[Update. USA Today has published a story on Chulym, by Joann Loviglio of Associated Press, that wins David Harrison’s seal of approval. (Via Mark Liberman at Language Log.)]
K. David Harrison, of the Swarthmore Linguistics Department, claims to have discovered is working on a new previously undescribed language in Siberia. The Swarthmore press release quotes him as saying:

“We went looking for a language we weren’t sure even existed… It had been misidentified and falsely lumped together with other languages in Russia for convenience and political reasons, and we didn’t know if any speakers were left. No scientists had visited them in 30 years, and no one had ever recorded a single word of the language.”

It continues:

Harrison says the Chulym people continue to practice their ancestral lifeways of hunting, gathering, and fishing, but because of a variety of social, political, and demographic factors are now clearly losing their ancestral language. “They live in six small, isolated villages, often intermixed with a majority Russian population,” he says. “Only 35 people out of a community of 426 still speak it fluently, and we didn’t find any fluent speakers under age 52. The remainder of the Chulym have switched to speaking only Russian. It’s now considered a moribund language.”
Harrison says the unique Chulym number systems, grammatical structures, and classification systems may be lost with the language. Their highly specialized knowledge of medicinal plants, animal behavior, weather signs, and hunting and gathering technologies is also threatened. “Not least of all,” he says, “their rich pre-literate oral tradition, including religious beliefs, stories, and songs, will soon be completely lost, both to themselves and to science.” Harrison hopes to preserve some of that tradition by returning in 2005 to produce a grammar of the language and a children’s storybook.

(News story based on the press release here.)
Now, I’m a little confused, because the only references I can find to a Chulym people (Chulym is the name of a couple of Siberian rivers) are to the Chulym Tatars, who speak a Turkic language. Ethnologue’s entry calls it a Western Turkic language and says “Closely related to Shor; some consider them one language.” But their entry for Shor has it as a Northern Turkic language, but goes on to say “Some sources combine Shor and Chulym.” Either way, if it’s Turkic, it’s not some amazing new language with a unique worldview, so presumably these aren’t the Chulym Tatars but some other group. I look forward to reading more about it.
(Thanks to Bonnie for alerting me to this story.)


  1. Does each language have a unique worldview, Mr. Sapir-Whorf? Or is it only “amazing new languages” that have such worldviews? (That’s certainly what granting agencies want to hear.) How well does a linguistic description (in whatever theoretical framework you prefer) capture such a worldview? Do multilinguals have fractal worldviews? Does preserving the last specimen in formaldehyde save a species (and its unique worldview)?

  2. Rrowr. Damn, Joel, did that touch a nerve or something?
    I don’t think you have to buy into strong Sapir-Whorf to think that a language /might/ represent a unique worldview — or at least be very closely associated with it.
    It’s very easy to turn these notions into straw men, sure. And, as you say, granting agencies.
    But still: it’s really rather sad to think that, in another twenty years or so, nobody will be able to sing a lullaby in Chulym. That’s putting aside whether this is or is not a great loss to humanity, and completely avoiding the question of evitability.
    (Is it “less sad” if it’s one member of a widely dispersed group with several close relatives, and “more sad” if it’s an isolate? Well, it seems so to me.)
    Doug M.

  3. “All Things Considered” had a segment about this last night, but for some reason it’s the only story from that broadcast that can’t be listened to online …

  4. On the question of Turkic origins, I listened to the NPR segment twice last night (OK, I’m an NPR junkie) and the researcher said quite plainly that “they are a Turkic people.” So I don’t think he’s claiming that they aren’t related to other Turkic peoples; as I understood the researcher, he was claiming that they had been mistakenly lumped in with other geographically contiguous groups but that their language wasn’t as close to nearby languages as had been previously assumed.
    His basis for doing this is that he claims to be the first person to actually record and document this language (which he referred to as Middle Chulym, not just Chulym) — that is, he’s claiming that earlier ethnolinguistic conclusions about Middle Chulym were based on inadequate data. In that case, LH, it may be that Ethnologue’s entry that claims Chulym is “Closely related to Shor; some consider them one language” might be based on the Russian ethnolinguistic materials that this researcher is trying to challenge.
    The researcher also claims that because this group of Middle Chulym speakers was always rather small, it was lumped in with other nearby ethnic groups in a classification that was more convenient for the Soviet administration. This may be where the notion of the “Chulym Tatars” comes from — sort of like the PRC classifications of Uighurs, Ewenks, etc. (actually, especially Uighurs), which gloss over all sorts of ethnic differences.
    The Chinese classifications were invented by government ethnographers in the 1950s whose specific task was to reduce the proliferation of hundreds of self-described ethnic groups to a manageable size (fifty-something groups, if memory serves). It is my impression that the Soviet ethnographers were charged with a similar task.
    For this reason, it seems to me that we should be even more careful of Chinese and Soviet government-sanctioned ethnonyms than of ethnonyms in general. They represent just one possible reification of a complex system, and in this case it’s a reification which was intended for the convenience of government administrators, rather than as a reflection of people’s self-definition. I’m not trying to valorize one approach over another — any attempt to pin down the boundaries of ethnic groups is doomed to a certain amount of approximation, and usually the approximations are somewhat self-interested — but just encouraging us all to be aware of the motivations behind particular classifications.

  5. Huh. That’s interesting; I’m glad you heard the NPR segment and are able to give an informed analysis. (My wife heard it and told me about it when I got home, but the bastards didn’t rebroadcast it while I was there to hear.) But even assuming it was wrongly lumped in with Shor or whatever, I still don’t understand how an accurate description of a Turkic dialect justifies all the talk about “unique Chulym number systems, grammatical structures, and classification systems.” Turkic languages (apart from Chuvash) are much of a muchness, aren’t they? How could “Middle Chulym” be such an anomaly? Is he just scrounging for grant money and/or fame?

  6. Well if it’s a Turkic language with a numerical classification system it’s cool. My impression from cruising around grammars of Central Asian Turkic languages is that they are rather less similar than the Turkic nationalists would have us believe, if you look past the core Turkic vocab and the heavy Persian/Arabic/Russian superstratum.
    And so what if he’s scrounging for grant money? I assume you have some idea of just how hard it can be to get money to work on endangered languages …

  7. Oh, I wasn’t putting him down, believe me; I’m all in favor of his getting grant money. I was just wondering if the uniqueness was being hyped for that purpose. I actually don’t know very much about Turkic languages, so I’m pleased to hear that they differ more than they’re alleged to—it makes me more interested in studying them.

  8. Many thanks — I’m listening to it now!

  9. I’d suggest that this is (yet another) case of an over zealous journalist seeking to present some findings as world-shattering. Notice that when the linguist is quoted, he says that not a single word has been ‘recorded’, which might mean electronically rather than written down, and the linguist is never quoted that the language itself is ‘an amazing new language’. I tell you, if I had a chance to work with a language that apparently has had as little work done on it as Chulym, especially one that could be a ‘missing link’, I’d be doing multiple back flips, especially if I were an assistant prof looking for tenure. Quite easy for a non-linguist to jump to the conclusion that this is ‘discovering’ something that no one knew about.
    Here’s another url about Chulym Turkik, by a Russian grad student at Rice University.
    The name is reminds me of ‘Cuman’, a Turkic group that united with the Russians to fight the Mongols and, after being routed, settled in Hungary. Turkic languages are really difficult to sort out because they stretch from Siberia to Eastern Europe and they’ve had such a diverse history.

  10. Dear friends, Thanks to all of you for your interest in my research and in the Middle Chulym people.
    Please note that the words “new”, “discover” and “amazing” do not appear anywhere in the press release, nor would I ever make such a claim.
    Unfortunately several of the news reporters who interviewed me (or their editors) could not resist putting those kinds of words into the headlines, or even attributing them to me in the body of the article. It’s unfortunate, but the alternative of not talking to the press at all is also not so great. Endangered language research is an urgent but little understood priority that can benefit from public awareness and media exposure. As hundrends of languages go extinct, we are facing not only a humanitarian but also a scientific catastrophe.
    Middle Chulym (the native name is “ös”) is most definitely Turkic, and most Turkic languages are fairly closely related. It was previously wrongly lumped together (both in Russian bureaucracy and in Soviet era ethnography) with Shor, and later with Xakas, two neighboring but quite distinct Turkic languages. The Middle Chulym were even dropped from the census as a distinct ethnic group for over 40 years. They recently regained their ethnic status and registered as a ‘tribe’ with 426 members (only 35 to 40 people still speak the language fluently).
    The Middle Chulym [ös] language is unique and distinct enough from Lower Chulym (the next closest language) to warrant its own Ethnologue entry. I will be communicating with the Ethnologue editors shortly to make the case for this and to send them exact statistics on the number of speakers and the state of the language.
    I’m also going to publish the first book ever in the language later this year: Two hunting stories and a shaman story told to me by Middle Chulym elders and illustrated by their grandchildren (who were told the stories in Russian so they could understand them).
    I want to emphasize that this is a very poor and marginalized community, and I am working with them not only to collect data for my own scholarly research, but also to assist them in their own goals for language revitalization. Any grant monies I may receive to support his work in the future will also go in part towards that goal. I am privileged to have the full approval and close supervision of the Middle Chulym tribal council for my work.
    The full press release (and links to stories and soundfiles) may be found at:
    Thanks again for your interest. It’s been great to have so many people respond positively to the NPR piece and other news stories.

  11. Hey, thanks for dropping by! I apologize for any flippancy that may have offended you (not that you show any signs of it, but one never knows) – it’s just my habit. I’m all in favor of the study of little-known languages, and I should have realized (as joe tomei did) that the hype was entirely added by the reporter. At any rate, I look forward to seeing the book (assuming the NYPL acquires a copy), and I’m glad this marginalized community has found an advocate.

  12. Well, since I took the first prize for flippancy up above, perhaps I should explain a bit about where it comes from. I’ve done fieldwork in Papua New Guinea, which still has more unrecorded languages than any area of equal size on earth. The themes that always pop up when any journalist does a story about the area are (1) cannibalism, (2) the stone age, and (3) first contact–all of which imply a uniquely alien mindset on the part of the exotic ‘Other’.
    Even a foreign entomologist who came out to collect specimens at the village I was staying in got all excited when he saw an old man bring out a stone adze to finish the inside of his dugout canoe. Well, the man had already done the bulk of the work with his metal adze blade, but switched to the stone adze because its beveled edge gave a smoother finish.
    My other kneejerk reaction was: Linguists don’t save languages; native speakers save languages (or fail to). The most a linguist can do is perhaps help design an orthography, demonstrate that the language has its own strict rules, and instill some pride in a unique linguistic heritage. Just seeing one’s marginal vernacular language in writing can do wonders. Some folks are a little harder to please, of course, and want to see a dictionary, a grammar, or a whole Bible translation.
    I’m thrilled to see a professional linguist committed to real fieldwork. Of all the things Chomsky has done for or to linguistics, his denigration of fieldwork and mere “descriptive adequacy” has been most pernicious at a time when languages really are disappearing at alarming rates. Too many professional linguists (like bloggers!) do armchair research, leaving the SIL/Wycliffe Bible Translators to do the bulk of new field research.
    Disclosure: I completed a Ph.D. in linguistics, but didn’t pursue it as a career (nor did it pursue me). I have recorded a new language for the first time, but have never saved one.

  13. I’m thrilled to see a professional linguist committed to real fieldwork. Of all the things Chomsky has done for or to linguistics, his denigration of fieldwork and mere “descriptive adequacy” has been most pernicious at a time when languages really are disappearing at alarming rates.
    A hearty Amen!

  14. This entire discussion is entirely fascinating as non-linguist. The Universal Grammar hypothesis would pretty much indicate that no single language is really worth saving: it is merely a possible permutation of a universal paradigm. Hence Chomsky’s lack of interest in linguistic diversity.

  15. I’m wondering too about degrees of uniqueness. Every language is unique in one sense, but no language is more unique than another either. I am a unique individual, but it doesn’t mean that all of my traits can’t be found in many other individuals as well.

  16. Is really true that Chulym has 50 words for fermented mare’s milk?

  17. Ah, a question for Radio Erevan!
    –Is it really true that Chulym has 50 words for fermented mare’s milk?
    –In principle, yes, depending upon the age of the milk, the mare’s rank in the herd, the owner’s rank in the tribe, and the time of day in which it is best consumed. Under socialism, 50 additional words were coined to denote the lack of fermented mare’s milk under each set of conditions, thus greatly enriching the language. Under capitalism, the latter set of words is dying out, leaving the Chulym once again in lexical poverty.

  18. Offtopic, but this seemed like the best place to post this link, which I thought you guys would find interesting – it’s an interview with a linguist working on Amazonian languages. I’d make it a live link if I knew how.

  19. I can’t get to the New Scientist site at the moment, but there’s a discussion of the interview at Language Log.

  20. I got in here with no problem. It’s amazing how the “unique world view” topos lives on in these reports.

  21. Can anyone provide some evidence (e.g. quotes) of Chomsky’s supposed “denigration of fieldwork”? The MIT Linguistics department (faculty and students) has a rich and on-going tradition of linguistic fieldwork, most especially embodied by the late Ken Hale, for whose work Chomsky has expressed nothing but respect. The concerns about “descriptive adequacy” as one of the criteria for theory-construction have nothing (so far as I can see) to do with the art and value of linguistic fieldwork. The leap from Chomsky’s theoretical remarks to a presumed view of the value of descriptive work looks (without substantiation) dangerously similar to the kind of sloganeering and misplaced hype that started off this particular thread in the first place.

  22. Wow, conversation on this still bubbling along. I’ll take a shot at explaining why people feel that Chomsky’s approach is problematic towards endangered languages. Certainly, there are no quotes (though there is the apocryphal quote from the era of deep structure that we only need to describe the deep structure of English, though that’s not attributed to Chomsky, I think) but the problem that I see is Chomsky’s insistence that his linguistic research is completely unconnected to his political writings and activity. Endangered language research is undergirded by a commitment to empowering the speakers of those languages so that they can maintain their language in the face of overwhelming pressure. If linguistics as a field simply seeks to ‘explain’ language rather than take an active stance in supporting these groups, I think that it has failed as a discipline. I certainly admit that there are a lot of linguists (probably a lot smarter than me) who would scoff at this, but this is not an empirical question, this is a real world problem of how we change hearts and minds. If people in the field that is charged with studying languages feel that they have no call to protect those languages from disappearance, well, there’s not much hope. This is a ways from Chomsky, and I don’t think Chomsky should be held up for particular attack, but I think that we can agree that he does carry a certain amount of weight and his position must influence others.
    A second problem is that if linguistic departments do not recognize the theory neutral description of languages as worthy of recognition (i.e. for a PhD dissertation), they are creating a value system that devalues descriptive work, and creates a situation where it becomes more difficult to support work endangered languages. Chomsky has noted that the standard for linguistics is not ‘descriptive adequacy’ but ‘explanatory adequacy’ From Aspects (a web page quote, my copy is at my office)
    “On one level (that of descriptive adequacy), the grammar is justified to the extent that it correctly describes its object, namely the linguistic intuition–the tacit competence–of the native speaker. In this sense, the grammar is justified on external grounds, on grounds of correspondence to linguistic fact. On a much deeper and hence much more rarely attained level (that of explanatory adequacy), a grammar is justified to the extend that it is a principled descriptively adequate system, in that the linguistic theory with which it is associated selects this grammar over others, given primary linguistic date with which all are compatible. In this sense, the grammar unjustified on internal grounds, on grounds of its relations to a linguistic theory that constitutes an explanatory hypothesis about the form of language as such.”(p 26)
    To say that a grammar is ‘unjustified’ is rejecting the notion of a descriptive grammar as a valid research goal. It is that rejection that I find problematic.
    My understanding of Minimalism (and the rejection of Optimality), is that there is no ‘place’ for diversity, except when the language acts to disconfirm proposed theories. The logical result of the notion that Language is a ‘perfect’ system is that only one instantiation of language is needed.
    Needless to say, one can argue that all this occurs on a completely different plane from the concerns of endangered languages, but I think a plausible argument can be made relating them.

  23. Maybe I can try to clarify my previous posting, on the suspicion that some of the rancour comes from terminological confusion.
    Descriptive adequacy seeks an answer to the question: what (i.e., what is the way things are?). Explanatory adequacy seeks an answer to the qeustion: why are they that way (and not some other way). Chomsky’s early work opened a new way of looking at the “why” question, with an emphasis on empirical, rather than philosophical, approaches. Note that the “why” question can’t be asked until we have answers to the descriptive question: in the quote from Aspects, explanatory adequancy is a metric for choosing among grammars that all meet the criterion of descriptive adequancy.
    There are numerous ways of describing a single language (writing a good “grammar”) all of which will be desciptively adeqaute (also lots of ways of doing it badly, of course). What the explanatory question asks us to consider is which among these actually describes what is happening “in the heads of the speakers”, a question of cognitive science.
    If *this* is your research goal, then a descriptive grammar is “just” data collection; a starting point. Pretty unexciting if intended as an answer to the why question. But potentially quite exciting, and certainly valuable, if your research goals are different, say language description, or pedagogy, or endangered language maintenance, or what have you. In that case, then writing a descriptive grammar may well suit your goals. It is certainly a valuable contribution to knowledge, and if it can be prepared in a way that is of use to the community (rather than collecting dust on Harvard library shelves as far too many do; kudo to David Harrison on these grounds) then it is all the more valuable a creation.
    Why the tension? I see no reason for the tension to arise from the nature of the discipline. The search for “universal grammar” as a cognitive system relies on careful and systematic descriptions of individual languages (see the quote from Aspects). I’ll concede that some “theoretical linguists” sneer at people who are more interested in description, if it will be conceded that some descriptive linguists sneer at people whose interests are not about languages themselves but rather some abstract cognitive system underlying language abilities (independent of the beauty of individual languages). And yes, some people will insist that the term “linguistics” describes only what they do, and not what the “others” do. But surely we can all imagine that researchers in some area might be prone to finding their research area more interesting (or even more valuable / important) than adjacent areas with a different perspective. Ever talk to theoretical and experimental physicists about the “other” half of their departments?
    As to the suggestion that theroretical (or generative, or whatever term suits you) linguistic research gets in the way of socially important goals. This is a fair argument, especially if you believe (though I don’t) that granting agencies like the National Science Foundation would otherwise give the money to activist causes. Funds are limited, defense (odd: usually this is for offensive uses) spending also gets in the way of socially valuable uses of economic resources. If your claim is that everyone should be actively engaged in empowering the down-trodden, then “hear, hear”. But I fail to understand the idea that people interested in the mathematical modeling of particular cognitive systems (in this case language) are doing something wrong if they don’t devote their time and energy instead to valuable social goals that happen to be brought out in connection with the object modeled. Should all vision researchers be active campaigners for the betterment of the visually impaired? Should all evolutionary biologists promote Greenepeace?
    Incidentally, I “do” happen to live in both of these worlds. I teach and work in theoretical linguistics (yeah, the Chomsky stuff), and I happen to be active in fieldwork and, less than I might like, in the endangered language maintenance area. I wholeheartedly agree that the situation with endangered languages reflects a dire situation for huge numbers of people the world over who are downtrodden and oppressed and the like, and that we should be up in arms about this. I write about this. I give talks about this. And I talk to the press about this whenever I can. But these views don’t come from being a theoretical linguist, nor are they inconsistent in any way that I see with being one.
    Too few people are involved in important social activist causes. The world is a horrible place for vast numbers of people. Why blame Chomsky (or his scientific writings) for that?

  24. Jonathan: Thanks for a couple of enlightening comments. I admit to having been prejudiced against Chomsky and his acolytes to the point of not wanting to hear anything good about them (if you’d tried to practice some other kind of linguistics in the ’70s, you might understand), and I’m glad to hear from somebody who “lives in both worlds” and can speak convincingly to these points. You’ve opened my bitter, hardened mind a bit (though I don’t intend to have any truck with transformations myself); thanks.

  25. Hi Jonathan,
    Thanks for the reply, and I’d like to discuss this a bit more, though discussing things in the comments of someone else’s blog always feels a bit strange to me (though others seem to have no compunction at other blogs I’ve visited). I want to emphasize that this is not a an anti-Chomsky or anti-MIT or anti-you post (and I would be remiss if I didn’t note your own work with the bibliography on language endangerment (info here ) but I think that you are a bit too quick to dismiss a very real problem.
    How many linguistic programs in the US would accept a theory neutral grammar of an endangered language as a PhD dissertation? (for the sake of argument, let’s set aside whether there can be a ‘theory-neutral’ grammar) How about something ‘pedagogical’? That underlying bias against descriptive work (and even stronger bias against pedagogy) seems to be inherent among theoretical linguists. Almost any academic linguist who wants to do work with endangered languages has to do it, as you do, on the side. Is that really where we should be as a field?
    I admit that this is a bit of an absolutist position, but the claim that supporting the maintenance of endangered languages by granting degrees necessitates feedback into those areas shouldn’t be considered a heretical goal, yet I’m hardpressed to think of one program that would accept that as a valid dissertation topic. In that sense, the argument that evolutionary biologists not supporting Greenpeace or vision researchers not dealing with vision impairments is a bit of a red herring. Vision researchers aren’t dealing with a situation where the object of their study is going to disappear and evolutionary biologists have a much wider field that simply what Greenpeace wants to protect. A better example might be archeologists not supporting the prevention of trafficking in antiquities, though even that is not as strong as it could be. This might be (no, I’m sure it is) too emotional an image, but my feeling is that it is akin Literature majors against book burning. How can you allow your object of study to disappear? Yet the fact that theoretic linguists can (and do) ‘sneer’ at descriptive work as well as not see the point of tackling language endangerment (cf. Peter Ladefoged’s opinion piece in Language) seems to me to indicate that there is a problem in the field, not that there are a few bad apples spoiling the bunch.
    On the other hand, the field is coming around (often pushed by informants and communities demanding some return), but it is far to slow for my tastes.
    Anyway, if you want to continue this discussion (which I think would entail discussing the state of the field of linguistics, which might be a bit outside of LH’s interest) I would be happy to continue.

  26. discussing things in the comments of someone else’s blog always feels a bit strange to me
    Other bloggers may feel differently, but I actively welcome discussions in my comment threads; it’s one of my favorite things about having the blog. To have people who know more than I about some subject sparked off my an entry of mine discuss it at length is enjoyable and educational for me. Pray don’t feel constrained!
    the state of the field of linguistics, which might be a bit outside of LH’s interest
    Au contraire: even though I haven’t been actively involved with linguistics for a couple of decades, I love the field and am always eager to find out how it’s doing.

  27. Thanks, Joe for the post, and LH for the invitation to continue the discussion. As the end of the weekend nears, my list of things to prepare for the week is beginning to be a much more mundane cause for alarm. So just very quickly:
    Joe asks: “How many linguistic programs in the US would accept a theory neutral grammar of an endangered language as a PhD dissertation?”
    Some useful links on that topic:
    The Linguistic Society of America: follow links to the Committee on Endangered Langages and their Preservation. LSA policy actively supports promoting the documentation of endangered languages, and degree-granting for descriptive grammars.
    And it is being done:
    (Soc’y for the Studies of the Indig. Langs. of teh Americas)
    See especially their dissertation index at the SSILA website (, 200 dissertations in the last 15 years, ranging from theoretical works with native american language data, through descriptive gramamrs, texts and dictionaries, from a variety of institutions.
    Note also the new MIT Indigenous Language Initiative:
    “The Department of Linguistics and Philosophy is in the process of creating a Master’s program in linguistics for speakers of threatened languages. Graduates of the program will be able to use their linguistic knowledge to help their communities keep their languages alive. In addition, the MIT Indigenous Language Initiative will offer expanded opportunities for MIT students and faculty to become involved in indigenous and endangered languages through work with native speaker linguists in the SM program and also with outside groups.” (from
    There’s certainly room for lots more to be done, and I would be the last to argue that we can just sit back and take it in stride. But I think much more is being done than people might think.
    I also think that one of the major hurdles lies in funding opportunities, of which there are precious few, both for endangered language maintenance in general and for PhD students seeking to do extended fieldwork projects.
    Finally, I’d like to echo an earlier sentiment posted here: “linguists don’t save languages…” The MIT initiative (and the many elsewhere that have come before it) are, I hope, far more likely to be of lasting value than linguists working in the old mould of writing descriptive grammars that may be “theory neutral” but are largely inaccessile to the communities that want to use them.
    OK, just glanced at the clock – would like to proof the above, and to continue the discussion, but work beckons…

  28. Thanks for the comments. I think it might be useful to review what we agree on before moving to what we disagree on. I think we agree on the following
    -It is (vitally?) important that we work to preserve endangered languages
    -There could be more done to do this
    At this point, we disagree about what linguistics as a ‘field’ can do. To work from your examples, you point out that SSILA has done work and has a catalogue of “200 dissertations in the past 15 years’. I’m trying to imagine the number of dissertations in linguistics over the past 15 years, and I would argue that this is a very very small number. But even that number is inflated. First of all, this is dissertations from _all_ fields, including the allied fields of anthropology and sociology (for ex. _Nahua Religious Oblications in Central Mexico, 1692-1810_, ). Second, to get that number, MA theses are included (this is not to put down MA theses, but it does inflate the numbers) Third, these are dissertations from all over the world. Fourth, many of the dissertations are topics on linguistics that simply have some language examples (_Spatial Deixis_ _Functional Universals of Tense-Aspect-Modality Morphology in SOV languages_) I would also add that SSILA is not a ‘major’ (and I say this as a SSILA member) linguistic organization as such. So this is not such a ringing defense of the field
    The call from the LSA, which _is_ a major organization, well, has it really been heeded? You argued that MIT has a long tradition of fieldwork, but how many descriptive grammars have been accepted as PhD dissertations? The MIT Indigenous Language initiative has not begun accepting students and will only offer MAs and will only be open to 2 students a year. The initiative is a good first step, and I am heartened by the discussion of the entrance requirements that note “In some cases admission will require a liberal interpretation of the provision that candidates for graduate work at MIT must have a B.A. degree or its equivalent. (snip)If we were to restrict our attention to persons holding the B.A., then we would defeat the very purpose of the program — namely, to engage the most talented and capable potential indigenous language scholars. The conferral of a degree upon successful trainees will greatly facilitate both their eventual hiring and their employment in serious language-related educational work within their home communities. The importance of this latter consideration was very evident to Ken Hale and others in past efforts to train native speaker linguists.” Great stuff, but the fact that this is coming about only after Ken Hale’s death underlines my feeling that moving in the right direction still doesn’t absolve the field.
    You note that ‘linguists don’t save languages’ and then you get a dig in at linguists writing ‘theory-neutral’ grammars that are inaccessible to the community. But it is not simply ‘theory neutral’ grammars that are inaccessible. And those grammars, by describing, at least present account for the data, even though it may not be in the most elegant way. Yet these descriptivists are often marginalized in departments because they cannot produce citable publications as quickly as those theoreticians across the aisle. Certainly this is not simply a linguistic department problem, but a problem with the academy as it stands, but it seems that in the face of such an overwhelming threat to the viability of minority languages, there would be some attempt at rewarding those descriptivists.
    In a sense, my problem is that a place like MIT has to set up a separate track rather than expand their notions of a degree internally. They have to set up a separate track because there is the overwhelming notion that theoretical contributions are the only appropriate contributions to the field. Unfortunately, what goes around comes around and the field now suffers because linguistics is primarily defined as a theoretical enterprise. It has made minimal inroads as an undergraduate degree, and increasingly larger efforts are being made by the LSA to raise awareness in the field.
    Unfortunately, I think that it is too late to change anything, and it saddens me, but I hope by pointing out the gap between words and deeds, perhaps those coming up in the field might reconsider what they adjudge as valid work in the field and readjust their criteria accordingly.

  29. quick post as work also beckons, but I do want to make a few quick points. (which, if need be, I’ll come back and defend later.)
    Firstly, I believe (and I’m not alone in this) that there is a dangerous fallacy in the term “theory-neutral description”. I speak as someone working on a description of a previously undescribed language for a PhD. A “description” is coloured by one’s underlying assumptions about how language works, whether they are presented in terms of minimalism, basic linguistic theory, hpsg, construction grammar, etc etc.
    secondly, in my experience of what’s “useful to the community”, even “theory-neutral” grammars have a huge amount of terminology and make many assumptions (see point above on “atheoretical” research) which make them difficult to decipher for non-linguists. Either that or they introduce no framework for description at all and it’s impossible for anyone to work out what’s going on. The Institute for Aboriginal Development’s series of learners guides seem considerably better than most, although I don’t know who uses them.
    Which brings me to my next point – many learners guides are written for “the community”, without clear analysis of who this “community” is, and without (sometimes) a clear recognition that a “community” is made up of many different interests and needs and a learners guide that tries to cater for children learning to read and speak their heritage language, adults who speak the language learning to read it, young adults transferring (maybe dodgily acquired) English literacy to a language they have passive knowledge of, etc etc, is probably going to fail.
    and finally on the “linguists don’t save languages” point. No, linguists don’t save languages without a willing community, and that’s the way it should be. Many communities with threatened languages have a long history of colonial oppression and continued racism, and it’s galling enough as it is for some community members that the descendants of the very people who contributed to them ceasing to speak their language in the first place (e.g. by stealing their children) are now the people saying “you gotta speak your languages, you can’t let them die,” etc etc. This irony is very keenly felt in some places. there is also a strong feeling in some places that continuing to speak a language that has been treated as “primitive” for so long holds that community to the past. Whatever you think of these opinions, you can see how they arise.
    I work on a language with 35 speakers, all over 60. My presence in the community (on and off over the last 5 years) has increased interest in the language a bit, amongst some people, and it’s provided the impetus for a few younger people to spend more time with their grandparents, but it hasn’t “saved” the language in any way, as most of the youngest generation want to speak English. I’m there because the old people wanted to tell oral history in their language.
    ok so quick post wasn’t so quick after all, but this topic has touched on something I think about a lot, and since I don’t yet have a blog of my own, thanks language hat for letting my take up some space on yours.

  30. Hi Claire,
    I think there are two ways to look at the term ‘theory-neutral grammar’. The first is as you do, to note that the term is vacuous, in that you have to have a theory of grammar in order to write a grammar. I agree, but I have seen this argument used to justify writing full-bore, pedal to the metal theoretical grammars that are completely opaque to not only the community but to the majority of linguists as well. ‘Hey, we have to adopt a grammatical theory, and I’m sure that this is right, so those descriptivists rwho pretend they don’t have a theory are off in cloud cuckoo land’. But there is another way to view the term ‘theory neutral’ (this is why I tried to put it in scare quotes, but I see that I didn’t succeed in catching it everytime) and that is that to be in a form that is maximally useful. This does not necessarily mean that needs to be accessible, but that the data can be extracted without demanding a a complete and total understanding of the theory behind it. For example, I’m thinking of Keren Rice’s Grammar of Slave that alternates (IIRC) chapters of ‘theory-neutral’ linguistic anthropology style grammar with theoretical description. But to call on that level of effort for a PhD is over and beyond the standards of even the most demanding dissertation supervisor. (note that Rice’s work was accomplished only after she had done a theoretical dissertation on Hare Phonology)
    Part of it is ‘mission creep’, as it were. As a concrete example, if we were to accept statistic-based dissertations that were acceptable 20 years ago, before computers, we would be accepting mediocrity. Yet rather than reconsider what is an appropriate dissertation/thesis, there continues to be the demand of a theory driven dissertation/thesis. Yet a descriptive grammar would not be acceptable if one has already been done for the language. And try to imagine a dictionary as an acceptable thesis, even though the lexicographic problems that one has to deal with dictionary in an exotic language are as challenging as anything one could think of for a grammatical analysis.
    Claire touches on some other points, and I will just recommend the volume _Linguistic Fieldwork_ (ed. Newman and Ratliff. Cambridge, 2001) which also has a very interesting essay by David Gil, whose work on Riau Indonesian has gotten some attention lately.

  31. Claire: I’m very glad you’re gracing my comments section, and I’ll link to your blog as soon as you start one. Your points are excellent, and anyone who’s working on a description of a previously undescribed language is a hero in my book.
    As for “theory-neutral” grammars, joe basically made my point for me. Of course technically there’s no such thing, but if you go back and read grammars done by structuralist linguists who thought they were describing “just the facts, ma’am,” sure, you can see their theoretical assumptions, but you can also use them to actually get a grasp of the language. Descriptions in the Chomsky era are full of complicated terminology and pseudo-mathematical presentation that has to be painfully learned and hacked through before you can actually figure out what’s going on with the language. Similar situation with literature: yes, the New Critics had all sorts of theoretical preconceptions, but they had useful insights into texts that they told you about in straightforward language. Nowadays, even if a critic has something interesting to say it’s a point of honor to say it in such a torturted, obscure way nobody who hasn’t read the same French philosophers can understand it. I exaggerate, of course, but I think I have a point.
    This is a great thread: keep it up, everybody!

  32. I’m rereading Harrison’s 2007 book When Languages Die, which talks quite a bit about Chulym/Ös. Particularly interesting is the orthography invented by Vasya Gabov, the youngest fluent speaker. Itmakes use of the Russian soft sign to indicate that that written Cyrillic a о y are pronounced /æ, ø, y/ rather than /a, i, u/. You only need one soft sign per word because of Turkic front/back vowel harmony; it’s placed after the coda of the first syllable. What’s particularly clever about this system is that Gabov worked out this system when he was young, before any contact with modern linguistics: he intuitively realized the relationship between palatalized consonants and fronted vowels, plus the existence of vowel harmony, with no outside help or encouragement (on the contrary).

    A 2003 paper by Harrison and Gregory Anderson explains the orthography as it stood then. Since then, Gabov has added Cyrillic ҥ (en-ge) and ғ (ge with stroke) for /ŋ, ɣ/ respectively, again without outside help or knowledge of other Turkic Cyrillic alphabets.

  33. Another Middle Chulym feature of interest is that it may have a Yeniseian substrate; at least, there are clear Yeniseian hydronyms in the historical Middle Chulym area. So it is not a Turkic “much of a muchness”.

  34. Fascinating, and I thank you for reviving this excellent thread, which had completely slipped from my memory.

  35. -Yeniseian substrate

    Early Russian explorers recorded dozens of ethnic groups in the area which no longer exist. It is thought that in 18-19th centuries, aboriginal population of South Siberia underwent a process of linguistc assimilation and switched to Turkic languages.

    Exactly why the Russian conquest had this surprising effect is not clear.

  36. David Marjanović says

    You only need one soft sign per word because of Turkic front/back vowel harmony; it’s placed after the coda of the first syllable.

    Shades of Crimean Tatar.

    Exactly why the Russian conquest had this surprising effect is not clear.

    Probably it didn’t, and the process was going on anyway…?

  37. Particularly interesting is the orthography invented by Vasya Gabov, the youngest fluent speaker. Itmakes use of the Russian soft sign to indicate that that written Cyrillic a о y are pronounced /æ, ø, y/ rather than /a, i, u/.

    Compare it with the Chechen Cyrillic alphabet.

  38. David Marjanović says

    A whole bunch of languages in the eastern Caucasus – East Caucasian as well as Turkic ones – use аь, оь, уь. But to put ь elsewhere in the word is a different convention.

  39. It seems that if the first syllable ends in a consonant, the soft-sign convention is used, but if it ends in a vowel, that vowel is iotated. See the first paper linked above.

Speak Your Mind