Posted: Apr 15, 2012 12:58 pm
by katja z
Zwaarddijk wrote:
One thing may be worth noting, though: these don't only happen as historical processes, but any one of them also occurs randomly in speech all the time. When some specific change starts relatively systematically happening in some context, the language is likely to change along that line.

I think the analogy with biological evolution is a useful one. Think of language as of a population. It isn't uniform - there is a certain degree of variation of speech produced by different groups of speakers, different speakers and even by the same speaker at different points and in different situations. Some mutations get fixed in the population, by selection or by drift. Some only get fixed in part of the population, and so language varieties diverge. The historical process is thus grounded in variation that exists at any point in time.

Re epenthesis, I'm not sure the English a/an is such a good example; if I remember correctly, it comes from a weak form of the numeral one, where the n has disappeared except when immediately followed by a vowel. I suppose you could argue that synchronically, it functions the way you've described. Still, if we're talking language change, then this is a different case than, say, the development of Fr. trembler from Lat. tremulare.