Alarodian.

A few weeks ago Hippophlebotomist mentioned the “Alarodian hypothesis”; having looked it up, I thought it was intriguing enough to give its own post. Wikipedia:

The Alarodian languages are a proposed language family that encompasses the Northeast Caucasian (Nakh–Dagestanian) languages and the extinct Hurro-Urartian languages.

The term Alarodian is derived from Greek Ἀλαρόδιοι (Alarodioi), the name of an ethnic group mentioned by Herodotus which has often been equated with the people of the kingdom of Urartu, although this equation is considered doubtful by modern scholars. A leading Urartologist, Paul Zimansky, rejects a connection between the Urartians and the Alarodians. Nearly nothing is known about the Alarodians except that they “were armed like the Colchians and Saspeires,” according to Herodotus. The Colchians and Saspeires are generally associated with the Kartvelians and/or Scythians, neither of whom spoke a Hurro-Urartian or Northeast Caucasian language

Historically, the term “Alarodian languages” was employed for several language family proposals of various size. Sayce (1880) employed the name for a small group that comprised Urartian (then called “Vannic”) and the Kartvelian languages (Georgian, Laz, Mingrelian, and Svan). In 1884, the German orientalist Fritz Hommel further included all languages of the Caucasus and the ancient Near East which did not belong to the Indo-European, Semitic, and the now obsolete Ural–Altaic language families, e.g. Elamite, Kassite. Later, he extended the Alarodian family to include the pre-Indo-European languages of Europe, e.g. Lemnian, Etruscan, Ligurian. Karel Oštir’s (1921) version of Alarodian included all aforementioned languages, further Basque, Sumerian, Egyptian, the Cushitic and Berber languages. The historical Alarodian proposal – especially Oštir’s maximal extension – was however not well-received by the majority of scholars (“Ce petit livre donne le vertige”—”This little book makes one dizzy”, A. Meillet), and eventually abandoned.

The term “Alarodian languages” was revived by I. M. Diakonoff for the proposed language family that unites the Hurro-Urartian and Northeast Caucasian languages. Work by I. M. Diakonoff and Starostin (1986) asserted the connection between “Nakh-Dagestanian” (NE Caucasian) and Hurro-Urartian on the basis of comparison of their reconstruction to Proto-Nakh-Dagestanian, later published in 1994 with Nikolayev.

I like the “vertige” quote — Meillet! thou shouldst be living at this hour: linguistics hath need of thee…

Also, this gives me an opportunity to link to Dravido-Korean languages, a piece of weirdness John Emerson (who used to propagate Dravido-Everything around these parts) shared with me recently. Enjoy!

Comments

  1. One of the originators of Dravidian-Koreanic, per the link, was Homer B. Hulbert. Please tell me Nabokov had never heard of him.

  2. David Eddyshaw says

    Exactly what occurred to me!

  3. J.W. Brewer says

    Most of the tv shows I watch are Alarodian.

  4. David Eddyshaw says

    And some of my best friends are Alarodian!

  5. Keith Ivey says

    Alarodian! Just can’t wait to get Alarodian.
    The life I love is making music with my friends,
    And I can’t wait to get Alarodian!

  6. This reminds me that I really need to make an appointment with my urartologist.

  7. ‘E looks at ur ‘art, does ‘e?

  8. I love pie Alarode!

  9. Any connection to the Mandalorian?

  10. One of the originators of Dravidian-Koreanic, per the link, was Homer B. Hulbert. Please tell me Nabokov had never heard of him.

    Hulbert is a familiar name to many Koreans due to his involvement in Korea, including his opposition to Japanese rule as well as the study of the Korean alphabet (he was one of the advocates for introducing spacing and modern punctuation in Hangul orthography).

    But we usually refer to him by his surname only as 헐버트 Heolbeoteu (I couldn’t have told you what his given name was off the top of my head), so we don’t get that alliteration that would trigger Nabokovian associations.

    The Dravidian-Koreanic hypothesis may be abandoned in scholarly circles, but you can find plenty of information online (including several videos) playing up the purported connection between Korean and Tamil, mostly relying on chance similarities in vocabulary.

  11. Hippophlebotomist says

    I’m not much of a fan of Quora, but I do think Prof. Thomas Wier’s overview of the strengths and weaknesses of the proposal is worth reading
    https://www.quora.com/How-credible-is-the-hypothesis-that-Northeast-Caucasian-languages-are-related-to-Hurrian-and-Urartian/answer/Thomas-Wier

  12. David Eddyshaw says

    I agree. It looks very sensible.

    The “strengths” seem actually to amount to a few typological similarities (alongside some very notable dissimilarities, as Wier points out.) That means nothing much at all when it comes to establishing whether languages are genetically related.

    (Yoruba is certainly related to Zulu. Goemai is certainly related to Hebrew. Welsh is quite closely* related to Lithuanian. Amharic and Quechua, on the other hand, are not Altaic** …)

    * Much closer than Goemai to Hebrew or Yoruba to Zulu.
    ** In fact, Altaic is not “Altaic” either.

  13. I’m a bit surprised that someone may think that Chechens are “not widely known”. Especially someone based in Tbilisi. Or is he speaking of the name “Nach-Dagestanian”? Then “Sino-Tibetan” is not much better (or not. it is better because Sino-Tibetan speakers use similar names for it… and Sino-TIbeetan speakers are numerous).

    But I’m even worse.
    Actually, I can’t even remember when if ever I was able to find SUCH a gap in my education: I did not know that Tiflis is Tbilisi.

    I’m not sure everyone understands: the reason is that I know both words very well since very early (like: 3, 4 or 5) age. Of course you hear names of the capital of one countries that make up your own country (the USSR and Russian empire) that are only “not really foreign” countries where people are allowed to travel VERY often.
    And I understood all ennumerable texts I read as referring to some other Georgian town.

    Same level as believing that “Rome” and “Roma” are two different cities.

  14. i don’t know whether it will be a comfort, but i was present many years ago at the dorm-room shmooze moment when a 19-year-old from a family of professors, with a solid primary- and secondary-school education in a college town in the era before the rollback of sex ed, found out that her womb and her uterus were the same thing.

  15. rozele, another such story I know is about elephants hatching out of eggs.

    I think as with Tiflis: a factoid learned in very early age (someone’s joke?) and never questioned. (after all, who can know if humans do it as well?).
    Once you have learned who are elephants, it is not difficult to undertand that they do not lay eggs. But for this you need to think about it, and this is exactly what does not happen:) A different layer of data. You just know that elephants lay eggs.

  16. David Marjanović says

    The specific connections of Hurrian-Urartian with East Caucasian and Hattic with West Caucasian are no longer accepted in the Moscow School either; there’s a paper somewhere on A. Kassian’s academia.edu page that I don’t have time to look for.

  17. David Eddyshaw says

    Of course Hattic is not related to West Caucasian! It’s actually related to Chinese:

    https://www.academia.edu/459794/Hattic_as_a_Sino_Caucasian_language_UF_41_2009_2010_

    SinoCaucasian *apālxq̇wĔ ‘leaf’ is clearly cognate with proto-Oti-Volta *vâ̰-kʊ; *wV ‘thou’ with Kusaal , and **xq̇(w)VrV́ ‘old’ with Kusaal kudug “old.”

  18. David Marjanović says

    That’s not the paper I meant, but it contains this useful warning:

    There is an old comparison of Slav. *medv-ědъ ‘bear’ (< ‘one who eats honey’) and OInd. madhv-ád- ‘Süßes essend’ (said of birds in Rig-Veda). But despite the exact phonetic regularity it is hard to reconstruct such a compound for the Proto-IE level, since tatpuruṣa madhv-ád- is formed after a synchronically regular and very productive model and there are not any reasons to suspect a Proto-Indic stem here rather than an occasional word-forming in a poetic text. We see the same situation with some previously proposed Hattic–WCauc. etymologies.

    Algonquian whisky words without any such special situation as whisky.

  19. David Eddyshaw says

    The issue turns up a fair bit with proto-Oti-Volta too, what with all the languages concerned having extremely productive nominal compounding.

    It only seems justifiable to assign compounds to the protolanguage when they not only correspond phonologically but also have some particular meaning that is not simply the obvious sum of their parts.

    It’s difficult to be sure in some cases: Kusaal tammɛɛd “builder”, for example, is tan “earth” (the material) compounded with the agent noun of “build”, and the same elements turn up in the word for “builder” all over Oti-Volta (it’s actually there in one of the few etymologically transparent ethnonyms, that of the Batammariba, the speakers of Ditammari.) But then, what else are you going to build houses out of in scrub savanna?

    Come to think of it, it’s an interesting methodological problem. You could, for example easily construct a proto-Oti-Volta compound meaning “white goat.” All the languages regularly compound nouns with attributive adjectives as the normal unmarked construction, and POV surely did likewise. So POV certainly actually had such a form, and the equivalents in the modern language can be derived from it quite regularly by the finest approved Neogrammarian methods. When is it legitimate to ascribe absolutely identical developments in all daughter languages to POV, and when, instead, do we have to ascribe them to parallel development in every individual language?

  20. When
    (a) the protolanguage is likely to form compounds as well
    (b) this particular compound is culturally significant – and the culture that made it likely is likely be the culture of speakers of the PL
    then indeed continuous use of it since the time of the PL is quite likely. But the level of certainty is different as is the nature of the argument.

    PS, I’m basically saying something similar to what DE said in the fourth paragraph (posted after I wrote my comment).

  21. I’m – as always – annoyed by how the question is formulated in the Quora link: as if we are discussing whether they could both have descended from the Proto-World or Proto-out-of-Africa or from some other Paleolithic horror – the sceptical alternative being that “no, no, no and there were several Proto-Worlds” (which would be quite a claim, not sceptical at all)

    And as always, I think that thinking in such terms does harm. Because in reality it is not even “PL within some fixed number of millenia (say, 15)”, no. And not about the order of branching (which matters, but we need three, not two branches to speak of it).
    It is about reconstructions: being able to find some featrues or material that have been retained in both since they were the protolanguage if they ever were.

  22. David Eddyshaw says

    This is part of the perennial question, what exactly is it that we are reconstructing when we reconstruct a protolanguage? Some very capable historical linguists have adopted quite different positions on this.

    The Bantuists (or some of them) draw a distinction between “proto-Bantu” and “Common Bantu” wbich (I think) is supposed to address some of these issues:

    https://biblio.ugent.be/publication/7245993/file/7245996.pdf

    [The part about Guthrie.]

    Though this is a somewhat dufferent issue, I suspect that it is not really possible to make a neat division like this. “Raw” data are not actually intelligible without some sort of prior theory. The relationship has to be more like a kind of dialectic. (Denying that you even have a prior theory just means that you have a whole set of unacknowledged preconceptions, which you will nof be able to correct if you are unaware that they even exist.)

  23. David Eddyshaw says

    Though it is certainly greatly to Guthrie’s credit (quite apart from his other astonishing achievements) that he saw that there was a significant methodological problem, even if he ddn’t magically conjure up a watertight solution. Was there much of a precedent for that at the time?

Speak Your Mind

*