Do Languages Change

Do languages change? Compare Chaucer to Shakespeare and Shakespeare to the Tellytubbies, and it seems obvious that they do over time, and space.

But I'm not so sure this is always the case. Now this is a bit left-field, and I don't know I believe it myself, but it's an idea that interests me.

Upon languages in abstract we impose structures, such as phonemes, morphemes, syntax, and so on. There is little intrinsic within a native speaker, or within the mysterious amalgam of native speakers that represents a language, that can be identified as a phoneme, or a morpheme, or participle. Often (particularly in syntax) occasional words will occupy wedge positions which don't quite work as intended, and unequivocal all-encompassing theories are usually (if not always) Towers of Babel. Though, to pick an example from the stable world of phonetics, Labov's proposed demerger phenomena aided by allophone recognition, represent a future change that should perhaps have remained part of the phonetic structure in the intervening period (though who could have predicted that?): a structure change rather than a language change.

I think it's important then to distinguish time-dependent structure, which we devise to match a language as it appears to transform in time, from a change in the language itself. For example, we might say one phoneme change is an artefact of our structuring and lack of future-knowledge. We can say either that such a symbol represents the time-invariant phoneme "d -> dh -> h -> zero", a transition as much part of the time-invariant language-in-abstract as static-in-time phonemes. Or we can say that the language has the "d" phoneme, and then the language itself changes to have a "dh" phoneme, and so on. In order to adopt the former we clearly must be humble and talk about English-to-the-present rather than Modern English.

How do we distinguish between these things? I think mutual intelligibility is a good measure of language change, as it is for synchronic definitions of languages; anything which compromises intelligibility is a change. Allophone changes are typically structure problems; they represent few impediments to intelligibility. As are many syntactic coinages, which slip unnoticed into the native speaker's grammar: I am going to go to eat pizza.. Theories are inadequate in that they cannot predict the future path of these constant language components, particularly given the vagaries of history and conquest, but this should be [might be] seen as humility in our powers of foretelling and structuring, rather than a change in the language in abstract.

Why is this distinction important? I think that a theory should be the best possible given the data. If we believe, erroneously, that some things are external and unknowable then we are inserting pointless ghosts into the nature of communication.

English, as a whole, the purée of the utterances of native speakers, has always had the capacity that, if a certain series of events were to happen, then, some non-intelligibility-reducing sound change would take place; that this was not evident in the time of Chaucer due to the predispositions of the language being buried deep within the minds of the language's speakers and that many many other potentials were never realized because the world didn't turn out that way, are matters for structure, not for saying the language has changed.

I find the dominant ideas of these perfect or perfectable theories and invisible hands of fate a bit strange, really. Just a thought. A procedure has been described to determine language change as opposed to structural change; that we cannot correctly model the future of language, like weather, is not a fault of the language, or of the weather, it is the fault of the model. Towards better models!

-- DanSheppard

I've edited this to make it as meaningful as I can, but I still can't make much sense of it. Is the proposition that there are latent potentials in natural language, and that the apparent evolution of language according to those latent potentials should be distinguished from more radical changes that arise from other causes, such as chance or foreign borrowings?

CategoryNaturalLanguage