Initial thoughts on historiographical revisionism
Unlike the Florida State Legislature, I understand that historical interpretations change over time. Historical interpretations of Marlborough and the War of the Spanish Succession have changed as well, but much much much much more slowly. How can we tell, you ask? Good question.
Historiography is largely invisible. It is dispersed across dozens if not hundreds of books, journal articles, book reviews, blogs… Reading up on a historiographical debate is something grad students do in seminars, and what practicing scholars do in the early phases of their research and as they try to keep up with their field. But the details of this preliminary research rarely make it onto the printed page. Every academic history book or journal article will start with a few sentences or pages explaining where their particular argument fits, answering the “So what?” question. If you submit (or review) a book proposal, you need to explicitly address which historiographies are at issue. Published book reviews should, ideally, place the work under review within a broader historiographical debate. Occasionally academic journals will even publish a ‘historiographical essay’ summarizing recent works on a particular subject – a scholar might even publish an entire book on some historiographical question. But these are rare – in fact, rumor has it that editors want you to delete most of that historiography when it comes time to publish your book. Whatever the reason, historians tend to save their ink for the details of their specific argument.
Historiography is challenging to keep up with not only because it is rarely discussed (and theoretically changes with each publication), but, as professional historians know, much if not most historical knowledge remains out of print. If there are a handful of books published on a topic, there will often be two or three-times as many journal articles/book chapters, and twice again as many conference presentations on the subject, spread out over years and presented at venues around the globe. Most of these conference presentations will never become a journal article, much less a book. Add to that all the information and argumentation that ends up on the cutting room floor during the editing process, and all that information that scholars retain in their noggins’ but fail to publish. You never know what other historians already know.
One result, at least in my experience, is that there is often wide disagreement over the shape of a particular historiographical debate. Take the “Military Revolution” (please, some say). Where does it stand now? Some say it’s dead, others say it should be dead, other say it hasn’t gone far enough, all while the most prominent proponent (Geoffrey Parker) is working on a new edition of his seminal book. The concept gets applied beyond its geographical heartland, converted into punctuated equilibrium, and even punctually equilibriated into a Revolution in Military Affairs (RMA). That’s a lot to keep track of, and it’s up to each historian to figure out where to draw the line when defining what works are relevant to the debate.
All this makes one wonder: how exactly do we measure a historiography? Is there a more ‘objective’ way to measure the state of the field? How do we determine what the majority of experts in the field agree upon? What questions they consider in need of more study?
Methodological questions abound when we try to summarize and reify the ideas of dozens of scholars. Do you emphasize the –graphy, i.e. the publication? If so, how does a dissertation that’s available online change a historiography – do you have to wait until it’s a book? It certainly seems to explain why many consider dissertations to have an expiration date – best to get it published before everyone adopts it via osmosis without bothering to cite you, or before the historiography bypasses your ideas altogether. Can an oral presentation alter the historiography by itself – Michael Roberts’ initial foray into military revolution in the form of a lecture in Belfast may have interested the audience members, but likely more important was its publication. Given how long it can take between a presentation and its final publication, historiography seems rather asynchronous. Historians always fret about how few accounts of the past ever get recorded for posterity – maybe historiography is the same way? Or, maybe historiography is only altered when a work (published or not) gets cited? [Ask universities and P&T committees about impact factors]. Does historiographical change require a certain critical mass? How does a lone voice – rejected or, worse yet, ignored – relate to the historiography? Are there dominant and subordinate historiographies on any given topic, and how do we determine which is which? By number of publications, by citation and sales figures, by status of the author? How many debates do we expect within a historiography, and is there a healthy ratio between the points under contention and the number of scholars working in the field? As with siege dates, do we measure the idea from its first public outing, its first publication, when it was first cited…? How does popular history relate to academic historiography? What do we make of a literature that is impervious to change? So many things to consider.
One new option is to take a digital humanities approach. Specifically, take advantage of the ability to have the full text of all those secondary sources available for searching and analysis. [Not that I would ever encourage anyone to copy a copyrighted work.] What scanning and textual analysis now allows is an ability to actually see the published historiography in its entirety – every word (taking OCR errors into account), every citation. We can discern patterns within authors’ sources, in their argumentation, in their organization (both chapter structure and the index), in the ideas and events they discuss and ignore, and in the specific language they use and refuse to use. We can of course perform such historiographical analysis without computers, we’ve done it for decades after all. But the results will necessarily be impressionistic rather than systematic, are likely to miss cases, and are impractical with a large number of works. This isn’t exactly the “distant reading” of literary DHers (à la Franco Moretti), but it does offer some interesting possibilities. Possibilities which I’ll explore (in my inimitably amateurish way) in a future post.
Ideas on how we can measure the change within a historiography?