On Academic Productivity

Over the next few months, anxious professors in the United States will compile tenure and promotion dossiers. Anxious graduate students, post-docs and professors will compile application materials for academic positions and fellowships. And many of these people—perhaps most—will be wondering: “Have I been productive enough”?

Academic “productivity” is an odd subject. Academic articles and books are not “products” in the sense that shirts or computers or pies are products. Producing more of them does not necessarily bring any benefit to anyone. Nor are they interchangeable. Most have only a moderate impact on their fields (or less), while a few change those fields forever. Yet evaluation committees still tally up the numbers, and judge whether the author is sufficiently “productive.”

Standards of productivity vary enormously, both between fields and within them. Economists are often seen as highly productive if they publish one paper a year. Some experimental scientists, working on large-scale projects with hundreds of collaborators, have their name on scores of papers a year. In my own field of history, most professors only publish a single book in their entire career. But according to the British Library catalogue, the military and diplomatic historian Jeremy Black has written or edited no fewer than 231.

Young scholars often get conflicting advice about productivity. Some advisors tell them not to publish anything until they have completed their research and can fully back up their conclusions. Others tell them to publish as much as possible. In the field of history, grad students are sometimes warned that if they publish too many articles out of their dissertations, they will have a hard time finding a publisher for their all-important tenure books. Others hear that a book without articles is insufficient for tenure.

For what it is worth, here are some thoughts based on my own experience both “producing” scholarship and evaluating the scholarship of others.

Most importantly: productivity, while not unimportant, is overrated. The quality and impact of publications matter far more than their raw number. Natalie Zemon Davis, one of the great historians of the past century, received her Ph.D. in 1959 and only published her first book, a collection of eight articles, sixteen years later. Any one of those articles did more for the field of history than 99% of the book-length monographs published in the same period, but by the formal standards of most US research universities, Davis might well have qualified as “unproductive,” and could conceivably have failed to get tenure as a result. She went on, by the way, to publish eight further books, and many more field-changing articles.

On the other hand, some great scholars have been fearsomely productive from the start. The British Library catalogue attributes 120 books to the late historian of medicine Roy Porter—not quite up to Jeremy Black’s Stakhanovite standards but still impressive. Not all Porter’s books were equally significant, but many were genuinely brilliant and amazing. All in all, I see little correlation between quality and quantity of publication.

The fact is that writing is a complex psychological process, and different people handle it in different ways. Some agonize over every sentence and need to get every clause exactly right before releasing a work into the wild. Others find that writing comes easily to them, and out of some combination of pleasure and compulsion end up producing scads of it. The novelist David Lodge deliciously captured what writing can represent to some especially productive academics in his portrait of the unpleasant Oxford don Rudyard Parkinson, in Small World: “Writing, his own writing, is… like sex: an assertion of will, an exercise of power, a release of tension. If he doesn’t write something at least once a day he becomes irritable and depressed—and it has to be for publication, for to Rudyard Parkinson, unpublished writing is like masturbation or coitus interruptus, something shameful and unsatisfying.”

I hope I don’t resemble Rudyard Parkinson, but (as this newsletter testifies) I do feel something of what the French call la démangeaison d’écrire—the itch, or compulsion to write. I enjoy writing, and work out my ideas best in the process of doing so. Also, I can’t imagine finishing a book without the feedback received in response to earlier published pieces of it. But not everyone operates in this manner, and why should they?

Much academic writing has a relatively short shelf life. But as long as it contributes usefully to ongoing discussions, it serves its purpose. Academic articles, at least in my field, don’t have to be definitive statements of a case. They can be literal “essays,” trial balloons of sort. For this reason, incidentally, I don’t agree with the advice that graduate students need to wait until they’ve completed all their research to start publishing. Aside from the fact that doing so is a patently bad strategy in the current academic job market, what better way to make connections and receive both attention and serious feedback? Obviously, everyone should have sufficient evidence to support their arguments. And anyone who falls into the “frequent publisher” category should make their writings as engaging as possible, so as not to pollute the cloud with unreadable sludge. But yes, if you have something to say: publish.

Incidentally, while I am on the subject of advice to graduate students, I can’t recall seeing a single case where a promising scholar failed to get a book published because too much of it had already appeared in article form. By contrast, I can think of many important books whose authors had previously published virtually every single chapter in article form. If a scholar’s arguments are at all important, readers will want to see the final version: the book. But if you know of counter-examples, please send them to me.

Needless to say, universities are anything but equal when it comes to helping scholars remain productive. It is one thing to publish frequently when you teach three or four small courses a year with ample research support and regular sabbaticals. It’s quite another to do so when you teach three or four large courses per semester—plus summer school, to make ends meet—with little or no research support.

But if productivity is overrated, we still can’t dismiss it entirely. For anyone teaching at a university level, it matters enormously to have had the experience of generating formal knowledge, at every stage from formulating a research question to responding to reviewers and completing the publication process. In the field of history, this means at the very least publishing several articles. And professors whose universities expect them to be active scholars need to remain part of ongoing discussions and debates. There are, of course, many ways to do this in addition to publishing books: digital projects, editing journals and book series, involvement with professional organizations, participating in conferences. But publishing is certainly one obvious and measurable way of remaining engaged.

Another reason we can’t ignore productivity entirely is that hiring and promotion decisions are always gambles of a sort—highly informed gambles, we hope, but gambles nonetheless, on how careers will evolve. If a candidate comes up for tenure in history at a research university with very sparse publication records—one or two articles, and an incomplete book manuscript—he or she may still, conceivably, go on to publish a significant book and become an important figure in the field. But most candidates who come up with this sort of record go on to publish very little. Very few historians who wait sixteen years to publish their first books are Natalie Davis, or anything close to her. I used to work at Johns Hopkins, back when that university only conferred tenure on full professors. I asked a senior colleague the reason for the rule and still remember his response verbatim: “If people are neurotic enough to publish two books under these conditions, they will keep publishing.” Brutal, but there was a certain truth to it.

In addition, as in any job, it matters for people to meet expectations. If a research university makes clear to assistant professors in the humanities that they need to publish a book to get tenure, and they don’t, it raises the question of whether they will meet expectations in other areas of their work, including teaching and service. Perhaps. There may be a good reason for the delay. But warning signs matter, especially when candidates have had long postdocs before even starting assistant professorships and come up for tenure a decade or so after finishing their PhD’s.

In short, productivity counts, but it shouldn’t count too much. When evaluating candidates, whether for promotion and tenure, for positions, or for fellowships, what has been published matters more than how much has been published. And while there are of course ways to measure the “impact” of publications—citations, the “impact factor” of the journal—nothing beats actually reading what the candidate has written.