States often fight to preserve their credibility. Whether they should or not is a much different question, one that has been debated for decades by scholars of international politics. Keren Yarhi-Milo takes a different approach to the issue in Who Fights for Reputation, which is an excellent contribution to this important literature and one that expands our understanding of how beliefs affect behavior. Yarhi-Milo is less interested in whether it is wise to fight to preserve one’s reputation for resolve than in the variation in leaders’ beliefs on the subject. The key, she reasonably surmises, may well be found in the psychological profiles they bring into office.
Yarhi-Milo’s key explanatory variable is a trait that has its origins in nature rather than nurture: self-monitoring, or the ability to regulate behavior to accommodate different social situations. High self-monitors are those “inclined to modify their behavior strategically in order to cultivate status-enhancing images” (p. 11). Low self-monitors, in contrast, are less likely to take the opinions of others into consideration when making decisions. They are more concerned with aligning their internal beliefs, rather than the views of others, with their external behavior. High self-monitors are those Zeligs willing to change to win the favor of their audience; low self-monitors have a very hard time projecting different versions of themselves. There is substantial variation in self-monitoring among individuals, including among those who inhabit the Oval Office and those who do not. Thus Yarhi-Milo does what often makes political scientists uncomfortable: she crosses a disciplinary boundary to borrow a concept from another field and applies it to her topic. Both fields are better off for her having done so.
Self-monitoring, according to Yarhi-Milo, is directly related to the concern for credibility. High self-monitoring leaders will pay more attention to their reputation for resolve (hereafter abbreviated as R4R, with apologies to R2P). Low self-monitoring presidents, in contrast, will be less willing to fight to maintain their R4R. High and low self-monitors come in both hawkish and dovish forms and are found across the political spectrum.
The theory makes good intuitive sense. Those most concerned with how others perceive them should also be most concerned with how their states are perceived during crises. Yarhi-Milo hardly relies on intuition, however; she employs an impressive array of methods to test her theory, beginning with a series of survey experiments, followed by a large-N statistical analysis of post–Cold War presidential decisions, and finally three in-depth historical case studies (of Presidents Carter, Reagan, and Clinton). Along the way she performs textual analysis of presidential speeches and conducts a survey of 68 presidential historians. Each method produces strong support for the notion that self-monitoring is linked to an actor’s concern for reputation and that high self-monitoring presidents will be more willing to use force to preserve their R4R. By the book’s end, the reader is left fully convinced that a real relationship exists and that no potential research avenue has been left unexplored.
Who Fights for Reputation is an excellent work of social science. It asks interesting and important questions; it lays out reasonable, testable hypotheses; it employs a series of quite diverse methods to evaluate its claims. It provides much to think about and certainly helps explain the variation in presidential concerns for credibility. Although any criticism will seem like nitpicking, there seems little point in having a “critical dialogue” without some limited contention. So I will highlight a few concerns.
The first nit could be picked from the experiments regarding the theory’s “microfoundations.” In a chapter cowritten with Joshua Kertzer, Yarhi-Milo reports the results of an experiment performed in the United States and Israel regarding the connection between self-monitoring in individuals and their willingness to fight to preserve their reputations. The crucial variation comes in the way participants were primed or the manner in which their tasks were described to them. A control group received straightforward instructions; another was told that if the United States failed to remain firm its enemies may doubt its resolve; a third was told the same thing, replacing the word “enemies” with “allies”; and a fourth group was primed with the phrase “the eyes of the international community are on us.” This wording supposedly allows for the sensitivity of participants to the opinions of others, and therefore their concern for reputation, to shine through. Although priming is a common technique in psychology experiments, it is not without perils for political science. The extent to which such phrases actually bring R4R to the forefront of participant concerns in a simulation regarding international politics is not fully clear. There is also a limit to the amount that such experiments can teach us about presidential decision making, because it is not obvious that participants in an experiment, even primed participants, react in ways similarly to national leaders. Jonathan Mercer spoke for many skeptics of experimental political science when he observed, “Bismarck was not an American college sophomore” (Reputation and International Politics, 1996, p. 70). Fortunately it is not necessary for the theory’s microfoundations to be on firm ground for it to be convincing.
Validity again becomes an issue for the quantitative analysis. Yarhi-Milo constructs a dataset to test the correlation between self-monitoring and concern for R4R and runs into some problems operationalizing her concepts. To represent presidential self-monitoring, she uses a score compiled from her survey of historians (which unfortunately we cannot assess, because there appears to have been an oversight regarding the online appendix—its address is not given, and Google was no help finding it). Concern for R4R is much more difficult to measure quantitatively so she uses proxies, both of which are drawn from the Militarized Interstate Dispute (MID) database: the number of MIDs the United States was involved in and the number it initiated. Neither is particularly satisfying. Although there is clearly some connection between resolve and the willingness to fight, MIDs involve much more than presidential concerns for credibility. The rest of the world gets a vote, at the very least. She attempts to control for variation in the security environment by including the worldwide rate of MIDs, but this is not a reflection of US national security concerns. The rate of violent disputes involving the United States is not merely a function of decisions made by its leaders nor is it the equivalent of their concern for resolve. The use of MIDs may be the best available way to approximate R4R, but it is an imperfect one that injects error into the model.
The self-monitoring scores for presidents are based to the extent possible on the standard measurements used by psychologists, but they correlate weakly with popular perceptions of their R4R. Eisenhower, Kennedy, Johnson, and Clinton all score substantially higher than Nixon, for instance, which does not comport with how they ran their foreign policies. I need more detail and explanation to be convinced that these presidents were more concerned with R4R than was Nixon, who prolonged a rather important MID (and engaged in other little ones, like the farcical Mayaguez “crisis”) largely to preserve US credibility and honor. Nixon’s America may well have engaged in fewer MIDs, but that was the result of relatively low levels of Soviet activity, not underattention to R4R.
Overall, the quantitative analysis produces the banalities standard to all MID-related research (“the United States was significantly more likely to be involved in and initiate military coercion against autocracies as well as less powerful countries,” p. 93), rather than persuasive evidence for the theory. The cases are much more interesting. Carter, Reagan, and Clinton were examined because they provide variation on both self-monitoring and hawkishness, and Yarhi-Milo was interested in seeing whether the measures were distinct. Again, one cannot help but think that Nixon would have made a more interesting subject.
Crises give the cases structure. For each presidency, Yarhi-Milo focuses on three to four instances where we might expect R4R to have factored into decision making. Like all good qualitative analyses, the cases follow the same pattern, asking similar questions in similar ways. The choice of crises is a bit puzzling, however. From the Carter era, she chooses three crises that involved the Soviet Union, whereas only one of the four Reagan-era crises did. These choices are certainly important, because the primary target for US reputational concerns during the Cold War was the Soviets, and we might expect different emphasis on R4R based on (1) the seriousness of the crisis and (2) the actors involved. The crises from the Clinton years include the intervention in Somalia and the Taiwan Straits of 1995–96, which were significantly different from both each other and the previous choices. Such wide variation can be a strength clearly, but it also has its weaknesses: Reputation for resolve has no meaning in the absence of an audience, and the composition of the target audience may well affect decision making. The cases here assume that concern for credibility is constant, as if policy makers cared equally about their reputations in the eyes of all observers. At the very least, the choice of crises is underexplained.
At the end of this generally convincing and comprehensive book, the reader is left wondering what the theory would have to say about the last two presidents. The current occupant of the White House is certainly both a high self-monitor and a reputation obsessive-compulsive. His predecessor, however, may pose some challenges. The United States has never had a president so uninterested in fighting for reputation. Like Eisenhower, he was willing to bluff and draw “red lines” that he had no intention of enforcing. “Dropping bombs on someone to prove that you’re willing to drop bombs on someone,” he told The Atlantic, “is just about the worst reason to use force” (Jeffrey Goldberg, “The Obama Doctrine,” The Atlantic, 317(3), 2016). Can it really be said, as Yarhi-Milo’s theory would predict, that Obama was also a low self-monitor? Did he not exhibit behavior that was highly responsive to social cues? Was he not willing to alter his words and actions when addressing, say, predominantly African American audiences?
While various debates over whether credibility is worth fighting for will rage on, Yarhi-Milo has essentially settled any question about who believes it to be so. Perhaps her insights about presidents also apply to scholars; perhaps the literature’s divides are more explicable by psychological traits than by the evidence. Perhaps the effects of self-monitoring have only begun to be explored.
