Truth, reality, caring less
Comments(12) This is essentially just an addition to the commenting going on under this LL post.
Perhaps a good potential contrast would be “could care less”, discussed many times on this blog [i.e., Language Log -RLG...ah, the perils of cut-and-paste] and elsewhere. The question that arises for many is whether this is licensed by the grammar of English, and thus an interesting case of negation not mattering (or rather, equivalent meanings across expressions that differ only in negation; it would be a mistake, I think, to say that in “i couldn’t care less,” that the “not” is meaningless), or if each and every token is the result of a common sort of processing/production error involving certain types of easy-to-miscompute meanings. That is probably a false dichotomy, but it’s nonetheless how I read ML’s “whether such cases are constructions, fixed expressions, idioms or whatever, as opposed to natural mistakes that people often make in using a psychologically difficult combination of elements and structures” (Caring Less with Stress, 2004.07.08).
One might point out that asking if something is “part of a language” in this context is asking a rather special sort of question that might well assume certain things about competence/performance, and at least the abstract existence of a language with a grammar that exists across speakers but also is individually instantiated in the mind of every speaker of “that language” (again, all useful analytic constructs, but you have to look at cases where they don’t capture exactly what you want to say).
Another (not incompatible I hope) question to ask is, how is language constituted in the mind of every individual, and how does that constitution change when exposed to tokens of “couldn’t care less” or “reality…truth”? If any part of what we do when we speak is imitate and approximate others (both based on distant and recent past experiences), something looked at in great detail for speech sounds, but also for lexical entries and syntactic configurations, then it need not be the case that every instance of over- or undernegation is either a performance error OR the trivial instantiation of a fully-licensed construct of the grammar. In fact, the latter option is, if I understand what people like Bob Port are saying, not really something to be considered for the individual. One might also want to bring in notions of hypo/hypercorrection from John Ohala’s work, though in the case of negation the hyper/hypocorrection story makes sense. With “reality being further from the truth,” you’d want some other notions.