7 June 2004


I was just reading a short comment rant over at the sometimes interesting dive into mark that touches on the prescriptivist/descriptivist debate. The more you understand about linguistics the tougher it is to be a language snob.

The last time this came up, a co-worker was making the trite observation that most people in this country don't speak correctly. He of course excluded himself, and I'm sure he had forgotten or excused his many uncaught verbal fumbles from previous conversations. Many people have their own grammatical buttons that can be pushed; mine is arrogance (which should be obvious from the average quality of my grammar).

(Descriptive linguistics is also (descriptively) called synchronic analysis--the language at one point in time. Synchronic features are one axis of a coordinate system that also includes diachronic features--language through history.)

David Foster Wallace wrote a characteristically lengthy article in the April 2001 issue of Harper's examining the battlefield of those two camps.

I submit, then, that it is indisputably easier to be dogmatic than Democratic, especially about issues that are both vexed and highly charged. I submit further that the issues surrounding "correctness" in contemporary American usage are both vexed and highly charged, and that the fundamental questions they involve are ones whose answers have to be "worked out" instead of simply found.

Many of the arguments of descriptivist apologists seem weakly democratic to me. For example, the loosened restrictions on double negatives that Jenny Cheshire defends in Language Myths (myth 14) through either marginalizing Boolean logic (It is very rarely appropriate to think in terms of logic when looking at language use.) or praising historical usage (Multiple negatives are frequent in Chaucer and Shakespeare's work.). She points out that double negatives are frequently used in conversations to re-emphasize the negative, as in the reply not to me it doesn't.

With that example, descriptivism seems appropriate, and could be labeled logical reduplication. However, she ignores the measurable extra time it takes to transform those not uncommon sentences where Boolean logic is relevant: "He didn't not want to go." This is clearly more cumbersome than its simplified Boolean identity: "He wanted to go." Grammar rules are not always arbitrary. There's a time for both a descriptvist approach and a prescritivist approach.

Others will defend split infinitives and dangling prepositions as semi-arbitrary adoptions from Latin rules. The more highly inflectional system of Latin made it impossible to split or dangle because a verb's inflection (conjugation) specified the infinitive form and a noun's inflection (declension) specified its case:

  • dative (to)
  • genetive (of)
  • ablative (from)
  • instrumental (by)
  • comitative (with)

The argument is made that because English can separate these words, the historical rules no longer apply. Against this argument I suggest that, like the prescriptive double negative rule, these forms often delay understanding. Excessive splits serve to subtly, frustratingly, yet effectively delay the action from the reader so that the list of adverbs loses meaning before it can be attached to the verb. Similarly, although dangling is sometimes a justifiable habit up with which you should put, excessive long-distance dependencies can force the reader or listener to transform the sentence before it is fully understood. It is those mental transformations that delay communication.

[ posted by sstrader on 7 June 2004 at 8:58:40 PM in Language & Literature ]