The Sinophone

« previous post | next post »

I think about the problem of the Sinophone every day, but I haven't written about it very often on Language Log (see "Readings" below).  We have Anglophone (English-speaking), Francophone (French-speaking), Hispanophone (Spanish-speaking), Germanophone or Teutophone (German-speaking), Italophone (Italian-speaking), Lusophone (Portuguese-speaking), Russophone (Russian-speaking),  Hellenophone (Greek-speaking), Arabophone (Arab-speaking), etc.  So why not Sinophone, since diasporic Sinitic speakers are spread widely around the world?

About fifteen years ago, several of us who were interested in the subject independently started to use the term "Sinophone", but credit is usually (and I think rightly) given to Shu-mei Shih for coining and popularizing it in written publications (2004 and especially 2007).

What prompted me to write this particular post was a hot topic on WeChat (China's "app for everything") during the last few days.  The debate was about the prescribed revision of the pronunciation of dozens of Chinese characters.  Some of the changes amounted to no more than a slight difference in tone, consonant, or vowel, while others involved a radically discrepant pronunciation, e.g., xún mázhěn instead of qián mázhěn 荨麻疹 ("nettle rash; hives; urticaria").

Two articles that described the required changes are here and here.  The second article was written by a linguist of the Chinese Academy of Social Sciences.

How all this relates to the matter of "Sinophone" was brought up by a bright M.A. student from the PRC at Penn who observed:

It is so amazing that many characters that we were required to remember how to pronounce one way turn out to be different now! It always seems that standardization of pronunciation has a great symbolic meaning in China, as linguistic uniformity signifies national unification — a tradition ever since the Qin dynasty (221-206 BC) when the standardization of characters and chariot tracks throughout the empire opened the first page of a unified dynasty.

However, linguistic hybridity has become so problematic that one standard can hardly be applied to a wide range of population. For instance, the films made in Chinese used to be called "Chinese films". But there are a large number of Singaporean films and Malaysian films also made in Chinese. Then those films are renamed as "Chinese language film". What about Tibetan films, Uyghur films, Cantonese films, Sichuanese films, etc.? The solution is the umbrella phase" Sinophone film". This appellation seems to have worked more or less, yet Chinese companies also invest a lot in English language films. Are they Sinophone then?

I guess many Chinese high school students are struggling to remember the right pronunciation of these unstable characters so that they can pass examinations, just as what I experienced years ago. Good luck to them!

It's remarkable that there's still such dissension over the pronunciation of so many basic characters.  On the other hand, it's natural that it should be that way with a population of nearly 1.4 billion, which is characterized by tremendous linguistic variety, even among Sinitic speakers, not to mention the more than fifty ethnic groups who speak non-Sinitic languages.  The Chinese government wants all the people to be speaking and writing the same language, but that is virtually impossible considering the low level of mutual intelligibility that exists among PRC citizens when they are speaking their own languages.


"Sinophone and Sinosphere:  Implications for Language and Literature in the 21st Century"

Lecture delivered at Charles University, Prague, Czech Republic on Friday, January 18, 2019

Victor H. Mair
University of Pennsylvania



It has scarcely been two decades since the concept of "Sinophone" was launched, so it is still a young concept and in need of clarification.  To whom does it apply, and how does it relate to the notion of "Sinosphere"?  We will focus primarily on linguistic and literary issues, but will also take into consideration political and cultural aspects of Sinitic, which subsumes both Sinophone and Sinosphere.


1. Sinophone (cf. Anglophone, Francophone) — Diasporic Sinitic

2. Sinosphere / James Matisoff / Hànzì wénhuà quān 漢字文化圈 ("Sinographic Culture Sphere")

3. Sinographic Sphere (Ross King, Wiebke Denecke, Peter Kornicki)

4. Sinoform (Tangut, Khitan, Jurchen, Nom)

5. Shu-mei Shih — first proponent of Sinophone

6. Sheldon Lu — first critic of Sinophone

7. Jing Tsu — early author on the concept of Sinophone

8. David Wang — early discussant of the concept of Sinophone

9. conferences — Harvard University and Boston University

10. Cambria Sinophone World Series — see link below

======linguistic and historical background========

11. Literary Sinitic / Classical Chinese

12. vernacularization

13. biànwén 變文 ("transformation text")

14. Middle Vernacular Sinitic (MVS)

15. Zhōnggǔ Hànyǔ 中古漢語 ("Middle Sinitic")

16. Zhu Qingzhi 朱慶之 — co-editor with VHM of a massive dictionary of MVS

17. fāngyán 方言 ("topolect"); tōngyán 通言 ("dialect")


“Sinographic Sphere Studies: Problems, Prospects, and Priorities”

Keynote lecture by VHM at "Seeking a Future for East Asia’s Past:  A Workshop on Sinographic Sphere Studies"

Boston University, Friday, April 27, 2018 from 8:30 a.m. to 6:30 p.m.



Cambria Sinophone World Series

Kornicki, Peter (2018).  Languages, Scripts, and Chinese Texts in East Asia.  Oxford University Press.

Lu, Sheldon Hsiao-peng (2007), "Review of Shih Shu-mei's Visuality and Identity: Sinophone Articulations across the Pacific," Modern Chinese Literature and Culture.

Mair, Victor (2012), "Sinophone and Sinosphere," Language Log (11/8/2012).

Shih, Shu-Mei (2004), "Global Literature and the Technologies of Recognition," PMLA 119.1, 16–30.

Shih, Shu-mei (2007), Visuality and Identity: Sinophone Articulations across the Pacific, University of California Press.

Tsu, Jing (2010), Sound and Script in Chinese Diaspora, Harvard University Press.

Tsu, Jing and David Der-wei Wang, eds. (2010), Global Chinese Literature: Critical Essays, Brill.

Wikipedia article on "Sinophone".

Sinophone Borderlands – Interaction at the Edges (major, multifaceted, multidisciplinary research project at Olomouc University, Czech Republic)



[Thanks to Qing Liao and Tong Wang]


  1. AntC said,

    March 1, 2019 @ 12:26 am

    the prescribed revision of the pronunciation of dozens of Chinese characters.

    In anything to do with language within PRC, I suspect 'big brother'.

    Are these revisions (sorry I can't read the links in MSM) more like peevery: ee-ther vs ay-ther or where to put the stress in 'controversy'? Or are they favouring one (regional) pronunciation of MSM over another? Or are they trying to root out influences from non-MSM topolects? (For example a Cantonese brogue in MSM.)

    I'm unconvinced about the idea of "Sinophone": with those other -phones you list, speakers are just about mutually intelligible if speaker/listener co-operate on not being too 'broad'.

    Whereas (say) MSM vs Cantonese don't even use the same characters. (I don't mean simplified vs traditional script; I mean there are words in Cantonese that have no counterpart in MSM, so need a different character.) Pointing this out is to counter what seems to be a common myth amongst the English-speaking nations that Chinese speakers can communicate by writing things down, because those characters are ideograms.

  2. Bathrobe said,

    March 1, 2019 @ 2:23 am

    This post is actually about the attempt of the linguistic authorities to 'regularise' the readings of Chinese characters, a process that I have pointed out in previous comments.

    As a part of language modernisation/standardisation, the authorities have over a long period of time attempted to move the writing system towards 'one character-one reading' by weeding out alternative readings. Early examples of this include the assignment of a single reading to 往 wǎng (previously wàng and wǎng) and the regularisation of 法国 to fǎguó (previously fàguó). This process has been taking place gradually but steadily, and is virtually invisible unless you remember the old readings or happen to have an older dictionary. Due to the fact that Taiwan is a separate "jurisdiction" (I will not say "de facto independent" as that immediately raises hackles among patriotic Chinese, but the effect is the same), this leads to divergences between Mainland and Taiwanese practice in 'reading characters'.

    Astute observers will notice that the process of standardising readings amounts to putting the written language ahead of the spoken. Changes involve not only individual characters but also words containing the characters, unless there is special provision to preserve older readings for certain groups of words. The spoken word is thus expected to change in order to accommodate newly authorised character readings.

    As one example of this, when the reading of 乘, previously chéng and shèng, was unified to chéng, the word for "Mahayana" changed overnight from dàshèng to dàchéng. The Wikipedia article on 大乘佛教 (Mahayana Buddhism) helpfully points out that the old pronunciation and current Taiwanese pronunciation are based on the reading shèng.

    Few people appear to notice the deep-rooted assumptions about the relationship between writing and pronunciation that this involves. Whereas there is some kind of awareness in the West that "x word is spelt a certain way" (although this breaks down quite frequently in the case of spelling pronunciations), in Chinese the standard locution is "x character is read this way". The character is regarded as "standard" and the pronunciation as secondary. That is perhaps why Chinese are often surprisingly docile in the face of the top-down standardisation of their language through the regularisation of character readings.

  3. AntC said,

    March 1, 2019 @ 3:26 am

    Thank you @Bathrobe for a clear explanation. All it does for me is raise more questions.

    move the writing system towards 'one character-one reading' by weeding out alternative readings.

    But why? We might complain about English orthography as compared with German or Italian. Having two readings for the characters "read" ('reed' vs 'red') is hardly an enormous drawback; no attempts at spelling reform have really caught on. Because a) we'd all have to retrain out reading/spelling habits; b) we'd still have to know the older pronunciation and spelling to read historical texts. (Just as in practice Chinese need to know both simplified and traditional characters.)

    Does this approach mean the authorities are introducing new characters to represent the reading (and its sense) that got weeded out? Having to memorise even more characters seems nuts.

    The character is regarded as "standard" and the pronunciation as secondary. That is perhaps why Chinese are often surprisingly docile …

    But that only applies to MSM(?) For most other topolects, there's no "standard" way to write them — especially for words that are not cognates of MSM words. For those topolects we have to say the pronunciation is "standard" — it's not very different to a pre-literate community.

    How is this affecting the pinyin? Did 乘 previously have two pinyin forms (chéng and shèng) and now has one?

    It's not just that dictionaries go out of date: all the apps on devices will need to learn the new pinyin; and the voice recognition. (I observe most people rely on those; and very often wouldn't know if Siri/Google had chosen the wrong character. And the recipient wouldn't know either: they use Google to speak the message.)

  4. Bathrobe said,

    March 1, 2019 @ 5:08 am

    But why?

    It's part of the language reform that has been going on for the past century. Characters were simplified, pinyin was introduced. The readings are also being simplified gradually to make MSM (Modern Standard Mandarin) easier to learn and read. It's language reform from the top down, and it does catch on because it's taught in schools.

    Does this approach mean the authorities are introducing new characters to represent the reading

    They are just simplifying the readings of existing characters. Why would they need new characters?

    For most other topolects…

    I think character readings of topolects are gradually falling by the wayside (although other commenters might have a better idea than me). Children are taught MSM, and taught in MSM, at school. Apart from HK, I don't think there is any systematic teaching of the topolectal reading of characters anywhere in China nowadays. I've witnessed this first hand, as I might have mentioned. I asked a lady from Fujian to pronounce 今日 'today' and 日本 'Japan' in her native topolect. She told me they called 日本 rìběn (Mandarin), although they pronounced 今日 in their own topolectal pronunciation. So for her there did not appear to be a systematic system of reading characters in her own topolect. We might have had our wires crossed, and there might be people who maintain the old sysems of reading, but I think that what is happening is that reading characters in topolectal pronunciations is falling by the wayside. The old idea that 'Chinese characters unify the disparate topolects' is being replaced, as I see it, by a 'standard vs dialect' situation. At any rate, that seems to be the current direction. So yes, topolects are being even further reduced to spoken patois than they used to be.

    Did 乘 previously have two pinyin forms (chéng and shèng) and now has one?

    In a word, yes. Shèng should have disappeared from dictionaries as an alternative reading (although I don't have access to recent dictionaries to confirm this). So 大乘 is now dàchéng. Full stop. I daresay that apps will be updated to confirm with the new standards.

  5. TIC said,

    March 1, 2019 @ 6:22 am

    This is a perfect example of the type of Sinitic-related LL post (most of them authored by Dr. Mair, of course) that I find fascinating and which, including their comments, I try hard to follow and comprehend… But doing so is inevitably difficult and frustrating (and, no doubt, ineffective/inaccurate) because, despite my layman's efforts to improve my basic understandings of many Sinitic-related topics (including, pertinent to this post, tonal variations), a good grasp of many such things just continues to elude me… Until now, of course, because all the blame is on me, I've resisted the urge to chime in to ask to have some (or sometimes even all!) concepts in a thread "dumbed down" for me…

    In case case, though, my intense interest forces me to make a exception… I promise to try harder to educate myself better in these areas… But would someone please help me by, if possible, attempting an analogy with what comparable efforts in English might look/sound like?… I get that this more than just an eee-ther vs. eye-ther (as a mandated pronunciation of "either") thing… But I just can't get a handle on what kind of change it might in fact be analogous with in English… To be honest, I'm not even clear on whether these efforts are focused on reducing (recognized) uncertainty/confusion/unintelligibilty, or on eliminating (unrecognized) misunderstandings/misinterpretations, in spoken communications… (Many thanks in advance to anyone who might respond and help me out!)…

  6. David Moser said,

    March 1, 2019 @ 6:47 am

    When are the Chinese language authorities going to adopt DESCRIPTIVE instead of PRESCRIPTIVE standards?? At least that agenda would have a chance of becoming more universally adopted.

  7. TIC said,

    March 1, 2019 @ 7:30 am

    I just reread this post and its comments (for at least the third time) and it occurred to me that my fundamental confusion is perhaps best illustrated by the fact that, even after carefully reviewing your excellent comments a few times now, Bathrobe, it remains entirely unclear to me whether, in referring to the "reading" of characters, you mean the interpretation/understanding — or the pronunciation — of them… Sorry for being so dense…

  8. Bathrobe said,

    March 1, 2019 @ 8:01 am


    I really should wait for Professor Mair on this one as he is very much an expert in these issues. But anyway…

    Towards the end of the last imperial dynasty at the end of the 19th century, China had a complex linguistic situation. To simplify:

    1) The scholar officials wrote in Literary Chinese (which was like European countries using Latin),

    2) However, quite a lot of literary production had come to be written in the spoken vernacular, mostly the northern vernacular based on Beijing,

    3) The scholar officials used Mandarin (Beijing-based vernacular) for oral communication,

    4) Ordinary people use their local 'patois' for communication.

    Mandarin was the most prestigious dialect, but each major local dialect (or topolect) had a tradition of reading Chinese characters in their own dialect pronunciation. That meant giving each character an appropriate reading in the local dialect. Check人#Pronunciation, which gives the pronunciation of the character 人 in different dialects. In Hong Kong, this is still the case — the locals read standard Mandarin Chinese with Cantonese pronunciation! There is a certain amount writing in the local Cantonese dialect, or something close to it, but it's not official and it's not 'high class'. You might compare it to Scottish people writing and speaking standard English — in a Scottish accent, of course, instead of "Scots Wha Hae" (by which I mean, pure Scottish dialect).

    At the end of the 19th century, however, the Chinese discovered that they were outclassed by the technologically advanced (and rapacious) Western powers, giving rise to a national crisis. Among other things they decided they had to completely modernise their language. That meant (after much deliberation):

    1. Deciding on a new vernacular standard to replace the old literary language. This, as I noted, already existed for writing vernacular literature.
    2. Deciding on a national standard pronunciation — Mandarin, based on Beijing
    3. Simplifying all those complex characters, which were fine if you were a scholar preparing for the imperial exams but pretty cumbersome for achieving universal literacy among your people.

    It's about a hundred years since this process began, and the Communists, who were naturally iconoclastic by temperament, carried it through boldly.

    One of the problems with Chinese characters is that they often have several different readings. One example is the Chinese poet that you might know by the name Li Po. In Chinese this is 李白. "Po" (actually ) is an old, “genteel”, reading of the character 白 'white'. This character is more normally read bái 'white' in everyday life. To simplify things, the Chinese have now got rid of the reading and settled for one single reading: bái. As a result, in China today Li Po is known as Li Bai. Very few people will will understand you if you call hime Li Po. People no longer have to learn multiple ways of reading the character 白, making it a lot easier on the memory. There are quite a few characters like 白 that can traditionally be read in multiple ways. By getting rid of multiple readings, the idea is to make the script more 'modern' and easier for ordinary people to learn. If this means that Li Po must be rechristened Li Bai, then so be it.

    That is basically what we are talking about.

  9. Bathrobe said,

    March 1, 2019 @ 8:14 am

    @ TIC

    Yes, the reading is the pronunciation. But different pronunciations can accord with different nuances of meaning.

    For example, 乘 chéng has several meanings, including 'to ride (a vehicle)' and 'to the power of' — don't ask me why it has these meanings, but it does!

    In certain words, however, like 大乘 dàchéng 'big vehicle' or 'Mahayana', the same character 乘 was traditionally read shèng. 'Hinayana' is called 小乘 xiǎoshèng 'little vehicle'. There are also a few other words in which 乘 is read shèng. To simplify things, the linguistic authorities decided, "To hell with it, read them all as chéng and have done with it. That will simplify things a lot. So from now on, it's dàchéng not dàshèng".

  10. Ken said,

    March 1, 2019 @ 8:23 am

    @TIC, I think the analogous change in English would be to require that each letter have only one pronunciation, but there would be no spelling reform. See Bathrobe's first comment above, on how the pronunciation of "Mahayana" changed.

    For example "caution" might be pronounced with /s/ (not /k/) for "c" and /t/ (not /ʃ/) for "t". It might end up with four syllables, if "au" and "io" were no longer diphthongs!

  11. J.W. Brewer said,

    March 1, 2019 @ 9:27 am

    Of course radically different readings of the same character are ubiquitous in Japanese, but that certainly did not prevent Japan from getting ahead of China (when it came to growing prosperity and coping with modernity) back in the Meiji era if not earlier and staying ahead ever since. Come to think of it, while VHM and others often try to debunk the "myth" that Chinese characters are primarily ideographic, I wonder if that is a more plausible and less mythical account of kanji in their Japanese context. Consider, as a simple example, how 中is the written representation of both chūō and naka. It's as if English orthography had a single glyph/character that was the written representation of both "center/central" and "middle," where the two readings are connected only semantically, rather than being divergent pronunciations both descended (via different sound changes in different topolects) from a common etymological ancestor.

  12. TIC said,

    March 1, 2019 @ 9:33 am

    Thanks very much, Bathrobe and Ken, for those patient and informative explanations… I'm definitely beginning to see the light… And I'll reread and further analyze/internalize those (and any additional) comments…

  13. Randy F McDonald said,

    March 1, 2019 @ 10:09 am

    Question: How does Arabic deal with the representation of its own internal linguistic diversity?

  14. Philip Taylor said,

    March 1, 2019 @ 10:09 am

    Bathrobe ["You might compare it to Scottish people writing and speaking standard English — in a Scottish accent, of course, instead of "Scots Wha Hae" (by which I mean, pure Scottish dialect)."] — Two points, if I may ? (1) I am not at all convinced that the Scots can "write standard English in a Scottish accent", but more importantly, when it comes to the parallel with the Cantonese-speaking regions of China, when the Scots "speak standard English in a Scottish accent", this is not something artificial, any more than when (for example) a West Indian pronounces Standard English with (e.g.,) a Barbadian accent. For both (and of course for all) English-speaking nations, it is the most natural thing in the world to speak/pronounce standard English with the local accent; thus I, for one, would accept that it is perfectly natural for the Cantonese-speaking Chinese to pronounce Mandarin/Putonghua with a Cantonese accent/pronunciation.

  15. Bathrobe said,

    March 1, 2019 @ 10:21 am

    @ Philip Taylor

    TIC asked me to come up with English-language parallels. I scratched my head and came up with that. Rather far-fetched but what can you do when the languages, writing systems, and situations are so different?

  16. TIC said,

    March 1, 2019 @ 10:49 am

    Bathrobe, I very much appreciate your effort to come up with something along the lines of an English language analogy for me… It helped greatly…

    Philip, I also appreciate your constructive critique of that analogy… It helped too…

    JW, I'm struggling with the analogy at the end of your comment… It seems to require one to imagine that, in English orthography, individual glyphs/characters represent (or, perhaps, conjure up?) ideas/concepts as opposed to sounds… What am I missing?…

  17. Scott P. said,

    March 1, 2019 @ 11:55 am

    @TIC, I think the analogous change in English would be to require that each letter have only one pronunciation, but there would be no spelling reform.

    Sounds like Roger Wright's theories on the shift from Latin to Romance, which he think was triggered by the Anglo-Saxon Alcuin coming to Carolingian France and being shocked that everyone didn't pronounce Latin the way he did (since, according to Wright, spelling lagged behind changed in vulgar speech). Charlemagne tasked him with ecclesiastical reform, which included pronunciation reform — the Church was now going to voice all those Latin letters that had become silent — making Church Latin a clearly different language than Romance.

  18. Victor Mair said,

    March 1, 2019 @ 1:42 pm


    Thank you for your good questions. I appreciate your candor. If we don't understand something, we should ask questions to clarify matters, not complain because we don't comprehend. Bathrobe is doing a heroic job of explaining things, and I'm grateful for his contributions. Here I'll just add a few notes of my own that I was planning on saying anyway.

    In the case of "readings" of characters, in Sinitic language studies we're talking mainly about their pronunciations. The problem, however, is that often these different pronunciations carry different meanings.

    For example, one of the items mentioned in the articles I cited is 骑 / 騎. Read / pronounced qí, it means "ride (a horse); sit on the back of; sit astride; to mount (someone in sex)". Read / pronounced jì, it means "horse or vehicle that one rides on; mounted soldier; cavalryman; (literary) classifier for soldiers or warhorses; surname​". It's obvious that qí is for verbal usages and jì for nounal usages.

    All well and good; makes a lot of sense: qí and jì are close cognates, but, alas, they're written with the same character. There's no indication in the way 骑 / 騎 is written — none whatsoever — that tells you when to say qí and when to say jì. Over time, people who are not learned in these matters tend to forget the less used variant readings and just collapse them into the most frequent one, so jì gradually has been getting eclipsed by qí. Many students, even from Sinophone countries, are unaware of the jì reading for nounal usages of 骑 / 騎.

    This is just one example of how variant pronunciations of characters gradually go by the wayside, but there are thousands of examples of this process that could be given. What's interesting is that the PRC government actually encourages this atrophication of variant readings, as in this case they want to "standardize" qí and jì as qí alone. For me, there is a meaningful distinction between qí as a verb and jì as a noun, and in Taiwan — where I learned the bulk of my Sinitic — the educational authorities pursue policies that tend to preserve the variant readings of characters.

    骑 / 騎 only has the two readings of qí and jì to worry about, but some characters have many more. It's not too rare to encounter characters that have four, five, or six variant readings, and I once encountered a character that had eleven variant readings (often some of the variants are strictly for local place names).

    I should also note that, while qí ("ride [a horse]", etc.) and jì ("mounted rider", etc.) are close cognates that derive from the same etymological root, other characters with multiple readings and meanings actually mask completely different morphemes or etymons for which they have been borrowed to write (see below for an example).

    Before going further, I should mention that characters with multiple readings are called pòyīn zì 破音字 ("character with variant pronunciations for different meanings; homograph") or duōyīn zì 多音字 ("polyphonic characters").

    Another very common character with multiple readings is 乐 / 樂, which may be pronounced lè, yuè, yào, lào, luò, and liáo. I will not go into as much detail with 乐 / 樂 as I did for 骑 / 騎, but will simply give the most basic meanings of the different pronunciations (most of the individual pronunciations also have a variety of meanings and usages of their own):

    lè — "happy; joy; pleasure"

    yào — "to be fond of; to enjoy; to appreciate"

    yuè — "music"

    lào — used in place names

    luò — used in compounds

    liáo — “to treat; to cure; to heal; therapy; treatment”

    It's evident that the first and second pronunciations and meanings are cognate (stative / adjectival verb and transitive verb), but most native speakers I encounter aren't familiar with the second pronunciation. In fact, the pronunciation of 乐 / 樂 as yào ("like; take delight in") seems to have dropped out of circulation on the Mainland nearly altogether (I polled about 40 of my Mainland graduate students, and only two of them knew it), though I learned it in Taiwan during the early 70s, and I think it is still alive there. The multiple pronunciations and meanings of 乐 / 樂 lead to much mischief in the reading of Chinese texts, especially Literary Sinitic texts where there is a much higher percentage of monosyllabic words than in modern Sinitic languages.

    Such are the delights / pleasures / joys ( lèqù 乐趣 / 樂趣 ["happy inclinations"]) of reading Sinitic.

    Once again, TIC, thanks for your intelligent questions. This is how we make progress in learning. Often it's good for the teacher — like Socrates — to ask questions and raise issues too, which is a method I frequently use in the classroom and in what I write. We should not pretend to know everything. Every day I learn something inestimably precious from my students. In this case, your patient, honest questions have helped me to see more clearly the issues surrounding the problem of variant readings and meanings of Sinographs.

  19. John Swindle said,

    March 1, 2019 @ 2:11 pm

    @TIC: Behind all of this lies the fundamental nature of the Chinese writing system: it's a vast, ultimately open-ended syllabary in which each character is conceived as having a certain meaning or range of meanings. The same syllable will be written differently if it's considered to have a different meaning. In this scheme of things a character, representing a syllable, will have only one pronunciation. But that relationship can slip. What's a totalitarian government to do?

  20. Huang Yu-hsiang said,

    March 1, 2019 @ 2:22 pm

    The “genteel” (literary) readings are just doublets. In Romance languages it would be equivalent to orally inherited words vs. modern words directly borrowed from Latin, e.g. direct (Latinate) vs. droit (inherited orally) in French or Spanish hierro (iron) vs. ferro- (as in ferrocarril). In Russian, such pairs also exist between modern Russian and Church Slavonic — город/град, for example. Standardization of Mandarin has eliminated many of these, but the practice of colloquial/literary pronunciation is alive and well in languages such as Southern Min and certain varieties of Wu. The Japanese kan-on, go-on, and tōsō-on readings can be compared to English borrowing from Norman French vs. modern French vs. Latin. Elimination of these pairs accounts for at least some of the changes. I hope this clears up some of the confusion.

    As far as reading written Mandarin aloud in Cantonese, Scottish English vs. Scots might not be the right comparison. Maybe it’s better to compare Cantonese-accented Mandarin to Scottish English (mutually intelligible), Mandarin-read-in-Cantonese to, say, a Scots Wikipædia article (yes, it exists) which due to the written format is similar to English and thus comprehensible but may not be understood when read aloud, and Cantonese to Scottish Gaelic (Cantonese is much more closely related to Mandarin than Scottish Gaelic to English but the point is they’re not mutually intelligible). The situations are still very different, but this is the best I can do with the English-language analogy.

  21. Chris Button said,

    March 1, 2019 @ 3:27 pm

    As one example of this, when the reading of 乘, previously chéng and shèng, was unified to chéng, the word for "Mahayana" changed overnight from dàshèng to dàchéng.

    Not making any justifications for the decision, but just in terms of the logic applied, I suppose it was because the "shèng" reading (originally derived from "chéng" at least in their respective Old Chinese forms) referring to the object that one rides (i.e. a chariot) was no longer in use in that sense.

  22. Jonathan Smith said,

    March 1, 2019 @ 5:37 pm

    Much of this is in origin popular "spelling pronunciation" and only later authoritarian fiat (e.g., 荨, 乘) — kind of like simplified characters that originate in "俗字".

  23. Su-Chong Lim said,

    March 1, 2019 @ 5:42 pm

    It seems to me that in this discussion, of the various people using examples of Scottish speech better to help discern and dissect speaking patterns in the greater Sinosphere, not all are using categorization of various forms of Scottish speech in the same way.

    In my usage, Scottish pronunciation of Standard English is how I would describe a person habitually using Scottish accent in daily speech reading from a script that would not be perceived to be irregular in any way to an English reader. It is only the phonic treatment of the English words that is changed from that of speakers from England (even though there would be a wide range of pronunciations in that category too!).

    "Scottish English” could mean different thing to different people. One meaning might be the Scottish dialect used by Robbie Burns (should I be writing “Robert”?) in his poetry, and one could rightly describe it as a distinct language, “Scottish”.

    The reference to Cantonese being equivalent to Scottish Gaelic in the Mandarin-Cantonese incomprehensibility conundrum is, in my opinion, misplaced. Scottish Gaelic is not closely related to “Scottish” (“Scottish English”) at all, being a Celtic language, whereas English and Scottish are Germanic languages.

  24. TIC said,

    March 1, 2019 @ 6:22 pm

    Many thanks to all who have contributed to my better (but still far from complete) understanding of the issues at play here… Especially, of course, to Dr. Mair… Not only for the very patient and thorough explanations… But also for the kind and reassuring words… I hesitated, out of profound respect for this forum and for those who frequent it, to share my confusion on this topic… I was reluctant to take the conversation backward, rather than to allow it to move forward among those who, unlike me, have significant knowledge and even expertise in the field… So I was extremely gratified by Dr. Mair's endorsement of my entry-level questions in the midst of this advanced-level course… Much appreciated…

  25. AntC said,

    March 1, 2019 @ 7:04 pm

    At the end of the 19th century, however, the Chinese discovered that they were outclassed by the technologically advanced (and rapacious) Western powers, giving rise to a national crisis. Among other things they decided they had to completely modernise their language.

    Did it occur to no-one that the 'outclassed' was because they had an ossified, corrupt system of government that (for example) squandered money intended for defence on a granite steamboat in the Empress's lake? Well, it did eventually: the Empire was overthrown; only to be replaced (after much turmoil) by an ossified, corrupt system of government.

    If language has much to do with it (which I doubt); did it occur to noone that Hong Kong modernised by leaving Cantonese in a state of 'benevolent neglect'?

    Did it occur to noone that none of those advanced Western powers have a writing system that's so burdensome on education and enterprise as Chinese characters; that even the stultifying schooling in the West in the C19th was not so inimical to learning as rote copying out of graphemes?

  26. Eidolon said,

    March 1, 2019 @ 8:37 pm


    On the contrary, it occurred to many people in China that all of those elements were at fault. The first target was the government – the Qing dynasty fell, to be replaced by a "Western style" Republic and then a "Soviet style" People's Republic; the second target was the language and writing system – vernacular Mandarin was introduced, alphabetization proposed, and eventually realized in the form of pinyin; the third target was culture – the reorganization of society from the ground up, first attempted by the Nationalists, but eventually enacted by the Communists via their massive land reforms, elimination of "feudal practices", and institution of peasant communes.

    The list goes on, and in this process, language reform was and is only one part of the transformation. No one ever thought that language was solely at fault, and relative to other measures adopted, it was one of the less affected. Culture was affected more: traditional Chinese institutions like the Confucian scholar official, legal polygamy, and foot binding have been all but eliminated, for example. By contrast, while Literary Chinese is basically dead, the Chinese characters still survive, as do almost all varieties of spoken Chinese.

    But identifying a problem and being able to implement a solution for it are two different concepts. What's happened to China since the 19th century is essentially a lesson in the limits of political reform. The revolutionaries lived long enough to see themselves become the reactionaries, and especially with the rise of Xi Jinping, political and cultural conservatism are coming back in force. The Communists once denounced Confucius as a "feudal evil," yet today they call their foreign influence centers, Confucius Institutes. China is returning to its old ways in many respects; but it's not like people didn't try to change them.

  27. Bathrobe said,

    March 1, 2019 @ 9:51 pm

    What Eidolon said.

    Language modernisation was pioneered in Japan with 言文一致 genbun-itchi, 'unification of the written and spoken languages', creating a new 国語 kokugo 'national language'. This became a model for China.

    The old linguistic order in both countries was tolerant of, perhaps you could say expected, linguistic diversity. The scholars and the elite used one language, the plebs another. The ability to read and write the prestige written language tended to be concentrated in the educated classes — of course, this simplifies the situation as there was actually quite a lot of literacy, especially in the vernacular, among ordinary people. The point, though, is that there were multiple literacies and no single unified language for the whole country. This fitted in fine with the older political structure but came to be seen as a liability if China was to be strengthened in the face of Western imperialism.

    Modernisers looked at the Western powers and saw how they practised universal education in a single standard language, ensuring national strength, unity, and an educated citizenry in basic areas like industry and science. They decided that they wanted the same for their own country. A single standard national language, a language that did not require an education in the classics but could be taught to everyone, a language that eliminated the barriers between different regions and different classes. And, of course, an updated vocabulary that eliminated the flowery and 'useless' language of the old order, including the stereotyped 八股文 bāgǔwén or 'eight-legged essay' of the old scholarly exams, and was adequate for representing the scientific concepts of the West.

    The consensus that gradually emerged — and it was contested every step of the way — is still with us today. A standard based on the vernacular, a 国语 guóyǔ 'national language' for the whole country (now renamed 普通话 pǔtōnghuà 'common language' on the Mainland), a drastic simplification of the characters to facilitate universal literacy, and a new style and vocabulary that owe much to Japanese and Western languages.

    Language modernisation thus had a strong political aspect and played a not-insignificant part in the move to modernise China.

  28. TIC said,

    March 2, 2019 @ 8:11 am

    After giving a lot of thought to everything said above, I've come to realize how naïve it was of me to ask for an English-language analogy for this Sinitic phenomenon… As Bathrobe suggested, a suitable and comprehensive parallel is impossible because "the languages, writing systems, and situations are so different"… I get that… And I also get that (unless/until I decide to put some concerted effort and study into it) I might never gain even a decent understanding of how factors as diverse as visual representations, ideas/concepts, syllables, sounds, pronunciations, etc. — compounded by the history and evolution of each of those — all seem to all factor into, and to interplay, in Sinitic writing, reading and speaking… Honestly, at this point, I think I lack even the most basic understanding of it all…

    I get that a good English-language analogy for the *mechanism(s)* of such a change is all but impossible… And I get that even a good — and comprehensive — analogy for the *effect* of such a change is equally impossible…

    But am I right in thinking that just *one* aspect of the effect is somewhat analogous to a decree that "there are too many words for (physically) 'small' in the English language, and it's inefficient and otherwise problematic for people to have to learn and remember all of them, so henceforth only 'small' will be used for this concept"?… And, as a result, all of the distinctions and subtleties and nuances and connotations (and joys!) of banished synonyms such as minute, teensy-weensy, microscopic, wee, Lilliputian, petite, runty, etc. are to be jettisoned and forgotten… Does this analogy even come close to reflecting one of the (seemingly problematic and objectionable) effects of such mandates?…

  29. Chris Button said,

    March 2, 2019 @ 8:58 am

    I suppose a kind of analogy in English for things like 騎 qí / jì and 乘 chéng / shèng where the difference originally stems from the presence of a final -s, may be found in English pairs like "project", "conduct", "permit" etc where the stress placement dictates if they are verbs or nouns.

  30. Bathrobe said,

    March 2, 2019 @ 11:03 am


    Actually, Chris Button has given an interesting example.

    It would be a little like English-language governments legislating that words like "project", "conduct", and "permit" were no longer to have dual pronunciations. The two pronunciations PRO-ject and pro-JECT would henceforth be unified as pro-JECT. This will save people from having to memorise two different pronunciations for the same word, even though the nuances are slightly different.

    Or perhaps a decree that 'vineyard' is no longer to be pronounced 'vinyard' since that puts a load on people's memories. Henceforth it must be pronounced 'vine yard', so that 'vine' is pronounced the same wherever it occurs (no random, confusing variation).

    Closer to the Chinese situation might be a ruling that '5' must no longer have the two readings 'FIVE' and 'FIF'. Henceforth it must be uniformly read as 'FIVE', so FIVE, FIVETH, FIVETEENTH, FIVETIETH…

    Nothing can fully capture the Chinese situation but these are possible analogies.

    It's not like outlawing words like 'teensy-weensy', 'microscopic', etc., which not even the Chinese government would attempt.

  31. Victor Mair said,

    March 2, 2019 @ 11:46 am

    We can be grateful to Eidolon and Bathrobe for their learned lectures on the history of language reform and language modernization during the last century and a half. Whenever we are confused about why Chinese intellectuals and governments felt compelled to propose radical changes in the way their compatriots were to write and speak, we can come back to these two comments for the historical, political, social, and cultural context:

  32. TIC said,

    March 3, 2019 @ 5:39 am

    Thanks, Chris Button and Bathrobe, for the helpful analogy to eliminating the distinct pronunciations of the noun and verb forms of words like 'project', 'conduct' and 'permit'… Dr. Mair's explanation about qi and ji (please forgive the imprecision of my keying) led me to momentarily consider that potential analogy… But I'd discarded it as too simplistic… I now get that it's decent as a partial/near analogy…

    Bathrobe, I continue to ponder your earlier point that "the process of standardising readings amounts to putting the written language ahead of the spoken … there is some kind of awareness in the West that 'x word is spelt a certain way' [whereas] in Chinese the standard locution is 'x character is read this way' [as if the] character is regarded as 'standard' and the pronunciation as secondary"…. My selective cut'n'paste paraphrase of what you wrote might be off-base, of course… And, if so, it might betray my confusion about whether you're suggesting (and, if so, my ambivalence about whether it jives with my way of thinking) that the opposite is true in English — that, in English, a word's spelling is regarded as 'standard' and the pronunciation as secondary…

    I'm hoping that someone might take a stab at addressing, if not answering, Randy F MacDonald's question (above): How does Arabic deal with the representation of its own internal linguistic diversity?… I suspect that at least part of the answer is that there is no analogous "governing" body (national or otherwise) that seeks to steer the evolution of the Arabic language(s)… But I wonder if there might be any near/partial analogies that might be illustrative…

    Dr. Mair, my (so far) endless rereading of this entire thread always takes me back to your opening line — "I think about the problem of the Sinophone every day"… Is the specific problem to which you were referring something along the lines of the view/fact that the concept of "Sinophone" is far more complex and problematic than, as examples, the concepts of "Anglophone" or "Francophone" — that "Sinophone" is more analogous to the (nonexistent?) concept of "Romanceophone"?…

  33. Victor Mair said,

    March 3, 2019 @ 6:57 am

    Is the specific problem to which you were referring something along the lines of the view/fact that the concept of "Sinophone" is far more complex and problematic than, as examples, the concepts of "Anglophone" or "Francophone" — that "Sinophone" is more analogous to the (nonexistent?) concept of "Romanceophone"?…

    Once again, TIC, you demonstrate your brilliantly productive and insightful inquisitiveness. The problem of the Sinophone is all that and more, including that of the unique (in the modern world) script.

    Within a day or two, I hope to have answers for Randy F MacDonald and you about how Arabic deals with "the representation of its own internal linguistic diversity".

  34. TIC said,

    March 3, 2019 @ 9:32 am

    Thanks again, Doc, for your (perhaps overly) kind words… But you need to knock it off… At best, you'll give me a swollen head… Worse, you might encourage me to too frequently interrupt expert/advanced-level discussions with layman's/entry-level questions… At worst, you might force me to reconsider my long-held and oft-stated view that "good questions are a dime a dozen — it's only good *answers* that are rare and valuable!"…

  35. Bathrobe said,

    March 3, 2019 @ 7:05 pm

    in English, a word's spelling is regarded as 'standard' and the pronunciation as secondary

    I was unclear. I believe that in English there is a general tendency to take the pronunciation as standard and regard spelling as somehow dancing around it.

    But there is the phenomenon of spelling pronunciations, whereby people change their pronunciation because they conclude (often from ignorance) that the spelling is correct and their pronunciation is wrong.

    But you gave me pause for thought. For the Chinese, the written word has long been regarded as almost sacred. But perhaps the very lack of a clear connection between characters and their pronunciation has led to an easier dissociation of written form and pronunciation, giving the characters greater flexibility to be pronounced in accordance with perceived needs. Different pronunciations in different dialects (topolects) is part of this, as is the move to abolish certain 'readings' in Mandarin. Of course, the greatest example of all is the use of Chinese characters to write a totally different language like Japanese.

    By contrast, spelling often drags English backwards as new generations fail to learn that (for instance) 'forecastle' is pronounced 'focsle or 'gunwhale' as 'gunnel', 'forehead' as 'forrid', or even 'often' as 'offen', resulting in the entrenching of spelling pronunciations.

  36. Philip Taylor said,

    March 4, 2019 @ 7:49 am

    I don't have access to the LPD in the office, but I have always believed that sounding the "t" in "often" is optional, and the OED supports me in this :
    Brit. /ˈɒf(ə)n/, /ˈɒft(ə)n/. But your nautical examples omit at least one classic : "boatswain" -> /ˈbəʊs(ə)n/.

  37. ~flow said,

    March 4, 2019 @ 9:46 am

    @Bathrobe it would seem to me that 'forecastle' is not quite so clear-cut example. The American Heritage Dictionary states:

    "forecastle (fōk´səl, fôr´kăs′əl) also focsle (fōk´səl)",

    IOW they allow, on the one hand, both the more conservative and the 'simplified' spelling ('focsle', which, in contrast to *'forrid', seems to be a thing).

    On the other hand they do not rule out the "fore-kassel" reading either. Interestingly, their 'ō' denotes the vowel of 'toe', whereas I would've guessed 'focsle' should have the short 'lock' vowel (i.e. 'focksl').

    I'm not sure whether I, as a second-language user, really like spellings a la 'forecastle', 'forehead', 'forfeit' and so on. English already has many, many words that have been contracted in both speech and writing, sometimes resulting in duplicates like 'phantasy' and 'fancy'. I guess in case I cast my vote in favor of spelling [foeksl] as 'forecastle' I must then also subscribe the motion to abolish 'fancy' (so both 'forecastle' and 'phantasy' will have a full and a contracted reading).

    It has been said above that the two ways to pronounce 'read' do not constitute a problem. I cannot second that. It is precisely in short phrases that a disambiguation would sometimes be welcome, and frankly, "I red a book", "I read a book" looks totally acceptable to me. In the same vein, cleanly separating "(to) lead" and "lead (the metal)" as "lead" and "led" would be most helpful. Curiously, people insist on "read" for [red] but the 100% analogous "lead" is already officially written the sane way, as "led".

  38. TIC said,

    March 5, 2019 @ 7:43 am

    Thanks very much, Bathrobe, for the clarification and further explanation… I find it ironic that your point about Chinese — that the character (analogous, in this respect, to the spelling in English) is primary and the pronunciation is secondary — is a good bit more understandable and readily acceptable to me than your view on the primacy of pronunciation over spelling in English… There's *so* much that I don't fully understand about Chinese (including the whole Are-Chinese-characters-ideograms-or-are-they-not? thing)… But one thing I do get is that the characters in no way represent/reflect pronunciation…

    I've been pondering, though, your stated belief that "in English there is a general tendency to take the pronunciation as standard and regard spelling as somehow dancing around it" for more than a day now… I'm leaning, at least at the moment, toward the view that my innate conception/belief is the opposite — that, for the most part, spelling takes primacy over pronunciation… Although it certainly isn't dispositive, the fact that common/acceptable pronunciations vary and evolve much more than do common/variant spellings is perhaps the chief factor that inclines me toward a spelling-is-primary view… I could, of course, be misunderstanding and/or mis-extending your point… Clearly, there are myriad factors at play here… Note that AmE is my native language and, to my profound shame, BrE is the closest thing I've got to a second language…

    – I'm extremely fortunate to be an excellent speller… I rarely have to give more than a moment's thought to the correct spelling of a word… But, to be honest, I'll bet that (up until the spelling was brought to my attention in this thread) it's quite possible that, in a spelling bee, I'd misspell 'vineyard' without the 'e'…

    – The word 'forecastle', as far as I know, is all but entirely frozen in time… I'm not aware of 'forecastle' (which in my head is pronounced like 'focsle') being used anywhere these days except in historical contexts… And, if I've ever encountered 'focsle' in writing, I've probably thought of it not as a standardized variant spelling but more like an eye-dialect thing… To a lesser degree, it seems to me, other nautical/maritime words such as 'gunwale', 'coxswain' and 'boatswain' have a bit of a frozen-in-time aspect, too… Of the three, only 'boatswain' has evolved a modernized/eye-dialect spelling, 'bosun', that I think of as having over time become almost as common/standard… (It's perhaps not insignificant that my go-to mental pronunciations of all of those seafaring words have a bit of a pirate's or swabbie's brogue to them!)…

    – I'm wondering whether 'forrid' is, in modern times, a common, BrE pronunciation of 'forehead'… I don't recall ever having heard it in AmE (except in nursery-rhyme with 'horrid')…

    – Here are a few pertinent pairs which, for whatever it's worth, I'll list in ascending order of how much I think of the latter (eye-dialect) spellings as having over time become almost as common/standard as the former — clapboard/clabberd; victuals/vittles; breeches/britches…

    – 'Often' is an interesting one… In casual/natural speech, I invariably ignore (elide?) the 't'… But, in some careful/formal-speech contexts, I might ("rightly" or "wrongly") pronounce the 't'… It's a lot like my pronunciation(s) of "either" — with 'eee-ther' being a casual/natural pronunciation and 'eye-ther' a more careful/formal one… Like the articles 'a' and 'the', intentional stress/emphasis within an utterance might also lead me (all-but-unconsciously) to go with the more "proper" pronunciations…

    – 'February' and 'Wednesday' have something in common for me… Unlike most folks, it seems to me, my routine pronunciations tend to reflect the spellings — i.e., the first 'r' in the former and the 'n' after, not before, the 'd' in the latter… And, in both cases, my go-to pronunciations tend to consistently — only very lightly (i.e., far from exaggeratedly) — acknowledge those oft-disregarded aspects of the spellings…

    – Naturally (because I was born and have spent all of my life almost midway between the two) my speech tends to be very, and even embarrassingly, dialectal — mid-Atlantic, with both New York and Philadelphia influences… And I try hard not to be too prescriptivist/judgmental with regard to other folks' dialects and pronunciations… But, try as I might, I just can't avoid looking down on 'liberry'-for-'library' and 'nuculer'-for-'nuclear' pronunciations…

    – I've gone on waaay too long already, so I'll forgo any opining on the whole BrE 'alumin(i)um' thing…

    That's it… Sorry for droning on… And for veering off topic here and there… But, for whatever it's worth, contemplation of each of those things factor into (if not support) my personal view that spelling takes primacy over pronunciation… But, as with most things, I remain open to — and welcome input that might lead me to — being persuaded toward a different view…

  39. Ellen K. said,

    March 5, 2019 @ 3:33 pm

    I wonder if non-rhotic speakers are more likely to see the "forrid" pronunciation of forehead, versus pronouncing it as spelled, as different. In my American English, the first syllable is the same either way (fore/for/four), with the second syllable in "forrid" reduced by dropping the h and getting rid of the secondary stress (and thus a less distinctly differentiated vowel, but not necessarily a different vowel). For a rhotic speaker, you've got an R vs. and H in the middle. Much more dramatic difference than having an H or not.

  40. Bathrobe said,

    March 5, 2019 @ 9:18 pm


    You may be right about the tendency of English speakers to privilege the spelling over the pronunciation. "Wednesday" is unsually pronounced "Wensday", no "d". Pronouncing it with a "d" is a correction back towards the spelling. That is: "Hey, why do people say it 'Wensday'? They must be wrong because that's not how it's spelt; I'm putting the 'd' back in". That is definitely privileging the spelling over the pronunciation.

    But there are many other cases where no one in their right mind would privilege the spelling. "Debt", or instance. How many people say "Hey, saying it as "det" is wrong! I'm putting the "b" back in!" (Actually, it was never there.) Or "love". "What's with this 'luv' thing? It should obviously be pronounced like 'loave', and that's how I'm going to pronounce it! And the only difference from 'loathe' will be the consonant!" Or "give". "Damn, what's with the 'giv' pronunciation? It ought to be pronounced to rhyme with 'jive'! I'm restoring the correct pronunciation!"

    These are ridiculous examples, but they demonstrate that even if you are committed to pronouncing words as they are spelt, there are limits.

    @ Ellen K

    I think "forrid" is dying out. You are right that there is a more dramatic difference in non-rhotic dialects than rhotic, but there is still a difference, and reinserting the "h" is still a spelling pronunciation.

  41. Ellen K. said,

    March 6, 2019 @ 1:00 pm

    @Bathrobe, I'm not convinced. Seems to me quite likely that, in American English, someone might hear /ˈfɔrəd/, and, from sound and meaning, understand that to be a reduction of /ˈfɔɹˌhɛd/, and thus pronounce it /ˈfɔɹˌhɛd/ without ever having seen the word in writing. No reason to conclude it's anything other than knowing the two morphemes in the word and pronouncing them.

  42. dainichi said,

    March 7, 2019 @ 1:44 am

    @Ellen K, I understand and agree with your point. Unfortunately it's probably hard to distinguish the "morpheme reconstruction" effect from the effect from hearing others use the spelling pronunciation.

    I'm curious about BrE, are there any dialects/registers where linking-r + h-dropping would cause 'forehead' to be pronounced 'forrid' without any phoneme replacement needed?

  43. TIC said,

    March 9, 2019 @ 7:32 am

    I've parsed and pondered it for a couple of days now, Bathrobe, and I remain confused about your comments (to Ellen) about pronunciations of the word 'forehead'… You said:
    I think "forrid" is dying out. You are right that there is a more dramatic difference in non-rhotic dialects than rhotic, but there is still a difference, and reinserting the "h" is still a spelling pronunciation

    Are you saying that, in BrE, the 'forrid' pronunciation seems to be giving way to 'forehead'?… Or to some other pronunciation such as, perhaps, 'for-ed'?…

    My confusion stems in part from the fact that, as I said earlier, I'm unclear on the common pronunciation(s) of the word in modern BrE…

    And I really have to challenge, or at least question, your assertion that — at least in AmE — 'forehead' is a 'spelling pronunciation' (i.e., at odds with standard/traditional pronunciations)… I don't think I've *ever* heard any pronunciation other than the full-blown 'forehead' in AmE (except, of course, in recitations of the old nursery rhyme in which it's paired with 'horrid'…

    So, are you suggesting that the *standard* AmE pronunciation is in fact a 'spelling pronunciation'?… In other words, are you perhaps suggesting that 'forrid' (or 'for-ed') had become standard in BrE by the time that AmE had begun to split off from BrE and that the AmE pronunciation reverted backward (to a more spelling-adherent version) sometime later?… If not, what am I missing here?… Thanks in advance…

  44. Bathrobe said,

    March 10, 2019 @ 5:45 pm

    What I meant is that "forrid" is giving way to "forehead". I've never heard "four'ed".

    I would be on shaky ground if I asserted categorically that "forehead" was a spelling pronunciation in the US because I don't know the detailed history (including dialect or regional usages in either Britain or the US), but I personally think that it probably is. Just as "waistcoat" is rather than the old "weskit".

    However, Ellen K may have a point. Namely that, rather than a spelling pronunciation, speakers may have deduced that the h had been transparently dropped and was accordingly restored by the listener. Was h-dropping common in the US?

  45. Bathrobe said,

    March 10, 2019 @ 5:55 pm

    "speakers may have deduced that the h had been transparently dropped and was accordingly restored by the listener"

    Garbled. People hearing 'fore'ead' may have automatically interpreted it as 'forehead', and restored the h purely on etymological rather than spelling grounds, as Ellen K said.

    If there was ever a stage where Americans said 'fore'ead' (which, I repeat, I am by no means sure of), it could conceivably have been a very long time ago. Incidentally, I don't think the 'split' between BrE and AmE was as neat as you appear to suggest.

    Standards come and standards go. Spelling pronunciations often become standard; there is nothing strange about it.

  46. Bathrobe said,

    March 10, 2019 @ 6:21 pm

    @TIC Incidentally, I'm not sure why Merriam-Webster give two pronunciations, one pronounced /fȯr-ˌhed/, the other /fär-əd, ˈfȯr- also ˈfȯr-ˌed/, which does not agree with your experience.

  47. TIC said,

    March 12, 2019 @ 6:06 am

    Thanks, Bathrobe (and others) for the responses…

    The pronunciation(s)-of-'forehead' issue, I've come to realize, is a particularly devilish one… It's unusually complicated, it seems to me, by a confluence of sometimes-competing influences: BrE and AmE, rhoticity and non-rhoticity, "spelling pronunciation" and "morpheme replacement", and, perhaps, by widespread (and early-in-life) exposure via Mother Goose to the forrid/horrid rhyme…

    I have, I believe, a pretty good ear for others' pronunciations… And I've got a slight update/correction to my earlier claim that I don't recall *ever* having heard any pronunciation other than the full-blown 'forhead' in AmE… It's occurred to me that I have in fact heard a pronunciation somewhat akin to 'farhead'… And, although I suspect that few folks are still monitoring following this thread, I'd love to see follow-up comments from others on variant pronunciations of 'forehead' in AmE…

    You're right, Bathrobe, that I significantly oversimplified things in referring to "the time [at which] AmE had begun to split off from BrE"… Let's just chalk it up to a(n all too rare) attempt at conciseness on my part!…

    Finally, you asked whether h-dropping was/is common in the US?… I haven't given it much thought, but I can't think of any examples of 'h'-dropping in AmE… But, on this point, too, I'd love to see input from others…

RSS feed for comments on this post