Caucasian Albanian.

The topic of the ancient polity known as Caucasian Albania has come up a number of times here at LH (first, I think, in 2004), and I’ve always found it intriguing; now De Gruyter Mouton has published Caucasian Albania: An International Handbook and made it Open Access, so we can all enjoy it. The section of most direct LH relevance is The Heritage of Caucasian Albanian, including the following chapters:

Jost Gippert, The Textual Heritage of Caucasian Albanian 95
Jost Gippert and Wolfgang Schulze, The Language of the Caucasian Albanians 167
Wolfgang Schulze and Jost Gippert, Caucasian Albanian and Modern Udi 231
Igor Dorfmann-Lazarev, The Udis’ Petition to Tsar Peter 261

The other sections are Caucasian Albania in Foreign Sources, The Caucasian Albanian Church, and Architecture and Archaeology. The conclusion of “Caucasian Albanian and Modern Udi”:

As we have seen, many divergences between Caucasian Albanian and the modern Udi language can easily be explained as diachronic changes that were induced either by system-internal factors or by the influence of neighbouring languages, and Albanian may thus well be regarded as an ancestor of Udi. This implies that for the question of their affiliation with other East Caucasian languages, Albanian must be taken as the starting point. However, with the abandonment of class agreement, the introduction of a system of person markers, the abundant use of clause subordination including relative clauses, and many other features, Albanian had already moved away considerably from what can be assumed to have been the common linguistic basis of the Lezgic subgroup of East Caucasian before the translations of biblical texts that we find in the palimpsests were accomplished.

Hooray for Open Access!


  1. Hooray for open access indeed. Speaking of the un-Lezgic appearance of Albanian, I am intrigued by how Indo-European-looking it is at times: intriguingly, in one key respect (masculine-feminine-neuter distinction in pronominal elements) it is Indo-European-looking WITHOUT being Armenian-looking.

    So: why is this?


    2-Influence from an Iranian language which had preserved masculine-feminine-neuter distinction in pronominals?

    3-(Most intriguing to me): Influence from a pre-Classical stage of Armenian, when the masculine-feminine-neuter distinction in pronominals still existed?

  2. Ooh, intriguing indeed!

  3. David Eddyshaw says

    Avar and Dargwa have masculine/feminine/neuter gender systems …

    The Northwest Caucasian languages seem to have started out with masculine, feminine and three distinct “other” genders, so a collapse to M/F/N is not a big stretch language-internally once you move from an arbitrary “grammatical gender” system to a “natural gender” system à l’anglaise.

    Cross-linguistically, it seems to be unusual for gender systems not to take account of biological sex, though Niger-Congo rather skews the total numbers by contributing an awful lot of individual exceptions. I suspect you also have to take into account a tendency by compilers of grammars not to call noun-class-systems-with-agreement “gender” systems unless the system distinguishes biological sex somewhere.

  4. More on the distribution of gender marking affixes in NEC here.

  5. David Eddyshaw says

    Yes: when I say “Northwest”, I mean “Northeast” …

    (Fortunately, I can tell left from right. It was one of the first things they taught us in medical school …)

  6. I have a synesthetic sense for the secondary compass directions, based on stereotypes of US regions. NW is wet and dark green. NE is lighter green and sandy beaches and blue skies. SW is desert and chapparal. SE is… what remains. Bayous maybe? I’ve spent very little time there. Anyway, that helps remembering them.

  7. 1-Cross-linguistically, I am not certain that a masculine-feminine distinction is that much more common than an animate/inanimate one: in the Algonquian language family, animacy-based grammatical gender is a feature which can be reconstructed back to the proto-language, which remained alive and kicking in every Algonquian language (minus pidgins), none of which (except MAYBE Michif), tellingly, created a distinct feminine gender either (despite close contact with, for instance, Iroquoian languages, many of which DO have a feminine gender).

    Indeed didn’t someone here (?Lameen?) recently point out that a distinctive feminine gender is a distinctive typological feature of Eurasia and Northern Africa, which elsewhere is a very rare feature in grammatical gender systems?

    2-I can tell left from right, I am proud to say, but East versus West remains a bit of a challenge, as had already been established almost a decade ago on this thread:

    (See my 12:27 and 6:34 comments of June 30)

  8. David Eddyshaw says

    a distinctive feminine gender is a distinctive typological feature of Eurasia and Northern Africa, which elsewhere is a very rare feature in grammatical gender systems?

    It’s not so: just off the top of my head, Lavukaleve, Yimas, the Gunwinyguan languages, Dyirbal*, Arawan …

    In fact, apart from Niger-Congo (a pretty big “apart from”, I admit) and Algonquian, I’m having some difficulty thinking of arbitrary-grammatical-gender-style systems in which biological sex is irrelevant.

    Unless you mean by “feminine”, a noun agreement class consisting exclusively of female human beings? We may be at cross purposes. I’m also not thinking of systems like Sumerian or Kusaal with a human/non-human or animate/inanimate distinction entirely predictable from the real-world reference, only systems (like Algonquian) where the gender is (in at least some cases) “arbitrary”, at least from the semantic point of view (though perhaps predictable from the morphological shape of the word) and just has to be learnt.

    * Women, fire and dangerous things …

  9. David Eddyshaw says

    Also Eastern* Nilotic, which can’t really be counted as Northern Africa. Khwe has a feminine gender too, but I think that’s English-style-predictable from real-world reference, so it doesn’t count for this purpose.

    * Quick check, looks at hands. Which one do I write with? Yes, Eastern.

  10. David Eddyshaw says

    Hittite and pre-proto-Indo-European come to mind as sex-indifferent grammatical gender systems, though. (I think we’ve had this discussion before …)

    And Danish …. maybe it’s a Eurasian thing … along with Scandi-Congo, of course …

  11. Since we piled up on poor ol’ Grambank a while back, here is what it has to say on “Is there a gender/noun class system where sex is a factor in class assignment?” And cf. WALS, much sparser, but it marks non-gender languages on the same map.

    At a glance, sex-based gender is common in IE, Afro-Asiatic, NEC, Dravidian, Non-PN Australian, Sepik River basin, and lowland Amazonia. Rare elsewhere.

  12. Here is a more eaborate Grambank map, showing languages with sex-based gender (red), animacy-based gender (green), both (black), and neither (light blue).

  13. David Eddyshaw says

    The Grambank map mixes up two rather different things with its green spots:

    (a) cases where there is a grammatical gender system, but it doesn’t involve sex (Niger-Congo etc)
    (b) cases where there isn’t a grammatical gender system at all.

    Note in particular that classifier systems (all the green splodges on China) are not the same thing as a grammatical gender system; while, admittedly, it is not possible to draw a hard-and-fast line between the two, that does not mean that they are all the same thing. Needless to say, Alexandra Aikhenvald has written a whole book about this …

    The WALS map is more what’s called for, but misses quite a few cases that I can think of without even looking anything up. Sparse indeed. I think I’d have to say it’s just a beginning.

    [EDIT in honour of “more elaborate Grambank map”]

    Getting there …
    Not a Eurasian peculiarity unheard of elsewhere, anyhow.
    It looks like the geographical distribution is very much secondary to the great expansion of particular language families in the Old World, especially Indo-European, Afroasiatic, Niger-Congo, and Sinitic.

    I’ve often been struck by how persistent grammatical gender systems are diachronically. They obviously answer to a deep human need …

    English is one of a tiny minority of Indo-European languages to have ditched it altogether, and Welsh, which has gone far out of its way to make itself as unIndo-European as posssible, nevertheless still has grammatical gender. Giving it up would evidently be a step too far …

  14. David Eddyshaw says

    Mind you, I notice that three out of the ten most widely spoken languages in the world either never had or have abandoned grammatical gender, and that far and away the most widely spoken Oti-Volta language, Mooré, is also one of the minority of Oti-Volta languages that have abandoned grammatical gender.

    The moral is clear: abandon grammatical gender and conquer the universe! Imperial Radch!

  15. English is a minimally inflected language. I have sometimes wondered whether it is a coincidence that case and gender persist in English grammar essentially only in pronouns. There is a seemingly “natural” reason why biological gender could persist in pronouns; could that have contributed to the retention of separate subjective and objective case pronouns as well?

  16. David Eddyshaw says

    Pronouns are pretty often morphosyntactically peculiar. Yer typical Australian language has ergative alignment with nouns as arguments but nominative-accusative with pronouns; the Atakora Oti-Volta languages are SVO but put both direct and indirect object pronouns before the verb (and I believe there are some obscure European SVO languages which do the same.)

  17. Stu Clayton says

    the Atakora Oti-Volta languages are SVO but put both direct and indirect object pronouns before the verb (and I believe there are some obscure European SVO languages which do the same.)

    Spanish, French and (in subordinate clauses) German took cues from them, I suppose.

  18. David Eddyshaw says

    Of course.

  19. Stu Clayton says

    In this transgender age, the preposition of direct/indirect object pronouns is one of the few things left that one can get upset about without hurting any feelings.

    On the related subject of postpositioning, here is the first corny high-brow mansplaining German pun that I learned in the early 70s:

    F: Was ist die juristische Definition von Homosexualität?
    A: Die Hintansetzung des eigenen Vorteils.

  20. Stu Clayton says

    Hintansetzung (“hinten ansetzen“) is a word that today you would probably hear, if at all, only as an item of legalese. It means “postponement” or “assignment of lesser importance in the given context”. Hintansetzung des eigenen Vorteils means “leaving aside (for the moment) considerations of how one could benefit” (there’s a shorter way of saying that in English, but words fail me).

    However, as already indicated, Hintansetzung can be understood as “positioning something right behind (something else) (and touching it)”, and Vorteil as “the front part”.

    So often the case with a foreign pun, it’s funny only when you don’t need to have it explained.

    As with the saying about the hedgehog and the quiver, you have to know what it means and also know what it means.

Speak Your Mind