Language Endangerment: Ideology or Fact?

Feb 13, 2015 by

I have been asked by a former student to comment on the article by Katie Wudel “Languages Are Going Extinct Even Faster Than Species”, published in the Good magazine online. Contrary to the name of the magazine, the article is anything but good—it is a sloppy ideological rant, scant on facts but full of sensationalism, which does nothing but a disservice to the cause of language maintenance and revitalization. The author’s carelessness is evident in the second sentence of the article that refers to “notable linguist and philosopher Dr. John Whorter”—but his real name is McWhorter. In the days of copy-and-paste, mistyping someone’s name, especially when linking it to an article by the said person, reveals a rather unforgivable lack of care.

IndigenousSiberiaThe author’s shoddy fact-checking carries over to the core part of the article. Wudel’s star example of a dying language is Tuvan, which in reality is rather a success story in language maintenance! Before we take a closer look at its survival status, let’s consider Wudel’s claim that Tuvan is spoken by “a mere 235,000 nomadic shepherds in Russia—many of whom also speak English and Russian”. Leaving aside the figure for the moment, let me correct two other mistakes in this brief quote. First, the characterization of Tuvinians as “nomadic shepherds”, which was correct historically, is no longer appropriate, as about 45% of the Tuvinians live in urban areas and do not raise livestock. Second, I would be very surprised if all that many Tuvinians spoke English, especially as compared to ethnic Russians or other ethnic groups in the Russian Federation. There is hardly any danger that Tuvinians will adopt English instead of their indigenous language any time soon. Rather, a significant number of Tuvinians speak Chinese, especially those who live in China or Mongolia. The article to which Wudel links her discussion of Tuvan claims that this language “is threatened by encroaching global dialects, like Russian and Mandarin” (highlight mine), but we already know that copy-and-paste is not Wudel’s strong suit.

Let’s now go back to Wudel’s quote. Her use of mere in “a mere 235,000” implies that this is a relatively small number, yet this is about 25 times larger than the median size of a world’s language. Thus, Tuvan is actually a rather populous language by world standards. It is also a relatively large language/ethnic group by Russian standards: according to the Russian federal law titled “On the guarantees of the rights of indigenous small-numbered peoples of the Russian Federation”, “indigenous small-numbered peoples” are defined as those that have no more than 50,000 members. (In connection with Wudel’s sloppiness, I will also note that she does not cite any source for her figure of “235,000”, which does not match either the figure of 254,000 speakers listed in the Ethnologue, nor the figure of 242,754 speakers from the 2010 Russian population census. Or, for that matter, the figure of 264,000 total Tuvan speakers, given in the article linked by Wudel. The differences are not great, but still.) However, in an earlier post, I have argued that sheer size of a language does not guarantee its survival (e.g. Belarusian with its population of about 4,000,000 speakers is nonetheless endangered), nor does the small size necessarily spell doom for a language. So what are the grounds for my claiming that Tuvan is relatively safe?

Let’s consider some more figures from the 2010 population census. Ironically, most so-called “titular nationalities” in the Russian Federation, that is ethnic groups that have their own autonomous units, do not constitute a majority in their titular republics or regions. Tuvinians are one of the rare exceptions to this pattern, as they constitute a significant majority, 79.7%, in the Republic of Tuva. Compare the Tuvinian situation to that of the Khakassians, who constitute only 12% of the population in Khakassia, or the Buryat, who constitute 28% of Buryatia’s population. Even the Yakut, who are the largest ethnic group in their titular republic, constitute only 46% of the population there. Overall, the survival of the indigenous language is closely correlated with the proportion in the total population: the more significant a given group is in the total population, the more likely it is to keep its language. Thus, only 69% of Khakassians know their indigenous language, whereas 83% of the Buryat do; recall that the Buryat are more than twice as large a group within the population of their republic. Among the Tuvinians, who, as mentioned above, constitute a majority in their titular republic, a remarkable 99.7% speak the indigenous language! Is this really an illustrative example of a dying language? Clearly not.

kart_index03And what of the gradual encroachment of Russian? It is indeed true that the majority of Tuvinians living in Russia speak Russian: 84.8%. However, this figure must be put in perspective, so let’s again compare with the Khakassians and the Buryat. Among the Khakassians, 98.1% know Russian, among the Buryat, the corresponding figure is 96.3%. Clearly, the Tuvinians are able to withstand the massive russification better than other Siberian groups. In fact, among the 180 ethnic groups listed in the census, the only ones with a lower figure in the “knowledge of Russian” column are: the “English” (probably, meaning “British”), the Americans, the Vietnamese, the Chinese, the Hindi-speaking Indians—all clearly recent immigrant groups—as well as Central Asian Gypsies (but not other Roma groups), Chechens, and Telengits (another Siberian group, closely related to Tuvinians, see map on the left). In other words, the Tuvinians are in fact among the least russified peoples in the Russian Federation.

Predictably, knowledge of Russian is more common among the Tuvinians living in urban areas (90.5%) than those residing in rural areas (80.4%). The same pattern is found all across Siberia: the more urban a given ethnic group is, the more they speak Russian and the less they are able to maintain their traditional language. For example, the Khanty people live primarily in rural areas, and approximately one in two of them speaks the indigenous language, compared to the predominantly urban-living Mansi, of which only a quarter speak the indigenous language.

All in all, Tuvan is not a “dying language”, as Wudel depicts it, but a well-maintained one. This characterization is supported by the Ethnologue, which classifies Tuvan as an educational language. So what is it about Tuvan that makes the issue of its survival so vital? According to Wudel, “the answer might lie in an untranslateable term like khoj özeeri”. She explains the term as follows:

“To the Tuvan, it’s a method for killing a sheep. It also implies kindness, decency, and a very specific ceremony by which a family so gently and neatly kills and prepares a sheep, they’re able to wear their finest attire because they don’t spill any blood. Like language itself, the word is a way to measure the character of the people who speak it.”

I have challenged the view that the death of a language means the death of its attendant culture and “ideas” (see here and here), but it is worth repeating that this position is based on a mistaken confusion of where language ends and culture begins. Non-linguists, such as Wudel, all too often fall into the trap of ascribing cultural notions to language, especially to words. (Confusing “words” and “language” is a hallmark of “amateurish linguistics”.) Regarding Wudel’s example, it is worth noting that various groups have words that mean ‘slaughter’ but because of their cultural understanding of what constitutes a proper slaughter procedure, these words are more often than not used to designate just such a proper procedure: dhabīḥah in Arabic, shechita in Hebrew, and jhatka in Hindi are but three examples.

More generally, an entire genre of supposedly “untranslatable word” lists has been popularized in the social media; I have seen at least a dozen such partially overlapping lists. The problem with them, however, is that they are rarely accurate. For example, one such list that I have recently seen translated the Italian passeggiata as ‘idle walk with friends on the street or the waterfront at sunset’. However, this word means simply a walk or a stroll—and the rest of the description comes from the Mediterranean custom rather than the meaning of the word itself. When my colleague and I take a stroll through the Stanford campus in the middle of the day, discussing our work, it is also a passeggiata. Moreover, similar words meaning a stroll—but frequently used for the customary leisurely stroll with friends on the street or the waterfront at sunset—exist in other Mediterranean languages: paseo in Spanish, volta in Greek, and so on. If in a historical reprise of Rome’s former glory, Italy manages to take over Spain and Greece, and its language replaces the local tongues, passeggiata will be a perfect replacement for the corresponding Spanish and Greek words. And if English or Russian comes to replace today’s Mediterranean tongues, stroll or progulka, which currently do not imply the idleness or the sunset or the accompanying friends, would be used to designate the venerable cultural tradition of the Mediterranean world. Cultures, after all, do not change to accommodate what a language is best suited to express. Exactly the opposite is true: cultures have their ways of twisting languages to serve their needs, making up new words and dropping old ones that are no longer needed. Two hundred years ago the Russian language had a whole list of words to designate specific types of animals—bulls, cows, horses—based on their sex, age, and fertility; today, the Russians have èsèmeska ‘SMS-message’, mobil’nik ‘mobile phone’, and sebjaška ‘selfie’.

An even bigger problem with Wudel’s article is her lack of understanding of what drives language shift and language death. She writes: “Economic growth is by far the biggest source of language loss. Why wouldn’t it be? Businesspeople and customers need to talk to each other.” Economic and political factors are indeed important to the phenomenon of language shift, as is illustrated perfectly by Itelmen. But “businesspeople and customers need[ing] to talk to each other” does not necessarily lead to language shift. Throughout history, people have found ways to trade with other linguistic groups without abandoning their indigenous languages: lingua francas, pidgins, even using the language of one of the two trading sides have all served the purpose. Sometimes even “dead” liturgical languages such as Hebrew were called upon to serve as a “commercial idiom”. Thus, trade in and of itself does not spell doom for any language. Nor does modernity, capitalism, or Western values serve as a necessary condition for language death. Contrary to the impression created by the frequent reports in the popular media on the tragic loss of various languages (e.g. Livonian, a Finnic language once spoken in Latvia), language endangerment is not a modern-era phenomenon. After all, Latin, Persian, and Arabic made languages like Celtiberian, Mogholi, and Phoenician disappear from the map long before the invention of airplanes and the Internet.

One final error in Wudel’s article that I feel compelled to correct is her characterization of Yiddish as “a dying language, or at least not exactly a living one, mainly because it hasn’t been widely embraced by the realm of commerce, and people rarely write it down”. Surely, it never had “an army and a navy”, as Max Weinreich quipped nearly a century ago, but Yiddish is spoken today by 1.5 million people around the world, and its literary tradition is evidenced in numerous books and magazines and culminated in the Nobel prize awarded to Isaac Bashevis Singer (1978). May I suggest that Wudel take a trip to Williamsburg, New York or Bnei Brak, Israel to see the next generation of Yiddish speakers?


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: