Chinese Orthographic Revolutionaries.

Joel at Far Outliers is quoting from Kingdom of Characters: The Language Revolution That Made China Modern, by Jing Tsu, and this post has a fascinating look at old attempts at reform of Chinese writing:

While working on his alphabet, Wang [Zhao] never strayed from the beliefs he had shared with the emperor back in 1898: China was losing its power because language was failing its people. Their low literacy and divided dialects impeded China’s ability to govern, negotiate with foreign powers, and keep pace with the outside world. China’s success as a nation and an international power hinged on the single issue of an accessible spoken and written language.

There had been others who shared Wang’s analysis of the problem, although they offered different answers to it. Lu Zhuangzhang, a Chinese Christian from Amoy (now Xiamen), developed the first phonetic system for a Chinese language by a Chinese. His 1892 Simple Script used fifty-five symbols, some of which were adapted from Roman letters to Chinese sound rules, to represent the southern dialect spoken in Amoy. Lu nearly went bankrupt in the process. Lu’s children would bemoan how he squandered the family’s livelihood financing his linguistic experiments.

Among those who followed in Lu’s footsteps was Cai Xiyong, an attaché to a Chinese diplomatic delegation to the United States. He developed his Quick Script for the major southern topolect group of Min, using a rapid writing system—a kind of shorthand—created by David Philip Lindsley. Shorthand, pioneered by Isaac Pitman for the English language in 1837, was a transcription system that used specially simplified notations to record phonemes, words, and phrases in real time.

The real innovator, many later thought, was Shen Xue, a brilliant medical student from Shanghai whose reputedly ingenious scheme, according to eyewitnesses, was originally written in English but has survived only as a preface printed by a Chinese journal. Shen devoted his life to propagating and offering free lessons on his Universal System at a local teahouse. He died a pauper at the tender age of twenty-eight.

Wang stood out from the rest in one important way: while he believed in giving people the power of literacy and the ability to connect with speakers of other dialects, he ultimately wanted to hold them to a single standard—Beijing Mandarin. He saw the critical importance of a unified language, and he was the first to propose Beijing Mandarin as the nation’s standard tongue. It would become the basis of the modern Mandarin, or Putonghua, that the Chinese speak today. For Wang, increasing literacy was only possible if one simultaneously created linguistic unity. To unite China’s hundreds of tongues with a single phonetic scheme would be like deconstructing China’s own Tower of Babel. Before Wang could tackle this problem, however, he had to contend with the native sound system that had been in place for centuries: a way of learning and teaching characters based on sounds called the reverse-cut method.

(See Joel’s post for links I’m too lazy to reproduce.) The fate of those whose good ideas don’t get accepted is sometimes a sad one…

Comments

  1. It’s not the point of the post, but “[s]horthand [was] pioneered by Isaac Pitman for the English language in 1837” seems a little suspect.

  2. J.W. Brewer says

    I’m not sure which of the various personalities you are referring to as having good ideas which didn’t get accepted, although I take it it’s not Wang, whose bad (if cross-culturally quite common) idea, viz. “let’s modernize by adopting a single Official National Language and then trying to squelch its regional rivals by illiberal means” DID get accepted by those in power. See Joel’s separate post https://faroutliers.com/2023/04/03/how-mandarin-became-the-standard/

  3. “good ideas”.
    My imression was that imposing Mandarin on everyone is believed to be a bad idea by some people here…

  4. Actually, I was thinking of Lu and Shen, though I may have been generous in calling their ideas good. They were new and interesting, at any rate.

  5. Obviously I agree that imposing Mandarin on everyone is a bad idea.

  6. Considering that Google only wants to offer me videos showing salon techniques, permit me to ask: what is the “reverse-cut method”?

  7. J.W. Brewer says

    Of course, for the author Joel quotes to write “Amoy (now Xiamen)” is just a surrender to that compulsory-Mandarinization dynamic. “Amoy” was a reasonable approximation of one version of what the locals call the place in their own language, which wikipedia advises me is now standardly romanized as “Ē-mûi.”

  8. Craig–

    The “reverse-cut method” is fǎnqiè 反切, which gives the pronunciation of a given character by using one character for the initial consonant and a second character for the rest of the syllable.

    Kind of like if you showed the pronunciation of English “quell” as cut+dwell. Or “fate” as frog+late, etc.

  9. It works reasonably well for pronunciation glosses, dictionary pronunciation guides, rimebooks, etc (as long as you happen to know the characters that are chosen)… but I don’t think anyone ever tried to use it for any sort of connected writing

  10. “Obviously I agree that imposing Mandarin on everyone is a bad idea.”

    Actually (and I’m objecting to myself, not you) even though presently I’m ready to call it a “bad idea”, but from 19s century…
    How do you imagine the “ideal” future in 19th century, IF you for some reason value language diversity (and language rights etc.)? It is not an easy question. And how do you imagine spread of literacy and rapid industrialisation? Multilingualism is compatible with literacy, of course, but when your goal is making everyone literate….Actually your other solution is just Wenyan.

  11. The blogpost in-between, Finding Chinese Character Sounds, describes the “reverse-cut” method.
    https://faroutliers.com/2023/04/02/finding-chinese-character-sounds/

  12. J.W. Brewer says

    The wikibio for Lu says: “When Lu later supervised a language school in colonial Taiwan, he realized the flaws with his Qieyin Xinzi and attempted to redesign the system on the basis of the Japanese kana syllabary, but there were already too many competing schemes.” I guess I don’t know to what extent would-be script reformers on the mainland were aware of kana, which seems like it should have been an obvious role model. It seems hard to believe that the devisers of bopomofo, which was first rolled out at least in beta version circa 20 years after Lu’s Simple Script, were not aware of kana. Although come to think of it I guess bopomofo largely implements the “reverse-cut” approach, with the exception that having a minority of syllables require 3 glyphs rather than 2 means you need a smaller total glyph inventory.

    And of course a bopomofo-type system that would work for non-Mandarin topolects like Southern Min might well need different/additional glyphs. Indeed, the eventual adaptation of bopomofo for Taiwanese Hokkien (close to interchangeable with Lu’s presumed native Amoy dialect) reportedly uses about two dozen glyphs beyond the set used for Mandarin, making Lu’s 55-glyph inventory seem parsimonious by comparison.

  13. I think I don’t understand these efforts.

    There had been others who shared Wang’s analysis of the problem…”
    The analysis: “Their low literacy and divided dialects impeded China’s ability to govern, negotiate with foreign powers, and keep pace with the outside world.

    All right, but would not a logographic system be a perfect choice for this exact problem?
    And why, for example, Communist China replaced Yi logograms with Yi syllabary?

    Perhaps from distance one may think that only exceptinally gifted people can learn Chinese characters, but modern Chinese do read and write this way.

  14. Here are two earlier passages that credit Manchu and Japanese practices, which I reluctantly omitted so as not to blog too much from that chapter:

    The Kangxi Emperor also ordered the compilation of the Kangxi Dictionary—the Chinese equivalent of the Oxford English Dictionary—which coined the Chinese term for “dictionary,” or “canon of characters.” Subtle Explorations, in contrast, was of a whole other genre and purpose. The handbook was intended to correct the age-old system of reverse-cut, updating it with Manchu phonetics to reflect modern speech sounds, the type of northern vernacular Mandarin that was spoken in the capital. The key point was that Manchu used a kind of alphabet. Unlike Western alphabets, which are classified mainly as vowels and consonants, Manchu is phonetically divided into consonant-vowel syllables as the phonetic base (pa, pe, pi, po, pu, for instance, as opposed to a, e, i, o, u). Manchu possessed the ability to notate the sounds of Chinese characters in a non-Western style. (pp. 27-28)

    Jolted by this discovery, Wang went back to his Mandarin Alphabet with revamped tools and refreshed understanding, unveiling a new version in 1903. His Mandarin Alphabet was composed of fifty symbols for sound initials and twelve other symbols for sound finals, making for sixty-two symbols in total. Remember: All Chinese syllables are made up of the opening consonant sound (an initial) and the sound that completes the consonant (the final). The fifty initials were borrowed, with further simplifications, from the Japanese derivations of Chinese characters, kana, a syllabic script developed in the eighth century. Wang’s twelve finals were modeled directly on the Manchu script’s twelve classes of letters and syllables. The Mandarin Combined Tone Alphabet made the transcription of Chinese fast and accurate to real-time speech. It used Chinese-derived symbols but essentially functioned phonetically like an alphabetic language, where the way a word sounds is generally evident in the way it is spelled. Using Wang’s Mandarin Alphabet, if you wanted to indicate the pronunciation of the character for “east” 東 (dong), you would use one of his character-derived phonetic notations from his fifty initials and one of his twelve Manchu-based finals: simply add the initial do and the final ng. That was it. (pp. 28-29)

  15. John Cowan says

    In order for that 50:12 ratio to make any sense, the medial glides must have been counted as part of the initials rather than (as nowadays) part of the finals.

  16. Yes I’m not following. And the internets seen curiously silent on ‘Mandarin Combined Tone Alphabet’.

    5 vowels times 5 tones (4 + neutral) in Putonghua has already blown 12; hasn’t covered nasalisation, nor glides as @JC says.

    Wang isn’t even trying to cover Southern Min or Cantonese, which have a bigger inventory.

    @Joel quoting the Kingdom book: All Chinese syllables are made up of the opening consonant sound (an initial) and the sound that completes the consonant (the final). Correct to say ‘syllables’; but how does this cope with polysyllable words? — which are legion, as VHM enjoys reminding everybody.

    Manchu possessed the ability to notate the sounds of Chinese characters in a non-Western style.

    Hmm, hmm. Western style copes with consonant clusters — word initial, word final, or medial. Didn’t some topolects still have vestiges of those in C19th?

    In short I don’t see how any of these schemes could unite the Kingdom. I suppose by then the ‘Middle Kingdom’ was trying to isolate itself from the rest of the world, or they could have observed how other kingdoms/conquerors coped with linguistic diversity.

  17. ‘In order for that 50:12 ratio to make any sense, the medial glides must have been counted as part of the initials rather than (as nowadays) part of the finals.’

    Much more sensible to count them as part of the ‘initial’, which then approximates the linguistic onset more precisely. It’s always struck me as bizarre that a word like guang, which is phonetically roughly [kʷɑŋ] or [kʷɑũ], would count the ‘final’ as -uang, when part of that ‘final’ is coarticulated with the ‘initial’!

    @AntC, I got the impression that tones are represented in a distinct way, not through separate final symbols. Still, I don’t see how twelve symbols would be enough anyway. Even trying to trim things as much as possible (assuming that the difference between xi and shi is driven by the onset, not an underlying vowel difference; treating chun as chuen; taking as underlyingly u where possible etc.), I still end up with 17 or so distinct finals:

    i, u, ü, e, a; ei, ou, ao, ai; in, ing, ong, en, eng, an, ang; er

    A phonemic script might get rid of entirely by treating it as /ui/ (which would have different realizations after different onsets), but that wouldn’t help in this style of script, as far as I understand it.

  18. I suppose by then the ‘Middle Kingdom’ was trying to isolate itself from the rest of the world, or they could have observed how other kingdoms/conquerors coped with linguistic diversity
    All the militarily and economically successful ones at that time were imposing a unified national language using a unified national orthography. Exactly what most of those Chinese attempts at reform were trying to imitate.

  19. J.W. Brewer says

    @Nelson: Perhaps you can see the wisdom in bopomofo doing guang (using tongyong pinyin which here agrees with hanyu) as ㄍㄨㄤ, rather than trying to break it into just two parts and thus having to decide where the break should be. By comparison two-glyph ㄍㄨ comes out as “gu” and two-glyph ㄍㄤ comes out as “gang,” so combining them unsurprisingly gets you “guang.” I guess the point is that ㄨ can be a “final” or “rime/rhyme” in some syllables but a “medial” in others.

  20. @J.W. Brewer, no objections to that kind of orthography, and I see that bopomofo also uses the same symbol for Pinyin w. The situation looks a lot like Classical Latin, where a single graph is taken as vocalic or consonantal as needed. Nicely economical. But not because it dodges some kind of analytical problem or genuine linguistic ambiguity.

  21. @AntC: The Wiki English bio of Wang Zhao (which I linked to) translates his titles differently: “It was called Mandarin Letters (Guanhua zimu) or the “Mandarin Chinese Harmonic Alphabet”, and was a stroke-style writing system, similar to pinyin, based on the Mandarin dialect.”

  22. Bathrobe says

    Obviously I agree that imposing Mandarin on everyone is a bad idea.

    You are speaking as a linguist (lover of language), anarchist (“don’t impose uniformity on me”), and lover of richness and diversity. European nations did the same thing, but what sets China apart is the sheer size and diversity of the country. To think, if the Europeans hadn’t set out to conquer and exploit the rest of the world, China might still be a model of enlightened diversity.

  23. There is another aspect: our ideas of language [policies, planning, preservation, development] are clearly based in misconceptions. It would be not bad to make it a science.

    Bashing CCP is all good, but.

    P.S. I do value diversity, and I agree that imposing Mandarin is a bad idea.

  24. J.W. Brewer says

    I’m as much of a CCP-basher as the next fellow, maybe more so, but … the key point here is that this drive to standardization and Mandarin supremacy was already underway in the immediate aftermath of the 1911 overthrow of the tottering Empire by a bunch of would-be reformers and modernizers who rapidly proved incompetent. It was the Manchus who had a relaxed attitude toward linguistic diversity in their domains, just as the ancien regime Bourbons in France didn’t really care too much if some of their subjects spoke Basque or Breton or Occitan or whatever rather than Paris-style French, with the drive for national uniformity/standardization only coming after 1789. So the PRC is just continuing an ROC-era project, albeit perhaps doing so with more brutality and less inefficiency. During the days of one-party authoritarian rule on Taiwan, the KMT was very Mandarin-supremacist, but these trends, and the general aspiration toward achieving modernization via illiberal methods, were already a thing well before the KMT ended up taking power in the mainland ROC in the late 1920’s.

  25. to my eye, it’s about the nation-state as a form, with the historical trajectories of different polities gradually converging as that model emerged and then became globally hegemonic. in europe, spain in 1492 marks one kind of inflection point, with the grammar of castillian and the expulsion of the jews, and the 1848 Springtime of Nations (and dreams of language academies) another.

    it seems to me that china’s distinct trajectory in terms of language standardization/mandarinization is different from those of european states on the one hand because its overall process of shift towards the nation-state model has followed such a different path, and on the other because in europe only russia built an empire contiguous enough to be transformed into a nation-state and encompassing a similar scale of linguistic diversity. i wonder whether the trajectories of, say, japan or thailand are more similar to the french or spanish models.

    but i don’t know that i’d describe either the bourbons or the qing as having a relaxed attitude towards linguistic diversity. neither may have been actively seeking to eliminate it, but there’s a lot of room between those two. to be a qing state employee you had to demonstrate fluency in a standardized form of what we call mandarin, but was endonymically called (if i remember right) “bureaucratese”: a very rigid privileging/mandating of the preferred lect of the rulers. and similarly, you weren’t gonna get far at versailles if you were a monolingual speaker of gascon, let alone breton or euzkara: the state ran as much as it could on a single lect. just because what the vast majority of the population spoke was of no particular interest to the state doesn’t mean that the rulers were relaxed about matters of language, just that as imperial rather than national rulers they didn’t consider peasants human enough to matter.

  26. A: you can talk to your goats in your dialect, but to study science you need excellent Persian and English.

    B: let’s encourage social mobility and promote education.

    For this purpose, let’s teach everyone to speak Persian and English. Don’t speak like a shepherd!

  27. @rozele in europe only russia built an empire contiguous enough to be transformed into a nation-state and encompassing a similar scale of linguistic diversity.

    hmm? In my “could have observed ” comment above, I was thinking specifically of the late C19th Austro-Hungarian Empire — which we’ve discussed at length in the Hattery. I think meets the criteria of contiguous and linguistic diversity(?) OK it disintegrated eventually to (linguistically-defined) nationalism, but that was later than the timescale here.

    What might have been a model more matching the ‘Middle Kingdom’s aspirations would be the Roman Empire. They seemed happy for the colonies to speak whatever, provided they paid their taxes/tributes with Latin paperwork.

  28. Well, A-H also tried to impose national languages, German in the Austrian half and Hungarian in the Hungarian half. That was one of the triggers that fueled the nationalisms of the other ethnics that resulted in its dissolution.
    And while the Roman empire didn’t really care what its subjected people spoke, still becoming part of the elite and participating in its politics meant speaking Latin (in the West) and Greek (in the East). It still ended up with most people speaking one of those two languages.

  29. I agree with rozele’s comment, and anyone who thinks premodern emperors/kings/whatever wouldn’t have imposed linguistic uniformity in a heartbeat if they’d had the opportunity is living in a dream world. States based on arbitrary power want uniformity in every available form.

  30. John Cowan says

    Up to a point, Minister. Plastic surgery has been around for a while, but no dictator has attempted to make all his people look the same.

  31. J.W. Brewer says

    It’s not that these older regimes had some sort of fundamental moral commitment to linguistic diversity, but it’s not just lack of opportunity; it’s lack of motive. You only need a comparatively tiny percentage of your subjects to be able to function as bureaucrats or courtiers or whatnot where you do care that they all have fluency in some approved language variety. You are not particularly interested in promoting social mobility, much less encouraging the unwashed masses to participate in nationwide dialogue with each other about how the polity should be governed; you are certainly not interested in funding universal public education; if you are smart you understand that top-down attempts to impose cultural transformation/assimilation on subject peoples are the sort of thing that often precipitate peasant revolts and why do you want to have to spend time and resources suppressing revolts rather than living the high life with your mistress(es)/harem as long as the various subject peoples pay their taxes?

    This gets us back to the recurrent problem or irony that the social/historical conditions under which multilingualism is long-term stable are in substantial tension with various things that modernity and liberalism (broadly construed) presume are axiomatically Good Things, like social mobility, universal education, and not having your career options and pool of potential marriage partners determined simply by reference to who your parents were. Ottoman-ruled Salonika was a place where at least six different ethnolinguistic groups could and did live cheek-by-jowel for centuries, most with varying degrees of multilingualism but all without losing their ancestral L1. Post-Ottoman Salonika, not so much.

    And of course when the late-Ottoman/Ataturkist Turks finally became interested in modernity and liberalism, they immediately set about to expel, exterminate, oppress, and/or forcibly assimilate their non-Turcophone neighbors in a way that no one had ever previously thought worth doing.

  32. It’s not that these older regimes had some sort of fundamental moral commitment to linguistic diversity, but it’s not just lack of opportunity; it’s lack of motive. You only need a comparatively tiny percentage of your subjects to be able to function as bureaucrats or courtiers or whatnot where you do care that they all have fluency in some approved language variety.

    Sure; my point is that if a fairy had appeared and said “If I wave my magic wand, all your subjects will instantly speak your imperial language instead of all those incomprehensible jabbers they currently use,” the offer would have been accepted instantly. The fact that eliminating diversity was too much trouble, and not of much value under the conditions of the day, does not in any way mean that rulers valued it.

  33. John Cowan says

    JWB puts his points negatively, but they can also be put affirmatively: it is in the interest of dictatorships to suppress communication between the dictatees. The Assyrian Empire went so far as to randomly move ethnic groups or parts of groups around the empire to keep them from conspiring against it; so did Stalin, of course. So I don’t think your magic-wand offer would necessarily be accepted.

  34. @AntC: i don’t see the hapsburg empire as having the scale of linguistic diversity that the russian and qing empires did: both of them claimed territory in regions (the caucasus; upland southeast asia) that are among the most diverse around. austria-hungary was pretty mixed by european standards*, but those standards are pretty impoverished compared to these kinds of fracture zones (to tip my hat to scott, as usual). and, on a slightly different tip, the hungarian empire used latin as its official language well into the 19thC – it was precisely the magyarizing move, as part of a shift towards nation-state-ness, that helped upset the imperial applecart.

    .
    * but not, i think, much more so than france or spain: slavic, germanic, finno-ugric, romance, albanian, turkic, indo-aryan vs romance, celtic, basque, indo-aryan, semitic [spain] or germanic [france]. i’m probably missing things on both sides.

  35. J.W. Brewer says

    @rozele, note that the Hapsburgs had good political divide-et-impera reasons to want to preside over speakers of as many different (and understood-to-be-different) Slavic languages and ethnicities as possible, since a conscious Pan-Slavic identity was likely to be troublesome for them. But in any event I think the Hapsburg domains were diverse enough from a practical perspective that the additional/incremental diversity of the Romanov domains didn’t much matter in terms of how challenging imposing uniformity would be. If 90%+ of your population/territory is speakers of various bits of a dialect continuum, even one where many pairs of points on the continuum are too distant from each other for mutual intelligibility, it’s probably easier to take the lumper tack that these are all merely local rustic dialects of the One Official National Language their children will henceforth all be taught in school, with how to handle outliers like Basques or Bretons off in the corner of the map a perhaps difficult but second-order question. That wasn’t an approach that was workable in the Hapsburg lands, or even either subpart of them considered alone.

  36. “States based on arbitrary power want uniformity in every available form.” – I’m not sure that “arbitrary power” as opposed to some other sort of power has to with this.

  37. The less arbitrary the power (i.e., the more the people subject to it have input into its decisions and can effectively object to changes they don’t like), the more diversity will be tolerated, pretty much by definition. People do not want to be molded into duplicates of each other, no matter how much that would please their rulers.

  38. @LH my intuitive objection is that I usually see language shift as an outcome of overall social architecture (1. urbanisation 2. universities speak Russian) and social pressure because people so often encourage speaking like educated. Or an unknown to me process that made Gaeltacht so small over the history of Irish independence.
    I also was very sceptical about Ukrainian in Ukraine.

  39. J.W. Brewer says

    Hat, qua anarchist, seems to have a uniform stereotype of unelected rulers that’s the mirror image of the classic “all anarchists are sketchy-looking bearded foreigners holding a bomb with a lit fuse.” Surely it’s more realistic to think that some such rulers are interested in compulsory uniformity (linguistic or otherwise) for its own sake, while others have other interests. Especially before the French Revolution, after which “reform and transform humanity for its own good” started to become a more common default answer to “what you ought to do with unconstrained political power.”

    Note FWIW that I didn’t claim that the Manchu rulers of China affirmatively valued linguistic diversity for its own sake,* only that they had a relaxed attitude toward it (outside of limited contexts/domains). Perhaps their attitude would have been different had they had access to fairies with magic wands who could change people’s languages costlessly, but that’s not really a testable hypothesis. Suppose they’d had a “this genie will grant you up to three wishes, but no wishing for more wishes” opportunity. What are the odds that magically creating linguistic uniformity would have been among their top three desires for enhancement of either their Empire or themselves?

    *Except I guess to the limited extent that some of them fought some rear-guard actions to try to preserve Manchu fluency among their own kind against the remorseless tide of Sinicization.

  40. A better image

  41. You may mean 1: simple rule of the majority, just where this majority affects decisions and maybe without brainwashing. Putin is popular in Russia. If you compare him to other dictators, he must be sensitive to public opinion. So we maybe need some other sort of majority rule. The problem here is that we can find many nationalist states “ruled by the majority”.

    But we imagine “brainwashing” as specifically political, not linguistical claims. Views on language are distributed differently, in school. So we should be ready to itnerpret stuff that you hear in school and even your culture as “brainwashing”.

    Or you can mean 2: a society where minorities play an important role. It is not clear how exactly it must be organised, but my intuitive objection is what I said above.

    I don’t know, maybe you are simply right. Maybe people who can affect political decisions will also take interest in what I called the “overall architecture” of their space (and in turn, rulers who are not authoritarian don’t initiate certain grand projects aimed at altering this architecture). Or maybe what you actually mean is decentralisation.

    There are still issues like some langauge becoming the lingua franca of local business.

  42. my point is that if a fairy had appeared and said “If I wave my magic wand, all your subjects will instantly speak your imperial language instead of all those incomprehensible jabbers they currently use,” the offer would have been accepted instantly.

    Why? Pre modern rulers had no interest in most of their subjects other than as sources of labor and taxes. We are talking about people who had no qualms enslaving and mutilating other human beings. I would be surprised if Augustus really cared whether the slaves being worked to death in the mines learned Latin, or if Mehmet II thought the villagers in Bosnia needed to speak Ottoman Turkish when the taxman came to collect. Imperial rulers did not see most of their subjects as equals, or even as real human beings. They would not have seen any utility in a single language, in fact a monolingual society might erase some very useful- to the rulers – markers of status and hierarchy. Literacy and the bureaucratic state seem to have convinced rulers of the utility of a single language.

  43. There are certain processes (include things that “happens on their own” and carefully planned things like the school system) that lead to massive language loss in modern times.

    Now, of course, an authoritarian regime may try to take control of these processes (not necessary, some authoritarian rulers see them as happening on their own).

    But the processes are there, whether authoritarian rulers control them or not. When people have power, the two questions are:
    – will they choose to intervene in these processes?
    – if they do, will this intervention be directed at preserving diversity or imposing uniformity?

    —-
    Historically there were regimes that very literally attempted to erase local barbarian tongues – and regimes that did not have this goal, but who took control of urbanisation, education and so on (USSR was like this) and thus took various decisions about roles of respective languages.

    It still appears to me that IF Dagestani people were asked once a year “do you want to stay with Russia” and IF they kept saying “yes, we do”, Russian would still remain important in Dagestan. Then if they say “no, we don’t” and become independent, it will be Avar or Dargwa and you would have to ask the same question “do you want to stay with Dagestan?” to Tabasarans.

  44. If 90%+ of your population/territory is speakers of various bits of a dialect continuum, even one where many pairs of points on the continuum are too distant from each other for mutual intelligibility, it’s probably easier to take the lumper tack that these are all merely local rustic dialects of the One Official National Language their children will henceforth all be taught in school,

    Here you offered a meaningful linguistic definition of “a language”:-)

    (No, I don’t think that we need to call language continuums “languages”, I don’t think “merely” makes sense, and I don’t think that children in such a space must be taught anything specific in school. I just think that “a small group of langauges” (equivalent reformulation of “a group of sufficiently close dialects”) isn’t a meaningful object and a “language continuum” is….

    In other words: I don’t approve renaming “language continuum > language”, but the resulting object will remain meaningful no matter how you call it).

  45. @JWB:

    “90%+ of your population/territory is speakers of various bits of a dialect continuum” wasn’t true of the russian empire*, the qing empire, or i think any of the other polities we’ve been talking about. that level of homogeneity is pretty much a post-WWII phenomenon, and it took a lot of very bloody work to engineer it.

    and sure, the habsburg and romanov empires had very different dynamics around language and the maintenance of imperial control. but for the russian autocracy to find pan-slavism useful, it had to already be moving towards a nation-centered idea of what a state should be. and for the hapsburgs to see either pan-slavism or local nationalisms** as a threat, they had to already be able to see the nation-state as a competing model. (and that’s not even touching the interesting can of worms that drasvi just opened****)

    which is my whole point: the advent and then hegemony of the nation-state as form is the main thing that drives language standardization projects, playing out in different ways in different places.

    i do think that the quantitative difference in diversity between austria-hungary or spain and russia or china does make a qualitative difference to these standardization processes. but since the most lect-diverse empires are also so different from each other, that would probably need to be argued case-by-case, with other comparisons as well (the roman, ottoman, abbasid, and inka empires come to mind).

    .
    * probably the easiest to demonstrate, since the 1897 imperial census asked about first languages, and (even after a few waves of russification policies) found under 70% for cradle-tongue speakers of russian, belarusian, and ukrainian combined (adding in polish*** doesn’t break 75%) combined.

    ** the difference between the two being only a matter of the scope of the hypothetical nation, with a large range of scales to choose from: slavs? illyrians? croats? dalmatians? janjevci? take your pick!

    *** same slavic continuum? different slavic continuum? same one in congress poland but different one in hapsburg galicia? not my circus.

    **** a language is a dialect continuum with an organization whose name begins with “Young”?

  46. Lars Mathiesen (he/him/his) says

    In Denmark, at least, the story is that the kings’ creation of a central administration was what drove the standardization of Danish. People in Copenhagen spoke a coastal dialect, but it was in the rich market towns in the center of Zealand that people could afford to send their sons to Latin school, and those sons formed the backbone of the growing bureaucracy. (And the clergy).

    Standard Danish is the language of Zealand farmers, not of any random group of courtiers; the king had to learn to speak like the farmers. (In the 500 years since then there was time for the areas around Roskilde and Ringsted to develop new local language traits),

  47. There’s a suggestion that there are North Slavic continuum (East Slavs and Slovaks and Czechs) and South Slavic continuum….

  48. David Marjanović says

    Remember: All Chinese syllables are made up of the opening consonant sound (an initial)

    …if you’re willing to accept zero-initials.

    (BTW, I’m convinced the Korean circle for the zero-initial is a 0.)

    ‘In order for that 50:12 ratio to make any sense, the medial glides must have been counted as part of the initials rather than (as nowadays) part of the finals.’

    As the example at the end of the quote makes clear, the 50 “initials” are not consonants, but (C)V sequences.

    “bureaucratese”

    That’s exactly the guānhuà of the Guānhuà zìmǔguān was “mandarin”, “bureaucrat-scholar-poet”.

    A-H also tried to impose national languages, German in the Austrian half and Hungarian in the Hungarian half.

    It’s not that simple. For example, Moravia ended up officially bilingual, and all the civil servants were given a few years to learn the other language.

    The whole story started with Joseph II (r. 1765–1790), an “enlightened absolutist” (meaning “everything for the people, nothing by the people”). He had a lot of simple, pragmatic ideas, over half of which ended up impractical to implement and/or unpopular enough that he had to take them back in part or in whole. One was to replace Latin as the language of bureaucracy by German in the whole empire. The Hungarian magnates were not amused and ended up keeping Latin until they got their own Half of the Empire and replaced it by Hungarian throughout that.

  49. I’m not going to make a separate post for it, but Joel also posts a very instructive account of How Mandarin Became the Standard. (Spoiler: it involved cheating and force majeure rather than rational discussion and compromise.)

  50. J.W. Brewer says

    Two things:

    1. David M.’s comment re the Josephine (or Josephinian?) reforms underscores that I was too hasty in blaming the Jacobins and the metric system for the rise of the notion that the proper use of absolute political power is the destruction of people’s traditional linguistic/cultural-etc. forms For Their Own Good. The so-called enlightened despots (and their intelligentsia henchmen) had already been pursuing that sort of agenda for a generation or three before the Jacobins arrived on the scene. And of course it had some considerable if intermittent pre-Enlightenment history, albeit usually in earlier times in connection with an explicitly religious agenda.

    2. I think rozele is right to focus on the rise of the nation-state as opposed to earlier forms of the geographical organization of rule, but I fear we have slightly talked past each other due perhaps to lack of clarity on my part in an earlier comment. My example re 90%+ of subjects being in a single “dialect continuum” (or however you want to label that phenomenon) was intended to apply to the examples of France and Spain, with my related point being that it wouldn’t work for the Hapsburg domains, with the notion that it perhaps wouldn’t work for the Romanov domains to an even greater degree being imho irrelevant once it wouldn’t work, period. Obviously there are e.g. plenty of local nationalists in Iberia who would vehemently deny (with more or less empirical linguistic evidence on their side) that Asturian or Catalan are merely “dialects” of the same thing that Castilian is a “dialect” (and indeed the exemplary/standard dialect) of, but you can at least assert that claim and maybe sound convincing if you have government power behind you, whereas no matter how much power you have behind you “Basque is simply another regional dialect of Spanish” is not a story anyone is going to buy.

  51. @David Marjanović: BTW, I’m convinced the Korean circle for the zero-initial is a 0.

    In today’s Korean, ㅇ is indeed a zero initial. But there are reasons to believe that it originally stood for a sound value in Middle Korean, perhaps [ɣ] or [ɦ].

  52. John Cowan says

    Ramsey tells the story in more detail in The Languages of China (1975). The relevant passage is quoted in the WP article on Wang Zhao:

    Few of the delegates at the 1913 conference on pronunciation seem to have had any idea of what they were up against. The negotiations were marked by frustratingly naïve arguments. “Germany is strong,” it was said, “because its language contains many voiced sounds and China is weak because Mandarin lacks them.” But if linguistic knowledge was in short supply, commitment to position was not. Passions were hot, and frustrations grew.

    Finally, after months of no progress, Wang Zhao, the leader of the Mandarin faction, called for a new system of voting in which each province would have one and only one vote, knowing full well that the numerically superior Mandarin-speaking area would then automatically dominate. Delegates in other areas were incensed. The situation became explosive. Then, as tempers flared, Wang Rongbao, one of the leaders of the Southern faction, happened to use the colloquial Shanghai expression for ‘ricksha[w],’ wangbo ts’o. Wang Zhao misheard it for the Mandarin curse wángba dàn ‘son of a bitch’ (literally ‘turtle’s egg’), and flew into a rage. He bared his arms and attacked Wang Rongbao, chasing him out of the assembly hall.

    Wang Rangbao never returned to the meetings. Wang Zhao’s suggestion to change the voting procedure was adopted, and after three months of bitter struggling, the Mandarin faction had its way. The conference adopted a resolution recommending that the sounds of Mandarin become the national standard.

    In some sense the result was structurally foreordained: the language being used at the conference was inevitably Mandarin, the only language that all of the delegates spoke and understood.

  53. J.W. Brewer says

    Perhaps unfortunate they had the conference in 1913 rather than 1919, by which time more data had accumulated bearing on the correlation if any between German’s consonant inventory and Germany’s national strength.

  54. Obviously there are e.g. plenty of local nationalists in Iberia who would vehemently deny (with more or less empirical linguistic evidence on their side) that Asturian or Catalan are merely “dialects” of the same thing that Castilian is a “dialect” (and indeed the exemplary/standard dialect) of…

    Implies that if you mix up the meanings
    1. “merely, simply, unworthy”
    2. “a variety”
    then you can find linguistical evidence against 1. Like “I have evidence that I’m not a mere woman”.

  55. John Cowan says

    mere woman

    Mere motion.

Speak Your Mind

*