Monday, June 26, 2017

Mathematics, Computing, and the Literary Mind

As some of you know, Willard McCarty has been hosting an informal online seminar on the digital humanities since 1987. One topic that comes and goes is the question of whether or not computing has more to offer the humanist than a set of practical tools. Could computing offer us a way of thinking about how the objects and processes of humanistic inquiry actually function? This topic has reappeared once again and I've decided to offer my two cents. Here's the note I posted to the seminar.

Moire Trio

Dear Willard et al.,

This business – mathematics & computing and what they offer humanists other than tools – is something I've been thinking about, off and on, since the late 1960s. Back then I wasn't interested in practical tools (for making a concordance, or stylometrics, or whatever), I was interested in thinking about how the mind worked. Of course lots of thinkers have pursued that line over the years, and while it's produced its share of nonsense, I don't think that should discredit the whole line of investigation, which is hardly unified and is still very much open-ended.

As far as I know the nature of computation is itself still very much under investigation. And I figure that literary studies (my particular corner of the humanities) may well have contributions to make. That is, understanding the computational properties of the literary mind is NOT (going to be) a matter of taking some existing ensemble of computational processes and fitting them to one text after another. Rather, we – someone – is going to have to create appropriate computational procedures.

Just how we – someone – get there from here, that's way beyond the scope of an email note, nor would I be able to chart a course given whatever scope I please. But I think we have to start with literary form and we must learn how to describe it.

I've got some general notes on this in a working paper, Description 3: The Primacy of Visualization:

Here's a somewhat more polished account (though unpublished): Sharing Experience: Computation, Form, and Meaning in the Work of Literature:

Some years ago I engaged in extensive correspondence with Mary Douglas, the anthropologist, and she got me interested in ring-composition. Texts with the form:

A, B, C...X, C', B', A

Why ring-composition? 1) Because it "smells" like something that requires a computational account. 2) It's something definite one can look for in a text. 3) Identifying and describing ring-composition in texts doesn't require any esoteric knowledge. But it does require the sort of feel for the phenomenon that comes only from paying close attention to texts.

Douglas has published short book on the subject (her last), based on a series of lectures she delivered at Yale: Thinking in Circles: An Essay on Ring Composition (Yale 2007). There's a chapter where she lists a set of identifying features of ring-composition.

I've produced a handful of working papers in which I describe ring-composition in a variety of texts. You can find those listed here:

If you're interested in reading around in that material, you might start with, Ring Composition: Some Notes on a Particular Literary Morphology:

One of the things I do in that working paper is gloss Douglas's diagnostic features as being aspects of a computational process.

Finally, it's worth remembering that ordinary arithmetic (which is fairly important in the theory of computation) is, after all, a linguistic process. The symbol set is highly restricted, as is the set of rules for its use (both sets are finite); but it is a creature of language.


Bill Benzon

No comments:

Post a Comment