Obituary: The Ubykh Language

Jan 25, 2012 by

While marking a birth of a natural language is a difficult matter (because of the dialect-language problem), a language death is often a more conspicuous event. Case in point: a language called Ubykh (Oubykh, in French sources). This language died nearly 20 years ago, on October 7, 1992, when its last native speaker, a man named Tevfik Esenç (see photo on the left) died in Turkey.



Ubykh belonged to the Adygean branch of the Northwest Caucasian language family (see chart). So how come its last known speaker ended up in Turkey? This is part of the tragic story of Ubykh. Until 1864 the Ubykh people lived along the eastern shore of the Black Sea in the area of Sochi, north-west of Abkhazia. It is difficult to estimate the number of Ubykhs in the middle of the 19th century, but some Russian sources have given a figure of 40,000-50,000. However, they were not ethnically homogeneous but were split up into a number of tribal communities. Until 1830, the Ubykhs did not have any serious clashes with the Russian Empire. However, the Adrianople treaty of 1829, in which Turkey ceded to Tsarist Russia all rights to the Black Sea coastal area of the Caucasus, opened the way for the Russians to attempt a conquest of the western Caucasus. Although the Russians occupied a few points along the coast of Abkhazia, they were eventually forced to abandon the campaign due to lack of familiarity with the local terrain and the fierce resistance by the Ubykh and other Northwest Caucasian groups in the region.

This resistance (some now even call it a “liberation movement”) continued throughout the 1830s, 1840s and 1850s, which peaks in the spring of 1840, when a terrible famine caused by a poor harvest and the loss of their cattle due to a particularly hard winter, and in 1857, when at the end of the Crimean War, the Tsarist forces resumed their conquest of the Caucasus. By 1861, the “elected elders” of the Ubykhs, Shapsugs and Abdzakhs decided to add political and diplomatic measures to their military fight. Prince Gagarin, the Kutaisi governor-general gives the following description of the situation:

“The Circassians have not lost either their heads or their hearts. On the contrary, they have decided to fight for their independence not only with arms, but with an energetic appeal to foreign powers. Whereas, of necessity, the main role in the matter of armed resistance has fallen on the Abadzekhs, the Ubykhs, who are no less enthusiastic over their joint cause, have taken the administrative and diplomatic initiative in keeping with their determination…”

He then proceeds to describe a letter from the “Circassians” (cover term for the Northwest Caucasian ethnic groups) to the Consul of Great Britain in Sukhum; part of the Circassians’ plan was apparently to appeal to international public opinion in order to receive aid from states hostile to Tsarist Russia. The defeat of Russia in the Crimean War was fresh in the memories of the Ubykhs and it strengthened their hopes of foreign intervention. It is very characteristic that the Ubykhs approached not the Turkish government, but England, realizing presumably the advantages of the British Empire over Turkey. However, their plans did not come to fruition as the letter fell into the hands of the Russians (that’s why Prince Gagarin is able to quote from it).

By February 1864, the Ubykhs were surrounded by a tight ring of Russian forces and on March 6, 1864, they ceased resistance. The Ubykh elders opened armistice negotiations with the Russians, who demanded, in accordance with the wishes of Tsar Alexander II, that

“those who wished to go to Turkey should assemble in encampments on the sea coast at the mouth of the rivers Shakhe, Vardane, and Sochi where Turkish ships may come. Those who wish to join us should begin at once to move to the Kuban where land will be allotted to them.”

During March 1864, in the angry glare of burning villages some 30,000 Ubykhs made their way to the coast to go to Turkey. A few families were removed to the Kuban and were later resettled in the Kostroma province. This same fate befell other Northwest Caucasian peoples: during the period of 1859-63, the Shapsugs, Natukhais and several other tribes were expelled, and in 1865 the Chechens, Ingush, Ossetians and Karachai-Balkars followed. On May 21, 1864 , the Tsar’s governor-general in the Caucasus, the Grand Duke Michael, reported to St. Petersburg “the end of the Caucasian war”.

Forty years after the expulsion of the Ubykh, the man to be the last speaker of the language, Tevfik Esenç, was born in the village of Haci Osman in Turkey, where most Ubykh speakers found themselves after the 1864 expulsion. It was there that he was raised by his Ubykh-speaking grandparents, served a term as the muhtar (mayor) of that village, before receiving a post in the civil service of Istanbul. And it was there that he was able to do a great deal of work with the French linguist Georges Dumézil to help record the Ubykh language. Luckily, Mr. Esenç was blessed with an excellent memory and he understood quickly the goals of Dumézil and other linguists who came to visit him. During his lifetime he had witnessed the decline in the use of the Ubykh language: the need to speak Turkish to be understood, plus competition from other Caucasian languages, made a knowledge of Ubykh useless. By 1980s, only Mr. Esenç had complete mastery of the tongue, though four or five other tribal elders could still remember some phrases.

To preserve as many scraps of the dying language as possible, linguists have taken Esenç to Oslo and to Paris, where he has been four times. Others have trooped rutted tracks to the farm village of Haci Osman where the last of the Ubykh speakers lives in a hut with a dirt floor. Mr. Esenç became the primary source of not only the Ubykh language, but also of the mythology, culture and customs of the Ubykh people. He spoke not only Ubykh but also Turkish and the Hakuchi dialect of Adyghe, allowing some comparative work to be done between the two languages. To elucidate some of the puzzling features of the language, Mr. Esenç even allowed himself to be X-rayed while articulating. One interesting issue raised by the necessity of working with just one speaker of the language is whether his way of speaking is representative of the language in general or is peculiar to him alone. In the case of Mr. Esenç, it turned out that he was a purist, and therefore his idiolect of Ubykh (i.e. personal way of speaking) is considered by some as the closest thing to a standard “literary” Ubykh language that existed.

The most striking structural feature of Ubykh is its large consonant inventory, consisting of the whopping 81 consonant phonemes (according to John Colarusso “How many consonants does Ubykh have?”, in George Hewitt, ed., Caucasian perspectives. Unterschleissheim: Lincom Europa, 145-55). It is one of the largest consonant inventories in the world, and probably the largest outside the Khoisan languages in southern Africa (which use click sounds, not found in Ubykh). It has consonants in at least eight, perhaps nine, basic places of articulation, distinguishing for example alveolar, post-alveolar, alveolo-palatal and retroflex affricates and fricatives. It also distinguishes plain, palatalized and labialized stops and fricatives. Its sound inventory contains 29 distinct fricatives, 27 sibilants, 20 uvulars and 3 different l-sounds, more than any other documented language. Ubykh also has the most disproportional ratio of phonemic consonants to vowels (though analyses of different scholars produce different vowel phoneme counts). Thus, as John Colarusso remarked, “any rigorous account of human phonetic perceptual capacity will have to take into account this precious marvel, Ubykh”. And luckily some of it is now recorded on Georges Dumézil’s file cards, on tape and in those X-ray images of Mr. Esenç. (You can hear some sample Ubykh sound files here.)

Still, there is little chance of resurrecting Ubykh as a living tongue, scholars and native speakers agree. As Tevfik Esenç himself said in an interview, “Turkish authorities aren’t interested, and our own young don’t want to learn it.” His own three sons are incapable of carrying on a conversation in their father’s tongue (the ethnic Ubykh community now speaks a distinct dialect of Adyghe, according to the Ethnologue website).

Mr. Esenç died in 1992 at the age of 88. Some years before his death he had written the inscription he wanted carved on his tombstone of white marble:

“This is the grave of Tevfik Esenç. He was the last person able to speak the language they called Ubykh.”

Like this post? Please pass it on:

Previous Post
| Next Post

Related Posts

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below:


  • Dm

    To a degree, the Muhajir movement of the Circassians was a joint Russo-Turkish project, with the Turkish government being quite eager to resettle the warlike Caucasian tribesmen in its territory, often among non-Osman peoples such as Arabs. Turkish Muhajirin High Commission negotiated the timing and the flow of emigration with the Russians; the resettlement machine continued to operate through the Balkan Wars, when the main source of Muhajir migration switched to Bulgaria.

    BTW I don’t think I can post under my Google Id now; the Discus host asks permission to access my personal information in order to do so :(

    • Asya Pereltsvaig

      Thank you for the interesting comment on the Muhajir Commission. I am particularly wondering about the continued work of the resettlement machine through the Balkan Wars. It’s curious that this is one issue on which the Russians and the Turks manage to cooperate despite the continued animosity.

    • Asya Pereltsvaig

      Here’s our techie’s reply regarding your second question:
      By personal information it is probably just getting permission to use
      the reader’s name and photo from their Google account. That’s pretty
      standard, and harmless to grant permission. If the reader doesn’t want
      to let Disqus get that information from the reader’s Google account, I
      recommend making an independent Disqus account then.

  • Steven Lytle

    I submit that any mention of Ubykh should also mention Rohan Fenwick, and his involvement with the language. 

    • Asya Pereltsvaig

      Yes, thank you for bringing it up.

    • Asya Pereltsvaig

      Yes, thank you for bringing it up.

    • Asya Pereltsvaig

      Yes, thank you for bringing it up.

    • Asya Pereltsvaig

      Yes, thank you for bringing it up.

  • Anonymous

    If they have brought back extinct languages such as Manx, Cornish, and Prussian, cannot they bring back Ubykh?  It will take a lot of organization from dedicated people, especially from people who are of Ubykh descent.  But it can be done.

    • Asya Pereltsvaig

      In theory it’s not impossible to resurrect Ubykh. My point is that it’s not likely to happen, given that there is little interest in the ethnic community itself or from the Turkish goverment. Even when it comes to the revival of such languages as Cornish, I’d be careful. It very much depends on one’s definition of what it means to revive a language. There are hundreds of thousands of people who know a few basic words in Cornish; mere thousands could have a conversation in the language. Hundreds (or perhaps a few thousands at best) are considered fluent in Cornish. But as for native speakers — which is what it means to be a “live” language — there are none reported that I’ve seen.

      • Anonymous

        I notice where they came from in Russia, is where the city of Sochi is at.  Sochi is where the next Winter Olympics are being held.  I wonder if the Winter Olympics cannot inform people who the Ubykh are, and how their language died.  I wonder how you say Winter Olympics in Ubykh?

        • Asya Pereltsvaig

          You are absolutely right about the homeland of the Ubykh (and related Northwest Caucasian groups) being around the city of Sochi, where the next Winter Olympics is scheduled for 2014. There is a great deal of controversy surrounding these Games, only a small part of which has to do with the indigenous groups like the Ubykh. This is considered in some detail by my partner Martin W. Lewis in our second blog:

          There will be an additional post there tomorrow on other aspects of the Sochi Olympic controversy, so stay tuned! (I will post the link to it here tomorrow).

  • Alex McMillan

    This post is a bit incidental and better suited for the LIN10 blog, but I cannot find how to post there.
    As noted in the original post, at the time Ubykh was described, it seems it had a single fluent speaker (and he may have been the sole speaker for a number of years).  As noted, the language as described may well be the informant’s idiolect.
    * The Australian language Dyirbal, important for its system of noun classes, is currently spoken by about 5 individuals; perhaps there were more speakers when first described, but the number must have been small.
    Should a caution similar to the idiolect be attached to Dyirbal?  With a small number of speakers, would a language be more likely to drifts?  If that is the case, should less emphasis be placed on this extreme case (and more on slassifications in other languages)?  Is it responsible to title a book on such a basis?

    • Asya Pereltsvaig

      Thank you for your comments!
      I’ve added a link to this post on the Moodle site for LIN10, which has a much narrower readership than this blog. (P.S. Only I am set up to post there).
      As for Dyirbal, you are correct in surmising that a smaller language is more likely to drift. At least, all we know so far seems to indicate so. However, when first described in the 1960s, it was more populous than now, with the number of speakers in the hundreds, perhaps even in the low thousands. So it is certainly not the same as the case of Ubykh, with no danger of being an idiolect. As for titling the book with a reference to Dyirbal categorization system, “extreme sells”, to paraphrase a known saying. Still, categorization in even more familiar languages, with millions of speakers (e.g. Russian or Swahili) are no less bizarre!

      • Rohan Fenwick

        While it is true that a very large portion of the documentation on
        was done using Tevfik Esenç’s idiolect, texts from about two dozen
        speakers have been published, and when the bulk of the descriptive work was done in the ’50s and ’60s there were still around thirty Ubykh-speakers surviving in three separate speech communities. That’s why in my
        2011 Ubykh grammar I tried to move towards a more pan-idiolectic
        approach rather than prescriptively follow Esenç’s idiolect as Georges Dumézil and Georges Charachidzé did (though naturally a
        focus on Esenç’s Ubykh is obviously impossible to avoid entirely, especially given Dumézil’s unfortunate willingness to have Esenç revise and “correct” texts from other speakers).

        Also, not all of the recording took place in the ’50s and ’60s; substantial texts go back to the late 1920s, with smaller corpora back to the 1860s and even a handful of recorded words from circa 1650. This allows for some really nice observations of
        how change and drift occurred, which can illuminate how the earlier and presumably less “decayed” forms of the language may have differed. (Rieks Smeets did a great paper in 1997 looking at changes from the 1860s through to the 1980s in Ubykh absolutive plural-agreement marking strategies.)

        • Asya Pereltsvaig

          Thank you for your detailed and illuminating comments!

  • Naci Kartal

    hello. i am an ubykh (dechen) john colarusso said me : ‘Ubykh was like a castle high on a mountain, above all others.  No one else
    could know what went on within its walls.’ it’s an arcaic language.Ubykh language is the most advanced alphabet with 2 phonemic vowels and 84
    phonemic consonants. and he said: vI have argued that Circassian, Ubykh, and Abkhaz – Abaza form a language family,
    Proto-North West Caucasian, that is distantly related to Proto-Indo-European
    (Aryan).  This original family, which I have called Pontic, would have existed
    7,000 – 9,000 years ago. what do you think about that? thank you.

    • Asya Pereltsvaig

      Thank you for your comment, Naci.

      When one works on a given language, especially one with a history like that of Ubykh, it is easy to fall in love with it and use terms like “archaic”, “the oldest”, “most advanced” etc. But such superlatives have no meaning in linguistic theory.

      Besides that it is indeed true that Ubykh had the highest consonant-to-vowel ratio of all known languages, as it had a large number of consonant SOUNDS and only two vowel PHONEMES. It’s not a matter of alphabet though (“alphabet” refers to a certain kind of writing system, not sounds of a language). It is not the highest number of consonant sounds cross-linguistically, as Khoisan languages typically have a much higher number, due to rich inventories of clicks. (I’ve written about this elsewhere in this blog.)

      As for the Ubykh’s relationships with other languages, it is certainly related to other Northwest Caucasian languages, as you mention. As for its relationships with Indo-European languages, the evidence is rather murky, which is generally the case with any links that go beyond 6,000 years or so…

      • Naci Kartal

        I want to ask you something. I prepare about a video documentary about Ubykh Language. What do you know about  the historical background of the Ubykh Language? I need much more information. How did it appear first.? Is it Arian or not? It’s interaction with other languages or cultures. etc… it’s different from other north caucassion languages. We says Adyge all the North Circassians. But Ubykh is not Adyge. I wonder that was it always there or came from somewhere else? thank you.

    • Ali Berzeg

      Hi Naci. I would like to know more about Dechen clan in Turkey. Can you give me some information?

  • Metin Sönmez

    George Hewitt’s Recordings from Turkey (1974) including Ubykh with Tevfik Esench:

    • Asya Pereltsvaig


  • TimUpham

    Keeping the Ubykh language going:

    Faxie Adigege zeneynsuguere letuq’e. Weneynsu geywigie sq’eni pxiedik’uin gexueceneyl. Zesuebletheweguerege zepxiedik’ulherixe let, aq’egii gaq’uq’e. Wepxiedik’uin sengiafi pselhxu xleyk’ieq’esegii. Sigue simguicaq’esie sigiin gets’elhi adic’ienin sicik’iewt, q’egli alesq’eyt’. Aneygnsu wepxiedik’u zenibyewme ayniwewtin lheq’ewq’e. Acebgiewisin ak’iegi dgejuepsewne zeqhasiigueren giiwin wesuwe wezeq’ale alesq’e.

    Once upon a time, in Circassia, there was a certain young man. That young man, having become of an age to marry, was looking for a young woman. In a certain distant country, he had heard them say, there is a renowned young woman. That young woman remained unmarried, saying to whoever came to court her, I will marry him who knows what is in my heart without my telling him. The young man took himself off to the young woman to marry her. Getting on his horse, he went, and when night fell, he entered a certain village and stayed there.