Linguistics and Indigenous Languages.

I don’t usually bring stuff here from the Log, since I figure most people interested in language follow it already, but this post by Christian DiCanio is important enough I’ll make an exception. It’s about UNESCO’s International Year of Indigenous Languages and calls for linguistic research on such languages; here are some of the points made:

Dutch phonetics and syntax are not inherently more interesting than Bengali phonetics and syntax. Bengali has a far more interesting consonant system (if you ask me as a phonetician). Even Bengali morphology, which is far more complex than Dutch morphology, is under-studied relative to Dutch. Dutch speakers just happen to reside in economically-advantaged countries where there has been active English-based scholarship on their language for many years. Bengali speakers do not. […]

But imagine if you were asked to review an abstract or a paper where the author chose to zoom in on the specific details of a particular syntactic construction in Seenku (a Mande language spoken by 17,000 people in Burkina Faso, see work by Laura McPherson) or how tone influences vowel lengthening in a specific Mixtec language (spoken in Mexico). These are minority and indigenous languages. Many linguists would agree that these topics are worthy of scholarship if they contribute something to our knowledge of these languages and/or to different sub-disciplines of linguistics, but where do we place the bar by which we judge?

In practice, linguists often think these topics are limited in scope – even though they are no more limited than topics focusing on the reflexive clitic ‘se’ in Spanish. A consequence of this is that those working on indigenous languages must seek to situate their work in a broader perspective. This might mean that the research becomes comparative within a language family or that the research is a case study within a broader survey on similar phenomena. Rather than magnifying more deeply, if they want their work to be considered by the field at large, linguists working on indigenous languages often take the “go wide” approach instead. […]

Whether intentioned or not, both people and languages can be granted privilege. Scholars working on well-studied languages benefit from a shared linguistic common ground with other scholars which allows them to delve into deep and specific questions within these languages. This is a type of academic privilege. Without this common ground, scholars working on indigenous languages can sometimes face an uphill battle in publishing. And needing to prove one’s validity is a hallmark of institutional bias.

DiCanio adds some questions for linguists to think about (“What languages get to contribute to the development of linguistic theory?”; “Do you quantify the number of languages or the number of speakers that a linguist works with?”), and the comment thread is unusually thoughtful and interesting, with contributions from Bathrobe, J.W. Brewer, and other familiar monikers. I hope it gets a lot of attention.

Comments

  1. This post appeared earlier on DiCanio’s own blog, but I’m glad it got more readership by being republished on LL.

  2. J.W. Brewer says

    I wonder how this is and isn’t affected by the sort of linguistic scholarship being done. It struck me that in the olden days when the discipline spent its first century or so primarily focused on reconstructing PIE things were otherwise. Everyone educated enough to become a professional academic in almost any discipline already knew Latin and Greek; you couldn’t make a living as an up-and-coming comparative philologist with those two tongues as your primary focus. The field gets founded by those who were willing to learn Sanskrit and maybe some archaic forms of Persian in addition to Latin and Greek, then as it develops it turns out to be a professional advantage to have dug deeper into the surviving manuscript fragments of Gothic or Old Irish than your competitors had. Then it turned out that some low-class no-account peasant language still extant and “indigenous” to one of the less posh bits of Eastern Europe (Lithuania) was a very good tongue for an ambitious Indo-Europeanist to gain expertise in before everyone else in the field did, and so on. Because there was as it were a discipline-wide puzzle to be solved and the way to make your name as a scholar was to find a piece of the puzzle that hadn’t already been comprehensively worked over and the way to do that was to gain expertise in one of the more obscure languages that was still plausibly part of the solution to the larger puzzle. Thus, a virtuous cycle was inspired, at least until the puzzle got sufficiently solved that interest drifted away from IE in other directions.

    Not sure what sort of sweeping research agenda would recreate that dynamic, but maybe others would have suggestions?

  3. I don’t follow LL, and I believe marie-lucie has said she doesn’t either. So thanks for posting this here!

  4. Ah, I’ll try to remember that!

  5. David Eddyshaw says

    I used to follow LL assiduously, but my appetite for mistakes in the English of Chinese speakers is really very limited. And I miss Geoffrey Pullum.

  6. January First-of-May says

    I used to follow LL assiduously, but my appetite for mistakes in the English of Chinese speakers is really very limited.

    Still better than mistakes (and other weirdnesses) in the English of Donald Trump, which was their favorite topic for long enough for me to decide that LL probably wasn’t worth following much.

    I still come there occasionally, of course, and now that my Language Hat (Commented-On Posts) archive binge had caught up to the post about Language Log (in September 2003), I’m seriously considering restarting the Language Log Twelve Years Later project on my blog… do you think it should be renamed to Language Log Fifteen Years Later?

  7. Christopher Culver says

    I’m another subscriber of this blog over RSS who doesn’t follow Language Log unless it is linked to from elsewhere. There just isn’t enough historical linguistics at LL, and the coverage there of other branches of linguistics rarely interests me.

  8. David Eddyshaw says

    On the topic of the actual post:

    This focus on language minutiae is (I would imagine) part of the mindset that the only proper goal remaining in linguistics is to fine-tune what is already (nearly) a wonderful all-compassing Theory of Everything that just needs a little tidying up around the edges; as all languages fundamentally work the same way, the relevant insights are most likely to come from deeper investigation into areas which are already well-studied. There is no point in collecting data just for the sake of it, an essentially naive pursuit: stamp collecting, not science. Such activity cannot be of serious interest unless it bears on existing theory: but if it does, it will either confirm the beauty of our deep insights, or perhaps even allow us to develop the Theory to make it even more beautiful and all-encompassing. Studies of exotic languages are, of course, welcome. We expect they will all help to forward the Great Endeavour.

  9. my appetite for mistakes in the English of Chinese speakers is really very limited. And I miss Geoffrey Pullum.

    Seconded on your second point (even though he seldom allowed commenting). On the first, and on Trumpery, it’s easy enough to skim the heads of the posts to filter out what you’re not interested in(?)

    Prof Mair (aside from the ‘lost in translation’, which I agree is tedious) does raise some interesting topics, with a wealth of detail from him and other experts in follow up comments.

    If I might abuse Hat’s welcome: I’d like to thank David M for his contributions on one recent very long thread in particular. I (as a rank amateur) learnt a lot, even if much of the o.p. was blimmin’ obviously wrong to said experts. That is to say, I learnt why it was wrong.

  10. AntC: Which thread was that?

  11. “An early fourth century AD historical puzzle involving a Caucasian people in North China”, starting January 25

    Also “Thai “khwan” (“soul”) and Old Sinitic reconstructions”, starting January 28.

    (I’m not posting links, because the WordPress/thing seems to barf if I post more than one. And please pass over my cack-handed contributions with tolerance.)

  12. David Eddyshaw says

    An early fourth century AD historical puzzle involving a Caucasian people in North China

    I saw the beginning of that, but stopped reading it when I saw that Mair was not interested in factual counterevidence, and he started using accusations of discourtesy in an attempt to shut down discussion.

    (I should perhaps declare a personal grievance: Mair once emailed me to say he’d corrected my spelling in a post on LL. His correction of my perfectly normal Scots form was in fact wrong …)

  13. David Eddyshaw says

    @Ryan:

    The actual paper is inaccessible, alas.

    On first principles, my Fisher-sense starts tingling when I read that somebody has proved something statistically based on a tiny unique sample.

  14. I haven’t looked at Language Log for years. I loathe Mark Liberman & Arnold Zwicky. Geoff Pullum is great, though; very smart and funny. I’m glad to read he’s quit.

  15. I saw the beginning of that, but stopped reading it

    To David M’s credit (which is why I particularly wanted to thank him), he [D.M.] barreled on giving detailed reasoning ‘never mind the torpedoes’.

    I got a similar treatment: at first Prof Mair complimented me for being polite, if sceptical; later when I grokked what D.M. was saying, I grew more sceptical; and was castigated for being less polite. (I think I was consistently applying the scientific method throughout.)

    So yes there’s a fair amount of ‘noise’ on that thread. And there isn’t so much robust discussion/’Liberty Hall’ tolerated there as here. For me, though, there’s a seam of gold that finally got me to understand the Comparative Method. For a long time (since I first stumbled across Edo Nyland’s stuff — which is clearly ravings), I’ve wondered how you avoid the occupational disease of knowing so many languages that you hear sound-alikes/sense-alikes all over the place.

    Of course YMMV if you already ‘get it’/actually know what you’re doing, unlike me.

  16. David Marjanović says

    Thanks, AntC. I still haven’t found the spoons to go back to these threads and see if Prof. Mair at least grasps that we’re having a culture shock.

    (Stop the loading of that page before the adblock detector kicks in.)

  17. David Marjanović says

    I got a similar treatment: at first Prof Mair complimented me for being polite, if sceptical; later when I grokked what D.M. was saying, I grew more sceptical; and was castigated for being less polite. (I think I was consistently applying the scientific method throughout.)

    Well, yes. You were asking the same questions I was asking, just in different words; the most visible difference was that I tried to answer the questions in the same comments. Yet all Prof. Mair could apparently see was absence vs. presence of attack. o_O

    I’ve wondered how you avoid the occupational disease of knowing so many languages that you hear sound-alikes/sense-alikes all over the place.

    Well, first, it’s only science: you’re not going to have metaphysical certainty, so be prepared to compare probability arguments.

    Then, try to find regular sound correspondences that make some amount of phonetic sense (it helps, of course, if a parallel is attested somewhere, and it usually is).

  18. I still haven’t found the spoons to go back to these threads

    There’s some more good material, incl Prof Vovin chiming in (suggest you start at the end and work backwards); but not from Prof Mair. I kinda expected he would know what the Comparative Method entails, but I see no acknowledgment of anybody’s culture shock.

    We should really continue the discussion there not here. I don’t see that more archaeological evidence about the Tocharians or Tarim mummies has any bearing. The 10-character riddle is not in Tocharian; Tocharian might be IE but it’s not Germanic.

    And the noise has rather obscured that the Chang Tsung-tung 1988 paper might have some sense to it. (It’s far more cautious and methodical.)

  19. Lars (the original one) says

    I stopped following LL because the distance between nuggets of new knowledge was ever increasing, and there are only so many hours in the week. I can still go back and read the old guest posts by Don Ringe, though.

  20. If I might abuse Hat’s welcome

    Nonsense, the only way you can abuse my welcome is by abusing other commenters. I welcome any and all discussion of anything, as long as it doesn’t get too contentious (which is why I try to steer clear of politics); you’re certainly welcome to discuss material from the Log, especially since (as has been pointed out) the professors who rule that roost are often hostile to contrary points of view (as professors often are in general, especially once they’re tenured and unused to dissent). I myself have seen Mair get increasingly hostile to, and threaten to ban, a commenter who was simply disagreeing with him. Anyway, carry on talking, I’m enjoying it all!

  21. I also no longer read the Log, so let me echo the thanks for “cross-posting”!

    On a slightly different topic, since I find it hard to imagine there’s very much overlap at all between LH readers and people NOT interested in the languages DiCanio seeks to promote, I have to say that I find truly regrettable the way he frames the problem:

    Whether intentioned or not, both people and languages can be granted privilege.
    ..needing to prove one’s validity…
    “What languages get to contribute to the development of linguistic theory?”

    Passing over the passive-aggressive conspiracy-mongering tinge to the phrase “get to” here, “privilege” is simply neither an accurate nor well-advised way to describe the problem.

    Ill-advised, because on a practical level, I just don’t think you can really expect much success from a strategy of guilting people into scholarship.

    Inaccurate, because there is nothing in the slightest pernicious, immoral, selfish, or indeed surprising in people taking an interest in those languages that have had the most chances to pique their curiosity. I don’t even think it would be fair to criticize the average grad student flailing about for a topic who shies away from such languages primarily because they would be difficult (and possibly expensive) to learn. Lamentable enough, but surely understandable.

    The thoughtful advisor might suggest such languages for research, or subtly promote them by using them in his own work and teaching, perhaps invite native-speaker informants to the department, etc.–any such strategies to catch interest are wholly laudable. But any advisor who would respond to a student genuinely interested in Dutch by asking if she was really sure she didn’t want to study a “less privileged” language like Bengali is just absolutely odious. Subtly implying such a critique of one’s student without saying it outright is even more odious.

    There are simply no moral-harassing shortcuts to getting your own topics of interest more status in your field of research–or should not be. That can only be done by hard work and demonstration: modelling yourself the scholarship you want to see more of. And still, sometimes even the worthiest research agendas simply take time to catch on: many a brilliant scholar has surely breathed his last before the world finally came to appreciate his work. It’s not a line of work you should go into, I would say, if you’re not prepared for that fate. DiCanio’s post radiates good intentions, and he has all my wishes for his success, but this is just not the right way to go about it. Such a framing is bound to secure applause and approval while others are watching and listening, but when it comes to getting students from Ohio to offer up their weekends and midnights for a decade of solitary study of a language in Burkina Faso that few colleages of theirs will ever have heard of, guilt-mongering is probably not going to cut it. For that you have to actually capture people’s imaginations.

    As someone who studies Japanese, despite it being a language far more well-represented and well-funded than the languages DiCanio points to, I will never have more than a small handful of colleagues from Las Vegas to Vladivostok who share my personal subfield. Even in Japan there are probably not enough to fill a single medium-sized lecture hall. This will always be frustrating, but it will never be a sign that English and Russian are the beneficiaries of unearned advantage in territories where speakers of those languages are actually thick on the ground.

  22. I should think Dutch (or other Germanic) dialects in the Netherlands are much more endangered than Bengali with its quarter billion speakers.

  23. David Eddyshaw says

    It’s not only endangered languages that are underdescribed. It’s not easy to find good modern grammars of many African languages even with millions of speakers. (Though things are steadily improving, I’m glad to say; I think the tide may be turning.)

    Reverting to my previous gripe, I get saddened by the proliferation of articles treating just some theoretically sexy tiny subset of the grammar of some exotic language by authors (often L1 speakers) who have never published anything treating the language at all extensively, despite presumably being in a unique position to do so. I don’t blame the authors themselves, who are evidently constrained by the perverse incentives of an academic culture which sees description merely as a means to an end. Hard enough to make a living in linguistics at all, without swimming against the flow …

    Languages are not mere showcases for isolated curiosa.

  24. I agree that the Log gets tediously narrow at times. It’s nice that they published DiCanio’s piece because, apart from Mair, it is so English-centric (even more so America-centric) that I’m often amazed anything else gets a look-in.

    Mair can get testy, but he’s at an age where he’s almost earned the right to be cantankerous about things he disagrees with. The “early fourth century AD historical puzzle involving a Caucasian people in North China” post struck me as crackpot and I didn’t hang around to see David’s comments. That is an amazing thread that left my head whirling.

    While Mair is not always very tolerant of different viewpoints and can be sharp when ‘correcting’ people, I’ve never had any great problems with him. I suspect that is because we are coming from a similar place. He has in common with Pullum one important trait: they are both fighting against prejudices that people have been taught during their linguistic education. Indeed, I sometimes wonder whether it is the eternal lot of linguists to struggle against the simple linguistic precepts that people are taught at primary school.

    In English this is generally related to prescriptivism and a few other beliefs in the inherent fitness of English for rational or scientific discourse. In Chinese it is a belief in the inherent superiority of the Chinese language over those of their smaller neighours, a belief that Chinese characters are the only logical way to write Chinese, a belief that local varieties of Sinitic are just ‘dialects’, and a few others besides. Since I don’t subscribe to these beliefs about either English or Chinese, I tend to agree with Pullum and Mair most of the time.

    Turning to DiCanio’s post, I personally feel that the concentration of scholarship on a few major languages, centring on European languages, tends to suck all the oxygen out of the room — a bit like prescriptivism — and leads to a lack of imagination about what riches lie out there. It seems have set in motion a spiral of staleness, like the one we see on the Log, where the same old languages are hashed over again and again without much new or interesting input. Japanese has been an exception. Japanese was fairly exotic when I started studying it, but for various reasons, including its ‘novelty’, it became a feeding ground for linguistic theory and has now become incorporated in the acceptable range of ‘languages that we know and refer to’.

    Personally I’ve only been coming to belatedly realise how narrow my own horizons have been and how many fascinating phenomena lie out there, outside the few languages that I’m familiar with.

  25. I still follow Language Log through my RSS feed, and I do read the posts by Mark Liberman. Victor Mair both posts a lot of boring stuff, and he is thin-skinned and obnoxious.

  26. @ elessorn

    Yes, I felt a little uncomfortable reading DiCanio’s post, and you’ve (partly) put a finger on the reason. In a way, I wish he had just come out and said what he wants to say, instead of soft-pedalling his position, subtly appealing to a version of the victim syndrome, and pleading nicely about things that should be said outright.

    Quite honestly, the way that scholarship has tended to concentrate on a few major languages (European languages) runs the risk of becoming stultifying. Nothing wrong with coming out and saying that. For example, how translation studies circled around a few West European languages for so long, which amounted to translating among a few ‘dialects’. Only when you widen the circle, as has happened when Chinese and Japanese are added to the mix, can you start understanding what sort of problems are really there. Not that comparing a few ‘dialects’ (I am exaggerating, of course) isn’t interesting, but it’s ultimately too narrow to gain a true picture of translation. DiCanio should have come out and said that studying a small group of languages is just too narrow a focus.

  27. To restate a point I’d made on the LL thread – there are so many languages and so few potential scholars that even if we could dictate people’s research topics better than we can and make them spread out a bit more, we can’t expect all languages out there to be covered in enough depth to be fruitful — fruitful in the sense that even if English and a few others are overcovered, there are benefits from having a certain critical mass of different scholars working on different facets of a single language whatever it is. I think, if only we could figure out how to get there, that my “modest” proposal to come up with a semi-arbitrary consensus list of hitherto understudied languages from around the globe (from different language families, different social/political circumstances, different typological features, etc.) and arrange for each of them to be given a critical mass of Western-university grad students / junior faculty tasked with working on them for the next few decades would be kind of awesome. Among other things, whenever someone came up with some grand Universal Theory claim that used data from English (or some other historically dominant/overstudied language) as the starting point, it would be pretty easy to check it against our Worldwide Language Sample baseline.

  28. But any advisor who would respond to a student genuinely interested in Dutch by asking if she was really sure she didn’t want to study a “less privileged” language like Bengali is just absolutely odious.

    I wouldn’t go as far as that. Australian linguistics students have for many years been encouraged, or “encouraged”, to do their field linguistics among the Aborigines, or in recent years in PNG. That doesn’t strike me as inherently worse than doing your thesis in crystallography if your advisor is a noted crystallographer, something which nobody thinks twice about. Of course, the Australian situation is partly about accessibility (the Netherlands is a lɔoooooooooooooooong* way away) rather than political correctness.

    Even in Japan there are probably not enough to fill a single medium-sized lecture hall.

    It could be worse: you could be one of the six students of hypertwistoploppic pseudotheomorphisms (h/t Douglas Hofstadter for inventing this term) in the world.

    There is no point in collecting data just for the sake of it, an essentially naive pursuit: stamp collecting, not science.

    Well, there’s a sense in which that is true. “About 30 years ago there was much talk that Geologists ought only to observe & not theorise; & I well remember some one saying, that at this rate a man might as well go into a gravel-pit & count the pebbles & describe their colours. How odd it is that every one should not see that all observation must be for or against some view, if it is to be of any service.” —Charles Darwin

    But language description definitely doesn’t fit in to that category, as a language is itself a system où tout se tient to be learned, not just a random collection of observations. I wish more linguists grasped this.

    Such activity cannot be of serious interest unless it bears on existing theory: but if it does, it will either confirm the beauty of our deep insights, or perhaps even allow us to develop the Theory to make it even more beautiful and all-encompassing.

    And if the books reaffirm the Qur’an, they are unnecessary; and if they contradict it….

    *One o per thousand km, and a half-o for another five hundred.

  29. “These are minority and indigenous languages.”

    Within Europe Dutch is certainly a minority language and obviously it’s indigenous. And it’s beginning to slide into the endangered category.

  30. January First-of-May says

    The “early fourth century AD historical puzzle involving a Caucasian people in North China” post struck me as crackpot and I didn’t hang around to see David’s comments.

    It was so obviously utter crackpot that it almost looked like a parody of utter crackpot, and I half-suspect that the guest author might well have snuck it up on Mair as a test of whether Mair would recognize the whole thing as the obvious joke it was.
    (The “examples showing the change from European f- to T[aiwanese] s-” were the final straw for me to be certain that there’s no way it could be serious. Any actual linguist worth the title would immediately see it was utter bunkum, while a sincere crackpot is unlikely to care about such fiddly details.)

    I did not hang around to see whether anyone else ended up with the same conclusions. Now that I think of it I should probably go and check now…
    (I did check just now and couldn’t find anything like that, but the thread was so long and meandering that I could have easily missed it.)

  31. Seeing Mair’s reference to Mallory seconded by DM as being up-to-date and useful seemed like the only point of agreement, so I started with the Tocharian origins paper to see if it helped me understand the LL comments. I like a good dive in the weeds, but it works best if you have a wildflower guide in hand when you descend.

    I’ve never read Mallory, but I like his writing there. Clear, but also occasionally surprisingly vivid for scholarly linguistics.

    Anyway, I think I’ve solved the problem – the mixed Germanic and Celtic roots for Tocharian being suggested by the end of the thread are just proxies for the actual ancient language. Clearly, when Caesar defeated the Belgians, they fled to Tocharia, speaking their otherwise lost independent IE language which originally grew up between Germania and Galli!

    Now, I’m working on a Belgic etymology for “Phew”, with it’s related meanings of relief in many English dialects, and disgust in Mittel Ohioische.

  32. David Eddyshaw says

    Evident nonsense. The Belgians have always spoken their indigenous language, Dutch, though it is increasingly threatened by Taiwanese, and must now be considered endangered. In a clear sign of impending language loss, it shows radical phonological and morphological simplification compared with its parent, Bengalian (or Old Belgian.)

  33. The Mair piece informs me, or reminds me, of what I don’t like about LL: in general, the person at the front of the lecture hall is really not interested in backtalk, especially lengthy backtalk with lots of evidence. He (rarely she these days) has a position to uphold, after all. So you have to be ‘at in ‘and and very ‘umble, sir, or you get slapped down.

    Needless to say, our beloved Hat is the antithesis of that.

    The Belgians have always spoken their indigenous language, Dutch

    Spinach! The indigenous language of the Belgians is of course Yiddish.

  34. David Eddyshaw says

    The indigenous language of the Belgians is of course Yiddish.

    It’s Impossible to Tell.

  35. Ridiculous. Yiddish is from Jutland.

  36. January First-of-May says

    Clearly, when Caesar defeated the Belgians, they fled to Tocharia

    In the (possibly mildly misquoted) words of Caesar himself, “I hear there’s a barbarian rebellion there” (referring to either Outer Mongolia or Lower Mongolia – sources differ).

  37. David Eddyshaw says

    Tibet shall be full of Roman policemen.

  38. Plus whatever that branch of Saxe-Coburg-Gotha speaks among themselves nowadays; Hičajqri, like as not.

    Lower Mongolia

    Lower Slobbovia, surely.

    full of Roman policemen

    With their eels.

  39. David Eddyshaw says

    whatever that branch of Saxe-Coburg-Gotha speaks among themselves nowadays

    Thubanic, according to the most reliable sources. Wake up, sheeple!

  40. You people are giving me some much-needed laughs. (I just got in from shoveling the walkway, and will have to do it again.)

  41. >Plus whatever that branch of Saxe-Coburg-Gotha speaks among themselves nowadays; Hičajqri, like as not.

    Would that be pronounced hitchhikery?

    Does it mean something? I googled it, and there are 4 citations – 1 page fails, 1 link doesn’t actually contain the word, and the other 2 are to LH comments in ’07 and ’17 from people who’ve also posted in the last 45 minutes.

    I take it to be a lexeme from whatever that branch of Eddyshaw/Cowans/First-of-May speak among themselves nowadays, but that they each conjugate it differently.

  42. It’s a creation of the later stages of this wondrous thread.

  43. Like many here, I have not regularly read LL in years, for reasons which others have described well enough upthread. Their stark humorlessness is also unattractive (that they disagreed about how to interpret David M.’s “Phew!”, all the while each was confidently interpreting a string of syllables some fifteen centuries old, is the sort of irony which occurred to none of them, I am certain).

    I must confess that this posting of Christian DiCanio’s gave me pause: is this real or a parody? His question (“What languages get to contribute to the development of linguistic theory?”) presupposes that the only purpose of any language is to serve as grist for a theoretical mill. That a language might actually be an interesting object of study *in and of itself* (whether the language is spoken by an ethnic minority or not is utterly irrelevant, I should think) seems not to enter the equation, in his mind.

    Today we often, when interpreting older written evidence, need to carefully examine the orthographic system/conventions, as well as the L1 phonology of the recorder(s), is order to deduce what the actual phonology behind the written representation actually was. Sometimes I wonder whether, a few centuries hence, it will be necessary for future philologists interpreting present-day “linguists'” description of various languages to be familiar with various theoretical fads of our time in order to deduce in what way(s) present-day “linguists” were likely to misunderstand/misinterpret the structure of the language they were working with. Perhaps, indeed, future philologists will prefer dealing with linguistic data taken down by linguistically untrained observers, since, untouched as they were (well, are/will be, from our point of view) by the theoretical fads in question there will be more confidence in the accuracy of the data.

  44. January First-of-May says

    Would that be pronounced hitchhikery?

    …Huh, it would, or at least very close. I never actually got that at the time.

  45. J.W. Brewer says

    A random datapoint that might be a reminder that “indigenous” languages were not always so marginal in big-name linguistics scholarship. 75 years ago, Noam Chomsky was some high school kid no one had ever heard of but Leonard Bloomfield was either the biggest name or a strong shortlisted candidate for biggest name among active linguistics scholars in American universities. Bloomfield did a lot of work on Algonquin languages, and even tried to pass on the results of that work to scholars on the other side of the Atlantic. Just on the eve of WW2 he published a piece on Menominee morphophonemics in a volume with the truly spectacular title “Etudes phonologiques dédiées à la mémoire de M. le prince N.S. Trubetzkoy, Travaux du Cercle Linguistique de Prague 8.”

  46. I reread that wonderous Hičajqri thread, and I admit I was a bit disappointed that nobody followed up on my “Khmer mercenaries” idea.

  47. Thubanic

    The language of reptilian humanoids from a planet orbiting Thuban (aka Alpha Draconis), I suppose?

    I never actually got that at the time.

    It originated from the twisted mind of David E from the hitchhiker on the highway from the original post (hence Hičajqri and the related language Hajwé).

    Khmer mercenaries

    The way I heard it, only the officers were Khmer: the shock troops were Ngamambo Face Dancers who expected to be paid entirely in loot. But of course this may have been a mere misinterpretation of the blackface used during the night attack to avoid reflected gleams off the killers’ bodies.

    Yiddish is from Jutland.

    Well, perhaps the Belgians are really Jutes rather than Jews after all, more or less as Jack Aubrey supposed Stephen Maturin to be saying that the Irish were really Jews.

  48. By the way, does anyone know what Geoff Pullum’s last post at LL was? The listing on his own website stops in January 2018, the “author pages” on LL don’t exist (when you click on the author name after “Filed by”), and both forward and backward manual search were in vain. Was it with a bang or a whimper?

  49. Trond Engen says

    @Brett: I’ll give you a little more context in a comment on the thread.

  50. Elessorn, you said the use of the word ‘privileged’ was:

    Inaccurate, because there is nothing in the slightest pernicious, immoral, selfish, or indeed surprising in people taking an interest in those languages that have had the most chances to pique their curiosity.

    …but I thought the whole point of the concept of ‘privilege’ in those sort of arguments was to focus on the elements on an imbalance that don’t depend on anything pernicious, immoral, selfish or, especially, surprising involved in perpetuating it.

  51. Yes, exactly. There’s nothing wrong with the proposal, but if one is unfamiliar with the discourse of “privilege” it might seem offputting.

  52. Privilege implies under-privilege. It’s hard for me to see the suffering inherent in not having your language analyzed by linguists.

    On a different note, it’s not much of a problem (maybe a first world problem, showing my privilege), but has anyone else noticed that if you tab out of the comments box, instead of going to the next box down, where you would sign your name to your comment, you’re taken to the top of the thread? I’ve gotten used to it, just Shift-tab to go backwards. But it’s a little wonky.

  53. OK, I think we’ve covered this ground already, maybe many times, but I’ve heard a real Belgian speak on TV and his native language was (somewhat) broken English with a smattering of French words, but only those words English speakers can be expected to remember after forgetting about 80% of what they’ve learned in school.

    What I think is Jutes (who are just Jews, but not middle eastern ones, the ones from the Khazars) after settling for awhile in England were forced out by Normans and left for Belgium.

  54. I love that the only accepted etymology, though not universally held, for Belgae/Belgian is that they had a tendency to bulge with anger. It helps me realize that some academics just make shit up. Or maybe they got it from the old nursery rhyme.

    Algie met a bear, the bear was Belgian, the bulge was angry.

  55. Lars (the original one) says

    @Ryan, that’s because the four text boxes have explicit tab indexes, but the comment textbox itself is set as index 4 and the three input fields below it are index 1 to 3. So tabbing forward from the comment box brings you to all the places that can take focus but don’t have explicit indexes, in source order, starting with the link on the banner image.

    (I think the name / email / website boxes used to be above the comment box, whatever made them move below it clearly didn’t bother to change the tab indexes).

    Still, if you have followed a link to a post (not a comment) a single tab will bring you to the name field which might be useful. However, the submit button does not have an explicit index so it’s between all the comment header links and the sidebar links, about as far from the comment box in the tab order as it can get. Bad WordPress theme, bad!

  56. David Marjanović says

    Vovin? That should be interesting. Too bad I’m too tired tonight.

    And the noise has rather obscured that the Chang Tsung-tung 1988 paper might have some sense to it. (It’s far more cautious and methodical.)

    Pessimistic as I am about it, I still have to read it.

    And the noise has rather obscured that the Chang Tsung-tung 1988 paper might have some sense to it. (It’s far more cautious and methodical.)

    (The “examples showing the change from European f– to T[aiwanese] s-” were the final straw for me to be certain that there’s no way it could be serious. Any actual linguist worth the title would immediately see it was utter bunkum, while a sincere crackpot is unlikely to care about such fiddly details.)

    No… I think we’re looking at a sincere crackpot who is, however, doing this for fun (as he flat-out says in that thread), not to prove something else, unlike most crackpots including the mentioned Goropius Becanus. Further unlike most crackpots, he has heard of regular sound correspondences, and he sort of tries to find some… for parts of some words anyway.

    As it so happens, Mair does have something to prove: he has long held (quite reasonably) that the amount of ancient contacts between China and its northwestern neighbors has been underestimated, so he jumps at this opportunity. The rest is history. Unfortunately it’s not historical linguistics.

    “Phew”, with it’s related meanings of relief in many English dialects, and disgust in Mittel Ohioische

    I believe this is what Kids Today call a ‘sick burn’.

  57. I actually like reading Liberman and Mair. Liberman, because he writes on a subject I’m interested in, phonetics, and shows up-close how a professional practitioner of the field is approaching some commonplace but non-trivial issues. With Mair, I don’t know anything about Chinese, but I like his enthusiasm, his encyclopedic knowledge, and his willingness to collect expert opinions from others on Shanghainese slang or Middle Chinese writing or whatever. Likewise, it wouldn’t occur to me to read a blog on Russian literature, but I’m glad Hat isn’t shy about blending the subject into this language-related blog, and I enjoy reading whatever in those posts I can understand.

  58. As it so happens, Mair does have something to prove: he has long held (quite reasonably) that the amount of ancient contacts between China and its northwestern neighbors has been underestimated, so he jumps at this opportunity.

    He jumps at every opportunity, no matter how far-fetched and implausible; I worry that he’ll discredit the whole idea by his indiscriminate enthusiasm.

  59. J.W. Brewer says

    Do the kids today still say “sick burn”? To judge by what I overhear of my daughters’ conversations with other teenagers passing through our house I infer that that expression’s current usage level may be way way down from a peak of three or four years ago.

  60. David Eddyshaw says

    There’s no fun in using teenage slang right.

    https://xkcd.com/166/


  61. By the way, does anyone know what Geoff Pullum’s last post at LL was?

    Not sure, but the last one I remember concerned the pronouns of choice for underprivileged speakers of the overpriveleged language. And, without getting into the substance of things, who can blame him for not blogging since.

    That said, he had it sorta comming. The flame war before that was about the -gg- word, which any person with IQ larger than 50 has to know is impossible to discuss without mental health consequences

  62. Mair does have something to prove: he has long held (quite reasonably) that the amount of ancient contacts between China and its northwestern neighbors

    This is part of what I feel is a personal mission to bring down certain “objectionable” aspects of the Chinese world view. His enthusiasm is a perfect mirror of Chinese pride that their civilisation is so ancient and venerable that it doesn’t owe anything to anyone, at least until Buddhism arrived. After that, of course, it was just minor contributions from the small, warlike, barbaric “minority nationalities”, who now belong to us, thank you very much, so we can make a small corner in our national ideology for their contributions, and then the arrogant Westerners who came and humiliated us and even had the temerity to suggest that we weren’t necessarily the font of civilisation.

    You can see where he is coming from. It fits in with other ideas like the refusal to see languages like Cantonese or Shanhainese as ‘dialects’, a fascination with the idea that the old rigidity of the Chinese script (resolutely impervious to outside influence) is breaking down as the language becomes multi-scriptal, the view that Chinese characters are actually a hindrance to learning Chinese, and the assertion that Literary Chinese does not need to be taught in Mandarin (which makes it a closed shop). His views are of a piece — consistent, coherent, and quite understandable if you have been exposed to Chinese cultural ideology over an extended period of time. He is often right in calling Chinese bullshit (let’s call a spade a spade, there is plenty of bullshit there, just as in most national ideologies), but his “enthusiasm” for his iconoclastic viewpoint sometimes drives him to embrace some dubious propositions. What is amusing about Mair, though, is that he knows more about the Chinese language and its writing system (e.g., the correct pronunciation of characters) than most Chinese do.

    Unfortunately, any challenge to the established world view is met on the Chinese side with huge ideological efforts to consolidate their position even further, such as the initiative that was adopted a couple of decades ago to set the chronology of Chinese history in stone, and the current ongoing, well-funded effort to rewrite the history of the Qing dynasty in a way that is favourable to the leadership’s thinking. In all this Mair is just a gadfly.

  63. I think it is widely agreed that Geoffrey Pullum’s attitude toward Language Log changed significantly with the death of his wife. I suspect that Victor Mair is also responding to his wife’s death in the way he posts. He has posted about how his wife, who was Chinese (from Taiwan, I think), was heavily invested in the idea that Chinese languages should be taught without traditional characters, and that the characters were actually an active impediment to learning the spoken tongue. I have personally suspected (although I have never met the man, and I am basing this wild-ass hypothesis purely on what I have read of his writing) that he may be overcompensating in trying to preserve his wife’s legacy.

    If this sounds snarky, I really don’t mean it to. My wife will be hundreds of miles away tonight, driving to her late uncle’s place in Ohio. I’m worried about her, and if she were in a sudden accident, I know I would spend the rest of my life wanting to tell people how she was the best electrical utility cartographer who ever lived.

  64. @David M As it so happens, Mair does have something to prove: he has long held (quite reasonably) that the amount of ancient contacts between China and its northwestern neighbors has been underestimated, so he jumps at this opportunity.

    @Hat He jumps at every opportunity, no matter how far-fetched and implausible; I worry that he’ll discredit the whole idea by his indiscriminate enthusiasm.

    Hear, hear. Towards @Bathrobe’s point: we’re only going to counter CCP’s ideology, and China’s huge economic heft, with scholarship, not more bogus science/anthropology/ ‘re-education’. (I take the point that scholarship, science, careful assessments of facts seems to be way out of favour all round the world.)

    @Y Mair … willingness to collect expert opinions from others on Shanghainese slang or Middle Chinese writing or whatever.

    In the case in point, the opinions in the o.p. came from someone who is no expert (at least not in linguistics or philology), and Mair was willing to collect other experts’ opinions only if they agreed with him. (None did.)

  65. By the way, does anyone know what Geoff Pullum’s last post at LL was?

    The last I can see (by searching for ‘Pullum’ on LLog) is a comment here in March 2018. Click on his name and it’ll take you to his website, with a link to an index of his posts.

    Must admit I hadn’t clocked that he’d stopped posting. Now you point it out, yes he’s been silent since January last year (and I don’t remember any announcement). He did post on Lingua Franca to note it was ceasing to be; and said he needed an editor to prod him into posting.

    He’s allowed to retire (I suppose …). I remember attending a lecture he gave circa 1978/79, and that was on GPSG, so must have been after his rock musician ‘career’.

  66. It’s not easy to find good modern grammars of many African languages even with millions of speakers. (Though things are steadily improving, I’m glad to say; I think the tide may be turning.)

    Then can anybody explain what on earth’s going on at the Institute of African Studies?

    Catherine Obianuju Acholonu has a wikipedia page. Seems she was sensible once upon a time(?) And what does Erich Fred Legner have to do with African linguistics?

    Is Igbo related to Khoisan? (wp doesn’t seem to think so.) Is this part of the lumper/splitter controversy? (Note that Greenberg got caught in the crossfire between Profs Mair and Vovin, along with collateral damage to Lyle Campbell.)

  67. January First-of-May says

    Is Igbo related to Khoisan? (wp doesn’t seem to think so.)

    I don’t think anything is thought to be related to Khoisan at the moment, except maybe a small bunch of (mostly click) languages in its immediate vicinity; IIRC, the current belief is that the (macro-)Khoisan branch probably separated from the rest of the world’s languages before any of those other languages split from each other.

    A few of the putative “proto-World” lists do include Khoisan, but even there that branch is an obvious outlier. The phonology is just too fundamentally different.

  68. Khoisan is not demonstrably related to Khoisan either.

    I believe it is currently held to be 3–5 independent language families (none of which is particularly related to Igbo).

  69. David Eddyshaw says

    @AntC:

    Your instinct is correct: the linked article is utter garbage.

    Igbo is part of Niger-Congo (which itself is somewhat problematic around the edges, but not as far as Igbo goes.)
    Only überlumpers of the sort who imagine that it is possible to reconstruct a single human protolanguage think Khoisan is demonstratably related to Niger-Congo, and as others have pointed out, Khoisan is not itself regarded as a genetic unity by any real mainstream expert.

    There’s some reason to think Central Khoisan didn’t arrive in southern Africa that long before Bantu.

    Greenberg’s classification of American languages is accepted only by those with no knowledge of or interest in the actual linguistic facts. He has tended to get something of a free pass with his lumping classification of African languages into just four groups partly because his classification was at least an improvement on predecessors, but also because there aren’t nearly as many good studies of African languages as American, or as many specialists in a position to point out the problems.

    West Africa is pretty certainly the homeland of Niger-Congo, which is the nugget of truth buried under all the farrago of fantasy and nonsense in the article. Igbo is the origin of Niger-Congo in exactly the same sense that modern Dutch is the origin of Indoeuropean.

  70. David Marjanović says

    I have now read up on the two LL threads, but they’re both closed. 🙁 I waited too long. But DHYY has said most of what I wanted to say.

    One important thing I’ve long wanted to mention there is that *mer- means “die” all over crown-group IE, but “disappear” in Anatolian; I don’t know if it’s attested in Tocharian, but the general root for “die” there is *wel-, which is shared with Anatolian. This looks like one of the innovations of the crown-group is that “disappear” became a euphemism for “die” and completely replaced the original word for “die”.

    Am I understanding this right – Chau Wu uses alignment algorithms from molecular phylogenetics? If so, the huge problem with that is that insertions and deletions happen at random in DNA, but not in language.

    Am I understanding this right – Chang’s paper (which I still haven’t read) says Pre-Proto-Germanic was an isolating language at one (convenient) point!?! It never was. The inflectional endings that can be reconstructed for Proto-Germanic on purely Germanic-internal evidence all go back straight to (crown-group) PIE by the usual regular sound changes; and the meanings that the endings express are likewise derived from PIE pretty effortlessly.

    The Sinitic numbers have cognates all over the rest of Sino-Tibetan. They can’t have been borrowed any later than Proto-Sino-Tibetan, then – and that language was spoken thousands of years before Proto-Germanic.

    A thesis on Ket syntax was recommended to me elsewhere because its introduction mentions the (huge!) spread of Yeniseian hydronyms. Also mentioned in that introduction, on p. 5, is that one of the Ket self-designations, “rarely used today, mostly by the older generation”, is “bright-/light-colored people”. In comparison to what, I wonder?

    I actually like reading Liberman and Mair. Liberman, because he writes on a subject I’m interested in, phonetics, and shows up-close how a professional practitioner of the field is approaching some commonplace but non-trivial issues. With Mair, I don’t know anything about Chinese, but I like his enthusiasm, his encyclopedic knowledge, and his willingness to collect expert opinions from others on Shanghainese slang or Middle Chinese writing or whatever. Likewise, it wouldn’t occur to me to read a blog on Russian literature, but I’m glad Hat isn’t shy about blending the subject into this language-related blog, and I enjoy reading whatever in those posts I can understand.

    All seconded.

    Khoisan is not demonstrably related to Khoisan either.

    The most optimistic take is that of George Starostin, who thinks that the 3 families in southern Africa are demonstrably, if quite distantly, related to each other, and yet more distantly to Sandawe – but the basic vocabulary of Hadza is Afro-Asiatic, so it could at most have a Khoisan substrate.

    Igbo is totally different.

  71. @ elessorn

    my personal subfield

    Just curious, if it is not too personal a question, what is your personal subfield?

  72. Your instinct is correct: the linked article [on Igbo/Khoisan] is utter garbage.

    There must be some sort of support for it from the Nigerian powers-that-be: Acholonu is a Fulbright Scholar, collaborated with the Rockefeller Foundation, UN, etc (see wikipedia biog).

    Is “the thesis of an Igbo origin of language” (that is, of all languages) some sort of cultural nationalism? Nobody doubts the ‘out of Africa’ origin. Does it matter where from Africa, specifically?

    (And yes anybody applying Edo Nyland’s “methodology” will produce only garbage.)

  73. Am I understanding this right – Chang’s paper (which I still haven’t read) says Pre-Proto-Germanic was an isolating language at one (convenient) point!?! It never was. The inflectional endings that can be reconstructed for Proto-Germanic on purely Germanic-internal evidence all go back …

    Chang is looking only at ‘contentful’ words. (Non-linguist speaking here: what I mean is not pronouns/prepositions/auxiliary verbs/… I.e. not the parts of speech that carry inflection/morphology in English.) No I don’t take him to be saying early Germanic was isolating. Neither is he saying Germanic grammar was borrowed, merely lots of vocab. I compared that to the huge Romance vocab borrowed into English.

    But since the Comparative Method relies on parallel morphology to ground the sound pattern changes, the evidence Chang works with is sound-alike/sense-alikes in vocab. Lots of vocab, and with plausible suggestions for sound changes. He does say Chinese was not tonal at the time of borrowing/did have syllable-final consonants. So he’s reconstructing the borrowed vocab from how today’s tones and vowel quality must have been derived from the Germanic consonantal coda.

    Is Proto-Sino-Tibetan thought to have had syllable-final consonants/no tones? IOW was the (hypothesised) borrowing into a receptive sound-alike environment?

    There’s still the question of why/how borrow so many everyday words into a language that (presumably) already had words for X.

    (Chau Wu’s paper and ‘methodology’ is best left ignored. Sadly, as Hat pointed out, it’ll only discredit the SPP series.)

  74. Tones in Sino-Tibetan are clearly independently developed after the breakup of the protolanguage. However, the same causes operating in different branches have often produced the same effects. Furthermore, it’s clear that tone has become an areal thing: only a few of the 150-odd Austroasiatic languages have tone, and those are the ones that have been longest under Chinese cultural domination, like Vietnamese.

    As for the origin of comprehensive borrowing, as well ask why so many loanwords for ordinary things from puppies to brushes exist in English, or why there are so many Chinese and English loanwords in Japanese, some of them clearly overlapping native Japanese words. Cultural dominance clearly explains much of it, but not why particular languages accept or reject loanwords in general (English accepts, Icelandic rejects; Serbian accepts, Croatian rejects).

  75. >as well ask why …

    I guess there are still some major theoretical questions for linguistics to explore! 🙂

  76. It may be more about culture than language, as the Serbian/Croatian example suggests.

  77. It’s history, not even culture.

    In 19th century, the Croats and Czechs lived in Austria-Hungary and had to deal with German/Magyar domination, so they developed a rather strong Slavic identity and dislike of foreign words polluting their pure language.

    Serbs didn’t have this problem having won their own state since 1804, so they were not as concerned with keeping their language purely Slav.

    For Poles, the main oppressor was Slavic too, so Slavic purism made no sense for nationalistic Poles.

    And Bulgarians, of course, borrowed their entire literary vocabulary from Russian after the liberation.

    So it’s all history and relatively recent nationalistic politics.

  78. Re “out of Africa,” we know virtually nothing non-speculative about the timing of the origin of language (not that we know very much non-speculative about the origin regardless of its timing!), so even if we were confident as to the dating of the diffusion of anatomically modern humans out of Africa (which apparently is subject to some dispute, as is the potential linguistic capacity of Neanderthals/Denisovians/etc) we can’t really pin the origin of language itself before that diffusion, can we? *Probably* before the initial settlement of Australia, but are we even 100% on that, or is that just the combined result of several plausible-sounding but not proven assumptions?

  79. My pet theory – language was invented by Neanderthals in Europe sometime before they came into contact with anatomically modern homosaps, around 40,000 BP or thereabouts.

    It didn’t help them to survive, but at least they managed to pass the language to their successors.

    All languages in the world are descendants of that Neanderthal language of Europe, including languages of Asia, Africa and Australia/New Guinea.

  80. Lars (the original one) says

    Are there any arguments in favour of anatomically modern humans being without language for more than a generation or so after their mental development allowed it? (Cf. the rapid de novo genesis of sign languages when deaf children are schooled together). Never mind earlier members of Homo, 50.000 years is a very long time in itself.

    So, even granting a single time and place for the ‘invention’ of language, be it 50.000 or 300.000 years ago, is there any reason to believe that all existing human languages share a common ancestor so recent that it is even meaningful to ask whether Khoisan ‘split off first’? We cannot measure time depths of maybe 30.000 or 40.000 years with current science, so the common ancestor would need to be much more recent.

    Of course the proposition that Khoisan split off first is either true or false, since we postulated away an independent origin, but if it is true it is most probably not knowable. (Somebody someday might come up with incontrovertible proof that Basque and Khoi-San are related at a certain time depth, thus falsifying it, though that would only serve to move the question one level down in the tree).

  81. And Bulgarians, of course, borrowed their entire literary vocabulary from Russian after the liberation.

    Since Russian had borrowed a great deal of literary vocabulary from Old Bulgarian = OCS, this lead to a rather complex situation.

  82. We have no idea. The notion that once the neurology made it possible, it immediately blossomed is as speculative as everything else, plus we don’t really know what the neurology was because we don’t have CAT scans and assuming e.g. as soon as skull size stopped changing the brain inside must have stopped evolving is not guaranteed to be true. And looking at changes in material culture and postulating that such-and-such advance could not have been achieved without language is equally speculative. We have no direct observations of the individual and social behavior of primates as neurologically advanced as just-on-the-brink-of-language-but-not-quite-there-yet humans, so we can’t really be sure what they would or wouldn’t be capable of. We have great difficulty imagining functioning without language, but so great is that difficulty we tend (if lay people rather than scientific specialists) to systematically assume the communicative behavior of dogs, cats, horses, birds, etc. is more language-like than it probably is, because if we were dogs we would of course be talking dogs.

  83. Lars (the original one) says

    @J.W. Brewer, those are good points. So there are no good reasons to suppose that humans had started talking 50.000 years ago, and no good reasons to suppose that they hadn’t. But by my logic that also means that there is no good reason to suppose that there was anything resembling a unitary human language within a time depth that we will ever be able to probe that Khoisan could split off from ‘first’.

    We can guess, sure. I prefer guessing that the situation the last 100.000 years or more has been a steady state of macrofamilies expanding and shrinking and splitting and going extinct, just like we see in the past we have evidence of. Assuming without evidence that we are close enough to the start of linguistic history to detect any traces of the founding event is the fallacy of exceptionalism.

    However, since you mention material culture, I would really like to have a good explanation for why it looks so static for so much of prehistory — is it simply a matter of very small population sizes, so that it could take a thousand years before somebody came along who was bright and curious enough to try to make a stone tool in a more efficient shape, and a thousand more before the idea spread to somewhere we can find it? A lack of language would of course explain much…. It could also just be an illusion brought on by the selection of artefacts in the archaeological record — maybe they used all their creative energy on word games 🙂

  84. David Eddyshaw says

    Is “the thesis of an Igbo origin of language” (that is, of all languages) some sort of cultural nationalism?

    I’ve no information about this case in particular, but I have noticed a deeply regrettable tendency in some perfectly sane and respectable African linguists to rubbish existing comparative work on the grounds that it’s all some colonialist construct, and only native speakers have any title to say how their languages are related (the more understandable given the actually pretty primitive and unsatisfactory state of mainstream comparative work on African languages, apart from a relatively few shining exceptions within fairly close-knit subgroups like Bantu.)

    Politicisation of issues properly belonging to comparative linguistics is by no means confined to Africa, of course (add your own ghastly examples.) It’s exacerbated by the fact that comparative linguistics is a niche sport even among linguists and by the fact that most of the educated general public (including editors of Nature) are unaware that it exists as a rigorous scientific endeavour at all.

  85. Are there any arguments in favour of anatomically modern humans being without language for more than a generation or so after their mental development allowed it?

    Classic great ape sign language argument.

    All great apes are apparently able to use sign language – if somebody teaches them.

    Anatomically modern humans likely faced same situation – having the necessary language hardware, but not the ‘software’.

  86. David Eddyshaw says

    All great apes are apparently able to use sign language

    No. They can learn some individual signs. That is not the same thing at all.

  87. but the basic vocabulary of Hadza is Afro-Asiatic

    I believe you’re thinking of Dahalo, a Cushitic language of Kenya, which has clicks, too. Hadza is out on its own (genetically, too).

  88. If people are not taught language, they will invent one. That’s why there are so many “village sign languages” around.

  89. David Eddyshaw says

    This is quite a good account of the status of “Khoisan.”

    https://www.researchgate.net/publication/319528342_Khoisan_linguistic_classification_today

  90. David Eddyshaw says

    From Catherine Obianuju Acholonu’ s Wikipedia page:

    …Igbo oral traditions confirm the findings of geneticists, that by 208000BC – 208000 BC – human evolution was interrupted and Adam, a hybrid, was created through the process of genetic engineering

    I believe that this particular Igbo oral tradition originates in the ancient doctrines of Wallace Fard Muhammad (1877-1934.) The genetic engineer in question was the evil Dr Yakub, who created white people on the Greek island of Patmos for the evulz.

    When President Rawlings invited Louis Farrakhan to Ghana (I was there at the time) the most vocal protesters were Muslims, who were understandably unhappy.

  91. Patmos? I thought it was a secret undiscovered island.

  92. January First-of-May says

    But by my logic that also means that there is no good reason to suppose that there was anything resembling a unitary human language within a time depth that we will ever be able to probe that Khoisan could split off from ‘first’.

    I think I should clarify: the claim is that (macro-)Khoisan was the first branch to split off from the crown group of world languages, i.e. the ancestor of all the extant* ones (at least assuming that there was such an ancestor, which of course becomes more likely if we assume a larger time depth for the initial origin of language).

    In other words (this is technically a slightly milder claim) – “Khoisan languages [plus a small amount of others] are not descended from the most recent common ancestor of all the other 6000-odd languages of the world”.

     
    *) or sufficiently recently extinct, though realistically it shouldn’t matter

  93. J.W. Brewer says

    It’s a standard claim that most Khoisan-speakers are a genetically distinct population whose ancestors branched off early from the common ancestors of absolutely everyone else. It would follow from that that it’s *plausible* that language likewise branched off that early, but I’m highly skeptical you can evaluate that claim in purely linguistic terms, i.e. by demonstrating that there are structural similarities in all non-Khoisan languages worldwide, not shared with Khoisan, that are best explained by hypothesizing a post-divergence most recent common ancestor of all non-Khoisan languages. Unless you assume that presence/absence of clicks is the most fundamental binary divide (and stable over tens of thousands of years) there could possibly be, and why would you assume that?

  94. @David Eddyshaw: That should be “Wallace Fard Muhammad (??-???).”

  95. David Eddyshaw says

    The three main Khoisan groups differ quite considerably from each other typologically (especially the Central group from the other two.) The really striking similarities are all phonological. That means little or nothing in terms of ultimate genetic relationships. Neighbouring Bantu languages have acquired the famous clicks too (during what must be really a very brief time period relatively speaking.)

    Sprachbund.

  96. President Rawlings

    I personally know someone who met him – one degree of separation.

    Maybe I should start keeping a list – it’s not the first time I encounter on LH mentions of people who know people I know.

    But someone from Ghana of all places!

  97. David Eddyshaw says

    I personally know someone who met him

    Never did myself; though whenever I travelled to Accra, my staff used to amuse themselves by asking me if I was going to call on my cousin (Rawlings being half-Scots, of course.)

  98. Just checked twitter of another acquaintance from Africa – full of photos from Davos and African Union summit in Addis Ababa.

    Apparently now I have one degree of separation to every head of state in Africa.

    {thinking} it’s very likely she met Rawlings too.

    Ghana keeps getting closer and closer

  99. David Eddyshaw says

    Ghana has a lot going for it as a country. Pay a visit!

    Ghanaians spend a lot of time banging on about how welcoming they are to strangers. Over the course of time this autobrainwashing has led to the happy result that they really are very welcoming to strangers.

    My own feeling is that there isn’t a lot to see in Ghana, comparatively. (Especially where I used to live. One bit of savanna looks very much like another bit of savanna.) What is endlessly interesting are the cultures and (of course) the languages. But you need to stay a while to appreciate things like that.

    There are, incidentally, quite a lot of Russian links (going back to Kwame Nkrumah’s day.) I used to come across a surprising (to me) number of Ghanaians who’d studied in Moscow.

  100. President Rawlings

    So that’s how a Scottish dictator acts. Seizes power, kills about 300 political figures, relinquishes to a democratically elected government, seizes power and relinquishes it again, gets himself re-elected in first a dirty election and then a clean one, retires after that.

    ““One national founder wrapped up his rebellion against the colonial power, not by using his supreme military rank to seize power in the newly independent country, but by retiring to his own lands for six years while others ruled. He then came to power through peaceful and democratic means, held office for eight years, and voluntarily retired again, this time for the rest of his life, despite every prospect that he could have become de facto President-for-Life if he had wanted to.” —my description of George Washington

  101. J.W. Brewer says

    Some Scottish dictators elsewhere on the same continent were less comparatively benign. https://en.wikipedia.org/wiki/The_Last_King_of_Scotland_(film)

  102. David Eddyshaw says

    For all kinds of reasons I don’t want to get into Ghanaian politics, but:

    Europeans and North Americans have little concept of the sheer difficulty involved in being any kind of effective ruler of an African state and consistently greatly underestimate the statecraft of those who’ve made any kind of go of it.

    A typical European/American politician in such a role wouldn’t last a week and would still leave his country worse off at the end of that week. And the worst kind (pick your own example), if placed in a position to do so, would out-Mobutu Mobutu. Or Amin.

  103. My description was intended to be complimentary (comparatively).

  104. David Eddyshaw says

    Ah. I was too touchy.

    Perhaps I should just have proposed Macbeth as a better example of a Scots dictator.

    The retirement bit is key, I think. Caesar famously said of L Sulla Sullam nescisse litteras, qui dictaturam deposuerit, that his laying down his dictatorship showed that he didn’t know his political ABC; but we know how that turned out for him.

  105. J.W. Brewer says

    Ghana is currently one of only 7 countries on the African mainland (not sure if lumping in some of the offshore island nations with “Africa” is all that conceptually helpful) that the Freedom House people think are “free,” and Rawlings’ example of stepping down (the last time) when he said he would despite the fact that his hand-picked successor had lost the election and the other guy won was a helpful step in getting them there. Rawlings’ various freely-elected successors lack the high profile and worldwide brand recognition of Nkrumah, but maybe the people of Ghana are better off for that.

  106. David Marjanović says

    I forgot: Goths coming from Austria!?! The whereabouts of the Goths were documented pretty much since Tacitus had them on the Polish coast. When the Romans showed up in Austria, it was all Celtic, except for Raetia in the west and maybe some Dacians in the Pannonian east.

    Is Proto-Sino-Tibetan thought to have had syllable-final consonants/no tones?

    Yes, as also all reconstructions of Old Chinese*, and even Early Middle Chinese on the phonemic level (as shown by contemporary descriptions, never mind modern reconstructions).

    * All, since the first ever in the 1950s, have an inventory of syllable-final consonants at the very least as large as that of Cantonese or Min, usually larger and containing clusters. And while some people have wondered if the difference between Type A and Type B syllables could have been tone (even though it has nothing whatsoever to do with the tones of Late Middle and later Chinese or modern Min), none of the published reconstructions makes that interpretation, AFAIK.

    My pet theory – language was invented by Neanderthals in Europe sometime before they came into contact with anatomically modern homosaps, around 40,000 BP or thereabouts.

    Australia was settled more like 60,000 BP, and yet even the Tasmanians spoke. My pet hypotheses are that language had a very gradual origin, and that the first fully modern language was spoken long before the last common ancestor of all attested languages (“Proto-World”).

    I believe you’re thinking of Dahalo, a Cushitic language of Kenya, which has clicks, too.

    Nope! Here’s the paper – in Russian. Hadza is not like Dahalo, it’s much weirder.

    The really striking similarities are all phonological.

    The similarities described in this paper (in English) may not be striking, but what do you think of them?

  107. In modern times, at least, I would never call something Scottish in order to insult it. Even “The Scottish Play” is superstition, not like, say, the curse of Scotland (the itch).

  108. David Eddyshaw says

    Never crossed my mind. You mean there are people out there who think “Scottish” is insulting? It’s a rum old world, and no mistake.

  109. >So there are no good reasons to suppose that humans had started talking 50.000 years ago, and no good reasons to suppose that they hadn’t.

    Given what we know of the genetics and what it suggests about movement of people, as well as what we know about linguistics, 50,000 years doesn’t seem like enough time to me. If you said 100,000 years ago, I’d concur.

    In particular, a 50,000 year cut-off for language would require something surprising in Australia. Geneticists seem to believe there was a foundational population more than 50k years ago, and they also find evidence of a population from India about 4,000 years ago. No other inflows have been detected genetically, which might not mean they didn’t happen, but would mean the newcomers didn’t have many offspring.

    So you would either posit that language only popped in Australia in 2000 BC. Or that a group that post-dated the foundational population made it to Australia at some point and introduced language, yet otherwise had no discernible genetic impact.

    It would seem extraordinary that a group with language didn’t have sufficient advantages to impose itself at least a little onto the genetic make-up of the continent. Yet had a big enough impact that everyone else learned how to do what they were doing.

    There are others here who know a lot more about this than I, but for me, 50,000 years doesn’t seem possible.

    The other thing is that the Khoisan genetic break from other populations is now thought to be more than 100,000 years.

    It would be a remarkable coincidence if the Khoisan/rest of the world genetic divide and the Khoisan/rest of the world linguistic both happen to be the deepest divisions among extant humans, but that it’s just by chance, with the linguistic divide only occuring tens of thousands of years later, after language had back-migrated there.

  110. David Eddyshaw says

    The neatest argument I’m aware of for relatively late development of Language ties it in with

    https://en.wikipedia.org/wiki/Behavioral_modernity

    However, I suspect that it’s one of those beautifully plausible ideas that unobligingly melts away the more facts you discover relating to the issue. But I speak from profound ignorance, only emboldened to do so by a suspicion that everybody else does too (quite possibly because I am afflicted by a variant form of the Dunning-Kruger effect …)

  111. J.W. Brewer says

    Re “remarkable coincidence,” it would be if both pieces were uncontroversially true, but I’m still stuck on what the actual basis is (other than clicks) for thinking that the linguistic divide between Khoisan (not necessarily an internally coherent grouping), on the one hand, and everything-else-in-the-world, on the other hand, is so profoundly deep? That’s objectively deeper than the divide between oh let’s say Algonquin and Austroasiatic because why?

  112. Goths coming from Austria!?! … When the Romans showed up in Austria, it was all Celtic,

    Thanks. (I think the suggestion was pre-Roman, but I’m suffering wikipedia overload. Yet again, if the Goths came south from Scandinavia, when/how did they get there in the first place?) So presuming the Austria thing is not the consensus, how do ‘minority’ dratted linguists get to hold so much sway on wikipedia? Or was that another bit of “Original Research”?

    Celtic was a recurring theme on that thread after it cleared away some of the nonsense — e.g. Chris Button thought he spotted a word that could not be Goth/could only be Celt or Balto-Slavic. Is it possible Celts and Goths collaborated on Eurasian trade/settlement? What sort of language/pidgin would they work in?

  113. David Eddyshaw says

    The similarities described in this paper (in English) may not be striking, but what do you think of them?

    Frankly, I’m not competent to judge; my knowledge (such as it is) all comes from reading papers by others. But I’m struck by the fact that actual Khoisanists (as opposed to professional long-rangers like Sarostin) nowadays seem rarely to be persuaded that the undoubted lexical similarities are due to common origin rather than borrowing. There’s no dispute over the fact that the southern African Khoisan languages are/were at any rate a Sprachbund, whether or not the three groups are ultimately related.

    Güldemann develops these themes in extenso in various articles; in a nutshell, he thinks that the Central Khoisan languages are comparatively recent arrivals in southern Africa (possibly related to left-behind Sandawe) which have absorbed a great deal (including lexicon) from the Tuu and Kx’a who preceded them (and which may be distantly related to each other.) Other chapters in the book that the article I linked to comes from deal with some of this (and some genetic issues too.)

    In fairness, I should say that Güldemann is a confirmed splitter (like me but with, like, actual knowledge); I would say that’s because he’s actually familiar with the languages he’s talking about (but then I would say that, wouldn’t I?) He’s very much into the theme of areal phenomena, and points out that modern understanding of such things just wasn’t there yet in the earlier days of trying to sus out language relationships in Africa: similarities got too readily attributed to common origin. To some extent, there’s no arguing with that; it doesn’t mean that attributions to common origin have to be wrong necessarily, of course, but it does mean that the evidence often needs reexamining.

  114. David,

    Starostin? Nnneeeah…

    I don’t know anything about these languages, but this jumps at me. In his Proto-Khoisan paper, he ignores tones completely. Considering the richness of tonal systems in these languages, you need to say something about correspondences, and some of his equations show mismatches which need explaining. This is at best poor scholarship.

    From what I can tell (I know hardly any Russian), the Hadza paper is Ruhlen-level grab and match, with no regard to regular correspondences, and with using an abundance of daughter languages to pick comparanda out of, rather than reconstructions. Nuh-uh.

  115. So it’s all history and relatively recent nationalistic politics.

    All too often, culture is history (shades of Hobsbawm), or at least phenomena, habits, and attitudes that have developed historically. Things are not ‘the way they are’: you have to understand how they got ‘the way they are’.

    Since they are two sides of the one coin, the propensity to borrow seems to me to be both historically and culturally determined.

    Why has English become such a bower bird, picking up shiny objects to decorate its nest? Did it start with the Norman Invasion, when a Germanic-speaking society was suddenly taken over by Romance speakers who now lorded it over the nest? Or did it happen later? The Ayenbite of Inwyt was written in the 14th century in very Saxonist prose, unlike Chaucer’s work, which used a lot of French words. Or was it later still, when the fad for “inkhorn terms” arrived? Perhaps it was the whole chain, starting with the Norman invasion, that made English into an importing language.

    On the other hand, Chinese has been more impervious to imported vocabulary. It’s not that Chinese haven’t borrowed foreign words — they have, right through history. But the Chinese writing system seems to discourage such words from taking root in respectable Chinese. It hangs round on the fringes, as dialect words, or it remains as a ‘phonetic rendering’; but even here, there is a tendency to replace foreign words with more “acceptable” Chinese renderings (麦克风 màikèfēng > 话筒 hùatǒng, 因特网 yīntèwǎng > 互联网 hùliánwǎng). There seems to be an attitude on the part of the arbiters of good prose that foreign words simply don’t belong. The cultural attitudes that this stems from seem to go back a very long way, and is, I suspect, closely related to the attitudes of the Confucian literati through history.

  116. Lars (the original one) says

    @Ryan, 50.000 years was what I found as the most common dating of the ‘anatomically modern human’. If the people in Australia came there earlier than that, and were ‘anatomically modern’ whatever that means (and I assume they were), then the date obviously has to be put earlier.

    @J.W. Brewer, thanks, you are saying what I mean only better — yes, it is possible that the speakers of Khoisan languages were as isolated linguistically as they were genetically for the last 100.000 years, but how can we ever prove that they didn’t take over some outside trading language 60.000 years ago? As people do.

  117. David Eddyshaw says

    Hausa is another one of the great magpie languages; as Hausa has gobbled up a lot of its neighbour languages in the relatively recent past (and is continuing to do so) and is moreover the language of a people rightly famed as soldiers, as scholars and as long-range merchants, it all seems not so much comprehensible as inevitable.

    But pretty much all of this applies to Chinese too …

    Classical Latin is notably purist, rarely adopting a loan when a calque will do (Cicero helpfully explained what he was up to explicitly in this.)
    Made no difference to the common folk, who borrowed whatever they felt like.

  118. I forgot to mention: There was one huge inrush of foreign vocabulary into Chinese in the 19th-20th centuries — from Japanese. But that mostly came about because the imported vocabulary wasn’t discernibly foreign, being written in Chinese characters. And most of it was formed on Chinese models. Still, quite a few words snuck in unawares, like 取缔 ‘crack down on’ and 取消 ‘cancel’, both purely Japanese, along with a lot of official bird names… And the bird names stuck because they looked a lot more like Literary Chinese than the plebeian names that local people gave to birds. (Ironic.)

  119. David E:

    “To be fair the Scottish ecomomy has its strengths – its chief exports being oil, whisky, tartan and tramps.” (Ray Winstone)

    “The currency of independent Scotland would be the Mars bar.” (Ian Hislop, not verbatim)

    “I do indeed come from Scotland, but I cannot help it.” “That, Sir, I find, is what a very great many of your countrymen cannot help.” (too notorious to attribute)

    To say nothing of anti-Highlander/anti-Jacobite sentiments and Ptolemaic comparisons of Scots with Scythians.

  120. David Eddyshaw says

    @JC:

    My sainted Scots grandmother felt, I think, deep down, that Highlanders were not really altogether human (a common enough feeling prior to the Disneyfication of the Highlands.) Quite a different matter from being anti-Scottish.

    Boswell was a sad, sad man. I think that that is all that needs to be said.

    As to the other examples, they all seem unequivocally positive to me. Hoots.

    (Deep-fried Mars bar, that would be.)

  121. I remember an article in the press from the time of the Bosnian war suggesting that if you didn’t understand the savagery and violence among the peoples of Yugoslavia, you should just remember what it was once like between the Scottish clans. Our current picture of Scotland seems to be idyllically Victorian.

    (Watching the The Royal Edinburgh Military Tattoo is quite a “hoot” — it is such an Empire, Victorian presentation of Scotland.)

  122. J.W. Brewer says

    It is to be regretted that Scottish cuisine was so late-blooming in its achievements that the earlier Scots emigres who crossed the Atlantic in the 18th and 19th centuries were not able to bring deep-fried Mars bars with them to contribute to American/Canadian cuisine. (My late maternal grandmother did pay culinary tribute to my maternal grandfather’s part-Scottish ancestry by always making shortbread at Christmas time with a wooden mold that imprinted a thistle on it, but I’m pretty sure she had gotten that device at some random upstate N.Y. garage sale circa the Truman administration — it had not been brought over the ocean in the days of the Highland Clearances and then handed down through the generations.)

  123. George Grady says

    All great apes are apparently able to use sign language

    No. They can learn some individual signs. That is not the same thing at all.

    Indeed. One of my sons is non-verbal and uses sign language, and has many friends in the local Deaf community. They find claims that Washoe, etc., were able to use sign language incredibly insulting.

  124. not altogether human

    The most barbarous isle of Lewis.

    a sad, sad man

    Well, no doubt. But this passage is not particularly evidence of that. Here’s the full context (emphasis added):

    Mr. Davies mentioned my name, and respectfully introduced me to [Johnson]. I was much agitated; and recollecting his prejudice against the Scotch, of which I had heard much, I said to Davies, ‘Don’t tell him where I come from.’

    —’From Scotland,’ cried Davies, roguishly.

    ‘Mr. Johnson,’ said I, ‘I do indeed come from Scotland, but I cannot help it.’ I am willing to flatter myself that I meant this as light pleasantry to soothe and conciliate him, and not as an humiliating abasement at the expense of my country.

    But however that might be, this speech was somewhat unlucky; for with that quickness of wit for which he was so remarkable, he seized the expression ‘come from Scotland’, which I used in the sense of being of that country; and, as if I had said that I had come away from it, or left it, retorted, ‘That, Sir, I find is what a very great number of your countrymen cannot help.’ This stroke stunned me a good deal; and when we had sat down, I felt myself not a little embarrassed, and apprehensive of what might come next.

    I have seen a suggestion that in the 18C this sense of come from was in fact a Scotticism. The OED doesn’t confirm this, though.

    Deep-fried Mars bar

    Well, it’s hard to use a cooked item as currency: its shelf life is short. I suspect, however, that it was Bruce Murrie (the second “M” in M & M’s) rather than Forrest Edward Mars Sr. (the first one) who was of Scots descent.

  125. My pet analogy of Scotland is Grand Duchy of Lithuania.

    Where the Highlanders, Lowlanders and English correspond to Lithuanians, Ruthenians and Poles.

    OK, the process didn’t finish in GDL due to Russians, but suppose partitions never happened and you would have almost completely Polish-speaking Belarus and Lithuania with Lithuanian and Belarussian spoken only in some isolated pockets in remote swamps.

  126. You mean Washoe the indigenous people or Washoe the chimpanzee?

  127. Obviously Washoe the chimpanzee.

  128. SFReader says: In 19th century, the Croats and Czechs lived in Austria-Hungary and had to deal with German/Magyar domination, so they developed a rather strong Slavic identity and dislike of foreign words polluting their pure language.

    This is a complex thing, and it goes back further than the 19th century. There are “puristic” tendencies in 16-18th century authors. However, I don’t believe it’s so much a “purism” issue as it is a tendency to slavicise concepts and words to make them easily understood and easily pronounced.

    Rather than being puristic (whatever that might mean), Croatian has been quite open to words from its own dialects and from other Slav languages. There are similar “puristic” tendencies in Czech, Hungarian and even German. So it could be a central European cultural area feature.

  129. Goths coming from Austria!?!

    Let’s not forget the pike that paid by the pools of Grafá. The Carpathians are only barely within modern Austria, but still.

    The Washoe are interesting people. Their language is the only non-Numic one in the Great Basin, and is an isolate (even people who believe in Hokan are now skeptical that it’s included), suggesting that they are in some sense aboriginal there. The name of Lake Tahoe is Washo (the less-misleading spelling used for the language), and although they rhyme in English, the Washo forms are dáʔaw ‘lake’ and wá:šiw (acute marks stress). Most Englishes lost [iw] long ago, so the deformation is not unnatural.

  130. David Marjanović says

    Good point about the tones.

    From what I can tell (I know hardly any Russian), the Hadza paper is Ruhlen-level grab and match, with no regard to regular correspondences, and with using an abundance of daughter languages to pick comparanda out of, rather than reconstructions. Nuh-uh.

    I’ll need to read it again, but there is no reconstruction of Proto-Cushitic (and the reconstructions of Proto-Afro-Asiatic disagree with each other a lot because of deficiencies in the data), so at the moment it may not be possible to do much better than this.

  131. David Marjanović says

    The lack of concern for the tones is explained on p. 430: of all Peripheral Khoisan languages (North + South), “adequate description of the tonal system is only available for ǃXóõ and Zhu|’hoan.”

    And then, within North Khoisan alone:

    All the 400 or so NK etyma for which dialectal data are available in that source can be loosely divided in two groups: «tonally stable», in which all or almost all of the dialects are in agreement on the tonal characteristics, and «tonally unstable», in which there are at least two or three different tonal patterns scattered throughout the dialects, with little hope of detecting any kind of distribution. […] In fact, outside of the items represented in Snyman’s short list, we do not even have any idea which NK roots are «tonally stable» in the first place

    […]

    I would, therefore, postpone a serious discussion of tonal oppositions in PPeK (and, in fact, in Khoisan overall) until a more or less acceptable reconstruction of segmental phonology has been effectuated. It is not excluded that there are areas of PPeK consonantism which are tightly linked with tones, such as, for instance, the «random shift» of voiced and voiceless reflexes of PPeK click effluxes and non-click consonants (see 4.2.2.2). On the whole, however, such interaction has not been shown to be very tight in any of the modern day Khoisan languages, and there is so far no reason to think the situation were to be any different in the proto-language.

    So there’s a lot to sort out within the uncontroversial groups first.

    Finally, in the general conclusions on p. 461:

    On the whole it must be said that the possibilities of historical work on Khoisan, both in terms of internal reconstruction and external comparison, are not only far from being exhausted, but, in fact, have so far been barely tapped. The majority of the problems associated with this work are of a purely technical nature — lack of linguistic data as well as not enough qualified specialists in the field, rather than any substantial theoretical obstacles that would somehow hinder the application of the classic comparative method to Khoisan.

  132. The majority of the problems associated with this work are of a purely technical nature — lack of linguistic data as well as not enough qualified specialists in the field, rather than any substantial theoretical obstacles that would somehow hinder the application of the classic comparative method to Khoisan.

    That’s refreshing! I’m tired of attempts to derail/downplay “the classic comparative method.”

  133. Good for him, but to me that says that the state of the art in Khoisan is not yet adequate for proving or disproving larger-scale hypotheses.

  134. Skipping back to OP despite the entire fascinating discussion…

    A consequence of this is that those working on indigenous languages must seek to situate their work in a broader perspective. This might mean that the research becomes comparative within a language family or that the research is a case study within a broader survey on similar phenomena. Rather than magnifying more deeply, if they want their work to be considered by the field at large, linguists working on indigenous languages often take the “go wide” approach instead. […]
    Scholars working on well-studied languages benefit from a shared linguistic common ground with other scholars which allows them to delve into deep and specific questions within these languages.

    This is basically true, though the mechanics of it are not quite as simple as some shadowy publishers complaining if you get too detailed on too small a language. You can definitely publish work on reflexive clitics in Washoe, it’s just that you will be then unlikely to have any colleagues at all who know what you’re talking about, and even very few who know how this contributes into the big picture of Washoe grammar. So if you want your work to be read and cited, you’ll have to add other hooks for context. For some areas this often means a comparative (historical or areal) angle; for others, a typological one.

    (And note that, while this may not be intrinsically a disprivilege, in today’s academic climate having your work go unread and uncited will soon transform into the rather real problem of being unable to secure funding.)

    In most cases the traditional context-generating method has been by-family or by-area philological sorting: work on reflexive clitics in Hadza would not necessarily be “Khoisanist” comparative research, but you’d likely be still publishing it thru Khoisanist or at least Africanist publication channels. Isolates and understudied families are often going to get a short stick of even this though. They may fare better along the typological route however, or sometimes even the universal-grammar route, as most people will understand that e.g. Hadza is not a priori expected to work in a typical Bantu or typical Cushitic fashion and might be therefore quite interesting. Or cf. how it’s isolates like Basque or Sumerian that still attract fanciful etymology research, while almost no one bothers to do this with anything like Brabantian Dutch anymore.

    This can also compound: small understudied subfamilies are often going to be “doubly marginalized”. Good luck finding anything at all about Nuristanian even in dedicated Indo-Iranian journals, let alone general Indo-European ones…

  135. David Marjanović says

    Where is Nuristân, and Who Cares?

    Historical IEists might care, because apparently those languages don’t have the full ruKi, only riK… but good luck finding more about that.

  136. Ha. I remembered having a fleeting interest in those languages years ago, so I dug out my battered and much-annotated copy of Lockwood’s Panorama of Indo-European Languages and turned to the appropriate section, where I discovered that I had crossed out Lockwood’s “West Dardic” and written in “Nuristani,” and had inserted between the pages a printout of that very webpage you link to!

  137. David Marjanović says

    I forgot where I know it from – quite possibly hereabouts.

  138. The printout was made on 2/21/02, so almost exactly seventeen years ago (and pre-LH). The site’s been around for quite a while.

  139. Looking up the recent three-volume and nearly 3000-page Handbook of Comparative and Historical Indo-European Linguistics from de Gruyter, Nuristani only gets three paragraphs / less than a page and a half, tacked onto the chapter on Indic dialectology. (Most of it spent on discussing its anomalous RUKI properties.) At least the closing line goes “…and future fieldwork is a desideratum.”

  140. @David M Am I understanding this right – Chau Wu uses alignment algorithms …

    It appears you’ve already read and critiqued his paper. (I must have missed that LLog round a couple of years ago.) You probably then blocked out the memory.

    I’ll say it out loud: what Wu is doing is pseudoscience.

    In the same thread Prof Mair repeatedly compliments Wu’s paper, and continues to in recent threads.

    I fell across that old post while working through another paper in the SPP series that seems to suffer the same methodological afflictions and produce equally nonsensical derivations. I have to ask: does Prof Mair know what Philology is supposed to be, and how the Comparative Method is supposed to work?

    I’m trying to be sympathetic to the idea there was linguistic interplay between IE and Chinese, but to Hat’s point Mair seems to be doing his durndest to discredit the whole idea, by sponsoring work that’s no better than crackpottery.

  141. David Marjanović says

    It appears you’ve already read and critiqued his paper. (I must have missed that LLog round a couple of years ago.) You probably then blocked out the memory.

    The actual “paper” is so long that I only skimmed it (and didn’t download it). It’s easily possible that I missed that part.

    what Wu is doing is pseudoscience.

    Well, yes. Pseudoscience with unusually benign motivations, but pseudoscience.

    does Prof Mair know what Philology is supposed to be, and how the Comparative Method is supposed to work?

    “Philology” perhaps, as the word tends to be used nowadays (i.e. as the study of old texts). Historical linguistics, including the comparative method – evidently not. Which surprises me greatly, given how many eminent historical linguists he knows personally and quotes correspondence from.

  142. And now the Scottish dictator is dead, peacefully in a hospital bed: BBC article in West African Pidgin English. A Rawlings quotation from a linked article: “E get time wey we see di white man, our colonial powers as our enemies but something wey dey worst than an enemy na traitor. Wen we win our freedom, some of di nobel nationalist of our time na dem lead our kontris but as time pass, some of dem become treacherous and betray our aspiration.” Amen.

  143. David Eddyshaw says

    I’d have voted for him if I’d had a vote (I was in fact encouraged to vote in the parliamentary elections by one of the candidates, who seemed to feel that my ineligibility was not an insuperable problem. Another story …)

    Ghanaians never forgot that he came to power in a coup (and put his predecessors to death); but it’s a reflection of Ghana’s good fortune that Ghanaians could afford to remember this about a man who did a great deal of good for his country and stepped down gracefully after losing fair elections.

    In his early years, there was repression; but by the time I lived in Ghana, the country boasted a free press, entirely unafraid to criticise Rawlings and his government often in quite scurrilous terms. I vividly recall crossing from Togo to Ghana by bus on one occasion and noticing that as soon as we crossed the border, people started to talk freely. It was as if everyone had stopped holding their breath.

    Rawlings never had a personality cult, and I never saw a picture of him in a government office. And he remained a Flight Lieutenant in his days of military rule.

    I was on the hospital board on one occasion when a dismissed employee had managed to get Rawlings’ (unpopular) wife onside to support her. My Ghanaian colleagues (of whom I was never prouder) refused to bend; but the episode also tells you something to the credit of Rawlings.

    And he was Scottish. RIP.

  144. Marianne Boelscher Ignace FRSC (born 1954) is a Canadian linguist and anthropologist. As a member of the Shuswap people, she is a Full professor in the departments of Linguistics and Indigenous Studies at Simon Fraser University (SFU), and Director of SFU’s Indigenous Languages Program and First Nations Language Centre. In 2020, Ignace was elected a Fellow of the Royal Society of Canada for her work in revitalizing and preserving indigenous languages.

    https://en.wikipedia.org/wiki/Marianne_Ignace

  145. But how is her surname pronounced?

  146. Fellowship of the Royal Society of Canada (FRSC) is an award granted to individuals that the Royal Society of Canada judges to have “made remarkable contributions in the arts, the humanities and the sciences, as well as in Canadian public life”.

    For a moment I thought it was some weird Interior Salish name.

    The professor’s surname is boring Ignace. I think if it’s French, it has to be pronounced as /i.ɲas/

  147. Athel Cornish-Bowden says

    Maybe I should start keeping a list – it’s not the first time I encounter on LH mentions of people who know people I know.

    I mentioned in another thread today that my aunt used to know Klaus Fuchs (the spy).

    She also worked with Anna Freud in the 1940s (who set her onto a career in child psychology), but I don’t know if she ever met Sigmund. She never married or had children, and my mother wasn’t too impressed with her ideas of child psychology. Her sister, another aunt, didn’t study child psychology, but had five children. My mother said she knew which one she’d consult on a question of bringing up children.

  148. The professor’s surname is boring Ignace. I think if it’s French, it has to be pronounced as /i.ɲas/

    You may find it boring, but I have no idea how to pronounce it — I can think of several plausible ways. And she was born in Germany, so no, not French.

  149. Athel Cornish-Bowden says

    In his early years, there was repression; but by the time I lived in Ghana, the country boasted a free press, entirely unafraid to criticise Rawlings and his government often in quite scurrilous terms. I vividly recall crossing from Togo to Ghana by bus on one occasion and noticing that as soon as we crossed the border, people started to talk freely. It was as if everyone had stopped holding their breath.

    Chile under Pinochet was in many ways less repressive than other South American dictatorships of the time. A couple of days after an attempt on his life in 1986 that almost succeeded I flew from Santiago to London. In the plane a Brazilian woman asked me what steps had been taken in the aftermath of the attempt. I said that the main thing was that the opposition magazines had been banned. She was amazed: you mean that there were opposition magazines? In fact when we arrived at the airport my wife’s cousin asked the man at the news stand if he had anything more interesting than government propaganda. Yes, he said, I have all of them, and produced six magazines from under the counter, which my wife’s cousin bought.

    Even in the worst period, still in progress when I was first in Chile in 1978, there was one opposition magazine, Mensajes, published by the Jesuits, that operated throughout the period of dictatorship. It was only rarely closed down (twice, I think, for short periods). I bought a copy on the street, because it said on the cover that there was an article about Sheila Cassidy, a British doctor who was arrested and tortured. At that time I knew almost no Spanish, but you don’t need to know much Spanish to know what words like tortura, eléctrica and vagina mean. (Much later I came to know the editor of Mensajes very well, and usually see him when we’re in Chile.)

    Anyway, returning to Jerry Rawlings, you have to admire someone who never promoted himself higher than Flight Lieutenant.

  150. SFU President’s Faculty Lecture: Dr. Marianne Ignace

    https://www.youtube.com/watch?v=l_lMf_Q5Yvw

  151. Thanks! So it’s /ˈɪgnɪs/ (“Igniss”), which was one of my guesses.

  152. David Marjanović says

    And she was born in Germany, so no, not French.

    That doesn’t follow – our attitude to seeing such names in writing is “doesn’t look German, looks French, so we’ll pronounce it in French”. Not that people will actually get the [ɲ] right, but I bet many would go straight for /ɪnˈjas/.

    The exception that comes to mind is Sarrazin, and that may in part be an attempt to escape the nasal vowel.

  153. I wasn’t saying it couldn’t be pronounced à la française; I was responding to “I think if it’s French, it has to be pronounced as /i.ɲas/.” If she was born in Germany, it’s a German name, even if of French origin. And that’s not obvious on its face; why couldn’t it be a Slavic name, etwa “ig-NAH-tse”?

  154. David Marjanović says

    Because -ce hardly occurs in surnames. (Unlike -ec, which does sometimes mislead people.)

    Ibiza, however, is always pronounced like a Czech or “Yugo” *Ibica in Austria.

  155. Ignace is specifically French version of the name.

    Slavic forms are different: Ignacy, Ignát, Ignác, Ignácz, Ignacij, Ihnat, Ihnot.

    Ignat is Russian (somewhat old fashioned name).

    Ignatiev is one of the very common Russian surnames. (Count Nicholas Ignatieff was involved in the Great Game and features prominently as a villain in Flashman novels. His great-grandson Michael Ignatieff lost an election and failed to become Prime Minister of Canada in 2011).

  156. Slavic forms are different: Ignacy, Ignát, Ignác, Ignácz, Ignacij, Ihnat, Ihnot.

    Given that variety, a Slavic Ignace certainly doesn’t seem implausible. I wasn’t saying there was such a name, just that without doing more research than I felt like doing, it seemed like a possibility.

  157. Ibiza, however, is always pronounced like a Czech or “Yugo” *Ibica in Austria.
    Same in Germany.

  158. Athelstan John Cornish-Bowden says

    Ibiza, however, is always pronounced like a Czech or “Yugo” *Ibica in Austria.

    Czech to the extent of having initial stress? How strange.

    What about the z: [z] as in English, [ts] as in German, [s] as in Catalan or Latin American Spanish, or [θ] as in Madrileño? Surely not [θ], as most German speakers can’t make that sound?

  159. I assumed DM was talking about the “z” being pronounced [t_s], not about stress. That’s anyway what my comment on German referred to. As for the stress, I’ve heard Germans stressing it on the first syllable, although stress on the second syllable is more frequent in my experience.

  160. David Marjanović says

    Both: z as [t͡s], which allows the Slavic analogy in the first place, and initial stress as in Czech and the FYLOSC standards but not as in Polish.

    most German speakers can’t make that sound

    Most under 40 can; [ð] is the one that continues to be out of reach for most.

  161. It can be Ignjat as well in “Slavic”.

    Going on a tangent: Ignaz Fuchs “croatianised” his name to Vatroslav Lisinski and became a famous composer.

  162. David Eddyshaw says

    [ð] is the one that continues to be out of reach for most.

    Why the difference from [θ]? Presumably not a question of frequency, or it would be the other way round …

    Almost certainly entirely irrelevant, but while I’ve never had much difficulty with [k͡p], I found [g͡b] a good bit trickier initially; doesn’t seem to be just me, either, inasmuch as there are a number of actual West African languages which only have the voiceless labiovelar stop. Now I think of it, [gʷ] seems to have presented more of a problem than [kʷ] for (the immediate linguistic predecessors of) the Romans and the Goidels too …

  163. David Marjanović says

    Why the difference from [θ]?

    I don’t actually know how far north that goes. Far enough south, the most obvious reason is that voiced fricatives (indeed voiced obstruents generally) are a foreign concept there. But beyond that, [ð] is less audible than [θ], and in a lot of Englishes it actually turns into [d] allophonically, though I haven’t seen that described and haven’t worked out the conditions myself… and even the most voicing Standard German accents are rather short on voiced fricatives: [z] is there, [ʒ] is a loanword phoneme some people have, [ɣ] was widespread in the northern half of the dialects but has been kept strictly out of the standard since the mid-20th century, [v] doesn’t real (it’s [ʋ] for most, and for at least some others – like me – it’s a [ṽ] nasal enough to lack audible friction even though it’s articulated as a fricative) and is not in any “counterpart” relation with /f/, unlike /z/ (with /s/) and unlike the English /v/ (or the Dutch one where it exists).

    [gʷ] seems to have presented more of a problem than [kʷ]

    Indeed, *[gʷʰ] was conspicuously rare in PIE.

  164. I knew an Urdu speaker who could produce [f] and [ʋ] (or was it [w]?) but never learned to make a [v].

  165. [z] as in English

    Depends on whose English. Brits seem to use /θ/; I myself would use /s/, but then I’m not ignorant enough to be authoritative on Merkin pronunciation. In any case the etymology is < Catalan Eivissa < Arabic يَابِسَة‎ yābisa < Latin Ebusus < Phoenician ????????????????????????‎ ybwšym /yibbōšīm/ ‘dedicated to the [Egyptian] god Bes’, defender of households and protector of mothers and pregnant women.

  166. Arabic yābisa < Latin Ebusus—how?

  167. David Marjanović says

    What is an Egyptian god doing in Phoenician?

  168. The region was closely allied with Egypt and therefore with its gods.

  169. Arabic yābisa < Latin Ebusus—how?

    I am not familiar with the facts of every stage of this etymology, but at first glance this particular stage looks like folk etymology of a Romance name.

    In Modern Standard Arabic yābisa, means “terra firma, land” (doubtless short for الأرض اليابسة al-ʾarḍ al-yābisa “dry land”). It would make sense as a designation for an island… I am unsure about the antiquity of this meaning in Arabic, however. It seems the earlier meaning was simply “arid, desiccated land”, which could also fit for Ibiza. In form, yābisa is simply the feminine of yābis “dry, arid, desiccated, compact, firm”.

    (For the ubiquity of such expressions for “dry land”, compare Irish tír and Latin terra.)

  170. Ah. And cf. Hebrew יַבָּשָׁה yabāšā ‘dry land’ in Genesis 1:9–10.

  171. @David Marjanović: While Bes is most famously associated with Egypt, he almost certainly originated elsewhere—making the dwarf a wandergott, if you will. His conventional depictions, even in Egyptian art, are extremely atypical (ithyphallic and shown from the front, rather than in profile) and only become prominent in the mid-to-late Middle Kingdom. I am not suggesting that he was Semitic in origin; scholarly opinion seems to be that it is most likely that Bes he came to Egypt from elsewhere in Africa, perhaps from the upper Nile. However, interest in Bes definitely extended beyond Egypt proper, and he makes plenty of appearances in the Egyptian-influenced Levant. Later on, Bes was one of the limited number of “Egyptian” gods who had prominent roles in Imperial Roman syncretism (although certainly Bes was far less popular than the Romanized version of Isis).

    Perhaps as an unintentional nod to Bes’s naturally syncretic character, the first edition Advanced Dungeons & Dragons Manual of the Planes ends its discussion of the Outer Planes with this note:

    The dwarven god of luck, Vergadain, is suspiciously similar to Bes in both appearance and portfolio (luck and wealth), yet he maintains a separate household and realm. Both deny that they are one Power, yet the times when they have been seen in one place are very few (and prove nothing when dealing with Powers of this caliber).

    Vergedain was a completely de novo creation (appearing in one of the several batches of “non-human deities” created by Dragon magazine contributor and later TSR employee Roger Moore), unlike most AD&D gods, who were based on real-world mythologies. There was also an earlier (although not actually very interesting) discussion in of the Dragon letter columns of whether Bes made sense as a deity to be worshipped by dwarven characters.

  172. David Marjanović says

    Fascinating.

  173. David Eddyshaw says

    Are you doing the eyebrow thing?

  174. I was wondering that too!

  175. David Marjanović says

    Wasn’t it just yesterday that I spelled out “*pretends being able to raise one eyebrow*”?

Trackbacks

  1. […] Hat takes note of some of the mechanisms by which linguistics can neglect the study of indigenous […]

Speak Your Mind

*