Re: Re[2]: [tied] Re: Short and long vowels

From: Patrick Ryan
Message: 39452
Date: 2005-07-27

----- Original Message -----
From: "Brian M. Scott" <BMScott@...>
To: "Patrick Ryan" <cybalist@yahoogroups.com>
Sent: Tuesday, July 26, 2005 11:08 PM
Subject: Re[2]: [tied] Re: Short and long vowels


> At 8:00:54 PM on Monday, July 25, 2005, Patrick Ryan wrote:
>
> > From: "david_russell_watson" <liberty@...>
>
> >> --- In cybalist@yahoogroups.com, "Patrick Ryan"
> >> <proto-language@...> > wrote:
>
> >>> --- In cybalist@yahoogroups.com, "David Russell Watson"
> >>> <liberty@...> wrote:
>
> >>>> You don't still believe that you can reconstruct the
> >>>> first human language, do you?
>
> >> You didn't answer this one.
>
> > Because it is not germane to the question of the
> > relationship of Afrasian to PIE.
>
> > But yes, I believe I have isolated the monosyllables that
> > made up the world's first language.
>
> The belief that this is possible is incompatible with an
> understanding of the timescale involved and the nature of
> linguistic change. Even if by some remarkable accident a
> little positive evidence still existed, it would be
> completely lost in the statistical noise.
>
> This is off-topic here, but since it underlies some of your
> arguments, I can't resist pointing out how badly some of the
> IE evidence has been handled.

***
Patrick:

I can hardly be angry with someone who has actually read what I have
written -- even if he strongly disagrees with it.

Now, you indirectly accuse me of being insensible to the timescale involved
and incapable of understanding the process of linguistic change. I am very
well aware of the timescale (circa 100,000 years) and am very familiar with
how languages change over time.

Firat off, there is nothing in my theory whch denies linguistic change over
time. When I propose that Nostratic /x/ is represented in Sumerian by <h>
and PIE by *gW, I am recognizing linguistic change. The implication of your
argument is that change will be so great as to obscure or make connections
impossible. I deny that; and my conversion tables show that it is simply not
true.

So, in my opinion, you are arguing from linguistic theory; and I am arguing
from linguistic facts.

Ringe and others have constructed theoretical models that are supposed to
'prove' that x-number of words of the vocabulary of an ancestor language
will be lost in descendent languages at a certain rate per year so that,
after 10,000 years or so have elapsed, a very small percentage of originally
common vocabulary items will be left in any two of the descendent
languages.

That is a nice theory, and may have some validity among some language
families, but the facts of Nostratic are different. I estimate that a
connection between Sumerian and PIE is at least 10,000 years old; yet I can
show that _most_ common words in Sumerian can be related through simple
rules of conversion to PIE though, of course, a few do _not_ have known
correspondents. Facts vs. theory.

PIE has been reconstructed well enough to be nearly a fact, and Sumerian, if
we can puzzle our way thorugh the writing system, is definitely a fact as is
hieroglyphoc Egyptian.

Nearly as old is the connection of PIE to Afrasian languages but root after
root can be successfully linked in any given two, and fewer though still an
impressive number in all three of the three.

Sometimes roots have wandered a bit semantically but they are still there if
we can recogniz them.

It seemed for a while as if every three months someone would come up with a
new formula for computing the percentage of vocabulary lost over time so, if
you are up on it, why do you not tell me what the probabilities are of
finding a root common to PIE, Sumerian, and Egyptian. Can you do it?

So, what are the odds on finding a word that means 'tremble' is Egyptian as
d3 (Arabic ta-tartara), Sumerian as dar-5, and PIE as *der-; and a word that
means 'shine' that is Egyptian b3, Sumerian bar-6/7, and PIE bher-?

I have literally dozens of these.

And I can show correspondences in the morphologies also.

***

>
> > You can see them if you wish at:
>
> > http://geocities.com/proto-language/ProtoLanguage-Monosyllables.htm
>
> Oh, dear. 'English has inherited a number of words that
> preserve the earliest meanings of the P[roto-]L[anguage]
> monosyllables: P?E, 'urine', in Modern English <pee>;
> P?E-P?E, 'urinate', in Modern English <peepee>.'


***
Patrick:

Amazing, is it not?

***


> Then there's this, from
> <http://geocities.com/Proto-language/ProtoLanguage-5.htm>.
>
> A strong confirmation of the original three-vowel contrast
> (really glide contrast) of the Proto-Language is found in
> the facts of Old Irish. ... When we see this pervasive
> system in an IE language that has perpetuated an ancient
> situation, it is difficult to believe that it was not a
> feature of, at least, earliest IE, and, most probably
> Nostratic, from which IE and Afrasian among others are
> descended.
>
> OIr <maicc>, gen.sing. of <macc> 'son', continues Ogam Ir.
> MAQ(Q)I, where Q, representing *kW, seems distinctly
> unlikely to have had original palatal quality, and of course
> the final consonant of the nom.sing. <macc>, unlike that of
> <maicc>, does not have palatal quality. The facts of OIr do
> *not* suggest that the system was inherited.


***
Patrick:

You rather conveniently neglected to mention the quote from Thurneysen, a
recognized expert on these matters. Since you did not see fit to mention it,
and seem to be attempting to deny it, I will quote it in full:

"In Old Irish every consonant may have three separate qualities:

1. palatal or i-quality,

2. neutral or a-quality,

3. u-quality.

Modern dialects retain only the first two, the u-quality having coalesced
with the
neutral, for which development see # 174."


Now perhaps you are more knowledgeable than Thurneysen; is that what you are
claiming?

***

> The current dogma among IEists is that palatal and velar
> articulations were occasioned by the presence of e or o
> after the consonant --- in spite of the fact that the
> palatalization or velarization of the consonants persists
> in other phonetic environments.
>
> Examples?
>
> 1) Let us look at the Modern Irish word for "noise
> (loud confused clamor, din)": <cullóid>. We recall
> that the reconstructed Nostratic form would be *kwal-.
> We saw in Greek <kaléo:> that the root was
> differentiated from other "kel"'s by a final -y. It
> begins to looks very much like <cullóid> was derived
> from an early IE kwa"lay (through kwl-"loy), and that
> the w-glide can still be seen reduced in the MI u.
>
> Except that this is EIr <callóid>, a borrowing of Latin
> <collatio> (according to the RIA's Dictionary of the Irish
> Language), from *kom and a suffixed, zero-grade form of
> *telh2- (PIE based on Watkins).

***
Patrick:

Mea culpa.

I wrote that essay seven years ago. During the last seven, I hope my thiking
has developed (and advanced?).

***

> 2) In Modern Irish <scal>, sting of a nettle, we see
> Nostratic s-mobile + *kal-; and in Modern Irish
> sceolang, "fleet, agile", we see s-mobile + Nostratic
> *kyal-.
>
> Except that OIr <sceolang> = <sceola> 'news-bringer;
> survivor (of a battle)', a derivative of <scél> 'story,
> narration, tale; news, tidings'. For modern Irish Dinneen
> gives the primary sense 'a fugitive, a deserter'; the
> adjectival sense 'fleet, agile' is clearly derivative.

***
Patrick:

I would not use these examples today.

***


> 3) Now, consider Old Irish gáu, "falsehood", which
> occurs also in Middle Welsh geu, and Middle Breton
> gou. Since every Irish vowel has one of three
> qualities, which quality does the initial g have? An
> a-quality because of gáu? An e-quality because of geu?
> Or an o-quality because of gou?
>
> 4) Current Celtic theory cannot give a good answer.
>
> Kenneth Jackson had no such problem, deriving all of the
> Brittonic forms from Late Brit. *ga:w- (LHEB 373). (And of
> course the OIr form would take precedence anyway, since the
> phenomenon is specifically Irish.)

***
Patrick:

Well, then we would agree. I would today derive *ga:w- and the others from
PIE *gha(:)wo-.

I agree with you on one thing: I should have never used Old Irish examples.

***

> >>>> That's significant though, because even if you both
> >>>> agree that Sumerian is Nostratic, your two voices don't
> >>>> constitute cumulative support of that when you differ
> >>>> in your actual reconstructions.
>
> >>> Excuse me, but that is ridiculous.
>
> >> No that's simple logic.
>
> > No, it is ignorant.
>
> No. It is correct as far as it goes, but it's incomplete.
> Two different reconstructions can be isomorphic or nearly
> so, in which case they are mutually supportive. (Someone --
> I think Richard -- already made this point in different
> words.) But when the two reconstructions are not nearly
> isomorphic -- when, that is, they are significantly
> different in phonological structure, say, and not just in
> phonetics -- then it is indeed ridiculous to suppose that
> they support each other.

***
Patrick:

Bomhard has used an inductive method to set up his phonological system for
Nostratic; following Anttila, I used a deductive method. Naturally, there
are differences due, primarily, to method.

But I heartily dispute the conclusion you draw from this.

Let us take the triad Egyptian b3, Sumerian bar(-6/7), PIE *bher-; 'shine'.

In my system, I believe they ultimately are traceable to PL *p?fa-ra,
Nostratic *bwar-.

Bomhard reconstructs *b/6r- but connects the two of the same three, missing
b3, and substituting br-g because it has not reached him yet that Egyptian
<3> is /R/, which I proposed 30 years ago among Egyptologists; and which
they finally accepted. This proposal was initiated, by the way, based on a
comparison of various Egyptian b3-roots with PIE *bher-roots.

His <b> is Egyptian <b>, Sumerian <b>, and PIE *bh; just as is my *p?f and
*bw are with the exception that I would modify bar to par.

Now I did not say that they support each other. _That would be ridiculous_.
But they both
support the claim that Egyptian, Sumerian, and PIE are all related through a
common ancestor.

Someone could come up with yet another Nostratic phonology, let us say, for
this root *fil-' as long as *f = Egyptian <b>, Sumerian <b>/<p> and PIE *bh,
etc., it would support the idea that the three are related through a common
ancestor if the correspondences could be repeated in many other roots.

***

>
> >> Your claim is that since both Bomhard and Ryan believe in
> >> the cognacy of Sumerian and P.I.E., that it must be so.
>
> > That is not my view, and it is not Bomhard's view.
>
> > The relationship is not a priori, it is a hypothesis that
> > he tested, and I tested; we both found it a tenable
> > hypothesis.
>
> This is relevant *only* if the detailed results of your
> testing are compatible in the sense explained above. If you
> arrive at significantly different reconstructions, you
> demonstrate only that at least one of you is significantly
> wrong (and that something is wrong with the associated
> testing or its interpretation).


***
Patrick:

Agreed. One of our reconstructions is less likely than the other; but they
both still support the hypothesis that Egyptian, Sumerian, and PIE are
related.

Can you not see that?

***
>
> >> However we have agreed that it is the quality of the
> >> argument itself and nothing else that matters, have we
> >> not? So if you and Bomhard are making the same argument,
> >> it amounts to _one_ argument, and one argument made no
> >> better simply because two share it. If on the other hand
> >> you and Bomhard make different arguments, they cancel
> >> each other out in so far as they differ.
>
> > You are really dense.
>
> > If a start with Egyptian d3, tremble', and compare it to
> > PIE *der-, 'tremble', it does not matter one little
> > turdlet if I say the proto-form of the first consonant was
> > /l°/ or /z/ or /þ/ as long as I can show that whatever
> > proto-consonant I propose consistently shows up in
> > Egyptian as <d> and in PIE as *d.
>
> This obviously has nothing to do with the point that David
> was making. He was addressing your claim that two voices,
> yours and Bomhard's, agreeing that Sumerian is Nostratic
> constitute cumulative support for that hypothesis. You are
> now discussing internal evidence for your reconstruction,
> which is a different matter altogether.
>
> Brian
>
> ***
Patrick:

I am not discussing "internal evidence"; this is a false characterization.

I am showing that Bomhard and I agree that Sumerian should be included in
the same family as PIE and AA.

Respectfully, I must disagree. That is the point as I see it. I am, however,
surprised that you do not see it.

***