January 29, 2005

Flag waving

As usual, I agree with almost everything Arnold Zwicky says, but I think explanatory adequacy is a suggestive example of a quirky linguistics meme for precisely the reason he says it is a bad example: as he puts it, someone who uses the expression might just as well be waving flags that say "CHOMSKY" and "MIT". Precisely! Using the term, I'm suggesting, has more to do with saying who we are than saying what we mean.

I'll buy, in humble pie contradistinction to what I said previously, that explanatory adequacy is a technical term. Quite apart from Chomsky's stated views on the meaning of explanatory, we see at least an ostensive definition in the form of explicit contrasts between theories which are explanatorily adequate, and those which are "merely descriptively adequate" as I recall Chomsky put it. (And the word merely is part of this meme.)

How productive is explanatory adequacy? Is it the case that something has explanatory adequacy if and only if it is an explanation which is adequate? If a linguist uses the phrase adequate explanation, is this then a technical term? What differs, I submit, is not the semantics, but the pragmatics. A linguist who uses the phrase adequate explanation to describe a theory would presumably deny that the theory was merely descriptively adequate. So the meanings of adequate explanation and explanatory adequacy must be closely related for linguists in general. Both phrases could be used with Chomsky's original acquisition-centered perspective in mind, although my guess is that in practice both are used without this slant: certainly the practice from Chomsky on has been to use them without heavy use of actual acquisition data. The main difference between a linguist that uses explanatory adequacy and one that uses adequate explanation is typically that the first one waves flags that the second one does not — at least, not as vigorously.

Logically, there are at least two ways that a term like explanatory adequacy could come to be widespread in the field. First, some users of a term they have seen before could be repeating it rather lazily, a boilerplate they hammer into a paper to show whether they approve of a given theory. It would then be snowclonic, to abuse the technical vocabulary of this blog. Alternatively, they could be repeating it because it really captures a feeling they take to be appropriate. And the question then is, do they have this feeing because they want to assert that some x is P, where x is a theory and P is the set of things that satisfy explanatory adequacy. Or do they have the feeling because they know that decorating their thoughts with a patina of Chomskyana will help establish their credentials and their sub-cultural affiliation? And are those linguists (I could name some) who fail to so decorate their papers failing to do so because the concepts do not refer appropriately, or because at some level they want to express their disaffectedness?

[Update: In my original post, I should have taken into account that hundreds of occurrences of explanatory adequacy are bibliographic citations of Chomsky (2001) Beyond Explanatory Adequacy. This does not affect the argument in a big way, since using Google I can put an upper limit on the number of these citations at less than 15% of the total. But perhaps it is time for me to move beyond explanatory adequacy, and figure out a better way to illustrate my pop-sociolinguistic Tensor-derived point, by finding more linguistic memes that don't fit neatly into the category of technical terms. Even better would be if someone does it for me! Suggestions sent to dib AT stanford DOT edu are welcome. So far the best I came up with is the word counterexemplify, which is used almost exclusively by linguists, although with a small minority of philosophical uses, and gets over 100 Google hits in various morphological forms. By comparison, counterexample gets hundreds of thousands of hits, mostly non-linguistic. So which linguist was it that first turned counterexample into a verb (or added a prefix to exemplify)?]

Posted by David Beaver at January 29, 2005 03:24 PM