Japanese Dialects.

Victor Mair at the Log posts about Kobayashi Takashi’s “Linguistic Treasures: The Value of Dialects“; it begins:

Japan has a great number of dialects. One scholar divides the archipelago into 24 areas with distinct regional linguistic forms. Yet, this is a very broad classification, and if one pays attention to variations in grammar and specific words, it is no exaggeration to say that there is a dialect for every city, town, village, and hamlet.

These were once looked down on for their association with uncultured, provincial speakers. In recent years, however, dialects have been increasingly appreciated for the pleasure they can bring to verbal interaction and their ability to draw people from a particular area closer together. This has led increasingly to their use in the names of local products, and their incorporation into plays and TV dramas.

Notably, the 2013 NHK morning drama Ama-chan (Little Diver) brought the expression of surprise jejeje to national attention, which was one element in its success. The phrase je comes from just one part of Kuji in Iwate Prefecture, although ja is in use over a much wider area centered on Iwate. Other unique ways to voice one’s amazement that are found in dialects but not standard Japanese include waiha, sāsa, ūu, and chopped off forms of da and ba. The wide range seen even in this category of utterance demonstrates how rich in dialects the country is.

Where did these differences originally arise? The main source of dialect forms comes through transmission of language from the center to the regions. New words in former capitals like Nara and Kyoto gradually spread through neighboring areas to the wider nation. These successive waves created regional linguistic differences, or dialects.

This means that many dialect words derive from standard terms in more ancient forms of Japanese. For example, the word menkoi in Tōhoku, equivalent to kawaii (cute, adorable), comes from megushi, a word seen in the Man’yōshū poetry collection from the Nara period (710–94). Churasan, an Okinawan word for “beautiful,” is related to kiyora nari, seen in works from the Heian period (794–1185) like The Tale of Genji.

Kobayashi goes on to discuss their role in people’s lives (“They build mutual understanding that goes beyond the use of language as a communication tool, and are essential to creating community ties”) and threatened dialects (“The Center for the Study of Dialectology is working to record conversations showing how dialect is used in everyday life”). Mair ends by mentioning variants of spoken language in China (“following Chinese custom I refer to them as fāngyán 方言”); I’ll end by passing along this beautiful chart of languages and dialects in late tsarist Russia (the Indo-European, Semitic, and Ural-Altaic families have languages, the rest merely нарѣчія ‘dialects’).

Comments

  1. Is it just me, or is Finland the only instance in that chart of a specific region being singled out geographically? I wonder why they didn’t just make Finnish part of Ural-Altaic instead. (Oddly, the chart also lists Swedish twice under both “Germanic languages” and “Finland”.)

  2. Huh. Well, Finland was a semi-autonomous part of the empire, so I guess that’s part of it.

  3. I felt that chart has a very 19th C. appearance (the colours perhaps) and lo! and behold, searching the bottom, where dates would appear on maps, found the date 1891.

  4. David Eddyshaw says

    By sheer free-association this reminded me of another rather pretty language map in Russian:

    https://upload.wikimedia.org/wikipedia/commons/f/f1/Gur-ru.png

  5. The main source of dialect forms comes through transmission of language from the center to the regions. New words in former capitals like Nara and Kyoto gradually spread through neighboring areas to the wider nation. These successive waves created regional linguistic differences, or dialects.

    I know that innovations (in vocabulary and otherwise) spread out in waves from centres of influence, but how accurate is this as an overall characterisation of the development of Japanese dialects? It seems just a little too simplified, but I am speaking as someone who doesn’t know enough about the development of Japanese dialects to say otherwise. (Words like “nation” make me uncomfortable, to be honest; they sound like projections from the present to the past. “Country” would make me feel less uncomfortable.)

  6. Is it just me, or is Finland the only instance in that chart of a specific region being singled out geographically? I wonder why they didn’t just make Finnish part of Ural-Altaic instead. (Oddly, the chart also lists Swedish twice under both “Germanic languages” and “Finland”.)

    They did include Finnish in Ural-Altaic, in between Karelian and Estonian. The Grand Duchy of Finland was an autonomous entity felt to be separate from the rest of the empire, so it was treated separately on this chart. So Finnish, Swedish, and Russian are all listed twice, once on the main chart and once for Finland.

    By the way, I was not familiar with the form of ѣ (yat) they used in the chart.

  7. @Bathrobe: I think “nation” is the right term in that context (in terms of meaning; I know nothing about the accuracy of the claim). To the extent that they mean different things at all, the word nation denotes a division of people, while country denotes a division of territory (and state one of governance).

  8. Mair ends by mentioning variants of spoken language in China (“following Chinese custom I refer to them as fāngyán 方言”);

    I was very puzzled by Mair’s remark when I saw it. (But I’m by now pretty much skunked by Prof Mair on LLog.) Perhaps I can ask the brains trust here.

    Mair actually added (“topolects”) as his translation of fāngyán. AFAICT that’s his own coinage; fāngyán is usually/Chinese custom is to translate as ‘dialects’. ‘dialects’ is very misleading for varieties of Chinese, because many of them are mutually incomprehensible (even where they descend from proto-Han). So they’re at least as diverse as Italian/Spanish/Portuguese/Sardinian/etc descended from Latin.

    Is that the same situation wrt these varieties of Japanese? I’m excluding languages in Japan that are clearly not related, like Ainu. Kobayashi seems to me to be talking about dialects and dialect words in the sense of ‘dialects’ familiar in (say) English, and not mutually incomprehensible ‘varieties’.

    There’s an extra complication with varieties of Chinese, that does justify avoiding ‘dialect’: they don’t have independent/well agreed writing systems; there’s a mish-mash of romanisation (with extra numerals to identify tones extra to those in putonghua), or repurposing Simplified and/or Traditional characters, or adding topolect-specific characters.

    Kobayashi’s article uses romanisation for the dialect words. Does that mean they can’t be represented in any Japanese script? Or that they’d be ambiguous between the dialect pronunciation and standard Japanese?

    Ah, I see the article Originally published in Japanese on February 26, 2020. How does the original represent the dialect words/pronunciations?

  9. Re: nation and dialects

    One main takeaway from my study of Japanese history which doesn’t get mentioned often, if at all, and which I had to figure out myself was that Japan throughout history only pretended to be a unified state like China or Korea, but in reality remained a conglomerate of de-facto independent or semi-independent principalities, recognizing nominal authority of emperor and shogun.

    Even Tokugawa shogunate, as late as 1860s, was just that – a sort of Far Eastern Holy Roman Empire. Granted Tokugawa shoguns were a bit more powerful than Habsburgs and grew even stronger over time (perhaps the policy of isolation worked).

    Only Meiji revolution of 1868 created a unified Japanese state.

    I believe this state of affairs has everything to do with development of Japanese dialects. Just like German dialects had a lot to do with political fragmentation of Germany until late 19th century.

  10. @AntC: Kobayashi seems to me to be talking about dialects and dialect words in the sense of ‘dialects’ familiar in (say) English, and not mutually incomprehensible ‘varieties’.

    Here is what he says:

    UNESCO maintains a list of world languages in danger that includes seven forms from different island clusters now part of Okinawa Prefecture and Hachijōjima, a far-flung part of Tokyo, considered to be dialects in Japan.

    So he concedes that what are considered “dialects” in Japan are called languages by authorities elsewhere. He clearly is including Ryukyuan varieties in his discussion of Japanese “dialects”. He mentions the Okinawan word ちゅらさん churasan, so he is not talking about the Okinawan dialect of Japanese (which is what most Okinawans speak today), but about the Okinawan language itself, which is not intelligible to Japanese speakers.

    @Bathrobe and @Brett: I also find “nation” a bit anachronistic in this situation, considering that Ryukyu was not annexed by Japan until the late 19th century. But it is common to talk about nations in this way, as if a German nation or a Russian nation has always existed.

  11. Dave Lovely says

    I’m minded of Max Weinreich’s quip “A language is a dialect with an army and a navy”…

  12. @SFR Just like German dialects had a lot to do with political fragmentation of Germany until late 19th century. and @Jongseong as if a German nation … has always existed.

    Thanks yes is the variation in German comparable to in Japanese? My understanding is that even if varieties of German are mutually incomprehensible (are they?), there’s a neutral variety or ‘Received Pronunciation’ as a lingua franca. That’s not the case with varieties of Chinese: many in Hong Kong could not understand putonghua until the PRC takeover became inevitable.

    And yeah we can’t take politics out of it. Was there always a German sprachbund? Ryukyuan is mutually unintelligible with, but a member of the Japonic family, says wp. So it’s more closely related than Ainu?

  13. David Marjanović says

    I know that innovations (in vocabulary and otherwise) spread out in waves from centres of influence, but how accurate is this as an overall characterisation of the development of Japanese dialects?

    All I know is that it’s accurate for the distribution of Kyoto-type tone systems as an island in the Tokyo-type ones: that dates from when Kyoto was the capital.

    Sometimes, of course, a “centre” of innovation is geographically marginal.

    By the way, I was not familiar with the form of ѣ (yat) they used in the chart.

    It’s the handwritten and italic form, but I had not seen it in normal print before.

    And in the italics on the chart, the ó-like form of б with the gap in the middle was new to me, too.

    I believe this state of affairs has everything to do with development of Japanese dialects. Just like German dialects had a lot to do with political fragmentation of Germany until late 19th century.

    There’s also the sheer size of Japan and all the mountains in it.

    My understanding is that even if varieties of German are mutually incomprehensible (are they?), there’s a neutral variety or ‘Received Pronunciation’ as a lingua franca.

    Knowledge of Standard German has always been limited to the educated classes. By now that’s everybody, though there are certainly people in Switzerland who are not downright fluent in it. Its vocabulary, grammar and sound systems are different from those of most or all dialects (the actual sounds used to fill in the abstract slots in the sound systems tend to be close to the average of dialects of larger regions, but even they are not all identical to any sounds of any particular dialect… plural for the different regional sound systems of Standard German, but they’re all close to each other).

    Because everybody is now familiar with Standard German, the mutual comprehensibility of the dialects is hard to assess now. The dialects of most of Austria and most of southern Germany are mutually comprehensible, likely with some help from everyone’s knowledge of the standard (which has different things in common with every dialect). But when the Swiss speak, I understand about half when I really pay attention – in other words, it might as well be Flemish.

    Was there always a German sprachbund?

    The short answer might be “only since Luther”; the long answer might be “yes, to a small extent”…

    So it’s more closely related than Ainu?

    Oh yes. Ainu is very different from… everything. Only long-rangers don’t simply consider it an isolate, and they regard it as part of the Austric macrofamily, together with Austronesian (the non-Chinese languages of Taiwan, and Malayo-Polynesian), Kra-Dai aka Tai-Kadai (e.g. Thai) (probably a member of Austronesian, actually), and Austroasiatic (e.g. Vietnamese, Khmer, and the Munda languages of India). English is closer to Japanese than that.

  14. Ainu is not related to any other language as far as we can tell. Ryukyuan languages are clearly related to Japanese, even if they are not mutually intelligible.

    Varieties of “German” are definitely mutually incomprehensible. They are not simply different pronunciations of the same language. But there is a Standard German ultimately based on the written language of Luther’s Bible. In many parts of Germany, Standard German has already replaced historical varieties, but in Switzerland for example people speak in Swiss German in their daily lives while reading, writing, and speaking to outsiders in Standard German. In terms of spoken language, this is not much different from Hong Kongers today using Cantonese in their daily lives while speaking to outsiders in Mandarin (even if the attitudes towards the languages are a bit different).

    The kingdom of Ryukyu was invaded by the Japanese feudal domain of Satsuma in 1609, who were acting independently of the Tokugawa Shogunate (and without their knowledge). So even as they became vassals of the Satsuma lords, they continued to speak Okinawan (specifically the Shuri dialect of the Island of Okinawa) as their main language as they had done before. Ryukyu had a tributary relationship with China while Japan didn’t, so it was in the interest of Satsuma to preserve the status quo. The influence of the Japanese language began in this period and Okinawan stopped being used in official documents, but people continued to speak Okinawan and other Ryukyu languages.

    After the Ryukyu Islands were formally annexed by Japan in 1879, the government started education in Japanese and actively discouraged the use of Okinawan. It took a generation after this for Okinawans to start identifying themselves as Japanese. Now, Okinawan is mostly spoken by old people and is considered threatened.

    Okinawan is traditionally written in hiragana, with kanji sometimes thrown in (but not as much as in Japanese). However, when Okinawan words are quoted in Japanese, they are most often written in katakana to mark them as foreign.

  15. David Marjanović says

    To solve the urgent needs of language communication between doctors from other parts of China and local patients, the medical assistance team of Qilu Hospital of Shandong University compiled The Guidebook of Wuhan Dialect for Medical Assistance Teams, Audio Materials of Wuhan Dialect for Medical Assistance Teams and The Handbook of Doctor-Patient Communication within 48 hours after the team arrived in Wuhan.

  16. I know that innovations (in vocabulary and otherwise) spread out in waves from centres of influence, but how accurate is this as an overall characterisation of the development of Japanese dialects?

    My intuition is that it could be a fairly good explanation for the Japanese branch of Japonic, used on the main islands of Japan. However, because of the distances involved, it probably does not work very well for the Ryukyuan branch. The settlement of the Ryukyuan islands, though having occurred within a short enough time-frame for the Ryukyuan languages to constitute a single branch, seems not to have occurred in a straightforward linear fashion of starting with the closest island from north to south. The internal complexity of the Ryukyuan languages can’t really be accounted for with the wave model.

  17. Speaking of Weinreich, it’s funny how the Russian chart has 5 million hebrew speakers.

  18. And in the italics on the chart, the ó-like form of б with the gap in the middle was new to me, too.

    There’s no gap; it’s just that that part of the letter is printed with a thinner line that shows up very faintly in reproduction. It’s just an italic form of the б you see in the other labels, which also has a very faint line in that place.

  19. I just noticed that the Russian chart also lists a single Japanese-Korean language.

    It’s just an italic form of the б you see in the other labels, which also has a very faint line in that place.

    At least for the sans serif typeface, you see the line definitely connecting in the label for Albanian.

  20. I just noticed that the Russian chart also lists a single Japanese-Korean language.

    Wow, I had missed that. Bizarre!

  21. Let me rephrase my concern.

    Is it right to say that the whole of Japan once spoke a single variety of “Japanese”, and that dialects exist because certain innovations from the centre of influence failed to reach areas further away? That is, dialects are really earlier forms of the language that haven’t undergone changes that took place at the centre of influence.

    If that were the case, surely Tsugaru dialect would be similar to Kagoshima dialect, which it isn’t.

    Kobayashi seems to me to be talking about dialects and dialect words in the sense of ‘dialects’ familiar in (say) English, and not mutually incomprehensible ‘varieties’.

    When dialects were still dialects, differences among them were quite marked, with a major impact on mutual intelligibility.

  22. Is it right to say that the whole of Japan once spoke a single variety of “Japanese”, and that dialects exist because certain innovations from the centre of influence failed to reach areas further away?

    No, any language that extends beyond a small territory is going to have dialects. This is about how innovations spread from one dialect to another.

  23. David Eddyshaw says

    the Russian chart also lists a single Japanese-Korean language

    I’ve read that during the brutal Japanese occupation of Korea, the official Japanese scholarly (!) line was that Korean was a dialect of Japanese (a sort of horrible reductio of Weinreich’s famous dictum.)

  24. No, any language that extends beyond a small territory is going to have dialects. This is about how innovations spread from one dialect to another.

    That’s my understanding, too. But he stated quite clearly:

    The main source of dialect forms comes through transmission of language from the center to the regions. New words in former capitals like Nara and Kyoto gradually spread through neighboring areas to the wider nation. These successive waves created regional linguistic differences, or dialects.

    This sounds to me like a slanted view of Japanese, one that sees dialects as museums for ancient vocabulary from Nara and Kyoto, indirectly lending them greater prestige than they are normally accorded.

    Taken together with his use of the word “nation”, I feel that this gives a rather distorted (from the author’s point of view, more attractive or romantic) view of the nature and origin of dialects in Japan.

  25. Yeah, that’s sloppy writing to be sure; it’s hard to know what he actually knows/understands and what is lost in the wash of rhetoric. People do get romantic about this stuff.

  26. Athel Cornish-Bowden says

    Many years (probably 60) ago I read an article about the Ainu in (probably) the National Geographic. It included the preposterous suggestion that the Ainu language was related in some way to English! Among the bits of “evidence” presented was that the Ainu for water is “wakka”. That’s the only one I remember, and checking a word list today I see that “wakka” does indeed mean water.

  27. David Eddyshaw says

    any language that extends beyond a small territory is going to have dialects

    In an area about 50 miles square, Kusaal manages to have two major dialects sharing only about 84% apparent lexical cognates, which I think is broadly similar to Castilian vs Catalan.

    (I think this is actually the result of the depopulation alongside the White Volta river caused by river blindness: the Bisa language to the direct north of Kusaal has a similar surprisingly large divergence between its eastern and western dialects.)

  28. David Eddyshaw says

    John Batchelor, who was one of the pioneers of Western study of Ainu, tried to connect it with Indoeuropean; I think it was all muddled up with the usual stuff confusing genetic origin and language origin, with the Ainu being assigned the role of “Aryans”, or “Caucasians” or what have you. On account of hairiness.

    The numbers “two, three” are tu, re, which is of course the sort of thing that immediately strikes amateur comparativists (like Batchelor) as highly significant.

  29. Behold, the Ainu my brother is a hairy man…

  30. David Eddyshaw: I’ve read that during the brutal Japanese occupation of Korea, the official Japanese scholarly (!) line was that Korean was a dialect of Japanese (a sort of horrible reductio of Weinreich’s famous dictum.)

    But would this already have been the case at the time of the 1897 census that the chart is based on? That was the same year that the Korean Empire was established and years before it was reduced to a Japanese protectorate (1905) and formally annexed to Japan (1910). William George Aston did suggest in 1879 that Japanese is related to Korean, but it’s difficult to find any claims regarding the genetic affinity of the two—whether in the context of the Altaic hypothesis or not—before the 20th century.

    Bathrobe: If that were the case, surely Tsugaru dialect would be similar to Kagoshima dialect, which it isn’t.

    The choice of the Tsugaru dialect in this example is an interesting one because of the question of when Japanese expanded into this area. The Tohoku (Northeast) region was originally inhabited by the Ezo or Emishi people. If they were indeed a separate ethnic group akin to the Ainu and spoke a non-Japonic language, then we can infer that Japanese has only been spoken in the area for about a thousand years or so.

    Viewing regional dialects as repositories of older words and expressions that are now lost in the standard language seems to be a reasonably widespread trope. I’m sure I’ve seen it in discussions of Korean dialects, often as a way of justifying the value of these regional forms. The way the author ascribes the dialectal differences as a consequence of transmission of new words from the centre to the regions, which is clearly problematic as Bathrobe points out, might be an extrapolation from this common idea of the regions as the museum of the old.

  31. Behold, the Ainu my brother is a hairy man…

    That was, by the way, how the Russians called them – “mokhnatye kuriltsy” (‘hairy men of the Kuril islands’)

  32. David Eddyshaw says

    As I think I may have vouchsafed before, when I was at university I knew a fellow-student of Ukrainian extraction called Mokhnaty. If only I had known! An opportunity now gone for ever …

  33. The Emishi people I mentioned in my previous comment, the original inhabitants of the Tohoku region and surrounding areas, were originally called 毛人 or “hairy people” in kanji, pronounced as えみし Emishi. Nowadays the ethnonym is written 蝦夷 and pronounced えぞ Ezo, believed to be a continuation of the same name with a shift in the pronunciation.

  34. From Ecclesiastical English: A Series Of Criticisms Showing The Old Testament Revisers’ Violations Of The Laws Of The Language: Illustrated By More Than 1000 Quotations: Being Part II Of The Revisers’ English by George Washington Moon (p. 68):

    Can any one tell me why the Revisers have described Esau as “a hairy man”, and Elijah as “an hairy man”? Was it because it was considered that, in Elijah’s case, the h should be dropped, “’airy” being a more appropriate description of him who “went up by a whirlwind into heaven”? 2 Kings ii. 11. For my own part, I consider such jokes as quite out of place in the Bible. Let us, then, leave this specimen of lack-wisdom clerical levity (no pun intended), and resume our criticisms.

  35. David Eddyshaw says

    I think I’d rather be a Hairy Person than a Prawn Barbarian, but that may be just me.

  36. I’m sure I once saw an encyclopedia entry or some such describing the hairy Ainus of Japan, but perhaps I’m making it up.

  37. January First-of-May says

    Viewing regional dialects as repositories of older words and expressions that are now lost in the standard language seems to be a reasonably widespread trope.

    Well, it’s true with the caveat that the same is also true the other way around; each variety of the language, including the standard one, preserved its own incomplete fraction of older words and (especially) expressions. It’s just that the ones that were preserved in the standard language aren’t perceived as “older”, because they’re also the current ones.

    IIRC, this is fairly easily seen in cases where two relatively closely related varieties ended up with their own standards, such as Czech and Slovak, or Spanish and Portuguese: either variety includes many words and expressions that are long obsolete in the other. To a lesser extent the same is true of British English and American English.

  38. David Marjanović says

    the Revisers have described Esau as “a hairy man”

    Is this about the Revised Version?

    In the KJV, I thought, the h-dropping is consistent (an hundred, mine hand…).

  39. There is some recent genetic evidence of a few similarities between Ryukyuans and Ainu, including a bit more hairiness IIRC, that suggests they were both original inhabitants pushed aside (north and south) by the new population that migrated from the Korean peninsula into Kyushu.

  40. Japan and Korea both have strong state ideologies of “one-language, one-nation” that lead many prominent linguists in each country to insist that the languages spoken in the Ryukyus and on Cheju Island are just dialects of Japanese and Korean, respectively. However, both governments are funding efforts to document those “dialects” that show how very different they are.

    Some of the old regional dialects of Japan are barely intelligible to speakers of Standard Japanese. Satsuma (Kagoshima) was famously so, partly by design, perhaps, to frustrate Tokugawa spies. The phonology of old Aomori dialect (northern tip of Honshu) is also very hard for most Japanese to understand, very truncated (not to waste warm breath in freezing air, according to some local lore).

  41. John Cowan says

    I think what is going on in Japan (this is bits and pieces put together and probably distorted by memory) is that almost all Japanese dialects are secondary dialects or what David M calls mesolects, the result of mixing the speech of the capital (whichever it is) with the local primary dialect, so that the results are closer together than they used to be.

    This is certainly what has happened in AmE, where the old dialect boundaries in the East have vanished as far as lexis is concerned, but remain in almost exactly the same place in the form of accent boundaries. Similarly, as is mentioned there, modern Romanesco is a secondary dialect derived from Tuscan settlement, hence the business about Standard Italian being a lingua toscana in bocca romana; varieties closely related to Old Romanesco are still spoken outside of Rome.

  42. I just noticed that the Russian chart also lists a single Japanese-Korean language.

    This reminds me of some commonly used designations involving groups speaking dissimilar languages (e.g. “Turco-Mongol” and “Anglo-Norman”). Nevertheless, after looking up “японский-корейский” and not finding any other references to this “Japanese-Korean language”, I think the term is probably just a one-off slip-up by a linguistically uninformed chart-maker.

  43. Sounds very likely.

  44. I find it amusing that the native language of the Anglo-Normans was not, as the name might suggest, West Germanic or North Germanic. Rather, it was an only distantly related Romance language. And most Turco-Mongols were probably Persianized Indo-European speakers, what’s more, with Arabic as their liturgical tongue.

  45. Can any one tell me why the Revisers have described Esau as “a hairy man”, and Elijah as “an hairy man”?

    I presume we are referring to an hairy man.

  46. A much-loved bit around this household, but he has transferred Elijah’s haspirate to Esau for what I believe is called comic effect. Let us all reflect upon the sacrifices we make daily, nay hourly, in the pursuit — the vain pursuit, it might all too often be called — of the approbation, the applause I might say, of our fellow man.

  47. Stu Clayton says

    Not even contrarians are spared ! A sobering thought.

  48. Europeans at one stage had an obsession with seeing Europeans lurking everywhere. One sometimes suspects it was a way of lowering other cultures’ position in the hierarchy. Whether it’s the Ainu in Japan, the Indo-Europeans in western China (who contributed so much ancient Chinese culture — Victor Mair loves this kind of speculation), or the Indo-Europeans in India (who became the Brahmin class), there is a strange fascination with this sort of thing. Of course it provokes a backlash, especially in places like China and India, because proud cultures don’t like to be told that even their history and origins belong to West Europeans.

  49. Stu Clayton says

    Due to the appearance and phonotactics of Indo-Europeans, they stuck out like sore thumbs in India and China. It must have been hard to lurk, so to the extent they succeeded they deserve some credit. Yet they are widely condemned for sticking their impudent hegemonic noses into other people’s business.

  50. David Eddyshaw says

    In West Africa it was the Fulɓe who had the dubious honour of being thought less local than other, less worthy, locals; though they were merely Hamitic, that was felt to be a step up from being hardcore African-African.

    The definitive rubbishing of this particular piece of nonsense seems to have been the work of August Klingenheben (who I’m sorry to say was an actual Nazi, but uncomfortably for moral tidiness was also a great scholar and – in this case – absolutely right.)

    Incidentally, the Fulfulde word for “ten” is sappo, which is transparently the origin of the Cantonese sahp; it surprises me that Prof Mair has not posted on this. I would be happy to provide a guest post on Language Log.

  51. David Eddyshaw says

    Hah! Esau is 蝦夷! It all fits, I tell you. Oh yes!

  52. Stu Clayton says

    You people always assume knowledge of Chinese, but I am forced to analyze the glyphs by ad hoc inspection guided by intuition. In 蝦夷 at the left I see a little guy kicking a man, presumably his father, who is kissing a woman (in a skirt). At the right are iron bars. Apparently this means “the wages of adultery are prison”. What this has to do with Japanese dialects is not clear.

  53. David Eddyshaw says

    I am innocent! I assume only knowledge of Japanese (reasonable enough, surely?)

    蝦夷 “Prawn Barbarian”, pronounced Ezo. Obviously.

  54. Stu Clayton says

    But Chinese glyphs have lots of strokes packed together, Japanese signs look like shorthand. Isn’t that how you tell them apart ? The Chinese bits are inserted in Japanese as “loan words” ?

    I hope I will not be obliged to revise everything I know about Asian cooking.

  55. David Eddyshaw says

    Only on Wednesdays.

    In a better and kinder world, the Japanese would have used Chinese glyphs only to write the multitude of Japanese words either borrowed from Chinese or made up from Chinese components on Japanese soil.
    Alas, in this sad sublunar sphere, the brave people of the Japanese archipelago long ago embarked on the adventure of writing native Japanese stems as if they were Chinese. You may well feel that this was unwise. But who are we to judge?

  56. Stu Clayton says

    Ah. No wonder I was misled. Seems as if, in the presence of ignorance, the best policy is silence.

  57. David Eddyshaw says

    You may be right. I have never tried that.

  58. Stu Clayton says

    By a happy stroke of providence, it is easier to discover ignorance in others than in ourselves. Imagine our plight if we had not even that to go on.

  59. Joel: Japan and Korea both have strong state ideologies of “one-language, one-nation” that lead many prominent linguists in each country to insist that the languages spoken in the Ryukyus and on Cheju Island are just dialects of Japanese and Korean, respectively. However, both governments are funding efforts to document those “dialects” that show how very different they are.

    Many Korean mainlanders still think of the speech of Jeju (Cheju) Island as a dialect of Korean, but the provincial government is promoting it using the name 제주어 濟州語 Jeju-eo “Jeju language”. For instance, in 2007 the Jeju Special Province government announced the Regulations for the Conservation and Promotion of Jeju Language (제주어 보전 및 육성 조례). This term only seems to have become mainstream in the 1990s, but is already used by most local organizations. So I think there is growing awareness in Jeju Island and in the linguistic community that Jeju is different enough to be considered a language of its own. But you can still encounter statements like “Jeju language is a dialect (bang’eon) of Korean”, so the change in terminology doesn’t automatically guarantee its status as a separate language.

    Not everyone has caught up, however. Just last November, the linguist Yi Sanggyu (이상규), who served as the director of the National Institute of the Korean Language from 2006 to 2009 and who is known for his work in Korean dialectology, spoke at a conference organized in Jeju Island on the subject of protecting and promoting the “Jeju language”. There, Yi called for a National Institute of Korean Dialects to be set up in Jeju Island to study Korean dialects like the endangered “Jeju dialect”. Newspaper reports said that Yi called for the protection of “Jeju language”, but his quotes in the article only talk about the “Jeju dialect”, so I assume that’s what he actually said. So a prominent linguist speaking at a conference that used the term “Jeju language” was still calling it “Jeju dialect” in 2019, perhaps out of life-long habit.

    But does this mean that Yi is denying that Jeju is a separate language? It’s not conclusive. The word he used in Korean is actually the Sino-Korean term 방언 方言 bang’eon which corresponds to fāngyán in Chinese. In its traditional usage in Korean linguistics, it should probably be understood as “regional speech”, much like in China (where the term originally even covered non-Sinitic speech). So in this sense, strictly speaking it is not contradictory to consider Jeju both a language and a bang’eon, even if it goes against the usual understanding that bang’eon is a subcategory of language. The outwardly baffling statement that Jeju language is a bang’eon of Korean should probably be understood in this traditional sense.

    Of course, with the growing influence of linguistic literature in English, bang’eon is being used more and more to mean “dialect” in the English sense. The shift in terminology from Jeju bang’eon to Jeju-eo may have been influenced by this development. I think however that the main reason Jeju islanders embraced the new term may have been one of prestige rather than linguistic considerations—”language” sounds more prestigious than “dialect”, which has the connotation of regional and backwards.

    During his directorship of the NIKL, Yi promoted the study of dialects. In 2006 there was a national survey of regional speech (지역어 地域語 jiyeok-eo), which included Jeju. It is perhaps notable that the survey used the term “regional speech” which sidestepped the issue of language vs dialect altogether.

  60. I find it amusing that the native language of the Anglo-Normans was not, as the name might suggest, West Germanic or North Germanic. Rather, it was an only distantly related Romance language. And most Turco-Mongols were probably Persianized Indo-European speakers, what’s more, with Arabic as their liturgical tongue.

    Another such term that recently came to mind is “Greco-Roman”. Looking up “Greco-Roman tongue” and “Greco-Roman language”, I found various sentences that in context seemed to refer to shared vocabulary rather than a unitary language (although in some of them, the phrase “Greco-Roman language” appeared to be used to refer to both languages as a whole).

    But I also came across this curious quote from a book published in 1878:

    The Grecian colonists, in common with all the other tribes above enumerated, came in the course of time to lose their original dialects; and, as early as the reign of Augustus, Latin was the spoken language of all Italy. The modern Italian, a soft, euphonious language, is more closely allied to the Latin than any other Greco-Roman tongue.

    I suppose in this case the word refers to a substratum-adstratum relationship, but one wonders what the “other Greco-Roman tongues” are supposed to be. Presumably they are the other Romance languages.

  61. I just noticed that there is another example in the chart of a single linguistic entry being described with multiple glossonyms; where Japanese and Korean are referred to as “Японско-Корейскій языкъ”, the description of Romanian includes a conjunction: “Молдаванскій и Румынскій”.

  62. Trond Engen says

    Where did these differences originally arise? The main source of dialect forms comes through transmission of language from the center to the regions. New words in former capitals like Nara and Kyoto gradually spread through neighboring areas to the wider nation. These successive waves created regional linguistic differences, or dialects.

    I was planning to add a “Surely this is exactly the other way around”, but I see the relevant points have been made already. So I go on to the actual examples:

    This means that many dialect words derive from standard terms in more ancient forms of Japanese. For example, the word menkoi in Tōhoku, equivalent to kawaii (cute, adorable), comes from megushi, a word seen in the Man’yōshū poetry collection from the Nara period (710–94).

    I don’t think this “comes from” can be any more accurate. Megu- seems like a possible descendant of menkoi-, but not the other way around. Common ancestry or the dialect form being more phonologically conservative than the received pronunciation of Nara poetry?

    Churasan, an Okinawan word for “beautiful,” is related to kiyora nari, seen in works from the Heian period (794–1185) like The Tale of Genji.

    Less to gripe with here. Chura- looks like a reasonable dialect (or descendant) form of kiyora-. I think the journalist may be using the terms “comes from” and “is related to” more or less interchangeably, and in about half the cases they fit.

    (But I know next to nothing about Japanese)

  63. I have no idea of the etymology of menkoi but megushi doesn’t look wrong.

    Megushi appears to be an adjective in the predicative form. In its attributive form I guess it would be meguki. In the modern language, adjectives have all changed to end in -i. (E.g., yoki (attrib.) and yoshi (predicative) are now yoi or ii). So megui would be the expected modern adjective.

    The second problem is the insertion of n. This is also found in certain Japanese doublets, like tobi ~ tonbi ‘kite’ and gobō ~ gonbo ‘burdock root’, the latter dialectal. I suspect that the insertion of n in mengui wouldn’t be considered that strange or abnormal. The only other problems are that of voicing and the second vowel, but these could also, I suspect, be resolved with reference to Tohoku phonology.

    At any rate, these are just points off the top of my head. Someone with proper scholarly credentials could give you a more definitive answer.

    There is a noun in standard Japanese, menkui, which refers to a strong preference for pretty women. I don’t think it is related since it appears to be etymologically derived from men meaning ‘face’ and kuu ‘to eat’.

  64. the insertion of n

    +kaba ~ kanba ‘birch’, amari ~ anmari ‘too (much)’

  65. Kisti is a Chechen dialect, not Ingush, although it’s complicated.

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

  66. More to the point (given that g is not a bilabial): togatte iru and tongatte iru ‘pointed’.

  67. I wonder if there is a good translation of togatta jinzai.

  68. “Togatta jinzai” refers to staff who are not boxed in by conventional thinking and can think and act outside the box — a type that is difficult for the traditional group-think of Japanese companies to accommodate.

    One website I found gives “unique, unconventional, avant-garde”.

    Another gives “shakers and movers”.

    Perhaps difficult to translate for cultural reasons.

  69. @Bathrobe: “Movers and shakers” doesn’t sound too bad for that meaning, but the words have to be in that order.

  70. (Don’t miss Jongseong Park’s comment above [March 28, 2020 at 2:26 am], which I just rescued from the spam bin.)

  71. David Marjanović says

    In West Africa it was the Fulɓe who had the dubious honour of being thought less local than other, less worthy, locals; though they were merely Hamitic, that was felt to be a step up from being hardcore African-African.

    Especially when “Hamitic” morphed into “migrated straight out of Ancient Egypt”.

    it surprises me that Prof Mair has not posted on this

    Contact Chau Wu, and that problem will be solved.

    I am forced to analyze the glyphs by ad hoc inspection guided by intuition.

    Most of them, unfortunately, never were pictures in the first place, but consist of a “classifier” part (the “radical”) and a hint at the pronunciation that made sense 3000 years ago but not so much nowadays (and of course not at all if used to write a native Japanese word). In 蝦, the left part is a squeezed version of the sign for “bug, varmin” and shows up a lot in the names of barbarian peoples.

    Japanese generally uses unmodified* Chinese characters for content words, the hiragana syllabary (drastically simplified from Chinese characters) for grammatical particles and rare words, and the katakana syllabary (drastically simplified from Chinese characters in a very different situation from the origin of hiragana) for emphasis, mostly for non-Chinese loanwords.

    * Except exceptions.

    I suppose in this case the word refers to a substratum-adstratum relationship, but one wonders what the “other Greco-Roman tongues” are supposed to be. Presumably they are the other Romance languages.

    Around that time (and occasionally later), there were actual historical linguists who thought Greek and Italic were each other’s closest relatives.

  72. The Russian map seems to have a fairly large assortment of language names no longer in use.

    “Чудское” among Finnic is presumably Veps. I do not recognize “Лопарское” but the numbers could match either Votic or Livonian, which are absent from the chart otherwise.

    “Жмудское” as the third Baltic language: is this perhaps an older Russian term for Žemaitian (Samogitian, Low Lithuanian?)

    The Turkic section has several names I don’t recognize, including “Тептярское” and “Сартское”.

  73. Лопарь is an old word for Sami (corresponding, I guess, to English Lapp). And yes, жмудь is an old word for Samogitians.

  74. I know Сарт as a (nowadays obsolete) Kazakh designation for the sedentary people of Central Asia, especially for the Uzbeks.

  75. Tajiks, Sarts, and Uzbeks at LH: 2011, 2015.

  76. /slaps forehead
    Yes, of course, I did think that sounded familiar for some reason… weird though that there’s no count from Finland then.

    Саамские I think has been standardized in Russian already in the 60s, i.e. decades earlier than the shift from Lappology to Sami studies in the Nordics.

    Another immediate sign of this being from the 19th century: Samoyedic is still given as an independent branch of Ural-Altaic.

  77. Jonseong Park: “By the way, I was not familiar with the form of ѣ (yat) they used in the chart.”

    That’s how my grandmother writes it in cursive. She still complains what a pain in the ass inkwells were in school, and the punishments for spilling ink. I picked up the old-style handwritten Я from her:
    “я” is an upside-down “ɤ” with a small loop at the end,
    and for the handwritten big yus (ѫ) you add an inside-curving “v” that crosses the main stroke on both sides at the bottom side.

  78. When I started school in 1990, our desks still had circular depressions for inkwells.

  79. “я” is an upside-down “ɤ” with a small loop at the end,

    For a moment I thought it would help me understand the squiggles in Овъёс written on a packet of Mongolian muesli. But alas, the offending letter is ъ, which looks like an ‘r’ in cursive style of the Latin alphabet, followed by a loop.

  80. Yes, handwritten ѣ and are sufficiently close to throw me for a loop first time I read a text written in “old orthography”. I do it so rarely that don’t remember ѣ‘s shape from one time to another.

  81. Stu Clayton says

    a packet of Mongolian muesli

    Is that a particular kind of mix ? Or merely a Mongolian packet of muesli ?

  82. Stu Clayton says

    Gemüse suggests that proto-Germans thought of vegetables as “mushies”, as something that must be cooked to a pulp to be edible. Müsli is from the same root, meaning dry mushies. An example of Swiss humor.

    I suppose this could be called instrumentalized perception – the Ding an sich can be seen only as its destiny. Like Wile E Coyote looking at the Roadrunner and seeing a roast fowl.

  83. Just ordinary muesli, made in Mongolia.

  84. PlasticPaddy says

    @stu, dm
    Could Gemüse have originally applied to cooked food/vegetables, as opposed to cold meals / salad? English and Dutch have a second generic “greens”, corresponding to Italian verdura. I think I have heard Grünzeug in German.

  85. including “Тептярское”

    https://fi.wikipedia.org/wiki/Teptjaarit

  86. It crossed my mind that it would be odd for 19th-century Russia to have ~5,063,000 native speakers of Hebrew, so I looked up other sources on the census. According to the Wikipedia article on the census, that number apparently applies to Yiddish, not Hebrew.

  87. Teptäärit would be much closer to the mark.

  88. @PP :
    According to Grimm, the original meaning was “gruel”, then “side dish served with bread” -> “side dish served with meat” -> “herbs and greens (as a side dish)” -> “greens”.
    Grünzeug exists, but it’s colloquial, slightly deprecative, and can also be used for animal fodder.

  89. Stu Clayton says

    @Hans:
    You are sidestepping the question of whether the “side dishes” were mushies. In “side dish served with bread”, the next “meaning” given after “gruel”, there is no claim or implication that the side dish was not gruel or gruel-like – mushies, that is.

    A more cautious interpretation of the second “meaning” is that a combination of two things was designated by a word that previously designated only one thing. Nothing is said about whether that one thing was one of the two things, or not.

    I presume the word you’re rendering as “side dish” is Beilage. Beilage says nothing about the consistency of the Beilage. It could be apple sauce (another mushy!), it could be a Caesar salad. Meat is unlikely to be called a Beilage to bread. But if it were, has Gemüse suddenly taken on the sense of “mix” instead of gruel ?

    I don’t see those “meanings” as saying anything about a decline in mushiness and an upsurge of interest in rabbit food.

  90. Овъёс written on a packet of Mongolian muesli

    The word is Russian, of course, and means simply “oats” or in this case “oatmeal”.

    Mongolian spelling reflects Mongolian pronunciation which in turn must be some sort of hypercorrection.

    I can just picture a grey-haired Mongolian agricultural sciences professor many years ago who convinced himself that he absolutely knows how this Russian word must be pronounced and taught this spelling and pronunciation to all his students.

  91. @Stu: I don’t know more about the matter than what I read in the DWB. At some point in their phase 2) or 3) the implication of things being gruelly or mushy was lost. As we all know, mushy Gemüse is nevertheless still a frequently encountered fact of life.

  92. Stu Clayton says

    What are these “lost implications”, given that overcooked carrots are still a fact of life ? They must have been so in days gone by, because Gemüse. Maybe the Old Ones actually liked them that way, knowing no other. Maybe the etymologista behind the DWB article have invested heavily in dandelion al dente, so they need to market it. They want people to believe that the sense of mushiness has gone, although the fact hasn’t.

  93. Stu Clayton says

    In one of the Indian “ready to eat” meals-from-a-pouch I was eating last week, there were a couple of fairly large round things that looked like seed pods from an exotic plant. There was surface structure, and the inside had the consistency of a fig.

    I had encountered these before. This time I took one apart. They’re not seed pods, but hand-made vegetaballs ! No mush there.

  94. Well, the thing is that at one point Gemüse came to be used also for fresh, uncooked vegetables, so no implied mushiness anymore. I agree that all examples under 2) may still have been mushy dishes.

  95. David Marjanović says

    I think I have heard Grünzeug in German.

    Literally “green stuff”; contemptive.

    In Vienna, markets for fruit and vegetables use the term Obst und Grünwaren. That latter word is unknown anywhere else to the best of my knowledge.

    It could be apple sauce (another mushy!)

    For those following along at home: Apfelmus. Few native speakers are aware that Mus is where Gemüse comes from.

  96. Lars Mathiesen says

    Grønsager/grönsaker is the neutral word in Danish and Swedish, but the supermarket section is called frugt og grønt. To a first approximation, grønsager is any (non-animal) produce you’d include in the salt kitchen, even fungi and seaweed, and frugt is for the sweet. Nuts are borderline. Botanical classifications are irrelevant.

    There is a concept of bladgrønt (‘leaf green’) where you can discuss if lettuce is a vegetable, but it’s green anyway so it goes in grøntafdelingen.

    Also, æblemos. And kartoffelmos. Swedes and turnips are best as mos too, I don’t blame the Old Ones. Any relation to E mash?

  97. Finländare says

    The numbers “two, three” are tu, re, which is of course the sort of thing that immediately strikes amateur comparativists (like Batchelor) as highly significant.

    As an interesting side note the Ainu adnominal numeral forms for one and two are sine and tu, and tupesan and sinepesan for eight and nine, the logic being “ten without two” (tup-e-san), “ten without one” (sinep-e-san).

    Eight and nine in Finnish also follow the same logic, eight being kahd-e-ksan (from kaksi, two), nine being yhd-e-ksän (from yksi, one).

    (Since it’s been established that modern Finns have a direct genetic link to ancient East Asian populations, could this be an obscure linguistic survivor of an ancient Northeast Asian Sprachbund affecting an earlier form of whatever these people spoke that has carried over to the Finnic languages? One can get fanciful at times, maybe due to too much self-isolation!)

  98. David Marjanović says

    Then I’d expect Yukaghir to be part of that. From looking at this list of 1–10 in 5000 languages (you have to load the whole list; then search it), it’s not.

  99. How do you get 19/29/39/etc from 20/30/40/etc, correspondingly?

    https://blogs.transparent.com/hindi/hindi-numbers-1-100/

  100. January First-of-May says

    Then I’d expect Yukaghir to be part of that. From looking at this list of 1–10 in 5000 languages (you have to load the whole list; then search it), it’s not.

    Instead 8 looks like it’s 6 combined with 4, which doesn’t make any sense to me at all. Is there some kind of underlying non-decimal system involved?

    Looks (a bit) like Arin and Pumpokol were part of it, though. So there’s that.

  101. Moin Davis says

    Here are all numbers in Hindi

    https://skillslelo.com/numbers-in-hindi/

  102. Instead 8 looks like it’s 6 combined with 4, which doesn’t make any sense to me at all.

    ‘6’ literally means ‘both sides 3’ (*mālə-jalo-), ‘8’ means ‘both sides 4′ (*mālə-jelek-) though syncope of the second components’ first vowels leaves this a bit nontransparent.

  103. January First-of-May says

    ‘6’ literally means ‘both sides 3’, ‘8’ means ‘both sides 4′

    …Oh. I see it now.

    I’m actually surprised how I didn’t figure it out the first time; something like this is the obvious reason why the word for 8 would look like the word for 4, but I’m guessing that the similarity to 6 confused me and I hadn’t thought to check if 6 relates to 3 in the same way (turns out it does).

Speak Your Mind

*