In the comments to Eric’s post about underrated historical novels, I pointed out that there is a problem with talking about the “historical novel” as a self-evident genre. I did not, however, go into much detail as to why, because I covered the topic on my qualifying exams and the less said about that experience the better. But since Eric asked so nicely, I will oblige and show you why this discussion’s so painfully tangled.

Short version: Its knots all sport thorns.

Long pedantic version:

The most basic definition of the historical novel is situational: if the moment the novel represents occurs prior to the moment in which it is written, the novel is historical. The conceptual problem with this definition is its expansiveness: any novel not written about the perpetual present or the future will be historical so long as it is sufficiently realist. (More on the necessity of limiting this discussion to realist novels shortly.) However, as the identifying characteristics of a given period become visible through age—think smile-lines or wrinkled foreheads—even those works set in the perpetual present become dated. Bright Lights, Big City was a novel of the 1980s when it was published in 1984; now, it reads like an historical novel about the 1980s, for the simple reason that any novel that sufficiently captures a zeitgeist also documents it for it future generations.  Such novels become a sort of historical artifact whose usefulness depends on the acuity and fidelity of their documentation: a perceptive author deeply indebted to the realist tradition will record an immense quantity of high-quality period detail. This being the case, these novelists of the perpetual present belong to a more anthropological than historical tradition.  An obvious distinction, but one which must be made lest we claim that all realist novels are historical by dint of being old.

So then, the historical novel is not historical because it belongs to history, but because its author makes a concerted effort to address historical material in a realist mode. It may seem odd to confine historical novels within the realist movement, but if we define the genre attitudinally—if the sole criterion for inclusion is a dogged historical disposition—our reckoning must include modernist novels that, though obsessed with history, nevertheless fail to conform to the fuzzy notion of the historical novel we’re refining here. Not that there aren’t some modernist works, like John Dos Passos’s U.S.A. trilogy, that warrant inclusion, but they do so on the basis of being “realist enough,” and in so doing confirm our belief that the historical novel is a subset of realism—or better yet, a mode of the realist genre that entails a particular distance from and attitude towards historical events. This means that historical romances like Rob Roy—written in 1817 about the England and Scotland in 1715—are excluded, but early realist novels like Huck Finn—written in 1884 about events in 1840 or thereabout—squeeze in.

The problem, of course, is that most people would include Rob Roy and exclude Huck Finn—the former on the basis of its reliance of historical events and personages, the latter because Mark Twain made it all up and novels consisting of unadulterated fictions are not rightly called historical. Of course, not all of Huck Finn is an invention: the time and its quarrels, the people and their prejudices, and the river and its course are all faithfully recreated in the novel, but somehow that is not enough to convince our gut that Huck Finn deserves to be called an historical novel. However, the desire to include Rob Roy speaks to another defining characteristic of historical novels: the presence of historical events and personages. It also demonstrates that their presence alone is insufficient—the narrative in which they appear must conform to a realist ethos or we risk including rousing yarns of how George Washington single-handedly defeated the Hessians with egg yolk and a tuning folk.

Historical novels, then, must depict historical events and personages while adhering to a realist ethos: its narrative need not be historically faithful, but at a minimum it must be historically plausible. Plausibility matters so much that even novels about far-flung futures can make a case for being historical novels. This is why, to my mind, categorizing historical novels according to their modal characteristics is more productive than doing so by their content. It may seem odd to say that the defining characteristic of historical novels is something other than their historical content, but categorizing them according to the manner in which that content is addressed allows us to include novels relating to present (Dreiser’s An American Tragedy) or future (Robinson’s Red Mars) histories and exclude novels whose manifest historical content is overwhelmed by romantic conventions (Scott’s Rob Roy) or implausible heroics (Weir Mitchell’s Hugh Wynne, Free Quaker: Sometime Brevet Lt. Col. of His Excellency General Washington).

In sum, defining the historical novel as a realist mode allows us to include the works we want to and exclude the ones we don’t and by “we,” of course, I mean “I.” You’re more than welcome to disagree, which is what my committee, none too impressed that I’d argued Rob Roy into the ether, did during my qualifying exams. Their main objections concerned my reliance on a pornographic notion of plausibility and an insufficiently articulated idea that historical novels address social evolution. If you’ve read this far, you know that I’ve again neglected to articulate that idea in the body of my argument, only this time I did so for a reason: it allows novels I’ve booted from the porch sneak back in through the barn. If I designate “the depiction of social evolution” to be a defining characteristic of the historical novel, Rob Roy and Huck Finn are back in; but if I claim that the depiction is incidental because it is impossible to accurately track historical developments forward in time without also capturing the response of a society to those developments, Rob Roy and Huck Finn are still out . . . which is where I believe they belong.

(x-posted here and here.)