review 97:4 (November 2003): 621-32, DOI: http://www.jstor.org/stable/3593027; Derren Filson and Suzanne Werner, “Bargaining and Fighting: The Impact of Regime Type on War Onset, Duration, and Outcomes,” American Journal of Political Science 48:2 (March 2004): 296-313, DOI: https://doi.org/10.1111/j.0092-5853.2004.00071.x; Robert Powell, “Bargaining and Learning While Fighting,” American Journal of Political Science 48:2 (April 2004): 344-361, DOI: https://doi.org/10.1111/j.0092-5853.2004.00074.x; James D. Fearon, “Fighting Rather than Bargaining,” Annual Meetings of the American Political Science Association, 2007; Dan Reiter, How Wars End (Princeton: Princeton University Press, 2009); Catherine C. Langlois and Jean-Pierre P. Langlois, “Does Attrition Behavior Help Explain the Duration of Interstate Wars? A Game Theoretic and Empirical Analysis,” International Studies Quarterly 53:4 (December 2009): 105101073, DOI: https://doi.org/10.1111/j.1468-2478.2009.00568.x; Branislav L. Slantchev, “Feigning Weakness,” International Organization 64:3 (Summer: 2010): 357-388, DOI: http://www.jstor.org/stable/40930440; Branislav L Slantchev, Military Threats: the Costs of Coercion and the Price of Peace (Cambridge: Cambridge University Press, 2011); Gideon Rose. How Wars End: Why We Always Fight the Last Battle (New York: Simon and Schuster, 2011); Jean-Pierre P Langlois and Catherine C. Langlois, “Does the Principle of Convergence Really Hold? War, Uncertainty and the Failure of Bargaining,” British Journal of Political Science 42:3 (July 2012): 511-536, DOI: http://www.jstor.org/stable/23274136.
 Andrew Kennedy, The International Ambitions of Mao and Nehru: National Efficacy Beliefs and the Making of Foreign Policy (Cambridge: Cambridge University Press, 2011).
 “Prime Minister’s Initiation to Chou Enlai: Reply to Lok Sabha Discussion,” P.N. Haksar Papers I-II, Subject File 25, Part E, NMML, February 22, 1960, 178 cited in Mastro, 80.
James H. Lebovic and Elizabeth N. Saunders. “The Diplomatic Core: The Determinants of High-Level US Diplomatic Visits, 1946–2010.” International Studies Quarterly 60:1 (2016): 107-123.
 Vernon J. Puryear. “New Light on the Origins of the Crimean War.” The Journal of Modern History 3:2 (1931): 219-234.
 Joseph Farrell and Robert Gibbons. “Cheap Talk Can Matter in Bargaining,” Journal of Economic Theory 48:1 (1989): 221-237; Robert F. Trager, Diplomacy: Communication and the Origins of International Order (Cambridge: Cambridge University Press, 2017).
 Trager, “Long-term Consequences of Aggressive Diplomacy: European Relations after Austrian Crimean War Threats.” Security Studies 21:2 (2012): 252.
 The New York Times, “The Eve of War: Four Days of Diplomacy,” 19 January 2011.
 Trager, Diplomacy, Chapter 7.
 Ian Kershaw, Fateful Choices: Ten Decisions that Changed the World, 1940-1941 (New York: Penguin, 2008).