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 Historical content remains viewable.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

Etruscan

13 views
Skip to first unread message

Paul Morrison

unread,
Aug 10, 1996, 3:00:00 AM8/10/96
to

To any Etruscologists out there: I am curious why nothing (as
far as I can tell) came of Zacharie Mayani's work relating
Etruscan to Albanian. His book "The Etruscans begin to speak"
came out in 1961, and I found much of it quite compelling, but I
have seen no mention of it in the material I have access to.
Admittedly, sometimes he goes pretty far out on a limb, but there
may be more recent archaeological evidence which could be used to
prove or disprove parts of his reasoning. I don't suppose anyone
has found Claudius' dictionary yet... <g>

Any comments?

Paul Morrison

--
J. Paul Morrison

Miguel Carrasquer Vidal

unread,
Aug 11, 1996, 3:00:00 AM8/11/96
to

Paul Morrison <74720...@CompuServe.COM> wrote:

I found the book quite amusing, but nothing in it compelling.
Etruscan is not an Indo-European language, much less a kind of
distorted Albanian.

I will not discuss Mayani's final appendix ("Twelve Etruscan Proverbs,
Aphorisms and Drinking-Songs"). Let's just take one inscription that
has been more or less deciphered, and see what Mayani makes of it.

A mirror depicting Hercules sucking Juno's breast (apparently as an
adoption ritual). Inscription says:

"eca sren tva ikhnac hercle unial clan thra sce"
"this image shows how Hercules Juno's son became?"

The last words ("thra sce") are still unclear. Is it "thras-ce", with
"thras" meaning "to become" (or "to suck", as others have suggested)?
At the risk of making Etruscan too Indo-Europeanish, one might even
interpret <clan thra> as "adoptive son" (cf. Latin fili-aster?) and
<s-ce> as "became" (?).

Anyway, here is Mayani's interpretation:

eca = "here is", Latin "ecca, ecce"
sren = Albanian <she"roj> "to heal", but if that comes from Latin
<sanare> (as it no doubt does), then Albanian <sh-lodh> "to relax,
repose" is another possibility.
tva = Albanian <due> "to like, to want"
ikhnac = Albanian <idhnak> "irascible, wrathful"
hercle unial clan = "Hercules son of Juno", straight from mainstream
Etruscology.
thra = "milk", Albanian <tra> "melted butter", Macedo-Rumanian <dzar>
"whey".
sce = Latin <suxit> "sucks", Albanian <thith> "to suck".

"Here is [how] wrathful Hercules, son of Juno, likes to restore
himself: milk he sucks".

Mayani's method, which stays more or less within the bounds of reason
in this example (where the image helps to interpret the text, and the
text itself is not too bizarre), goes completely wild when these
conditions are not met, e.g. the inscription on an Etruscan missile:
"strevc", interpreted as "Stretch out like a wolf [running]!" or "Leap
like a wolf!". Sure, Albanian <shtrij> = "to stretch", and <ujk> is
"wolf"! Not to mention the missile which reads "thkikth" (nice
palindrome), [quoting:] "Thk" is Albanian <dzak> "blood, to bleed";
"ik" is "run!"; final "th" is "in". Thus: "Draw blood in your
course!".


==
Miguel Carrasquer Vidal ~ ~
Amsterdam _____________ ~ ~
m...@pi.net |_____________|||

========================== Ce .sig n'est pas une .cig


Hans Kamp

unread,
Aug 11, 1996, 3:00:00 AM8/11/96
to Miguel Carrasquer Vidal

Can someone explain the similarities between Etruscan and Albanian?

Hans Kamp.

Miguel Carrasquer Vidal

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to

Hans Kamp <hans...@introweb.nl> wrote:

>Can someone explain the similarities between Etruscan and Albanian?

Which similarities? Name one.

Stephen C Carlson

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to

In article <4ugjvu$f9d$1...@mhafc.production.compuserve.com> Paul Morrison <74720...@CompuServe.COM> writes:
>To any Etruscologists out there: I am curious why nothing (as
>far as I can tell) came of Zacharie Mayani's work relating
>Etruscan to Albanian.

Probably for the same reason why nothing came of the book I bought
in high school claiming that Etruscan is a degenerate from of
Sanskrit: it's bogus.

Stephen Carlson
--
Stephen C. Carlson, George Mason University School of Law, Patent Track, 4LE
scar...@osf1.gmu.edu : Poetry speaks of aspirations, and songs
http://osf1.gmu.edu/~scarlso1/ : chant the words. -- Shujing 2.35

Richard M. Alderson III

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to

In article <320EC7...@introweb.nl> Hans Kamp <hans...@introweb.nl> writes:

>Can someone explain the similarities between Etruscan and Albanian?

Easily: Both are (or were) languages spoken on the northern Mediterranean
coast.

End of similarities.
--
Rich Alderson You know the sort of thing that you can find in any dictionary
of a strange language, and which so excites the amateur philo-
logists, itching to derive one tongue from another that they
know better: a word that is nearly the same in form and meaning
as the corresponding word in English, or Latin, or Hebrew, or
what not.
--J. R. R. Tolkien,
alde...@netcom.com _The Notion Club Papers_

Miguel Carrasquer Vidal

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to

Hans Kamp <hans...@introweb.nl> wrote:

>Miguel Carrasquer Vidal wrote:


>>
>> Hans Kamp <hans...@introweb.nl> wrote:
>>
>> >Can someone explain the similarities between Etruscan and Albanian?
>>

>> Which similarities? Name one.

>You named (at leas I saw) some of them:

>Albanian Etruscan

>Sheroj Sren
>Due Tva
>Idhnak Ikhnac
>Tra Thra
>Dzak Thk

>Otherwise, I did not understand, what you were comparing.

*I* wasn't comparing, so help me, Zacharie Mayani was. I was merely
pointing out the absurdity of some of his claims. Reading the book,
one realizes that his method consists of:

1. take an Etruscan inscription
2. look up the words in an Albanian dictionary, using a fuzzy search
(d=dh=t=th=z, etc.; add vowels at will)
3. see if some sense can be put into the string of words you've found
4. if unsuccessful, break up the words into smaller pieces and goto 2.
5. if we still miss some words to complete the sentence, add a few
"borrowings" from Latin, Greek, Osco-Umbrian etc.
6. if we have a sentence, publish it, else goto 1.


OK, let's prove that Basque is related to Dutch. I swear I took a
sentence at random from an introductory text, and simply applied the
method outlined above:

Eguraldi ona dagoenez, dendarik denda joan gaitezke.

<egur> is Dutch "eekhoorn", squirrel (Old Germanic "aikur"). <aldi>
is "altijd", always. <ona> might be a preposition "om, op", on,
around. <da-go-en> is "dakgoten", roof-gutter, and <ez> is Dutch
"is", is. <dendarik denda> sounds onomatopaeic, like Dutch
"dender-de-dender", <joan> sounds like Dutch "gaan", to go, and
<gaitezke> is "geiten", dial. "geites", goats, with postfixed <-que>,
"and, too", borrowed from Latin. Thus: "The squirrel is always on the
roof-gutters, 'dender-de-dender' he goes, and the goats too."

I won't "decipher" the second sentence in the textbook, because it
contains the Basque phrase <behar nuke>, obviously a cognate of Dutch
"B.H's neuken", to fuck bra's: those Basques!

Of course, what the sentence really means in Basque is "Because the
weather is good, we can go shopping". Note that the "translation" of
<joan> was right on the spot! You can't always be wrong.
(Oh, and <behar nuke> really means "I would need").

John E Koontz

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to

In article <320EC7...@introweb.nl>, Hans Kamp <hans...@introweb.nl> writes:

|> Miguel Carrasquer Vidal wrote:
|> > I will not discuss Mayani's final appendix ("Twelve Etruscan Proverbs,
|> > Aphorisms and Drinking-Songs"). Let's just take one inscription that
|> > has been more or less deciphered, and see what Mayani makes of it.
|> >
|> > A mirror depicting Hercules sucking Juno's breast (apparently as an
|> > adoption ritual). Inscription says:
|> >
|> > "eca sren tva ikhnac hercle unial clan thra sce"
|> > "this image shows how Hercules Juno's son became?"

I thought the whole problem with Herakles was that he wasn't Hera's
son?

|> Can someone explain the similarities between Etruscan and Albanian?

Well, none, apart from being both obscure. That's Carrasquer's point.

----
John E. Koontz (koo...@bldr.nist.gov)

Disclaimer: Views and recommendations, express or implied, are my own, and
do not reflect the opinion or policy of my employers.

Hans Kamp

unread,
Aug 12, 1996, 3:00:00 AM8/12/96
to Miguel Carrasquer Vidal

Miguel Carrasquer Vidal wrote:
>
> Hans Kamp <hans...@introweb.nl> wrote:
>
> >Can someone explain the similarities between Etruscan and Albanian?
>
> Which similarities? Name one.

You named (at leas I saw) some of them:

Albanian Etruscan

Sheroj Sren
Due Tva
Idhnak Ikhnac
Tra Thra
Dzak Thk

Otherwise, I did not understand, what you were comparing.

Hans Kamp.

Anthony West

unread,
Aug 13, 1996, 3:00:00 AM8/13/96
to
>>Can someone explain the similarities between Etruscan and Albanian?
>
>Easily: Both are (or were) languages spoken on the northern Mediterranean
>coast.
>
>End of similarities.
>--
>Rich Alderson
>
Can anyone name the scholar who (1970-90?) attempted to
derive Etruscan from Hittite? I read his effort and
thought it slightly less nutty than most such moves;
his methodology had a professional cast and I think his
credentials were somewhat mainstream.

Strikes against him: he was arrogant, and in the business
of tying up genetic relationships for *all* poorly
attested early languages from the Classical period - not
just Etruscan.

In his favor (my notes, not his): (1) both Classical
historiography and the Lemnian sister-language of 500
B.C. point Etruscanists to the NW corner of Anatolia ca.
1000 B.C. The Anatolian branch of IE is an obvious
candidate for comparison. Anatolian languages diverged
early from the mainstem of IE evolution and are shot w/
non-IE vocabulary items, making the (mostly) non-IE
vocabulary of Etruscan less anomalous than a comparison
w/ Albanian would.

(2) Etruscan shows a synthetic, case-suffixed noun
system of typical IE proportions, with 2 or 3
declensions and 2 or 3 genders. Most of its case
endings compare easily w/ those of other evolved IE
daughter tongues - including the genitival /-l/, known
mostly from Anatolia. And its gender oppositions -
animate/neuter, w/ traces of masc./fem. - are also
Anatolian in style, I believ.

The way nouns are organized and related strikes me as
a more fundamental (genetic) property than any lexical
set. Thus, the fact that Etruscan used non-IE numbers
is not a genetic problem. Modern Japanese mostly uses
Sino-Tibetan numbers, not Altaic numbers. But Japanese
compares well structurally w/ Altaic, not at all w/
Sino-Tibetan.

The "lumpers" most in fashion today - Russia's Proto-
World theorists - do not group Etruscan closely w/
IE. The "splitters" say: nothing can yet be said about
Etruscan affiliations. Obviously, at least one group
of scholars is missing the boat. I wonder if perhaps
each is right about the other.

-Tony West aaw...@CritPath.org


Miguel Carrasquer Vidal

unread,
Aug 13, 1996, 3:00:00 AM8/13/96
to

aaw...@CritPath.Org (Anthony West) wrote:

>Can anyone name the scholar who (1970-90?) attempted to
>derive Etruscan from Hittite? I read his effort and
>thought it slightly less nutty than most such moves;
>his methodology had a professional cast and I think his
>credentials were somewhat mainstream.

Not sure this is the one you mean: Francisco Adrados, well-known
Spanish Indo-Europeanist, was investigating this. I don't know what,
if anything, came of his research. His theories on PIE itself are
slightly off-mainstream (he reconstructs 9 laryngeals), but otherwise
sound.

>In his favor (my notes, not his): (1) both Classical
>historiography and the Lemnian sister-language of 500
>B.C. point Etruscanists to the NW corner of Anatolia ca.
>1000 B.C. The Anatolian branch of IE is an obvious
>candidate for comparison. Anatolian languages diverged
>early from the mainstem of IE evolution and are shot w/
>non-IE vocabulary items, making the (mostly) non-IE
>vocabulary of Etruscan less anomalous than a comparison
>w/ Albanian would.

>(2) Etruscan shows a synthetic, case-suffixed noun
>system of typical IE proportions, with 2 or 3
>declensions and 2 or 3 genders. Most of its case
>endings compare easily w/ those of other evolved IE
>daughter tongues - including the genitival /-l/, known
>mostly from Anatolia. And its gender oppositions -
>animate/neuter, w/ traces of masc./fem. - are also
>Anatolian in style, I believ.

No traces of masc./fem. in Anatolian, but neither are there in
Etruscan, as far as we know.

>The way nouns are organized and related strikes me as
>a more fundamental (genetic) property than any lexical
>set. Thus, the fact that Etruscan used non-IE numbers
>is not a genetic problem. Modern Japanese mostly uses
>Sino-Tibetan numbers, not Altaic numbers. But Japanese
>compares well structurally w/ Altaic, not at all w/
>Sino-Tibetan.

>The "lumpers" most in fashion today - Russia's Proto-
>World theorists - do not group Etruscan closely w/
>IE. The "splitters" say: nothing can yet be said about
>Etruscan affiliations. Obviously, at least one group
>of scholars is missing the boat. I wonder if perhaps
>each is right about the other.

There seems to be a willingness, even outside the Russian Nostratic
camp, to admit the possibility of IE and Uralic being related (e.g.
Bjoern Collinder, the Uralicist, and J.P. Mallory in "In Search of the
Indo-Europeans"). A while ago, a short list of possible Uralic-IE
cognates was posted here on sci.lang, to which I replied the
following:

*I was reminded of a similar list I drew up for IE and Etruscan.
*There are some interesting parallels:
*
*
* IE Etruscan
*
* *-m -n (pron.) 'accusative'
* *-s -s(i) 'genitive'
*(Hitt.) *-l (pron.) -l(a) 'genitive'
* *-i -i 'locative/dative'
* *-nt- -nth 'pres. ptc.'
* *-to- -tha- 'pf. ptc.'
* *-kwe -c 'and'
*(Hitt.) *-ma -m 'and, but'
* *-bhi -pi 'for, at, ...'
* *me- (nom.) mi 'I, me'
* (acc.) mini
* *ko- i-ca 'this, that'
* *to- i-ta 'this, that'
* *'ag- ac- 'to drive, to do'
* *hanti- hanthe 'before'
* *leudh- laut-n 'free(dman), people, family'
* *nepot- nefts 'nephew, grandson'
* *sed- sut(h)-,sat(h)-'to sit, to set'
* *dei-n- tin 'day'
* *dei-u- tiu(r) 'light > moon, month'
* *doh- tur 'to give' (cf. Grk. do:-r-on)
* *dem-/dom- tmia 'house, temple'
* tham- 'to settle, to establish'
* *sahu-l- usil 'sun'

Some of the above may be borrowings, but for others borrowing is an
unlikely possibility, and we are faced with the option of taking them
to be cognates, or mere chance resemblances (same options for the IE -
Uralic list). The really interesting thing is that the Etruscan list
is nearly as long and as convincing as the Uralic list, despite the
fact that so little is known about Etruscan, and less about Lemnian or
Rhaetic. To me, this suggests that the closest relative of
Indo-European (if any, I should add, for the sake of prudence) is
Tyrrhenian (Etruscan-Rhaetic-Lemnian), rather than Uralic. Not
surprising, this, given the fact that I am rather unconvinced by the
Mallory/Gimbutas hypotheses, and would rather go along with Renfrew
part of the way [back to the Balkans to be precise, not into
Greece/Anatolia, where there be Tyrrhenians].

Loren Petrich

unread,
Aug 13, 1996, 3:00:00 AM8/13/96
to

In article <Dw1wy...@critpath.org>, Anthony West <aaw...@CritPath.Org> wrote:

>Can anyone name the scholar who (1970-90?) attempted to

>derive Etruscan from Hittite? ...

>In his favor (my notes, not his): (1) both Classical
>historiography and the Lemnian sister-language of 500
>B.C. point Etruscanists to the NW corner of Anatolia ca.

>1000 B.C. ...

>The "lumpers" most in fashion today - Russia's Proto-
>World theorists - do not group Etruscan closely w/
>IE. The "splitters" say: nothing can yet be said about
>Etruscan affiliations. Obviously, at least one group
>of scholars is missing the boat. I wonder if perhaps
>each is right about the other.

Try this one out for size:

The hypothesis that Etruscan is related to the North Caucasian
languages. This is proposed by some of those Russian lumpers, and the
details are to be found in one of Shevoroshkin's books on on Nostratic and
related subjects (I forget exactly exactly which one).

Also of interest is Diakonov and Starostin's book proposing that
Hurro-Urartian is related to the Northeast Caucasian languages.

So could Etruscan and Lemnian be related to North Caucasian and
Hurrian-Urartian?

The acid test of this hypothesis is how well it does on
interpreting the more obscure vocabulary, as well as how well it does
with the grammar.

--
Loren Petrich Happiness is a fast Macintosh
pet...@netcom.com And a fast train
My home page: http://www.webcom.com/petrich/home.html
Mirrored at: ftp://ftp.netcom.com/pub/pe/petrich/home.html

H. M. Hubey

unread,
Aug 13, 1996, 3:00:00 AM8/13/96
to

aaw...@CritPath.Org (Anthony West) writes:


>The "lumpers" most in fashion today - Russia's Proto-
>World theorists - do not group Etruscan closely w/
>IE. The "splitters" say: nothing can yet be said about
>Etruscan affiliations. Obviously, at least one group
>of scholars is missing the boat. I wonder if perhaps
>each is right about the other.

A Sufi/Zen master was out for a walk with his
students. They saw some protesters and some policemen
clubbing them. The disciples asked what they should do.
The Master said "Help the mental cripples". "But "
said the students "which ones are cripples?"
"They both are" said the Master "one group can't help but
enforce the law, no matter what, and the other group
can't help but protest it in the streets."

The "lumpers" and "splitters" also exist in paleontology.
And they have the same problem as the logicians. They
only accept two values, instead of a continuum. Most things
in life come in degrees. The lumpers, in this sense must
be closer to the truth.


--
Regards, Mark

http://www.csam.montclair.edu/~hubey hu...@pegasus.montclair.edu

T.T. Gerritsen

unread,
Aug 14, 1996, 3:00:00 AM8/14/96
to

Hans Kamp <hans...@introweb.nl> wrote:

[snip]

>Can someone explain the similarities between Etruscan and Albanian?

I'm not an Etruscologist, but I'm afraid there aren't many, since
Etruscan is certainly not an IE language. The examples Miguel quotes
are evidence that Mayani is the sort of dilettante "linguist" whose
derivations and comparisons lack any sort of systematic approach. In
proper Dutch: volksetymologie!


Tanno Gerritsen
T.Ger...@inter.nl.net


Hans Kamp

unread,
Aug 14, 1996, 3:00:00 AM8/14/96
to T.Ger...@inter.nl.net

Asking this question, I was only looking to the similarities between Etruscan and
Albanian words. It is not less important also to look al the meaning of the words. And
that is what I forgot.

Now I think, that I have to agree with you: "There are not many similarities between
Etruscan and Albanian words."

BTW, there are lots of lots of similarities between - for example - Dutch and Afrikaans.

Hans Kamp.

Peter k Chong

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

Dear fellow curious people,

I have read that Etruscan is probably related to the ancient Lydian
tongues of Asia Minor. Unfortunately, my source (Encyl. Britannia) didn't
back it up with any correspondaces and I stress the book said there is
only good reason to believe in a relationship. Other authors have
postulated that Etruscan might be an offshoot of Sumerian or a related
Turanian language like Magyar, Turkish or Khalkha Mongolian. However
there is something noteworthy in this theory. I was once shown samples of
the numbers 1-10 written in three scripts: Ancient Hungarian, Etruscan
and Roman. The similarity among all of them was quite striking.

The author of the book went on to say that the Hungarian system was
the oldest, the Etruscan one was borrowed from the Magyars and the Roman
one was a refinement of the Etruscan one. Of course this whole assumption
is based on the probable (I'm not kidding) placement of the Hungarians or
a related people in the Carpathian Basin since 1000 BC. Thus the
Hungarians are not a Finnish-type people but instead are more likely to
be one of the Scythian/Persian/Turanian equestrian peoples and settled in
the Carpathian Basin from Scythia before the founding of Rome.

The source of the ancient scripts can be found in Sandor Nagy's
book "The Forgotten Cradle of Hungarian Culture", Patria Pub. 1972. I
know this theory sounds fairly wild-ass considering that the few examples
of Etruscan Etruscologists have run across bear little to the Hungarian,
Turkish and Sumerian word lists I have and it seems only the numerical
script is related. NEway, I'd like to hear from someone's input...

Peter Chong


Loren Petrich

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

In article <4uugil$9...@news.inforamp.net>,

Peter k Chong <pet...@inforamp.net> wrote:
>Dear fellow curious people,
>
>I have read that Etruscan is probably related to the ancient Lydian
>tongues of Asia Minor.

That's probably from Herodotus claiming that the Etruscans had
come from Asia Minor. The Northeast Caucasian hypothesis of Diakonov and
Starostin would fit very well with that.

... Other authors have

>postulated that Etruscan might be an offshoot of Sumerian or a related
>Turanian language like Magyar, Turkish or Khalkha Mongolian.


... However

>there is something noteworthy in this theory. I was once shown samples of
>the numbers 1-10 written in three scripts: Ancient Hungarian, Etruscan
>and Roman. The similarity among all of them was quite striking.

Writing system != language!!! The Arabic numerals are used to
write numbers in *many* different languages -- *without* affecting their
pronunciation in *any* of them, with the exception of words for "zero",
which tend to be borrowed.

Stephen C Carlson

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

In article <ALDERSON.96...@netcom16.netcom.com> alde...@netcom.com writes:
>The problem with the Lydian connection for the Etruscans is that Lydian, like
>Lycian and a number of other minor langauges of Asia Minor, is demonstrably
>Indo-European, and in fact demonstrably Anatolian (that is, closely related to
>Hittite, Luwian and Palaic of 1000 years earlier).
>
>Etruscan is demonstrably *not* Indo-European.

Is this really the modern consensus?

Nostraticist Bomhard seems to think otherwise, that Etruscan is either
IE or closely related to it. It seems that our knowledge of Etruscan
is too scant to demonstrate much about Etruscan, but there are things
that point to Etruscan being IE. The major evidence against it is that
the numbers do not seem to be IE, but numbers can be borrowed.

Miguel Carrasquer Vidal

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

Peter k Chong <pet...@inforamp.net> wrote:

>I have read that Etruscan is probably related to the ancient Lydian

>tongues of Asia Minor. Unfortunately, my source (Encyl. Britannia) didn't
>back it up with any correspondaces and I stress the book said there is
>only good reason to believe in a relationship.

Lydian is a member of the Anatolian branch of Indo-European.
The best (and so far the only) reason to believe Etruscan is related
to Anatolian (Hittite, Luwian, Lycian, Lydian) is the genitive ending
in -l. There are some other similarities between the little we think
we know about Etruscan and what we know about Indo-European, but at
the same time more than enough dissimilarities to rule out the
possibility that Etruscan is in fact a daughter language of IE (or
Anatolian, I believe). See previous articles in this thread.

>I was once shown samples of
>the numbers 1-10 written in three scripts: Ancient Hungarian, Etruscan
>and Roman. The similarity among all of them was quite striking.

Gee, lemme guess:

I II III IIII V VI VII VIII VIIII X

(the V and X might vary a bit).

And please tell me that "Ancient Hungarian" is not the script of the
Tartaria tablets...

> The author of the book went on to say that the Hungarian system was
>the oldest, the Etruscan one was borrowed from the Magyars and the Roman
>one was a refinement of the Etruscan one. Of course this whole assumption
>is based on the probable (I'm not kidding) placement of the Hungarians or
>a related people in the Carpathian Basin since 1000 BC.

Ah good, it can't be the Tartaria tablets then (ca. 4500 BC).

> Thus the Hungarians are not a Finnish-type people but instead are more likely to
>be one of the Scythian/Persian/Turanian equestrian peoples and settled in
>the Carpathian Basin from Scythia before the founding of Rome.

"Before the founding of Rome" :-) Oh my God, it's all about
Transylvania again!

For the true and much more fascinating story on the Hungarians, see my
article "Re:California Indian Languages Ugric?" here on sci.lang not
so long ago.

Richard M. Alderson III

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

In article <4uugil$9...@news.inforamp.net> Peter k Chong <pet...@inforamp.net>
writes:

>I have read that Etruscan is probably related to the ancient Lydian tongues of
>Asia Minor. Unfortunately, my source (Encyl. Britannia) didn't back it up with
>any correspondaces and I stress the book said there is only good reason to
>believe in a relationship.

The problem with the Lydian connection for the Etruscans is that Lydian, like


Lycian and a number of other minor langauges of Asia Minor, is demonstrably
Indo-European, and in fact demonstrably Anatolian (that is, closely related to
Hittite, Luwian and Palaic of 1000 years earlier).

Etruscan is demonstrably *not* Indo-European.

>Other authors have postulated that Etruscan might be an offshoot of Sumerian


>or a related Turanian language like Magyar, Turkish or Khalkha Mongolian.

Magyar is a Finno-Ugric language (in fact, the "Ugric" in the term is derived
from the Russian word for "Hungarian"), shown to be related to Finnish in a
1752 monograph which predates the clear demonstration of the Indo-European
family by 34 years. The Finno-Ugric family is in turn a subfamily of the
Uralic languages.

The Turkic languages such as Osmanli Turkish, the Mongolic languages such as
Khalkha, and the Tungusic languages such as Manchu, make up the central core of
the Altaic family, with pretty good evidence for the inclusion of Japanese and
Korean as well.

It was at one time thought, without proof, that the Uralic and Altaic families
were closely related; that they are not has become clear as more research on
the question has been pursued.

Sumerian has no relation to *any* of the above.

Miguel Carrasquer Vidal

unread,
Aug 15, 1996, 3:00:00 AM8/15/96
to

scar...@osf1.gmu.edu (Stephen C Carlson) wrote:

>>The problem with the Lydian connection for the Etruscans is that Lydian, like
>>Lycian and a number of other minor langauges of Asia Minor, is demonstrably
>>Indo-European, and in fact demonstrably Anatolian (that is, closely related to
>>Hittite, Luwian and Palaic of 1000 years earlier).
>>
>>Etruscan is demonstrably *not* Indo-European.

>Is this really the modern consensus?

>Nostraticist Bomhard seems to think otherwise, that Etruscan is either
>IE or closely related to it. It seems that our knowledge of Etruscan
>is too scant to demonstrate much about Etruscan, but there are things
>that point to Etruscan being IE. The major evidence against it is that
>the numbers do not seem to be IE, but numbers can be borrowed.

If it were only the numbers...

Kinship terminology is non-IE. And it's not even Hittite (where
Hittite deviates):

The first 4 don't prove much, true:
father: apa
mother: ati
grandfather: papa
grandmother: teta

But these are more serious:
son: clan
daughter: s'ec
brother: ruva, sister: ???
wife: puia

The plural is in -r (clan, clenar). Definitely un-IE (OK, I must
admit clan/clenar has an awfully German ring to it:-)

The 3rd p. past tense is in -ce (turu-ce), and there is no sign
anywhere of typical IE endings such as -m(i), -s(i), -t(i), -nt(i).

The only personal pronoun we know is "mi", I, which actually counts
against the Anatolian connection (where I = ammu(k)).

Jacques Guy

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to

Stephen C Carlson wrote:

> Probably for the same reason why nothing came of the book I bought
> in high school claiming that Etruscan is a degenerate from of
> Sanskrit: it's bogus.

Has anyone made hay of the *striking* similarity between
Etruscan and Irish yet? viz clan "son".

All roads lead to Rome, they say. All comparisons lead to...
Proto-World! I await, with some trepidation, a proof that
Etruscan and Mayan are related (their speakers had great
big noses, and, since Luigi Cavalli-Sforza, as we know
that genes and languages are related, so there you are)

Jacques Guy

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to

Miguel Carrasquer Vidal, taking Hans Kamp to task about
similarities between Etruscan and Albanian:

> Which similarities? Name one.

Both are written in related alphabetical systems.
Both are or were spoken in Southern Europe.

I claim the prize!

Peter k Chong

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to m...@pi.net

Dear Mr. Vidal,
"MI" = I eh? For some funny reason, I find many words that denote a
first person pronoun or first person possessive construction in many
languages. EX:

MY, MINE, ME (English), MON, MA, MES (French), MEIN (German), ME, MI
(meanings are usually equivalent to "my" or "mine" in Romance languages)
ÉN = "I"; ENYÉM = "mine"; -M, -OM, -EM = first person
personal possessive suffixes cf. "My...", MI = "We" (Hungarian), MINä = I
(Finnish), MIN, ME, MAN' = "I" (Mari, Komi, Samojed), MON = "I" (Mordvin,
Udmurt, Lappish), MEN = I (Karchai, Kyrgyz), MIN = I (Yakut/Sakha), MAN =
I (Old Turkish), MäN = "I" (Uygur), BI = "I" (Mongolian tongues and
Tungusic tongues). Of course there's also AMMU from Lydian.

I admit I don't know other IE tongues or Semitic ones, but why the do we
find similar 1st person pronouns (with an "M" [sometimes "B"] or "N"
sound?) such as these betwenn some IE and Ural-Altaic tongues? Maybe
Nostratic kinship?

Puzzled

Peter Chong


Paul Morrison

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to

Mayani's book that I mentioned has 8 pages of them! I'll just
pick one I especially like. On a grave painting showing two
serving women in a hurry preparing for a banquet:
remiismethumfs (using th for theta)

Albanian:
re - new, again
mish - meat
me - with
tumb - cream (Etr. f often corresponds with b in other languages)

So it's a fair bet it means "another meat with cream". Mayani
says you could ask for that in Scutari, and they might not
approve of your taste in food, but they would understand you!

Simon Buck

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to

Ysgrifennodd Richard M. Alderson III (alde...@netcom16.netcom.com):

The problem with the Lydian connection for the Etruscans is that
Lydian, like Lycian and a number of other minor langauges of Asia
Minor, is demonstrably Indo-European, and in fact demonstrably
Anatolian (that is, closely related to Hittite, Luwian and Palaic of
1000 years earlier).

Etruscan is demonstrably *not* Indo-European.

How do you go about that then?

I can imagine that one might fail to show it to be IE, or show it to be
something else, or even show that it can not be shown to be anything (on
lack of evidence, say), but how do you show it is not IE?

------------------------------------------------------------------------
Simon.Buck @ Computing-Service.Cambridge.AC.UK
Gwasanaeth Cyfrifiadurol Prifysgol Caergrawnt, CB2 3QG, Y Deyrnas Unedig

Miguel Carrasquer Vidal

unread,
Aug 16, 1996, 3:00:00 AM8/16/96
to

Paul Morrison <74720...@CompuServe.COM> wrote:

>Mayani's book that I mentioned has 8 pages of them! I'll just
>pick one I especially like. On a grave painting showing two
>serving women in a hurry preparing for a banquet:
>remiismethumfs (using th for theta)

>Albanian:
>re - new, again
>mish - meat
>me - with
>tumb - cream (Etr. f often corresponds with b in other languages)

>So it's a fair bet it means "another meat with cream".

How much are you willing to bet?

The inscription next to it, according to Mayani, says "thrama
mlithuns" (*). I won't swear to it, but by the etymological method
that yields "a weight of honey", from Greek "drachma" and "melit-".
It is not too far fetched to think the Etruscans might have borrowed
these two words from Greek. "remiismethumfs", seems parallel: "thrama
mlithuns" <==> "re(e)miis methumfs". If we merrily follow along with
the honey-bee, and forget about that strange <f> in there, this might
contain <methu(m)>, "mead". "A *r(e)emiis* of mead". [I don't know
why Mayani puts one e in brackets].
We know from other inscriptions that "mead" was <math> in Etruscan.
<Math> vs. <methu> is well within the range of temporal/regional
variations that we find in Etruscan. The inscriptions Mayani
discusses are from Orvieto, 4rd/3rd century. I don't have the data
about where and when <math> was used for "mead", so we'll have to
leave it as a tentative hypothesis for now.
Anybody know of an antique liquid measure or vessel type sounding like
"re(e)miis"?


(*) Mayani, after conjuring up a cafetaria atmosphere with "Another
meat with cream!", translates this as: <thra ma mli, thunz>, "Milk
with honey, twice!" ("Deux steak frites, deux!")

Miguel Carrasquer Vidal

unread,
Aug 17, 1996, 3:00:00 AM8/17/96
to

Peter k Chong <pet...@inforamp.net> wrote:

The two possibilities are kinship or coincidence (we can safely
exclude borrowing here). What are the chances that a language will
have m, n, b or w in its first person singular pronoun (or affix or
whatever)? We select on the properties [nasal] and/or [labial] (what
I would call the "fuzzemes" {N}, {M}, {B}, {W}), excluding non-nasal
[dental], [palatal] and [velar], roughly the "fuzzemes" {T}, {S}, {L},
{R}, {Y}, {K} and {H}. There is also the zero fuzzeme {0}, because
every sound can also disappear. In this case, we expect a 4:8 chance
(one in three). But it's more, because many languages (but by no
means all) use different unrelated roots to express first person
singular, and so have more than one possibility of hitting a nasal or
a labial. Let's put the average at 2 roots for "I", giving a 50-50
chance overall.

Now let's put it to the test, using Campbell's "Compendium of the
World's Languages". The compendium is of course biased in that it
lists almost all Indo-European languages, and almost no Nilo-Saharan,
Australian or Papuan, for instance, but it's all I have.
My list is also biased in that the "nasal" forms are given in full,
whereas lack of them is rendered simply as --. [Note that I was
primarily looking for nasals, so I might not have counted all b's and
w's]


KHOISAN
!Kung mi
Nama --

NIGER-KORDOFANIAN
Kpelle nya', nga', me^
Mende ngi', nya', mu'
Bambara ne
Fulani mi, am
Wolof man, ma
Gurenne mam, ma, m-
Akan (Twi) me~, me~ye, mi~-
Ewe nye, me
Yoruba mo, ng, mi, e`mi
Igbo mU, m
Efik a`mi, n, ng, m
Proto-Bantu *ni
Swahili ni, -angu, mimi

NILO-SAHARAN
Maasai nanu
Nubian --
Dinka --
Shilluk --

AFRO-ASIATIC
Semitic *-(n)i
Berber n@kk
Hausa ni, mini, -na

BASQUE ni, na-

SUMERIAN nga
[Emesal: me]

ELAMITE --
DRAVIDIAN
Gondi (n)anna:, -o:n
Telugu ne:nu, na:
Kannada na:n(u), na:, -e:ne
Tamil na:n_, en_-
Malayalam n~a:n
[Second person is mostly *ni]

BURUSHASKI --

NAHALI ??
[no data]

URARTIAN (oblique:) me
HURRIAN --

KARTVELIAN
Georgian me, m-, mi
Svan mi
Mingrelian ma

ETRUSCAN mi, mini

INDO-EUROPEAN *me, *-mi
etc.

URALIC
Finnish mina"
etc.

YUKAGHIR met, -ng

ALTAIC
Turkish ben, -m
etc.

AINU kuani (?), en-
KOREAN na
JAPANESE --
[Korean and Japanese have suffered from "East Asian" pronoun-loss]

NIVKH n'i

CHUKCHI-KAMCHATKAN
Chukchi G@m (?)
Koryak gImmo (?)
[Very doubtful]

ESKIMO-ALEUT
Aleut thing (?), -ng
Inuit nanga, -nga

NORTH CAUCASIAN
Kabardian --
Abkhaz --
Ubykh --
Chechen --
Agul --
Avar --
Lezgi --
Dargva nu
Lak na

KET b-

SINO-TIBETAN
Yue (Cantonese) ngo5
Mandarin wo3
Tibetan nga
Ladakhi ng@
Tangut nga
Karen --
Garo anga, ang-
Naga (Ao) ni
Kachin ngai, nye
Burmese --

MIAO-YAO ??
[No data]

AUSTRO-ASIATIC
Mundari ing
Santali in', -n'
Khasi nga
Vietnamese --
Bahnaric inh, anh
Khmer --
Mon --
Niccobarese --
[Mon, Khmer and Vietnamese have suffered pronoun-loss]

AUSTRO-TAI
Thai --
Lao --
Zhuang --
Chamorro --
Ilocano --
Tagalog --
Cebuano --
Bugis --
Makassarese --
Malagasy --
Indonesian --
Polynesian --
etc.

TASMANIAN mi(na)

ANDAMANESE --

PAPUAN
Veri ne
[Second person is n@]

AUSTRALIAN
Western Desert Lg. -rna (?)
[Second person is -n]

NA-DENE
Haida --
Tlingit --
Chipewyan --
Apache --
Navajo --

SALISH
Squamish 'n, -(a)n

WAKASHAN
Kwakiutl --

ALGONQUIAN
Blackfoot nistoa, ni(t)-, no-
Cheyenne na-
Arapaho ne, ne', n-
Cree niya, ni-
Menominee na"-

SIOUAN
Dakota miye, ma, wa-
Crow wih-/bi-, ba-

IROQUOIAN
Seneca --
Cherokee --

ZUNI --

CHOCTAW ano (?)

MAYAN
Yucatec ten, en, in
Quiche in, nu
Mam n-

UTO-AZTECAN
Hopi 'ine, ne'-, ney
Nahuatl nehhua(tl), no, ni-, nech

OTO-MANGUEAN
Mixtec --
Zapotec --

SOUTH-AMERICA
Bribri --
Miskito --
Chibcha --
Chimu moin~, main~, -in~
Quechua n~oqa, -ni
Aymara naya (?)
Mapudungu in~che, n~i, -n~, -n
Guarani --
Tupi --
Carib --

I won't count the languages, because that would be meaningless anyway.
Looking at the genetic groups, Nilo-Saharan [the little data I have],
North Caucasian, Austro-Tai, Na-Dene and several Amerind groups score
well below average. But most other groups are scoring above average.
What does it mean? Proto-World? Not reconstructably. All human
languages are genetically related? Probably.

fred hamori

unread,
Aug 17, 1996, 3:00:00 AM8/17/96
to

In article 4v0cvp$k...@news.xs4all.nl,
m...@pi.net (Miguel Carrasquer Vidal) said:
Interesting, but apparently regardless of how little is known abut a language people still try
to force it to be I.E. in this group because the others are so unpopular. I am not trying to prove
what it is but only find it interesting that many of the words you listed are UralAltaic and maybe
there is something to the idea that the ancient language of Europe was not I.E. but U.A.

>If it were only the numbers...is non IE /might be interesting to list these again, since I missed it./

>Kinship terminology is non-IE. And it's not even Hittite (where Hittite deviates):

A lot of Hittite religious terms were also from early Messopotamian languages and not I.E.
Hatti after all preceeded it and was not I.E.

>The first 4 don't prove much, true:

>father: apa /Hungarian apa/
>mother: ati
>grandfather: papa /Bulgar Turkic baba =ruler,patriarch , also Scytian papa-ois/


>grandmother: teta
>
>But these are more serious:

>son: clan /Scythian colax comes to mind as the youngest son/
>daughter: s'ec /Ugrian ses =sibling, Sumerian ses =sibling, Hungarian ocs=younger brother/
>brother: ruva, sister: ???
>wife: puia /Hungarin p>f fele-(seg) =wife where seg is suffix only.


>
>The plural is in -r (clan, clenar). Definitely un-IE (OK, I must
>admit clan/clenar has an awfully German ring to it:-)
>
>The 3rd p. past tense is in -ce (turu-ce), and there is no sign

>anywhere of typical IE endings such as -m(i), -s(i), -t(i), -nt(i) /are you saying the past tense varies?
> /interesting that t, nt is called I.E. suffixes
/since T is also in Hungarian as past tense.

>The only personal pronoun we know is "mi", I, which actually counts

>against the Anatolian connection (where I = ammu(k)). /mi=we in Hungarian and U.A./

Again I did not interject to proove what Etruscan is, just put in my two cents worth with some extra information. It is however known that the Etruscans came to N.Italy by way of the Balkans, Hungary and so on .. before it was that of course. But there seems to be a near eastern Anatolian connection or origin to it.
For example Tarkinius name seems to be much like the Scythian Targitar-(us) or the Parthian, Turkic, Hungarian TARKAN, TARXAN, TARHAN,TORGYAN etc which were the leading lords. Naturally linguists need to give plenty of latitude to neighbooring cultural influences of such names and titles. I dont want to
get into the age old chicken and egg question now.
Fred H.


Alexander Pruss

unread,
Aug 17, 1996, 3:00:00 AM8/17/96
to

Peter k Chong <pet...@inforamp.net> wrote:

> "MI" = I eh? For some funny reason, I find many words that denote a
>first person pronoun or first person possessive construction in many
>languages. EX:

>MY, MINE, ME (English), MON, MA, MES (French), MEIN (German), ME, MI
>(meanings are usually equivalent to "my" or "mine" in Romance languages)
>ÉN = "I"; ENYÉM = "mine"; -M, -OM, -EM = first person
>personal possessive suffixes cf. "My...", MI = "We" (Hungarian), MINä = I
>(Finnish), MIN, ME, MAN' = "I" (Mari, Komi, Samojed), MON = "I" (Mordvin,
>Udmurt, Lappish), MEN = I (Karchai, Kyrgyz), MIN = I (Yakut/Sakha), MAN =
>I (Old Turkish), MäN = "I" (Uygur), BI = "I" (Mongolian tongues and
>Tungusic tongues). Of course there's also AMMU from Lydian.

>I admit I don't know other IE tongues

Polish: Ja, mo'j, mnie.

> or Semitic ones,

Hebrew: ANI
Aramaic: ANA

Alex.

Jacques Guy

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

Loren Petrich wrote:

> Try this one out for size:

> So could Etruscan and Lemnian be related to North Caucasian and ^^^^^^^

That must be a typing mistake.

*Lemurian*, not Lemnian.

Charles Rezk

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

m...@pi.net (Miguel Carrasquer Vidal) wrote:

>And please tell me that "Ancient Hungarian" is not the script of the
>Tartaria tablets...

[etc.]

>Ah good, it can't be the Tartaria tablets then (ca. 4500 BC).

What exactly are the Tartaria tablets?

just curious,
Charles Rezk
re...@math.mit.edu

Miguel Carrasquer Vidal

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

re...@math.mit.edu (Charles Rezk) wrote:

>What exactly are the Tartaria tablets?

They are three clay tablets found in Tartaria (Romania), dating, I
believe, from the Vinc^a culture period of the Balkans (5500-4000 BC).
The tablets are scribbled upon, and some of the same scribbles are
also found, if I remember correctly, on pots and figurines of the same
era. I also have a picture of a similar tablet dug up at the site of
Karanovo, Bulgaria.
Some have considered them typologically (or even genetically) related
to early Sumerian writing, others consider them to be simple tallies
or even random scribbles. They are probably older than the earliest
Sumerian writing known to us, so the Sumerian import theory has not
been very popular since the tree-ring calibration of radiocarbon dates
set Vinc^a at least a millennium back in time. Looking at the
Karanovo tablet, I don't think they look like random scribbles (for
one thing, the "text" is neatly divided into four lines, in the style
of the Cretan Linear A/B tablets). There are tallies there allright
(the first line looks something like III I <V-like sign> III), but it
looks like there are also signs for what is being counted. Nothing
recognizably pictographic, though, so we may just be dealing with an
at least partially phonetic script. Unfortunately, I don't think
there's a ghost of a chance of them ever being deciphered, unless lots
more suddenly turn up in a dig.

Miguel Carrasquer Vidal

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

Ivan A Derzhanski <i...@banmatpc.math.acad.bg> wrote:

>Miguel Carrasquer Vidal wrote:

>> CHUKCHI-KAMCHATKAN
>> Chukchi G@m (?)
>> Koryak gImmo (?)
>> [Very doubtful]

>What is doubtful about them? In Chukchi _gym_ is `I' and _gyt_ is `thou',
>so the _m_ : _t_ contrast is there. (It's also there in the plural, btw.)

You're right, I noticed the 2nd sg. forms only after the article had
been sent.

>> CHOCTAW ano (?)

>Hardly. Note that the _-no_ appears in the other personal pronouns also.

Yes, it wasn't on my original list, but then I thought I should
include it, fair is fair, if all I was counting was nasals and labials
in 1st person morphemes (the "statistical" approach). Generally, the
question mark (e.g. Ainu <kuani>? because of affix <-ku>, Aymara
<naya>? because of affix <xa>), indicates that this form was included
for "statistical" reasons only. (But I was wrong on Chukchi/Koryak,
and you can also add Itelmen (Kamchadal): <k@mma>, <k@zza>).

Some "w" forms I missed were Old Japanese <wa>, <ware>, Modern
Japanese <watasi>. Also, a single Austronesian language (Trukese) has
"I" = <ngaang> (and also <wu>, but that, like Hawaiian <wau>, looks
like it's derived from *kau (maybe something like *o-?au or *u-?au),
so I didn't count it: I never had a head for statistics...).

Ivan A Derzhanski

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

Miguel Carrasquer Vidal wrote:
> Peter k Chong <pet...@inforamp.net> wrote:
> >[...] why the do we find similar 1st person pronouns (with an "M"

> >[sometimes "B"] or "N" sound?) such as these betwenn some IE and
> >Ural-Altaic tongues? Maybe Nostratic kinship?[...]

> Now let's put it to the test, using Campbell's "Compendium of the
> World's Languages". The compendium is of course biased in that it
> lists almost all Indo-European languages, and almost no Nilo-Saharan,
> Australian or Papuan, for instance, but it's all I have.
> My list is also biased in that the "nasal" forms are given in full,
> whereas lack of them is rendered simply as --. [Note that I was
> primarily looking for nasals, so I might not have counted all b's
> and w's]

Then you may have missed some cognates. The Nostratic idea seems to be
that *_mi_ is a Nostratic 1sg pronoun, which made it into the IE, Uralic,
Altaic and Kartvelian languages (the shift _m_ > _b_ being predictable in
Altaic), but not the Dravidian or the Afro-Asiatic ones. The Semitic _n_
is thus not related to the _m_ found elsewhere; none the less, some other
_n_s are, or may constitute a second bunch of cognate forms.

The questions are, then, (1) why _m_ turns up so often in non-Nostratic
languages, (2) why _n_ does so too and (3) if there can be any relation
between the two.

> CHUKCHI-KAMCHATKAN
> Chukchi G@m (?)
> Koryak gImmo (?)
> [Very doubtful]

What is doubtful about them? In Chukchi _gym_ is `I' and _gyt_ is `thou',
so the _m_ : _t_ contrast is there. (It's also there in the plural, btw.)

> CHOCTAW ano (?)

Hardly. Note that the _-no_ appears in the other personal pronouns also.

--
"mIw'e' lo'lu'ta'bogh batlh tlhIHvaD vIlIH [...]
poH vIghajchugh neH jIH, yab boghajchugh neH tlhIH"
(Lewis Carroll, "_Snark_ wamlu'")
Ivan A Derzhanski <i...@banmatpc.math.acad.bg, i...@cogsci.ed.ac.uk>
Dept for Math Lx, Inst for Maths & CompSci, Bulg Acad of Sciences
Home: cplx Iztok bl 91, 1113 Sofia, Bulgaria

H. M. Hubey

unread,
Aug 18, 1996, 3:00:00 AM8/18/96
to

m...@pi.net (Miguel Carrasquer Vidal) writes:

>They are three clay tablets found in Tartaria (Romania), dating, I
>believe, from the Vinc^a culture period of the Balkans (5500-4000 BC).

Doesn't this date and place put the tablets in the putative homeland
of the great PIE folks?

Why not try to see if anything matches any of the reconstruced *PIE?

Is there enough of the script to produce a few sentences?

Miguel Carrasquer Vidal

unread,
Aug 21, 1996, 3:00:00 AM8/21/96
to

hu...@pegasus.montclair.edu (H. M. Hubey) wrote:

>m...@pi.net (Miguel Carrasquer Vidal) writes:

>>They are three clay tablets found in Tartaria (Romania), dating, I
>>believe, from the Vinc^a culture period of the Balkans (5500-4000 BC).

>Doesn't this date and place put the tablets in the putative homeland
>of the great PIE folks?

Well, not according to Gimbutas, who considers the "Old European
culture" (the highlight of which was Vinc^a) exactly the antithesis of
Indo-European. The Vinc^a towns, according to this, the most
generally accepted theory on IE origins, would indeed have suffered
destruction from invading "Kurgan" horsemen, who introduced IE in the
Balkans. Now, I think Vinc^a *was* Indo-European (Pre- or
Para-Anatolian, I'd guess), so I would like nothing more than to know
how the tablets were sounded out...

>Why not try to see if anything matches any of the reconstruced *PIE?
>Is there enough of the script to produce a few sentences?

As I said in the preceding article: "Unfortunately, I don't think


there's a ghost of a chance of them ever being deciphered, unless lots
more suddenly turn up in a dig".

==

Miguel Carrasquer Vidal

unread,
Aug 21, 1996, 3:00:00 AM8/21/96
to

Ivan A Derzhanski <i...@banmatpc.math.acad.bg> wrote:

> [I wrote:]


>
>> Note that I was
>> primarily looking for nasals, so I might not have counted all b's

>> and w's.

>Then you may have missed some cognates. The Nostratic idea seems to be
>that *_mi_ is a Nostratic 1sg pronoun, which made it into the IE, Uralic,
>Altaic and Kartvelian languages (the shift _m_ > _b_ being predictable in
>Altaic), but not the Dravidian or the Afro-Asiatic ones. The Semitic _n_
>is thus not related to the _m_ found elsewhere

It might yet be. Apart from unadorned *mi/*me, a very common form is
*min/*men or *mini/*mene. This last form might easily be reduced to
*mni (compare Latin 'illu > Fr. il, Spa. e'l and il'lu > Fr. le, Sp.
lo), and then to *ni. This seems preferrable to positing *ni="I" for
A-A, especially since *ni="thou" (at least, it is in Basque (-n,
2sg.fem.), Dravidian (*ni), Uralic (*n-: Samoyed, Mansi-Xanty), Altaic
(*-n), Japanese (na), Sino-Tibetan (*na/*ni-), Athabaskan (*ni),
Nilo-Saharan (*ini), Tasmanian (ni), Andamanese (Ni) and Australian
(*Nin). Afro-Asiatic may itself have it in *'an-ta, but I doubt that,
in view of *'an-a:ku "I").

Paul J Z Kriha

unread,
Aug 24, 1996, 3:00:00 AM8/24/96
to

In article <4v0b25$j...@news.xs4all.nl>, m...@pi.net (Miguel Carrasquer Vidal) wrote:
>Peter k Chong <pet...@inforamp.net> wrote:
>
>>I have read that Etruscan is probably related to the ancient Lydian
>>tongues of Asia Minor. Unfortunately, my source (Encyl. Britannia) didn't
>>back it up with any correspondaces and I stress the book said there is
>>only good reason to believe in a relationship.
>
>Lydian is a member of the Anatolian branch of Indo-European.
>The best (and so far the only) reason to believe Etruscan is related
>to Anatolian (Hittite, Luwian, Lycian, Lydian) is the genitive ending
>in -l. There are some other similarities between the little we think
>we know about Etruscan and what we know about Indo-European, but at
>the same time more than enough dissimilarities to rule out the
>possibility that Etruscan is in fact a daughter language of IE (or
>Anatolian, I believe). See previous articles in this thread.
>
>>I was once shown samples of
>>the numbers 1-10 written in three scripts: Ancient Hungarian, Etruscan
>>and Roman. The similarity among all of them was quite striking.
>
>Gee, lemme guess:
>
>I II III IIII V VI VII VIII VIIII X
>
>(the V and X might vary a bit).

Yes, it wasn't just Romans who used to count on fingers, :-)
then draw the fingers, palms, and fists for numerics.

Our 'arabic' digits also developed out of old (Hindu ?)
pictograms of fingers.


>And please tell me that "Ancient Hungarian" is not the script of the
>Tartaria tablets...
>

>> The author of the book went on to say that the Hungarian system was
>>the oldest, the Etruscan one was borrowed from the Magyars and the Roman
>>one was a refinement of the Etruscan one. Of course this whole assumption
>>is based on the probable (I'm not kidding) placement of the Hungarians or
>>a related people in the Carpathian Basin since 1000 BC.

Tricky blighters these Magyars. They must have been hiding in the
woods for 2000 years. The local populations of many nations and tribes
knew nothing about them. For last several centuries before the appearance
of Magyars the Southern and Western Slavs were connected until they got
separated by wedges of Germanic tribes from the West and Magyars from the
East. So the Magyars were there all this time and suddenly before the end
of the 1st millenia AD they suddenly crawled out of woodwork.

Aphewmazing.


>Ah good, it can't be the Tartaria tablets then (ca. 4500 BC).
>

>> Thus the Hungarians are not a Finnish-type people but instead are more likely to
>>be one of the Scythian/Persian/Turanian equestrian peoples and settled in
>>the Carpathian Basin from Scythia before the founding of Rome.

And they managed to keep their horses well hidden as well. :-)


>"Before the founding of Rome" :-) Oh my God, it's all about
>Transylvania again!
>
>For the true and much more fascinating story on the Hungarians, see my
>article "Re:California Indian Languages Ugric?" here on sci.lang not
>so long ago.
>
>

>==
>Miguel Carrasquer Vidal ~ ~
>Amsterdam _____________ ~ ~
>m...@pi.net |_____________|||
>
>========================== Ce .sig n'est pas une .cig


Paul JK.

-- _
Planets appear to move slowly; Trees, in the winter, are bare; \__O_
Demons possess more than toasters; Socks behave strangely in dryers. (;)\
---- These are all things I believe in ---- _/ \_

0 new messages