From: Richard Wordingham
Message: 66798
Date: 2010-10-24
> --- In cybalist@yahoogroups.com, "G&P" <G.and.P@> wrote:There's also the evidence that sound changes spread through vocabulary word by word. Perhaps there should be some quantification on that, at least for literate peoples. The change [jU@] > [j3] (conventional notation for English, not IPA) in my idiolect while I was at university seems to have affected all words with no arguable syllable division in the [jU@]. (Thus, it affects <pure> but not <skua>.) However, that still leaves the fact that even dialects are bundles of idiolects.
> > Famously, certain Latin words do not show the expected sound change.
> > The reason is that they are borrowings from a rural dialect that did
> > not share those changes. They tend to be words for farm things:
> > cow, plough, and so on.
> And for sacrificial technology, says Douglas, if I understood him correctly.
> > This process can surely be found in almost any language? It does
> > not mean "optional sound change"; it means more than one dialect
> > as a source of vocabulary.
> I think that it means in this context, where we are discussing procedure, that if you argue for an split sound change, you should provide the sociological provenance of each alternative, because if you don't, your sound change will be an optional sound change, which we don't want.