Re: THEORY: derivation question

   Date:         Wed, 24 Mar 1999 21:49:10 -0600
   From: dunn patrick w <tb0pwd1@...>

   could such a thing as [p] -> [k] ever happen, or is that just
   silly? It's probably a stupid question, but I want to know.

It just so happens that a case of /p/ > /k/ was discussed on the
IndoEuropean list around the new year. In an Austronesian language
called Skikun, it seems that all instances of word-final /-p/ have
been changed to /-k/ in the space of two generations. There are no
conditioning factors such as dissimilation present, just a steady
replacement of forms with /-p/ to forms with /-k/ in the lexicon of
each successive age group of speakers.

By the way, this is something shown by recent research into how sound
changes actually happen: A sound change does not usually happen all at
once, or by gradual modification of the phonetical realization through
generations until some end point is reached.

Rather, a change --- often phonetically abrupt --- starts in a small
number of words, with a small number of speakers, and spreads through
both the population and the lexicon. Or perhaps it doesn't --- it may
disappear again, or become a feature of a dialect or sociolect, or
spread only to some words where a certain condition is satisfied, or
even remain in a few words as unexplained exceptions to sound laws.

And people do adopt new pronunciations throughout their lives. Speaker
age does play a role --- but if older people always continued to speak
exactly as they did when they were twenty years old, the difference
between generations would be much larger than actually observed.

Lars Mathiesen (U of Copenhagen CS Dep) <thorinn@...> (Humour NOT marked)