In my last post, which was ostensibly an attempt to "Harvard" a book -- it didn't take, really; I got too interested in specific arguments the book was making -- I touched on Carson Schutze's attempts to reform grammaticality judgment methodology in linguistics. I ILL'ed a monograph by William Labov that he references, and have been flipping through it. Schutze is great for getting a sense of what GJs are and why they are important (and worth reforming) in "traditional" linguistics; Labov is solidly a sociolinguist (he basically invented sociolinguistics while he was a grad student), so his purposes with GJs are slightly different than those of syntacticians -- he, like other sociolinguists, is more interested in studying patterns of variation that make up various dialects.
The word "sociolinguistics" is now used in a somewhat loosey-goosey way compared to what it meant in the 1960s. In my field of TESOL, it's sometimes invoked whenever anything involving sociocultural context of language is brought up. (You could argue that we're following the more Hymes/Gumperz currents of sociolinguistics, I guess, but we often aren't.) But for Labov and his followers, it is essentially the study of variation in language through the use of empirical evidence.
His emphasis on empirical evidence leads Labov to be rightly skeptical of the reliance that some linguists allegedly place on their own intuition when making statements about language. Labov favors observation as linguistic data, for many reasons which I won't go into here. He makes some suggestions about how to reform the use of introspective metalinguistic data in his 1975 monograph What is a Linguistic Fact?
Here are the things he recommends for "continued exploration of grammatical judgments" :
1. 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 languageNext time I'll look more closely at these -- especially #4 and its implications for the notion of 'acceptability' in sociolinguistics, world Englishes, and written texts. ultimately, my goal is to show that the concerns of Schutze and Labov are appropriate for their fields but not necessarily so for researchers who (like me) have more socioculturally-oriented, discourse-based goals. Or something like that. Stay tuned.
2. 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. 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. 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.