Selling Schelling Short: Reputations and American Coercive Diplomacy after Syria

Abstract

The notion that credibility is important in international politics, and maintaining it requires following through on threats and commitments to establish a reputation for resolve, has been standard diplomatic savoir-faire for decades, with its most influential articulation in the work of Thomas Schelling. According to a new consensus it also is wrong; credibility is linked exclusively to the relative capabilities and interests a state can bring to bear. The new consensus was invoked to criticize the Obama administration's coercive strategy against Syrian chemical weapons use in 2013. This article revisits Schelling to determine whether critics have an accurate depiction of coercive diplomacy theory. The analysis show that many of Schelling's specific arguments related to American–Soviet Cold War rivalry were mistakenly offered as general statements about reputation in all coercive bargaining encounters. A more nuanced interpretation stresses commitment of reputation operating within the complexity of the particular bargaining situation. Reputations are relevant but do not determine credibility in international politics; they matter more, relative to other factors, in iterated encounters (and the expectation of future crises) between the same two actors, a situation that approximates Schelling's ‘continuous negotiation’, as well as across fundamentally similar crises between an adversary and a third party. This qualified position was missing in the debate over Syrian chemical weapons in 2013, and should be embraced to better manage tense diplomatic relationships and periodic crises with other potentially hostile world powers. Schelling's work shows that reputation can be an ingredient for peace, and not merely a pretense for war.

Introduction

The notion that credibility is important in international politics, and that maintaining it requires following through on threats and commitments so as to establish a reputation for resolve, has been standard diplomatic savoir-faire for decades, having found its most influential articulation in the work of noted economist and game theorist Thomas C. Schelling, particularly his seminal The Strategy of Conflict (1960) and Arms and Influence (1966).¹ The problem, according to many critics, is that Schelling's arguments about the relevance of reputation are fundamentally wrong. At least, so say a series of social scientists (and their proponents) who regard the concept of credibility (insofar as it is determined by past behaviour) as a myth, a shibboleth of strategic and diplomatic parlance that remains unsupported by empirical evidence and potentially disastrous as a foreign policy priority. In late summer 2013, this view featured prominently in discussions regarding the American response to Syrian President Bashar al-Assad's violation of President Barack Obama's red-line warning against the use of chemical weapons (CW) in Syria's ongoing civil war. Abandoning that red-line (essentially acquiescing to Assad's use of chemical weapons), critics argued, would have no deleterious effects on American credibility in this or future crises, because new research in political science had effectively demonstrated that reputational considerations are overblown. This argument has important implications for American coercive diplomacy beyond Syria. According to this view, leaders must re-evaluate long held beliefs about the importance of following through on commitments or the dangers of being called in repeated bluffs. Of course, if this position is wrong, leaders could face disastrous consequences for abandoning such principles. In a world in which American deterrent and coercive capabilities undergird a significant measure of global security, the stakes of the debate are incredibly high.

This article takes issue with how the ‘conventional wisdom’ of, specifically, Thomas Schelling (as the forbearer of rational deterrence theory) is represented by critics who propose an alternative understanding of how past behaviour and reputations for resolve influence the credibility of coercive threats in the context of international crises. The first section situates the argument through a brief discussion of the Syrian chemical weapons crisis of 2013. The purpose here is to highlight the parameters of the policy debate regarding coercive diplomacy and credibility in that case. Policy arguments as to the irrelevance of reputations figured prominently amongst administration critics, and were buttressed by reference to scholarly work challenging the received wisdom associated with Schelling. In the second section, I introduce and evaluate many of the criticisms this scholarship levies at Schelling, distilling the essence of the objections and what they say about critics’ perceptions of his arguments. The third section returns to Schelling's work to determine whether or not the critics have offered an accurate depiction of his core logic and theory. This analysis suggests that many of Schelling's specific arguments about the American–Soviet Cold War rivalry are mistakenly offered as general statements about his views on reputations in all coercive bargaining encounters. A more nuanced and novel interpretation is presented in which the commitment of one's reputation is understood as operating within and through the complexity of the particular bargaining situation. On the basis of this reading, the fourth section reorients the debate away from unhelpful extremes—the question is not whether reputations and past actions are ‘always’ or ‘never’ relevant (both views, Schelling would argue, are mistaken). The more important questions are when, how and why do reputations matter?

This article suggests tentative answers. Reputation (or, more specifically, the commitment of one's reputation) is likely to be relevant in those bargaining encounters that approximate Schelling's designation of ‘continuous negotiations’—that is, when actors (in this case, states) have a reasonable expectation that future encounters will occur with the same adversary over the same or different stakes. Given the protracted nature of most international crises (including, for example, Syria), maintaining a reputation for resolve is likely to enhance one's bargaining position in the context of iterated encounters. This position is enhanced, moreover, because one's bargaining position in the present is weakened or strengthened depending on the plausibility of one's commitment of reputation; while it is unlikely to be definitive (the balance of capabilities and interests at any given point will always be relevant, and indeed can change through the course of a protracted crisis), to the extent that commitment of one's reputation is an available tactic in a strategic bargaining situation, it is in one's interest to maintain its efficacy. More broadly, reputation is potentially relevant given the prospect of future encounters with the same or similar adversaries—states may reasonably make reputational inferences based on one's past behaviour in like-situations.² The important point is that Schelling considers reputation not, as many critics allege, as definitive of credibility but rather as a component of a more complex strategic bargaining situation; his logic underscores the continued relevance of reputation for American coercive diplomacy (in Syria and beyond) and the need for policy-makers to be aware of both its utility and its limitations in managing international crises. These lessons are likely to be particularly relevant as the United States navigates tense diplomatic relationships (and periodic crises) with Iran, China, Russia and other world powers in the years to come.

Syrian Chemical Weapons and the ‘Cult of Credibility’

Since 2011, Syria has been engulfed in a bloody and intractable civil war as insurgent forces attempt to overthrow the regime of President Bashar al-Assad. With violence escalating steadily in the weeks and months following the initial outbreak of conflict—and as government forces increasingly targeted civilian, non-combatant populations—pressure mounted on the United States to intervene (either directly against the Assad regime or indirectly by supporting elements of the insurgency) and stop the bloodshed. The Obama administration, while forcefully condemning the Assad regime, largely resisted this pressure, likely wary of becoming involved in further mid-east conflict following nearly a decade of war in Iraq (as well as the ongoing conflict in Afghanistan). Nonetheless, tentative reports regarding the potential mobilization of the Syrian government's chemical weapons arsenal generated real concern from the administration.³ In response to a question at a press conference on 20 August 2012, President Obama said the following with respect to potential Syrian chemical weapons use:

We have communicated in no uncertain terms with every player in the region that that's a red line for us and that there would be enormous consequences if we start seeing movement on the chemical weapons front or the use of chemical weapons. That would change my calculations significantly.⁴

Almost exactly one year later, on 21 August 2013, the Syrian military launched a series of sarin gas attacks against rebel strongholds throughout the Ghouta agricultural belt east of Damascus, killing some 1,400 Syrians, including roughly 400 children. This blatant violation of the August 2012 red-line warning sparked an intense debate about whether airstrikes were needed to protect American ‘credibility’ and associated ‘reputation for resolve’. Several established political and foreign policy commentators, such as David Ignatius of the Washington Post and Roger Cohen of the New York Times, stressed the importance of maintaining credibility by following through on red-line threats.⁵ This concern was echoed by many policy-makers, including Senators Lindsey Graham and John McCain, as well as former Senators Jon Kyl and Joseph Lieberman, all of whom argued that inaction in Syria would lead to the erosion of America's reputation for toughness in dealing with intransigent adversaries across the globe.⁶ Similarly, the Obama administration, as articulated by Secretary of State John Kerry, was clear in its belief that following through on threats in Syria ‘matters deeply to the credibility and the future interests of the United States of America and our allies'.⁷ These arguments reflected the long-standing belief, widely associated with Thomas Schelling, that credibility in international politics is established and protected, at least in part, by following through on commitments and threats.

Many critics advocating against air strikes, conversely, did so by explicitly challenging the notion that enforcing a red-line threat was important for American credibility. In a series of blog posts and commentaries on the Syria crisis, the scholarly works of Ted Hopf, Jonathan Mercer and Daryl Press cited the resolved question of credibility and reputation in international politics.⁸ Commentators such as Stephen Biddle, Stephen Walt and Fareed Zakari denounced the ‘credibility argument’, pointing to ‘extensive’ political science research and ‘a generation of scholarship’ that rendered concerns about credibility in Syria ‘silly’.⁹ Commentators also addressed the folly of the ‘cult of “credibility”’ vis-à-vis Syria; the ‘conventional wisdom’ that past behaviour helps establish a reputation for resolve which in turn generates credibility for future coercive threats had been superseded by new research (exemplified by the work of Hopf, Mercer and Press) in which bluffing and backing down carried absolutely no consequences.¹⁰ Failing to enforce a red-line in Syria, therefore, would have no deleterious effects on American credibility in this or other crises, because states look only to the relevant balance of power and interests, and not past behaviour, in determining the credibility of a coercive threat.

Nonetheless, the Obama administration continued to make the case for, and signal its intentions to employ, air strikes against Syrian government targets even though, the critics argued, the tangible interests that would make such a threat credible (and therefore successful) were absent. On the strength of this threat, the United States was able to secure a disarmament deal in which the Assad regime agreed to accede to the Chemical Weapons Convention (CWC) and dismantle and destroy its chemical weapons stockpiles. As a result of this clear coercive success, American credibility vis-à-vis future confrontations with the Assad regime was maintained; while red-lines will of course not be definitive or automatic deterrents, the efficacy of their issue has not been completely undermined.¹¹ For future American opponents in a similar position to Assad (across a multitude of potential dimensions), moreover, red-lines will continue to have at least some significance; again, not as fait accompli deterrent postures but as components of a broader and more complex coercive encounter the outcome of which will still be contingent on additional factors related to the specific dynamics of the crisis in question, including, of course, relative power and interests.¹²

That being said, those arguing that future credibility in other crises was sufficient in and of itself to warrant intervention in Syria largely overstated the transferability-of-reputation argument, ironically undermining the case for action insofar as the logic of that position was so easily refutable.¹³ The invocation of a vague, ethereal and everywhere-and-always present American ‘reputation for resolve’ was understandably frustrating to many of the war weary—this line of reasoning inevitably supported (at least the threat of) military action. Indeed, that such a position was couched in justifications beyond the immediate imperatives of the Syria crisis (that is, in terms of hypothetical future crises with different opponents under unspecified circumstances) underscored the perception that arguments about credibility were inherently tied to a hawkish American foreign policy overly disposed to international intervention. (The legacy of Vietnam, and the so-called ‘domino theory’ of Soviet expansion, loomed large in this regard.) The logical implication of justifying airstrikes against the Assad regime exclusively in these terms was that reputations are definitive of credibility, and must, therefore, always be defended, whatever the costs in blood and treasure. For those opposing air strikes, this position was a straw man, an easy point of entry for the counter-argument that past behaviour doesn't really matter and fighting to protect one's reputation is never, under any circumstances, justified.

Had Obama followed the critics’ advice and abandoned his red-line by acquiescing to Assad's chemical weapons use, the present imbroglio in Syria/Iraq would be compounded by the looming spectre of a large and unaccounted for chemical weapons stockpile. Recent developments challenge but do not definitively undermine this achievement. A fact-finding report released by the Organization for the Prohibition of Chemical Weapons (OPCW) in December of 2014 found compelling evidence that chlorine gas had been used as a chemical weapon in Syria. While the report did not assign blame, witnesses reported the use of helicopters, strongly indicating government involvement.¹⁴ The use of chlorine as a weapon is troubling, and challenges the success of the 2013 disarmament deal. Nonetheless, because chlorine has many industrial purposes it is not considered a chemical weapon under the CWC (though its use as a chemical weapon is banned by international law) and was, therefore, not listed as part of the government's chemical weapons stockpiles to be dismantled. While using chlorine for hostile purposes violates the spirit of the 2013 disarmament deal, the possession of chlorine gas itself is not a violation, marking out somewhat of a grey area moving forward. Ultimately, the success of the initial deal cannot be dismissed, even if future negotiations are required to address Syria's use of chemical agents not covered under the CWC. While the present situation is far (far) from good, that it could have been demonstrably worse absent the robust coercive diplomacy of the Obama administration should give pause to those critics who had been calling, essentially, for the president to ‘just let it go’.

Schelling and his Critics

Underlying many of the arguments to stay out of the Syria crisis was the notion that following through on red-line threats would not enhance American credibility (in Syria or elsewhere) because the scholarly evidence did not support such a claim. This left aside the issue of whether responding to Assad's chemical weapons use or intervening in Syria more generally was in and of itself in the national interests of the United States (that is, whether intervention would have been justified even in the absence of the initial red-line threat). In this sense, it was not the critics political preferences but an objective invocation of extant academic work—what Christopher Fettweis has characterized as the ‘mountain of research from political science to suggest that [reputation] is an illusion, that credibility earned today does not lead to successes tomorrow and therefore is never worth fighting for’—that determined their policy advice.¹⁵ In this most pressing of issue areas—the application of military force in support of coercive diplomacy—the knowledge gap between practitioners and scholars must be closed. As Fettweis argues elsewhere:

The IR scholarship on reputation and credibility in international politics has evolved through a couple of clear stages. In the first, early-deterrence theorists like Thomas Schelling argued that actions are interdependent, that potential adversaries and allies routinely learn fundamental lessons about the basic disposition of states based upon their behavior in other arenas. In practice, this meant that remaining engaged in otherwise peripheral, pointless ventures (Vietnam, Korea, etc.) made sense as attempts to send messages about American resolve. Four decades of empirical and theoretical work have brought that wisdom into doubt; today, the dominant view in the academy (if not yet among practitioners) is that actions tend to be independent, that other states rarely learn the lessons we hope to teach. The current consensus—one that has been supported in recent years by the work of Jonathan Mercer, Ted Hopf, Daryl Press, and many others—is that, despite what practitioners may believe, reputations do not decisively affect the actions of others [emphasis added]. Furthermore, states can do little to control their reputations for resolve; target states form their own views, ones that are largely unaffected by efforts to control them.¹⁶

According to this view, anyone emphasizing this consensus in the context of policy debates is simply bringing policy-makers up to date as to the current state of knowledge within the academy (‘if not yet among practitioners’). As suggested by Dafoe and his co-authors in an extensive review of the literature, there is reason to be skeptical of this assessment.¹⁷ What is clear, however, is the extent to which critics of the Obama administration's coercive strategy in Syria embraced this ‘new consensus’ in their policy critiques. In this section, we engage with the scholarly basis of that consensus—in particular its treatment (and general dismissal) of the foundational work of Thomas Schelling.

Fettweis’ characterization of the origins of reputation and deterrence theory in the above extended quotation is relatively conventional. Paul Huth's well-known survey of reputation and deterrence similarly positions Schelling as the father of theoretical development in the field:

As with so much of the deterrence literature, the work of Thomas Schelling has exerted considerable influence in the development of this theoretical position. It is probably accurate to describe the argument of Schelling as the conventional wisdom about reputations in deterrence theory.¹⁸

It is not surprising, then, that both Hopf and Press begin their books with epigraphs from Arms and Influence.¹⁹ Presumably, they do so to distil the spirit of the argument they subsequently seek to refute:

Hopf: ‘We lost thirty thousand dead in Korea to save face for the United States, not to save South Korea for the South Koreans, and it was undoubtedly worth it. Soviet expectations about the behavior of the United States are one of the most valuable assets we possess in world affairs'.²⁰

Press: ‘When we talk about the loss of face that would occur if we backed out of Formosa under duress, or out of Berlin, the loss of face that matters most is the loss of Soviet belief that we will do, elsewhere and subsequently, what we insist we will do here and now'.²¹

Tellingly, each excerpt refers to ‘saving face’—as we shall see, this concept is central to critics’ interpretations of the implications of Schelling's arguments regarding reputation.

‘Saving face’ is the basis for the ‘strong-interdependence-of-commitments’ position Huth associates with Schelling:

Schelling's argument is that a defender's past behavior in a broad range of geographic situations creates strong prior beliefs in a potential attacker about the defender's expected behavior. Thus, the credibility of deterrence policies is linked over time and reputations have a powerful causal impact on whether both general and immediate deterrence succeed or fail … scholars in this school have tended to accept these conclusions as a defining feature of a deterrent relationship between two parties, by positing that both the defender and potential attacker share the belief that a retreat results in a considerableloss of face for a state, and thus significantly weakens the credibility of future threats.²²

Huth is generally critical of the logical and theoretical foundation of this position, noting that ‘the weakest element in the argument is the claim that reputations are a powerful causal variable relative to what I have termed case-specific determinants of credibility’.²³ These case-specific determinants refer to ‘the specific configuration of military capabilities, interests at stake, and political constraints faced by a defender in a given case’.²⁴ Huth goes on to explain:

The argument of this [case-specific] school is that potential attackers are not likely to draw strong inferences about defender resolve or capabilities from prior conflicts. Potential attackers therefore do not believe that past behavior is a reliable guide to predicting the future behavior of a defender. Instead, potential attackers reason that a defender's behavior is largely a function of its interests at stake and relative capabilities in a particular dispute … ²⁵

This view offers a ‘strong logical position’ because its expectations are clear and readily identifiable (at least in a theoretical sense). The credibility of a coercive threat is linked exclusively to the relative power (capabilities) and interests a state can bring to bear in a particular crisis.

Press claims to offer a decisive empirical test between these two competing positions. The ‘strong-interdependence-of-commitments’ thesis ascribed to Schelling is dubbed ‘Past Actions theory’ and pitted against the ‘case-specific’ thesis which Press calls ‘Current Calculus theory’:

In contrast to Past Actions theory—which envisions leaders assessing credibility by peering into the past—Current Calculus theory posits that decisionmakers treat crises as if they were sui generis … Future commitments will be credible if—and only if—they are backed up by sufficient strength and connected to weighty interests. ²⁶

Press examines each theory in a series of crises (Munich 1938–1939; Berlin 1958–1961; Cuba 1962) and concludes that leaders rarely (in fact, never) made reference to an adversary's past behaviour while calculating the credibility of its current threat. Putting aside several important theoretical and logical problems with these findings, it is sufficient for our purposes to emphasize the extent to which Past Actions theory constitutes a clear, coherent and parsimonious position with regards to reputation: it is definitive.²⁷ That is, Press’ formulation of the thesis relies exclusively on a direct and discrete (that is, precluding interactive effects) causal link between prior behaviour and current credibility. This is consistent with Huth's characterization of the ‘strong-interdependence-of-commitments’ position, and hence is directly traceable to the ‘conventional wisdom’ associated with Schelling.

This attribution is consistent across other critics’ work. Hopf, for example, extracts lengthy passages from Arms and Influence to underscore Schelling's emphasis on ‘face’ in the context of his (Hopf's) discussion of the ‘expectations’ of conventional deterrence theory (pp. 2–7). In assessing the implications of these expectations on American foreign policy during the Cold War, Hopf writes:

Regardless of the trivial nature of a given piece of territory, it was assumed that the United States must act in that part of the world if it was to maintain a reputation for defending its interests elsewhere … The critical argument is that U.S. interests are not to be equated with the strategic value of a piece of territory but are to be expanded to include the protection of the reputation of the United States.²⁸

In this sense, America's reputation is perceived to outweigh and override all other considerations in the determination of strategic value, to the point where otherwise ‘trivial’ issues are afforded major importance. This emphasis is reflected in Hopf's operationalization of rational deterrence theory (RDT) for the purposes of empirical testing, in which adversary assessments of credibility hinge entirely on the ‘lessons’ they learn from previous conflicts.

Mercer, similarly, makes clear Schelling's putative emphasis on reputation over and above case-specific variables in determining credibility (and therefore coercive success). He quotes from Arms and Influence:

The main reason why we are committed in many of these places is that our threats are interdependent. Essentially, we tell the Soviets that we have to react here because, if we did not, they would not believe us when we say we will react there.²⁹

He then summarizes and paraphrases what he perceives Schelling's logic to be: ‘Most disputes are not about the issue at hand, but about expectations of future behavior: if we are timid now, they think we will be timid in the future, so we must be tough'.³⁰ As Mercer correctly points out, the perceived interdependence of behaviour is essential to this logic: ‘Without the “fact” of interdependence, the “tool” of reputation is denied and with it an important means of imparting credibility to threats which may otherwise seem incredible'.³¹ Indeed, Mercer's characterization takes the claim further: not only must interdependence (that is, reputation) be a fact, it must also be decisive vis-à-vis what he calls ‘situational’ variables (the dynamics of the crisis at hand).³² In Mercer's logic, based on in/out group psychology, behaviour is attributed either situationally or dispositionally. When disposition is used to predict future behaviour, a reputation is said to have formed. Again, interdependence is crucial: ‘A reputation forms only when people perceive commitments as interdependent or coupled'. For the purposes of theory testing Mercer, like Hopf and Press, offers a simplified version of RDT: ‘a state that yields should be viewed as irresolute and a state that stands firm should be viewed as resolute; however a state behaved in the last crisis should govern others’ expectations of that state'. Mercer contends that this formulation ‘fits with the beliefs of deterrence theorists (and of decision-makers) about how reputations form'.³³

Hopf, Mercer and Press all assess the importance of reputation in international crises by testing a version of deterrence theory that takes as its core logic the complete interdependence of commitments. That they find little to no support for this position is, in fact, consistent with much prior work in the field. As Huth summarized in his assessment of empirical findings of studies of deterrence, ‘[t]here is weak support for the strong-interdependence-of-commitments argument’—an observation he included under the subheading: ‘Lack of Support for Schelling’.³⁴ But to suggest that there is little evidence to support the hypothesis that the interdependence of commitments makes reputations and past behaviour determinative of credibility only repudiates Schelling if that thesis is a fair representation of the arguments he was actually making. Clearly, a strong perception exists that this is the case. Mercer strikes a powerful rhetorical note on the first page of his book by equating Schelling (and other early-deterrence theorists) with claims so ‘outdated’ they stretch back to ancient Greece:

Since Thucydides, leaders have worried that both friend and foe sit in judgment over their every act. Any sign of irresolution will, they fear, cause allies to defect and adversaries to challenge. Deterrence theorists transformed these fears of decision-makers into strategic theory. The crucial importance of reputation is now common sense. It is also wrong.³⁵

Similarly, that the opening epigraphs to both Press and Hopf's volumes highlight Schelling's use of the term ‘face’ is meant, one would imagine, to call attention to his use of what would now likely be considered a rather silly and outmoded concept (at least in the context of contemporary strategic studies and social science). In the course of these and other critiques, Schelling and his peers are treated with wry reverence, respect for their pioneering efforts balanced by bemusement as to their quaint and antiquated arguments. But are these characterizations fair? Did Schelling really mean to say that no sacrifice of lives was too large, no crisis too peripheral, to justify backing down and losing ‘face’ in the international community? Or does a re-engagement with his work suggest a more nuanced interpretation in which reputation represents an important but not definitive component of successful coercion?

Schelling Re-visited

As mentioned, Schelling's critics repeatedly highlight his use of the term ‘face’ in the course of their commentary. Yet they do not provide the context in which this term is actually used. Schelling, in fact, also recognized (as evidenced by his use of inverted commas) that ‘face’ is ‘a frivolous asset to preserve, and that it is a sign of immaturity that a government can't swallow its pride and lose face'.³⁶ He goes on to point out, however, that there is a more ‘serious kind of “face” that relates not to a country's “worth” or “status” or even “honor”, but to its reputation for “action”’.³⁷ It is this kind of ‘face’ that Schelling believes is worth preserving—not an ethereal or romantic idealization of national ‘honour’ but rather a bargaining reputation in the context of iterated strategic encounters. Moreover, Schelling is clear that

 … the value of this ‘face’ is not absolute. That preserving face—maintaining others’ expectations about one's own behavior—can be worth some cost and risk does not mean that in every instance it is worth the cost or risk of that occasion.³⁸

That is to say, there are limits to the extent to which face should be protected; limits related to the ‘cost or risk of that occasion', that is, to case-specific variables through which a concern for reputation must be filtered. The example Schelling uses—the Korean War—is, therefore, anecdotal, tied specifically to the dynamics of the Soviet–American relationship, and not necessarily generalizable to other bargaining encounters between other sets of states:

We lost thirty thousand dead in Korea to save face for the United States, not to save South Korea for the South Koreans, and it was undoubtedly worth it. Soviet expectations about the behavior of the United States are one of the most valuable assets we possess in world affairs.³⁹

In this instance, the case is an example of Schelling's logic in a particular situation. The mistake critics make is equating the use of an example to an exposition of the logic of the general argument. Press, for example, quotes Schelling on ‘the loss of face that would occur if we backed out of Formosa under duress’, again referring to American reputation vis-à-vis the Soviet Union. Recall also Mercer's ‘restatement’ of Schelling's logic detailed above; the specific (‘we tell the Soviets’) is recast as the general. Huth, similarly, commits this error. Following a quotation in which Schelling clearly emphasizes the importance of Soviet expectations Huth paraphrases: ‘Schelling's argument is that a defender's past behavior in a broad range of geographic situations creates strong prior beliefs in a potential attacker about the defender's expected behavior’.⁴⁰ To be fair, distinguishing Schelling's underlying logic from his discussion of the Soviet–American relationship is difficult, particularly in Arms and Influence, which is very much a book about bargaining and reputation in the context of the Cold War. The key task is identifying what is general and what is specific within that analysis.

Robert Ayson has identified a similar problem with regards to Schelling's exposition of ‘stability’ vis-à-vis nuclear deterrence.⁴¹ Considering Schelling's essay ‘Surprise Attack and Disarmament’ in isolation allows one to understand his arguments regarding stability and the ‘rapidly evolving military technology of the nuclear age'.⁴² ‘From this basis', Ayson writes, ‘one might conclude that the requirements of nuclear deterrence explain the concept'.⁴³ That is to say, Schelling's treatment of ‘stability’ as such finds its expression through his analysis of nuclear stability during the Cold War. ‘However', he says, ‘Schelling's own treatment of stability in terms of the vulnerability of retaliatory forces is a single, albeit very important, case of his original concept’.⁴⁴Ayson's subsequent analysis extracts ‘a concept of stability applicable to a wide range of strategic situations where the danger of haste is a common factor (and which, by implication, is not restricted to the nuclear balance between the Cold War superpowers)’.⁴⁵ In other words, nuclear deterrence is an instantiation of Schelling's understanding of stability but is not definitive of that understanding itself; instead, it is the underlying logical principle—the ‘danger of haste’—that serves to unite disparate instances of ‘stability’ of which nuclear deterrence between the Soviet Union and United States is one example.

A similar analysis is possible with respect to Schelling's understanding of reputation and credibility. While critics draw primarily from Arms and Influence, a re-engagement with Schelling's other, earlier influential book The Strategy of Conflict suggests that many have misunderstood Schelling's underlying logic. This work includes his famous ‘Essay on Bargaining’ which outlines many of the key principles which were to become so central to the subsequent development of game theory. Unlike in Arms and Influence, Schelling's analysis here is explicitly general, not specific: ‘To study the strategy of conflict is to take the view that most conflict situations are essentially bargaining situations'.⁴⁶ The emphasis is on explicating the way two or more actors interact such that the behaviour of one is contingent on the behaviour of the other. It is in this sense that Schelling suggests ‘we may call our subject the theory of interdependent decision'.⁴⁷ The purpose is to outline, in broad terms, ‘a tactical approach to the analysis of bargaining’ in which the strategic (or interdependent) is emphasized:

The subject includes both explicit bargaining and the tacit kind in which adversaries watch and interpret each other's behavior, each aware that his own actions are being interpreted and anticipated, each acting with a view to the expectations he creates.⁴⁸

While examples of specific cases and instances are peppered throughout the narrative, they are used primarily to demonstrate the constant element of bargaining that operates within otherwise diverse contexts:

In economics the subject covers wage negotiations, tariff negotiations, competition where competitors are few, settlements out of court, and the real estate agent and his customer. Outside economics it ranges from the threat of massive retaliation to taking the right of way from a taxi.⁴⁹

As Schelling concludes: ‘These are situations that ultimately involve an element of pure bargaining—bargaining in which each party is guided mainly by his expectations of what the other will accept'.⁵⁰

This, then, is the structure or logic of the situation, the ‘solution’ or outcome of which is contingent not on ‘mutual adjustment’ toward an endpoint or ‘resting place’ but rather coercion toward a ‘concession’ by one side (these situations being defined as ‘distributional’ or ‘zero-sum’). As Schelling points out, the nature of pure bargaining creates

 … some range of alternative outcomes in which any point is better for both sides than no agreement at all. To insist on any such point is pure bargaining, since one always would take less rather than reach no agreement at all, and since one always can recede if retreat proves necessary to agreement. Yet if both parties are aware of the limits to this range, any outcome is a point from which at least one party would have been willing to retreat and the other knows it! There is no resting place.⁵¹

That is, the outcome relies on the effective use of strategy, and is not in the end discernible from the structure of the bargaining situation itself: ‘There is … an outcome; and if we cannot find it in the logic of the situation we may find it in the tactics employed'.⁵²

In the context of pure bargaining, reputation constitutes one of the manipulable aspects of the strategic encounter, and is, therefore, related to the ‘tactics’ by which an outcome (concession) can be ‘won’ by one or the other party: ‘A potent means of commitment, and sometimes the only means, is the pledge of one's reputation'.⁵³ Similarly, the relevance and impact of reputation changes based on the parameters of the bargaining situation, sometimes more and sometimes less important as a tactic by which an opponent may be forced (or more precisely induced) to concede. For example, Schelling writes of ‘intersecting negotiations’:

If a union is simultaneously engaged, or will shortly be engaged, in many negotiations while the management has no other plants and deals with no other unions, the management cannot convincingly stake its bargaining reputation while the union can. The advantage goes to the party that can persuasively point to an array of other negotiations in which its own position would be prejudiced if it made a concession in this one.⁵⁴

And again on ‘continuous negotiations’:

A special case of interrelated negotiations occurs when the same two parties are to negotiate other topics, simultaneously or in the future. The logic of this case is more subtle; to persuade the other that one cannot afford to recede, one says in effect, ‘If I conceded to you here, you would revise your estimate of me in our negotiations; to protect my reputation with you I must stand firm.’ The second party is simultaneously the ‘third party’ to whom one's bargaining reputation can be pledged.⁵⁵

If this particular scenario sounds familiar, it is because it mirrors the logic underpinning Schelling's later exposition, in Arms and Influence, of Cold War bargaining between the Soviet Union and United States: ‘Essentially we tell the Soviets that we have to react here because, if we did not, they would not believe us when we say that we will react there'.⁵⁶ In this sense, the ‘interdependence’ of commitments that generated reputational concern between the Americans and the Soviets was a function of continuous negotiation; of the recognition that other topics, issues and crises were to be negotiated between ‘the same two parties … simultaneously or in the future’. This understanding as to the ‘interdependence’ of reputation is therefore considerably more nuanced than the version critics have been presenting.

It may be countered that ‘intersecting negotiations’ as outlined here re-injects an overly broad applicability of reputation (that is, complete interdependence of commitments), insofar as one party points to other negotiations as the basis by which concession in the present instance is not possible. This objection confuses the logic of the bargaining situation and the strategic signals being communicated—in this case, the outcome is not contingent on, or necessarily related to, having stood firm in other negotiations; the appeal to one's bargaining reputation is pure tactic vis-à-vis the present negotiation; it is designed to convey to opponents that your reputation matters to you (in light of future negotiations). The plausibility that you value your reputation can influence the outcome of the present encounter in one's favour. Whether the opponents in ‘other’ crises or negotiations subsequently infer lessons from that encounter—and, therefore, rely on the ‘fact’ of one's ‘reputation’ in assessing credibility—is beyond the scope of the argument; they may, or they may not—this is contingent on the specific conditions of that negotiation and the tactics one or both sides employs within it (along with, of course, the attendant balance of power and interests). Logically, one would assume that repeated concessions would gradually erode the plausibility of the assertion that one's reputation matters to oneself, which is the basis of the efficacy of the tactic of staking one's bargaining reputation in a particular negotiation (an assertion indirectly supported by many of the empirical studies cited in the review by Dafoe and his co-authors in which bluffing, backing down, etc. were found to carry consequences). In the context of ‘intersecting negotiations’, however, in which the list of plausible ‘situational’ factors (most fundamentally, different opponents) is significant, so long as one does not always concede (bluff, back down) there is the possibility of maintaining enough reputation for resolve to make its commitment in a particular negotiation at least somewhat effective.

Continuous negotiations are subtly, though not fundamentally, different; because the present and future opponent is the same, the plausibility of future appeals rests even more squarely on present behaviour. That is, if one wishes future commitments of reputation to be effective one should stand firm now, otherwise this tactic will have less influence in the next encounter, gradually eroding one's bargaining position in the overall negotiations; the key is that because negotiations are ‘continuous’, ‘concession’ is defined much more broadly than in one-off encounters. As such, maintaining a reputation for resolve by standing firm in one instance is not ‘interdependent’ in the sense that one sacrifices resources (in the context of international politics—blood, treasure) for some future, undefined circumstance in which one's ‘face’ or reputation will be relevant; instead, maintaining a reputation for resolve by standing firm matters within the present strategic encounter. Commitments of reputation are ‘interdependent’ to the extent that each such commitment is but a single iteration of a continuous and ongoing negotiation.

Re-reading Schelling's discussion of the interdependence of commitments in Arms and Influence reveals the extent to which the Cold War competition between the Soviet Union and United States conforms to this underlying logic. The key is the tactical manipulation of ‘a need to stand firm’ for the purposes of future confrontations with the same opponent. It is in this sense that ‘[t]here is no way to let California go to the Soviets and make them believe nevertheless that Oregon and Washington, Florida and Maine, and eventually Chevy Chase and Cambridge cannot be had under the same principle'.⁵⁷ This is because ‘[o]nce they cross a line into a new class of aggression, into a set of areas or assets that we always claimed we would protect, we may even deceive them if we do not react vigorously’.⁵⁸ The point is that reputational considerations (demonstrating resolve by following through on threats/commitments), has influence over bargaining negotiations even as they are filtered through other, tangible situational factors. Understanding Schelling's emphasis on the tactical relevance of standing firm helps contextualize many of the quotations Press, Mercer, Hopf and others extract for the purposes of highlighting his ostensible over-emphasis on the interdependence of reputation. Reputation does not necessarily supersede interests, capabilities or other case-specific determinants (as made clear when Schelling caveats, as quoted above, that although maintaining expectations about one's behaviour ‘can be worth some cost and risk [it] does not mean that in every instance it is worth the cost or risk of that occasion’). Yet insofar as it is a manipulable quality and employable tactic within a strategic encounter, Schelling counselled Americans to protect it's reputation during the Cold War presumably because, as a good American, he was interested in the United States winning its negotiations with the Soviets.⁵⁹

This brings us to a final point of criticism, put forward most forcefully by Jonathan Mercer, which the present analysis can rebut. As mentioned, Mercer takes issue with respect to Schelling's assumption of the ‘fact’ of reputation; it assumes too much, he argues, to suggest that one can impart the ‘appropriate’ lesson to one's adversary—such that the lessons one wants learned, are learned—and thereby steer negotiations in the direction one intends. ‘No one spells out exactly how to get one of these reputations', Mercer writes, ‘[p]resumably, a state will be given the appropriate reputation for behaving [a certain way]'.⁶⁰ Instead, he argues, reputations are fundamentally ‘in the eye of the beholder’; that is, states do not exert decisive (or indeed any) control over how their past behaviour is perceived.

In fact, Schelling is acutely aware of the limitations associated with ‘signalling’ a desired reputation, and of the inherent problem relating to an opponent's obstreperous perceptions. To return to his example of California, Schelling writes ‘it helps to remind us that the effectiveness of deterrence often depends on attaching to particular areas some of the status of California'. As he goes on to explain, however:

The principle is at work all over the world; and the principle is not wholly under our control. I doubt whether we can identify ourselves with Pakistan in quite the way we can identify ourselves with Great Britain … To identify is a complex process. It means getting the Soviets … to identify us with, say, Pakistan in such a way that they would lose respect for our commitments elsewhere if we failed to support Pakistan … In a way, it is the Soviets who confer this identification; but they do it through the medium of their expectations about us and our understanding of their expectations. Neither they nor we can exercise full control over their expectations. ⁶¹

Schelling is thus quite clear: reputation is largely in the eye of the beholder, but this does not preclude the possibility that such perceptions can be influenced (though not wholly determined) by one's strategic behaviour. Indeed to suggest, as Mercer does, that reputations form in a largely deterministic way actually contradicts the logic of the ‘eye of the beholder’ argument (if reputation is in the eye of the beholder, there is no reason to expect it to be always interpreted the same way, that is, that no dispositional attributions will ever be made). In contrast, this logic reinforces Schelling's position, insofar as his emphasis on the tactic of committing one's reputation accounts for the (mis)perceptions of the opponent. Recall that the essence of Schelling's emphasis on bargaining is the interactive contingency of the encounter between two actors. In this way, the probability of successful deterrence and/or compellence is augmented but not guaranteed by the effective use of strategy; it is suggestive and contingent, not determinative.

Reorienting the Debate

Through a re-engagement with Schelling's core logic and theory, the preceding analysis has called into question the standard practice of linking Schelling with a ‘strong-interdependence-of-commitments’ position in which reputation supersedes all other factors and exclusively determines the credibility of coercive threats. To the extent that this constitutes the ‘conventional wisdom’ with regards to reputation and credibility in international politics, critics’ have been engaged with a drastically simplified and overstated position in their recent efforts to ‘overturn’ it.

That being said, it must be noted that many critics, in discussing Schelling, simultaneously point to the strong emphasis policy-makers have traditionally placed on reputation and credibility. Press provides numerous quotations from American decision-makers during the Cold War to highlight the extent to which they believed the strong-interdependence-of-commitments positions so often (wrongly) associated with Schelling. Jonathan Mercer, likewise, peppers his pages with references to policy-makers and their justifications of intervention and war on the basis of credibility and reputation. In this sense, it might be argued that it doesn't really matter what Schelling was ‘actually’ saying—what matters are the lessons that were extracted and internalized by people in positions of power such that they were subsequently employed in the conduct of American foreign policy. For example, Press quotes Secretary of State Dean Rusk as to why the United States became involved in Vietnam:

The integrity of the United States commitment is the principal pillar of peace throughout the world. If that commitment becomes unreliable, the communist world would draw conclusions that would lead to our ruin and almost certainly to a catastrophic war.⁶²

The so-called ‘domino theory’ of American foreign policy was based largely off of perceptions such as this; indeed, this theory required a complete-interdependence-of-commitments position in order for it to make sense. Hopf's book is quite explicitly a reaction to this (domino) application of the reputation argument, and as such is an important challenge to a misguided foreign policy legacy.⁶³

The logic behind this legacy is reflected, in many ways, in some of the statements issued by contemporary policy-makers vis-à-vis the Syria crisis (such as Graham and McCain), something which might help explain how vocal critics’ became during the course of this debate, as visions of Vietnam and the associated loss of American lives were conjured by the sounds of the beating war drum. To the extent that justification for intervention in Syria was couched in terms of some vague and always-and-everywhere applicable American ‘reputation for action’ these criticisms were undoubtedly warranted. The notion that credibility in Ukraine was somehow a product of action in Syria, as Anne-Marie Slaughter argued, is as wrong as the perception that backing down never carries consequences; both positions overstate their claims.⁶⁴

A fundamental problem with this debate is it's either/or structure: either reputations never matter or they always do. As Dafoe et al. point out in their review article, there are better ways to approach the issue: ‘Much scholarship has focused on the broad question of whether reputation and status matter, but we recommend scholars also investigate the nuanced and productive questions related to when, how, and why they matter'.⁶⁵ That is, if a reasonable middle-ground can be occupied, in which the influence of reputation is recognized, as Schelling intended it to be, as contingent on the confluence of other, case-specific variables related to the structure, context and parameters of the particular bargaining situation, then the important and productive task of further specifying the ‘when, how, and why’ of that influence can be engaged. The present analysis suggests some preliminary answers to these questions.

First, reputations are likely to matter the greater the link and connection between one round of bargaining and the next. This link is most apparent in ‘continuous negotiations’ in which each party is aware that future encounters will occur. In the context of international relations, the characterization of ‘enduring’ or ‘strategic’ rivalry is potentially pertinent, as these relationships are based, in part, on the ‘anticipation of future confrontations (the expectation of a continuing conflict relationship)'.⁶⁶ The scholarly study of rivalry suggests that coercive diplomacy in the context of enduring conflictual relationships may operate differently from conventional interactions between states; lessons exert greater influence given repeated confrontations and expectations of future conflict.⁶⁷

Even outside international rivalry, however, one can identify situations in which the logic of Schelling's ‘continuous negotiations’ may apply. A common mistake in the RDT literature is to conceptualize and code coercive encounters as single data points, one-off instances of either success or failure. In reality, however, many (if not most) international crises are protracted, with multiple encounters occurring between the same states over time. The success or failure of coercion, as Frank Harvey points out, is best understood as developing through the course of the crisis, as challenges and probes provide opportunities to display resolve and signal commitments thereby enhancing, potentially, credibility.⁶⁸ In this sense, crises such as Syria—in which there were initial red-line threats against chemical weapons use, followed by probes (limited but escalating chemical weapons use) by the Assad regime, the tipping-point of the August 2013 chemical attacks in Ghouta, and the subsequent re-establishment of a coercive threat by the Obama administration—conform in many ways to the basic structure of a continuous negotiation. Even with current events in the region, the impasse between the Assad and the United States is less ‘resolved’ then it is ‘paused’—both sides recognize the potential for future confrontation. In the event such a confrontation occurs, the United States will enjoy greater credibility having not backed down following the Ghouta attacks. The problem for critics was that they became too focused on rebutting weak claims that action in Syria was necessary for American credibility elsewhere; they missed the fact that mounting a robust coercive threat vis-à-vis the Assad regime mattered for American credibility in that crisis, against that opponent, for the purposes of the future deterrence of chemical weapons use against civilians in that country.

Of course, to emphasize the immediate, within-case importance of reputation above cross-case or ‘transferable’ reputation is not to suggest that the latter does not or cannot have influence. If understood in terms of gradations and not absolutes, the extent to which crises approximate one another along a variety of dimensions (opponents, power balances, issue areas, geographical region etc.) can serve to either augment or diminish the relative importance of ‘outside’ (or general) reputation in any particular case. Logically, prior behaviour will be most relevant given identical crises (between the same parties over the same issues etc.). Because no two crises, even between the same opponents, are ever truly identical, however, the influence of reputation will be maximized in those crises that most closely approximate this ideal. (Note that the tactical commitment of reputation in Schelling's ‘intersecting negotiations’ similarly relies on implicit similarities between negotiations—the illustrative definition above [p. 10] not only suggests like-actors [management; unions] but also specifies that the commitment of reputation will only be effective if the case can be ‘persuasively’ made that other—simultaneous or future—negotiations will be influenced; undoubtedly the persuasiveness of any such argument will rely in whole or in part on effective and compelling comparisons and parallels between present and future bargaining situations.) Naturally, this occurs most readily when the two opponents are the same, but similarities across other dimensions can also push the approximation; the key is to identify similarities between cases such that the across-case transfer of reputation is logically plausible.⁶⁹ Absent this specification, vague appeals to general American credibility, for example, are likely to ring hollow. Yet a careful comparison of Syria and Iran (Middle Eastern power; authoritarian regime; crises over weapons of mass destruction) suggests that some lessons may be transferable. Again, not as definitive or determinative, but as complementary and additive components of credibility (that is, as ‘part of the puzzle’ in Tehran's broader assessment of American resolve vis-à-vis their nuclear program and any coercive threats pertaining thereto).

Finally, how and why would such reputation for resolve matter? Again, our engagement with Schelling provides potential answers. It would matter because one's bargaining position in the present is weakened or strengthened depending on the plausibility of one's commitment of reputation; while it is unlikely to be definitive (the balance of capabilities and interests at any given point will always be relevant, and indeed can change through the course of a protracted crisis), to the extent that commitment of one's reputation is an available tactic in a bargaining situation, it is in one's interest to maintain its efficacy. That is to say, a strong reputation for resolve amplifies the effectiveness of the commitment of reputation as a tactic in bargaining because adversaries who believe those commitments (that is, who believe that our reputation matters to us) are more likely to concede given that our intention to stand firm has been effectively communicated. Of course, because international politics are never ‘pure’ bargaining—with interests, capabilities and other variables related to context and time and place and circumstance constantly in play—there is reason to be cautious as to the extent to which these principles are applicable in any given situation; nonetheless, there is an element of pure bargaining in any confrontation, particularly so between sovereign states in the international system. As such, it would be foolish to dismiss the implications of Schelling's analysis altogether, and to characterize his logic and his arguments as inapplicable to the post-Cold War, 21st Century conduct of international relations.

Conclusion

This assessment, in which both positions of the either/or debate regarding the relevance of reputation are dismissed in favour of a qualified, contextual recognition that reputations occasionally do matter (sometimes more, sometimes less), is certainly less provocative than critics’ claims as to the dismantling of a decades old ‘conventional wisdom’.⁷⁰ No doubt, criticisms as to the misapplication of reputation and credibility arguments are fostered by frustration at the ‘[b]lood and treasure expended in unnecessary wars’ such as Vietnam and Iraq.⁷¹ ‘Before another politician declares that America's reputation requires us to spend lives and money today to ensure peace tomorrow', writes Mercer, ‘we need to think carefully about the logic and evidence behind the argument'.⁷² There is no question as to the urgency of this directive. Yet tempering policy-makers’ overemphasis on reputations and credibility, while important, should not come by overstating the obverse extreme. Careful analysis of the contributions of Schelling, so often the touchtone of contemporary criticisms regarding credibility and reputation, reveals that a complete dismissal of reputational ‘currency’ is largely unwarranted. The intuitive notion that following through on commitments now might have some impact on the credibility of similar commitments later remains sound. There is something to be gained by maintaining a reputation for resolve—a better bargaining position in the future—and this can have a tangible impact on peace and security in the world. Intransigent states with weapons of mass destruction (WMD) will hesitate to cross red-lines if they perceive them to be real and meaningful. The key word here is ‘hesitate’—there are many other variables related to time and place and power and interest that may cause them to proceed anyway. Many if not most of these variables are not subject to manipulation, particularly in the immediate term over the course of coercive bargaining in a particular crisis. A reputation for resolve is not, in other words, a magic bullet, and balanced debate and discussion as to when and where it should be protected must focus on the merits and dynamics of the particular situation, not on recourse to either/or positions related to a ‘conventional wisdom’ or a ‘new consensus’.

This conclusion has broad implications for American foreign policy moving forward. Most immediately, implementation of the P5+1 Joint Comprehensive Plan of Action (JCPOA) with Iran will require a credible threat of American retaliation in the face of potential Iranian violations. Yet, as discussed above, the imperative to establish a reputation for resolve cannot supersede all other considerations—the credibility of the threat regarding sanctions ‘snapping’ back into place is likely to be undermined by resistance from, China and Russia among others, particularly if there is no international consensus as to the ‘severity’ of any particular violation.⁷³ In this scenario, the United States should not chase a reputation for resolve by escalating an otherwise minor impasse solely to maintain credibility. More serious violations, up to and including a full-blown ‘breakout’ towards a nuclear weapon, would present a different challenge; here a clear and definitive red-line has been (repeatedly) set, and American reputation (and associated credibility) in the context of WMD proliferation (as well as in ongoing negotiations with Tehran on major security issues, including its military activities in the region) would be severely undermined if no action were taken.

The same case-by-case logic should be applied to American interactions with potential great power adversaries China and Russia. In many ways, the Syria crisis itself can be considered as part of an extended (or continuous) encounter with Russia—accession to the disarmament deal as much successful coercion of Moscow as Damascus. Yet as has been made clear, the notion that crises in Syria and Ukraine were directly linked with respect to credibility overstates the relevance of reputation; the differences too great and similarities too shallow to warrant such transfer (or to render it plausible). Because Russia remains such a prominent player in the mid-east, however, the 2013 Syria disarmament success does establish a pattern of behaviour and thereby a potential basis for further engagement with Moscow in the region, including possible confrontation with respect to Iran and implementation of the Iranian nuclear deal (as discussed above). Russian provocations in Eastern Europe, conversely, will unfold according to its own strategic logic, with Western credibility most clearly and obviously linked to a NATO ‘tripwire’ such that Russian incursions into a NATO member state prompts intervention from NATO members. In the event that this line is crossed and no such intervention occurs, the credibility of this long-established red-line would be significantly (if not fatally) undermined. A potential analogue with respect to American–Chinese relations is the difference between proxy relationships in Africa, on the one hand, and Chinese provocations in the South China Sea, on the other. As (primarily economic) competition between China and the United States intensifies on the African continent, lessons from past behaviour and the commitment of reputations will likely be germane to bargaining and negotiations with respect to third party African states; similarly, more direct (though here also the United States will potentially work through and in the interests of regional allies such as Japan or other South East Asian nations) and potentially high-stakes confrontations in the oceans of Asia (whether over Taiwan, the Senkaku islands or other issues) will be connected over time such that the logic of reputations outlined here will be relevant.

Parsing these differences, and the even more fine-grained dynamics with respect to each crisis or confrontation in a chain of negotiations, is admittedly complex, leaving room for reasonable disagreements as to when, in a particular instance, specific reputational considerations are in play, or how influential they might be relative to other factors. Yet given the importance of these encounters, and the implications for global security connected to the successful management of American relations with China, Iran, Russia, Syria and others, it is incumbent on scholars and analysts of international politics to embrace rather than eschew this complexity. Dismissing reputation in the context of international crises is dangerous. Committing one's reputation, as Schelling argued, remains an important component of a nation's bargaining strategy. ‘War at best is ugly, costly, and dangerous’ Schelling wrote, a ‘ … persuasive threat of war may deter an aggressor; the problem is to make is persuasive, to keep it from sounding like a bluff'.⁷⁴ We must remember, in other words, that reputation can be an ingredient for peace, and not merely a pretense for war.

ACKNOWLEDGEMENTS

Thanks to Frank Harvey, panel participants at the International Studies Association North-East Annual Convention, 8 November 2014 in Baltimore, and the anonymous reviewers for Contemporary Security Policy for their helpful comments and suggestions on earlier drafts of this article.

Disclosure statement

No potential conflict of interest was reported by the author.

Funding

I would like to acknowledge the Killam Trust for its financial support.

Notes

1. Thomas C. Schelling, The Strategy of Conflict (Cambridge: Harvard University Press, 1960); Schelling, Arms and Influence (New Haven, CT: Yale University Press, 2008 [1966]).

2. For an empirical evaluation of this argument see Mark Crescenzi, ‘Reputation and Interstate Conflict', American Journal of Political Science, Vol. 51, No. 2 (2007), pp. 382–96.

3. For a discussion of the administration's concerns regarding Syrian chemical weapons through 2012–2013, see Ken Dilanian and David S. Cloud, ‘US Saw Yearlong Rise in Chemical Weapons use by Syria', Los Angeles Times, September 6, 2013.

4. ‘Remarks by the President to the White House Press Corps', The White House, Office of the Press Secretary, 20 August 2012 (emphasis added). The precise impetus for these remarks is unclear, but likely represents a mix of domestic and international pressures. On one hand, the administration was certainly keen on enforcing the well-established international norm against the use of chemical weapons. At the same time, however, a strong stand on chemical weapons may have been designed to placate domestic opponents calling for the administration to intervene more forcefully in Syria. Whatever its origins, however, the basic point remains: once the red-line was issued, the parameters of the debate over whether it was necessary to enforce it to maintain American ‘credibility’ were established.

5. David Ignatius, ‘In Syria, U.S. Credibility is at Stake', The Washington Post, 28 August 2013; Roger Cohen, ‘Red lines matter', New York Times, 3 September 2013.

6. John McCain and Lindsey Graham, ‘Obama is Failing the Middle East, and U.S. Interests There', The Washington Post, 25 October 2014; Joe Lieberman and Jon Kyl, ‘Inaction on Syria Threatens U.S. Security', The Wall Street Journal Online, 30 August 2013. It must be noted, however, that these were journalists and policy-makers making foreign policy arguments. As such, their positions were not regularly reinforced by a careful reading of deterrence scholarship/theory/research. Proponents of the alternative position (that reputations don't matter) on the other hand, positioned their argument as more informed, scientific, research and theory-based policy recommendations. In this sense, there was really no academic debate.

7. US Department of State, ‘Statement on Syria—Remarks by John Kerry Secretary of State', 30 August 2013, Treaty Room Washington, DC.

8. Daryl Press, Calculating Credibility: How Leaders Assess Military Threats (Ithaca, NY: Cornell University Press, 2005); Jonathan Mercer, Reputation and International Politics (Ithaca, NY: Cornell University Press, 1996); Ted Hopf, Peripheral Visions: Deterrence Theory and American Foreign Policy in the Third World, 1965–1990 (Ann Arbor: Michigan University Press, 1994).

9. Fareed Zakaria, ‘U.S. Credibility is Not on the Line in Syria', The Washington Post, 8 May 2013; Stephen Biddle, ‘Congressional Testimony: Assessing the Case for Striking Syria', Council for Foreign Relations, 10 September 2013; Stephen Walt, ‘We're Going to War Because we Just Can't Stop Ourselves', Foreign Policy, 27 August 2013; Walt also offered up his Foreign Policy blog to Daryl Press and Jennifer Lind for a guest spot, see Daryl Press and Jennifer Lind, ‘Red Lines and Red Herrings', Foreign Policy, 6 May 2013.

10. See Daniel Larison, ‘The Desperation of the Syria Hawks’ ‘Credibility’ Argument’, The American Conservative, 6 May 2013; Daniel Larison, ‘The Cult of “Credibility”’ The American Conservative, 13 May 2013; and Daniel Larison, ‘The “Credibility” Argument for Syrian Intervention is Nonsense’, The American Conservative, 29 April 2013; Benjamin Friedman, ‘Credibility over “red-lines” no reason for war’, CATO Institute, 26 August 2013; Max Fisher, ‘The red line fallacy: What everyone gets wrong about why the U.S. would strike Syria’, Washington Post, 28 August 2013; Dylan Matthews, ‘Why Obama Shouldn't Care About Backing Down in Syria’, Washington Post, 12 September 2013; Jonathan Mercer, ‘Bad Reputation: The folly of going to war for “credibility”’ Foreign Affairs, 28 August 2013; Press and Lind, ‘Red Lines and Red Herrings'.

11. It is important here to be very clear regarding what is meant by ‘success’. Obviously the broader conflict in Syria has yet to be resolved, and in many ways has continued to deteriorate since 2013. The Assad regime remains in power, and has even purportedly used chlorine gas as a weapon against civilians (see endnote 14). Moreover, there is a common perception (particularly among political opponents) that the Obama administration ‘failed’ to follow through on its red-line threat by not intervening militarily following the Ghouta attacks. This article takes a narrower view as to the parameters of the case—the intent of the coercive strategy was to eliminate (and thereby prevent the future use of) Syria's chemical weapons stockpile (as defined under the CWC). This goal was largely achieved through the disarmament deal. In this sense, ‘success’ is defined much more narrowly than a complete cessation of violence in Syria or the resolution of the broader conflict. Similarly, the threat of military strikes resulted in the elimination of Syria's chemical weapons—the red-line against chemical weapons was therefore enforced, even if the strikes never actually occurred. That is to say, the purpose of the initial red-line was to prevent the use of chemical weapons, not to communicate that the use of chemical weapons would trigger an intervention to solve the broader crisis by toppling the Assad regime (though a limited intervention might very well have escalated—given the escalatory logic of coercive diplomacy and the American predilection for mission creep—and eventually threatened the regime, a possibility to which Assad [and Russia] was likely keenly aware).

12. The key insight here is that a combination of factors collectively combine to determine the credibility of a coercive threat. Reputation for resolve is but one of four prerequisites for successful deterrence/compellence, the others being effective communication, clear commitment (interests), and the capability (power) to carry out the threat. See Frank Harvey, ‘Rigor Mortis or Rigor, More Tests: Necessity, Sufficiency, and Deterrence Logic’, International Studies Quarterly, Vol. 42, No. 4 (Dec. 1998), pp. 675–707.

13. For an example see Anne-Marie Slaughter, ‘Stopping Russia starts in Syria', Project Syndicate, 23 April 2014.

14. . Note by the Technical Secretariat. Third Report of the OPCW Fact-Funding's Mission in Syria. S/1230/2014 (The Hague: Organization for the Prohibition of Chemical Weapons, 18 December 2014).

15. Christopher Fettweis, ‘Wrong Beliefs About the Ukraine Crisis: No.1, that the Stakes are High for U.S.’ Foreign Policy, 30 April 2014.

16. Christopher Fettweis, ‘Review of “The Shadow of the Past: Reputation and Military Alliances Before the First World War” by Gregory Miller', Perspectives on Politics, Vol. 10, No. 4 (Dec. 2012), p. 1130.

17. Allan Dafoe, Jonathan Renshon, and Paul Huth, ‘Reputations and Status as Motives for War', Annual Review of Political Science, Vol. 17 (2014), p. 372. For additional work challenging the notion that reputations don't matter in international politics see Paul Huth, Extended Deterrence and the Prevention of War (New Haven, CT: Yale University Press, 1988); Paul Huth and Bruce Russett, ‘Deterrence Failure and Crisis Escalation’, International Studies Quarterly, Vol. 32, No. 1 (1988), pp. 29–45; Pauk Huth, Christopher Gelpi, and Scott Bennett, ‘The Escalation of Great Power Militarized Disputes: Testing Rational Deterrence Theory and Structural Realism’, American Political Science Review, Vol. 87, No. 3 (Sept. 1993), pp. 609–23; Elli Lieberman, ‘What Makes Deterrence Work? Lessons from Egyptian-Israeli Enduring Rivalry’, Security Studies, Vol. 4, No. 4 (Summer 1995), pp. 851–910; Frank Harvey, ‘Deterrence and Ethnic Conflict: The Case of Bosnia-Herzegovina, 1993–94’, Security Studies, Vol. 6, No. 3 (1997), pp. 180–210; Frank Harvey, ‘Rigor Mortis or Rigor, More Tests: Necessity, Sufficiency, and Deterrence Logic’, International Studies Quarterly, Vol. 42, No, 4 (Dec. 1998), pp. 675–707; Joshua Goldstein and Jon Pevehouse, ‘Reciprocity, Bullying, and International Cooperation: Time-series Analysis of the Bosnia Conflict’, American Political Science Review, Vol. 91, No. 3 (Sept. 1997), pp. 515–29; Anne Sartori, Deterrence by Diplomacy (Princeton, NJ: Princeton University Press, 2005); Barbara Walter, ‘Building Reputation: Why Governments Fight Some Separatists but Not Others’, American Journal of Political Science, Vol. 50, No. 2 (2006), pp. 313–30; Mark Crescenzi, ‘Reputation and Interstate Conflict’, American Journal of Political Science, Vol. 51, No. 2 (2007), pp. 382–96; Douglas Gibler, ‘The Costs of Reneging: Reputation and Alliance Formation’, Journal of Conflict Resolution, Vol. 52, No. 3 (2008), pp. 426–54; Dustin Tingley and Barbara Walter, ‘The Effect of Repeated Play on Reputation Building: An Experimental Approach’, International Organization, Vol. 65, No. 2 (2011), pp. 343–65; Gregory Miller, The Shadow of the Past: Reputation and Military Alliances Before the First World War (Ithaca, NY: Cornell University Press, 2012); Timothy Peterson, ‘Sending a Message: The Reputation Effect of US Sanction Threat Behavior’, International Studies Quarterly, Vol. 57, No. 4 (Dec. 2013), pp. 672–82; Alex Weisiger and KerenYarhi-Milo, ‘Revisiting Reputation: How Past Actions Matter in International Politics’, International Organization, Vol. 69, No. 2 (2015), pp. 473–95;Frank Harvey and John Mitton, Fighting for Credibility in Syria: US Reputation Building in Asymmetric Conflicts, 1991–2013. Unpublished manuscript (2015).

18. Paul Huth, ‘Reputation and Deterrence: A Theoretical and Empirical Assessment’, Security Studies, Vol. 7, No. 1 (1997), p. 83, emphasis added.

19. Jonathan Mercer, for his part, extracts the same passage as Hopf, but places it within his text (p. 5).

20. Taken from Thomas Schelling, Arms and Influence (New Haven, CT: Yale University Press, 2008 [1966]) pp. 124–25.

21. Ibid., pp. 55–6.

22. Huth, ‘Reputation and Deterrence, p. 84, emphasis added.

23. Ibid., p. 84.

24. Ibid., p. 85.

25. Ibid., p. 85.

26. Press, Calculating Credibility, p. 3, emphasis added.

27. For critiques of Press see Dafoe, Renshon, and Huth (2014); Weisiger and Yarhi-Milo (2015); Walter (2006); Tingley and Walter (2011); Tomz (2011); and Harvey and Mitton (2015).

28. Hopf, Peripheral Visions, pp. 2–3.

29. Schelling, Arms and Influence, p. 55.

30. Mercer, Reputation and International Politics, p. 18.

31. Ibid., p. 19.

32. Similar to Press’ emphasis in Current Calculus theory on power and interests, Mercer's posits that adversary's often make ‘situational attributions’ such that behaviour is explained by reference to the dynamics of the case at hand. ‘A situational attribution’, Mercer writes, ‘means that most people in the same situation would behave similarly'. As such, ‘situational attributions do not have cross-situational validity’; that is, they preclude the possibility that past behaviour or reputations are relevant (pp. 6–7).

33. Mercer, Reputation and International Politics, p. 43.

34. Huth, ‘Reputation and Deterrence’, p. 91.

35. Mercer, Reputation and International Politics, p. 1, emphasis added.

36. Schelling, Arms and Influence, p. 124.

37. Ibid., p. 125.

38. Ibid., p. 125, emphasis added.

39. Ibid., pp. 124–25, emphasis added.

40. Huth, ‘Reputation and Deterrence’, p. 84, emphasis added.

41. Robert Ayson, Thomas Schelling and the Nuclear Age: Strategy as Social Science (London: Taylor and Francis, 2004).

42. Ibid., p. 87.

43. Ibid., p. 87.

44. Ibid., p. 88, emphasis in original.

45. Ibid., p. 88.

46. Schelling, The Strategy of Conflict, p. 5.

47. Ibid., p. 16.

48. Ibid., p. 21.

49. Ibid., p. 21.

50. Ibid., p. 21.

51. Ibid., p. 22, emphasis in original.

52. Ibid., p. 22.

53. Ibid., p. 29.

54. Ibid., p. 30.

55. Ibid., p. 30.

56. Schelling, Arms and Influence, p. 55.

57. Ibid., p. 56.

58. Ibid., p. 56, emphasis in original.

59. Consider, for example, Schelling's discussion at the outset (p. 3) of The Strategy of Conflict: ‘A study of conscious, intelligent, sophisticated conflict behavior—of successful behavior—is like a search for rules of “correct” behavior in a contest-winning sense … we all are, in fact participants in international conflict, and we want to “win” in some proper sense'.

60. Mercer, Reputation and International Politics, p. 25

61. Schelling, Arms and Influence, pp. 56–7, emphasis added.

62. Quoted in Press, Calculating Credibility, p. 2

63. Schelling, it should be noted, was initially supportive of limited involvement in Vietnam but quickly—by 1967—was clear in his belief that America should withdraw. The costs, in that case, were simply too much and could not justify continued involvement.

64. Anne-Marie Slaughter, ‘Stopping Russia starts in Syria’.

65. Dafoe, Renshon, and Huth, ‘Reputation and Status’, p. 372, emphasis added.

66. Paul F. Diehl, The Dynamics of Enduring Rivalries (Urbana: University of Illinois Press, 1998), p. 3.

67. See, for example, Paul Huth and Bruce Russett, ‘General Deterrence Between Enduing Rivals: Testing Three Competing Models’, American Political Science Review, Vol. 87, No. 1 (1993), pp. 61–73; Janice Gross Stein, ‘Deterrence and Learning in an Enduring Rivalry: Egypt and Israel 1948–73’, Security Studies, Vol. 6, No. 1 (1996), pp. 104–52; Russell Leng, ‘When Will They Ever Learn? Coercive Bargaining in Recurrent Crises’, Journal of Conflict Resolution,Vol. 27, No. 3 (Sept. 1983), pp. 379–419; Russell Leng, Bargaining and Learning in Recurring Crises: The Soviet-American, Egyptian-Israeli, and Indo-Pakistani Rivalries (Ann Arbor: Michigan University Press, 2001); ZeevMaoz and Ben Mor, Bound by Struggle: The Strategic Evolution of Enduring International Rivalries (Ann Arbor: Michigan University Press, 2002); William R. Thompson and David Dreyer, Handbook of International Rivalries 1494–2010 (Washington, DC: CQ Press, 2011); Michael P. Colaresi, Karen Rasler, and William R. Thompson, Strategic Rivalries in World Politics: Position, Space, and Conflict Escalation (New York: Cambridge University Press, 2008); Paul F. Diehl, The Dynamics of Enduring Rivalries (Urbana: University of Illinois Press, 1998); and Paul Diehl and Gary Goertz, War and peace in international rivalry (Ann Arbor: University of Michigan Press, 2001).

68. Harvey, ‘Rigor Mortis or Rigor, More Tests’, pp. 699–700.

69. Imagine, for example, that one is involved in an altercation with a bully. In a subsequent encounter with the same bully lessons from the previous confrontation will be germane (though less so if the bully has somehow broken an arm between then and now, in which case his/her diminished capacity will dominate any lessons derived from the first fight). Next, imagine that one witnesses a fight between a bully and one's twin sibling. If confronted by that same bully, the lessons of that prior altercation will be relevant, though not as relevant as if the initial fight had involved one directly. Such lessons will similarly be relevant, though again slightly less so, if the initial fight was between the bully and a sibling of similar age. These lessons are of diminishing relevance if the fight occurs between the bully and a cousin of the same age; a friend of similar size and disposition; a stranger with superficial similarity (height/weight); and so forth until the relevance of the past encounter is virtually zero, one having witnessed the bully beat up a donkey (though this might confirm, in the final analysis, that one is not dealing with a rational actor or, as helpfully pointed out by an anonymous reviewer, that the adversary has a clear preference for violence). The logic of this argument is supported by the findings from Weisiger and Yarhi-Milo: ‘ … we observe that the effects of past actions remain, but are weaker, when the subsequent interaction less closely resembles the dispute in which the country in question earned its reputation'. Weisiger and Yarhi-Milo, ‘Revisiting reputation: How past actions matter in international politics’, p. 492.

70. For those familiar with Huth's 1997 review, it is important here to distinguish the present argument from the one he calls the ‘qualified-interdependence-of-commitments’ position. Huth's position posits that reputations will form when states act against expectations. My own is that reputation must be seen as a component of credibility, neither definitive nor irrelevant, and that working through the particulars of a specific case is necessary to understand the relative impact of reputation for that coercive encounter.

71. Press, Calculating Credibility, p. 160

72. Mercer, Reputation and International Politics, p. 1

73. The JCPOA attempts to mitigate potential Chinese and Russian obstinacy in this regard by allowing an American veto to enforce the re-implementation of sanctions; in practice, however, absent a broader international consensus a return to the full sanctions regime is likely to be difficult, particularly the more (re)integrated Iran becomes into the international economy.

74. Schelling, Arms and Influence, p. 35