Near the end of his important, recent history of the American historical profession, That Noble Dream: The "Objectivity Questions" and the American Historical Profession, Peter Novick offers a sardonic view of sessions at the American Historical Association annual meetings in the 1980s: "Typically there would be three tenuously connected papers, edited with a cleaver so as to be (almost) deliverable in the allotted time. The shredded remnants were read aloud as rapidly as the speakers' lips and tongues could move, while pretending not to notice the chairperson pointing at the clock. There followed one or two "prepared" comments cobbled together at the last moment because the paper had only just arrived. Then, if time allowed, there would be a couple of usually rambling and off-the-point remarks from the floor. The conclusion was often a plea to the audience (friends and family of the speakers, those on the search committees sampling the merchandise, and a collection of incurable innocents in search of enlightenment) to exit the room as rapidly as possible because the hotel staff had to arrange it for a luncheon now overdue."

Novick concedes only that his description might be a "slight" exaggeration and that exceptions to it are "rare." Others may disagree. Certainly, many of us have attended AHA sessions in which we have heard beautifully delivered versions of genuinely important papers or listened to impassioned debates over issues of great scholarly or political weight. Yet most of us would also have to admit that we have attended more than a few sessions that pretty much match Novick's description. Moreover, the problem of sessions in which papers are hastily read to small or inattentive audiences, in which paper givers and commentators routinely exceed their time limits, and in which there is little or no substantive discussion, is not confined to the annual meetings of the AHA. Such panels can be found at most major historical gatherings, from the OAH (Organization of American Historians) to the OHA (Oral History Association).

Novick is not the first historian to criticize the format of these scholarly rituals. Reviewing the papers presented at the 1925 AHA Annual Meeting, American Historical Review Editor J. Franklin Jameson complained that "much of the program was devoted to subjects of limited range. In consequence of this, there could be little real discussion of the papers, few auditors having the temerity to dissent publicly from the views of one who has apparently made himself the master of a small, or perhaps obscure, portion of history." "But," he concluded with an air of resignation, "the absence of lively discussion from our annual meetings is an old story and has been dwelt upon perhaps to satiety, by one who is now presenting his twenty-fourth of these annual chronicles."

The quarter century that Jameson had chronicled in his reports on the annual meetings had seen the AHA's annual meetings congeal into a format that (with only one significant alteration) has dominated our public academic discourse to this day. In the earliest years of the AHA's history, it seems, scholarly papers were not always read in their entirety. Of the numerous contributions presented to the 1895 meeting, for example, "some were read by title only" and "others were greatly abridged in reading." The formal sessions often focused on professional and curricular issues rather than the discussion of academic papers. But by 1900, the reading of papers had become standard as had the practice of running more than one session in a single time slot. The rushed presentation of papers was also familiar to our early twentieth-century predecessors, who packed four papers into most sessions and allotted the paper givers, what Jameson calls, "the usual twenty minutes" for their performance.

The single "innovation" in the entire twentieth-century history of the history conference came in the late 1920s, perhaps in response to Jameson's complaints about the absence of "lively discussion." Near the end of that decade, it became increasingly common for someone to be designated on the program as "discussion" leader—a practice that had been tried sporadically in previous years. By 1930, every session on the program specified the papers and then "Discussion led by" a particular individual. Based on the descriptions of the sessions, these discussion leaders apparently took the role of today's "commentators," although that term did not come into use until 1951.

Thus, the AHA annual meetings had not only reached their "modern" format but that format had also been found unsatisfactory before most of the currently active historians in the United States were born. Why has this arrangement outlived generations of complaints? Part of the reason is simply its longevity; it has become one of our entrenched and unquestioned rituals. But convention sessions also serve some real functions. They are generally an effective showcase for new work; they offer a way to find out about scholarship going on in places and fields outside the immediate experience or acquaintance of an individual scholar. Less effectively (given the format), they furnish an occasion for people working on common topics and problems to meet, debate, and exchange sources, approaches, and ideas. Unfortunately, convention sessions are probably most successful in fulfilling objectives that have little to do with scholarship: they give people a line on their vitas (important for finding jobs, winning tenure, or getting raises) and they provide a rationale for faculty members to get their universities to pay their conference expenses.

Is discussion possible at an academic meeting? Those conferences with lively intellectual interchange usually do not include the formal readings of papers. Generally, papers are distributed and read in advance. Sometimes, such meetings even build in opportunities for relatively small groups (ten to twenty people, for example) to sit down and argue about a paper or an idea. Such approaches are common in both specialized conferences and in the meetings of some other professional organizations. Such formats do not themselves guarantee a stimulating conference, of course. You still need good papers and thoughtful participants. But it helps.

The notion that the prior circulation of papers makes for a better conference—like the complaints about the reading of papers—is not a new (or particularly startling) one. In his report on the 1928 AHA Annual Meeting in Indianapolis, Jameson noted the AHA had decided to repeat at two sessions an "innovation" tried at the first meeting in that city eighteen years earlier, "when for all the papers presented at the session for ancient history outlines were distributed in advance." "The general opinion," he reported of the experience in 1928, "seemed to be that such discussions were especially worthwhile."

Many people will immediately object that the prior circulation of papers is not feasible. To be sure, the logistics of doing this for annual meetings that attract about six times the number of participants that they did sixty years ago are formidable. Yet it should quickly be added that the technologies for producing, reproducing, and circulating manuscripts (e.g. photocopying, wordprocessing, faxing, computer networking, and express mailing) have probably also improved more than six fold. When the AHA repeated its experiment with prior circulation in 1929, it had to use the expedient of sending around proof sheets of a forthcoming article by Dixon Ryan Fox. (Again, "the result was a lively session with some good-tempered intellectual skirmishing.")

Some solutions to this problem are as simple today as they were sixty years ago. In 1928, one of the "discussion" sessions focused on an article by Ulrich B. Phillips that had been published in the October issue of the American Historical Review. Some might argue that the point of a conference is to present fresh, unpolished, in-progress work. But the discussion of recently published work is at least as important as talk about unpublished manuscripts. Moreover, the distinction between the two is not always so hard and fast. Many articles are themselves part of larger, in-progress projects. And some conference "papers" turn out to be work that is on the verge of appearing in print. Indeed, one of the problems of many conference presentations is precisely that most papers are really intended for publication and are written with a reader rather than a listener in mind.

In any case, the circulation of unpublished papers is not really quite as difficult as it might appear. The program for the AHA meeting is published almost three months before the annual meeting. Papers are supposed to be written and sent to commentators almost six weeks before the conference. That would give more than ample time for those interested in attending a session to write to the paper givers and request (and receive) a copy of the paper. But what about the expense? That is actually less of an issue that it might at first seem. Let's say that a typical session attracts fifty participants. Let's also say that a typical paper is fifteen pages (the length of a thirty-minute paper). Since most historians now work on wordprocessors, a single-spaced, eight-page version of the paper could be easily printed. Photocopied back-to-back, the paper would only take up four sheets. That means that copying costs could be less than a quarter and mailing costs would be a single twenty-five cent stamp. Sending out fifty papers would only cost $24.50. The clerical work required would be equally modest if those requesting papers provided self-addressed, stamped envelopes. Moreover, perhaps the time is not too distant when most historians will make use of free, interuniversity computer networks like BITNET, which will eliminate the copying and postage costs and most of the clerical work as well as greatly speed up the whole process.

Even if it were decided that the prior circulation of papers was not feasible, why not simply make the papers available on the first day of the conference? That would dispose of the postage charges, most of the clerical work, and the need for advance planning by participants. A modest charge for the papers could defray the costs and discourage people from taking papers for sessions they do not plan to attend. Each morning of the conference could provide a one and one-half hour "reading period," which would provide sufficient time to read the papers for all that day's sessions (as long as people adhered to the current limits for length). The time for the reading period could be made up by shortening the sessions; given the time saved by eliminating the formal reading of papers, there would still be substantially more time left for discussion.

None of this is meant to be a panacea. There would still be boring or pointless sessions. And changing the format of history conferences does not address other even more important issues about their content. Surely, there is also a need for more sessions that address broad historiographic, theoretical, methodological, comparative, professional, and curricular questions. Small group seminars on such questions organized around pre-assigned readings—the format that most of us use in graduate teaching—would, for example, be one of many welcome innovations. There is no dearth of new (or old) ideals for change. I would urge others to contribute their suggestions to this and similar publications. I would also propose that the program committee of the AHA take the small step of designating, say, one-third of the sessions at the next conference as "experimental" to see what does and doesn't work. It should not be difficult to improve modestly on what we have been doing so poorly for almost a century.

—Roy Rosenzweig is associate professor of history at George Mason University and serves on the AHA's Membership Committee.