John Wilkins does it again, with his post, The classification of clouds, which interests me because you can’t classify what you can’t describe. And, as you may have surmised, description is one of my current hobbyhorses. The only literary studies (and similar disciplines) are going to ride out of the current hermeneutic swamp is on the back of description.
So what does Wilkins say? For openers:
Clouds were regarded as so subjective, fleeting and resistant to classification that they were a byword for the failure of empirical classification, until Luke Howard in 1802 proposed the foundation for our present system of cloud classification (in competition, although he did not know it, with others in Europe, and on the heels of Hooke and later meterological language proposals including one by Lamarck the same year.
Now, as a practical matter, neither classification nor description loom large in the literary studies tool kit. There are recognized genres—poems, lyrical and dramatic; dramas, comic, tragic, and otherwise; novels, historical, autobiographical, the Bildungsroman, and so forth; and there are pesky things like the pastoral. But classification schemes are not much discussed. Nor, for that matter, do I see it as an immediately pressing matter. But description is, and, perhaps after a generation or two of competent descriptive work, the problem of literary classification will appear more tractable, or disappear in an unexpected way.
Back to Wilkins. He continues:
Howard’s proposal, like Lamarck’s, was driven solely by empirical observations. No experiment was possible with clouds (although there were some schemes for building cloud producing machines early on), and there was no real theory as such, just a desire to, as Lamarck said, note that “clouds have certain general forms which are not at all dependent upon chance but on a state of affairs which it would be useful to recognise and determine”... In short, this is an example of a classification scheme without much if anything in the way of Theory.
Nor can critics much experiment with texts, though writers, of course, do so all the time. As for theory, whatever capital-tee “Theory” is in literary studies, it is NOT a theory OF literature; it is more like a grab bag of abstract ideation critics use in concocting readings or interpretations.
Wilkins then lays out that standard classification—cirrus, cumulus, and so on—and notes that Howard’s “criteria used apparent density, elevation, height, and whether it produced rain.” The criteria are, of course, the dimensions, if you will, of usable description. They aren’t the only possible terms; one could always describe clounds in terms of the images they evoke: elephants, sunflowers, the Christ, and so forth. But that’s useless. And can easily go on forever and ever and get nowhere.
Howard’s criteria were not useless. With them in hand "now meteorologists could communicate and seek explanations." I believe that a coherent and sensible method of textual description will have similar benefits for literary study.
What are the useful dimensions of literary description? We don’t know, though we’ve got clues all over the place. What we’ve got to do is take those clues (you know, things story vs. plot, binary oppositions, line length, and so on through a very long list) and run them up against the newer psychologies (cognitive, neuro, evo) and see what turns out in the realm of practical description of actual texts. We need those first, lots of them. Then we can begin the task of constructing generalizations over them.