Re: CHAT: Tepa & CHAT: Re: minimum phonemes, was Re: vrindo

Matt:
> And Rosta quoth:
>
> >There is a reputable but neglected Sino-Austronesian hypothesis,
> >according to which Chinese is Austronesian. I forget now whether
> >the Tibetan half of Sino Tibetan also goes in with Austronesian.
> >The problem with all such proposals like these is that they tend
> >to be published in obscure and inaccessible places, with available
> >books on classification either defeastistically conservative or
> >based on imagination rather than scholarship.
>
> With regard to your final comment:
> What do you think of Merritt Ruhlen's (sp?) recent book on language
> classification?  I've only just skimmed it myself, but it seems to
> be quite sensible - neither too conservative nor too outlandish.
> Any thoughts?
Do you mean the one published about 10 years ago with a second edition
more recently, called _A guide to the world's languages: Vol 1,
Classification_?
If so, then my main criticism would be:that he pays far too little
attention to the
quality of the works proposing classifications, and to the "security" of
classifications. Given that he's a fanatical lumper, he is overly
sympathetic
to many larger-scale groupings. Basically, you can read Ruhlen & get an
impression of very many of the classifications that have been proposed
(though with no indication of the scholarly quality of the proposals),
but
very little impression of what our current state of knowledge is, and
how
confident we can be about its various constituents.
Ideally, I would like information about how patchy our knowledge of the
languages being classified are, about how strong the evidence of
commonalities between languages are, and about how strong the evidence
is for such commonalities implying a genetic relationship.

 --And.