Re: Q (Caucasian Elf)

At 10:16 pm -0500 24/2/01, Herman Miller wrote:
>On Sat, 24 Feb 2001 17:19:38 -0600, Danny Wier <dawier@...> [...]
>>from Larry Trask in an article on Basque and Nostratic:
Is there any natlang or hypothesized natlang reconstruction that has _not_
be connected by someone with Basque!

>>1)  !Xu~ (Khoisan) 95
>>2)  Lak (North Caucasian) 60
>>3)  one dialect of Arabic (Afro-Asiatic) 56
>>4)  Panjabi (Indo-European > Indo-Aryan) 59
>>5)  Kabardian (North Caucasian) 48
>>6)  Haida (Na-Dene) 46
>>7)  Nazahua (?) 45
>>    Shilha (Afro-Asiatic) 45
>>9)  Irish Gaelic (Indo-European > Celtic) 6
>>10) Igbo (Niger-Congo) 43
>>    Tlingit (Na-Dene) 43
>>12) Sui 42
>>13) Otomi 41
>>14) Hindi-Urdu (Indo-European) 40
>
>Irish Gaelic? That doesn't seem right. I know it's got 2 versions of most
>consonants (palatalized and velarized), and 3 of some, but I can't see how
>you'd get more than 35 or so in all.
I've tried also and can account for only 36 or so - I can't get the
inventory up to 40, let alone the 44 which, I assume, is meant by the list
above.

>I wonder why Sindhi isn't on the list.
..or any of the Nguni languages, which certainly have high inventories of
consonants.  A quick and conservative count I've made of Xhosa (Bantu >
Nguni), e.g. yields 43 (I think a more accurate count might give something
closer to 50).

Being set to NOMAIL last week, I've missed the beginning of this thread.
Is the list above just given out of interest or what?

Ray.



=========================================
A mind which thinks at its own expense
will always interfere with language.
                   [J.G. Hamann 1760]
=========================================