Text Patterns - by Alan Jacobs

Wednesday, December 23, 2009

taste-testing literary style

Nicholas Lezard looks at a new scientific “formula” that can, it is claimed, identify a given author’s stylistic “fingerprint”: the key, it appears, is “the frequency with which authors use new words.” Lezard is not impressed:

For any reasonably well-read person should be able to tell whether a text is by Hardy, Melville or Lawrence almost at a glance even if they haven't read it before. . . . Do you remember when, years ago, some dismal piece of doggerel (which began, as I recall, with the lines "Shall I die?/Shall I fly?") was, on the basis of word-frequency, claimed to be a hitherto undiscovered work by Shakespeare? A few people were impolite enough to point out that it was far too shit for Shakespeare to have written, but on the whole news agencies and those with a tin ear for poetry went along with the assertion. It even made it into an edition of his collected works, but I think now has been quietly dropped.

Leaving aside the question of whether this new formula is reliable — I don't know enough about it to have an opinion, and I doubt that Lezard does either — I wonder about Lezard’s claim that a given writer’s style is so readily identifiable by “any reasonably well-read person.” ("Almost at a glance"?) That would depend, I guess, on several things: first, how distinctive that style is — Dickens is more highly mannered than George Eliot, Pynchon more so than Philip Roth — , but also how large a sample one would get to work from, and what the sample happened to be. Distinctive as Dickens is, I could find paragraphs from his novels that would be hard to distinguish from Trollopian passages of similar length. Sometimes even the most peculiar writer just has to get on with the business of telling a story. (Similar caveats would need to be made about the evaluation of poetic style.)
When someone makes the kind of sweeping claim that Lezard makes here, I find myself wanting to put him to the test. So tested, he might discover that styles are more elusive than he thinks. In this context I am reminded of wine tasting: experts pronounce with great confidence on the traits of various wines, but their judgments are highly inconsistent and their palates easily fooled. Robert Parker himself was recently reminded of the perils of blind tasting. There’s no reason to think that the “tasting” of literary styles is any more reliable.

2 comments:

  • Aren't there questions of this sort on the Literature GRE?

  • It's been the better part of a century since I took the GRE, but I think those questions either use selections with a unique (and therefore highly identifiable) style, or have content that the test-taker is supposed to recognize: a reference to Barsetshire, say, or the name of a character. Or both.

Post a Comment

[Basic HTML tags can be used in this comment field.]