Last spring, a surprising letter arrived in the mail of an acquaintance of mine. He is an able young Ph.D. from one of the nation's top universities, who at the time held a temporary position and was seeking his first tenure-track job. To improve his chances of employment, he organized a session for the upcoming convention and submitted his proposal to the AHA Program Committee. The committee chair, Professor Linda Hall, responded as follows:

Thank you for your very interesting panel proposal for the 1991 American Historical Association meetings in Chicago. However your proposal fails to meet all the AHA requirements we need in order to consider it for presentation. Specifically, your panel is not gender balanced. ... If you can find a female participant we will be pleased to consider your proposal.

Most historians who read this letter react predictably. First, upon getting the gist of it, they look up in amazement and say, "Surely this isn't AHA policy!" Then, realizing that they are not quite sure what their Association's policy is, they back down a bit, saying, "Whatever the AHA's policy is, surely it was not meant to be carried out in this way." The conversation then typically focuses on the wording of the letter. "Too literal-minded," "too rigid," are typical comments. Most readers are eager to express their support for the principle of gender integration, but they believe it can be achieved less coercively. "Giving preference to gender-balanced panels is desirable, but completely rejecting single-sex panels, refusing even to consider them—that's going too far," is the conclusion most commonly reached.

As it happens, Professor Linda Hall, the Program Committee chair, fully agrees that the letter went too far. She informs me that the letter erred in making gender balance a prerequisite for consideration and she accepts full responsibility for the misunderstanding that resulted in that error. Only a handful of such letters went out, unnoticed in the avalanche of mail that crosses the desk of any Program Committee chair. The letter did not accurately reflect the policy of the 1991 Program Committee, which in practice never treated gender balance as an absolute prerequisite for consideration or acceptance—though in other ways the committee did strive to enforce Association policy on gender balance "strictly," in accordance with the language of the relevant policy guidelines and the advice of AHA staffers. Once she discovered the error, Professor Hall sought to correct it through telephone communications to those concerned, including the recipient of the above letter. Although my acquaintance's proposal was turned down, both he and I accept Professor Hall's assurances that it received fair consideration and was rejected for reasons other than gender balance.

Professor Hall generously accepts personal responsibility for misunderstandings that might better be attributed to another source: the AHA's policy statement on gender balance, Program Committee guideline 6d. Although ambiguous, that guideline not only authorizes, but instructs, the Program Committee to engage in conduct only marginally less coercive than that threatened by the letter. The guideline reads as follows: "The Program Committee will actively seek to avoid gender-segregated sessions. It shall encourage proposers of individual sessions to ensure that whenever possible sessions include members of both sexes." Presumably as a result of this policy, the proportion of single-sex panels appearing in the annual program has fallen, by my count, from thirty-five in 1986, the year before the first version of 6d was adopted, to no more than six at the 1990 convention. The impact of the policy is felt principally by males: in each of these years, all but one single-sex panel was male. As we shall see, this is unsurprising, since only one of every five or six historians today is a woman, making it much more difficult to include a woman on every panel than a man.

The language of guideline 6d is open to a wide range of interpretations. One can, of course, interpret it noncoercively by focusing on the word "encourage." Encouraging—even "actively" encouraging—a proposer to conform to a test of gender balance is presumably not the same thing as requiring conformity as a condition for further consideration. But the statement also lends itself to a much more coercive interpretation, for it gives the Program Committee ample authority to exclude a panel if its proposer cannot show that balance was impossible to achieve. The task assigned to the Program Committee is to "actively ... encourage" proposers to "ensure ... whenever possible" that both sexes are included. These are strong words, and they place the burden of proof squarely on proposers. The tacit assumption is that when someone submits a panel not balanced by gender, the most likely explanation is bias. That this presumption of guilt is refutable does not render it unobjectionable.

The letter received by my acquaintance was not so much a departure from AHA policy as a guilelessly explicit expression of its most coercive features. After all, how is the committee to "actively encourage" gender balance if not by excluding, or threatening to exclude, those panels that fail to meet the test? How is the committee to satisfy itself that the "whenever possible" condition has been met, except by interrogating proposers of single-sex panels, requiring them to testify that they have tried and failed to find participants of the requisite gender? And how can this testimony be elicited unless the committee warns proposers that their panel is likely to be excluded if it is not forthcoming? The letter sent to my acquaintance was an "accident" in the sense that it interpreted 6d more inflexibly than Professor Hall and her committee intended, but even on the most flexible interpretation the guideline encourages the committee to treat proposers as guilty until proven innocent. This was an "accident" that was waiting to happen. What set the stage for it is the evasively coercive language of guideline 6d.

Upon seeing the letter, both a colleague and I called AHA headquarters to protest, taking care not to divulge the name of the recipient. After conferring with Professor Hall, AHA staffers assured us that our acquaintance's proposal would receive due consideration, and reported that the committee did not regard gender balance as an absolute prerequisite, all appearances to the contrary notwithstanding. The recipient of the letter received the same assurances directly from Professor Hall a few days later, when he called to inquire cautiously about the fate of his proposal. For obvious reasons he did not identify himself as the subject of our protest. Although Professor Hall assured him that his proposal would be considered, she did not expressly repudiate the language of the letter he had received, and so her assurances seemed pro forma. What does it mean, we wondered, for anyone to promise over the telephone that a single-sex panel will receive "due consideration," when the committee has already specified in writing that gender balance is a prerequisite to any consideration at all, and when its standing instructions direct it to "ensure ... whenever possible" that panels be gender-balanced?

Coercive measures do, I believe, have their appropriate uses in rectifying deep-rooted discrimination. If there is any justification at all for restricting single-sex panels, I suppose it lies in the belief sometimes voiced in feminist circles that males typically are unable or unwilling to recognize merit in the work of their female colleagues. On this interpretation, guideline 6d evidently fights fire with fire, prejudice with prejudgment, sexism with sweeping assumptions about the ethical and intellectual incapacities of the male majority of the Association. But even people firmly convinced, as I am, of the reality and harmfulness of sexism among males might think this suspicion of male judgment and motivation excessive in the case at hand, given the fact that last year seven of the nine elective positions on the governing Council of the AHA (leaving aside the ex officio positions reserved for the president, president-elect, and past president) were occupied by women. How threatening can male chauvinism be in an organization that is more than 80 percent male, and yet elects women to 78 percent of the regular positions on its governing board?

Still more to the point, how threatening can male chauvinism be in an organization in which women are already overrepresented on panels? No doubt this will surprise most historians, as it did me. Not women, but men, are underrepresented on AHA convention panels. At the 1990 convention, for example, women held 34 percent of all panel positions, twice what one would expect, given that they currently constitute little more than 17 percent of academically employed history Ph.D.'s. Nor is the overrepresentation of women a recent development. Women were also overrepresented in 1986, the year before the earliest version of guideline 6d was adopted by the AHA Council. At that convention 28 percent of panelists were women, although women at that time made up only about 16 percent of academically employed Ph.D.'s. Even a decade ago, back in 1980, when women accounted for 15 percent of the profession, they held 23 percent of panel positions at the annual convention, half again the rate of participation one would expect on the basis of their numbers.*

Ironically, women already have much readier access to panel participation than would be produced by even the most radical affirmative action measures now being debated on the national scene. If, for example, the AHA were to adopt the much-hated "quota" system, gearing panel participation directly to the proportions of men and women in the profession, the number of women panelists at future annual meetings would have to be reduced—to half its current level. For many years it has been obligatory for candidates for AHA office to demand that our annual meetings "reflect accurately the real diversity of the profession." I wonder how many have realized that this goal, if taken seriously, would require the Program Committee to reverse course and take steps to promote greater participation by males?

Professional associations are porous organizations, easily influenced by any dedicated interest group. It is neither surprising nor disturbing, in view of feminism's current momentum in academic circles, that women should temporarily hold disproportionately large shares of both panel positions and professional offices. But with accession to power the case for affirmative action evaporates. Men need not bend over backwards to be fair in areas in which women are already being treated more than fairly.

What justification is there, then, for the retention of guideline 6d? No doubt avoidance of gender segregation is a worthy goal, and the guidelines should continue to articulate that general sentiment. But what reason is there for thinking that adequate progress toward the goal cannot be achieved voluntarily? Why try to micromanage the composition of particular panels, diluting the intellectual criteria that ought principally to govern panel selection, when women's overall rate of panel participation is already higher than that of men?

Why place a special burden of proof on the proposers of all-male panels? Surely anyone can see that all-male panels need not signify anything sinister, but only the natural workings of statistical distribution in a population that is more than 80 percent male. Panels typically have about five members; fewer than one in five historians is a woman. Arithmetic alone makes it difficult to include a woman on every panel. Given the large role that women already have in our annual meetings, guideline 6d is not only needless, but counterproductive, for it indiscriminately penalizes even gender-blind decision making and forces male proposers to engage in "tokenism" on behalf of a group that is already more than adequately represented. No one wants to play the role of token. Why not trust each other's good judgment and leave both the Program Committee and individual proposers free to choose panelists, not because of their sex, but because of the cogency and intellectual relevance of the contribution they are likely to make to the session?

Strong as the case for change may seem, 6d is not without defenders. In April 1991 the AHA's Research Division, which oversees the Program Committee, met and considered a proposal to revise 6d. The proposal was prompted by the letter sent to my acquaintance, a copy of which was furnished to the Division. The revision proposed was the least confrontational solution imaginable. It consisted of a three-word parenthetical addition to the existing language of the guideline, authorizing the Program Committee to "encourage (but not require) proposers of individual sessions to ensure that whenever possible sessions include members of both sexes."

Mild though it was, this revision was vigorously rejected. Instead, at its next meeting, in October 1991, the Research Division reaffirmed the language of the guidelines and merely supplemented it by adding several ambiguous sentences to the Program Committee's informal "lore" book, which does not bind anyone. The first of the new sentences commendably warns committee members not to "misconstrue or misrepresent the AHA's guidelines." But this sound advice is then counterbalanced by an affirmation of the Committee's unqualified "authority to set goals or requirements that go beyond that [sic] stated in the AHA guidelines ... [as long as] all announcements and correspondence ... clearly state when a goal or requirement is a committee policy rather than an AHA policy." As an afterthought, the Division added to this ambidextrous statement a third and final sentence that points in the right direction, though with little effect: "For example, guideline 6d should not be interpreted as requiring all sessions to be gender-integrated."

The result of all this foggy language is that future Program Committees are specifically deprived of authority to treat gender balance as an AHA requirement, but left completely free to adopt it as a committee requirement—a distinction without a difference. And of course even when gender balance is no longer treated as a flat "requirement," it remains in place as a criterion for preferring some panels over others. The Program Committee is still under orders to discriminate against single-sex panels unless their proposers can show that it is not "possible" to include members of both sexes. How rigorously 6d is to be enforced is, in effect, left up to each committee, and is likely to vary unpredictably from one year to the next.

Instead of mandating ritual displays of mutual respect, let's try relying on the real thing, voluntarily expressed. In the absence of hard evidence of discrimination, coercive measures, no matter how well intended, do more harm than good. The time has passed for fine-tuning the hopelessly vague and misguided language of guideline 6d. In place of 6d (and 6e, which parallels it in matters of age, ethnicity, and race), I urge the Council to adopt a new Program Committee guideline reading as follows:

In the selection of panelists the aim should be to avoid parochialism in all its forms. No simple formula will insure this result, but panels made up of a single gender, race, religion, age cohort, political persuasion, or school of thought are, on average, less likely to be intellectually productive than panels that give voice to diverse perspectives. Panelists should, of course, be selected mainly for the intellectual cogency and relevance of their expected contribution to the topic under discussion. Other things being equal, the Program Committee will give priority to panels that display a spirit of cosmopolitanism, both in the kinds of questions asked and in the range of perspectives brought to bear on them.

* My figures for the proportion of women in the profession come from the same source routinely cited in these pages by the Committee on Women Historians: the biennial "profiles" published by the National Research Council, titled Humanities Doctorates in the United States. They give the proportions of women among academically employed history Ph.D.'s, based on an annual census in which NRC personnel have great confidence. AHA staffers tell me that the proportion of women on AHA membership rolls is higher, ranging somewhere between 25 and 30 percent, but they admit that this is only an estimate; surprisingly, as of this writing, no one has ever counted AHA members by gender. Even if this estimate were used instead of NRC's hard data, women would still have been overrepresented on 1990 AHA panels by as much as a third. Readers may be surprised that women constitute less than 20 percent of the profession, since the proportion of women among new Ph.D.'s, fresh out of graduate school, has been above 30 percent for nearly a decade; but the population to which these new recruits are being added is large (it includes people who got their degrees as many as forty years ago) and so it changes slowly. The proportions of panel participants are based on names of known gender that appear in the "Index of Participants" in each annual meeting Program. Space limitations preclude a fuller account of the sources employed, but anyone interested is invited to correspond with the author.

Thomas L. Haskell is a professor of history at Rice University.