--- In cybalist@..., "Richard Wordingham" <richard.wordingham@...>
wrote:
> Alert men in white coats!
>
> Warlpiri (Pama-Nyungan) ------------------------- maligi
> Lavukavele (East Papuan) ------------------------ mitakeu
> Russian (Indo-European) ------------------------- sobaka
Disallowed:
Pali (Indo-European) ---------------------------- sunakHa
From Piotr:
>Yes, it's another famous false cognate. The language is Mbabaram
(now extinct; the last living informant was discovered by Bob Dixon
some time ago), and the word in question is <dOg> (with an open [O]
to make it even more English-sounding). It derives from *gudaga, a
well-attested word with numerous cognates (Mbabaram was one of the so-
called initial-dropping Australian languages).
No. 5 in the list! I wonder if it is related to the Anindilyakwa
word below?
> On the other hand:
> Anindilyakwa (Australian) ---------------------- wuruwad.e
> Ama (Left May) --------------------------------- VlOwO:u
> Portuguese (Indo-European) --------------------- cachorro
>
> Reckoning 20% probability of a guttural, the probability of getting
> one of a 4-3, 5-2 6-1 or 7-0 split on whether the third syllable
> begins with a guttural is only 3.3%. Striking out Sanskrit
s'unaka / Pali sunakHa on the grounds of possible cognacy with sobaka
does raise the probability to 9.9%.
Score now 4-3. Possibility of 4-2, if 'wuruwad.e' derives
from 'gudaga'. Unless, of course, English 'dog' established itself
because of the 'g', in which case *gudaga is biased and must be
disallowed, for a root yielding [dOg] stands a reasonable chance of
starting its third syllable with a guttural.
I disallow Sanskrit s'unaka because there is a strong possibility
that the final syllable was interpreted as the same vague diminutive
in both the Pali form and Russian sobaka. There's also the
possibility that they are doublets.
Richard.