homeprev issuesexecpublic affairsstaffasa home

ASA, Professors Submit Brief to U.S. Appeals Court on “Don’t Ask, Don’t Tell” Policy

by David Segal, University of Maryland-College Park

The American Sociological Association recently joined a group of social science professors, including me, in submitting an amicus curiae brief (Cook v. Rumsfeld) supporting former military personnel who have brought suit against the Departments of Defense and Homeland Security after having been discharged from military service because of their sexual orientation. Using social science research, the brief challenges the assertion, made by supporters of the ban on gays in the military, that allowing gays and lesbians to serve openly would undermine unit cohesion and adversely impact military performance. The ASA Council had unanimously endorsed ASA’s signing on to the brief.*

In 1993, when former President Bill Clinton attempted to fulfill his campaign promise to lift the ban on gays in the military, both houses of Congress held extensive hearings on the issue. The cohesion argument was advanced by numerous proponents of the ban, who rooted their position in three pieces of social science research conducted during World War II to determine why soldiers fight: (1) combat historian S.L.A. Marshall’s after-action combat interviews with soldiers; (2) surveys conducted by Samuel A. Stouffer as part of the American Soldier project; and, most important, (3) Edward A. Shils and Morris Janowitz’s study of cohesion and disintegration in the Wehrmacht. The cohesion argument had previously been advanced in the late 1940s to delay the racial integration of the military and again in the 1970s and 1980s to delay gender integration.

Research Foundation

The brief argues that the cohesion argument as stated lacks any scientifically validated empirical support, and that research on cohesion conducted in the United States, as well as research on military forces that do not discriminate on the basis of sexual orientation, in fact belie the assertion. Questions have been raised about whether Marshall actually conducted the interviews on which his conclusions are based. Stouffer and his colleagues did ask American soldiers if they wanted to serve in racially integrated units and found that they did not. They also asked soldiers what kept them going in battle, and one of the most frequently named motivations was “their buddies,” but this operationalization of cohesion was not expressed by a majority of soldiers, nor was it linked to the socio-demographic homogeneity of their units. The Shils and Janowitz data were not based on sociological surveys but on intelligence interrogations of German Prisoners of War (POW), who if they said they were motivated to go on fighting because of loyalty to their fellow soldiers were likely to be released earlier than if they said that they were committed to the Nazi cause. Research on POWs is no longer allowed by federal regulations protecting human subjects. None of these studies sought to empirically link sexual orientation to either cohesion or performance, although the Shils and Janowitz study did refer to homoerotic ties between German officers and their soldiers.

These studies have been interpreted as supporting the importance of social cohesion (read homogeneity) in units. However, in recent research, social cohesion has been linked as often to poor performance as to good performance. On the other hand, task cohesion— bonding on the basis of contribution to group efforts—has been linked to effective performance, but the causal link goes from performance to cohesion, not the reverse.

No research has actually been conducted on American military forces to determine the impact of sexual orientation on cohesion or performance. However, in comparative research on military forces that do not discriminate on the basis of sexual orientation, including forces that are reputed to be extremely effective, such as those of Great Britain and Israel, no negative effects have been noted. And retired general and Joint Chiefs chair John Shalikashvili, who had supported the U.S. policy, stated in a January 2, 2007, New York Times op-ed that a changed social context mandates a reconsideration of Don’t Ask, Don’t Tell.

* The brief, submitted on behalf of ASA and social scientists by the Washington, DC, law firm of Covington and Burling, is accessible at; click on “Press” in the upper righthand corner. The case likely will be argued in early spring in the U.S. Court of Appeals for the First Circuit.