Sum: Yakhontov's principle

[Larry Trask]

----------------------------Original message----------------------------
On Thu, 21 Jan 1999, Alexander Vovin wrote:

> Larry,
>
>     Hmmm... The following puzzles me:
>
> >Claim 2: If the proportion of phonetic resemblances in the 35-word list
> >is higher than the proportion of phonetic resemblances in the 65-word
> >list, then this is evidence that the languages are related.
>
> >This second claim I have big problems with.  I myself do not believe
> >that phonetic resemblances are of any significance at all in comparative
> >linguistics, at least in the absence of a rigorous statistical
> >underpinning.  Given the current conspicuous disagreements among the
> >linguists working on statistical approaches to comparison, and the plain
> >absence of any statistical approach that is generally regarded as valid
> >and effective, there can be no such underpinning at present.
>
>     I also have a big problem with this claim, and to the best of my
> memory, I do not recollect it to be present in the Iakhontov's handout
> that I had before, although he might have added it later. Is that
> something more recent?

I don't know.  This second claim appears to be imputed to Yakhontov by
Starostin in the relevant passage in Starostin's book.  But the wording
is not very explicit, so perhaps this is Starostin's own addition to
Yakhontov's ideas.

All this started with an open exchange of letters between Starostin and
me in the pages of the American journal Mother Tongue.  In his second
letter, Starostin explicitly attributed only the first claim to
Yakhontov, but then went on to invoke the second claim as though it were
the same claim, a maneuver for which I chided him at the time.


Larry Trask
COGS
University of Sussex
Brighton BN1 9QH
UK

larryt at cogs.susx.ac.uk