Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

more from Hill-Tout

18 views
Skip to first unread message

Yuri Kuchinsky

unread,
Feb 14, 1999, 3:00:00 AM2/14/99
to

In this post I will look at some of the quibbles about Hill-Tout's
evidence Ross Clark has posted recently. Most of them are completely
without merit, and in some cases they're actually at odds with what Ross
has been saying himself. Once again, his objections demonstrate his
willful blindness and inability to understand new data that is in conflict
with his cherished faith and preconceptions.

I will consider the following 4 items.


MOON

mahina, Polynesian

Ross wrote on 1999/02/05:

> > Polynesian "mahina' (moon) may resemble
> > something Hill-Tout writes as "ma-hin" in some interior Salish
> > language.

This is a fine example of failing to get the point. First of all, there
are also some seeming cognates with other NW Coast languages:

ma-qha-ten = moon, Ntlakap.
mehme-khtl, = moon, Bilq.
ma-ma = light (widely shared)

And Hill-Tout actually wrote in his paper (that Ross claims to have read
carefully -- yeah, right!) that some of the best parallels with
Austronesian actually exist in interior Salish -- as well as in Bilqula, a
coastal Salish language. How could he have missed it?

But Hill-Tout doesn't present so much material for "moon". So let's go to
the next item.


TOOTH

Ross wrote on 1999/02/14:

> > I think Greg would be
> > referring to such words as /gigi/ for "tooth", identical in Malay and
> > Kwakiutl. Amazingly, this word completely disappears in Polynesian,

But maybe not? In any case, this is not the main point.

> > and
> > instead what turns up is /nifo/, which by an amazing coincidence just
> > happens to have cognates throughout Oceanic.

Bullocks! As it happens, gigi has much better cognates throughout Oceanic
-- and this is very important.

It is as clear as can be, based on Hill-Tout, that good cognates for gigi
are common not only in Oceanic languages, but also on NW Coast. This is a
very far ranging similarity indeed!

(Here, and in the following cases, I'm just presenting some selections
from the copious data collected by Hill-Tout. See his paper for more
cognates.)

Oceanic:

gi gi, Salayer, Baju
ngisi, Menado
nisi-nen, Massaratty

NW Coast:

tshi-tshi-sh, Nootka
gi-geis, Thatl.
rei-tshi-min, Lill.
Aei-te-men, Okana.

(The last example is particularly interesting, since we have here a NW
Coast language that does not have those consonants that Mike recently
claimed are so distinctive, special, and omnipresent on NW Coast.)

So what do we have here? We have clear and rather obvious wide parallels
between NW Coast and Austronesian non-Polynesian languages. The parallels
are much better between NW Coast and Austronesian non-Polynesian languages
than between Polynesian languages and other Austronesian languages. But
this is exactly what my hypothesis has predicted! So my hypothesis has
demonstrable predictive power. QED.

And the same thing exactly happens with the following example that Ross
also missed completely. Willful blindness?

> FIRE (general sense)
>
> In Polynesia:
>
> 1. Rotto, hai
> 2. Mysol (coast), lap

ROSS WROTE:
Yuri, do you have any clue at all where "Rotto" and "Mysol" might be? Do
you actually think they are "in Polynesia"?

MY REPLY:
And you failed to see the forest behind the trees, Ross.

> On NW Coast:
>
> 1. Kwakiutl, hai
> Snan., haiuk
>
> 2. Ntlakap, lap, or aap [sunset, evening]

Again, this is evidence that NW Coast languages are connected very closely
to some Austronesian non-Polynesian languages. This supports my hypothesis
that some NW Coast languages are Austronesian.


NEGATIVES

Once again, Ross Clark has been stonewalling. He refuses to go to the
cited sources, and demands that I spoon-feed him with more and more
details.

Well, here they are -- very easy to look up. The blind will always remain
blind, or so it seems.

Polynesian negatives
--------------------

1. te, Maori
tai, Tongan

2. aua, Maori

3. i, according to Fornander, this is the basic Polynesian form. Many more
examples are given by Hill-Tout.


NW Coast negatives
------------------

1. taa, Shew.
taa, Kull.

2. aua, Songes
whaa, Nisk.

3. i, according to Hall, is basic Kwakiutl negative.

All this evidence that I've presented supports my theory wonderfully. Ross
Clark missed all of this repeatedly, because he lacks objectivity. Much
other evidence I've presented has still been ignored by everybody, because
it is uncomfortable to all those who are wedded to prevailing false
dogmas.

Regards,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku

Man will occasionally stumble over the truth, but most of the time he
will pick himself up and continue on -=O=- Churchill's Commentary on Man

Anthony West

unread,
Feb 14, 1999, 3:00:00 AM2/14/99
to

Yuri Kuchinsky wrote in message <7a79sv$eom$1...@whisper.globalserve.net>...

>
>In this post I will look at some of the quibbles about Hill-Tout's
>evidence Ross Clark has posted recently
>
>I will consider the following 4 items.
>
[snip more amateur stabs at linguistic research]

>So what do we have here? We have clear and rather obvious wide parallels
>between NW Coast and Austronesian non-Polynesian languages. The parallels
>are much better between NW Coast and Austronesian non-Polynesian languages
>than between Polynesian languages and other Austronesian languages. But
>this is exactly what my hypothesis has predicted! So my hypothesis has
>demonstrable predictive power. QED.
>

Yes. Your hypothesis demonstrates that an untrained reader who keeps
re-reading and repeating the failed comparisons of 100-year-old,
long-abandoned scholarship, without bothering to learn anything that modern
scholars have learned since 1898, will keep on making the same mistakes
forever. There is no substitute for study.

Since you have no background at all in linguistics, Yuri, and perhaps are
too poor to afford a course in linguistics, try tracking this book down:
"Principles of Historical Linguistics" / Hans Henrich Hock, 2nd ed.,
rev. and updated; Berlin, New York, Amsterdam; Mouton de Gruyter, 1991. ISBN
3-11-012962-0.

Hock provides an introduction to comp ling that is detailed and up-to-date,
while quite clear throughout. I find it an excellent entrée for beginners.
Yet it's old enough that you can probably find it at used-book rates now.
Try a search on abebooks.com.

If you read this book, your sci.lang posts will improve immeasurably in
quality.

Regards,

Tony West
Philadelphia aaw...@critpath.org


Miguel Carrasquer Vidal

unread,
Feb 14, 1999, 3:00:00 AM2/14/99
to
On 14 Feb 1999 19:58:23 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Ross wrote on 1999/02/14:
>
>> > I think Greg would be
>> > referring to such words as /gigi/ for "tooth", identical in Malay and
>> > Kwakiutl. Amazingly, this word completely disappears in Polynesian,
>
>But maybe not? In any case, this is not the main point.
>
>> > and
>> > instead what turns up is /nifo/, which by an amazing coincidence just
>> > happens to have cognates throughout Oceanic.
>
>Bullocks! As it happens, gigi has much better cognates throughout Oceanic
>-- and this is very important.
>
>It is as clear as can be, based on Hill-Tout, that good cognates for gigi
>are common not only in Oceanic languages, but also on NW Coast. This is a
>very far ranging similarity indeed!
>
>(Here, and in the following cases, I'm just presenting some selections
>from the copious data collected by Hill-Tout. See his paper for more
>cognates.)
>
>Oceanic:
>
>gi gi, Salayer, Baju
>ngisi, Menado
>nisi-nen, Massaratty

Selayar and Menado are on Sulawesi, you moron. These are not
Oceanic languages.


=======================
Miguel Carrasquer Vidal
m...@wxs.nl
Amsterdam

Mike Cleven

unread,
Feb 15, 1999, 3:00:00 AM2/15/99
to
On 14 Feb 1999 19:58:23 GMT, yu...@globalserve.net (Yuri Kuchinsky)
wrote:

>


>In this post I will look at some of the quibbles about Hill-Tout's
>evidence Ross Clark has posted recently. Most of them are completely
>without merit, and in some cases they're actually at odds with what Ross
>has been saying himself. Once again, his objections demonstrate his
>willful blindness and inability to understand new data that is in conflict
>with his cherished faith and preconceptions.

Well, we'll see what objections - or even validations - might come
from the Salishan experts I've just forwarded your whining to, Yuri.
Hill-Tout made numerous errors in his recording of languages, but
we'll see if anything below is actually even correct. You've cited
incorrect sources before, after all......

The fun part, for the moment, will be to see if Yuri can actually
identify the languages referred to by Hill-Tout's
abbreviations.....the _modern_ names, Yuri, and which language
families they belong to, and where they are. Show us you actually
know something. And don't go running to the Encyclopaedia Canadiana
for help! (it's probably out of date, anyway)

>
>ma-qha-ten = moon, Ntlakap.
>mehme-khtl, = moon, Bilq.
>ma-ma = light (widely shared)
>
>And Hill-Tout actually wrote in his paper (that Ross claims to have read
>carefully -- yeah, right!) that some of the best parallels with
>Austronesian actually exist in interior Salish -- as well as in Bilqula, a
>coastal Salish language. How could he have missed it?

There's an error here, used by Yuri. Do you think he has any idea
what it is?


<snip>

>
>tshi-tshi-sh, Nootka
>gi-geis, Thatl.
>rei-tshi-min, Lill.
>Aei-te-men, Okana.
>
>(The last example is particularly interesting, since we have here a NW
>Coast language that does not have those consonants that Mike recently
>claimed are so distinctive, special, and omnipresent on NW Coast.)

To this point, the simplest reply is to ask Yuri how he thinks that
the sounds in "tshi-tshi-sh" are comparable/interchangeable with the
voiced guttural of "gi-geis", or to the nasal-labial of "Aei-te-men".
And to point out that there are just as many variants of t, t!, t',
tl, th, thl, tlh, etc. as there are of the gutturals and palatals
previously listed.........


>
>> On NW Coast:
>>
>> 1. Kwakiutl, hai
>> Snan., haiuk
>>
>> 2. Ntlakap, lap, or aap [sunset, evening]

>------------------
>
>1. taa, Shew.
>taa, Kull.
>
>2. aua, Songes
>whaa, Nisk.
>

Mike Cleven
http://members.home.net/ironmtn/

The thunderbolt steers all things.
- Herakleitos


Ross Clark

unread,
Feb 15, 1999, 3:00:00 AM2/15/99
to
Yuri Kuchinsky wrote:
>
> In this post I will look at some of the quibbles about Hill-Tout's
> evidence Ross Clark has posted recently. Most of them are completely
> without merit, and in some cases they're actually at odds with what Ross
> has been saying himself. Once again, his objections demonstrate his
> willful blindness and inability to understand new data that is in conflict
> with his cherished faith and preconceptions.

And my compliments to your mother, Yuri.

>
> I will consider the following 4 items.
>
> MOON
>
> mahina, Polynesian
>
> Ross wrote on 1999/02/05:
>
> > > Polynesian "mahina' (moon) may resemble
> > > something Hill-Tout writes as "ma-hin" in some interior Salish
> > > language.
>
> This is a fine example of failing to get the point. First of all, there
> are also some seeming cognates with other NW Coast languages:
>
> ma-qha-ten = moon, Ntlakap.

CV resemblance

> mehme-khtl, = moon, Bilq.

C resemblance

> ma-ma = light (widely shared)

CV resemblance, doesn't mean "moon", what does "widely shared"
mean? I see two neighbouring Interior Salish languages with this word,
plus a certain amount of vague Hilltoutian handwaving

So you really think all of these are "seeming cognates", ie derived from
a single original form? Like to make a guess at what that proto-form
might be?


> And Hill-Tout actually wrote in his paper (that Ross claims to have read
> carefully -- yeah, right!) that some of the best parallels with
> Austronesian actually exist in interior Salish -- as well as in Bilqula, a
> coastal Salish language. How could he have missed it?

I didn't miss it. He said it. I did not say he didn't say it. What the
hell are you on about? Is it that you simply have such an uncontrollable
itch to vilify other people that you must accuse them of duplicity every
few lines even when there's no basis for it?


> But Hill-Tout doesn't present so much material for "moon". So let's go to
> the next item.
>
> TOOTH
>
> Ross wrote on 1999/02/14:
>
> > > I think Greg would be
> > > referring to such words as /gigi/ for "tooth", identical in Malay and
> > > Kwakiutl. Amazingly, this word completely disappears in Polynesian,
>
> But maybe not? In any case, this is not the main point.
>
> > > and
> > > instead what turns up is /nifo/, which by an amazing coincidence just
> > > happens to have cognates throughout Oceanic.
>
> Bullocks!

Thundering down the track! Two by two!

>As it happens, gigi has much better cognates throughout Oceanic
> -- and this is very important.
>
> It is as clear as can be, based on Hill-Tout, that good cognates for gigi
> are common not only in Oceanic languages, but also on NW Coast. This is a
> very far ranging similarity indeed!
>
> (Here, and in the following cases, I'm just presenting some selections
> from the copious data collected by Hill-Tout. See his paper for more
> cognates.)
>
> Oceanic:
>
> gi gi, Salayer, Baju

Identical to gigi, not Oceanic -- Western Malayo Polynesian,
spoken in Sulawesi, Indonesia.

> ngisi, Menado

V...V resemblance, not Oceanic (a local Sulawesi dialect of
Malay)

> nisi-nen, Massaratty

V...V resemblance, don't know this language, not much use asking
Yuri. Probably same area, judging from the list in Tryon



> NW Coast:
>
> tshi-tshi-sh, Nootka

Nootka really is related to Kwakiutl, so this could be a real
cognate, with palatalization and devoicing of *g, I guess

> gi-geis, Thatl.

CVC resemblance, don't know what language this is

> rei-tshi-min, Lill.

One syllable looks like Nootka, but since Lillooet is up in the
interior, borrowing seems unlikely,

> Aei-te-men, Okana.

Could be related to the previous item since this is another
interior language; all resemblance to "gigi" has vanished at this
point



>
> (The last example is particularly interesting, since we have here a NW
> Coast language that does not have those consonants that Mike recently
> claimed are so distinctive, special, and omnipresent on NW Coast.)

What nonsense. If Mike did say they were "omnipresent", only someone as
comprehensively ignorant of linguistics as yourself could have taken that
literally. And in any case, your sources are so primitive that you would
have no way of knowing if various of these consonants were glottalized or
whatever.

>
> So what do we have here? We have clear and rather obvious wide parallels
> between NW Coast and Austronesian non-Polynesian languages. The parallels
> are much better between NW Coast and Austronesian non-Polynesian languages
> than between Polynesian languages and other Austronesian languages.

I think we really are dealing with the Eyes of Faith here, folks. Yuri,
like Hill-Tout, looks at this ragbag and seas "clear and rather obvious
wide parallels". Never mind that none of the languages he listed are
Oceanic. Never mind that the phonetic resemblances fade away as fast as
you look at them. If I showed him 200 languages with cognates of
Polynesian /nifo/ he would find some way to ignore them. Nothing to be
done.

> But
> this is exactly what my hypothesis has predicted! So my hypothesis has
> demonstrable predictive power. QED.

Light the fireworks.

>
> And the same thing exactly happens with the following example that Ross
> also missed completely. Willful blindness?

Again, I didn't miss it. Get help for that Tourette's before you get
yourself arrested.


> > FIRE (general sense)
> >
> > In Polynesia:
> >
> > 1. Rotto, hai

Probably means Roti, which has /ha?i/

> > 2. Mysol (coast), lap

Don't find this language listed anywhere

>
> ROSS WROTE:
> Yuri, do you have any clue at all where "Rotto" and "Mysol" might be? Do
> you actually think they are "in Polynesia"?
>
> MY REPLY:
> And you failed to see the forest behind the trees, Ross.

Translation: No, I don't have any clue where they are. Uh-oh, I guess
they're not in Polynesia. But look at the Big Picture!....

>
> > On NW Coast:
> >
> > 1. Kwakiutl, hai

Not in my Kwakiutl dictionary. Could be intended for initial
/xi-/ in various words to do with fire & burning

> > Snan., haiuk

I assume this is the island dialect of Halkomelem. Could be
something like Cowichan /hay?qw/.

> >
> > 2. Ntlakap, lap, or aap [sunset, evening]
>
> Again, this is evidence that NW Coast languages are connected very closely
> to some Austronesian non-Polynesian languages. This supports my hypothesis
> that some NW Coast languages are Austronesian.

More feeble scraps. CV resemblance between Roti and Snan, maybe CVV if
we're being generous. Could be CVC between "Mysol" and Ntlakap., except
that you switched meanings.

>
> NEGATIVES
>
> Once again, Ross Clark has been stonewalling. He refuses to go to the
> cited sources, and demands that I spoon-feed him with more and more
> details.

You're missing the pedagogical purpose here, Yuri. I realize I'm probably
wasting my time, but it is an attempt to get you to take a more
responsible attitude towards your data if you are ever going to grow up
to be a scientist.



> Well, here they are -- very easy to look up. The blind will always remain
> blind, or so it seems.
>
> Polynesian negatives
> --------------------
>
> 1. te, Maori

/te:/

> tai, Tongan

/ta'e/
>
> 2. aua, Maori
(and dozens of other forms reflecting PPN *kaua or *'aua,
negative imperative)


>
> 3. i, according to Fornander, this is the basic Polynesian form. Many more
> examples are given by Hill-Tout.

Well, that's what he gets for taking Fornander as a linguistic authority.
And in fact he doesn't give any examples at all of Polynesian languages
with /i/ as a negative, because there aren't any. He gives a few negative
words with /i/ in them...and a few with /a/ in them, and /u/ in
them...there are only five vowels after all.

>
> NW Coast negatives
> ------------------
>
> 1. taa, Shew.
> taa, Kull.

I have no way of checking these.

>
> 2. aua, Songes
> whaa, Nisk.

The Coast Salish languages I have material on have a negative
verb something like /?aw@/ or /?@w@/

> 3. i, according to Hall, is basic Kwakiutl negative.

My guess is that "the Rev.Mr.Hall, for many years missionary among the
Kwakiutl" would be a linguistic authority of the same style and same
degree of reliability as Fornander. My sources suggest that /k'i/ is the
basic negative root in Kwakw'ala.

Scoring the negatives:

1. Remains to be checked
2. Better than average, could be CVCV
3. Nil


> All this evidence that I've presented supports my theory wonderfully.

This is a delusion.

>Ross
> Clark missed all of this repeatedly, because he lacks objectivity.

This is a lie.

Much
> other evidence I've presented has still been ignored by everybody, because
> it is uncomfortable to all those who are wedded to prevailing false
> dogmas.

This is just routine raving.

Ross Clark

GKeyes6988

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to

Greg Keyes wrote:

.Central Malayo-Polynesian

.DA''A NGISI
.ROTI NISI-K


Sorry -- DA'A is not central, but Western MP.

-- Greg Keyes

GKeyes6988

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
Yuri Kuchinsky wrote:

>TOOTH
>
>Ross wrote on 1999/02/14:
>
>> > I think Greg would be
>> > referring to such words as /gigi/ for "tooth", identical in Malay and
>> > Kwakiutl. Amazingly, this word completely disappears in Polynesian,
>
>But maybe not? In any case, this is not the main point.
>
>> > and
>> > instead what turns up is /nifo/, which by an amazing coincidence just
>> > happens to have cognates throughout Oceanic.
>
>Bullocks! As it happens, gigi has much better cognates throughout Oceanic
>-- and this is very important.
>
>It is as clear as can be, based on Hill-Tout, that good cognates for gigi
>are common not only in Oceanic languages, but also on NW Coast. This is a
>very far ranging similarity indeed!
>
>(Here, and in the following cases, I'm just presenting some selections
>from the copious data collected by Hill-Tout. See his paper for more
>cognates.)
>
>Oceanic:
>
>gi gi, Salayer, Baju
>ngisi, Menado
>nisi-nen, Massaratty

And, as has been pointed out, none of these is Oceanic. Yuri doesn't know what
"Oceanic" is, as he has demonstrated elsewhere. But then, the less he knows
about something the easier it is to be sure of the Truth, I suppose.


>NW Coast:
>
>tshi-tshi-sh, Nootka
>gi-geis, Thatl.
>rei-tshi-min, Lill.
>Aei-te-men, Okana.
>
>(The last example is particularly interesting, since we have here a NW
>Coast language that does not have those consonants that Mike recently
>claimed are so distinctive, special, and omnipresent on NW Coast.)

Dealt with by Ross.

>So what do we have here? We have clear and rather obvious wide parallels
>between NW Coast and Austronesian non-Polynesian languages. The parallels
>are much better between NW Coast and Austronesian non-Polynesian languages
>than between Polynesian languages and other Austronesian languages. But
>this is exactly what my hypothesis has predicted! So my hypothesis has
>demonstrable predictive power. QED.

Pretty nice, using Hill-Tout and Heyerdahl to "test" what Hill-Tout and
Heyerdahl claim.

But no, lets review. Hypothesis 1:

A.Very early Austronesians came to the NW
B. After millenia, some went on to Hawai'i and became Polynesians.

What this predicts (Yuri failed to catch the real implications) is

(1.) the POLYNESIAN languages should resemble NW coast languages more than they
resemble other Astronesian languages and

(2)that POLYNESIAN languages should be more like Austronesian languages from SE
Asia (*) than they are like their nearest geographical neighbors in Micronesia
and Melanesia (speakers of Oceanic languages). This would be because, say,
Woleaian and Hawa'ain would have been seperated from proto-Austronesian times
(since Polynesians were in Vancouver and not Near Oceania) and would not share
the same innovations.

*(Yuri says Japan but there aren't any Austronesian languages there - yah, I
know the Austro-Tai /Japanese possibility, but that changes the time depth a
bit, among other things)

Let me try a tree diagram of the prediction:

Austronesian Somewhere in Island SE Asia
I
I__________________________
I I I

I I I
I (lots of languages) I
I I
I___________ I
I I I
I I I
I I I

"NW Polynesian Oceanic languages
Coast languages"

Hypothesis # 2

Austronesian speakers moved west from SE Asia.

This predicts:

A.Polynesian is most closely related to the Oceanic languages, less closely
related to the central Malayo-Polynesian languages, and so on.

B. It's not related to NW coast languages at all, and no systematic comparison
of vocabulary will reveal a regular set of cognates between them.

(I''ll also throw in another fact that's been brought up and which Yuri ignores
-- Kwakiutl and Salishan, for instance, have no known genetic afiiliation.
Using "NW coast languages" in a search for cognates with Hawai'ian is like
using a combined vocabulary of English, Chinese, and Choctaw to find cognates
with !Kung. Comparing all NW coast languages to all Austronesian ones as a
means of turning up "cognates" is like using Dakota, Japanese, and Malagasy to
compare with any and all of the Indo-European languages.)

Ah well, the word.

TOOTH.
(Polynesian)
Tongan NIFO
Samoan NIFO
Tahitian NIFO

(other Oceanic)

Kwaio NIFO (-na)
Lau LIFO
Roviana LIVO
Tawala NIWO-NA
Raga LIVO-
Port Sandwich RIVON

SHWNG

Sawai NGANG-O

Central Malayo-Polynesian

DA''A NGISI
ROTI NISI-K

Western Malayo-Polynesian

Indonesian GIGI
Balinese GIGI
Uma NGIHI?
Wolio NGINCHI

Madagascar:
Malagasy NIFI

(Borneo)
Murut DIPON

(Phillipines)
Tagalog IPIN
Isnag NgIPAN

Formosan languages
Atayal &I?NUX
Tsou XISI
Paiwanese ALyIS

($= voiced velar fricative)


In the list of eighty terms I checked, nine had CVCV agreement with GIGI.
All were in Indonesia, in the center of the western MP region. Move
NWcoastward -- Borneo, the Phillipines, Taiwan -- and you get forms that less
and less resemble GIGI, never mind the other NW coast forms given. Move into
Oceania, and the resemblance vanishes. Oddly, at the other end of the
Austronesian world, we find a cognate pretty close to the Oceanic form.

Maybe the reconstructed form would be GIGI, I don't know. If so, I have a
suggestion at the end of this section.

How does this bode for Yuri hypothesis? Well, we have 0 resemblance between
the NW coast and Polynesian -- very bad, because that's where the closest
relationship ought to be, given that Polynesians were supposed to have come
from there. Between the CMP GIGI forms and the NW coast we have (at best)
forms like Ngipan, which if we really stretch have a CV resemblance. That's bad
too, because NW coast languages ought to be something like the near Asian ones
-- and more to the point (since NW coast language have no systematic
resmeblance to ANY Austronesian language) Polynesian also ought to be closer
to those than it is to Oceanic languages like Raga. Instead we see the
Polynesian terms are most like Micronesian and Melanesian ones and less like
those elsewhere in the Austronesian area (except in Madagascar).

Here's an interesting test, Yuri. Since Kwakw''ala GIGI is identical with
Balinese GIGI, it's clear that both languages have conserved your putative
proto-Polynesian/Kwakiutl /G/ as the other Austronesian languages have not. So
you ought to be able to find scads of Balinese/Kwakiutl "G" cognates.

All thing considered, this word is a good test of hypothesis #2.

>And the same thing exactly happens with the following example that Ross
>also missed completely. Willful blindness?
>
>> FIRE (general sense)
>>
>> In Polynesia:
>>
>> 1. Rotto, hai
>> 2. Mysol (coast), lap

This language is not included in the eighty-language list in the COMPARATIVE
AUSTRONESIAN DICTIONARY . Nor does the form LAP appear in any of those which
do appear there.



>ROSS WROTE:
>Yuri, do you have any clue at all where "Rotto" and "Mysol" might be? Do
>you actually think they are "in Polynesia"?
>
>MY REPLY:
>And you failed to see the forest behind the trees, Ross.

And Yuri fails to know what a Polynesian language is, much less an Oceanic one.



> On NW Coast:
>
> 1. Kwakiutl, hai
> Snan., haiuk

According to my dictionary, the Kwakw'ala word for Fire is XIKELA. XI is a
suffix for words to do with fire, but is not free-standing.


> 2. Ntlakap, lap, or aap [sunset, evening]

Which is hardly a word for fire. If you want to open the semantic realm like
that, you need a set of regular sound correspondances to make it plausible. And
what lexicon is this from?

.Again, this is evidence that NW Coast languages are connected very closely
.to some Austronesian non-Polynesian languages. This supports my hypothesis
.that some NW Coast languages are Austronesian.

And, again Yuri fails his logic test, not understanding the implications of his
own hypothesis. See above for more specfic comments

FIRE
Polynesian

Tahitian AUAHI
Tongan AFI
Samoan AFI


Other Oceanic

Woleaian YAFI
Rotuman ReHI
Motu LAHI
Raga $ABI
Takia YA
Dami YA
Lewo Kapi

SHWNG
Sawai LUTEN

Central MP
Roti HA?I

Western MP
Balinese API
Da'a API
Konjo API
Malagasy AFU

Murut APUY
Tagalog APOY

Formosan:
Tsou PUZU
Atayal HAPUNIK


Notice how many of these look like Kwakiutl XI? Of the eighty terms, only one
had the shape HA?I, or anything close. Bear in mind that ? Is a consonant.
And as for Lap -- it doesn't mean "fire", and if it does exist in
Austronesian, its form isn't very common.


Again, it's pretty easy to see where the Polynesian languages fall out, here.

-- Greg Keyes

GKeyes6988

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
.According to my dictionary, the Kwakw'ala .word for Fire is XIKELA. XI is a
.suffix for words to do with fire, but is not .free-standing.


And here I meant prefix, of course.

-- Greg Keyes

Yuri Kuchinsky

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Sun, 14 Feb 1999 21:16:27 GMT:
: On 14 Feb 1999 19:58:23 GMT, yu...@globalserve.net (Yuri
: Kuchinsky) wrote:

: >Oceanic:


: >
: >gi gi, Salayer, Baju
: >ngisi, Menado
: >nisi-nen, Massaratty

: Selayar and Menado are on Sulawesi, you moron. These are not
: Oceanic languages.

Thank you Miguel for your kind correction. I should have said
"Austronesian".

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

Reality is that which, when you stop believing
in it, doesn't go away -=O=- Philip K. Dick

Yuri Kuchinsky

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
Ross Clark (d...@antnov1.auckland.ac.nz) wrote on Mon, 15 Feb 1999 17:01:28 +1200:
: Yuri Kuchinsky wrote:

> > MOON
> >
> > mahina, Polynesian
> >
> > Ross wrote on 1999/02/05:
> >
> > > > Polynesian "mahina' (moon) may resemble
> > > > something Hill-Tout writes as "ma-hin" in some interior Salish
> > > > language.
> >
> > This is a fine example of failing to get the point. First of all, there
> > are also some seeming cognates with other NW Coast languages:
> >
> > ma-qha-ten = moon, Ntlakap.
>
> CV resemblance
>
> > mehme-khtl, = moon, Bilq.
>
> C resemblance
>
> > ma-ma = light (widely shared)
>
> CV resemblance, doesn't mean "moon", what does "widely shared"
> mean? I see two neighbouring Interior Salish languages with this word,

Nope. Bilqula is not interior. There are three NW Coast languages with
this word. How many do you want?

> So you really think all of these are "seeming cognates", ie derived from
> a single original form? Like to make a guess at what that proto-form
> might be?

I have no idea. The logical place to look for a proto-form would be NE
Asia.

> > And Hill-Tout actually wrote in his paper (that Ross claims to have read
> > carefully -- yeah, right!) that some of the best parallels with
> > Austronesian actually exist in interior Salish -- as well as in Bilqula, a
> > coastal Salish language. How could he have missed it?
>
> I didn't miss it. He said it. I did not say he didn't say it. What the
> hell are you on about? Is it that you simply have such an uncontrollable
> itch to vilify other people that you must accuse them of duplicity every
> few lines even when there's no basis for it?

No, I'm just commenting on your apparent inability to perceive the
significance of these pretty clear similarities.

> > TOOTH

> > It is as clear as can be, based on Hill-Tout, that good cognates for gigi
> > are common not only in Oceanic

I should have said "Austronesian" here.

> > languages, but also on NW Coast. This is a
> > very far ranging similarity indeed!
> >
> > (Here, and in the following cases, I'm just presenting some selections
> > from the copious data collected by Hill-Tout. See his paper for more
> > cognates.)
> >

> > Oceanic [rather Austronesian]:


> >
> > gi gi, Salayer, Baju
>
> Identical to gigi, not Oceanic -- Western Malayo Polynesian,
> spoken in Sulawesi, Indonesia.
>
> > ngisi, Menado
>
> V...V resemblance, not Oceanic (a local Sulawesi dialect of
> Malay)
>
> > nisi-nen, Massaratty
>
> V...V resemblance, don't know this language, not much use asking
> Yuri. Probably same area, judging from the list in Tryon
>
> > NW Coast:
> >
> > tshi-tshi-sh, Nootka
>
> Nootka really is related to Kwakiutl, so this could be a real
> cognate, with palatalization and devoicing of *g, I guess
>
> > gi-geis, Thatl.
>
> CVC resemblance, don't know what language this is
>
> > rei-tshi-min, Lill.
>
> One syllable looks like Nootka, but since Lillooet is up in the
> interior, borrowing seems unlikely,

Ross is begging the question here.

> > Aei-te-men, Okana.
>
> Could be related to the previous item since this is another
> interior language; all resemblance to "gigi" has vanished at this
> point

That's because you're refusing to go to the cited source and to find many
more intermediate forms. You're still stonewalling.

The general point is pretty clear. This Austronesian word is also rather
widely shared by NW Coast languages. I don't see why Ross would try to
deny the obvious.

> > > FIRE (general sense)
> > >
> > > In Polynesia:
> > >
> > > 1. Rotto, hai
>
> Probably means Roti, which has /ha?i/
>
> > > 2. Mysol (coast), lap
>
> Don't find this language listed anywhere

This is an island near New Guinea. Other islands in the area are Aru,
Waigiou, and Jobie.

> > > On NW Coast:
> > >
> > > 1. Kwakiutl, hai
>
> Not in my Kwakiutl dictionary. Could be intended for initial
> /xi-/ in various words to do with fire & burning
>
> > > Snan., haiuk
>
> I assume this is the island dialect of Halkomelem. Could be
> something like Cowichan /hay?qw/.

Snanmth village of the Sto:lo people? Salish (Halq'emiylem).

> > >
> > > 2. Ntlakap, lap, or aap [sunset, evening]
> >
> > Again, this is evidence that NW Coast languages are connected very closely
> > to some Austronesian non-Polynesian languages. This supports my hypothesis
> > that some NW Coast languages are Austronesian.
>
> More feeble scraps. CV resemblance between Roti and Snan, maybe CVV if
> we're being generous. Could be CVC between "Mysol" and Ntlakap., except
> that you switched meanings.

Again, perhaps it would help to consult Hill-Tout's paper for more
intermediate forms?

> > NEGATIVES

> > Polynesian negatives
> > --------------------
> >
> > 1. te, Maori
> /te:/
>
> > tai, Tongan
>
> /ta'e/

Thanks for providing modern transcriptions.

> > 2. aua, Maori
> (and dozens of other forms reflecting PPN *kaua or *'aua,
> negative imperative)
> >
> > 3. i, according to Fornander, this is the basic Polynesian form. Many more
> > examples are given by Hill-Tout.
>
> Well, that's what he gets for taking Fornander as a linguistic authority.
> And in fact he doesn't give any examples at all of Polynesian languages
> with /i/ as a negative, because there aren't any. He gives a few negative
> words with /i/ in them...and a few with /a/ in them, and /u/ in
> them...there are only five vowels after all.

Somehow I doubt that Fornander failed to be informed about such a basic
matter. I don't see why I should take your word over Fornander's.

> > NW Coast negatives
> > ------------------
> >
> > 1. taa, Shew.
> > taa, Kull.
>
> I have no way of checking these.

Again, you can consult Hill-Tout for more.

> > 2. aua, Songes
> > whaa, Nisk.
>
> The Coast Salish languages I have material on have a negative
> verb something like /?aw@/ or /?@w@/

Thanks for providing modern transcriptions.

> > 3. i, according to Hall, is basic Kwakiutl negative.
>
> My guess is that "the Rev.Mr.Hall, for many years missionary among the
> Kwakiutl" would be a linguistic authority of the same style and same
> degree of reliability as Fornander.

And I suspect your dismissal of Hall is even less valid than that of
Fornander.

> My sources suggest that /k'i/ is the
> basic negative root in Kwakw'ala.

I don't know why I should take your word over Hall's. After all, he lived
and worked in the area for a long time.

> Scoring the negatives:
>
> 1. Remains to be checked
> 2. Better than average, could be CVCV
> 3. Nil

Your objectivity level is nil, correct -- as was already known. The
evidence I've presented seems valid.

Regards,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

GKeyes6988

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
Yuri Kuchinsky wrote:
<snip>

> So you really think all of these are "seeming cognates", ie derived from
> a single original form? Like to make a guess at what that proto-form
> might be?

(Yuri)
.I have no idea. The logical place to look for a .proto-form would be NE
.Asia.


<snip>

If you are claiming this is an Austronesian word, that's going to be mighty
difficult, as there are no Austronesian languages in NE Asia. How about Taiwan
or the Phillipines? Shall I look there?

But congratulations on finally grasping -- however vaguely -- the concept of
"Oceanic", after being corrected on this point after (only) five or six
extremly confident and extremely wrong posts. Kind of cute how you try to
imply that this was a consistent typo or something on your part, rather than
simple ignorance of even the most basic facts concerning the Austronesian
languages.

-- Greg Keyes

Miguel Carrasquer Vidal

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
On 16 Feb 1999 19:53:58 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Sun, 14 Feb 1999 21:16:27 GMT:
>: On 14 Feb 1999 19:58:23 GMT, yu...@globalserve.net (Yuri
>: Kuchinsky) wrote:
>

>: >Oceanic:


>: >
>: >gi gi, Salayer, Baju
>: >ngisi, Menado
>: >nisi-nen, Massaratty
>

>: Selayar and Menado are on Sulawesi, you moron. These are not
>: Oceanic languages.
>

>Thank you Miguel for your kind correction. I should have said
>"Austronesian".

Non-Oceanic Austronesian.

Miguel Carrasquer Vidal

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
On 16 Feb 1999 20:53:30 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Ross Clark (d...@antnov1.auckland.ac.nz) wrote on Mon, 15 Feb 1999 17:01:28 +1200:

>> So you really think all of these are "seeming cognates", ie derived from
>> a single original form? Like to make a guess at what that proto-form
>> might be?
>
>I have no idea. The logical place to look for a proto-form would be NE
>Asia.

Precious!

GKeyes6988

unread,
Feb 16, 1999, 3:00:00 AM2/16/99
to
>
Miguel Carrasquer Vidal wrote:

>On 16 Feb 1999 20:53:30 GMT, yu...@globalserve.net (Yuri
>Kuchinsky) wrote:
>
>>Ross Clark (d...@antnov1.auckland.ac.nz) wrote on Mon, 15 Feb 1999 17:01:28
>+1200:
>>> So you really think all of these are "seeming cognates", ie derived from
>>> a single original form? Like to make a guess at what that proto-form
>>> might be?
>>
>>I have no idea. The logical place to look for a proto-form would be NE
>>Asia.
>
>Precious!

Yes, I sort of passed over the fact that Yuri has no idea what Ross means,
here. I thought it sort of spoke for itself.

-- Greg Keyes

Ross Clark

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to
Yuri Kuchinsky wrote:
>
> Ross Clark (d...@antnov1.auckland.ac.nz) wrote on Mon, 15 Feb 1999 17:01:28 +1200:
> : Yuri Kuchinsky wrote:
>
> > > MOON
> > >
> > > mahina, Polynesian
> > >
> > > Ross wrote on 1999/02/05:
> > >
> > > > > Polynesian "mahina' (moon) may resemble
> > > > > something Hill-Tout writes as "ma-hin" in some interior Salish
> > > > > language.
> > >
> > > This is a fine example of failing to get the point. First of all, there
> > > are also some seeming cognates with other NW Coast languages:
> > >
> > > ma-qha-ten = moon, Ntlakap.
> >
> > CV resemblance
> >
> > > mehme-khtl, = moon, Bilq.
> >
> > C resemblance
> >
> > > ma-ma = light (widely shared)
> >
> > CV resemblance, doesn't mean "moon", what does "widely shared"
> > mean? I see two neighbouring Interior Salish languages with this word,
>
> Nope. Bilqula is not interior. There are three NW Coast languages with
> this word. How many do you want?
>

But the Bilqula (Nuxalk) form is *not* "ma ma". HT gives it as
"mehme-khtl"; my modern Nuxalk vocabulary shows it as /micmik'lh/.

> > So you really think all of these are "seeming cognates", ie derived from
> > a single original form? Like to make a guess at what that proto-form
> > might be?
>
> I have no idea. The logical place to look for a proto-form would be NE
> Asia.

Add this to the file of evidence that Yuri doesn't know what a
"proto-form' is.

>
> > > And Hill-Tout actually wrote in his paper (that Ross claims to have read
> > > carefully -- yeah, right!) that some of the best parallels with
> > > Austronesian actually exist in interior Salish -- as well as in Bilqula, a
> > > coastal Salish language. How could he have missed it?
> >
> > I didn't miss it. He said it. I did not say he didn't say it. What the
> > hell are you on about? Is it that you simply have such an uncontrollable
> > itch to vilify other people that you must accuse them of duplicity every
> > few lines even when there's no basis for it?
>
> No, I'm just commenting on your apparent inability to perceive the
> significance of these pretty clear similarities.

Your "comments", as so often, consist of imputations of dishonesty and
stupidity.

If you think the Lillooet borrowed the word (or the syllable) from
Nootka, why don't you say so?


>
> > > Aei-te-men, Okana.
> >
> > Could be related to the previous item since this is another
> > interior language; all resemblance to "gigi" has vanished at this
> > point
>
> That's because you're refusing to go to the cited source and to find many
> more intermediate forms.

By "cited sources" you mean Hill-Tout? You lie. I have it here before
me. If there are further sources you know about, why not tell us? Your
impudence is amazing. It is people like me and Greg who are seeking out
additional (and better) sources to check up on Hill-Tout's data. You sit
there with the sacred ancient text before you and accuse us of refusing
to go to the sources!

I am looking at Hill-Tout and there are no "intermediate" forms between
the two above and the rest. Their only connection is having a single
syllable in common with the Nootka form.


You're still stonewalling.
>
> The general point is pretty clear. This Austronesian word is also rather
> widely shared by NW Coast languages. I don't see why Ross would try to
> deny the obvious.

HILL-TOUT's COMPLETE LIST OF NW COAST WORDS FOR "TOOTH"

WAKASHAN

Gigi, kyiky (in syn. -hsia) Kwakiutl
Tshi-tshi-sh, Tchi-tchi-tchi Nootka

SALISHAN

Gi-geis Thatl.
Yenas Snan.
Yinis Pent.
Yinis Skquam.
Tsenes Song.
Yelis Sumas etc.
Dzudis Nisk.
Obsin Tsehalis
Rei-tshi-min Lill.
Aei-te-men Okana.
Helah Shew.
Hioh, pl. hi-hi-oh Ntlak.


Now we can all look at these and decide which ones are "this
Austronesian word".


>
> > > > FIRE (general sense)
> > > >
> > > > In Polynesia:
> > > >
> > > > 1. Rotto, hai
> >
> > Probably means Roti, which has /ha?i/
> >
> > > > 2. Mysol (coast), lap
> >
> > Don't find this language listed anywhere
>
> This is an island near New Guinea. Other islands in the area are Aru,
> Waigiou, and Jobie.

A long way from Polynesia.

>
> > > > On NW Coast:
> > > >
> > > > 1. Kwakiutl, hai
> >
> > Not in my Kwakiutl dictionary. Could be intended for initial
> > /xi-/ in various words to do with fire & burning
> >
> > > > Snan., haiuk
> >
> > I assume this is the island dialect of Halkomelem. Could be
> > something like Cowichan /hay?qw/.
>
> Snanmth village of the Sto:lo people? Salish (Halq'emiylem).
>
> > > >
> > > > 2. Ntlakap, lap, or aap [sunset, evening]
> > >
> > > Again, this is evidence that NW Coast languages are connected very closely
> > > to some Austronesian non-Polynesian languages. This supports my hypothesis
> > > that some NW Coast languages are Austronesian.
> >
> > More feeble scraps. CV resemblance between Roti and Snan, maybe CVV if
> > we're being generous. Could be CVC between "Mysol" and Ntlakap., except
> > that you switched meanings.
>
> Again, perhaps it would help to consult Hill-Tout's paper for more
> intermediate forms?

Intermediate between what and what?

>
> > > NEGATIVES
>
> > > Polynesian negatives
> > > --------------------
> > >
> > > 1. te, Maori
> > /te:/
> >
> > > tai, Tongan
> >
> > /ta'e/
>
> Thanks for providing modern transcriptions.
>
> > > 2. aua, Maori
> > (and dozens of other forms reflecting PPN *kaua or *'aua,
> > negative imperative)
> > >
> > > 3. i, according to Fornander, this is the basic Polynesian form. Many more
> > > examples are given by Hill-Tout.
> >
> > Well, that's what he gets for taking Fornander as a linguistic authority.
> > And in fact he doesn't give any examples at all of Polynesian languages
> > with /i/ as a negative, because there aren't any. He gives a few negative
> > words with /i/ in them...and a few with /a/ in them, and /u/ in
> > them...there are only five vowels after all.
>
> Somehow I doubt that Fornander failed to be informed about such a basic
> matter. I don't see why I should take your word over Fornander's.

Modesty forbids me to suggest why you should, Yuri. But doesn't it tell
you something that Hill-Tout does not cite a single such form? What
happened? Did Fornander keep them in a locked vault or something?

No, there is a simpler explanation. Fornander believed that somehow
beneath the actual diversity of negative forms in Polynesian there must
be some simple (probably short), fundamental form from which they were
all derived. For some reason he thought this was "i". But there's no
language that actually has a negative of this form, nor does comparative
evidence suggest that there ever was. On straightforward matters
relating to the Hawaiian language, I would give Fornander the weight of
an authority. But his theorizing about "fundamental forms" is a
different matter.

>
> > > NW Coast negatives
> > > ------------------
> > >
> > > 1. taa, Shew.
> > > taa, Kull.
> >
> > I have no way of checking these.
>
> Again, you can consult Hill-Tout for more.

That's not the point. I want to check whether the forms he gives for
these particular languages are valid.

>
> > > 2. aua, Songes
> > > whaa, Nisk.
> >
> > The Coast Salish languages I have material on have a negative
> > verb something like /?aw@/ or /?@w@/
>
> Thanks for providing modern transcriptions.
>
> > > 3. i, according to Hall, is basic Kwakiutl negative.
> >
> > My guess is that "the Rev.Mr.Hall, for many years missionary among the
> > Kwakiutl" would be a linguistic authority of the same style and same
> > degree of reliability as Fornander.
>
> And I suspect your dismissal of Hall is even less valid than that of
> Fornander.
>
> > My sources suggest that /k'i/ is the
> > basic negative root in Kwakw'ala.
>
> I don't know why I should take your word over Hall's. After all, he lived
> and worked in the area for a long time.

See remarks above about Fornander. And it isn't my word, it's what can
be found in modern descriptions of the language.

>
> > Scoring the negatives:
> >
> > 1. Remains to be checked
> > 2. Better than average, could be CVCV
> > 3. Nil
>
> Your objectivity level is nil, correct -- as was already known. The
> evidence I've presented seems valid.

Of course.

Ross Clark

Peter T. Daniels

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to
Ross Clark wrote:

> > > So you really think all of these are "seeming cognates", ie derived from
> > > a single original form? Like to make a guess at what that proto-form
> > > might be?
> >
> > I have no idea. The logical place to look for a proto-form would be NE
> > Asia.
>
> Add this to the file of evidence that Yuri doesn't know what a
> "proto-form' is.

What Ross is too polite, or too reticent, to mention is that you don't
"look for" proto-forms, you *reconstruct* them.
--
Peter T. Daniels gram...@worldnet.att.net

Yuri Kuchinsky

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to
Ross Clark (benl...@ihug.co.nz) wrote on Wed, 17 Feb 1999 11:42:55 +1300:
: Yuri Kuchinsky wrote:

[Ross:]
: > > ... all resemblance to "gigi" has vanished at this
: > > point

It didn't, of course.

: > That's because you're refusing to go to the cited source and to find many
: > more intermediate forms.

: By "cited sources" you mean Hill-Tout?

Correct.

: You lie.

??? Really?

: I have it here before me.

: I am looking at Hill-Tout and there are no "intermediate" forms between


: the two above and the rest.

??? Surrealism all over again...

: Their only connection is having a single


: syllable in common with the Nootka form.

Aha! Some connection, after all?

[Yuri:]
: > The general point is pretty clear. This Austronesian word is also rather


: > widely shared by NW Coast languages. I don't see why Ross would try to
: > deny the obvious.

: HILL-TOUT's COMPLETE LIST OF NW COAST WORDS FOR "TOOTH"

: WAKASHAN

: Gigi, kyiky (in syn. -hsia) Kwakiutl
: Tshi-tshi-sh, Tchi-tchi-tchi Nootka

: SALISHAN

: Gi-geis Thatl.
: Yenas Snan.
: Yinis Pent.
: Yinis Skquam.
: Tsenes Song.
: Yelis Sumas etc.
: Dzudis Nisk.
: Obsin Tsehalis
: Rei-tshi-min Lill.
: Aei-te-men Okana.
: Helah Shew.
: Hioh, pl. hi-hi-oh Ntlak.

: Now we can all look at these and decide which ones are "this
: Austronesian word".

I think one could say pretty safely e.g. that hi-hi-oh for "tooth" is
cognate with gi-gi? Why would anyone try to deny the obvious?

Thanks for beating me to posting the evidence that refutes you, and proves
my point, Ross. Your original statement that "all resemblance to "gigi"
has vanished at this point" was obviously wrong.

That's why I like arguing with you -- since you always manage to prove
yourself wrong in the end, and end up arguing _against yourself_...

Regards,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku

"Contrariwise," continued Tweedledee, "if it was so, it might be, and
if it were so, it would be; but as it isn't, it ain't. That's logic!"
-- Lewis Carroll, "Through the Looking Glass"

Yuri Kuchinsky

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to
GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
(EST) in article <19990215214842...@ng37.aol.com>...

> But no, lets review. Hypothesis 1:
>
> A.Very early Austronesians came to the NW
> B. After millenia, some went on to Hawai'i and became Polynesians.
>
> What this predicts (Yuri failed to catch the real implications) is
>
> (1.) the POLYNESIAN languages should resemble NW coast languages more
> than they resemble other Astronesian languages and
>
> (2)that POLYNESIAN languages should be more like Austronesian
> languages from SE Asia (*) than they are like their nearest
> geographical neighbors in Micronesia and Melanesia (speakers of
> Oceanic languages). This would be because, say, Woleaian and Hawa'ain
> would have been seperated from proto-Austronesian times (since
> Polynesians were in Vancouver and not Near Oceania) and would not
> share the same innovations.
>
> *(Yuri says Japan but there aren't any Austronesian languages there -
> yah, I know the Austro-Tai /Japanese possibility, but that changes the
> time depth a bit, among other things)

So far so good.

> Hypothesis # 2
>
> Austronesian speakers moved west from SE Asia.
>
> This predicts:
>
> A.Polynesian is most closely related to the Oceanic languages, less
> closely related to the central Malayo-Polynesian languages, and so on.

This oversimplifies of course. The Pacific has been a mixing bowl of
cultures for millennia, so no such simple deliniation can be expected. The
real relationships are far more complex, I'm sure.

> B. It's not related to NW coast languages at all, and no systematic
> comparison of vocabulary will reveal a regular set of cognates between
> them.
>
> (I''ll also throw in another fact that's been brought up and which
> Yuri ignores -- Kwakiutl and Salishan, for instance, have no known
> genetic afiiliation.

This whole area is quite disputed (see Sapir's Mosan hypothesis). There're
many far-ranging similarities.

> Using "NW coast languages" in a search for cognates with Hawai'ian is
> like using a combined vocabulary of English, Chinese, and Choctaw to
> find cognates with !Kung. Comparing all NW coast languages to all
> Austronesian ones as a means of turning up "cognates" is like using
> Dakota, Japanese, and Malagasy to compare with any and all of the
> Indo-European languages.)

Nonsense.

But maybe not? In any case, this is quite meaningless, even if true.

> never mind the other NW coast forms given.

What about them?

> Move into Oceania, and the resemblance vanishes.

Not quite.

> Oddly, at the other end of the Austronesian world, we find a cognate
> pretty close to the Oceanic form.

Your efforts ended up in a dead end, it seems.

> Maybe the reconstructed form would be GIGI, I don't know. If so, I
> have a suggestion at the end of this section.
>
> How does this bode for Yuri hypothesis? Well, we have 0 resemblance
> between the NW coast and Polynesian --

Questionable. Hill-Tout gives some possible cognates. But it makes no
sense to go into this, since this would be disputed (he uses the meaning
"to laugh, show teeth"). Biased critics like you will not be satisfied
with anything less than obvious cognates, as I well know.

I'm quite satisfied that NW Coast language area shows clear cognates with
non-Polynesian Austronesian. Good enough for me at this point...

> very bad, because that's where the closest relationship ought to be,
> given that Polynesians were supposed to have come from there.

Your test is inconclusive. Big deal.

> Between the CMP GIGI forms and the NW coast we have (at best) forms
> like Ngipan, which if we really stretch have a CV resemblance. That's
> bad too, because NW coast languages ought to be something like the
> near Asian ones --

Questionable assumption.

> and more to the point (since NW coast language have no systematic
> resmeblance to ANY Austronesian language)

Question-begging.

> Polynesian also ought to be closer to those than it is to Oceanic
> languages like Raga. Instead we see the Polynesian terms are most
> like Micronesian and Melanesian ones and less like those elsewhere in
> the Austronesian area (except in Madagascar).

I've been through this already with Ross. Your assumption is that
Polynesian-related languages of Melanesia have been there ever since. But
the reality is more likely to be that they are relatively recent in that
area.

Try to realize that this is the crucial point in this whole discussion.
You simply cannot expect me to grant you the validity of such an
assumption. This is neither realistic nor logical.

> Here's an interesting test, Yuri. Since Kwakw''ala GIGI is identical
> with Balinese GIGI, it's clear that both languages have conserved your
> putative proto-Polynesian/Kwakiutl /G/ as the other Austronesian
> languages have not. So you ought to be able to find scads of
> Balinese/Kwakiutl "G" cognates.

Never mind...

> All thing considered, this word is a good test of hypothesis #2.

Baloney. You got yourself completely tangled up with these pontifications,
that's all...

...

> >> FIRE (general sense)
> >>
> >> In Polynesia:
> >>
> >> 1. Rotto, hai
> >> 2. Mysol (coast), lap
>
> This language is not included in the eighty-language list in the
> COMPARATIVE AUSTRONESIAN DICTIONARY .

I already explained where it is.

> Nor does the form LAP appear in any of those which do appear there.

If you consult Hill-Tout, you will find quite a few cognates, such as

apoi, Silong
api, Solor

and others.

> > On NW Coast:
> >
> > 1. Kwakiutl, hai
> > Snan., haiuk
>
> According to my dictionary, the Kwakw'ala word for Fire is XIKELA.

And H-T transcribes this as hai-kala. So?

> XI is a suffix for words to do with fire, but is not free-standing.
>
> > 2. Ntlakap, lap, or aap [sunset, evening]
>
> Which is hardly a word for fire.

I see that you haven't read H-T yet. Why not?

He explains that _in both areas_ the words for "fire" and "sunset" are
commonly related. Try to clue yourself in about such things, Greg...

> If you want to open the semantic realm like that, you need a set of
> regular sound correspondances to make it plausible. And what lexicon
> is this from?

Wakie-wakie, Greg! H-T already did that.

> .Again, this is evidence that NW Coast languages are connected very closely
> .to some Austronesian non-Polynesian languages. This supports my hypothesis
> .that some NW Coast languages are Austronesian.
>
> And, again Yuri fails his logic test, not understanding the
> implications of his own hypothesis. See above for more specfic
> comments

You're very confused.

About half? Because Xi = hai (as Ross already suggested)!

This should be a very good object lesson for you, Greg. Now you see how
you jumped into this whole debate without adequate preparation. There's a
reason why these relationships are unknown at this time. That's because
these languages are very difficult to analyse. The phonetics are very
complex.

> Of the eighty terms, only one had the shape HA?I, or anything close.

Get yourself a clue, Greg.

> Bear in mind that ? Is a consonant. And as for Lap -- it doesn't mean
> "fire", and if it does exist in Austronesian, its form isn't very
> common.

You're too confused for words. You simply misunderstood the
transliteration used by H-T...

You've flunked this one, Greg, since you've completely failed to grasp, or
to do justice to these comparisons, I'm afraid.

Yours,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

Miguel Carrasquer Vidal

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to
On 17 Feb 1999 17:11:04 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>I've been through this already with Ross. Your assumption is that
>Polynesian-related languages of Melanesia have been there ever since. But
>the reality is more likely to be that they are relatively recent in that
>area.
>
>Try to realize that this is the crucial point in this whole discussion.

Too bad you've misunderstood it then.

Here's a list of terms for you to learn:

AUSTRONESIAN: a language family spoken from Madagascar to Easter
Island and from Taiwan to New Zealand.

MALAYO-POLYNESIAN: a synonym for "Austronesian". Sometimes used
to denote the subgroup of Austronesian which consists of the
whole of Austronesian minus the Austronesian aboriginal languages
of Taiwan.

OCEANIC: a major subgroup of Austronesian (Malayo-Polynesian),
comprising most of the Austronesian languages spoken in New
Guinea/Melanesia, Micronesia and Polynesia (but the Austronesian
languages around Geelvink Bay, Irian Jaya, are not Oceanic).

POLYNESIAN: a subgroup of Oceanic, spoken roughly in the area
between Tonga, Hawaii, Easter Island and New Zealand.

POLYNESIAN OUTLIERS: Polynesian languages spoken in some parts of
what is geographically Micronesia and Melanesia, being relatively
recent arrivals to the area from Tuvalu in Polynesia.

Anthony West

unread,
Feb 17, 1999, 3:00:00 AM2/17/99
to

Ross Clark wrote in message <36CB2F...@ihug.co.nz>...

>Yuri Kuchinsky wrote:
>>
>> Ross Clark (benl...@ihug.co.nz) wrote on Wed, 17 Feb 1999 11:42:55
+1300:
>> : Yuri Kuchinsky wrote:
>>
>> [Ross:]
>> : > > ... all resemblance to "gigi" has vanished at this
>> : > > point
>>

>> It didn't, of course.
>
[snip]

>>
>> I think one could say pretty safely e.g. that hi-hi-oh for "tooth" is
>> cognate with gi-gi? Why would anyone try to deny the obvious?
>
>Add "cognate" to the list of terms Yuri doesn't understand.
>
>Let's see now: What "Gigi" and "hi-hi-oh" have in common is...two
>successive identical syllables with /i/. We note that /hihioh/ is the
>plural of /hioh/, evidently formed by reduplication. So in order to make
>these cognate we are going to have to compare what (as far as we know)
>is a single root morpheme in Kwakiutl with a reduplicated initial
>syllable in Ntlak., ignoring the /oh/ part. This is not impossible, and
>might even be justifiable if we knew more about these languages; but it
>is a little less than "obvious". Next we have to look at the "g" and
>"h". Although Hill-Tout's "h" could mean any number of things, the two
>are phonetically in the right neighbourhood that they could both be
>derived from some velar/uvular original. Now all we need is a couple of
>additional words with this kind of resemblance showing Kwakiutl /g/:
>Ntlak. /h/, and we'll have the beginnings of a case...
>
This destroys Yuri's case in the fourteenth separate way.

Yuri's source Hill-Tout did not even understand, or report accurately, that
the true form of the Kwak. word is "hioh," with reduplication. /hioh/ looks
much less like /gigi/.

In order to demonstrate a possible relationship, as Ross says, Hill-Tout and
Yuri would have to produce a systematic series of words in which /g/ in the
one language equals /h/ in the other. You can't do this with one word. But
one word is all Yuri has.

Absent a system of g>h correspondences, Yuri is stuck with the fact that if
/hihi/ "parallels" /gigi/, it also parallels /haha/. Need I point out it
also parallels /pipi/ ?

That's why linguistics threw out Hill-Tout and all work of that ilk. None of
it has any standing in the profession any more.

Tony West
Philadelphia aaw...@critpath.org

>
>>
>> Thanks for beating me to posting the evidence that refutes you, and
proves

>> my point, Ross. Your original statement that "all resemblance to "gigi"
>> has vanished at this point" was obviously wrong.
>
>No, you just didn't understand it. See above. If we throw in "hi-hi-oh",
>we have four out of fourteen languages with forms which might just be
>cognate with /gigi/. So the rest are just to add weight?
>
>Ross Clark

Ross Clark

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Yuri Kuchinsky wrote:
>
> Ross Clark (benl...@ihug.co.nz) wrote on Wed, 17 Feb 1999 11:42:55 +1300:
> : Yuri Kuchinsky wrote:
>
> [Ross:]
> : > > ... all resemblance to "gigi" has vanished at this
> : > > point
>
> It didn't, of course.

I guess it's your Special Glasses, is it, Yuri? When you look through
them at "Gigi" and "Aei-te-men", you see similarity, do you? For those
of us who don't have this gift, can you try to paint a word-picture of
just what this similarity is?

>
> : > That's because you're refusing to go to the cited source and to find many


> : > more intermediate forms.
>
> : By "cited sources" you mean Hill-Tout?
>

> Correct.
>
> : You lie.
>
> ??? Really?

Yes.

>
> : I have it here before me.

>
> : I am looking at Hill-Tout and there are no "intermediate" forms between


> : the two above and the rest.
>

> ??? Surrealism all over again...
>

> : Their only connection is having a single


> : syllable in common with the Nootka form.
>

> Aha! Some connection, after all?

If one syllable in common is what you mean by an "intermediate form",
then I guess that's what it is. Clearly we are all going to have to
learn Yurian definitions for a lot of familiar terms

>
> [Yuri:]
> : > The general point is pretty clear. This Austronesian word is also rather


> : > widely shared by NW Coast languages. I don't see why Ross would try to
> : > deny the obvious.
>
> : HILL-TOUT's COMPLETE LIST OF NW COAST WORDS FOR "TOOTH"
>
> : WAKASHAN
>
> : Gigi, kyiky (in syn. -hsia) Kwakiutl
> : Tshi-tshi-sh, Tchi-tchi-tchi Nootka
>
> : SALISHAN
>
> : Gi-geis Thatl.
> : Yenas Snan.
> : Yinis Pent.
> : Yinis Skquam.
> : Tsenes Song.
> : Yelis Sumas etc.
> : Dzudis Nisk.
> : Obsin Tsehalis
> : Rei-tshi-min Lill.
> : Aei-te-men Okana.
> : Helah Shew.
> : Hioh, pl. hi-hi-oh Ntlak.
>
> : Now we can all look at these and decide which ones are "this
> : Austronesian word".
>

> I think one could say pretty safely e.g. that hi-hi-oh for "tooth" is
> cognate with gi-gi? Why would anyone try to deny the obvious?

Add "cognate" to the list of terms Yuri doesn't understand.

Let's see now: What "Gigi" and "hi-hi-oh" have in common is...two
successive identical syllables with /i/. We note that /hihioh/ is the
plural of /hioh/, evidently formed by reduplication. So in order to make
these cognate we are going to have to compare what (as far as we know)
is a single root morpheme in Kwakiutl with a reduplicated initial
syllable in Ntlak., ignoring the /oh/ part. This is not impossible, and
might even be justifiable if we knew more about these languages; but it
is a little less than "obvious". Next we have to look at the "g" and
"h". Although Hill-Tout's "h" could mean any number of things, the two
are phonetically in the right neighbourhood that they could both be
derived from some velar/uvular original. Now all we need is a couple of
additional words with this kind of resemblance showing Kwakiutl /g/:
Ntlak. /h/, and we'll have the beginnings of a case...


>

Ross Clark

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
I warned Greg privately that he would be wasting his time presenting
Yuri with comparative Austronesian data. One way or another, Yuri would
manage to reject it, misinterpret it as supporting his theories, or just
pretend it didn't exist. I take only small satisfaction in being proved
right...

I'll comment on just three points.

(1)

Yuri Kuchinsky wrote:
>
> GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
> (EST) in article <19990215214842...@ng37.aol.com>...
>
>
>

> > Polynesian also ought to be closer to those than it is to Oceanic
> > languages like Raga. Instead we see the Polynesian terms are most
> > like Micronesian and Melanesian ones and less like those elsewhere in
> > the Austronesian area (except in Madagascar).

[Yuri:]


>
> I've been through this already with Ross. Your assumption is that
> Polynesian-related languages of Melanesia have been there ever since. But
> the reality is more likely to be that they are relatively recent in that
> area.
>
> Try to realize that this is the crucial point in this whole discussion.
> You simply cannot expect me to grant you the validity of such an
> assumption. This is neither realistic nor logical.

"Been through it", all right. This consisted of Yuri saying (much as he
does here) "I won't believe it, and you can't make me". He babbles about
"Polynesian-related languages of Melanesia", because that was the way
they talked 100 years ago. If it was good enough for Hill-Tout and his
contemporaries, it was good enough for Heyerdahl; and if it was good
enough for Heyerdahl, you know it's good enough for Yuri. Praise the
Lord. In fact, he has no idea what languages he is talking about or what
the evidence indicates about their relations. Along with Thor, he has
taken the Vow of Ignorance, and to that he remains faithful.

(2)

> > > 2. Ntlakap, lap, or aap [sunset, evening]
> >
> > Which is hardly a word for fire.
>
> I see that you haven't read H-T yet. Why not?
>
> He explains that _in both areas_ the words for "fire" and "sunset" are
> commonly related. Try to clue yourself in about such things, Greg...

Yes, Greg, learn your Austronesian from Hill-Tout. That's the scientific
way.

This particular example is one that regularly fools beginners, who look
at, say Maori /ahi/ "fire" and /ahiahi/ "afternoon,evening", and say
"Aha!". (The second word doesn't mean "sunset" of course -- that's a
kind of stepping-stone meaning by which we make the connection.) In fact
the two are from completely different Austronesian originals: *apuy
"fire" and *Rabi(Rabi) "evening". It just happens that in all Oceanic
languages *uy > i and *b merges with *p, and in many *R > 0, which leads
to homophony.

(3)

Hm. Let's see if we can reverse-engineer Yuri's Magic Similarity Glasses
so that "about half" of these 20 items look like "xi" or "hai".

Feature: Number of items: Comment:

Has /x/ in it 0 Not enough
Has /i/ in it 13 Too many
Has /h/ in it 5 Not enough
Has /a/ in it 14 Too many


/i/ and /h/ 5 Not enough
/h/ and /a/ 4 Not enough
/a/ and /i/ 12 Better

I think we might be getting close here. Can anybody get exactly 10?

Ross Clark

GKeyes6988

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to

Yuri Kuchinsky wrote:
>
>
>GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
>(EST) in article <19990215214842...@ng37.aol.com>...

(Greg)


>> But no, lets review. Hypothesis 1:
>>
>> A.Very early Austronesians came to the NW
>> B. After millenia, some went on to Hawai'i and became Polynesians.
>>
>> What this predicts (Yuri failed to catch the real implications) is
>>
>> (1.) the POLYNESIAN languages should resemble NW coast languages more
>> than they resemble other Astronesian languages and
>>
>> (2)that POLYNESIAN languages should be more like Austronesian
>> languages from SE Asia (*) than they are like their nearest
>> geographical neighbors in Micronesia and Melanesia (speakers of
>> Oceanic languages). This would be because, say, Woleaian and Hawa'ain
>> would have been seperated from proto-Austronesian times (since
>> Polynesians were in Vancouver and not Near Oceania) and would not
>> share the same innovations.
>>
>> *(Yuri says Japan but there aren't any Austronesian languages there -
>> yah, I know the Austro-Tai /Japanese possibility, but that changes the
>> time depth a bit, among other things)

(Yuri)
>So far so good.

Okay. Though you seem to have an attack of schizophrenia about this later in
your post.

(Greg)


>> Hypothesis # 2
>>
>> Austronesian speakers moved west from SE Asia.
>>
>> This predicts:
>>
>> A.Polynesian is most closely related to the Oceanic languages, less
>> closely related to the central Malayo-Polynesian languages, and so on.

(Yuri)


>This oversimplifies of course. The Pacific has been a mixing bowl of
>cultures for millennia, so no such simple deliniation can be expected. The
>real relationships are far more complex, I'm sure.

As sure as ignorance of the linguistic facts can make you, yes.

(Greg)


>> B. It's not related to NW coast languages at all, and no systematic
>> comparison of vocabulary will reveal a regular set of cognates between
>> them.
>>
>> (I''ll also throw in another fact that's been brought up and which
>> Yuri ignores -- Kwakiutl and Salishan, for instance, have no known
>> genetic afiiliation.

(Yuri)


>This whole area is quite disputed (see Sapir's Mosan hypothesis). There're
>many far-ranging similarities.

No demonstrable genetic affiliation, and people have been trying for a long
time. Which means if there is one, the time depth is considerable. If
Wakashan and Salishan languages were both originally Austronesian, they should
be as readily related to one another as the Austronesian languages in Asia.
More so, actually, since they would have separated later.

But then, these are your hypothetical, unattested NE Asian Austronesian
languages, aren't they?

(Greg)


>> Using "NW coast languages" in a search for cognates with Hawai'ian is
>> like using a combined vocabulary of English, Chinese, and Choctaw to
>> find cognates with !Kung. Comparing all NW coast languages to all
>> Austronesian ones as a means of turning up "cognates" is like using
>> Dakota, Japanese, and Malagasy to compare with any and all of the
>> Indo-European languages.)

Yuri
>Nonsense.

Well, it must be from your point of view, because its the only way you can come
up with anything. But it remains an unsound method, your "expert" opinion
notwithstanding.

(Greg)

(Yuri)


>But maybe not? In any case, this is quite meaningless, even if true.

Maybe not? You have GIGI from Taiwan or the Phillipines, or from your
non-existent NE Asian languages? Why don't you cite them, then?

(Greg)


>> never mind the other NW coast forms given.

(Yuri)
>What about them?

They are even less like any Austronesian form, but you knew that.

(Greg)


>> Move into Oceania, and the resemblance vanishes.

(Yuri)
>Not quite.

Quite.

(Greg)


>> Oddly, at the other end of the Austronesian world, we find a cognate
>> pretty close to the Oceanic form.

Yuri


>Your efforts ended up in a dead end, it seems.

This is a truly bizarre assertion.

(Greg)


>> Maybe the reconstructed form would be GIGI, I don't know. If so, I
>> have a suggestion at the end of this section.
>>
>> How does this bode for Yuri hypothesis? Well, we have 0 resemblance
>> between the NW coast and Polynesian --

(Yuri)


>Questionable. Hill-Tout gives some possible cognates. But it makes no
>sense to go into this, since this would be disputed (he uses the meaning
>"to laugh, show teeth"). Biased critics like you will not be satisfied
>with anything less than obvious cognates, as I well know.


While you seem to be satsfied with any two words HT or Campbell put together,
while at the same time somehow being unsatisfied with the well-attested
cognates I've posted. I can't really account for this without assuming some
sort of bias on your part.

Yuri


>I'm quite satisfied that NW Coast language area shows clear cognates with
>non-Polynesian Austronesian. Good enough for me at this point...

Oh, of course it's good enough for you. But according to you (and the scenario
you agreed to above) the Polynesian languages ought to be closer to NW coast
languages than to any of the languages in Austronesia.

(Greg)


>> very bad, because that's where the closest relationship ought to be,
>> given that Polynesians were supposed to have come from there.

(Yuri)


>Your test is inconclusive. Big deal.

Ah. But it's a test of your hypothesis. Mustn't test the hypothesis, eh?
Just keep repeating it until it becomes real, like the Velveteen Rabbit?

(Greg)


>> Between the CMP GIGI forms and the NW coast we have (at best) forms
>> like Ngipan, which if we really stretch have a CV resemblance. That's
>> bad too, because NW coast languages ought to be something like the
>> near Asian ones --

(Yuri)
>Questionable assumption.

Care to embarrass yourself by explaining why?

(Greg)


>> and more to the point (since NW coast language have no systematic
>> resmeblance to ANY Austronesian language)

(Yuri)
>Question-begging.

No, this is the heart of the question.

(Greg)


>> Polynesian also ought to be closer to those than it is to Oceanic
>> languages like Raga. Instead we see the Polynesian terms are most
>> like Micronesian and Melanesian ones and less like those elsewhere in
>> the Austronesian area (except in Madagascar).

(Yuri)


>I've been through this already with Ross. Your assumption is that
>Polynesian-related languages of Melanesia have been there ever since. But
>the reality is more likely to be that they are relatively recent in that
>area.

Does this statement make any sense to anyone here? How can the other Oceanic
languages be closely related to Polynesian (more closely than either is to the
other Austronesian languages) if they have been separated since before Yuri's
putative Kwakunesians left Asia for the NW coast?

(Yuri)


>Try to realize that this is the crucial point in this whole discussion.
>You simply cannot expect me to grant you the validity of such an
>assumption. This is neither realistic nor logical.
>

Now the schizophrenia sets in. You granted it at the top of the post. You had
to. If Polynesian was separated from the other Austronesian languages for
millenia, there ought to be a jarring disjunction between them. No way around
it, period. If Polynesian came from the NW coast, it ought to be related to
some NW coast language more closely than it is to, say, Woleaian. Your only
way around that is to start changing your claim -- which I see you have been
doing in the past day or so. Now you say the NW coast languages might not be
Austronesian at all. But that makes us wonder even more why we need a trip to
the NW coast to place speakers of Austronesian in Hawai'i, doesn't it? So your
only choice is to waver bewteen the two positions, posit languages you have no
evidence for (NE Asian ones) and pretend that Polynesian isn't an Oceanic
language -- now that you finally know what an Oceanic language is.

Oh, and your major strategy -- insult people in the hopes that they will go
away.

(Greg)


>> Here's an interesting test, Yuri. Since Kwakw''ala GIGI is identical
>> with Balinese GIGI, it's clear that both languages have conserved your
>> putative proto-Polynesian/Kwakiutl /G/ as the other Austronesian
>> languages have not. So you ought to be able to find scads of
>> Balinese/Kwakiutl "G" cognates.


(Yuri)
>Never mind...

Yes, perish the thought that you should get a dictionary and do something
yourself. I suspect you don't even know what I'm getting at, here. This is a
perfectly serious suggestion. If the languages are related, as you claim, it
ought to bear fruit. And it's so easy! You don't have to catalogue any
sound-shifts or explain how this phoneme became that phoneme -- because there
wasn't a sound shift at all!

(Greg)


>> All thing considered, this word is a good test of hypothesis #2.

(Yuri)


>Baloney. You got yourself completely tangled up with these pontifications,
>that's all...

Really? Maybe you can explain how the very obvious close connection between
Polynesian languages and the other Oceanic ones doesn't bear out hypothesis #
2.

> ...
(Yuri)


>> >> FIRE (general sense)
>> >>
>> >> In Polynesia:
>> >>
>> >> 1. Rotto, hai
>> >> 2. Mysol (coast), lap

Greg

>> This language is not included in the eighty-language list in the
>> COMPARATIVE AUSTRONESIAN DICTIONARY .

Yuri


>I already explained where it is.

Greg


>> Nor does the form LAP appear in any of those which do appear there.
>

Yuri


>If you consult Hill-Tout, you will find quite a few cognates, such as
>apoi, Silong
>api, Solor
>
>and others.

I repeat. No LAP. I gave plenty of cognates like the ones you just posted,
and its clear where they fall out in relation to the Polynesian languages.

(Yuri)


>> > On NW Coast:
>> >
>> > 1. Kwakiutl, hai
>> > Snan., haiuk
>>
>> According to my dictionary, the Kwakw'ala word for Fire is XIKELA.
>
>And H-T transcribes this as hai-kala. So?

So /h/ and /x/ represent different sounds. I understand the whole orthography
thing is very confusing to you, Yuri. You've been showing a lot of that
confusion, lately.

(Greg)
>> XI is a suffix (meant prefix, here) for words to do with fire, but is not
free-standing.

Yuri

>> > 2. Ntlakap, lap, or aap [sunset, evening]

Greg


>> Which is hardly a word for fire.

Yuri.


>I see that you haven't read H-T yet. Why not?
>
>He explains that _in both areas_ the words for "fire" and "sunset" are
>commonly related. Try to clue yourself in about such things, Greg...

But not so commonly that the words are actually the same or similar, I see, or
he would have cited a semantically cleaner cognate.

(Greg)


>> If you want to open the semantic realm like that, you need a set of
>> regular sound correspondances to make it plausible. And what lexicon
>> is this from?
>
>Wakie-wakie, Greg! H-T already did that.

Really? What are these regular sound correspondances, then? You do know what
I mean, right?

(Yuri)


>> .Again, this is evidence that NW Coast languages are connected very closely
>> .to some Austronesian non-Polynesian languages. This supports my hypothesis
>> .that some NW Coast languages are Austronesian.

Even though I see you're backing off of this now, claiming that this "very
close relationship" doesn't actually entail them being Austronesian.

(Greg)


>> And, again Yuri fails his logic test, not understanding the
>> implications of his own hypothesis. See above for more specfic
>> comments

(Yuri)
>You're very confused.

Oh, I'll let other's judge which of us is confused, Yuri.

Greg

What in heaven's name are you talking about? The Kwakiutl is /XI - /, or in
actual fact XIKELA , since the /XI/ doesnlt occure free-standing. What Ross
suggested was that HT didn't represent the sounds very well. You want the "h"
back. Sorry, no can do.

And note that it least resembles, once again, any Polynesian word, though the
Polynesian words look an awful like other Oceanic ones.

(Yuri)


>This should be a very good object lesson for you, Greg. Now you see how
>you jumped into this whole debate without adequate preparation. There's a
>reason why these relationships are unknown at this time. That's because
>these languages are very difficult to analyse. The phonetics are very
>complex.

Oh, yes, so complex that your 'cognates' don't look like cognates whereas mine
do. But if you keep staring and staring at hundred-year old orthographies, and
don't confuse yourself by trying to find out what the words SOUND like, then
eventually it all comes clear.

You are so very cute, Yuri, when you try to be condescending about something
you clearly know so little about.

Greg


>> Of the eighty terms, only one had the shape HA?I, or anything close.

Yuri


>Get yourself a clue, Greg.

Greg


>> Bear in mind that ? Is a consonant. And as for Lap -- it doesn't mean
>> "fire", and if it does exist in Austronesian, its form isn't very
>> common.

Yuri


>You're too confused for words. You simply misunderstood the
>transliteration used by H-T...

Not confused enough to think that Roti /h/ and Kwakw'ala /x/ are the same sound
simply because HT used and "H" in his transliteration. That seems to be your
confusion.

(Yuri)


>You've flunked this one, Greg, since you've completely failed to grasp, or
>to do justice to these comparisons, I'm afraid.

Cuter and cuter!

-- Greg Keyes


Anthony West

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to

GKeyes6988 wrote in message <19990217190530...@ng15.aol.com>...

>
> Yuri Kuchinsky wrote:
>>
>>
>>GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
>>(EST) in article <19990215214842...@ng37.aol.com>...
>
>(Greg)
>>> Hypothesis # 2
>>>
>>> Austronesian speakers moved west from SE Asia.
>>>
>>> This predicts:
>>>
>>> A.Polynesian is most closely related to the Oceanic languages, less
>>> closely related to the central Malayo-Polynesian languages, and so on.
>
>(Yuri)
>>This oversimplifies of course. The Pacific has been a mixing bowl of
>>cultures for millennia, so no such simple deliniation can be expected. The
>>real relationships are far more complex, I'm sure.
>
>As sure as ignorance of the linguistic facts can make you, yes.
>
The charm of Yuri's raving can only be fully appreciated by linguists. The
internal structure of the Austronesian family is one of the largest and best
worked out taxonomies in the world. It involves far greater numbers of
languages than Indo-European, for instance. While any project of that size
always has work that remains to be done, the outlines of its chief internal
divisions have been thoroughly worked out for a long time. They are
available to anyone who actually wants information.

In some ways, the project was helped by the huge number of languages
involved. The cognates literally leap off the page in group after group, as
Greg showed in his detailed demonstration of the Swadesh list -- unlike
Hill-Tout's vague and loony "parallels," which Yuri keeps pawing at and
snuffling over.

Notice again how Yuri commits one of his favorite fatal flaws in reasoning:
he confuses cultural with linguistic evidence. Because "the Pacific has been
a mixing bowl of cultures for millennia ... no such simple deliniation [sic]
can be expected." Translation: Yuri didn't expect it. But who cares?
Linguists don't base taxonomies on "cultural mixing bowl" criteria, but on
linguistic criteria.

Oceania constitutes what Johanna Nichols calls a "spread zone" (cf.
Linguistic Diversity in Space and Time), wherein a striking linguistic unity
reigns over a vast territory, and to all paleolinguistic appearances has so
reigned since deep in the past.

Tony West
Philadelphia aaw...@critpath.org


Anthony West

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to

Yuri Kuchinsky wrote in message <7aet78$s9e$1...@whisper.globalserve.net>...

>GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
>(EST) in article <19990215214842...@ng37.aol.com>...
>
[snip]

>This whole area is quite disputed (see Sapir's Mosan hypothesis). There're
>many far-ranging similarities.
>
Correction: *was* disputed. Mosan has fairly few far-ranging similarities;
that's why, after 70 years, it's still not regarded as sound.

>> Using "NW coast languages" in a search for cognates with Hawai'ian is
>> like using a combined vocabulary of English, Chinese, and Choctaw to
>> find cognates with !Kung. Comparing all NW coast languages to all
>> Austronesian ones as a means of turning up "cognates" is like using
>> Dakota, Japanese, and Malagasy to compare with any and all of the
>> Indo-European languages.)
>
>Nonsense.
>

No, that is precisely what Yuri is doing. Because he is fundamentally
ignorant of the principles of linguistic classification, he thinks that all
NW Coast languages are somehow related, so that you can compare them as a
body to Hawai'ian, or to Austronesian languages as a body. But in fact, they
fall into several families which are as unrelated as English, Chinese and
Choctaw are -- or as Dakota, Japanese and Malagasy are. Yuri is smart enough
to realize that those two groupings are "nonsense." But he had never
actually thought much about NW Coast linguistics until recently, so he
simply took it for granted that because they form a geographical group, they
also form a genetic group.

But his "NW Coast" group is as nonsensical as either of the ones Greg Keyes
just made up for us.

Tony West
Philadelphia aaw...@critpath.org


Anthony West

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to

GKeyes6988 wrote in message <19990217190530...@ng15.aol.com>...
>
> Yuri Kuchinsky wrote:
>>
>>
>>GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500
>>(EST) in article <19990215214842...@ng37.aol.com>...
>
[snip. Greg provides a thorough and methodical search for Austronesian forms
like GIGI "tooth," which the dotty Hill-Tout was convinced had been
implanted into some Salishan language --

and then Yuri watches his case go up in smoke. Again]

>>> In the list of eighty terms I checked, nine had CVCV agreement with
>>> GIGI. All were in Indonesia, in the center of the western MP region.
>>> Move NWcoastward -- Borneo, the Phillipines, Taiwan -- and you get
>>> forms that less and less resemble GIGI,
>>
>
>(Yuri)
>>But maybe not? In any case, this is quite meaningless, even if true.
>
>Maybe not? You have GIGI from Taiwan or the Phillipines, or from your
>non-existent NE Asian languages? Why don't you cite them, then?
>

[snip, to cut to the chase a few bars later]


>
>Yuri
>>I'm quite satisfied that NW Coast language area shows clear cognates with
>>non-Polynesian Austronesian. Good enough for me at this point...
>
>Oh, of course it's good enough for you. But according to you (and the
scenario
>you agreed to above) the Polynesian languages ought to be closer to NW
coast
>languages than to any of the languages in Austronesia.
>
>(Greg)
>>> very bad, because that's where the closest relationship ought to be,
>>> given that Polynesians were supposed to have come from there.
>

Here Yuri illustrates for us another peril faced by amateur linguists:
over-reliance on broad geographical comparisons when you're actually trying
to explore a narrow one.

Yuri was trying to show that Polynesians came from British Columbia, via
Hawai'i. For this to work, he needs Salishan words like GIGI to appear in
Hawai'ian, or at least some other Polynesian tongue. But they don't. So he
leans on Hill-Tout's GIGI, on the assumption that all non-Western languages
look alike and amount to the same thing.

Alas, the Austronesian GIGI-like forms are pretty much confined to Western
Malayo-Polynesian, which is located in Indonesia.

Yuri was trying to ferry his NW Coast Indians, on the good ship Gigi, from
Vancouver to Oahu. Unfortunately, the boat took him to Java instead.

Oops!

Tony West
Philadelphia aaw...@critpath.org


Anthony West

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to

GKeyes6988 wrote in message <19990217190530...@ng15.aol.com>...
>
> Yuri Kuchinsky wrote:
[snip]

>>> Notice how many of these look like Kwakiutl XI?
>
>>About half? Because Xi = hai (as Ross already suggested)!
>
>What in heaven's name are you talking about? The Kwakiutl is /XI - /, or
in
>actual fact XIKELA , since the /XI/ doesnlt occure free-standing. What
Ross
>suggested was that HT didn't represent the sounds very well. You want the
"h"
>back. Sorry, no can do.
>
> And note that it least resembles, once again, any Polynesian word, though
the
>Polynesian words look an awful like other Oceanic ones.
>
>(Yuri)
>>This should be a very good object lesson for you, Greg. Now you see how
>>you jumped into this whole debate without adequate preparation. There's a
>>reason why these relationships are unknown at this time. That's because
>>these languages are very difficult to analyse. The phonetics are very
>>complex.
>
They are so difficult and so complex, Yuri explains, that only a complete
ignoramus can analyze them. Someone who has never studied any Austronesian
language, any American Indian language, or linguistics in general. And who
refuses to study them, moreover. Where, oh where can we find such a savior?
One who has, how should we put it -- "adequate preparation"?

I've got it! Let's call Yuri! Can I have Directory Assistance in Toronto,
please?

>Oh, yes, so complex that your 'cognates' don't look like cognates whereas
mine
>do. But if you keep staring and staring at hundred-year old orthographies,
and
>don't confuse yourself by trying to find out what the words SOUND like,
then
>eventually it all comes clear.
>
>You are so very cute, Yuri, when you try to be condescending about
something
>you clearly know so little about.
>
>Greg

>>> Bear in mind that ? Is a consonant. And as for Lap -- it doesn't mean
>>> "fire", and if it does exist in Austronesian, its form isn't very
>>> common.
>
>Yuri
>>You're too confused for words. You simply misunderstood the
>>transliteration used by H-T...
>
>Not confused enough to think that Roti /h/ and Kwakw'ala /x/ are the same
sound
>simply because HT used and "H" in his transliteration. That seems to be
your
>confusion.
>
>(Yuri)
>>You've flunked this one, Greg, since you've completely failed to grasp, or
>>to do justice to these comparisons, I'm afraid.
>

Yuri is a walking, typing compendium of beginner's errors today.

An excellent starting rule for all: NEVER rely on faulty transcriptions when
better ones exist.

NW Coast languages have in common that they are full of phonemes that do not
occur in most European standard languages and were not well represented by
the Latin alphabet when linguists started to work on them 100 years ago. It
took quite a while before they figured out how to represent these sounds
with any clarity.

Hill-Tout, alas, came before this development. It is apparent from a glance
that his use of "H," for instance, or any other letter, is absolutely
unreliable as a notation of any particular Salishan or Wakashan sound. (One
should also note his *omissions* of letters: often he simply ignores a sound
in an Indian word because he can't think of any way to spell it.)

Scientists have a technical term for this. It's "bad data." The only
solutions for bad data are either to pitch it out, or to recheck every
single item in it against more correctly recorded vocabularies of the same
languages.

Fortunately, linguists no longer need to pore over Hill-Tout's timeworn
wordlists. The real job of checking out NW Coast languages for possible
Austronesian relationships was done much later than 1898, using tools and
methods that are currently regarded as valid, with access to vastly larger
and finer resources of information than Hill-Tout had at his disposal
(basically, the poor fellow, like Yuri, had only a one-book source for all
of Austronesian -- and no contact with any linguists who actually knew these
languages).

The job was done. And the results were negative.

So it is very easy to do justice to Hill-Tout's comparisons today. Pitch
them.

Anybody who would keep them around for play, though, must first figure out
how to transcribe accurately every word that Hill-Tout cited.

Re-ga-rds,

Tony West
Philadelphia aaw...@critpath.org


Yuri Kuchinsky

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Thanks, Miguel, but I already knew all this. I've made a mistake when
arguing with Greg when I labelled some languages Oceanic (I wasn't even
sure at that point about the exact geographical locations of some of these
obscure languages), whereas I should have said Austronesian. This was a
minor misstatement, provoked by Greg's general confusion about this whole
discussion. Now, of course he and other bumblers will harp about this
until Kingdom come. Personal attacks is the only weapon of a looser?

Yours,

Yuri.

Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Wed, 17 Feb 1999 17:42:00 GMT:
: On 17 Feb 1999 17:11:04 GMT, yu...@globalserve.net (Yuri
: Kuchinsky) wrote:

: >I've been through this already with Ross. Your assumption is that


: >Polynesian-related languages of Melanesia have been there ever since. But
: >the reality is more likely to be that they are relatively recent in that
: >area.
: >
: >Try to realize that this is the crucial point in this whole discussion.

: Too bad you've misunderstood it then.

: Here's a list of terms for you to learn:

: AUSTRONESIAN: a language family spoken from Madagascar to Easter
: Island and from Taiwan to New Zealand.

: MALAYO-POLYNESIAN: a synonym for "Austronesian". Sometimes used
: to denote the subgroup of Austronesian which consists of the
: whole of Austronesian minus the Austronesian aboriginal languages
: of Taiwan.

: OCEANIC: a major subgroup of Austronesian (Malayo-Polynesian),
: comprising most of the Austronesian languages spoken in New
: Guinea/Melanesia, Micronesia and Polynesia (but the Austronesian
: languages around Geelvink Bay, Irian Jaya, are not Oceanic).

: POLYNESIAN: a subgroup of Oceanic, spoken roughly in the area
: between Tonga, Hawaii, Easter Island and New Zealand.

: POLYNESIAN OUTLIERS: Polynesian languages spoken in some parts of
: what is geographically Micronesia and Melanesia, being relatively
: recent arrivals to the area from Tuvalu in Polynesia.


: =======================
: Miguel Carrasquer Vidal
: m...@wxs.nl
: Amsterdam

--

Yuri Kuchinsky

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 10:09:02 +1300:
: Yuri Kuchinsky wrote:

[Ross:]
: > : HILL-TOUT's COMPLETE LIST OF NW COAST WORDS FOR "TOOTH"


: >
: > : WAKASHAN
: >
: > : Gigi, kyiky (in syn. -hsia) Kwakiutl
: > : Tshi-tshi-sh, Tchi-tchi-tchi Nootka
: >
: > : SALISHAN
: >
: > : Gi-geis Thatl.
: > : Yenas Snan.
: > : Yinis Pent.
: > : Yinis Skquam.
: > : Tsenes Song.
: > : Yelis Sumas etc.
: > : Dzudis Nisk.
: > : Obsin Tsehalis
: > : Rei-tshi-min Lill.
: > : Aei-te-men Okana.
: > : Helah Shew.
: > : Hioh, pl. hi-hi-oh Ntlak.
: >
: > : Now we can all look at these and decide which ones are "this
: > : Austronesian word".

: >
: > I think one could say pretty safely e.g. that hi-hi-oh for "tooth" is


: > cognate with gi-gi? Why would anyone try to deny the obvious?

: Add "cognate" to the list of terms Yuri doesn't understand.

You do blather a lot of late, don't you, Ross? For a hypocrite like you
this sure would indicate that you're up the creek without a paddle? Taking
some lessons from Tony the Brain-Dead?

: Let's see now: What "Gigi" and "hi-hi-oh" have in common is...two


: successive identical syllables with /i/. We note that /hihioh/ is the
: plural of /hioh/, evidently formed by reduplication. So in order to make
: these cognate we are going to have to compare what (as far as we know)
: is a single root morpheme in Kwakiutl with a reduplicated initial
: syllable in Ntlak., ignoring the /oh/ part. This is not impossible, and
: might even be justifiable if we knew more about these languages; but it
: is a little less than "obvious". Next we have to look at the "g" and
: "h". Although Hill-Tout's "h" could mean any number of things, the two
: are phonetically in the right neighbourhood that they could both be
: derived from some velar/uvular original. Now all we need is a couple of
: additional words with this kind of resemblance showing Kwakiutl /g/:
: Ntlak. /h/, and we'll have the beginnings of a case...

Very fine. Now, don't try to claim that Your Opinions on this are the only
one possible.

: > Thanks for beating me to posting the evidence that refutes you, and proves
: > my point, Ross. Your original statement that "all resemblance to "gigi"
: > has vanished at this point" was obviously wrong.

: No, you just didn't understand it.

Yadda, yadda.

: See above. If we throw in "hi-hi-oh",


: we have four out of fourteen languages with forms which might just be
: cognate with /gigi/. So the rest are just to add weight?

Some of the rest are also arguable. And you were still wrong.

Yuri.

Yuri Kuchinsky -O- http://www.globalserve.net/~yuku -O- Toronto

You never need think you can turn over any old falsehoods without a
terrible squirming of the horrid little population that dwells under
it -=O=- Oliver Wendell Holmes

Miguel Carrasquer Vidal

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
On 18 Feb 1999 16:40:26 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Personal attacks is the only weapon of a looser?

Yes.

Yuri Kuchinsky

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 11:42:44 +1300:

: I warned Greg privately that he would be wasting his time presenting


: Yuri with comparative Austronesian data. One way or another, Yuri would
: manage to reject it, misinterpret it as supporting his theories, or just
: pretend it didn't exist. I take only small satisfaction in being proved
: right...

Yadda, yadda, yadda...

: I'll comment on just three points.

: (1)

: Yuri Kuchinsky wrote:
: >
: > GKeyes6988 <gkeye...@aol.com> wrote on Wed, 17 Feb 1999 01:37:12 -0500


: > (EST) in article <19990215214842...@ng37.aol.com>...

: >
: >
: >
: > > Polynesian also ought to be closer to those than it is to Oceanic


: > > languages like Raga. Instead we see the Polynesian terms are most
: > > like Micronesian and Melanesian ones and less like those elsewhere in
: > > the Austronesian area (except in Madagascar).

: [Yuri:]
: >
: > I've been through this already with Ross. Your assumption is that


: > Polynesian-related languages of Melanesia have been there ever since. But
: > the reality is more likely to be that they are relatively recent in that
: > area.
: >
: > Try to realize that this is the crucial point in this whole discussion.
: > You simply cannot expect me to grant you the validity of such an
: > assumption. This is neither realistic nor logical.

: "Been through it", all right. This consisted of Yuri saying (much as he


: does here) "I won't believe it, and you can't make me". He babbles about
: "Polynesian-related languages of Melanesia", because that was the way
: they talked 100 years ago. If it was good enough for Hill-Tout and his

: contemporaries, it was good enough for Heyerdahl; and if it was good
: enough for Heyerdahl, you know it's good enough for Yuri. Praise the


: Lord. In fact, he has no idea what languages he is talking about or what
: the evidence indicates about their relations. Along with Thor, he has
: taken the Vow of Ignorance, and to that he remains faithful.

Yadda, yadda, yadda...

: (2)

: > > > 2. Ntlakap, lap, or aap [sunset, evening]


: > >
: > > Which is hardly a word for fire.
: >
: > I see that you haven't read H-T yet. Why not?
: >
: > He explains that _in both areas_ the words for "fire" and "sunset" are
: > commonly related. Try to clue yourself in about such things, Greg...

: Yes, Greg, learn your Austronesian from Hill-Tout. That's the scientific
: way.

Yadda, yadda, yadda...

: This particular example is one that regularly fools beginners, who look


: at, say Maori /ahi/ "fire" and /ahiahi/ "afternoon,evening", and say
: "Aha!". (The second word doesn't mean "sunset" of course -- that's a
: kind of stepping-stone meaning by which we make the connection.) In fact
: the two are from completely different Austronesian originals: *apuy
: "fire" and *Rabi(Rabi) "evening". It just happens that in all Oceanic
: languages *uy > i and *b merges with *p, and in many *R > 0, which leads
: to homophony.

Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
are still to explain how and why _on NW Coast too_ the words for "fire"
and "sunset" are _also_ commonly related.

So try again.

: (3)

: >
: > > FIRE

: >

: Hm. Let's see if we can reverse-engineer Yuri's Magic Similarity Glasses


: so that "about half" of these 20 items look like "xi" or "hai".

Yadda, yadda, yadda...

: Feature: Number of items: Comment:

: Has /x/ in it 0 Not enough
: Has /i/ in it 13 Too many
: Has /h/ in it 5 Not enough
: Has /a/ in it 14 Too many


: /i/ and /h/ 5 Not enough
: /h/ and /a/ 4 Not enough
: /a/ and /i/ 12 Better

: I think we might be getting close here. Can anybody get exactly 10?

Neat statistical calculation without any clear results.

Is this all you have to throw at me now?

Yours,

Yuri Kuchinsky

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
GKeyes6988 <gkeye...@aol.com> wrote in article

<19990217190530...@ng15.aol.com>...
> Yuri Kuchinsky wrote:

> >I'm quite satisfied that NW Coast language area shows clear cognates with
> >non-Polynesian Austronesian. Good enough for me at this point...
>
> Oh, of course it's good enough for you. But according to you (and the
> scenario you agreed to above) the Polynesian languages ought to be
> closer to NW coast languages than to any of the languages in
> Austronesia.

You still misundertand this whole debate, Greg... I've explained your
errors in another post already.

> (Greg)
> >> Between the CMP GIGI forms and the NW coast we have (at best) forms
> >> like Ngipan, which if we really stretch have a CV resemblance. That's
> >> bad too, because NW coast languages ought to be something like the
> >> near Asian ones --
>
> (Yuri)
> >Questionable assumption.
>
> Care to embarrass yourself by explaining why?

You've laid out your basic lack of comprehension of this whole problem
already abundantly. Your assumption is questionable because your "near
Asian ones" is quite vague and meaningless.

> (Yuri)
> >I've been through this already with Ross. Your assumption is that
> >Polynesian-related languages of Melanesia have been there ever since. But
> >the reality is more likely to be that they are relatively recent in that
> >area.
>
> Does this statement make any sense to anyone here? How can the other
> Oceanic languages be closely related to Polynesian (more closely than
> either is to the other Austronesian languages) if they have been
> separated since before Yuri's putative Kwakunesians left Asia for the
> NW coast?

I've explained your error already.

> (Yuri)
> >Try to realize that this is the crucial point in this whole discussion.
> >You simply cannot expect me to grant you the validity of such an
> >assumption. This is neither realistic nor logical.
> >
>
> Now the schizophrenia sets in. You granted it at the top of the post.
> You had to. If Polynesian was separated from the other Austronesian
> languages for millenia,

But which languages?

> there ought to be a jarring disjunction between them. No way around
> it, period. If Polynesian came from the NW coast, it ought to be
> related to some NW coast language more closely than it is to, say,
> Woleaian.

False assumption.

> Your only way around that is to start changing your claim -- which I
> see you have been doing in the past day or so. Now you say the NW
> coast languages might not be Austronesian at all.

At this time, maybe not. For my theory to work, Austronesian languages
merely had to be there _2000 years ago_. That's all. Even if I find _some
traces of them_ on NW Coast at this time, this will support my theory
already. Understand now?

> But that makes us wonder even more why we need a trip to the NW coast
> to place speakers of Austronesian in Hawai'i, doesn't it? So your
> only choice is to waver bewteen the two positions, posit languages you
> have no evidence for (NE Asian ones) and pretend that Polynesian isn't
> an Oceanic language -- now that you finally know what an Oceanic
> language is.

You're quite confused.

...

> Yuri
> >If you consult Hill-Tout, you will find quite a few cognates, such as
> >apoi, Silong
> >api, Solor
> >
> >and others.
>
> I repeat. No LAP.

I repeat. Some possible cognates of LAP.

> I gave plenty of cognates like the ones you just posted, and its clear
> where they fall out in relation to the Polynesian languages.

Even if so, why do they _have to be_ similar to Polynesian languages in
any case? Since I can show similarities of NW Coast and Austronesian, this
strengthens my theory already.

...

> >> According to my dictionary, the Kwakw'ala word for Fire is XIKELA.
> >
> >And H-T transcribes this as hai-kala. So?
>
> So /h/ and /x/ represent different sounds.

But ha and x represent the same sound. This is what you've missed! But now
you seem to be finally getting clued in...

> (Yuri)
> >This should be a very good object lesson for you, Greg. Now you see how
> >you jumped into this whole debate without adequate preparation. There's a
> >reason why these relationships are unknown at this time. That's because
> >these languages are very difficult to analyse. The phonetics are very
> >complex.
>
> Oh, yes, so complex that your 'cognates' don't look like cognates
> whereas mine do.

You're dreaming. Your cognates don't prove your theory -- whatever it is
-- at all.

Take care,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

For every credibility gap, there is a gullibility fill

Miguel Carrasquer Vidal

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
On 18 Feb 1999 16:53:19 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 11:42:44 +1300:

>: This particular example is one that regularly fools beginners, who look
>: at, say Maori /ahi/ "fire" and /ahiahi/ "afternoon,evening", and say
>: "Aha!". (The second word doesn't mean "sunset" of course -- that's a
>: kind of stepping-stone meaning by which we make the connection.) In fact
>: the two are from completely different Austronesian originals: *apuy
>: "fire" and *Rabi(Rabi) "evening". It just happens that in all Oceanic
>: languages *uy > i and *b merges with *p, and in many *R > 0, which leads
>: to homophony.
>
>Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
>are still to explain how and why _on NW Coast too_ the words for "fire"
>and "sunset" are _also_ commonly related.
>
>So try again.

Read again. "Fire" *apuy and "evening" *Rabi are _not_ related
in the Oceanic langauges.

Ross, "you're wasting your time presenting Yuri with comparative
Austronesian data. One way or another, Yuri will manage to reject


it, misinterpret it as supporting his theories, or just pretend

it doesn't exist."

GKeyes6988

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Yuri Kuchinsky wrotE

Translation: No, Yuri doesn't know what a cognate is, and demonstrates it
almost continually. But as he's trying to come off as an expert in this matter,
he really resents Ross pointing this out. Therefore, he scientifically insults
Ross, imagining this will make things better.


>: Let's see now: What "Gigi" and "hi-hi-oh" have in common is...two
>: successive identical syllables with /i/. We note that /hihioh/ is the
>: plural of /hioh/, evidently formed by reduplication. So in order to make
>: these cognate we are going to have to compare what (as far as we know)
>: is a single root morpheme in Kwakiutl with a reduplicated initial
>: syllable in Ntlak., ignoring the /oh/ part. This is not impossible, and
>: might even be justifiable if we knew more about these languages; but it
>: is a little less than "obvious". Next we have to look at the "g" and
>: "h". Although Hill-Tout's "h" could mean any number of things, the two
>: are phonetically in the right neighbourhood that they could both be
>: derived from some velar/uvular original. Now all we need is a couple of
>: additional words with this kind of resemblance showing Kwakiutl /g/:
>: Ntlak. /h/, and we'll have the beginnings of a case...
>
>Very fine. Now, don't try to claim that Your Opinions on this are the only
>one possible.

Translation: Yuri isn't sure exactly what Ross means, and he doesn't want to
make that obvious by trying to analyse it. For one thing, he's worried this
will make it very clear indeed he doesn't know what a cognate is. So instead
he will merely cast doubt on Ross' analysis without actually giving any
linguistic reason for that doubt, or offering a contrary analysis.

>: > Thanks for beating me to posting the evidence that refutes you, and
>proves
>: > my point, Ross. Your original statement that "all resemblance to "gigi"
>: > has vanished at this point" was obviously wrong.
>
>: No, you just didn't understand it.
>
>Yadda, yadda.

A compelling argument indeed. Yuri knows he didn't understand it, but he's
hoping some myopic lurker will miss that.

>: See above. If we throw in "hi-hi-oh",
>: we have four out of fourteen languages with forms which might just be
>: cognate with /gigi/. So the rest are just to add weight?
>
>Some of the rest are also arguable. And you were still wrong.

But Yuri won't argue them because he can't. If you look back through this
post, there isn't a shred of analysis or lingusitic argument to the affect that
Ross was "wrong", but instead evasion and re-assertion.

-- Greg Keyes

Brian M. Scott

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
On 18 Feb 1999 16:58:51 GMT, yu...@globalserve.net (Yuri Kuchinsky)
wrote:

>GKeyes6988 <gkeye...@aol.com> wrote

>> Your only way around that is to start changing your claim -- which I
>> see you have been doing in the past day or so. Now you say the NW
>> coast languages might not be Austronesian at all.

>At this time, maybe not. For my theory to work, Austronesian languages


>merely had to be there _2000 years ago_. That's all. Even if I find _some
>traces of them_ on NW Coast at this time, this will support my theory
>already. Understand now?

And this theory is exactly what? Are you capable of giving us a clear
and accurate summary of the claims that your theory currently makes?
If not, you have no theory.

Brian M. Scott

Doug Weller

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
In article <7ahfpq$qbl$1...@whisper.globalserve.net>, on 18 Feb 1999 16:40:26
GMT, yu...@globalserve.net said...

> Personal attacks is the only weapon of a looser?
>

Doug
--
Doug Weller Moderator, sci.archaeology.moderated
Submissions to: sci-archaeol...@medieval.org
Doug's Archaeology Page: http://www.ramtops.demon.co.uk
Co-owner UK-Schools mailing list: email me for details

Holoholona

unread,
Feb 18, 1999, 3:00:00 AM2/18/99
to
Yuri Kuchinsky wrote:

> You do blather a lot of late, don't you, Ross? For a hypocrite like you
> this sure would indicate that you're up the creek without a paddle? Taking
> some lessons from Tony the Brain-Dead?


Yuri Kuchinsky also wrote:

> Personal attacks is the only weapon of a looser?
>


Shouldn't this end the thread, folks?

Ross Clark

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Yuri Kuchinsky wrote:
>
> Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 10:09:02 +1300:
> : Yuri Kuchinsky wrote:
>

> You do blather a lot of late, don't you, Ross? For a hypocrite like you
> this sure would indicate that you're up the creek without a paddle? Taking
> some lessons from Tony the Brain-Dead?
>
>
>

> Very fine. Now, don't try to claim that Your Opinions on this are the only
> one possible.
>

> Yadda, yadda.
>

> Some of the rest are also arguable. And you were still wrong.
>

Yes, the Content-O-Meter is reading a flat zero.
Let's see what the manual says:
- Check that power is on
- Run calibration test
- If reading is still zero, the poster has nothing to say.

Ross Clark

Ross Clark

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Yuri Kuchinsky wrote:
>

> Yadda, yadda, yadda...
>

>
> Yadda, yadda, yadda...
>
>
> Yadda, yadda, yadda...

[Ross Clark wrote:]


>
> : This particular example is one that regularly fools beginners, who look
> : at, say Maori /ahi/ "fire" and /ahiahi/ "afternoon,evening", and say
> : "Aha!". (The second word doesn't mean "sunset" of course -- that's a
> : kind of stepping-stone meaning by which we make the connection.) In fact
> : the two are from completely different Austronesian originals: *apuy
> : "fire" and *Rabi(Rabi) "evening". It just happens that in all Oceanic
> : languages *uy > i and *b merges with *p, and in many *R > 0, which leads
> : to homophony.
>
> Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you

> are still to explain how and why _on NW Coast too_ the words for "fire"
> and "sunset" are _also_ commonly related.

Perhaps I need to simplify even further. What I have just pointed out is
that (i) the Austronesian words mean "afternoon, evening" and not
"sunset", (ii) the words for "fire" and "afternoon, evening" in
Austronesian are *not* related, but are accidentally homophonous in many
languages. This unfortunately puts paid to Hill-Tout's claim that "This
feature must be regarded as furnishing evidence of a high order of a
psychical character."
Hill-Tout cites some words from British Columbia languages that do show
similarity, eg Lillooet "Rulap" (fire) "Rap" (evening). Whether these
words are historically connected or merely accidentally similar could be
investigated. Unfortunately, it's not the sort of thing we can just take
Hill-Tout's word for.


[Yuri continued:]

>
> Yadda, yadda, yadda...
>

Ross Clark

GKeyes6988

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to

Yuri Kuchinsky wrote:

>GKeyes6988 <gkeye...@aol.com> wrote in article
><19990217190530...@ng15.aol.com>...
>> Yuri Kuchinsky wrote:

(Yuri)


>> >I'm quite satisfied that NW Coast language area shows clear cognates with
>> >non-Polynesian Austronesian. Good enough for me at this point...

(Greg)


>> Oh, of course it's good enough for you. But according to you (and the
>> scenario you agreed to above) the Polynesian languages ought to be
>> closer to NW coast languages than to any of the languages in
>> Austronesia.

(Yuri)


>You still misundertand this whole debate, Greg... I've explained your
>errors in another post already.

Sure you did, Yuri. Sure you did.

>> (Greg)
>> >> Between the CMP GIGI forms and the NW coast we have (at best) forms
>> >> like Ngipan, which if we really stretch have a CV resemblance. That's
>> >> bad too, because NW coast languages ought to be something like the
>> >> near Asian ones --

>> (Yuri)
>> >Questionable assumption.

(Greg)

>> Care to embarrass yourself by explaining why?

(Yuri)


>You've laid out your basic lack of comprehension of this whole problem
>already abundantly. Your assumption is questionable because your "near
>Asian ones" is quite vague and meaningless.

Ah. So it's not my assumption that's questionable -- you just don't understand
what I was talking about. In other words, you have no linguisitic argument
behind your statement, you just want to pretend I wasn't clear about which
languages I was refering to -- even though I gave you a list of them. Though
it's clear you are stalling, I will reiterate what I'm sure everyone else here
remembers, and which they only have to go back a few posts to find.

1. You yourself have evaded questions about where your Kwakunesians came from,
though you hint at NE Asia. There are, of course, no Austronesian languages in
NE Asia.

2. This leaves us with the Austronesian languages closest to NE Asia and hence,
by your proposed route, the NW coast. Those would be the Austronesian
languages of Taiwan, which is where the highest level of linguistic diversity
within the Austronesian languages exist. Generally, this is considered to be
the dispersal point for Austronesian. I mentioned some of these in an earlier
post.

3. The Astronesian languages of Taiwan do not share the innovations of the
Malayo-Polynesian languages (Austronesian sans the Formosan languages), and
they are even more divergent from the Western Malayo-Polynesian languages you
cited (GIGI et al) came from. The cognate list I posted demonstrated this
pretty well. I'm sorry it confused you, but it didn't seem to confuse anyone
else. Sorry to say, I think the confusion comes from your head, not from my
presentation.

4. Now -- follow this closely -- Whether your Kwakunesians took off from your
unlikely NE Asian homeland or from Taiwan -- or even from the Phillipines if
this happened a long time ago (a few thousand years) then they left WITHOUT the
innovations that occured later and far away in the region of Sulawesi. So
while the Kwakunesian languages in the NW coast would have developed their own,
distinctive innovations, they would not possess those present around Sulawesi
unless they CAME from there AFTER those innovations developed. Kwakunesian's
closest affinities would be to the languages they last shared an ancestor with,
those present in Taiwan or the Phillipines or wherever. Sure, innovation
happens in both places -- but what you could never get under these
circumstances is a NW coast language with closer affinities to Konjo than to
Rukai, or whichever Taiwanese dialect they were putativly speaking. Your
comparisons don't reflect that.

Let's clean this up and follow it a little further, shall we?

5. The Kwakunesian languages hang out in the Pacific NW for untold generations.
They develop their own special character, undergo their own sound changes,
morphological oddites, vocabulary replacement, and so on. Then, circa 2000
years ago (or whenever you are now saying this happened) they head off for
Hawai'i. Then (as you seem to be going now) all Austronesian languages in the
NW vanish, leaving only a few words floating around like measles among
languages they aren't genetically related to.

6. Now the Kwakunesians are in Hawai'i, speakers of an Austronesian language
which has been separated from the other Austronesian languages for Millenia.
meanwhile, other Austronesian speakers are pushing their way into near Oceania
and Micronesia. Again, they speak their own distinctive brand of Austronesian.
This group of languages, Oceanic, has in its history the innovations that
occured in Malayo-Polynesian, but it now has its own set of innovations in
phonology, vocabulary, and so forth built on top of that.

7. Eventually they bump into the Kwakunesians, who have the same shared
innovations, though seperated by thousands of years and miles, most of which
was spent in an environment wholly unlike Polynesian. It's a miracle!

8. Meanwhile, the floating terms in the NW coast don't look anything like
Hawai'ian or Samoan.

And no, your "borrowing" scenario won't fly just because you want it to, and
only someone wholly ignorant of the literature and data involved would ever
suggest it. You don't think the Oceanic languages (including the Polynesian
ones) reflect a genetic unity? Take it up with a hundred years of scholarship
-- not with me. Explain why the regular sound correspondances etc. aren't,
somehow. In fact, lay it all out here, so we can debate it. As I said
earlier, the genetic relationship of Oceanic is no more in question than that
Norse is Germanic. Wishing it away won't help.

This leaves only two possibilities:

1. The Kwakunesians sailed to Near Oceania, junked their original
Austronesian language and adopted an Oceanic one wholesale, stem to stern.
They then sailed back to Polynesia, where their language aquired the
distinctive innovations of the Polynesian branch of Oceanic.

2. The Kwakunesians were only recently separated from other Oceanic speakers,
and came from the same direction that the rest of them did -- from the
direction of Asia. They were never on the American NW coast.

I leave it to the discerning reader to figure out which one makes more sense.


>> (Yuri)
>> >I've been through this already with Ross. Your assumption is that
>> >Polynesian-related languages of Melanesia have been there ever since. But
>> >the reality is more likely to be that they are relatively recent in that
>> >area.

(Greg)

>> Does this statement make any sense to anyone here? How can the other
>> Oceanic languages be closely related to Polynesian (more closely than
>> either is to the other Austronesian languages) if they have been
>> separated since before Yuri's putative Kwakunesians left Asia for the
>> NW coast?

(Yuri)


>I've explained your error already.

You've explained that you think all Oceanic languages -- Melanesian and
Micronesian -- borrowed huge chunks of vocabulary, grammar, and phonology while
at the same time retaining the wholesale apperance of regular sound
correspondances and so forth. Given the fact that you have demonstrated pretty
much no knowledge of the languages in question -- to the point of not knowing
what I meant when I refered to them -- I think we can pretty much conclude they
you are talking from out of a very tall hat, perhaps wirth a very long nose
thrown in. Of course, since you don't understand linguistics either, you
probably don't understand that we could TELL if this was due to massive
borrowing (one word can be a problem -- whole vocabularies, no chance) but I
suspect you don't care. Ross it right -- you have your Truth, and damn any
data -- no matter how copious -- and it's very copious -- that contradicts it.


>> (Yuri)
>> >Try to realize that this is the crucial point in this whole discussion.
>> >You simply cannot expect me to grant you the validity of such an
>> >assumption. This is neither realistic nor logical.
>> >

(Greg)

>> Now the schizophrenia sets in. You granted it at the top of the post.
>> You had to. If Polynesian was separated from the other Austronesian
>> languages for millenia,

(Yuri)
>But which languages?

Sigh. The only question is whether you are really this confused or just
pretending to be. I'm going to guess both.


(Greg)


>> there ought to be a jarring disjunction between them. No way around
>> it, period. If Polynesian came from the NW coast, it ought to be
>> related to some NW coast language more closely than it is to, say,
>> Woleaian.

(Yuri)
>False assumption.

Well, I suppose not if you are now postulating all NW coast Austronesian
languages are dead. But your borrowed words in NW coast languages ought to
look more like Polynesian, and they don't. And see above, again. Repeat.


(Greg)


>> Your only way around that is to start changing your claim -- which I
>> see you have been doing in the past day or so. Now you say the NW
>> coast languages might not be Austronesian at all.

(Yuri)


>At this time, maybe not. For my theory to work, Austronesian languages
>merely had to be there _2000 years ago_. That's all. Even if I find _some
>traces of them_ on NW Coast at this time, this will support my theory
>already. Understand now?

Yes, I understand that you are frantically trying to salvage something from
this you just might possibly be right about. It's odd to me that you would
rather appear dumb than admit you were wrong, but you seem hell-bent on it.


(Greg)


>> But that makes us wonder even more why we need a trip to the NW coast
>> to place speakers of Austronesian in Hawai'i, doesn't it? So your
>> only choice is to waver bewteen the two positions, posit languages you
>> have no evidence for (NE Asian ones) and pretend that Polynesian isn't
>> an Oceanic language -- now that you finally know what an Oceanic
>> language is.

(Yuri)
>You're quite confused.

Nope, that's the lay of the land. Even if you were to make some of your
cognates fly, it still doesn't give any linguisitc reason to think Polynesians
came from the NW coast -- in fact, very much the contrary. If you don't see
why by now, you're pretty much hopeless. The migration from the NW coast to
Polynesia is the keystone of this little fantasy, and no amount of handwaving
over GIGI or hypothesis shifting can fix that. You seem to be presenting us
with a sort of linguistic scorched-earth policy, first wiping out all of the
Austronesian speakers in NE Asia and now obliterating those on the NW coast in
the vague hope this will somehow make us forget that it is the POLYNESIAN/NW
coast connection you must demonstrate, not some vague maybe-coulda connection
between the NW coast and Sulawasi.
.

>> Yuri
>> >If you consult Hill-Tout, you will find quite a few cognates, such as
>> >apoi, Silong
>> >api, Solor
>> >
>> >and others.

(Greg)
>> I repeat. No LAP.

(Yuri)


>I repeat. Some possible cognates of LAP.

(Greg)


>> I gave plenty of cognates like the ones you just posted, and its clear
>> where they fall out in relation to the Polynesian languages.

(Yuri)


>Even if so, why do they _have to be_ similar to Polynesian languages in
>any case? Since I can show similarities of NW Coast and Austronesian, this
>strengthens my theory already.

Nope. Polynesian languages are Oceanic. They are very similar to the other
Oceanic languages and fairly different from the ones you are trying to
establish cognates with. I think you get this by now.

(Greg)


>> >> According to my dictionary, the Kwakw'ala word for Fire is XIKELA.
>> >

(Yuri)


>> >And H-T transcribes this as hai-kala. So?

(Greg)

>> So /h/ and /x/ represent different sounds.

(Yuri)


>But ha and x represent the same sound. This is what you've missed! But now
>you seem to be finally getting clued in...

You've snipped the part where I made this clear. Do you really not understand
this yet? Yes, HT used /h/ to represent /x/, but that doesn't make Kwakw'ala
/x/ the same sound as Roti /h/.

>> (Yuri)
>> >This should be a very good object lesson for you, Greg. Now you see how
>> >you jumped into this whole debate without adequate preparation. There's a
>> >reason why these relationships are unknown at this time. That's because
>> >these languages are very difficult to analyse. The phonetics are very
>> >complex.

(Greg)


>> Oh, yes, so complex that your 'cognates' don't look like cognates
>> whereas mine do.

(Yuri)


>You're dreaming. Your cognates don't prove your theory -- whatever it is
>-- at all.

Umm -- that would be hypothesis #2, which I laid our pretty clearly.

-- Greg Keyes

GKeyes6988

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
>
>Thanks, Miguel, but I already knew all this. I've made a mistake when
>arguing with Greg when I labelled some languages Oceanic (I wasn't even
>sure at that point about the exact geographical locations of some of these
>obscure languages), whereas I should have said Austronesian. This was a
>minor misstatement, provoked by Greg's general confusion about this whole
>discussion. Now, of course he and other bumblers will harp about this
>until Kingdom come. Personal attacks is the only weapon of a looser?

Actually, the first time this came up was when I said that Polynesian
languages were related to the other Oceanic languages. You said I was babbling
and then asked which particular languages I meant, as if Oceanic didn't
designate a grouping. Then you cited several "Oceanic" languages which weren't
and so on. Shall I assemble your posts in chronological order? Would that
clear things up?

The point isn't so much that you made a mistake, but that you're trying to
carry on a conversation about something which you don't even have the most
basic knowledge about.

Pointing out that someone doesn't know what they are talking about, I suppose,
could be considered a personal attack. But that pales, as everyone here knows,
next to the sort of vitriol you daily unleash.


--- Greg Keyes

George Black

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
In article <36CC9B...@qualcomm.com>, bmo...@qualcomm.com wrote:
>Yuri Kuchinsky wrote:
>
>> You do blather a lot of late, don't you, Ross? For a hypocrite like you
>> this sure would indicate that you're up the creek without a paddle? Taking
>> some lessons from Tony the Brain-Dead?

I have followed (with interest) the linguistic thread.
Sadly Yuri has learnt nothing in spite of the lengthy explanations.

>Yuri Kuchinsky also wrote:
>
>> Personal attacks is the only weapon of a looser?

Out of the mouths .......................


Tomorrow is only a day away.

George

Jonathan Stone

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
In article <7ah75h$emn$1...@tensegrity.CritPath.Org>, "Anthony West" <aaw...@critpath.org> writes:
|> The charm of Yuri's raving can only be fully appreciated by linguists.
[... ]

Not really. Yuri's at his most charming when he's throwing away 2,500
years of physics by dissing Archimedes' principle.


Btw, i note that you are now a Kuchinksy "Loon". As I am still the
Kuchinksy "King of Loons", I guess that makes you my Kuchinksy "Subject".
appalling thought.

Keep up the good work. Maybe you can make "Prince of Loons".

Peter T. Daniels

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
> (Greg)
> >> So /h/ and /x/ represent different sounds.
>
> (Yuri)
> >But ha and x represent the same sound. This is what you've missed! But now
> >you seem to be finally getting clued in...
>
> You've snipped the part where I made this clear. Do you really not understand
> this yet? Yes, HT used /h/ to represent /x/, but that doesn't make Kwakw'ala
> /x/ the same sound as Roti /h/.

Clearer (except to Yuri) might be: HT used <h> to represent [x], but
that doesn't make K. [x] the same sound as R. [h].
--
Peter T. Daniels gram...@worldnet.att.net

Yuri Kuchinsky

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Thu, 18 Feb 1999 18:21:41 GMT:
: On 18 Feb 1999 16:53:19 GMT, yu...@globalserve.net (Yuri
: Kuchinsky) wrote:

: >Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 11:42:44 +1300:
: >: This particular example is one that regularly fools beginners, who look


: >: at, say Maori /ahi/ "fire" and /ahiahi/ "afternoon,evening", and say
: >: "Aha!". (The second word doesn't mean "sunset" of course -- that's a
: >: kind of stepping-stone meaning by which we make the connection.) In fact
: >: the two are from completely different Austronesian originals: *apuy
: >: "fire" and *Rabi(Rabi) "evening". It just happens that in all Oceanic
: >: languages *uy > i and *b merges with *p, and in many *R > 0, which leads
: >: to homophony.
: >
: >Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
: >are still to explain how and why _on NW Coast too_ the words for "fire"
: >and "sunset" are _also_ commonly related.

: >
: >So try again.

: Read again. "Fire" *apuy and "evening" *Rabi are _not_ related
: in the Oceanic langauges.

That's according to Ross. But he may be wrong? What if this is not a case
of homophony?

Now, if you scratch your head and think a little, you will see that I'm
asking Ross to explain if on NW Coast this is also a case of homophony.
Just a simple coincidence? Or maybe something else?

So I recommend you think before you post.

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku

"Contrariwise," continued Tweedledee, "if it was so, it might be, and
if it were so, it would be; but as it isn't, it ain't. That's logic!"
-- Lewis Carroll, "Through the Looking Glass"

Yuri Kuchinsky

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Ross Clark (benl...@ihug.co.nz) wrote on Fri, 19 Feb 1999 09:26:56 +1300:
: Yuri Kuchinsky wrote:
: >
: > Ross Clark (benl...@ihug.co.nz) wrote on Thu, 18 Feb 1999 10:09:02 +1300:
: > : Yuri Kuchinsky wrote:
: >

: > You do blather a lot of late, don't you, Ross? For a hypocrite like you
: > this sure would indicate that you're up the creek without a paddle? Taking
: > some lessons from Tony the Brain-Dead?

: >
: >
: >
: > Very fine. Now, don't try to claim that Your Opinions on this are the only
: > one possible.
: >

: > Yadda, yadda.
: >

: > Some of the rest are also arguable. And you were still wrong.
: >

: Yes, the Content-O-Meter is reading a flat zero.

Funny that my Content-O-Meter was giving exactly the same reading when I
went through your original post...

Yuri Kuchinsky

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Ross Clark (benl...@ihug.co.nz) wrote on Fri, 19 Feb 1999 09:40:56 +1300:

[Yuri:]


: > Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
: > are still to explain how and why _on NW Coast too_ the words for "fire"
: > and "sunset" are _also_ commonly related.

: Perhaps I need to simplify even further. What I have just pointed out is


: that (i) the Austronesian words mean "afternoon, evening" and not

: "sunset", (ii) the words for "fire" and "afternoon, evening" in


: Austronesian are *not* related, but are accidentally homophonous in many
: languages. This unfortunately puts paid to Hill-Tout's claim that "This
: feature must be regarded as furnishing evidence of a high order of a
: psychical character."

That's _if_ your explanation is correct.

: Hill-Tout cites some words from British Columbia languages that do show


: similarity, eg Lillooet "Rulap" (fire) "Rap" (evening). Whether these
: words are historically connected or merely accidentally similar could be
: investigated. Unfortunately, it's not the sort of thing we can just take
: Hill-Tout's word for.

And, unfortunately, it's not the sort of thing that I can just take _your_
word for.

Do these strange NW Coast languages tend to homophony exactly where
Austronesian languages also do? Just a simple coincidence? Or maybe
something else?

Yours,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

Miguel Carrasquer Vidal

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
On 19 Feb 1999 17:23:31 GMT, yu...@globalserve.net (Yuri
Kuchinsky) wrote:

>Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Thu, 18 Feb 1999 18:21:41 GMT:


>: Read again. "Fire" *apuy and "evening" *Rabi are _not_ related
>: in the Oceanic langauges.
>
>That's according to Ross.

No, that's according to the body of evidence for the
reconstruction of Austronesian as it has been accumulated over
the last centuries by people that:

a) on average were smarter than you
b) were aware of how to do linguistics
c) knew what the fuck an Autronesian language was

Yuri Kuchinsky

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Peter T. Daniels (gram...@worldnet.att.net) wrote on 19 Feb 1999 15:39:07 GMT:
: > (Greg)
: > >> So /h/ and /x/ represent different sounds.
: >
: > (Yuri)
: > >But ha and x represent the same sound. This is what you've missed! But now
: > >you seem to be finally getting clued in...
: >
: > You've snipped the part where I made this clear. Do you really not understand
: > this yet? Yes, HT used /h/ to represent /x/,

No, Greg, clean your glasses and read again. I said ha and x represent the
same sound. And that's the way it is.

: but that doesn't make Kwakw'ala


: > /x/ the same sound as Roti /h/.

And I never claimed it does. But Kwakw'ala ha = x is probably the same
sound as e.g. in Roti ha?i, Garam hai, etc.

: Clearer (except to Yuri) might be: HT used <h> to represent [x], but
: that doesn't make K. [x] the same sound as R. [h].

I'm glad you're managing to follow some of this discussion, Peter.

Regards,

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

Most of the evils of life arise from man's being
unable to sit still in a room || B. Pascal

Eva Kifri

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
In sci.archaeology Yuri Kuchinsky <yu...@globalserve.net> wrote:
(I think Greg wrote this part:)

: : but that doesn't make Kwakw'ala


: : > /x/ the same sound as Roti /h/.

: And I never claimed it does. But Kwakw'ala ha = x is probably the same


: sound as e.g. in Roti ha?i, Garam hai, etc.

You never claimed that it does but then go on to claim that
it does????? HT and Campbell use <h> to represent a what is more
commonly represented by <x>. The Roti sounds they represent by <h> is
NOT the same sound as the Kw. sound. That is why they are normally
represented by different characters. Campbell and Hill-Tout were
struggling to represent unfamiliar sounds. This is not a matter that
you can dismiss as dialectal difference. There is such a thing as
accuracy in transcription.
: Regards,

: Yuri.

: Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

-eva
: Most of the evils of life arise from man's being

Brian M. Scott

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to
Yuri Kuchinsky wrote:

> Peter T. Daniels (gram...@worldnet.att.net) wrote on 19 Feb 1999 15:39:07 GMT:

> : > (Greg)


> : > >> So /h/ and /x/ represent different sounds.

> : > (Yuri)
> : > >But ha and x represent the same sound. This is what you've missed! But now
> : > >you seem to be finally getting clued in...

> : > You've snipped the part where I made this clear. Do you really not understand
> : > this yet? Yes, HT used /h/ to represent /x/,

> No, Greg, clean your glasses and read again. I said ha and x represent the


> same sound. And that's the way it is.

> : but that doesn't make Kwakw'ala


> : > /x/ the same sound as Roti /h/.

> And I never claimed it does. But Kwakw'ala ha = x is probably the same


> sound as e.g. in Roti ha?i, Garam hai, etc.

Well, no. The <h> in the Roti word represents [h]; the <h> (or <ha>, if
you prefer, though I doubt that your interpretation is the same as
Campbell's) in Campbell's transcription represents [x]. And [h] and [x]
are quite different sounds.

> : Clearer (except to Yuri) might be: HT used <h> to represent [x], but
> : that doesn't make K. [x] the same sound as R. [h].

> I'm glad you're managing to follow some of this discussion, Peter.

It was nice of you to confirm his estimate of your level of
understanding, Yuri.

Brian M. Scott

Anthony West

unread,
Feb 19, 1999, 3:00:00 AM2/19/99
to

GKeyes6988 wrote in message <19990219194202...@ng01.aol.com>...
>(Yuri Kuchinsky) wrote.
>>Date: 2/19/99 3:01 PM EST
>>Message-id: <7akfup$13k$1...@whisper.globalserve.net>

>>
>>Peter T. Daniels (gram...@worldnet.att.net) wrote on 19 Feb 1999
15:39:07
>>GMT:
>>: > (Greg)
>>: > >> So /h/ and /x/ represent different sounds.
>>: >
>>: > (Yuri)
>>: > >But ha and x represent the same sound. This is what you've missed
>>: >

>>: > You've snipped the part where I made this clear. Do you really not
>>understand
>>: > this yet? Yes, HT used /h/ to represent /x/,
>>
>>No, Greg, clean your glasses and read again. I said ha and x represent the
>>same sound. And that's the way it is.
>>
>>: but that doesn't make Kwakw'ala
>>: > /x/ the same sound as Roti /h/.
>>
>>And I never claimed it does. But Kwakw'ala ha = x is probably the same
>>sound as e.g. in Roti ha?i, Garam hai, etc.
>>
>>: Clearer (except to Yuri) might be: HT used to represent [x], but

>>: that doesn't make K. [x] the same sound as R. [h].
>>
>If anyone at all is confused about this, I will have another go at it. I
>thought I was pretty clear, as I thought Ross was earlier. Whether Yuri is
>confused or simply lying at this point is pretty moot.
>
Yuri is simultaneously confused and lying. It's part of his condition.

"Ha and x represent the same sound. And that's the way it is. And I never
claimed [Kwakw'ala /x/ is the same sound as Roti /h/]. But Kwakw'ala ha=x is
probably the same sound as Roti ha?i ...."

This is what raving sounds like, when it is transcribed. I never use this
term loosely, as an insult, but clinically. This is what an intelligent
madman sounds like, in a mental hospital, when his thoughts start getting
all tangled up. I've worked with a few. In Yuri's case, which we are now
watching, Yuri is experiencing the delusion that he understands several very
difficult and unusual languages perfectly, which he has in fact never
studied at all -- that he has merely been fingering wordlists from,
obsessively repeating their letters like a mantra, that seem to him filled
with some cosmic meaning meant for his ears alone.

Psychotropics produce the best results. But one should not reinforce Yuri's
delusions anyway.

Tony West
Philadelphia aaw...@critpath.org


Ross Clark

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
Yuri Kuchinsky wrote:
>
> Ross Clark (benl...@ihug.co.nz) wrote on Fri, 19 Feb 1999 09:40:56 +1300:
>
> [Yuri:]
> : > Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
> : > are still to explain how and why _on NW Coast too_ the words for "fire"
> : > and "sunset" are _also_ commonly related.
>
> : Perhaps I need to simplify even further. What I have just pointed out is
> : that (i) the Austronesian words mean "afternoon, evening" and not
> : "sunset", (ii) the words for "fire" and "afternoon, evening" in
> : Austronesian are *not* related, but are accidentally homophonous in many
> : languages. This unfortunately puts paid to Hill-Tout's claim that "This
> : feature must be regarded as furnishing evidence of a high order of a
> : psychical character."
>
> That's _if_ your explanation is correct.

As Miguel has already pointed out, Yuri, it is not "my" explanation.
Calling it that is simply your way of preserving your linguistic
virginity (as the Vow of Ignorance requires) by not having to consider
the body of evidence on which it is based.

It's not that anybody wants to force you to take an interest in
comparative Austronesian. But it would be more honest if you did not
make far-reaching claims of support for your theories based on areas of
science which you refuse to learn about.

>
> : Hill-Tout cites some words from British Columbia languages that do show
> : similarity, eg Lillooet "Rulap" (fire) "Rap" (evening). Whether these
> : words are historically connected or merely accidentally similar could be
> : investigated. Unfortunately, it's not the sort of thing we can just take
> : Hill-Tout's word for.
>
> And, unfortunately, it's not the sort of thing that I can just take _your_
> word for.

Nobody asked you to.

>
> Do these strange NW Coast languages tend to homophony exactly where
> Austronesian languages also do? Just a simple coincidence? Or maybe
> something else?

Woooo-hooo...


Ross Clark

GKeyes6988

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
(Yuri Kuchinsky) wrote.
>Date: 2/19/99 3:01 PM EST
>Message-id: <7akfup$13k$1...@whisper.globalserve.net>
>
>Peter T. Daniels (gram...@worldnet.att.net) wrote on 19 Feb 1999 15:39:07
>GMT:
>: > (Greg)
>: > >> So /h/ and /x/ represent different sounds.
>: >
>: > (Yuri)
>: > >But ha and x represent the same sound. This is what you've missed! But
>now
>: > >you seem to be finally getting clued in...
>: >
>: > You've snipped the part where I made this clear. Do you really not
>understand
>: > this yet? Yes, HT used /h/ to represent /x/,
>
>No, Greg, clean your glasses and read again. I said ha and x represent the
>same sound. And that's the way it is.
>
>: but that doesn't make Kwakw'ala
>: > /x/ the same sound as Roti /h/.
>
>And I never claimed it does. But Kwakw'ala ha = x is probably the same
>sound as e.g. in Roti ha?i, Garam hai, etc.
>
>: Clearer (except to Yuri) might be: HT used to represent [x], but
>: that doesn't make K. [x] the same sound as R. [h].
>
>I'm glad you're managing to follow some of this discussion, Peter.
>
>Regards,
>
>Yuri.


If anyone at all is confused about this, I will have another go at it. I
thought I was pretty clear, as I thought Ross was earlier. Whether Yuri is
confused or simply lying at this point is pretty moot.

-- Greg Keyes

GKeyes6988

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to

Yuri Kuchinsky wrote:

Ross Clark (benl...@ihug.co.nz) wrote on Fri, 19 Feb 1999 09:40:56 +1300:

[Yuri:]
: > Nice dumbing-down try, but I'm afraid it doesn't quite work. Because you
: > are still to explain how and why _on NW Coast too_ the words for "fire"
: > and "sunset" are _also_ commonly related.

Ross


: Perhaps I need to simplify even further. What I have just pointed out is
: that (i) the Austronesian words mean "afternoon, evening" and not
: "sunset", (ii) the words for "fire" and "afternoon, evening" in
: Austronesian are *not* related, but are accidentally homophonous in many
: languages. This unfortunately puts paid to Hill-Tout's claim that "This
: feature must be regarded as furnishing evidence of a high order of a
: psychical character."

(Yuri)
.That's _if_ your explanation is correct.

: Hill-Tout cites some words from British Columbia languages that do show
: similarity, eg Lillooet "Rulap" (fire) "Rap" (evening). Whether these
: words are historically connected or merely accidentally similar could be
: investigated. Unfortunately, it's not the sort of thing we can just take
: Hill-Tout's word for.


(Yuri)
.And, unfortunately, it's not the sort of thing that I can just take _your_
.word for.

You don't have to. Go read some books on the internal relationships of the
Austronesian languages. The changes he walked you though are all well attested
-- loss of initial R being one of the better-known ones because it has an
interesting distribution.


(Yuri)
.Do these strange NW Coast languages tend to homophony exactly where
.Austronesian languages also do? Just a simple coincidence? Or maybe
.something else?

1. You're passing over something even simpler -- that the Polynesian word's
don't mean "Sunset", as Ross had said twice now, and which you have twice
ignored. My Hawai'ian dictionary gives NAPO'O 'ANA O KA LA for sunset. The
sunset/fire connection in Polynesia was in HT's mind (and now firmly stuck in
yours, like the weird orthographic confusion you maintain), and so whether or
not a language or languages on the NW coast make this connection becomes
entirely moot, sinces it has no "parrallel" in Polynesia.

As for the NW, someone really ought to check and see what HT is really giving.
Two of his NW favorites don't show this fire/sunset connection at all, which
makes us wonder just how widespread this was on the NW coast.

Kwakw'ala

Fire XIKELA
Evening DZAKWA
Sunset LENDZASA AS TL'ISELA

Puget Salish

Fire HUD (U)
Evening KwICADI
Sunset SelhAXIL

-- Greg Keyes


Jonathan Stone

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
In article <7al7bf$ems$1...@tensegrity.CritPath.Org>, "Anthony West" <aaw...@critpath.org> writes:

|> This is what raving sounds like, when it is transcribed. I never use this
|> term loosely, as an insult, but clinically. This is what an intelligent
|> madman sounds like, in a mental hospital, when his thoughts start getting
|> all tangled up. I've worked with a few. In Yuri's case, which we are now
|> watching, Yuri is experiencing the delusion that he understands several very
|> difficult and unusual languages perfectly, which he has in fact never
|> studied at all -- that he has merely been fingering wordlists from,
|> obsessively repeating their letters like a mantra, that seem to him filled
|> with some cosmic meaning meant for his ears alone.

This is an interesting observation. Yuri's own writing suggests,
fairly strongly, that we sound the same way to Yuri.

|> Psychotropics produce the best results. But one should not reinforce Yuri's
|> delusions anyway.

Ummmm, you mean in mending the worldviews of those so afflicted, or of
emulating such world-views? (Or both?)

George Black

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
In article <19990219202516...@ng142.aol.com>, gkeye...@aol.com
(GKeyes6988) wrote:


For Maori
I've got evening as ahiahi
fire as ahi
firewood as wahie

Day as ra
sky as rangi
above as runga

The relationships between words appears to be quite solid (as has been pointed
out)

Yuri Kuchinsky

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Fri, 19 Feb 1999 17:54:45 GMT:
: On 19 Feb 1999 17:23:31 GMT, yu...@globalserve.net (Yuri
: Kuchinsky) wrote:

: >Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Thu, 18 Feb 1999 18:21:41 GMT:
: >: Read again. "Fire" *apuy and "evening" *Rabi are _not_ related
: >: in the Oceanic langauges.
: >
: >That's according to Ross.

: No, that's according to the body of evidence for the
: reconstruction of Austronesian as it has been accumulated over
: the last centuries by people that:

: a) on average were smarter than you
: b) were aware of how to do linguistics
: c) knew what the fuck an Autronesian language was

Yes, but unfortunately this does not answer my question if on NWC this is
also a case of homophony. Because, if so, then my comparison is still very
much valid.

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku

"Contrariwise," continued Tweedledee, "if it was so, it might be, and

Yuri Kuchinsky

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
Eva Kifri (e...@idiom.com) wrote on 19 Feb 1999 22:04:06 GMT:

: In sci.archaeology Yuri Kuchinsky <yu...@globalserve.net> wrote:
: (I think Greg wrote this part:)

: : : but that doesn't make Kwakw'ala


: : : > /x/ the same sound as Roti /h/.

: : And I never claimed it does. But Kwakw'ala ha = x is probably the same


: : sound as e.g. in Roti ha?i, Garam hai, etc.

: You never claimed that it does but then go on to claim that
: it does?????

Get yourself a clue, Eva. You're amazingly clued out. Can you see a
difference between /h/ and /ha/?

: HT and Campbell use <h> to represent a what is more


: commonly represented by <x>.

Nope.

I repeat, HT uses /ha/ to transcribe /x/.

xikela = fire (Kwakw'ala, according to a modern transcription)

hai-kala = fire (Kwakw'ala, according to Hill-Tout's transcription)

Can I make this any clearer for you?

Regards,

Yuri.

Yuri Kuchinsky -O- http://www.globalserve.net/~yuku -O- Toronto

You never need think you can turn over any old falsehoods without a
terrible squirming of the horrid little population that dwells under
it -=O=- Oliver Wendell Holmes

Yuri Kuchinsky

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
Brian M. Scott (BMS...@stratos.net) wrote on Fri, 19 Feb 1999 20:04:14 -0500:
: Yuri Kuchinsky wrote:
: > : > (Greg)

: > : > >> So /h/ and /x/ represent different sounds.

: > : > (Yuri)
: > : > >But ha and x represent the same sound. This is what you've missed! But now
: > : > >you seem to be finally getting clued in...

: > : > You've snipped the part where I made this clear. Do you really not understand
: > : > this yet? Yes, HT used /h/ to represent /x/,

: > No, Greg, clean your glasses and read again. I said ha and x represent the
: > same sound. And that's the way it is.

: > : but that doesn't make Kwakw'ala


: > : > /x/ the same sound as Roti /h/.

: > And I never claimed it does. But Kwakw'ala ha = x is probably the same


: > sound as e.g. in Roti ha?i, Garam hai, etc.

: Well, no. The <h> in the Roti word represents [h]; the <h> (or <ha>, if


: you prefer, though I doubt that your interpretation is the same as
: Campbell's) in Campbell's transcription represents [x].

You're beginning to catch up, Brian, but you still can't seem to get the
cast of characters straight. We're talking about Hill-Tout. Not Campbell.
Too many bunglers on this thread...

: And [h] and [x] are quite different sounds.

This is a tautology of course. But what you neglected to mention is that
Hill-Tout's /ha/ is probably the same sound as e.g. in Roti ha?i, Garam
hai, etc. This can be assumed to be so until shown otherwise.

Regards,

Anthony West

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to

Yuri Kuchinsky wrote in message <7amlnl$err$4...@whisper.globalserve.net>...

>Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Fri, 19 Feb 1999 17:54:45
GMT:
>: On 19 Feb 1999 17:23:31 GMT, yu...@globalserve.net (Yuri
>: Kuchinsky) wrote:
>
>: >Miguel Carrasquer Vidal (m...@wxs.nl) wrote on Thu, 18 Feb 1999 18:21:41
GMT:
>: >: Read again. "Fire" *apuy and "evening" *Rabi are _not_ related
>: >: in the Oceanic langauges.
>: >
>: >That's according to Ross.
>
>: No, that's according to the body of evidence for the
>: reconstruction of Austronesian as it has been accumulated over
>: the last centuries by people that:
>
>: a) on average were smarter than you
>: b) were aware of how to do linguistics
>: c) knew what the fuck an Autronesian language was
>
>Yes, but unfortunately this does not answer my question if on NWC this is
>also a case of homophony. Because, if so, then my comparison is still very
>much valid.
>
First, Yuri, other posters have shown that the words for "fire" and "sunset"
are NOT the same or similar in some of the NW Coast languages that you have
been citing. Your sources are dreadfully obsolete and full of primitive
errors.

The burden falls on you to demonstrate, using _contemporary_ language
materials, that such homophones exist in some Salishan or Wakashan
languages. At present, you have not shown that any exist.

After, and only *after* you find some homophones for "fire" and "sunset" can
you then proceed to ask if they are related or reflexes of different roots.

Again, Step One is to pitch Hill-Tout and Campbell and use accurate
material. No linguist will touch junk from the ancient past when reams of
much better material are available.

Tony West
Philadelphia aaw...@critpath.org


GKeyes6988

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to
Yuri Kuchinsky wrote:

>Brian M. Scott (BMS...@stratos.net) wrote on Fri, 19 Feb 1999 20:04:14
>-0500:
>: Yuri Kuchinsky wrote:
>: > : > (Greg)
>: > : > >> So /h/ and /x/ represent different sounds.
>
>: > : > (Yuri)
>: > : > >But ha and x represent the same sound. This is what you've missed!
>But now
>: > : > >you seem to be finally getting clued in...

Greg


>: > : > You've snipped the part where I made this clear. Do you really not
>understand
>: > : > this yet? Yes, HT used /h/ to represent /x/,
>

Yuri


>: > No, Greg, clean your glasses and read again. I said ha and x represent
>the
>: > same sound. And that's the way it is.

>: > : but that doesn't make Kwakw'ala
>: > : > /x/ the same sound as Roti /h/.
>
>: > And I never claimed it does. But Kwakw'ala ha = x is probably the same
>: > sound as e.g. in Roti ha?i, Garam hai, etc.
>

>: Well, no. The in the Roti word represents [h]; the (or <ha>, if


>: you prefer, though I doubt that your interpretation is the same as
>: Campbell's) in Campbell's transcription represents [x].

Yuri


>You're beginning to catch up, Brian, but you still can't seem to get the
>cast of characters straight. We're talking about Hill-Tout. Not Campbell.
>Too many bunglers on this thread...

>: And [h] and [x] are quite different sounds.

Yuri


>This is a tautology of course. But what you neglected to mention is that
>Hill-Tout's /ha/ is probably the same sound as e.g. in Roti ha?i, Garam
>hai, etc. This can be assumed to be so until shown otherwise.

Oh -- well, that's easily done. We just look in the table of phonetic symbols
for each language. That is, to see what sounds the orthographic conventions
represent. Clearly an alien thought to Yuri, who seems to think the letters
ARE the sounds.

In all of the Austronesian examples I cited, including Roti H?AI, the /h/ is a
voiceless glottal fricative (Tryon, vol 1, fasc. 1, pp xviii)

The Kwakw'ala dictionary says of the "X" in XIKELA, 'fire'.:

x = /x/ voiceless palatalized velar fricative. Palatalization assimilates
progressivly before /i/
(Grubb 235)

So, not the same sound -- the glottis and velum are different point of
articulation (that means they SOUND different, Yuri). Let me add the
following: In Kwakw'ala, there are 2 distinct voiceless velar fricatives and
two voiceless uvular ones. To understand the language, you must be able to
distinguish between them.

There is also a voiceless laryngeal fricative in Kwakw'ala, represented by H
/h/. This would be the closest sound to that represented by English H and the
/h/ in Roti. It is not the sound found in XIKELA, fire. A Kwakw'ala example
would be HIK'ALA, "noise". Compare this to HT's rendering of "fire",
"HAI-KALA". You start to see why linguists carry on about orthographic
conventions, yet?

Hill-Tout used H to represent the palatalized velar fricative /x/. What did he
use to represent the Laryngeal fricative /h/, or the uvular fricatives? Likely
he couldn't hear the difference and collapsed most, if not all, into the same
symbol, H. Too bad, because In Kwakw'ala the differences are crucial.

Not the same sound. But this little discussion has given us further insight
into Yuri's linguistic expertise.

-- Greg Keyes

References:

Tryon, Darell, ed.
1995 COMPARATIVE AUSTRONESIAN DICTIONARY. Trends in Linguistic Documentation
10, Winter and Rhodes series eds. Berlin, New York: Mouton de Gruyter.

Grubb, David
1977 A PRACTICAL WRITING SYSTEM AND SHORT DICTIONARY OF KWAKW'ALA. National
Museum of Man Mercury series, Canadian ethnology service paper no. 34. Ottawa:
National Museums of Canada.

Anthony West

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to

Jonathan Stone wrote in message <7al8ii$hf2$1...@nntp.Stanford.EDU>...

>In article <7al7bf$ems$1...@tensegrity.CritPath.Org>, "Anthony West"
<aaw...@critpath.org> writes:
>
>|> This is what raving sounds like, when it is transcribed. I never use
this
>|> term loosely, as an insult, but clinically. This is what an intelligent
>|> madman sounds like, in a mental hospital, when his thoughts start
getting
>|> all tangled up. I've worked with a few. In Yuri's case, which we are now
>|> watching, Yuri is experiencing the delusion that he understands several
very
>|> difficult and unusual languages perfectly, which he has in fact never
>|> studied at all -- that he has merely been fingering wordlists from,
>|> obsessively repeating their letters like a mantra, that seem to him
filled
>|> with some cosmic meaning meant for his ears alone.
>
>This is an interesting observation. Yuri's own writing suggests,
>fairly strongly, that we sound the same way to Yuri.
>
Probably so, at least when he's in a psychiatric crisis. He may have better
days and worse days. Since he is completely solipsistic when in crisis, he
cannot understand why other people say the things they do. He cannot listen
to any of their explanations, because they would quickly lead to the
conclusion that he himself is inadequate. And that's what he's trying not to
think about! One can hardly blame him, really.

>|> Psychotropics produce the best results. But one should not reinforce
Yuri's
>|> delusions anyway.
>
>Ummmm, you mean in mending the worldviews of those so afflicted, or of
>emulating such world-views? (Or both?)

Well, I wouldn't want to spend five minutes of my life inside Yuri's
worldview, that's for sure. But to each his own .... No, I particularly
think that it never helps mentally ill people to "play along" with their
delusions. They need reality testing from healthy people.

Tony West
Philadelphia aaw...@critpath.org


Anthony West

unread,
Feb 20, 1999, 3:00:00 AM2/20/99
to

Yuri Kuchinsky wrote in message <7ammvc$kvi$1...@whisper.globalserve.net>...

>Brian M. Scott (BMS...@stratos.net) wrote on Fri, 19 Feb 1999
20:04:14 -0500:
>: Yuri Kuchinsky wrote:
[snip]

>This is a tautology of course. But what you neglected to mention is that
>Hill-Tout's /ha/ is probably the same sound as e.g. in Roti ha?i, Garam
>hai, etc. This can be assumed to be so until shown otherwise.
>
Let's see. In your response to Eva Kifri, you claimed that in Hill-Tout's
orthography, "ha" = /x/.

Now you're claiming that the correct transcription of Roti ha?i in ASCII IPA
is /x?i/.

You are wrong. This can be assumed to be so. You cannot show otherwise.

Tony West
Philadelphia aaw...@critpath.org


Yuri Kuchinsky

unread,
Feb 21, 1999, 3:00:00 AM2/21/99
to
GKeyes6988 <gkeye...@aol.com> wrote in article
<19990220143012...@ng111.aol.com>...
> Yuri Kuchinsky wrote:

> >But what you neglected to mention is that
> >Hill-Tout's /ha/ is probably the same sound as e.g. in Roti ha?i, Garam
> >hai, etc. This can be assumed to be so until shown otherwise.
>

> Oh -- well, that's easily done.

Really, Greg?

> We just look in the table of phonetic symbols
> for each language. That is, to see what sounds the orthographic conventions
> represent. Clearly an alien thought to Yuri, who seems to think the letters
> ARE the sounds.

You're too confused.

> In all of the Austronesian examples I cited, including Roti H?AI, the /h/ is a
> voiceless glottal fricative (Tryon, vol 1, fasc. 1, pp xviii)
>
> The Kwakw'ala dictionary says of the "X" in XIKELA, 'fire'.:
>
> x = /x/ voiceless palatalized velar fricative. Palatalization assimilates
> progressivly before /i/
> (Grubb 235)

So you go and do this nit-picking analysis to show that the sounds are not
the same but similar... How nice.

...

> What did he [HT]


> use to represent the Laryngeal fricative /h/, or the uvular fricatives? Likely
> he couldn't hear the difference

So what did you prove, Greg, really? Your own blindness and clued out
state?

1. Are you assuming that the sounds have to be _exactly the same_ in order
for this comparison to be valid?

2. Are you assuming that your SHORT DICTIONARY OF KWAKW'ALA has "The One
and Only True Pronunciation" of XIKELA? Perhaps you're so naive that
you've never heard that in fact several dialects of Kwakq'ala are known,
and many varieties of pronunciation? Again, you seem to have missed the
long preamble to this discussion where all this has been covered already?

3. And the most important, now you've "debunked" Kwakq'ala XIKELA, and you
think you can rest on your laurels? This is really not very bright... No,
my dear Greg, your debunking is not even close to finish yet. Just barely
started! That's because you now have to move on to THE NUMEROUS OTHER
WORDS on NWC meaning "fire"! Such as

hauk, Nootka,
haiuk, Snan., Sumas,

and many more, apparently!

The only thing I can say to summarise is

Doh!

Yuri.

Yuri Kuchinsky -=O=- http://www.globalserve.net/~yuku -=O=- Toronto

Peter T. Daniels

unread,
Feb 22, 1999, 3:00:00 AM2/22/99
to
Yuri Kuchinsky wrote:

> > x = /x/ voiceless palatalized velar fricative. Palatalization assimilates
> > progressivly before /i/
> > (Grubb 235)
>
> So you go and do this nit-picking analysis to show that the sounds are not
> the same but similar... How nice.
>
> ...
>
> > What did he [HT]
> > use to represent the Laryngeal fricative /h/, or the uvular fricatives? Likely
> > he couldn't hear the difference
>
> So what did you prove, Greg, really? Your own blindness and clued out
> state?
>
> 1. Are you assuming that the sounds have to be _exactly the same_ in order
> for this comparison to be valid?

Look, another word Yuri doesn't know ... and apparently doesn't even
know he doesn't know: "phoneme"!

Of course, the concept of "phoneme" was not widespread in the time of
Hill-Tout (it was apparently invented in the 1880s by Kruschevsky), and
Yuri hasn't looked at any more recent work on linguistics than that.

0 new messages