After a detour into Patty Lather's article about paradigm proliferation, which was very difficult to read until the last 5-10 pages, I'm back and ready to talk a bit about Labov's stance on grammaticality judgments and where I want to go with them.
As a reminder, here are Labov's principles:
1. The CONSENSUS Principle: If there is no reason to think otherwise, assume that the judgements of any native speaker are characteristic of all speakers of the language
2. The EXPERIMENTER Principle: If there is any disagreement on introspective judgments, the judgments of those who are familiar with the theoretical issues may not be counted as evidence.
3. The CLEAR CASE Principle: Disputed judgments should be shown to include at least one consistent pattern in the speech community or be abandoned. (If differing judgments are said to represent different dialects, enough investigation of each dialect should be carried out to show that each judgment is a clear case in that dialect.)
4. The VALIDITY Principle: When the use of language is shown to be more consistent than introspective judgments, a valid description of the language will agree with that use rather than introspections.
While I don't have the book in front of me, so I can't refer back to how he unpacks some of these (he's not uncritically accepting of all of them, nor does he expect all linguists to be, if I recall correctly), I do want to mention that Labov is working very much as a 'traditional' linguist here. There is nothing at all wrong about this. I have to stress both of those things, because the direction I suggest for AJTs is totally different than Labov's, but it doesn't mean that I think he is wrong or that my suggestion is the only way to make AJTs better. Labov is a descriptive sociolinguist looking for patterns of variation in speech communities, mostly by means of quantitative research methods. (E.g., surveys.)
On the other hand, the work I am doing is mostly situated in a qualitative tradition. I am still, to a point, interested in empirical evidence of language usage (though perhaps would be more likely to rely on corpus data for this), but
I also take judgment data as empirical evidence: not of how the language is actually used (which is another matter), but how it is viewed/understood/accepted/rejected
by stakeholders.
The world Englishes tradition takes "attitude" very seriously. Users' beliefs about the legitimacy of language, at both the conceptual (macro) and "real usage" (micro) levels, is of paramount importance in determining the status of a variety of English at home and abroad, as it were. We can use other terms besides attitude -- ideology -- but the fact is that people's judgements about language usage are very salient social facts that play an important role in the researcher's (and public's?) understanding of how language(s) exist and operate in society.
(Though this is not strictly related, I am thinking of Eric Henry's article on Chinglish from a couple of years ago which manages to show convincingly that labelling language use as "Chinglish" has much less to do with the linguistic 'stuff' of an utterance than it does with the position of the speaker and social/power differences between the perpetrator of Chinglish and the labeller of Chinglish. He discusses the language in question, but also
how the language is reported and the reporter's attitude toward it/the speaker and ultimately how the latter is the most relevant in the labelling of the utterance as Chinglish.)
Language attitude research has a long history of quantitative studies, while language ideology research relies more on ethnographic methods. (The one coming from (socio)linguistics, the other from linguistic anthropology,
as I touched on here.) I've chosen to use interviews and a kind of qualitative AJT to get at data about attitude/ideology toward Chinese Ss' English use, so I want to address how I modify Labov's last two principles.
#3: Disputed judgments should be shown to include at least one consistent pattern in the speech community or be abandoned.
Because I am not using judgment data as a guarantor of actual usage or of proof that a particular dialect has a particular grammar, I would modify this one to something like : "
Disputed judgments may point to shifts in attitude and/or practice; the dispute is evidence of something interesting happening, and it should be pursued further."
#4: When the use of language is shown to be more consistent than introspective judgments, a valid description of the language will agree with that use rather than introspections.
Again, this is totally right for researchers whose concerns are similar to Labov's. However, I would say something like "
When the use of language is shown to be more consistent than introspective judgments, the judgments should be further investigated to probe the source of the inconsistency."
If my #3 and #4 sound similar -- well, they are. Because ultimately I think that studying the connection between judgment (which I take as a kind of "attitude") and usage
in and of itself -- rather than taking judgment as a kind of backup for usage data, both in service of linguistic description -- is worthwhile, interesting, complicated, and ultimately quite important for understanding how language works in society.
Which, in the end, probably makes me somewhat closer to the sociology of language than it does to sociolinguistics -- but as I said, 'sociolinguistics' covers a multitude of approaches nowadays.