Caucasian languages and Asia Minor

Wolfgang Schulze W.Schulze at
Thu Feb 4 11:03:32 UTC 1999

Rick Mc Callister schrieb:

>         Hattic, Hurrian & Urartian are often said to be N. Caucasian
> --which in modern times is almost always found on the north slope and north
> of the Caucasus.
>         I seem to remember reading that formerly, N Caucasian languages
> were spoken even farther to the north. [which obviously doesn't preclude
> them having a larger territory to the south as well].
>         It's possible that Kartvelian may have divided Hattic et al.
> [assuming they are NC] from N. Caucasian and pushed them to the west &
> south, perhaps even as far as Greece --where they are some claims that
> "pre-Pelasgian" languages may have been from the Caucasus. Kaska --just a
> name on the map as far as I know-- has also been claimed as Caucasian.

Let me first say this: The claim that Hattic and Hurro-Urartian should
beclassified as "North Caucasian" first presupposes that we do have something
like "North Caucasian". However, the genetic relationship between West
Caucasian (West- and East Circessian, Ubykh, Abaza, and Akbkhaz) and East
Caucasian (about 29 languages) ist far from being proven (if ever such a proof
is possible). Today, "North Caucasian" is strongly advocated for by S.L.
Nikolayev and S.A. Starostin (North Caucasian Etymological Dictionary, Moscow
1994 (Asterisk) [NCED]), and many people refer to this work just as THE final
word on this question. Yet, the NCED is full of methodological errors,
incorrect data, false integration of loan words etc. (see current reviews (e.g.
in _Diachronica_)). There are many others proposals concerning this question,
e.g. that East Caucasian once was located in the Kuro-Alazani region of (now)
Westerns Azerbajdzhan, whereas West Caucasian had a trong orientation towards
the Ponctic region (see Schulze 1998:169-186 and the references mentioned
there). From this it would follow that the "North Caucasian" hypothesis merely
results from the phantasies of the "trained comparativists [...] trained by
aficionados of the Nostratic school" as Dixon has put it (Dixon 1997:135, fn.

    But this is only the one side of the problem. Rick says that "N Caucasian
languages were spoken even farther to the north". What is this assumption based
on? Loan words? Anthroponyms? Place names etc.? Ethnonyms (a very dangerous
criterion, as everybody should know)? As far as I know, only variants of the
(now) northern West Caucasian languages (that is, Circessian) are liable to
have be spoken more nothernly (in the Eastern Pontic area). The northernmost
East Caucasian language, namely Chechen, had probably never been spoken north
of the Terek river (which is understandable if we assume that the speakers of
Nakh (Chechen-Ingush and Bac) probably immigrated from the South East (remeber
the Old Armenian name for "Chechen language" _naxch'amatyan_ which can perphaps
be read as chech. _nouxchiyn mout:_ ("language of the Chechen") as well as the
name for the region of Nakhichevan, perhaps more than a folk etymology often
referred to by Chechens).

    The question of  "a larger territory to the south" occupied by speakers of
"North Caucasian" has never been substantiated. The only thing that we know for
sure is that the southernmost East Caucasian language (Udi) once covered a
broader areal in Western Azerbajdzhan, perhaps down to Nagornyj Karabax. But
that' all. No place names, no (older) loans words in the southern languages
that could identified as "East Caucasian". The problem is surely linked with
the status of Hurro-Urartian. Diakonoff/Starostin 1986 have claimed that this
language cluster qualifies as East Caucasian (but see Schulze 1987). This claim
still is maintained in the NCED. But the qualitiy of the "proof" is VERY poor.
According to my opinion, the sound laws proposed by Diakonoff/Starostin are
rather ad hoc and very often based on a hapax legomenon; their comparative
treatement of the grammars of (actual!) East Caucasian languages and
Hurro-Urartian is a complete failure.

    Again we can observe this fatal trend in historical linguistics or
omnipcomparativism: A language isolate cannot be accepted; such a status would
be an offense or at least a challenge to the researchers. Every language in the
Ancient Middle East etc. has to belong to one of the major language groups
(with all the Nostraticism behind). Naturally every language has its history
and it is clear that every language has its ancestors. But why should this
ancestor be represenetd by one of the language groups we know of today?
Language diversity probably was much stronger in ancient times than nowadays,
and Asia Minor was a melting pot of the descendants of many such "ancestors".
There is no scientific need to put all the languages (or language fragments or
rumors on language fragments stemming from ancient sources via ethnonyms) into
one or two baskets.

    What I have said for Hurro-Urartian is even more valid for Hattic. Exepct
for one or two pseudo-etymlogies which have the qualitity of comparing English
_cat_ and Lak (East Caucasian) _ch:it:u_ ("cat", as all of you know a
wanderwort), the claim of Hattic as then "West Caucasian" is based on
typological aspects (prefix-agglutination, some kind of ergativity etc.). But
if we (errously) accept typology as a feature of genetic affiliation (which we
should NEVER do) it would be much better to compare West Cauacsian to the
Non-Pama-Nyunga languages of Australia. Though we cannot prove
"non-relationship" out of methodological reasons it seems rather unlikely that
Hattic has something to do with any kind of "North Caucasian".

    The less we know about a language the more it is likely that the language
is subjected to the kind of claims discussed above. This is especially true for
"Pelasgian". I never have seen any serious treatement of Pelasgian (or even
"pre-Pelasgian") elements by someone who has an explicit knowledge of East or
West Caucasian. Again, we have to deal with a rumor, an on-dit, already
established in the Early 20th Century by people who had to fill in gaps in
those times of romantic comparativism (which obviously realizes a strong
revival today).

    What has been said so far is also (and the more) valid for all assumptions
about "Kaska" (nothing but a place name!) or other so-called "languages" etc.
E.g., the Gutian language (bab. the peole _q'uti:m_;  midth of the 2nd
millenium in Babylonia) is sometimes identified as a ancestors of what now is
Udi, just becasue there are Aremian and Urartian sources that mention a region
_q'uti_ etc. in South East Caucasia. It is so simple to claim the loss of
initial *q'- in Udi (which by no means is a sound law in Udi) and then to
propose the match _q'ut'-_ / _udi_. All we know from Gutean is a list of 14
king names (and two or three very very doubtfull "loans" into Babylonian): None
of these king names has ever been identified as Udi (even only approximately)
[I confess I tried to, but it doesn't work].

To conclude: Whatever the linguistic situation in Asia Minor, Greece etc. has
been before or by the time of the arrival of IE speakers: Up to now there is NO
serious and scientifically substantiated evidence that North Caucasian
languages have ever been spoken in the area. It may well have been, who knows
[though I doubt]! But it is little helpfull to rely on rumors etc. without any
thoroughfull knowledge of West and East Caucasian, both snychronically and -
much more important - diachronically. Remember that Historical Linguistics with
respect to these two language groups still is in its infancy. It has the
quality of IE linguistics in the beginnings of the last century. We work hard
to improve the situation but it will take its time. Referring to historical
aspects of these languages today in order to subtantiate any claim of
affiliation to language isolates is nothing but (in a metaphorical sense)
nostratic science fiction. And this surely isn't a good basis for any serious
discussion of the major problem of IE urheimat.



Diakonoff, I.M. & S.A. Starostin 1986. _Hurro-Urartian as an East Caucasian
Language. Muenchen: Kitzinger.
Dixon, R.M.W. 1997. _The rise and fall of languages_. Cambridge: CUP.
Nikolayev, S.L. & S.A. Starostin. _A North Caucasian Etymological Dictionary_
Moscow: Asterisk.
Schulze, W. 1987. Review of Diakonoff/Starostin 1986. _Kratylos_ 32:154-159.
Schulze, W. 1997. Review of Nikolayev/Starostin 1997. _Diachronica_
Schulze, W. 1998. _Person, Klasse, Kongruenz. Fragmente einer
Kategorialtypologie des einfachen Satzes in den ostkaukasischen Sprachen. Vol 1
(in two parts): Die Grundlagen._
Muenchen: LINCOM Europa.

Prof. Dr. Wolfgang Schulze
Institut fuer Allgemeine und indogermanische Sprachwissenschaft
Universitaet Muenchen
Geschwister-Scholl-Platz 1
D-80539 Muenchen
Tel: 089-21802486 (secr.)
       089-21802485 (office)
Email: W.Schulze at

More information about the Indo-european mailing list