French girl sells crêpes in a Taiwan market

« previous post | next post »

Scene at a Taichung night market:

In a recent Mín Shì Xīnwén 民視新聞 (Formosa TV News network) report (dated March 28) about a French girl selling crepes (kělì bǐng 可麗餅) in a Taichung night market, two words pop up that we have touched on here at Language Log:  sūcuì 酥脆 ("crunchy; crispy") and Qtán Q彈 ("chewy; springy; elastic"), the latter being biscriptal.  The way she says "QQde QQ的" is really cute.


Selected readings

[Thanks to Chau Wu]


  1. M. Paul Shore said,

    April 9, 2020 @ 8:38 am

    As a very much lapsed student of Chinese, I've spent almost zero time in my life watching Chinese-language television news stories; and one thing I found fascinating watching this one was that, just as happens in English-language and at least some other European-language television and radio news stories, the reporter dropped the pitch of her voice in the last sentence, and perhaps gave other non-verbal vocal signals as well, indicating that the story was coming to its end. (In English-language television and radio news stories, these non-verbal indications sometimes come in the last sentence and sometimes are spread over the last two sentences.) Is this just inevitable natural human speech behavior? Or is it something that reflects the influence of the broadcasting customs of English and some other European languages?

  2. Philip Taylor said,

    April 9, 2020 @ 9:04 am

    Very perceptive, MPS — I heard what you heard, but you noticed it and I did not.

  3. Luke said,

    April 9, 2020 @ 1:46 pm

    It's interesting to see France being pronounced in Taiwan as fa4guo2 rather than the usual fa3guo2. I never knew his variation exists.

  4. Michele Sharik Pituley said,

    April 9, 2020 @ 2:09 pm

    > the reporter dropped the pitch of her voice in the last sentence, and perhaps gave other non-verbal vocal signals as well, indicating that the story was coming to its end.

    I noticed that, too. In terms of pitch, it sounded just like a US newsperson's sign-off. "This is Bailey Quarters, WKRP news."

  5. Bathrobe said,

    April 9, 2020 @ 7:17 pm

    It's interesting to see France being pronounced in Taiwan as fa4guo2 rather than the usual fa3guo2.

    "fa4guo2" is the original pronunciation. "fa3guo2" is a later spelling pronunciation (as it were), based on the fact that the character used to write it is 法 (fa3).

    The original pronunciation has pretty much died out in China (although someone in Beijing told me once that they'd heard it from an old person) due to language standardisation efforts aimed at eliminating variant pronunciations of characters.

    Taiwan appears to have kept the original pronunciation.

  6. Victor Mair said,

    April 9, 2020 @ 7:29 pm

    I certainly learned it as fa4guo2 half a century ago in the United States and in Taiwan. When I heard people from the mainland saying fa3guo2, it sounded very strange, as did a thousand other mainland usages.

  7. Jonathan Smith said,

    April 9, 2020 @ 8:42 pm

    The name for France goes back to a phonetic loan; i.e., the character "法" was selected based on sound to begin with. So what really calls for explanation is the separation between fa3 'law' and fa4[guo2] 'France' in the older standard language including in Taiwan: this is an older innovation than restoration/leveling to fa3 (if that's indeed what happened). Special treatment of old rusheng characters in late imperial Mandarin? Dunno…

    正妹 etc. kinda trashy in this news story… :( sigh

  8. Michael Watts said,

    April 10, 2020 @ 1:43 am

    An earlier LL post had an image of a Qing document which referred to America using a 口字旁-modified 美 character. This makes sense in terms of the old idea that you put a 口 radical on a character to indicate that it represents sound without meaning (an idea that modern Chinese emphatically reject). It's not standard now, and indeed the character itself does not appear to exist in Unicode.

    I'm curious what happened to erase this character from official existence, and whether other country names used the same technique. Was France originally 口+法?

  9. Michael Watts said,

    April 10, 2020 @ 1:46 am

    It seems like 佛朗机 should also get a mention.

  10. Leo said,

    April 10, 2020 @ 2:24 am

    Question from a non-Sinophone: why are the interviewees subtitled – is this routine practice in Taiwan, or is it something specific about how those people spoke?

  11. Michael Watts said,

    April 10, 2020 @ 11:02 am

    I don't think it's anything specific about how the interviewees spoke. Subtitling speech is pretty common in China (for example, TV shows are routinely subtitled), presumably because historically many people would be unable to understand any given person's speech. But in this case, the announcer isn't subtitled, so it would seem to be more of a vestigial practice.

  12. Bathrobe said,

    April 10, 2020 @ 11:26 pm

    if that's indeed what happened

    Elimination of reading variants is indeed a thing in China, as I've mentioned here several times. It perhaps flies under most Sinologists' radar because it's not as radical as character simplification.

    I first became aware of it when I discovered a sheet interleaved into a Chinese-Japanese dictionary I bought in Japan many years listing all the changes in character readings. The ones I still remember (somewhat vaguely) are 往 (wàng to wǎng) and 呆 (ái to dāi), as well as (discovered later) 乘 (shèng to chéng). The sheet seemed to be a temporary measure to catch changes that had not yet been incorporated in the body of the dictionary.

    I'm still mystified why people who are familiar with Chinese doubt that this kind of standardisation ever took place.

  13. Jonathan Smith said,

    April 11, 2020 @ 12:02 am

    @Bathrobe Not at all; of course leveling of the kind you're referring to is pervasive across Sinitic. I meant that we might not want to assume that at coinage, the character "法" was selected for "France" yet pronounced differently than the common morpheme 'rule, law' for some reason / by some fiat. Is this your thinking? That is of course possible… but if not, no, it would not be as simple as to say that fa4guo2 is "the original pronunciation."

  14. Bathrobe said,

    April 11, 2020 @ 2:53 am

    It might be "persuasive across Sinitic", but it is a part of deliberate language planning policy in the PRC. That is why these old usages tend to be preserved in Taiwan rather than the Mainland.

    I meant that we might not want to assume that at coinage, the character "法" was selected for "France" yet pronounced differently than the common morpheme 'rule, law' for some reason / by some fiat.

    Then why was it pronounced differently? I would suggest that the concept that the common morpheme 'rule, law' had a fixed, invariant pronunciation is putting the cart before the horse. There was quite a bit of variation in readings in the pre-standardisation era, and, while I could be wrong, I would suggest that maybe the idea of pronouncing 法 differently in 法国 was not quite as strange as modern sticklers for unified readings think it is. That fàguó somehow represented some kind of 'drift' away from the original fǎguó seems strange to me. If fǎguó drifted away, why didn't other words containing the morpheme 法?

    In fact, I would even suggest that Chinese who pronounced 法国 as fàguó might have been perfectly aware that it did not carry the meaning 法 'law, rule' — that is, that it was being used for purely phonetic reasons, to represent the name of a Western country. Even now many Chinese have an awareness that characters representing foreign words are being used for purely phonetic purposes and that tone is not as important as it is in native words.

  15. Philip Taylor said,

    April 11, 2020 @ 4:13 am

    "Persuasive" (BR) or "pervasive" (JS) ?!

  16. Bathrobe said,

    April 11, 2020 @ 5:04 am

    'Pervasive'. Silly mistype.

  17. Victor Mair said,

    April 11, 2020 @ 7:37 am


    "why are the interviewees subtitled – is this routine practice in Taiwan, or is it something specific about how those people spoke?"

    Good question! Sorry I didn't have a chance to respond to it till now.

    This is a phenomenon that exists throughout the Sinosphere. When / One (!!) I first encountered it in films, on television, in theaters for drama and opera, etc., it seemed very odd to me. Later, it slowly dawned on me that this practice bespoke two things:

    1. how few people speak (and understand by hearing) the national / standard / common language

    2. how prevalent the topolects are and how few the people are who understand them outside the areas where they are spoken

    The irony of it all is that a goodly percentage of the viewers are illiterate, so they can't read the subtitles either.

    Mainly, then, I suppose one could say that the subtitles are meant for literate speakers of the national / standard / common language who can't understand what the interviewees are saying.

  18. Philip Taylor said,

    April 11, 2020 @ 7:51 am

    "[VHM] Mainly, then, I suppose one could say that the subtitles are meant for literate speakers of the national / standard / common language who can't understand what the interviewees are saying".

    Is that conclusion true ? If if we restrict the universe of discourse to those speakers of any Chinese topolect who are sufficiently literate to be able to read hanzi, and if (as is widely asserted) the written Chinese language can be understood by native speakers of most (if not all) of the Chinese topolects, whereas the spoken "standard" language (Putonghua) can be understood by relatively few (other than as a second language), then why would not the conclusion of your analysis be "Mainly, then, I suppose one could say that the subtitles are meant for literate speakers of all Chinese topolects who cannot understand what the interviewees are saying but who can read the hanzi" ?

  19. Victor Mair said,

    April 11, 2020 @ 8:12 am

    The subtitles are written in the national / standard / common language.

  20. M. Paul Shore said,

    April 11, 2020 @ 8:20 am

    In discussing the pronunciation fa4guo2, have we ascertained, using the largest historically-oriented dictionaries, that that's the only fourth-tone use of 法 on record? If not, what did the other use or uses mean?

    Also, given all the problems that have been discussed here, why did the coiners of the spelling "法国" make that choice? Could they have been merchants or others interacting with the French who were only imperfectly literate? And why couldn't the word for "France" have been written "珐国" (fa4guo2), "the country of enamel", or "髮国" (fa4guo2), "the country of hair"?

    Finally, why were the presumably hardcore PRC types who, decades ago, began pushing the fa3guo2 pronunciation comfortable with doing that seeing as it seems to flatter France, a member of the evil First World, by calling it "the country of laws", an impression that was already present in the spelling? (Or did they think that for a country to be a country of laws was a bad thing?) Repeating my point in the previous paragraph, why couldn't they, instead of promulgating an altered pronunciation, have promulgated a new spelling "珐国" or "发国"?

  21. Philip Taylor said,

    April 11, 2020 @ 9:01 am

    "The subtitles are written in the national / standard / common language" — OK, understood, and agreed (I did not seek to dispute it). But let us consider a hypothetical native speaker of Shanghainese, who for some reason has never learned any Putonghua whatsoever. If he (or she) were to read those subtitles, and ignoring what mental sound images they would create in his/her mind, what fraction of what was being glossed in the subtitles would he/she understand ? And if the answer is non-zero, to what extent would that also be true for speakers of the other 56 or so Chinese topolects ?

  22. Ed M said,

    April 11, 2020 @ 11:03 am

    It appears she's making a crêpe filled with a squared slice of, what, ham? Or is it filled with Spam? (The food, not the email.)

    Is a hammy/Spammy crêpe a thing in Taiwan?

  23. Jonathan Smith said,

    April 11, 2020 @ 11:55 am

    To be clear, I'm not claiming fa3 is "original" vis-a-vis fa4, but rather that neither this formulation nor its opposite can do justice to the complexity of the problem. To give a couple "early Mandarin"-ish data points…

    According to Coblin's (2007) handbook, 'send out', 'hair', and 'law' are identical in 'Phags-pa spelling (14th c. ish) — all entering-tone words spelled "HWA [fa]". And according to Pulleyblank's lexicon (1991), these three words developed from late medieval "faːt", "faːt", and "faːp" respectively to homophones "fǎ" in the Zhongyuan yinyin (14th c.; note entering tone is gone in this standard.)

    Also compare, e.g., Cantonese, where these three words are homophones faat3 (so 'enamel' mentioned by M. Paul Shore above.)

    So the problem involves the notoriously complicated development/mixture of old entering tone words into Mandarin(s). The mainland "standard" pronunciations of these three words are of course now fā 'send out', fǎ 'rule', and fà 'hair' (but I believe in Taiwan it is fǎ 'hair'…) Maybe certain Mandarin varieties display more consistency here.

    A lot depends on the exact when/where/who of the use of "法" to write "France". So my thought for Bathrobe was that it's of little use to focus overly on "original pronuncition" vs. "sticklers"/"standardizers" (after all, in many cases the latter two types of folks very much work at cross-purposes…)

  24. Calvin said,

    April 11, 2020 @ 1:06 pm

    In 海國圖志 (first published in 1840's,, France was referred to as 佛蘭西, with alternative translations listed: "即佛郎機,一作佛朗西,一作拂蘭祭,一作法蘭西,一作和蘭西,一作勃蘭西". Besides "法", all other "Fa" phoneme characters ("佛", "拂", "和", "勃") are pronounced in second tone.

  25. Jonathan Smith said,

    April 11, 2020 @ 2:23 pm

    Ah interesting, Michael Watts mentioned 佛朗机 above. It looks like coda -t or some reflex of such was preferred in the first syllable of these transcriptions…

  26. Mango said,

    April 11, 2020 @ 2:30 pm

    @Michael Watts: I have seen 19th century texts which add 口 to more or less all foreign geographic names. If I remember correctly (but I'd need to check!) José Martinho Marques did this in his 歐羅巴各國總敘. I don't remember how he treats France, but I remember seeing for England.

    As to the pronunciation fa4guo2, there's a parallel case: Russia, which is É in the PRC but È in Taiwan…

  27. Mango said,

    April 11, 2020 @ 2:35 pm

    In my last comment, it should have been "but I remember seeing (口英) for England". The character is apparently not supported on the forum.

  28. Victor Mair said,

    April 11, 2020 @ 3:21 pm

    When I played the French Horn half a century ago, my recollection is that people referred to it in Chinese as Fàlánxī hào 法蘭西號.

    Now we're supposed to call it "horn".

    Incidentally, Wiktionary gives Manchu ᡶᠠᠯᠠᠨᠰᡳ (fa lan si) for "France".

  29. M. Paul Shore said,

    April 11, 2020 @ 4:15 pm

    The Wikipedia articles "Horn", "French horn", "German horn", and "Vienna horn" contain some illuminating discussions of the varieties and nomenclature problems of modern horns. It turns out that one of the main reasons, perhaps the main reason, that the term "French horn" has fallen into a certain degree of disfavor is that what's usually called a French horn in English is actually a German horn, with a true French horn being something else. (And by the way, just in case anyone should feel tempted to bring up the matter of English horns, they're irrelevant to all this, since they belong to a different instrument family.)

  30. Bathrobe said,

    April 11, 2020 @ 5:17 pm

    Japanese also uses 仏 futsu 'Buddha' for France without making claims that France is Buddhist.

    As Johnathan Smith suggested, speculation about notorious problems of entering tones etc. is less relevant than knowing "who" (Cantonese speaker? Mandarin speaker?) created these terms and "when" (16th century? 19th century?).

    Dismissing the change of fà to third tone as of little use when we don't know the so-called "original" pronunciation is not very helpful. I was discussing the issue of standardisation, which is pertinent linguistic issue. From what I understand, the usual pronunciation in Beijing, the seat of the modern standard, was originally Fàguó, and this was changed. The historical process that led to the pronunciation fàguó is a different issue (one that is being fuzzily presented in terms of "data points") and doesn't invalidate or render irrelevant observations on the process of standardisation/regularisation. Perhaps we can disengage from this thicket of talking at cross purposes if I modify my formulation to "the originally accepted pronunciation in the Beijing-based standard was fàguó". You are then free to pursue your speculations on how the character was chosen and how it might originally have been pronounced in particular varieties of Chinese.

  31. Bathrobe said,

    April 11, 2020 @ 5:33 pm

    In the video in question, it is clear that the pronunciation fàguó is being used because this was the originally accepted pronunciation in the Beijing-based standard, not because of the notoriously complicated development of entering tone in various types of Mandarin. Interesting as that question is, it is not directly relevant to why fàguó is used in the video.

  32. Jonathan Smith said,

    April 11, 2020 @ 8:00 pm

    ^ I divided these two issues into (1) "the separation between fa3 'law' and fa4[guo2] 'France' in the older standard language" (fa4guo2 seems to be traceable to, e.g., the 國音常用字彙 of 1932) and (2) the relatively recent "restoration/leveling to fa3 (if that's indeed what happened)" in my first comment (where incidentally, fa3guo2 seems also to have been the Taiwan standard since at least the 80's.) My basic position on all this, without digging further, is that there has long been variation fa3~fa4 in Mandarin pronunciations of 'France' dating to dunno when and caused by dunno what but might relate to rusheng results. That's why I wrote "if that's indeed what happened." And I do acknowledge that references to "original pronunciation" vs. perceived corruption whether due to standardization or other are probably triggering for me :D IMO the real reason for most of the leveling you have been referring to in Sinitic is not standardization per se at all but rather literacy itself.

  33. Bathrobe said,

    April 12, 2020 @ 12:08 am

    I do acknowledge that references to "original pronunciation" vs. perceived corruption whether due to standardization or other are probably triggering for me :D IMO the real reason for most of the leveling you have been referring to in Sinitic is not standardization per se at all but rather literacy itself.

    I'm sorry, trigger or not, but I think you're wrong. If you were right, there would be no differentiation between the pronunciation in Taiwan and that on the Mainland. All things being equal they should be the same. But there is a differentiation, despite starting from the same place (especially in the city of Beijing), and if you believe that this is unrelated to educational and linguistic policies, then I would ask you to show cause.

    My basic position on all this, without digging further, is that there has long been variation fa3~fa4 in Mandarin pronunciations of 'France' dating to dunno when and caused by dunno what but might relate to rusheng results.

    "Dating to dunno when" and "caused by dunno what" and "might relate to rusheng results" are your response to any claim that educational / standardisation policies might have had a hand. Ok.

  34. Bathrobe said,

    April 12, 2020 @ 2:21 am

    Incidentally, I did not use the word "corruption". I remarked on a change.

    As for rusheng results causing variation between fà and fǎ just in the word 法国 and no other words using the morpheme 法, please explain.

  35. Michael Watts said,

    April 12, 2020 @ 3:08 am


    It seems to me that Jonathan Smith is saying (and I agree) that — at the time when 法 was chosen to represent France — it is unlikely that 法 was chosen and a new pronunciation was then innovated for 法 in the sense "France". Rather, since the character was selected with the pronunciation in mind, the pronunciation that developed into fà was probably already current for 法 in its other senses when this choice was made.

    Because if the existing pronunciation(s) of 法 did not suit the word "France", the obvious thing to do would be to pick a different character with a more appropriate reading, not to pick an inappropriate character and then give it a new reading.

    I'm surprised that you seem to disagree?

  36. Bathrobe said,

    April 12, 2020 @ 3:38 am

    I am sceptical of the claim that 法國 fàguó definitively represents an already existing divergence in character readings. I am of the view that assigning characters to write foreign words at the time was already looser than it was in representing native morphemes. The rigid tendency to one character-one reading that is beloved of standardisers (and, it seems, Jonathon Smith) was less prevalent in pre-modern times than it is now, which would possibly have encouraged tolerance of variant tone readings.

    It am quite happy to be proved completely and absolutely wrong on this, but only deeper research into historical sources would reveal the truth, which I am not in a position to do at the moment since I don't have the resources. But vague suggestions of special treatment of old rusheng characters in late imperial Mandarin don't cut it for me.

    Where I do stand firm is the fact that the split in usage between the Mainland and Taiwan exactly matches the split between recognition of fàguó and fǎguó in the officially recognised pronunciation in each jurisdiction. JS is dismissing this as irrelevant but has provided no grounds for this other than something has "triggered" his dislike.

  37. Leo said,

    April 12, 2020 @ 6:33 am

    @Victor Mair: thank you for your answer! In the video, are all the interviewees both speaking and being subtitled in Mandarin?

  38. Michael Watts said,

    April 12, 2020 @ 11:44 am

    The rigid tendency to one character-one reading that is beloved of standardisers (and, it seems, Jonathon Smith)

    The rigid tendency to one character-one reading that is beloved of standardisers (and, it seems, Jonathon Smith)

    But I don't see this in the comments at all. Rather, I see the question of:

    1. We observe that, in Taiwan, the character 法 has two readings, fǎ and fà. They are differentiated by sense.

    2. The reading fà is exclusive to the sense "France", while every other sense uses the reading fǎ.

    3. The sense "France" is a very recent use of the character.

    Now, the question:

    4. Does the alternation in reading, fà vs fǎ, predate or postdate the use of the character in the sense "France"? That is to say, did the character have multiple readings which then segregated by sense, or did the character first develop multiple senses and only then develop an alternate reading for the new sense?

    The first option, that multiple readings exist and are later sorted into obviously distinct senses, makes much more sense to me than the second option.

  39. Bathrobe said,

    April 12, 2020 @ 5:32 pm

    法國 fàguó is an abbreviation of 法蘭西 Fǎlánxī, one of a number of names that were coined to write the word 'France' (Calvin has already pointed out 佛郎機 Fúlángjī, 拂蘭祭 Fúlánjì, 佛朗西 Fúlǎngxī, 和蘭西 Hélánxī, and 勃蘭西 Bólánxī. Japanese gives us 仏蘭西, in Mandarin Fúlánxī. Modern Taiwan gives us 法蘭司 fǎlánsī). 佛 can also be read .

    All of these are attempts to render 'France' phonetically. Since the Chinese can't pronounce 'fr', the first consonant cluster was broken up into two, as fǎlán, fúlán, fúlǎng, hélán, and bólán. The assignment of characters would, I suggest, have been dependent on the native pronunciation of the person choosing them. But the intent was to somehow capture the pronunciation 'fr' using Chinese characters. It is also reasonable to assume that the people choosing these names did not necessarily intend for the first syllable to be pronounced fully and clearly, with full tone, as it would be in a Chinese word. They were attempting to capture impressionistically what would have sounded to a Chinese speaker like 'felan', 'fulan', or 'falan'. (Interestingly, none of them use characters like 然 rán to capture the 'ran' in 'France'.)

    The name that appears to have become most commonly used is 法蘭西 Fǎlánxī. Since this was intended as an impressionistic rendering of the French word, it was quite possibly pronounced with the first syllable as , or even fa, de-emphasising the first syllable since it is secondary to the second in pronouncing the word 'France'. It is possible that even the second syllable was not conscientiously pronounced as lán.

    法蘭西 was then abbreviated to 法国 in line with common practice in Chinese. In pronouncing 法国, it is quite possible that the pronunciation was adopted out of a consciousness that 法蘭西 was not actually fǎlánxī but something more like falanxi. If, as Jonathon Smith suggests, there was an alternative reading for 法, this might have helped in the choice of this character, and also given rise to the reading 法國 as fàguó. But it is not essential.

    The above is merely surmise on my part. But it does not seem to have any less basis than the suggestion that reflexes of the entering tone were still current in late imperial China and resulted in the pronunciation fàguó for 'France'.

    I suggested that Jonathan Smith is taking a rigid approach to the pronunciation of characters because he insists that by using 法 Chinese speakers were irrevocably committed to using the pronunciation , and that only by positing a second reading of can we explain the pronunciation fàguó.

    I welcome comments from people who are better versed in this than I am (and I pretend to no expertise in the history of Chinese phonology) to throw some light on the issue. Three data points drawn from the 1980s, 1930s, and the 14th century respectively aren't very convincing to me.

  40. Neil Kubler said,

    April 12, 2020 @ 10:30 pm

    Historically, as others have pointed out, 法 is an "entering tone" syllable and since Modern Standard Mandarin (MSM) has no entering tone, its tone was (subconsciously, by speakers, over decades and perhaps centuries) "assigned" across the existing 4 tones in MSM, so that it is now the only character in the Chinese language that can be pronounced with all 4 different tones! So we have mei far "no way" (Tone 1), mei fazi "no way" (Tone 2), banfa(r) "way of doing things" (Tone 3), and Faguo (Tone 4 followed by neutral tone). Some older speakers in Beijing still pronounce the Fa- in Faguo with Tone 4. This pronunciation was brought to Taiwan in 1945. Later, language planning people in the PRC, in an effort to "simplify," accepted the (technically incorrect) "reading pronunciation" of many less educated people and non-Beijing speakers of Faguo with Tone 3 for Fa-, so that most younger PRC Chinese (including even Chinese teachers from Beijing) now pronounce Faguo with Tone 3. In Taiwan, after increased contact with PRC since the 1990s, the pronunciation Faguo with Tone 3 for Fa- has actually become more common, with the result that in Taiwan one now hears both Tone 4 and Tone 3 for this word. Since there are users of both in the various Chinese-speaking societies, we need to help learners comprehend both; they can use whichever they prefer in their own speech.

  41. Bathrobe said,

    April 12, 2020 @ 10:36 pm

    Thank you, Neil Kubler. I realise you are not taking sides, nor even backing me up. I do, however, appreciate the clarity you've brought to this, which was previously lacking.

  42. Bathrobe said,

    April 12, 2020 @ 10:44 pm

    Incidentally, I'm familiar with all the tones you mentioned (没法儿做 méi fār zuò, 没法子 méi fázi, 没办法 méi bànfǎ, and of course 法国 fàguo), without consciously noticing that they were all pronounced with different tones. They just sounded 'right'. Thank you for pointing it out.

  43. Victor Mair said,

    April 13, 2020 @ 8:05 am

    From Alain Peyraube:

    I have also always learned that the FA of FAGUO should be a fourth tone.

    And when I was in Beijing in the beginning of the 1970's, it was indeed the case for all old people in the countryside !

  44. Victor Mair said,

    April 13, 2020 @ 8:11 am

    From Tsu-Lin Mei:

    I watch three Chinese language TV programs every day. 中天,CCTV, & Phoenix 鳳凰台,Taiwan, Peking, & Hong Kong. 法囯 is for me FA (tone 3).

    When I was teaching at Harvard, Iris Pian sometimes convened the staff to determine what we say and what we should teach our students. For these occasions, I usually passed. My Mandarin has been corrupted by 8 years of residence in Shanghai. Iris also does not qualify. Her Mandarin was Y.R. Chao's official version. So we deferred to Mrs. Lin to be the arbiter.

    My student Sun Chaofen is professor at Stanford and leads the Chinese language program at Stanford. He told me he cannot find a single speaker in Peking who speaks in the way prescribed by Mandarin Primer. I remember when I was teaching at Harvard, I used to listen to the phonographic records of Mandarin Primer before going to class. For your question, you should ask Perry Link. He was our chosen standard bearer for Y.R.Chao's Mandarin.

  45. Victor Mair said,

    April 13, 2020 @ 11:27 am

    From San Duanmu:

    T3 is indeed commonly used on the mainland, while T4 can still be heard, slightly old fashioned perhaps. The variation is likely due to the fact that when Ru disappeared in Mandarin, its members split among three other tones,known as 入派三声. Specifically, syllables with voiceless onsets split among T1, T2, and T4, often unpredictably. Therefore, both fa3 and fa4 could have been an expected choice. As to why there is a shift from T4 to T3 remains unclear to me.

  46. gds555 said,

    April 13, 2020 @ 12:20 pm

    Hmmm, Peyraube . . . Bathrobe . . .Could the latter be a pseudonym for something like Balthasar Peyraube, Alain's secretive, scrappy Sinological-rival relative?

  47. Victor Mair said,

    April 13, 2020 @ 2:12 pm

    From San Duanmu:

    The fact that different tones are still in use for 法, such as 没法儿看(T), 没法子(T2), and 没办法(T3), suggests that which new tonal category a Ru syllable goes into may be context dependent. The same is true for 一, as in 第一(T1), 一个(T2), and 一豌(T4). It is possible that when France was translated, several tones of 法 were still in use (as 法 and 一 are today), and the translator picked T4, probably because the word France has a falling intonation (in English). Then, there is a competition between tradition and leveling. It is worth noting that, in some dialects, Ru syllables did not split among other tones, but mostly merged with one tone. Speakers of such dialects could have given rise to the leveling consequences on the mainland, and the lack of such speakers in Taiwan may be the reason for the lack of leveling there.

  48. Chau said,

    April 13, 2020 @ 4:26 pm

    In Taiwan the fourth tone pronunciation of 法 is found strictly in translated terms: 法國 French Nation, 法蘭西 France, 法語 French, 法郎 Franc, 法蘭克 Franks, 法蘭絨 flannel, 法國號 French horn, 法網賽 French Open, and 法西斯主義 Fascism. According to this rule of usage, the 法 in 法蘭克福 (Frankfurt am Mein) should be in the fourth tone too, but I heard it almost always in the third tone. The fourth tone pronunciation is a mark of the pre-1949 Republican MSM brought over to Taiwan in 1947. Since then there is a split of MSM, each going its own way as Neil Kubler explained it well. I find it interesting that 'computer' is translated as 電腦 'electric brain' in Taiwan but as 计算机 in China. In contrast, 'calculator' is translated as 計算機 in Taiwan but as 计算器 in China.

    In addition to the 法 of 法國, there is another idiosyncrasy in translating Russia as Mango brought it up earlier. Normally 俄 is pronounced with the second tone é such as 俄頃 éqĭng 'a short while' and 俄然 érán 'suddenly'. But in translating Russia, 俄 is given the fourth tone: 俄國 Èguó and 俄羅斯 Èruósī. However, for translating Ohio, Oregon, and Oklahoma, it reverts back to the second tone.

    Another mark of divergence between Taiwan and China is in the tone of 亞 as in 亞洲 'Asia'. 亞 is pronounced in the fourth tone in China. In Taiwan the Mandarin dictionaries give third and fourth tones as alternatives. But in reality, 亞 is pronounced in the third tone almost uniformly. 亞軍 yăjūn 'the runner-up', 亞洲 Yăzhōu 'Asia', 亞熱帶 yărèdài 'subtropical belt', etc. Even when 亞 is followed by another third tone morphosyllable, the rule of tone Sandhi applies, by which 亞 is shifted to the second tone. For examples, in 亞馬遜河 Yămăsūnhé 'Amazon River' and 亞理斯多德 Yălĭsīduōdé 'Aristotle', the 亞 is read as Yá.

  49. Victor Mair said,

    April 13, 2020 @ 6:09 pm

    From Perry Link:

    Here's a little note on 法: Y.R. Chao, in Concise Dictionary of Spoken Chinese, counts a first-tone pronunciation as authentic Mandarin in the phrase 没法儿 [Verb]. That, together with 法子, which he counts as an authentic second-tone usage, not tone sandhi, means that 法 is a very rare (unique?) example of a character that can be read in all four tones.

  50. Bathrobe said,

    April 13, 2020 @ 8:16 pm

    Thank you to our learned commentators for throwing more light on this.

    I was working on the assumption that the word 'France' must first have come into the speech of the Mandarin class and circulated among them in spoken form, as falanxi or something similar. Since this was a word from a toneless language, the assignment of tones may have been haphazard or idiosyncratic. A number of different ways arose to write it, all reflecting in some way the phonotactics and range of syllables available to Mandarin, and the indeterminacy of the pronunciation, especially the tones.

    As Jonathon Smith rightly pointed out, the disappearance of the entering tone gave rise to tonal variants of 法 (not two but three or four) in Mandarin, making 法 one choice for writing 'France' (i.e., 法蘭西). Fàlánxī was likely the intended pronunciation because (not ) was perceived as close to how 'France' was pronounced by foreigners.

    However, 法國 was incorrectly but popularly read as fǎguó among the populace (levelling), which was sanctioned by putonghua language authorities in the PRC to simplify character readings. This was not followed by their counterparts in Taiwan, contributing to the general disappearance of fàguo on the Mainland but its continued survival on Taiwan. However, fàguo is under threat on Taiwan as fǎguó continues to gain ground through levelling and influence from the Mainland.

    Is this a reasonable summation of the situation?

  51. Bathrobe said,

    April 13, 2020 @ 8:44 pm

    With regard to 计算机, as a terminological issue this appears to be the 'officially-approved' term for a computer on the Mainland. But while some people conscientiously adhere to this usage, and while it is de rigueur in more official contexts, 电脑 is extremely common on the Mainland.

  52. Jonathan Smith said,

    April 13, 2020 @ 9:04 pm

    Waha this became quite the conversation 8D
    @Bathrobe More or less IMO, except the idea that at the outset "Fàlánxī was likely the intended pronunciation" is entirely unclear given the above info and depending on time/place of coinage may not even be a meaningful claim. The point re: so-called "character readings" in Chinese is really *variation* across time and space vs. countervailing *leveling* that tends to remove it (esp. "among the populace" as you say). And the same thing continues to happen to 没法儿, 法子 (now it seems rarely pronounced 2nd tone), and countless others without any accompanying diktat… though needless to say diktat has a role to play…

  53. Bathrobe said,

    April 13, 2020 @ 9:33 pm

    depending on time/place of coinage may not even be a meaningful claim

    I was assuming that it became current among the Mandarin-speaking class, which may or may not be right. Still I'm curious about the reason for the final 西, which doesn't match the French pronunciation or the English.

  54. Michael Watts said,

    April 13, 2020 @ 9:59 pm

    Still I'm curious about the reason for the final 西, which doesn't match the French pronunciation or the English.

    It has always been my strong impression that Chinese transcriptions of foreign names are chosen by some other means than trying to match the sound closely.

    Just in this thread, I see no reason for the first syllable in 亞理斯多德 yă-lĭ-sī-duō-dé 'Aristotle', when 阿基米德 ā-jī-mǐ-dé 'Archimedes' starts with a more appropriate "a".

  55. Bathrobe said,

    April 13, 2020 @ 10:10 pm

    Yet in Japanese 亞 is a….

  56. Jonathan Smith said,

    April 14, 2020 @ 2:58 am

    For those still following :D, the best I can do for now / from here seems to be the second edition (1912) of Gile's A Chinese–English Dictionary (first ed. 1892) at Under the entry for 法 (pp. 410-411) we find a number of words with specific notation by the author indicating deviation from Tone 3, to wit:

    沒法fa2子 there is no help; no remedy
    法fa4場 an execution ground
    法fa2碼 standard weights
    法fa4旨 the precepts of the Buddhist law
    法fa4語 word of the Law; words of admonition
    法fa4寶 the Sutras
    法fa3 or fa4國 or 法蘭西 or 法朗西 France
    法fa4蘭 enamel [with cross-reference to 琺]

    So Giles encountered the variation I referred to above in the standard language of his time. I suppose we might further speculate that fa4, standard (?) in fa4lang2 琺瑯 'Frankish Empire > Frankish enamel', thence spread also to the alternate writing fa4lan2 法蘭 found in Giles and further, somewhat erratically, to 法蘭西/法朗西 and 法國.

  57. Jonathan Smith said,

    April 14, 2020 @ 3:12 am

  58. Victor Mair said,

    April 14, 2020 @ 8:35 am

    From South Coblin:

    This issue revolves around the so-called qīngrù 清入 question in Pekingese pronunciation.

    Briefly, QYS rùshēng words with earlier voiced initials tended to develop different modern tonal reflexes in closely related northern Mandarin dialects. And since Peking was a major demographic metropolis into which people from various surrounding areas tended to immigrate, competing pronunciations appeared in the city vernacular, which was originally itself of course just one of the numerous northern type dialects. In many cases, there tended to develop a sort of consensus as to which tone should be used in particular environments. But in other cases, the competition continued and is sometimes still in play to this day. In the case of 法, there were at least three competitors, i.e., fá, fǎ, and fà. Now, a consensus developed in the case of the word fázi in the vernacular. And fǎ came to be the agreed upon form in compounds like fǎlǜ and Fófǎ. But in the word for "France", which was in any case a later foreign loanword rather than a received form, the competition continued.

    In the course of my life, I have heard both Fǎguó and Fàguó from native speakers of Pekingese. Li Fang-kuei said Fǎguó. Jing [VHM: South's wife], who is of course from Taiwan, also says Fǎguó, but she knows that in modern Taiwan Guoyu they say fàguó. You may have noted a somewhat similar situation in the case of "hair" in Modern Standard Chinese. In Taiwan one says lǐfǎdiàn, but in prescribed Mainland PTH one finds lǐfàdiàn. Sir Thomas Wade noted that both fǎ and fà were used for "hair" in Peking of his time. Situations of this type are common in metropolitan patois.

    The late Prof. Albert Marckwardt, who was my first teacher of historical linguistics at Princeton and was an Anglicist, told an amusing story that is perhaps pertinent here. One time, somebody asked Dr. Johnson regarding the word "either", "Which is right, [iðər] or [aiðər]?" Dr. Johnson replied, [nᴇðər] "neither" (where the first syllable was pronounced like the neigh- of "neighbor"). Thus, in the standard London pronunciation of that period, there were *three* variant pronunciations of "either" and "neither", only two of which survive today. Chaque mot a son histoire.

  59. Chris Button said,

    April 14, 2020 @ 10:54 pm

    Far more vexing for me than the Mandarin tone of 法, is its graphic composition as an abbreviation of 灋 (it now looks like it should be pronounced as if 去 is its phonetic) and its broader etymological relationships…

  60. Victor Mair said,

    April 15, 2020 @ 7:14 am

    From Tsu-Lin Mei:

    Standard Mandarin has tone sandhi. If two 3rd tone syllables are in a row, then the first 3rd tone syllable changes to 2nd tone. E.g. 買馬 "buy a horse" sounds the same as 埋馬 "bury a horse". I propose a test. 美法聯盟 "America-France alliance" occurred during the War of Independence , when the fledgling America was allied with France, against England. 美 & 法 are both in the 3rd tone according to my pronunciation. So tone sandhi occurs. On the other hand, if 法 is in the 4th tone (departing tone), then there would be no tone sandhi.

RSS feed for comments on this post