Unfriendly Fire

by Dr. Nathaniel Frank

  The reliance of the pro-ban case on military opinion was suspect from the beginning, for reasons already discussed. As the GAO itself noted, “tabulations of self-initiated letters are not valid” when “stronger evidence is available in the form of more technically sound, public opinion poll evidence.”39 But again, the real question is what is the relevance of opinions to begin with? Charlie Moskos seemed to believe that opinion polls constituted hard evidence. In his congressional testimony, he announced that the social scientific community had “empirical data” showing that sex integration had “degrade[d] military effectiveness” during the Gulf War. What were his data? “According to a Roper poll taken in Desert Shield,” he testified, “45 percent of those who served in mixed-gender units in the Gulf said that there was enough sexual activity to degrade military performance. That is a very high number.”

  “A Roper poll? An opinion poll?” asked Judith Stiehm incredulously, when presented with Moskos’s data in an interview. Opinion polls, she said, “are only opinions.” To have empirical evidence on cohesion, “you have to measure cohesion.” But Stiehm said that when military officials “constructed studies to systematically measure this, they kept getting studies which found no correlation, and so they stopped doing the studies.” It’s a criticism echoed by other scholars, many of whom believe that Moskos abused his academic credentials. “It’s an incredibly insidious role that Moskos has played in the policy process,” complained Aaron Belkin, a political scientist who founded the Palm Center at the University of California, Santa Barbara, which studies gays in the military. “He used his academic credentials to pretend that the policy is based on academic evidence when in fact it’s based on homophobia.” Belkin said that “when you look at the evidence they use and start to scrutinize it, you realize it’s not evidence.”40

  But of course, offering military opinions as empirical evidence is par for the course for this sociology expert. In a 1993 essay, “From Citizens’ Army to Social Laboratory,” Moskos rejected the comparison of racial integration to allowing gays in the military. But rather than make an argument for how the analogy failed, he opted to simply cite Colin Powell’s opinion on the matter. In place of models or measurements to account for his concern over “morale and group cohesion,” Moskos offered up a pithy quote from a highly decorated veteran who writes for Newsweek: “One doesn’t need to be a field marshall to understand that sex between service members undermines those critical factors that produce discipline, military orders, spirit, and combat effectiveness.” Never mind that no one was talking about “sex between service members,” only about the military service of homosexuals—a group of people many Americans still can’t seem to discuss without thinking of sex. Moskos continued to defend the ban years after it was passed using the same nonevidence. In 1997, he suggested that critics of his argument about the privacy rights of straights check into the policies of major universities regarding “what really happens in freshman dorm assignments when open gays and straights are assigned the same room and someone objects.” Moskos claimed that, at Northwestern, where he taught at the time of his remark, straight students were permitted to change room assignments for this reason alone. But a spokesperson for Northwestern denies that the university ever had such a policy. Either way, should the policies of private American universities dictate—or even help explain—the laws governing U.S. military personnel? Then, when asked in a 1999 radio debate if he could provide actual evidence showing that gay service undermined military effectiveness, he said, “If you want data, we have survey data on this question and there is . . . a vehement opposition by the majority of the men. If that isn’t data, I don’t know what is.”41

  But as we have discussed, that is not evidence, at least not the kind that should be convincing to honest, rational people. As it turns out, however, an enormous amount of research has been conducted on the connection between cohesion and performance, research that offers overwhelming evidence about whether the gay ban hurts or helps the military mission. What does this research say? Does it bear out the vivid images that ban defenders painted of a nation’s military undermined by gays serving openly and cohesive fighting forces torn asunder? What exactly is the connection between homosexuality, unit cohesion, and combat performance?

  Unit cohesion is, like love, not an easy concept to pin down. Loosely associated with what sociologists call “primary group solidarity,” it refers to a kind of mystical association that ties soldiers together through a shared identification with a leader. The most famous research on military bonding comes from the era after World War II, when American thinkers turned to Freudian insight hoping to understand Nazi soldiers’ compliance with Hitler. In 1948, Edward Shils and Morris Janowitz launched the field of military sociology when they published their pioneering study, “Cohesion and Disintegration in the Wehrmacht in World War II.” The article argued that camaraderie and leadership, not ideology or cause, motivated the German soldier. Enlisted men form emotional ties with their peers through joint idealization of a leader, the authors explained. The relationship to the leader that the troops share develops into comradeship, inspiring empathy, self-sacrifice, and ultimately a willingness to die for the group.42

  A generation later, U.S. involvement in Vietnam spawned new inquiries into the social and psychological sources of combat motivation, which largely reiterated the central role of group loyalty as the only force strong enough to compel individuals to risk death. By the time Congress held its hearings in the spring of 1993, a consensus seemed to be emerging around the idea that cohesion was the “essence” of a fighting force, the glue that binds soldiers together in “coordinated action.”43

  Such conclusions were absorbed and echoed by the military brass. The Joint Chiefs’ MWG report had written in its findings that “military operations are team operations—units win wars, not individuals. The rights and needs of the group are emphasized while individual rights and needs are often set aside or sacrificed for military necessity.” General Powell asserted that, “to win wars, we create cohesive teams of warriors who will bond so tightly that they’re prepared to go into battle and give their lives.” Together, the reports and testimony of the national security community helped to enshrine an interdisciplinary orthodoxy in military and civilian circles alike which assumed a causal relationship between unit cohesion and effective militaries: Strong cohesion made a fighting force effective, while weak cohesion made it ineffective.44

  Yet the actual impact of unit cohesion on combat effectiveness is not nearly so simple. The Shils and Janowitz scholarship was heavily critiqued in the years since its publication. Of particular note was a broad reassessment in the 1980s that suggested that their work gave “disproportionate attention” to the issue of primary group solidarity. For instance, one 1989 study argued that, contrary to Shils and Janowitz, primary group cohesion cannot explain the combat motivation in Hitler’s army because close-knit groups never existed: German soldiers continued to fight despite staggering personnel losses, which meant human turnover was far too high for them to have ever maintained close cohesive ties. The 1980s research on group behavior, which came mostly from military sociology and psychology, made a crucial distinction between two different kinds of unit cohesion. “Social cohesion” refers to bonds of friendship and affinity among group members. When individuals enjoy spending time together, when they feel emotional ties of loyalty and companionship, they form a socially cohesive group. “Task cohesion,” by contrast, refers to the group solidarity that results from the collective efforts of individuals dedicated to achieving a common goal. Members of a legal firm, a theater group, or an assembly line may have no desire to socialize together before or after their work is done, but their mutual devotion to winning a case, producing a play, or building a car can generate superb task cohesion for the purpose of completing their respective missions.45

  Moskos said in a 2000 interview that he has “never understood the significance of the distinction between task and
social cohesion. It is immaterial to my policy prescription of don’t ask don’t tell.” He later said, “To me it’s all psychobabble. ‘We’re going to take this hill even though we all hate each other’? I can’t conceive of achieving a task when you all hate each other.”46

  But as usual, his folk wisdom does not stand up to actual research. By almost all accounts, any positive correlation between unit cohesion and military performance relates to task, not social, cohesion. In institutional settings, the relevant question is not whether group members like each other, but whether they are mutually committed to the task at hand. As Judith Stiehm pointed out in a 1992 article, “trust and confidence develop not from homogeneity, but shared experience.” When members of different backgrounds come together in the armed forces, “the military assumes the job of training them to behave as a team. It has many powerful tools to develop desired responses.” In 1994, Brian Mullen and Carolyn Copper, of Syracuse University, conducted the most complete meta-analysis to date on the cohesion-performance relationship. Their research indicated that, after controlling for task cohesion, social cohesion had no connection to performance.47

  In 1996, Robert J. MacCoun, a Berkeley psychologist and contributor to the 1993 Rand study on gay service, published his results of an extensive review of fifty years of research, covering nearly two hundred publications. MacCoun concluded that “it is task cohesion, not social cohesion or group pride, that drives group performance. This conclusion is consistent with the results of hundreds of studies in the industrial-organizational psychology literature.”48

  Finally, military psychologists have made a distinction between comradeship and friendship, categories which roughly parallel task cohesion and social cohesion, respectively. While comradeship develops by subordinating individual needs to those of the group, friendship is all about individuality and close affective ties, which pose a danger in combat environments no matter if you’re gay or straight. As a result, leaders have always maintained a rule of “don’t get too close.” The question is, under what conditions is comradeship likely to become friendship? For soldiers in combat, the intensity and duration of battle make emotional intimacy a constant threat, “a burden of love,” as an American bomber pilot poignantly expressed it, “that couldn’t flourish.” By this reasoning, the goal of military training is to build group cohesiveness at the expense of individuality, “through a regime that makes prior identities irrelevant.” This is the reason for basic training. In other words, the very essence of military training is designed to manage the very problems that critics of gay service claim is a gay problem alone.49

  Not only does social cohesion fail to augment performance; it can frequently impair it. According to many studies, including some conducted or cited by Charles Moskos himself, social bonding and combat performance are inversely correlated: The more soldiers bond with each other, the more combat performance suffers. Fraternizing, desertion, fraggings (the killing or maiming of a unit leader), sexual harassment—these collective acts of insubordination are not products of lax social ties but of ties that bind too much. It is not, then, the presence of gay soldiers that threatens military effectiveness but the invitation they represent to get too close. Some advocates of the gay ban, it seems, fear that service members in units with open gays will get along too poorly, while others feel they’ll get along just a little bit too well.50

  This reassessment of the unit cohesion theory, and the more subtle appreciation of the difference between social and task cohesion, was also absorbed by the U.S. military. In fact, despite the military’s endless invocation of unit cohesion, the Pentagon has a long history of trying to minimize social cohesion and encourage individualism in its personnel. In 1985, a Rand report prepared for the Pentagon warned against “too much affective cohesion,” because it “might interfere with the critical appraisal of performance that is needed to maintain quality output, as members become concerned with supporting each other and raising group morale instead of concentrating on the task at hand.” Between the 1950s and the 1980s, the army experimented with a buddy system in which units were trained together and then sent into combat. Evaluations by the Walter Reed Army Institute of Research of the unit manning system, called COHORT, concluded that “military cohesion has not been valued as a combat multiplier in the U.S. Army.” They found that cohesion is “a byproduct, not a core goal leaders need be trained to create and maintain,” and that “there is as yet no commitment in the Army to building and maintaining group cohesion.”51

  This may be why Moskos’s real reasons for opposing an end to the gay ban had nothing to do with unit cohesion—whether task or social. While his public voice continued to emphasize the centrality of unit cohesion to combat performance, his private focus was quite different. “Fuck unit cohesion,” he said in a 2000 interview. “I don’t care about that.” In Moskos’s view, the rationale for the double standard was about a discomfort with thinking about sexuality, which, for him, boiled down to the rights of straight soldiers not to be watched with eyes of desire. “I’ve offered all kinds of arguments against the policy,” he said, “but the privacy one is where it breaks down.” Indeed, Moskos felt so strongly about the privacy issue that he viewed mandatory gay-straight cohabitation as tantamount to Nazism: “I would not want to fight for a country in which privacy issues are so trampled upon,” he said. “Those are the conditions of concentration camps.”52

  One has to wonder why Moskos took the charge he was given as a social scientist and used it to give cover to the assertion by generals and politicians that unit cohesion—not sexual discomfort—was the central basis of the gay ban, if that is not what he believed. But the concentration camp analogy is not wholly irrelevant, as the ghosts of Nazi Germany have maintained a strange and persistent presence in the dialogue on homosexuality and the military. A little-discussed section of Shils and Janowitz’s famous study on Wehrmacht soldiers sheds unwelcome, but inescapable, light on the need to maintain a regime of sexual repression in the armed forces. Revealing the “dark side of cohesion,” this research highlights not only the homosocial bonds of enlisted men but the homoerotic tendencies of military culture.53

  According to the scholarship, it is the unacknowledged erotic bonds of military men that actually underlay the primary group cohesion achieved in the German army during World War II. In 1948, the noted psychiatrist William Menninger characterized the wartime soldier’s bond as one of “disguised and sublimated homosexuality.” Some scholars suggested that the popular war song “My Buddy” (“I miss your voice and the touch of your hand, my buddy”) helped neutralize fears of unexpected longings, making the military a safe place for same-sex intimacy. As long as there were no homosexuals present, soldiers could have intense same-sex relationships and not worry about being gay.54

  At the 1993 Senate hearings, the sociologist David Segal, who studied under both Shils and Janowitz, described the World War II research to the frequently befuddled assemblage of silver-haired senators. Cohesion in the Wehrmacht, he explained, was based in part on a “latent homosexual subculture.” There was, of course, no sanctioning of actual same-sex sexual activity in the Nazi army. But while the Wehrmacht tolerated no avowed gay soldiers, homoerotic attachments seem to have been quite common. As Segal put it, “there was a hard core of enlisted personnel in the Wehrmacht who were attracted to the company of other men. They did not necessarily behave homosexually; indeed, they probably did not. But they preferred the company of men.”55

  Segal reported to Congress that he had never seen this particular piece of Shils and Janowitz research cited before. “I discovered it by accident,” he said, “while I was grading some midterm exams.” But its implications were clear. “It basically suggests that what we have more recently called ‘male bonding’ may well have been in the Wehrmacht this propensity to seek other males as erotic objects, although not acting on that.” These conclusions, he explained, “throw into question” the assertion that homosexual tendencies undermine
unit cohesion. In fact, rather than undermining unit cohesion, the presence of men with quiet same-sex longing appears to have enhanced it by bringing together groups of people with the propensity for intense bonding. The key to successfully cohesive fighting units, it turns out, is that homosexual—or at least homoerotic—affection should be present but repressed: that is, unspoken and, if possible, unacknowledged.56

  Moskos’s testimony before Nunn expanded on this theme: “Precisely because there are homoerotic tendencies in all male groups,” he explained, referencing the “sexual insecurities” of straight men, “this is exactly why [we need] the ban. Once these homoerotic tendencies are out, the cat is out of the bag, then you have all kinds of negative effects on unit cohesion.” But for Moskos, “the point is that in the Nazi army, you could not be a gay.” He struggled to reconcile the presence of latent homosexual desire with what he regarded as “probably the most barbaric system toward gays” in human history. “You have these erotic tendencies operating at one level,” he concluded, “but at the same time, the system is the most repressive ever known,” an arrangement which has historically “worked for a good fighting army.”57

  In other words, the problem with acknowledging the presence of gays in the military is that it could burden with added meaning the low-level homoerotic behavior that is normally operating among all service members. It would force even men who are effectively straight to come face-to-face with buried strands of their own same-sex desire, feelings that do not make them homosexual but whose very presence is nevertheless a threat to their fragile heterosexual identity.