Abstract
One of the central puzzles in the study of diplomacy is why some interactions between leaders result in positive social bonds, while others are mired in distrust and hostility. Recent research in the field of microsociology, the study of everyday interactions, most notably the pioneering research of American sociologist Randall Collins, suggests several critical ingredients for a successful interaction, including bodily copresence. In this article we interrogate this claim and provide theoretical reasons why textual communication may serve as a proxy for copresence in leader interactions. We demonstrate that while copresence, in the form of face-to-face interaction, is required for strong bond formation, mediated interaction in the form of letters can serve to create weak social bonds. The strength of weak bonds is in the reduction of distrust and the gradual development of trust that can be critical to the de-escalation of crises. Empirically, we explore our argument in two hard cases for social bond formation: the letters exchanged between Nikita Khrushchev and John F. Kennedy during the Cuban Missile Crisis and the interactions, both textual and face-to-face, between Mikhail Gorbachev and Ronald Reagan at the end of the Cold War. We conclude by pointing toward an interdisciplinary research agenda on the determinants of social bond formation in diplomacy.
Introduction
In March 1983, US President Ronald Reagan called the Soviet Union an “evil empire.” Five years later, he was walking alongside Mikhail Gorbachev, shaking hands in the streets of Moscow. How did what had once been a relationship of enmity become transformed into a relationship of cooperation? Reagan and Gorbachev’s own answer to this question was the personal relationship of trust they forged together through first letter writing, a breakthrough face-to-face meeting in Geneva, and follow-up summits in Reykjavik, Washington, and Moscow interspersed with continued communication via letter. The study of face-to-face interaction as an instrument of reassurance diplomacy has been an area of intense recent interest (see, for example, Hall and Yarhi-Milo, 2012; Yarhi-Milo, 2014; Wong, 2016). Intriguingly, however, existing research suggests that diplomats, ambassadors, and heads of state have long been able to create bonds with one another and develop sophisticated social networks, through non-face-to-face modalities of interaction. This creates a puzzle. Existing research highlights, as in the Reagan–Gorbachev case, the critical importance of face-to-face interaction for reassuring adversaries, yet text-based communication also played a supporting role. And historically, this modality of communication appears to have played a singular role in such reassurance.
Prior to the advent of telephony and relatively expeditious international travel, letters, cables, and telegrams represented, in many cases, the only form of interaction heads of state or diplomats would have with one another. The first records of diplomatic communication, the Amarna Letters, sent between the Egyptian court and other kings in the region, encoded signals and meaning in clay tablets (Jönsson, 2000). Much later, in Renaissance diplomacy, Italian ambassadors, diplomats, and leaders engaged in intricate and complex signaling through many channels, including the development of the structured and strategic diplomatic letter. The advent of the telegram in the 19th century ushered in rapidity of speed previously unmatched, causing leaders and diplomats to adapt to a new temporal reality (or as British Foreign Minister Lord Palmerston put it upon receiving his first telegram in the 1804s, “My God, this is the end of diplomacy!” (cited in Jönsson and Hall, 2005: 91). In the modern period, diplomats are often much more likely to interact over social media or instantaneous messaging services on mobile phones than they are to meet physically in person (Bjola, 2015). In short, historically, text communication had to be able to clarify intentions, build trust, and settle disputes, for it was, in some cases, the only way for parties to interact.
This article investigates whether bonds can form in text-based communication as they do in face-to-face diplomacy. We define a process of social bonding as one that involves two individuals changing their understanding of their own and the other’s identity and interests through a process of positive and mutual identification. We argue that existing research in a variety of disciplines, from psychology to social information processing theory, highlight the ways in which text can serve as a stimulus to imagination, empathy, and behavior adaptation, causing readers to place themselves in the shoes of unseen others. Utilizing these insights, we argue that text can serve as a proxy for face-to-face interaction and ultimately produce social bonds. However, the proxy is not a perfect corollary to face-to-face meetings, and therefore the bonds formed are weaker than their face-to-face counterparts.
We assess textual bond formation in two separate Cold War hard cases for social bonding. The first case is the letters exchanged between President John F. Kennedy and his Soviet counterpart Nikita Khrushchev during the Cuban Missile Crisis (CMC). What makes this a hard case is that Kennedy and Khrushchev had met face-to-face at the 1961 Vienna summit, prior to their exchange of letters during the CMC, but this was a negative emotional encounter that created deep distrust between the leaders. Khrushchev’s belief that the youthful US president was weak was a precipitating factor in the US–Soviet crisis over Berlin later that same year (Kempe, 2011; Wong, 2020). Existing explanations of the peaceful resolution of the CMC focus on Khrushchev’s willingness to accede to US demands to remove the missiles because of the asymmetrical balance of conventional (Bundy, 1988) and nuclear capabilities (Kroenig, 2018). However, Lebow suggests a further explanation for the peaceful ending of the crisis that has not figured prominently in previous accounts. He argues that a “unique bond” formed between the two leaders “based in part on having guided the destinies of their respective nations through the most acute confrontation of the nuclear age” (Lebow, 1990: 491). Jervis (2015: 30–31) has gone so far as to claim that trust was critical to the peaceful resolution of the crisis. But neither Lebow nor Jervis have provided a detailed explanation of how the leaders overcame their preexisting relationship of distrust and developed a new bond of trust in the midst of what to date is the most dangerous crisis of the nuclear age.
We intervene by explaining the bond that formed between Kennedy and Khrushchev during the crisis as an outcome of what microsociologists term an interaction ritual. Utilizing a model of bond formation initially proposed by Randall Collins (2004), we demonstrate that the social interaction between the two leaders produced a bond that comprised a relatively low, though important, level of solidarity between Kennedy and Khrushchev. Collins argues that “moral solidarity. . . is another word for ‘trust,’” (2004: xv; see also Boyns and Luery, 2005; Bramsen and Poder, 2018 on solidarity/trust as a successful interaction ritual outcome) and one that can vary in strength, depending on the interaction (Collins, 2004: 105). This leads us to label this positive emotional energy as “trust as solidarity.” However, we argue that the solidarity created in the CMC case was relatively weak, because it lacked a face-to-face element. As a result, the trust between Kennedy and Khrushchev coexisted alongside hedging and risk calculations regarding the current and future intent of the other.
Our second case study represents a similarly difficult one for bonding, as it pitted against each other two leaders of two states in a relation of enmity: Ronald Reagan and Mikhail Gorbachev. The key difference here, however, is a variation in the modality of interaction. As with Kennedy and Khrushchev, there was an exchange of letters, but crucially also several face-to-face interactions. The result, we demonstrate, was a bonding process that resulted in a strong level of solidarity. The element of risk calculation progressively became smaller between Reagan and Gorbachev, leading to a rare type of trust that has been called “suspension” (Wheeler, 2018; see also Möllering, 2001, 2006 who first developed the concept). Here, solidarity is achieved at the highest level, and there is a complete absence of risk calculation between two leaders who have bonded in this way. Our aim with the cases is not to compare them in a focused comparison, but rather highlight the comparative role of letters in explaining how social bonds and levels of solidarity and trust may vary in strength with different modalities of interpersonal interaction.
We proceed as follows. In the next section we briefly introduce the Collins model and existing International Relations (IR) applications of it. We then explore the arguments of those social theorists who argue that social bond formation can develop in non-face-to-face interaction through text-based communication. We probe the importance of bodily copresence in relation to the letters sent between Kennedy and Khrushchev during the CMC and the letters and face-to-face interaction between Reagan and Gorbachev. We pay particular attention to within-case variation and use process tracing techniques to tease out when bonding occurred, or did not occur, in each interaction. The conclusion explores the implications of our argument and identifies future research questions.
Social bonding through text: Overcoming lack of copresence and barriers to outsiders
In previous work (Holmes and Wheeler, 2020), we asked why leaders often “hit it off” and develop “personal chemistry” with one another, whereas other times the interaction remains mired in distrust or apprehension. Building on the work of leading microsociologist 1 Randall Collins, who draws upon a rich tradition in sociology of investigating dyadic or “micro” interactions (see, for example Goffman, 1959), and has developed an interaction ritual model of bond formation, we argued that four ingredients need to be present for a positive bond to develop: bodily copresence, mutual focus of attention (which we conceptualize building on Booth and Wheeler 2008 as shared security dilemma sensibility (SDS)), shared mood (conceptualized as positive identification of interests and humanization), and a barrier to outsiders. 2 The puzzle is how social bonds can develop in text-based communication where one of the ingredients, bodily copresence is not possible and another, barriers to outsiders, is challenged. In this section, we develop an argument, drawing from several diverse literatures, to suggest that leaders in reassurance diplomacy can satisfy both conditions by substituting some of what face-to-face interaction provides through text and excluding outsiders, linguistically and symbolically, through the use of what Collins and Hanneman (1998) have termed “cultural membership capital.”
Turning to the bodily copresence condition first, it is perhaps unsurprising that Collins has been, at times, deeply skeptical regarding the notion that bonding is possible without bodily copresence. For Collins, physical copresence allows for individuals to engage in rhythmic entrainment, where nervous systems become mutually attuned (2004: 64). This, in turn, brings productive effects in establishing shared mood and mutual focus of attention. Intriguingly, however, in subsequent work building upon the original model, his stance has been a bit more optimistic regarding the possibilities for bonding through, for example, digital technology. In a recent interview Collins admitted: “Some areas are genuine challenges to the theory, especially in the era of electronic media. Do you actually have to meet face-to-face to experience intense solidarity?” (quoted in Van der Zeeuw et al., 2017; emphasis ours). Here, Collins opens the door to the possibility of some bonding occurring through non-bodily copresence. Nevertheless, he goes on to note that “People in important economic and political positions almost always meet face-to-face for important business. Their success depends on being able to exert emotional influence over somebody and to pick up clues of what they are dealing with” (quoted in Van der Zeeuw et al., 2017). Collins’ intervention leads us to develop the notion that social bonds might be differentiated by strength, where the strongest bonds will be products of interaction rituals where all four ingredients are present and the outcome variables exist at relatively intense levels due to rhythmic entrainment and the feedback processes that strengthen shared mood and mutual focus of attention. Weaker bonds, on the other hand, are products of interaction rituals where bodily copresence is absent and barriers to outsiders challenged, but mutual focus of attention and shared mood are present. We begin by investigating reasons why text-based communication is believed to result in weaker bonds and then turn to theoretical reasons why some of the limitations of text-based communication can be overcome.
Text-based communication makes bonding difficult because of the lack of ability to perceive, and respond to, others. The temporal nature of text-based communication, even digitally, means that such synchronicity is challenged. As Collins contends, “In the strongest forms of social solidarity, the body is a necessary ingredient” (quoted in Van der Zeeuw et al., 2017). This is because text-based communication “lack[s] the flow of interaction in real time; even if electronic communications happen within minutes, this is not the rhythm of immediate vocal participation. There is little or no buildup of focus of attention in reading an email, no paralinguistic background signals of mutual engrossment” (Collins, 2004: 63–64). Text-based communication is therefore asynchronous one-channel communication with relatively low information richness, low feedback, little or no propinquity, and low entrainment. As DiMaggio et al. (2019: 84) put it, this is one reason why the Internet has not replaced face-to-face interaction: “. . . the Internet matters not as a site of interaction rituals but because it facilitates bringing people together, increasing the speed and scope of face-to-face mobilization.” To the extent that digital technology aids bonding, it seems to be in a supportive role.
On the other hand, scholars in a variety of fields and disciplines have long noted the ability of text to substitute, in some ways, for “real” experience. Samuel Taylor Coleridge in 1817 coined “willing suspension of disbelief,” in order to explain why people enjoy reading fiction or poetry despite the fact that the activity seemed to include little more than reading words on paper (see DiMaggio et al., 2019). Imagination, or placing oneself “into” the text, allows individuals to have experiences that involve many of the same psychological effects as those that occur with the body in “real life.” For example, Gerrig (1993) argues that when we read text we engage in a mental process of “narrative transportation”: “Someone (‘the traveler’) is transported, by some means of transportation, as a result of performing certain actions. The traveler goes some distance from his or her world of origin, which makes some aspects of the world of origin inaccessible. The traveler returns to the world of origin, somewhat changed by the journey” (Gerrig, 1993: 10–11). Anyone who has “gotten lost in a story” can attest to the aptness of the metaphor of travel, but for psychologists who study reading and imagination, the notion of transporting one’s mental state from one place to another is an important aspect of how text provides meaning. Green and Brock (2000), for instance, demonstrate that readers are more likely to believe a text, or change their beliefs, when they are more engrossed in the story. As DiMaggio et al. (2019) point out, this is tantamount to “a kind of entrainment in absentia.”
Further, a wealth of evidence suggests that through reading individuals can improve “theory of mind” capabilities, such as the ability to understand the emotional states of others, or empathy because reading about others requires it (see Kidd and Castano, 2013). “Empathy is at the heart of the novel” (Nafisi, 2003: 111). When reading Tolstoy’s Anna Karenina, for example, one is required, in some sense, to have an empathic response: one “considers what it is like to be caught up in Anna’s situation, one considers her quandaries, her perplexities and moral dilemmas, and arguably as a result, one is able to experience the dread and hopelessness of her situation” (Margolis, 1984: 58). Reading fiction is therefore not only narrative transportation, but an exercise in empathy. And, as a series of recent studies have shown, skills in empathy practiced through reading may be beneficial in social life as well. As The New York Times recently put it, “For Better Social Skills, Scientists Recommend a Little Chekhov.” Practicing empathy through fiction and the resultant imagination, it turns out, aids individuals in reading others and leads to better understanding of their emotional and cognitive perspectives (Kaufman and Libby, 2012; Keen, 2007; Kidd and Castano, 2013). Yet, it may be one thing to argue that fictional texts can produce empathy; it is quite another to argue that non-fictional portrayals produce similar outcomes. However, there is evidence that individuals can assess mental and emotional states of others through non-fictional text communication by “reading between the lines” and deducing the thoughts and emotions of others through interpretation of the text (see Engel et al., 2014; Alecsandru, 2013 who uses the CMC as a case study). The subfield of “presence studies” in contemporary media theory has investigated how letters, emails, and other forms of text-based communication can create a sense of co-presence through intense intimacy and a “sense of the other’s presence.” As Esther Milne (2013: 2) puts it, “[p]resence is a term that need not refer always to material, corporeal presence. Rather, presence is an effect achieved in communication (whether by letters, postcards or email, for example) when interlocutors imagine the psychological or, sometimes, physical presence of the other.”
Relatedly, insights from social information processing theory (Walther, 1992), which is an interdisciplinary approach that draws upon psychology and cognitive science, are insightful. One of the early findings in the study of computer-mediated communication suggested that mediated communication is not necessarily detrimental to interpersonal relations, but rather might enhance a sense of connection between individuals. The reason has to do with behavioral adaptation to the medium. “[W]hen using a medium with limited availability of visual cues or other channels of communication, interlocutors adapt their behaviors in order to connect more effectively using that particular medium” (Sherman et al., 2013). For example, individuals engaged in text-based communication will employ behavioral strategies to make up for the deficit of visual cues. With digital technologies these might include the use of emoticons or images in order to convey affect; with text-only mediums these might include taking more care to include words devoted to explain one’s mental state or the use of words with particular sentiment affiliation—such as “empathic constructions” or “emotional speech”—in order to convey affectation (see Sherman et al., 2013). Joseph Walther (1992, 1996), the original creator of social information processing theory, notes that in some situations text-based communication can be superior to relationship development face-to-face. Walther (1996: 33) argues this occurs “when users experience commonality and are self-aware, physically separated, and communicating via a limited-cues channel that allows them to selectively self-present and edit; to construct and reciprocate representations of their partners and relations without the interference of environment reality.” Walther (1996: 33) privileges the temporal aspect of the interaction, noting that it is “Perhaps more so when this communication is asynchronous and/or when the [computer-mediated communication] link is the only link there is.” One noted complication, however, with non-immediate forms of communication, such as letters, is accounting for the spontaneity of thought of collective effervescence, something more easily measured with instantaneous message delivery in a laboratory than with letters sent over a period of time.
The upshot from social information processing theory and presence studies is that the medium of text need be “no less personal than FtF” (Walther, 1996: 33). If individuals utilize words and specific constructions to serve as proxies for the visual cues obtained in bodily copresence, then it is conceivable that bonding may take longer, but ultimately is possible, to some degree, through text. While there have been few direct and systematic applications of the Collins model to text-based communication, Ling’s (2008a) study of cell-phone usage is instructive. Utilizing interviewing and ethnographic methods, Ling finds that while face-to-face interaction creates stronger solidarity, cell phone usage (both text messaging and voice calls), strengthens solidarity. “I am not suggesting that co-present interaction is losing its place as a central context for the development of social cohesion. . . It is clear that face-to-face or body-to-body. . . interaction is the most efficient way of generating Collins’ focused emotions and mutual recognition. . . I do, however, assert that social bonds can be maintained and nurtured through mediated interaction. Once a bond is forged, however, mediated interaction is often as effective as co-present interaction in the development and maintenance of the interaction” (Ling, 2008a: 6–11). In other work (2008b: 170–174), Ling goes further and suggests the possibility for weak bonds to form through mediated interaction (such as cell phone text messaging) and then strengthened with subsequent face-to-face interaction. While Ling does not identify one central mechanism by which the mediated interaction substitutes for bodily co-presence, his rich empirical work suggests a strong role for imagination and “narrative transportation,” placing oneself closer to the interlocutor through the type of imaginative empathy discussed earlier. Or, as DiMaggio et al. (2019: 88) put it, it is conceivable that individuals “imaginatively construct their interlocutors by importing prior experience into their reception of the utterances they read.”
Overcoming copresence limitations is not the only problem for text-based interaction rituals. Barriers to outsiders implies that individuals in an interaction are able to delineate those in the interaction from those that are not. In bodily copresent rituals this may be relatively straightforward. Diplomats and leaders are often accompanied by translators, note-takers, advisors, photographers, and so on. While each of these individuals tend to occupy the same physical space, leaders can exclude them from the interaction itself. Sometimes this exclusion is subtle, such as a whispered conversation in the back of the room. Other times, it is more overt, such as the thirty minutes North Korean leader King Jung-un and South Korean President Moon Jae-in spent alone sitting on a park bench in Panmunjeom, North Korea, at the May 2018 Inter-Korean Summit. In either case, the ability to exclude who is involved in the interaction is achieved.
Crucially, however, barriers can also be created through non-material means as well. Discursively, individuals can utilize linguistic stratagems whose true meaning might only be ascertained by individuals within a particular interaction. Overt examples might involve a cryptolect, argot, or anti-language, where a dyad or group utilizes either new language or existing language in new ways, to exclude outsiders from grasping meaning. The language of medicine, spoken informally by physicians, for example, allows for easily communicating potentially upsetting news between doctors while shielding patients from its meaning (Hukill et al., 1961: 147). More subtle examples include reference to ideas or concepts that only those in an interaction could fully understand or appreciate, perhaps because of the affect-laden nature of their meaning. High-performance surgical teams, for example, communicate not only in cant, or code, but by ascribing particular understanding to language (de Rond, 2017). Meaning is understood by group membership; only individuals with the responsibility of the lives of other human beings can fully appreciate the meaning of the terms used. Carol Cohn’s (1987) groundbreaking work on the symbolic language used by nuclear defense officials, for example, highlights the group-membership aspect of particular (in this case very gendered) diction that is only completely understood by those within the narrow confines of group borders.
This cultural membership capital, or knowledge of group symbols, represents a form of barrier creation, as outsiders will be unfamiliar with the symbols that have developed over time in the group. We argue that this type of barrier creation could also exist in text-based communication. The easiest case would be where two leaders retained exclusive control over the letters they sent to each other, as with private individuals who may choose to keep a bond forged with a significant other through letters and other text-based communication secret from partners, loved ones, and close friends. However, this type of overt barrier to outsiders is rare in diplomacy, and in both of our cases, the letters that were sent were drafted by small groups of close and trusted advisors. Nevertheless, the absence of an overt barrier does not mean that barriers did not exist between the leaders and their team of drafters. Bringing these insights together, a testable proposition develops. While bodily copresence and a clear barrier to outsiders are required ingredients for a strong social bond, individuals are capable of forging weaker social bonds and maintaining those bonds in the absence of bodily copresence and situations where overt barrier construction may not be possible. Individuals can substitute for bodily copresence by being present in the interaction through careful and precise textual self-representation of their beliefs, mental states, and emotions, while imaginatively placing oneself into the position of the other through the exercise of empathy. A barrier to outsiders can be created through cultural membership capital. In short, while individuals may not be able to physically interact together in private, they can still interact and develop mutual focus of attention, shared mood, and exclude outsiders.
The Cuban Missile Crisis: Social bond formation through the exchange of letters
The CMC represents a fascinating puzzle of how two leaders, on the brink of nuclear war, came through the exchange of letters to trust one another such that they could de-escalate the crisis. In this section we focus on the 10 letters exchanged between Kennedy and Khrushchev, beginning with Kennedy’s first letter on 22 October 1962 announcing the naval “quarantine” and ending with Khrushchev’s letter of October 28 publicly accepting the terms of Kennedy’s letter of October 27. We explain this puzzle by arguing that during these six days of text-based interaction, a weak bond developed between the two leaders that contributed significantly to the peaceful ending of the crisis.
In assessing sources, we adopt Alexander George’s dictum that “requires. . . each document be evaluated relative to what is known about the actors, their intentions, their interactions, and the situation they found themselves in” (quoted in Thies, 2002: 357). This requires placing the memoir or other document within the greater social context and set of relationships in which it was created. In our investigation of the CMC, we have triangulated the data based on several sources. Although we have drawn on Khrushchev’s memoir and his reflections in Khrushchev Remembers (1971), we have also relied on contemporaneous notes, as well as oral and critical oral histories. 3 We have also used the official record of the Foreign Relations of the United States (FRUS) to triangulate between divergent pieces of data.
Crucial to our empirical strategy is the ability to demonstrate that the ritual ingredients come together to form the bond, rather than are a result of the bond. For instance, with respect to the theory, one might ask why the sharing of emotional states is not an effect of a social bond rather than a cause of it. In order to avoid concerns of endogeneity—that it is the bond that creates shared mood (humanization and positive identification of interests) and mutual focus of attention (shared SDS)—it is crucial that the process of bond formation be teased out in order to carefully establish causality. This requires detailed and incremental step-by-step analysis of the interaction to establish how shared mood and mutual focus of attention obtained, generating collective effervescence, and ultimately solidarity. 4
Before turning to the crisis itself, it is important to recall the baseline relationship of the two leaders heading into October 1962. Four months earlier, the two leaders met in Vienna for a summit that was perceived differently by each side. Kennedy, perhaps because he was unprepared for what was to come in the interaction, remarked afterward that “[Khrushchev] just beat the hell out of me,” whereas for his part Khrushchev was in “bubbling good spirits” (Reynolds, 2009: 211), because he had been able to dominate Kennedy and, from all accounts, won most of the arguments. The two leaders therefore begin their interactions over Cuba not from a position of having a positive relationship, but rather from an attitude of deep distrust and suspicion of each other’s intentions, making this a hard case for bonding.
In the leadup to the crisis, Khrushchev had made what Kennedy viewed as “solemn assurances” (October 25, FRUS 1961–1963: Document 64) that “no offensive weapons were being sent to Cuba” (October 25, FRUS 1961–1963: Document 64; Blight et al., 1993: 463–464). However, on 14 October 1962, a U-2 reconnaissance aircraft took photos of Soviet activity in Cuba that showed the Soviet Union was secretly installing Medium-Range Ballistic Missiles and preparing sites for the deployment of Intermediate-Range Ballistic Missile Kennedy was informed of Khrushchev’s action on October 16 and he said to McGeorge Bundy, Special Assistant to the President for National Security Affairs, “He can’t do this to me” (Dobbs, 2008: 6). The weak bond that we will argue developed between Kennedy and Khrushchev has to be viewed against Kennedy’s baseline feeling at the beginning of the crisis that, in Soviet Ambassador Anatoly Dobrynin’s words, Khrushchev had perpetrated “a deliberate deceit” (1995: 81).
In his first letter to Khrushchev on October 22, Kennedy sent him an advance copy of the speech he was to give at 1900EST (FRUS, 1961–1963: Document 60). In his speech, Kennedy announced a naval “quarantine” around the island of Cuba (Kennedy, 1962). The quarantine was aimed at signaling US resolve and he called upon the Soviet leader to turn back from this course of action and dismantle the missile sites.
Although in a letter sent two days later, the Soviet leader described Kennedy’s actions as an “ultimatum” and rejected his “arbitrary demands” (FRUS, 1961–1963: Document 63); this is also the moment in the crisis when we see the first signs of an imaginative identification between the two leaders and the beginnings of social bond formation. Khrushchev wrote: You, Mr. President, are not declaring a quarantine, but rather are setting forth an ultimatum and threatening that if we do not give in to your demands you will use force. Consider what you are saying! And you want to persuade me to agree to this! No, Mr. President, I cannot agree to this, and I think that in your own heart you recognize that I am correct. I am convinced that in my place you would act the same way. When you confront us with such conditions, try to put yourself in our place and consider how the United States would react to these conditions. I do not doubt that if someone attempted to dictate similar conditions to you—the United States—you would reject such an attempt. And we also say—no (FRUS, 1961–1963: Document 63).
In his letter, Khrushchev reached out to his US counterpart and invited him to imagine how he would feel had Khrushchev issued a similar ultimatum: He asked, “How would you have reacted to this? I think that you would have been indignant at such a step on our part. And this would have been understandable to us” (FRUS, 1961–1963: Document 63). Khrushchev wrote, “No, Mr. President, I cannot agree to this, and I think that in your own heart you recognize that I am correct. I am convinced that in my place you would act the same way” (FRUS, 1961–1963: Document 63).
Kennedy’s response was sent the following day (October 25) and in it, the US leader emphasized to Khrushchev his sense of betrayal and regret, urging the Soviet leader take the necessary steps, crucially the removal of the missiles, to end the crisis. Kennedy wrote: In early September I indicated very plainly that the United States would regard any shipment of offensive weapons as presenting the gravest issues. After that time, this Government received the most explicit assurance from your Government. . . both publicly and privately, that no offensive weapons were being sent to Cuba. And then I learned beyond doubt what you have not denied—namely, that all these public assurances were false and that your military people had set out recently to establish a set of missile bases in Cuba. I repeat my regret that these events should cause a deterioration in our relations. I hope that your Government will take the necessary action to permit a restoration of the earlier situation” (FRUS, 1961–1963: Document 64).
Kennedy’s deep feelings of betrayal and regret at Khrushchev’s action drove out the exercise of SDS for his Soviet counterpart that Khrushchev had appealed for in his previous letter. However, as the fear increased for each leader that events were slipping out of control, a positive moral identification would grow between the two leaders as each came to positively identify with the other’s responsibility to avoid nuclear war.
Khrushchev’s response came the following day in the most important letter exchanged between the two leaders during the crisis, ultimately providing the basis for its de-escalation. This letter was a critical one because it created the mutual focus of attention (shared SDS) and shared mood (positive identification of interests and humanization) between Kennedy and Khrushchev that are essential ingredients in a process of social bond formation. What is striking about this letter is how Khrushchev captured the emotion of fear that Kennedy was experiencing at the same time. This letter provides clear evidence of narrative transportation. Khrushchev framed the crisis, and the risk of nuclear annihilation it had created, as a responsibility they both shared. He wrote, “the preservation of world peace should be our joint concern, since if, under contemporary conditions, war should break out, it would be a war not only between the reciprocal claims, but a world wide cruel and destructive war” (FRUS, 1961–1963: Document 65). He told Kennedy, referring to Kennedy’s letter of October 25 that, “I got the feeling that you have some understanding of the situation which has developed and (some) sense of responsibility. I value this.” He went on, “I see, Mr. President, that you too are not devoid of a sense of anxiety for the fate of the world. . . and of what war entails” (FRUS, 1961–63: Document 65). Khrushchev imaginatively transported himself into Kennedy’s shoes and invited the American president to do the same with regard to him. The Soviet leader appealed to Kennedy in this letter to show restraint in language that is directed to him as a fellow human being and not just as the office of President of the United States. He and Kennedy had to find a peaceful way out of the crisis otherwise they would “come to a clash, like blind moles, and then reciprocal extermination will begin” (FRUS, 1961–1963: Document 65). Khrushchev’s desperation that the two military machines he and Kennedy sat atop of were heading for a cataclysmic collision can be seen in his emotional plea to Kennedy that: “If. . . you have not lost your self-control and sensibly conceive what this might lead to, then, Mr. President, we and you ought not now to pull on the ends of the rope in which you have tied the knot of war, because the more the two of us pull, the tighter that knot will be tied. And a moment may come when that knot will be tied so tight that even he who tied it will not have the strength to untie it, and then it will be necessary to cut that knot, and what that would mean is not for me to explain to you, because you yourself understand perfectly of what terrible forces our countries dispose. (FRUS, 1961–1963: Document 65)
Faced with the possibility of a nuclear conflict, Khrushchev offered Kennedy a way out of the crisis: the Soviet Union would remove the “armaments which you call offensive” if “assurances were given by the President and the Government of the United States that the USA itself would not participate in an attack on Cuba and would restrain others from actions of this sort” (FRUS, 1961–1963: Document 65).
In his memoir of the crisis, Bundy reflected that given the time difference, Khrushchev’s letter “did not get the same-day answer for which it cried out” (Bundy, 1988: 443–444). The problem was that before Kennedy could reply to Khrushchev’s letter, the Soviet leader sent a second letter, this time broadcast on Radio Moscow at 5 p.m. local time. Now Khrushchev upped the ante by tying the promise to remove the missiles from Cuba to the removal of US Jupiter missiles that were based in Turkey. Despite the Soviet leader escalating his demands in his public letter of October 27, the limited solidarity that was developing through their communication was in further evidence in this letter.
Khrushchev publicly appealed to Kennedy to join him in exercising their shared responsibility to avoid a nuclear conflict. He wrote: Then the people could breathe more easily, considering that statesmen charged with responsibility are of sober mind and have an awareness of their responsibility combined with the ability to solve complex questions and not bring things to a military catastrophe (FRUS, 1961–1963: Document 66).
A weak social bond was forming between the two leaders as each came through their written correspondence to imaginatively realize that the other was experiencing the same anxieties about the specter of nuclear annihilation. The two leaders, through their textually mediated interaction, were able to create three of the ingredients necessary to develop weak solidarity around the idea, in the later words of Reagan and Gorbachev, that “a nuclear war cannot be won and must never be fought.” If this knowledge had been transparently obvious to Kennedy and Khrushchev from the outset, then they would not have needed to write these letters. It was their text-based interaction which created the mutual focus of attention and shared mood that produced the bond. Kennedy and Khrushchev experienced this bond, and the solidarity it comprised, as in the words of Sachs (2013: 36) a “feeling [of] responsibility that only the other could understand,” invoking their shared group membership. Evidence for this can be seen in Khrushchev thanking Kennedy in his letter of October 28 for his “sense of proportion,” “understanding,” and “responsibility,” in handling the crisis (FRUS, 1961–1963: Document 68).
Both of Khrushchev’s letters on October 26 and 27 gave expression to his growing fear that events were slipping dangerously out of his and Kennedy’s control (Markwica, 2018: 171), a position shared by Kennedy whose fear was amplified when he was informed in the early afternoon of October 27 that a U2 had inadvertently strayed into Soviet airspace and then later that same afternoon that a U2 had been shot down over Cuba (Blight and Lang, 2012: 137; Dobbs, 2009: 230–293). Khrushchev gave voice to Kennedy’s fears at this time in his letter to Kennedy the following day agreeing to the removal of the missiles. He asked about Kennedy’s relation to the U2 incident: “How should we regard this. What is this: A provocation? One of your planes violates our frontier during this anxious time we are both experiencing, when everything has been put into combat readiness. Is it not a fact that an intruding American plane could be easily taken for a nuclear bomber, which might push us to a fateful step?” (FRUS, 1961–1963: Document 68).
The shared mood of fear between the two leaders enabled them to navigate their way out of the crisis. Kennedy agreed on October 27 in a letter, publicly broadcast so that it would reach Khrushchev before events could spin any further out of control, to Khrushchev’s offer in his letter of October 26 that, “If assurances were given by the President and the Government of the United States that the USA itself would not participate in an attack on Cuba. . . Then too, the question of the destruction, not only of the armaments which you call offensive, but of all other armaments as well, would look different” (FRUS, 1961–1963: Document 67).
The trust that developed between Kennedy and Khrushchev as a result of their text-based interaction was a limited one. Nevertheless, Jervis (2015: 30) argued that the final settlement of the crisis relied on “the trust that Kennedy placed in Khrushchev.” This was a remarkable transformation in their relationship when it is remembered that Kennedy began the crisis feeling betrayed by Khrushchev. As we discuss in the Appendix, Kennedy secretly agreed through the face-to-face meeting between his brother and Dobrynin on Saturday evening (October 27) that he would remove the Jupiter missiles in Turkey within a few months, but he trusted the Soviet leader to keep the deal confidential. Kennedy had no guarantee that Khrushchev would not go public, but Khrushchev honored his trust, never mentioning the deal in public or in his later memoir and reflections. Kennedy also trusted Khrushchev when he agreed to remove the missiles and what US intelligence presumed rightly to be the nuclear warheads that had been deployed there. The Joint Chiefs of Staff (JCS), for example, viewed the Soviet offer as a ruse, worrying that Khrushchev would seek to hide some of the missiles and warheads (Blight and Lang, 2012: 143; Dallek, 2013; Freedman, 2002: 219; Stern, 2003: 385). The views of the JCS illustrate how the two leaders maintained a barrier to outsiders here in that the JCS were not part of the process of social bonding and trust emergence between Kennedy and Khrushchev, and hence they calculated the risks very differently to their political leaders.
Yet, trust was not only one-way as suggested by Jervis. Khrushchev also had to trust Kennedy to honor his pledge not to invade Cuba and remove the Jupiters as he did a few months later. According to his son Sergei Khrushchev’s reflection on the 40th anniversary of the crisis, Khrushchev returned “constantly to the premise that Kennedy’s word should be trusted” (Khrushchev, 2001: 623; see also Dobrynin, 1995: 90). Khrushchev later recalled, “I trusted him in the sense that he would keep his word” (Khrushchev, 2007: 348). At the same time, Khrushchev showed his awareness that there were others in the US government who believed Kennedy was being too weak in not launching an attack against Cuba (Blight and Lang, 2012: 141–143). Khrushchev wrote in his letter of October 28 that, “I regard with respect and trust the statement you made in your message of 27 October 1962 that there would be no attack, no invasion of Cuba. . . [but] there are irresponsible people who would like to invade Cuba now and thus touch off a war” (FRUS, 1961–1963: Document 68).
The solidarity between Kennedy and Khrushchev may have been “unusual” (Jervis, 2015: 31), but it would be misleading to exaggerate the strength of the bond. The process of bonding through text-based interaction coexisted uneasily during the crisis with continued hedging and risk calculation on the part of both leaders. We have documented in the Appendix Khrushchev’s fears that Kennedy was preparing to attack Cuba, but what Kennedy did not know it seems, or at least fully realize, is that there were Soviet tactical nuclear weapons on Cuba and whatever orders Soviet commanders may have had not to use them without direct orders from Moscow, there was always the possibility of this eventuality in the event of a US invasion.
As we discussed earlier, Lebow (1990: 491) has recognized that a personal bond formed between Kennedy and Khrushchev, but he does not provide an argument for how this developed. We have highlighted the importance of the letters that were exchanged between Kennedy and Khrushchev; it was the textually mediated interaction that, in substituting for bodily copresence and creating a subtle barrier to outsiders, made possible a level of solidarity between Kennedy and Khrushchev that was so crucial to the peaceful ending of the crisis.
The end of the Cold War: Social bonds face-to-face and through letters
One of the central puzzles of the Cold War is how the conflictual relationship between the United States and Soviet Union could be transformed in such short order. In 1983, Ronald Reagan famously called the Soviet Union the “evil empire” and less than five years later was extolling the partnership in peace the two states had forged, including a historic agreement to ban a large swathe of nuclear weapons. In addressing this puzzle, scholars have investigated a number of factors, from environmental (such as the deteriorating economic condition of the Soviet Union) to dispositional (Gorbachev’s “new thinking;” Reagan’s profound fear of nuclear war etc.). More recently, the interactional element has been a site of inquiry. Holmes (2013, 2018) argues, for example, that face-to-face diplomacy between Reagan and Gorbachev allowed both leaders to clarify their intentions, ultimately resulting in a level of reassurance that allowed Reagan and Gorbachev to be comfortable taking on more vulnerability vis-a-vis the other leader. Wheeler (2018) demonstrates that in addition to intention understanding, the two were able to bond, and eventually, reach a level of trust where both leaders suspended risk calculations in relation to the other’s intent and integrity (“trust as suspension”). Ultimately, this relationship of “bonded trust” (Wheeler, 2018: 45) meant that the two leaders could effectively disregard hawks in their own administrations and focus on transforming the relationship of the two states from a competitive one to one characterized by partnership.
The importance of the Reagan–Gorbachev case in our argument is twofold: first, it provides evidence for bonding and trust-building face-to-face, and second, it supports our core claim that while weak bonds can form through non-face-to-face interaction, as in the exchange of letters in the CMC, strong bonds depend on face-to-face interaction. In contrast with the CMC, the letters exchanged between Reagan and Gorbachev both preceded and followed the face-to-face diplomacy. As such, the series of correspondences, 38 letters in total, that beginning in March 1985 Reagan and Gorbachev sent each other, provide significant insight into how both sides sought to build and maintain social bonds.
Reagan writes to Gorbachev the same day that the latter takes office as General Secretary. The letter is noteworthy both for its conciliatory tone as well as the deep desire conveyed to build a personal relationship. Reagan notes that, “Our differences are many, and we will need to proceed in a way that takes both differences and common interests into account in seeking to resolve problems and build a new measure of trust and confidence. But history places on us a very heavy responsibility for maintaining and strengthening peace, and I am convinced that we have before us new opportunities to do so” (Saltoun-Ebin, 2010: 268). Gorbachev’s reply echoes the conciliatory tone and notes the value of exchanging letters: “We value the practice of exchanges of views between the leaders of our two countries . . . In this context I attach great importance to the exchange of letters” (Reagan Library, Head of State File). Gorbachev agreed with Reagan’s view, expressed in his letter, regarding the “new opportunities which are opening up” for better relations and emphasized his and Reagan’s shared responsibility—echoing the Kennedy–Khrushchev exchanges—“not to let things come to the outbreak of nuclear war which would inevitably have catastrophic consequences for both sides.” Gorbachev ended by suggesting that a face-to-face meeting was essential if the two leaders are to reach “mutual understanding on the basis of equality and taking account of the legitimate interests of each other.”
Reagan shared Gorbachev’s intuitive feeling that a face-to-face meeting was necessary for a breakthrough in the conflict. He reflected in his memoir that if he could only get “in a room alone and try to convince him [a Soviet leader] we had no designs on the Soviet Union and Russians had nothing to fear from us” (Reagan, 1990: 589) then it might be possible “to break down the barriers of mistrust that divided our countries” (Reagan, 1990: 12). As both Reagan and Gorbachev had anticipated, their first summit meeting in Geneva in November 1985 produced the advance both was seeking. Officials present at the meetings in Geneva note how the summit encouraged both leaders to firmly believe that they could work together (Matlock, 2004: 169, 173; Shultz, 1993: 606–607). Reagan recalled that as he “shook hands with Gorbachev for the first time,” and “looked into his smile,” he sensed that he “had been right [to think that a breakthrough might be possible] and felt a surge of optimism” (1990: 12). Recalling in his memoir, Gorbachev noted that he and Reagan had been able to develop an “intuition” about the other (Gorbachev, 1996: 405, 408).
The importance of the Geneva meeting is that Reagan and Gorbachev came to positively identify with the other’s anxiety (shared mood) regarding the risk of nuclear war. This does not mean that they did not disagree profoundly on big positional/tactical issues, such as the fate of the Strategic Defense Initiative (SDI). However, what is critical is that beneath these important disagreements, there was over the course of five hours of meetings a growing mutual focus of attention on the part of Reagan and Gorbachev that there could be no security if each acted on the belief they could only be secure if the other was insecure. Put differently, Reagan and Gorbachev were able to convey to each other their “intention and capacity” (Booth and Wheeler, 2008: 7) to exercise SDS. Gorbachev and Reagan were able to impress upon the other that not only did they understand that the two states they represented may each be culpable in the insecurity of the other, but crucially they both came with peaceful intent.
Given the importance our model places on humanization in social bond formation, it is significant that the process of realizing their interests were aligned and they were in this together occurred in the context of Reagan and Gorbachev coming to see the humanity in one another. The two leaders, their wives, and closest advisors shared dinners together during the summit and on these occasions the two leaders who held the power of global nuclear devastation in their hands, talked not of what divided them, but instead bonded over shared experiences of life as a father and a grandfather (Wheeler, 2018: 155).
The letters that were exchanged between the two leaders after the Geneva meeting nurtured the bond that had developed in and through their first face-to-face meeting. However, their emerging solidarity showed some signs of faltering as 1986 progressed. Gorbachev perceived a growing gap between Reagan’s sincere intent and integrity towards him and his capacity as president of the United States to adopt policies that challenged the “enemy image” of the Soviet Union that had guided previous US policy. Gorbachev became convinced that if real progress was to be made, it had to come from the top, and critical to this was another face-to-face meeting between him and Reagan.
The meeting in Reykjavik in October 1986 demonstrates the importance of bodily copresence in giving renewed energy and dynamism to a process of bonding. This was the turning point in the Reagan–Gorbachev relationship because the process of bonding (mutual focus of attention and shared mood) that developed during the summit produced a level of solidarity between the two that meant both were in a mental state of trust as suspension. As noted earlier, the trust that emerges out of a strong bond of this kind exists when neither individual feels the need to hedge against the intent and integrity of the other. The contrast is clear with the weak bond that developed between Kennedy and Khrushchev through their textual mediated interaction in the CMC, where both leaders continued to hedge against the future intentions of the other. Below, we set out what might count as evidence against our claim that a strong bond of trust existed between Reagan and Gorbachev (trust as suspension) and why this evidence is not persuasive in this case.
Empirical indicators of suspension
We identify two empirical indicators of suspension: (i) discursive materials that evidence a process of bonding and (ii) the suspension of risk calculations by one actor in relation to the intent and integrity of another actor with whom that actor has created a bond of high solidarity. If there is discursive evidence that Reagan and Gorbachev were calculating risks in relation to the other’s intent and integrity, then this would be evidence that would falsify the theory of suspension. It might be argued that the disagreements over SDI that continued between Reagan and Gorbachev throughout the three years of their tenure as leaders demonstrate that each was still seeking to achieve security at the expense of the other. Reagan and Gorbachev may have placed less emphasis on the competitive aspects of the US–Soviet relationship than their predecessors, but for all their personal warmth toward each other, they were leaders of two superpowers that viewed each other with suspicion and continued to prepare for the possibility of nuclear conflict.
The problem with this criticism of the suspension claim is that it conflates hedging at the interstate level, where we recognize enemy images of both sides persisted in key elements of the national security bureaucracy, and the interpersonal relationship where such hedging and risk calculation was absent. For example, Caspar Weinberger, Reagan’s Secretary of Defense at the time of the Reykjavik summit, having not been a party to the process of bonding that had taken place between Reagan and Gorbachev, viewed the risks of trusting Gorbachev very differently from Reagan. In the aftermath of Reykjavik, Weinberger argued that the US Government should break off negotiations with the Soviet Union until it had deployed SDI (see Wilson, 2014: 200). Had Reagan accepted this counsel from Weinberger, he would have betrayed Gorbachev’s trust.
It is true that Reagan and Gorbachev failed to reach agreement on a historic nuclear arms reduction agreement at Reykjavik because they could not bridge their positions over SDI. However, it is erroneous to argue this stemmed from a lack of trust between the two leaders. Instead, the obstacle to an agreement was that neither could not be sure that their successors would honor this trust. Gorbachev said to Reagan, “If I go back to Moscow and say that despite our agreement on deep reductions of nuclear weapons, despite our agreement on the 10-year period [of non-withdrawal from the ABM Treaty], we have given the United States the right to test SDI in space so that the US is ready to deploy it by the end of that period, they will call me a fool and irresponsible leader” (cited in Savranskaya and Blanton, 2016: 232). Reagan also had the problem of the future in mind. He trusted in Gorbachev’s peaceful intent and integrity, but he wanted SDI as an insurance policy in case Gorbachev’s successors harbored aggressive intent.
Our key contention is that as a result of their face-to-face interactions at Geneva and Reykjavik, there developed a solidarity between Reagan and Gorbachev that enabled them to suspend risk calculations in relation to each other. The most important diplomatic outcome of this solidarity was the 1987 Intermediate Nuclear Forces (INF) Treaty that was negotiated through 1987, culminating in the signing of the Treaty by the two leaders at their summit in December of that year in Washington, DC.
Evaluating mediated interaction: The strength of weak bonds
We have argued that while face-to-face interactions may be required for strong social bonds to develop, leaders, through the text may form weak social bonds, even in the absence of bodily co-presence and an overt barrier to outsiders. To assess this argument, we examined social bond formation during the Cold War, in two of, arguably, the unlikeliest places it should be found: the CMC and nuclear arms reduction negotiations. We find evidence for weak social bonds in letters and strong social bonds in face-to-face interaction. With the former, the relationship was characterized by a lower level of solidarity that coexisted with continued suspicion and hedging behavior. With the latter, the relationship was characterized by a high degree of solidarity and the absence of risk calculation, or trust as suspension. In the end, we argue that while bonds produced through the mediated interaction of letter exchange might be weak, there is strength in weak bonds to the extent it allows leaders to reduce distrust and even, as in the case of Kennedy and Khrushchev, develop enough trust to de-escalate and resolve a nuclear crisis. And, importantly, this solidarity lasted beyond the CMC, making possible agreements on a Hotline to improve channels of trusted communication in any future crisis and the signing of the first nuclear arms control agreement between the superpowers, the 1963 Partial Test Ban Treaty. This does set up the fascinating counterfactual as to whether Kennedy and Khrushchev might have developed even stronger solidarity had they met face-to-face in the aftermath of the CMC. Our wager would be yes.
There are several empirical difficulties in studying social bond formation. First, ideally researchers will be able to independently observe and measure the interaction ritual ingredients, causal mechanism, and outcome variables of interest, while also holding confounds constant. However, there are several reasons why this is difficult to achieve. First, while qualitative process tracing techniques are useful in piecing together what happened in a given case, it remains difficult to pinpoint bonding processes. One reason for this is that observational data does not provide direct access to the existence or non-existence of bonds. Leaders rarely refer directly to “positive energy” or “hitting it off,” or even “solidarity.” Researchers must instead use empirical indicators of these concepts or proxies, which may be prone to error. Second, given the temporal nature of interactions in international politics, which may take place over a condensed time period, as in the CMC, or several years, as in the end of the Cold War, it is impossible to account for, or control, all confounds, such as a changing material environment. Nevertheless, despite these limitations, process tracing allows for demonstrating the plausibility of bond formation and to understand its effects. Similarly, we have posited a number of different mechanisms by which letters might serve as a proxy for copresence, including empathy, imagination, solidarity, and the way words are used to represent the self. While it is difficult to definitively test these qualitatively in historical case studies, laboratory experiments (that gain in internal validity what they give up in external validity) represent a promising area for future research where the ability to isolate and test for specific mechanisms may yield significant dividends. Finally, in some respects our cases may be classified as empirically straightforward. Both US and USSR archives are believed to be relatively complete, with declassification procedures that have largely kept the historical record intact. This is relatively rare. Researchers investigating bond formation where the archival record is erratic or incomplete will face significant difficulty in establishing the presence or non-presence of the four ritual ingredients. In these cases, interviews, oral histories, participant observation, and ethnography may be the only realistic methods available for assessing bonding, which poses significant challenges (see, for example, Stoll, 2020 who uses participant observation and biography interviews to delineate social milieus in Nairobi, Kenya).
There are also theoretical difficulties in interrogating bond formation in diplomacy. While the Collins model privileges that which occurs within the interaction, rather than reducing outcomes to the dispositions of the actors or the environment in which the interaction takes place, such interactions can never be isolated from previous interactions nor are the dispositions of the actors unimportant. Collins himself notes their crucial importance: “Social order is produced on the micro level: that is to say, all over the map, in transient situations and local groups, which may well be stratified by class, race, gender or otherwise divided against each other . . . the history of their chains—what sociologists have conventionally referred to as their positions in the social structure—is carried along in emotions and emotion-laden cognitions that become the ingredients for the upcoming encounter” (2014: 105). Khrushchev and Kennedy’s backgrounds, gender, race, and prior interactions, not only with one another but with significant others in their personal histories and environments, led them to both where they are, and who they are, during the CMC. The next theoretical move is to interrogate how a narrowly defined interactional order, as we have done here, can be broadened to account more fully for social structure. Finally, other theoretical approaches, such as recognition theory (see Duncombe, 2016; Duncombe, 2019 for example), complement the approach we have taken here and may be useful in explaining how we can make the move from dyadic interactions at the leader level to broader Self/Other identity understandings and narratives, such as that between the US and USSR, at the state level.
Another issue this study has raised is the extent to which bonds both replicate and aggregate to foreign policy decisions. We argue that Kennedy and Khrushchev bonded, yet it is also true several individuals were part of the overall decision-making process on both sides. Even if these internal actors were not integral to the bond formation between the two protagonists, we know empirically that these same outsiders gave advice, pushed particular policy proposals, and this included on both sides, calls for military action. As Lebow (1990: 491–492) argues, at the end of the day Kennedy and Khrushchev were able to not only “become as much allies as adversaries,” but they did so, in part, by “conspir[ing] with each other against their respective internal opponents.” We see a similar dynamic with Reagan and Gorbachev, who eschewed hawks in their own administrations in favor of privileging the personal bond they developed dyadically. The interaction between bonded leaders and their non-bonded advisors may be a fruitful area for future research. What types of arguments do bonded leaders make to outsiders who question the bond? What is the role of their leadership in changing the views of individuals and sub-cultures within a state bureaucracy? Lastly, our focus here has been on the micro-interactions of leaders, but microsociology also investigates the ways in which such interactions scale to create broader social structures. This is a critical question that we hope to take up in future research, as it is particularly relevant for explaining how cultures of enmity may change over time, in this case the US-USSR, but it is also crucial for understanding how structures are created that then give (potential) rise to new micro-interactions.
Our focus on letters as the form of mediated interaction captures a time when cable-wire and letter-exchange were ubiquitous means of interaction among leaders. This continues today, of course, but in the modern period leaders also engage through a variety of different media: telephone, video conference, or even Tweet. During the COVID-19 pandemic, leaders and diplomats became all too familiar with the intricacies of these various technologies, often lamenting the inability for them to replace bodily copresent face-to-face diplomacy. Theoretically, from a Collins perspective, new technology should increase synchronicity and make collective effervescence easier to achieve. Put another way, the fact that we find the existence of weak bonds in letters, which are heavily temporally mediated, bodes well for the notion that bonds may form through more immediate forms of technology. On the other hand, as social information processing theory suggests, faster is not always better: careful crafting of self-presentation may take thought and care, two qualities not always present in tweets, for example. Future research may investigate how telephone, texts, or video conferencing, to name a few technologies, affect the imagination process. It may be, for instance, that what tweets or text messages gain in speed they lose in length and nuance, affecting the ability for parties to imaginatively place themselves inside the interaction. In the end, the means by which leaders reduce distrust and mistrust about one another, through a process of social bonding, represents a potentially fruitful, and interdisciplinary, research agenda.
Footnotes
Acknowledgements
We are extremely grateful to Ana Alecsandru, Richard Herrmann, Jeffrey Kaplow, Benoît Pelopidas, Len Scott, Catriona Standfield, participants at a pre-ISA workshop (Toronto, 2019), ISA panel (virtual, 2021), and the anonymous reviewers and editors of EJIR for detailed comments and constructive criticism on earlier drafts. The order of author names reflects the principle of rotation.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
