Return to Op-Ed page

Cass R. Sunstein, "At Unease" in The New Republic, at 41 (September 6, 1999)

Book Review of Don't: A Reader's Guide to the Military's Anti-Gay Policy by Janet E. Halley (Duke University Press, 159 pp., $14.95)

In 1993, Jamie Gorelick, general counsel of the Department of Defense, testified before a subcommittee of the House Armed Services Committee. Defending President Clinton's new "Don't ask, Don't tell" policy, Gorelick was asked to explain how someone alleged to be homosexual could establish that she was not, in fact, homosexual.

Gorelick answered with an example: "In one instance a woman was alleged to have said that she was a lesbian. She came in and said essentially that the statement had been misunderstood. And actually, she introduced evidence-she brought in boyfriends who testified essentially that she was heterosexual-and the conclusion was reached that she would be retained." Gorelick offered this illustration to show how accused members could prove their heterosexuality and defend themselves against losing their job.

At that moment in the House hearings, you might suppose that the ban on homosexuals in the military would be in serious trouble. Is it really possible to imagine a trial in which a member of the American military has to defend herself against a charge of lesbianism by asking former boyfriends to testify on her behalf? What would the boyfriends be expected to say? Would they be subject to cross-examination? If the issue is whether the defendant is really homosexual, what particular questions might be asked, by way of resolving that contested issue? Surely-you might suppose-Gorelick's commendably candid response would raise doubts about the legitimacy of the policy that she was attempting to defend. Yet there was no visible reaction to Gorelick's anecdote. It was as if the proceeding that she described was perfectly acceptable, a matter of common sense, entirely routine.

Since 1982, the Department of Defense had maintained a policy requiring separation of any servicemember deemed to be "homosexual," defined as "a person, regardless of sex, who engages in, desires to engage in, or intends to engage in homosexual acts." (Branches of the military had set out similar bans for decades before.)

In 1991, in his campaign for the presidency, Bill Clinton made this prohibition an issue, and in late 1992, President-elect Clinton said that he intended to "lift the ban" on homosexuals in the military. In the President's very first week in office, however, the Joint Chiefs of Staff (including the immensely popular Colin Powell) threatened to resign in protest over any such effort, and George Mitchell, the Senate Majority Leader, told Clinton that there were seventy votes against the President's proposed decision. This was the beginning of the "don't ask, don't tell" policy (sometimes called "don't ask, don't tell, don't pursue")-a compromise reached among the White House, the Joint Chiefs, and the Senate opposition.

Two documents eventually emerged: a new Defense Department policy and a new statute, the latter putting a ban on homosexual conduct into federal legislation for the first time. The governing regulations, continuing on the books, begin by proclaiming that "homosexuality is incompatible with military service. The presence in the military environment of persons who engage in homosexual conduct or who, by their statements, demonstrate a propensity to engage in homosexual conduct, seriously impairs the accomplishment of the military mission." A homosexual is defined exactly as before. Separation from the service will come from either homosexual conduct or statements indicative of homosexuality. A member of the service may be discharged if he has stated that he is homosexual or bisexual, "unless there is a further finding that the member is not a homosexual or bisexual." Homosexual acts will result in discharge unless there are further findings--unless it is shown, among other things, that the conduct is unlikely to recur (the "queen for a day" exception) and that (despite the acts in question) the soldier does not desire or intend to engage in homosexual acts. At the same time, the new policy announces that commanders would not ask members of the service about their sexual orientation; there would be no "asking" and no "pursuit." The Department of Defense also recognizes that "individuals with a homosexual orientation have served with distinction in the armed services of the United States." Sexual orientation is not a bar to service without conduct; but a statement of homosexuality counts as conduct.

Later in 1993, Congress enacted a statute, growing out of the new Pentagon policy but significantly less protective of gay and lesbian servicemembers. The statute--it is the first actual law on the subject-is rooted in a congressional finding that the "presence in the armed forces of persons who demonstrate a propensity or intent to engage in homosexual acts would create an unacceptable risk to the high standards of morale, good order and discipline, and unit cohesion that are the essence of military capability." Under the statute, members must be separated from the armed forces under two circumstances. The first is "conduct," and here the statute is essentially the same as the Defense Department policy, but with an important proviso: a homosexual act is not merely same-sex erotic bodily contact, but also "any bodily contact which a reasonable person would understand to demonstrate a propensity or intent to engage in a" same-sex sexual act. The second is when a member "tells," and here the statute is more complicated.

The statute says that if a service member makes a statement that "he or she is a homosexual or bisexual," discharge must follow, unless there is a " further finding" that he or she "is not a person who engages in, attempted to engage in, has a propensity to engage in, or intends to engage in homosexual acts." Under the Defense Department directive implementing this statute, moreover, a "statement by a member that demonstrates a propensity or intent to engage in homosexual acts is grounds for separation not because it reflects the member's sexual orientation but because the statement indicates a likelihood that the member engages in or will engage in homosexual acts." Sexual orientation itself "is considered a personal and private matter," and not a bar to continued service.

What accounts for the shift from the old policy to the new one, and then to the new statute? What is the meaning of this shift? Janet Halley attempts to answer these questions not on the basis of interviews or conventional investigative techniques, but with close readings of the new policy and law, and with an exhaustive examination of the public record (hearings, press conferences, speeches, congressional debates). Halley thinks that this system is "much, much worse than its predecessor," and that a principal culprit is the Department of Justice, which, in her view, is the likely source of "the most alarming innovation in the new policy--a new set of rules that allows homosexual conduct to be inferred from supposed homosexual status."

Halley's target--what she means to challenge as false, indeed fraudulent-- is the widespread view that the new policy is more lenient than the old one, and that there has been a large-scale shift from discharging servicemembers for their "status" toward discharging them for their "conduct." Much of her discussion is dedicated to showing that the new rule does not bar status- related discharges--that status plays a large if sometimes subtle role under the new policy, in large part by requiring accused members to defend their heterosexual status. The result of all this is a kind of culture of surveillance, in which members must avoid giving anyone a hint of "conduct that manifests a propensity."

Halley acknowledges that unlike the new policy, the old policy was explicitly based on status, authorizing "military officials to determine not only what servicemembers did but what they desired and intended, all with the aim of determining who they were." President Clinton attempted to obtain a departure from this policy; having been blocked in his effort to lift the ban, he sought to shift regulatory attention from status to conduct. Thus Secretary of Defense Les Aspin explained that the purpose of the revision was to focus attention on "not what they are, but what they do," and President Clinton insisted publicly that "the emphasis should be always on people's conduct, not their status." After Congress enacted the statute, many people, including Clinton and Aspin, claimed that it merely reiterated and codified the new Defense Department policy, embodying the shift from status to conduct.

Halley thinks that this is all quite misleading, even a form of deceit. To be sure, the president "had no choice" but to retreat, and homosexual status alone (whatever that may mean) is no longer a basis for separation from military service. "But the 1993 revisions, even if they do not target people on the basis of their status, are nevertheless deeply, pervasively, and creatively ascriptive." Perhaps the most important point is that the "Clinton administration did introduce several elements of more or less simple conduct regulation, but Congress ... eliminated every single one of them in favor of heavily status-inflected revisions."

For one thing, Congress refused to recognize, as did the Defense Department, that homosexuals had served, and are serving, with distinction in the military. (Many military officials, including Colin Powell, so testified before Congress.) For another, Congress rejected Clinton's proposal to focus exclusively on sexual misconduct, and it showed no interest in his (fairly radical) suggestion that the grounds of sexual misconduct should be applied without regard to sexual orientation. Thus Congress refused to follow the Defense Department's suggested policy requiring commanders to investigate allegations of sodomy "in an even-handed manner without regard to whether the conduct alleged is heterosexual or homosexual."

Congress also defined homosexual acts to include not only bodily contact " for the purpose of satisfying sexual desires," but also bodily contact which a reasonable person would understand to demonstrate a "propensity or intent to engage in" acts designed to satisfy sexual desires. (Might this include hugging or holding hands?) And finally Congress also refused to require " don't ask" in the statute itself. The consequence has been that the "don't ask" idea--the only remaining trace of protection for gay and lesbian service members--is the current executive branch policy, but it is not the law, and it may be changed by another administration.

Halley thinks that Clinton was able to seek only "the appearance, not the reality, of a 'don't ask' reform." Even under the Defense Department policy, a range of indirect questions can be--and are--asked about sexual orientation, without violating the ban on direct questions. In any case, both the policy and the statute allow an investigation to result from "any evidence of conduct that manifests a propensity to engage in homosexual conduct." As we shall see, commanding officers have a great deal of discretion to interpret this phrase.

One of Halley's most interesting discussions involves the post-1993 emphasis on this idea of "propensity," a genuine innovation, not to be found in any predecessor provision. With some interesting and suggestive (though not at all conclusive) sleuthing, Halley attributes the innovation to the Department of Justice. What is the underlying idea here? Halley identifies two possibilities, which she labels "the actuarial model" and the " psychometric model." Under the actuarial model, the idea of propensity points simply to a statistical likelihood: "detecting people with a propensity is merely a prudent way of identifying servicemembers who are likely to engage in harmful conduct and getting them out of the military before they can actually harm its essences." Under the psychometric model, by contrast, a " propensity" is not a statistical matter but "an inner, intrinsic, even natural or innate characteristic." This model "attributes a pathological personal trait to each individual homosexual." Halley believes that the " psychometric model of homosexual propensity is a hidden premise of the actuarial model," and that the result of the focus on "propensity" is to obliterate any simple distinction between status and conduct.

Attorney General Janet Reno, among others, has claimed that the law is concerned only with conduct because members have an "opportunity to rebut" a presumed propensity to act; but Halley thinks that this is "particularly disingenuous." The reason is that the presumption--based on statements or other evidence of "propensity"--imposes on accused members a "heavy handicap," thus achieving "substantive outcomes under the guise of a merely technical change in procedure." The system of presumption-and-rebuttal also produces a high degree of arbitrariness, as individual commanding officers have considerable discretion to decide what statements and what acts show a " propensity."

The definition of homosexual acts at work here includes bodily contact which (in the words of the 1993 law) "a reasonable person would understand to demonstrate a propensity or intent to engage in" erotic, same-sex bodily contacts. Halley thinks that this is "the most important innovation" of the 1993 reforms. Who is this reasonable person, and what inferences are diverse (and reasonable?) people permitted to draw? Halley contends that members must "imagine a lexicon of conduct-that-my-commander-thinksto-manifest-a- propensity." To counteract the threat of discharge, therefore, "acting viciously anti-gay is probably the best way" to provide insulation against suspicion.

All this seems to me a bit of hyperbole on Halley's part; but in practice the results have been a mess. Compare two cases, both of which involved actual trials. In one case, Air Force Captain Richard Richenberg testified that he had recently realized that he was homosexual, but that he had not engaged in, and he did not intend to engage in, prohibited conduct. Richenberg made a sharp distinction between homosexual "attraction" and " doing it," and said that he disapproved of casual sex on moral grounds, and added that at the time when he considered himself heterosexual, he had honored his belief that sexual intercourse should not occur until marriage. But he was discharged, with the trial court saying that it "is of no constitutional significance that the plaintiff may in fact have no propensity to engage in homosexual conduct." By way of contrast, consider the servicemember who said that he was a "homosexual," that he had "feelings toward men and women" but "sexual feelings" only for women, and that he had decided not to engage in same-sex conduct because he feared "catching a disease." In this case it was held that he had rebutted the presumption and could retain his job. The two cases exemplify a set of extremely messy results.

For Halley, the most general problem is that "if you have been found to have engaged in conduct that manifests a propensity to engage in same-sex erotic acts, you are going to be discharged unless you can prove that you have no such propensity," and to prove that, "you need to prove your status." The result is that "heterosexuality prevails" under the policy--at the expense, in practice, of homosexuals, who are at risk of discharge, and of heterosexuals, who are at risk of false accusation. This, then, is Halley's basic analysis of "don't ask, don't tell": that it is not a shift from status to conduct, and not a safe refuge for discreet homosexual service members, but a recipe for widely variable and improvised diagnoses of "propensities," and a system that ties "status to conduct and conduct to status in surprising, devious, ingenious, and perverse ways."

An obvious issue arises. Is the policy constitutional? In a brief discussion, Halley notes that the Constitution is taken to prevent government from punishing people for their "status" alone, and many of the constitutional challenges insist that discrimination has occurred "based on their status, not their conduct." But Halley thinks that this strategy "is an insult to the personal sexual dignity of most servicemember clients." She wants those who challenge the policy to accept, even to "revel in" the (hardly surprising) fact that "self-identified homosexuals do engage in more same-sex erotic conduct than those who identify otherwise." She would also like courts to strike down the policy under either the First Amendment or the Equal Protection Clause.

Courts might conclude, for example, that "the 'don't tell' rules regulate speech according to its content, without excuse when the content itself-- homosexual status--is expressly permitted." Or they could conclude "that it is irrational," and hence violative of the equal protection clause, "to protect against conduct-based harm to military effectiveness when homosexuals engage in the conduct but not when heterosexuals do, or when open homosexuals rather than closeted ones do it." More broadly, courts "could observe that unit cohesion, privacy in the barracks and the showers, and sexual tranquility are eroded, not promoted, by the policy's joint legitimation of servicemembers' homophobic sensibilities and of secretive homosexual presence in the military." Thus Halley simultaneously challenges a leading strategy of gay rights organizations (to contest the policy as "status" regulation) and urges courts to invalidate the policy on a variety of constitutional grounds.

This is a clear, careful, and illuminating account of the various shifts in the substance of the policy, and of who lost (mostly gay rights advocates and Clinton) and who gained (mostly supporters of the old policy). Halley also shows, quite convincingly, that while the new policy does not target status " alone," it is hopelessly inadequate to describe it as a regulation of conduct rather than status. Indeed, her discussion of the "propensity" issue is a tour de force. For anyone interested in the "don't ask, don't tell" policy, this is an excellent place to start. But there are problems.

Halley's discussion is very much focussed on public texts, and her analysis is above all a textual analysis, with comparatively little discussion of the real-world effects of the 1993 revisions and their predecessor. Owing to the limitations of her textual focus, I do not believe that Halley establishes her claim that the new policy is "much, much worse than its predecessor," or even worse at all. It is not easy to see how the new approach, or for that matter any approach, can be worse than one that requires discharge of anyone " who engages in, desires to engage in, or intends to engage in homosexual acts. " Perhaps the new policy is more misleading, more deceptive, more unpredictable, more confusing; but is it really worse than a policy that requires discharge of anyone who desires or intends to engage in homosexual acts?

To answer this question, we need to know more, and not only more about texts. What have been the trends for dismissal of homosexuals from the military? Are the numbers going down or up? To the extent that they are changing, to what extent is the reason for the change the policy itself? Halley offers extremely little statistical evidence about the empirical questions, and this is a large omission. In fact, the data support her basic position. From 1982 to 1994, there was a decline every year in the number of people discharged for homosexuality: 1,998 in 1982, 1,660 in 1985, 996 in 1989, 730 in 1992, 682 in 1993, and 617 in 1994. Yet the numbers have gone up steadily since that time: 757 in 1995, 858 in 1996, 997 in 1997, and 1,145 in 1998-the largest number since 1987.

But all this still does not tell the whole story. To evaluate these discouraging numbers, it would be valuable to know the number of homosexuals who have entered the military at various points--and also to know the circumstances of the discharges in various periods. Were the discharges in 1982-1990 systematically different from those in the last five years? Has there been a greater trend toward discharging people based on unrebutted " propensities"? Have relevant military officials become more intensely opposed to homosexuals in the last five years? Halley does not discuss these matters.

Nor does she deal much with the series of abuses, recently acknowledged by the Pentagon, that have followed on the heels of the policy. These abuses have included various forms of intimidation and harassment--some of them cases in which service members complaining of harassment have found themselves, rather than their harassers, under investigation. In a widely publicized event, a soldier rumored to be gay was recently killed with a baseball bat at Fort Campbell, Kentucky. It is far from clear whether these events are a product of the new policy; what is clear is that the new policy has not prevented them. The Defense Department has recently announced guidelines designed to prevent the most egregious abuses, but it remains to be seen whether these guidelines will do much good.

Perhaps the greatest oddity of Halley's instructive book is that, while her main goal is to show that the new policy is not based on "conduct," she also believes that the policy would be little better if it were in fact based only on conduct. She thinks, plausibly, that a policy based only on conduct would be an insult to the dignity of homosexual servicemembers. And this point is connected to another surprising feature of the book: Halley's thin discussion of the underlying questions of policy and law. We get little sense of why the military was so strongly resistant to Clinton's proposal--of what interests underlie the military's reluctance to lift the ban entirely, and of whether those interests are legitimate. Halley shows that those who sought some kind of ban believe that "unit cohesion" is at risk from open homosexuality, and that recruitment and morale would be endangered if soldiers were aware that they were serving with homosexuals, partly because people must live in such close quarters. It is clear that Halley finds these arguments weak, but she does not analyze them in any detail.

The relevant anxieties take several forms. The first is that sexual relationships and sexual impositions-above all, involving superiors--are intolerable in the military setting. A prime concern is sexual harassment. To be sure, there is an irony in seeing the law of sexual harassment invoked by people not normally so enthusiastic about that law; some defenders of the policy seem to fear that homosexual men will frequently treat heterosexual men the same way that heterosexual men sometimes treat women, and they are eager to protect heterosexual men against that kind of treatment. Undoubtedly many people do believe that sexual harassment is intolerable, and that all steps should be taken to ensure that it does not occur in the military setting. If sexual impositions create the difficulty, however, then the natural solution is to ban sexual impositions of any kind, not to forbid gays and lesbians from serving their country.

Perhaps the problem is not sexual harassment, but sexual relations. Perhaps the military setting is incompatible with even voluntary sexual relations among members of the service, at least when they must work together in close quarters. But if this is so--and it certainly seems reasonable to think that it is so--then the solution is to ban the relevant relations. Indeed, current policy imposes restrictions on "fraternization," and any such restrictions could be applied to homosexual relationships.

Perhaps the strongest argument against a true reform of this discriminatory policy--and what really seems to underlie the idea of "unit cohesion"--is that even without sexual harassment and sexual relations, it would be very hard for many young men and women to know that they are serving with open gays and lesbians. Of
course there is, in the military, a degree of segregation between young men and young women, partly to prevent the kinds of entanglements, tensions, and misunderstandings that might genuinely undermine military preparation and effectiveness. The ban on open homosexuality is designed to prevent the same sorts of difficulties. And if the lifting of that ban really would hurt morale and recruitment, and cause serious internal problems, it would not be so easy for outsiders to dismiss the military's underlying concern. When Colin Powell and other members of the Joint Chiefs of Staff resisted President Clinton's initiative in 1993, this was their fundamental objection. And there is a genuine interest in sexual privacy as well, raised most vividly, perhaps, in the context of communal barracks and showers.

Yet this argument is not without its own problems. If some young people find it hard to work in close quarters with homosexuals, this is often because of a form of ignorance and prejudice (and sometimes hatred) that is not ordinarily a legitimate basis for denying people employment opportunities. Giving into these feelings tends to perpetuate and to ratify them. The argument also depends on highly speculative empirical claims about the harmful effects of a nondiscriminatory policy. It is important to remember that an exceptionally wide range of countries do not exclude homosexuals from the armed services. There are no bans, no analogues to "don't ask, don't tell, " and apparently no problems for "unit cohesion" in Denmark, Norway, Finland, Belgium, Spain, the Netherlands, France, Canada, Germany, Switzerland, Sweden, and Israel. Nor must we look abroad for similar evidence. Gays and lesbians fought for their country in World War II, the Korean War, and the Vietnam War, and their military records show that they served the nation as well, on average, as heterosexuals.

To be sure, sexual privacy is a legitimate interest, and perhaps a nondiscriminatory policy would require special efforts to protect privacy. But if military commanders were genuinely committed to make nondiscrimination work, it is hard to imagine that such a policy would create serious problems. Does this mean, then, that courts should strike down "don't ask, don't tell," as Halley urges, under the First Amendment or the Equal Protection Clause? There are hard questions here. The military imposes a wide range of legitimate restrictions on the speech of servicemembers, and in the military context it does not violate the First Amendment to use statements as evidence of conduct, if the conduct is independently unacceptable. If the military forbids the use of cocaine, it is perfectly entitled to take a member's statement, "I like to use cocaine," as evidence of cocaine use; indeed, there would be no First Amendment problem with taking the statement, "I am a cocaine user," or "I would like to use cocaine," as evidence of an intention or a propensity to use cocaine, at least if the member is permitted a rebuttal. The "don't ask, don't tell" policy reaches somewhat further, but when combined with the general power of the military to regulate speech by members of the service, this point is probably sufficient to defeat a First Amendment challenge.

Despite the policy's regulation of "telling," the real question is not one of free speech. It is whether homosexuals have been denied the equal protection of the laws. As Halley acknowledges, and as recent courts have emphasized, members of the service are not entitled to the same rights as civilians. When the military is involved, courts should be extremely cautious about rejecting the considered judgments of the elected branches of government. But it is one thing to adopt a general posture of deference; it is quite another for courts to permit the government to exclude, to stigmatize, and to humiliate a class of citizens whose members have long served the nation with distinction.

In the aftermath of a large number of documented abuses, the Defense Department is now embarking on an effort to ensure against harassment, violence, and witch-hunts--to provide anti-harassment training and to require investigations to be launched only on the basis of solid evidence. But even if the reforms make things better, there is something farcical and bizarre, and more than slightly Orwellian, about this earnest and highly publicized effort to ensure that people will be investigated, as acknowledged or practicing homosexuals, only if there is actual evidence that they are, in fact, acknowledged or practicing homosexuals. The most fundamental problem with the policy is not the occasional abuses, egregious as some have been, but its very core.

The best solution would be for Congress, which has an independent obligation to the Constitution, to reject the "don't ask, don't tell" policy on equal protection grounds, and to work with the military to accommodate its concerns. But this is one of the exceedingly rare cases, I think, in which courts would be justified in invalidating a military practice under the Constitution.

In 1992, in Sex and Reason, Richard Posner observed that "it is terrible to tell people they are unfit to serve their country, unless they really are unfit, which is not the case here." Since he wrote those words, more than 5, 056 members of the American military have been discharged under "don't ask, don't tell," a number of them after being placed under oath and asked, before a trier of fact, whether they have a "propensity" to be interested in men, women, or both. Numerous other members of the military, now serving their country and prepared to risk their lives on its behalf, face a similar fate. The "don't ask, don't tell" policy is an atrocity. Thesooner we are rid of it, the better.


Return to Op-Ed page