Tones and the alphabet

« previous post | next post »

The question of whether tones are added to alphabet words used in Sinitic languages arose in the discussion that followed this post:

"Papi Jiang: PRC internet sensation" (4/25/16)

One commenter remarked:

It's surely an odd (meaningless) name, not easy to say for a Chinese, which suggests that it wasn't particularly chosen to popularize in the first place. The choice of the homophone to her surname may have to do with 酱 being used as a cute contraction (informal, may come from Taiwan) for zhèyàng 这样.

To which another replied:

What's difficult to say about Papi酱? Pa, pi, and jiang are all chinese syllables?

This led the first commenter to revise their opinion:

True. I was reading pi in fourth tone, and another fourth tone after that makes it a little 拗口. To ammend: it is a little awkward to say for ME.

I was just thinking the other day that when an English word or acronym has entered Chinese (should I say Mandarin), we'd automatically attach a tone for each syllable. For example, DNA is dì ēn ēi, and CT is sēi tì.

My personal experience is that Chinese do not add tones to letters in a consistent fashion, and many do not add tones at all.  To check my own impression, I asked a number of students and colleagues.  Here's a sampling of the replies I received:

1. a native of Hunan

I usually won't add tones to the letters when I speak Mandarin. But when I speak the topolect of my hometown, I usually speak the letters like Chinese characters with tones, such as "大不留梯哦“ [for WTO] and "怕皮酱" [for Papi Jiang].

2. a teacher of Mandarin

First tone or neutral tone?

First tone, I think.

3. a native of Sichuan

Thank you for raising the question, because I never noticed this phenomenon. Yes, people do add tones when speaking these alphabetic words, Otherwise, it would sound really strange.
W(∨)T (-) O (-).
Pa(-)pi(light tone 轻声) 酱(﹨)

VHM:  there is a fallacy here — she claims that she pronounces "W" in the third tone, but that is impossible since "W" has three syllables; how can you spread a third tone over three syllables? how can you spread any tone over three syllables?

4. an American who is proficient in Mandarin

"A Grammar of Mandarin" by Jeroen Wiedenhof, p. 369-370 has a fascinating discussion on this topic, including a table with these pronunciations:

ēi bī xī dī yī áifu jī èichi āi/ài zhèi kèi áilu áimu ēn ōu pī keyōu ár áisi tī yōu wéi dàboyōu àikesī wài sáide/[ziː]

It says "The system is relatively stable, including the citation tones given. Other tones occur for Roman letters in fixed expressions …". One example given for "other tones" is bí chāo or bì chāo for B超 'medical sonography'. Also, "Mandarin speakers with good English competence may omit the tones given in the table … Intonation will then follow English usage, which often has the greatest fall in pitch on the last letter being spelled."

5. a Chinese language professor who is a native speaker of MSM

High tone?

6. a native Beijinger

I, like most Chinese people, would say WTO with T pronounced in the fourth tone and  O pronounced in the first tone. It sounds like "W ti4 ou1". I am not sure what Papi酱 is…but I would say pa1 pi 酱.

In abbreviations, if the last letter is O, U, A, N, C, H, V, it is usually pronounced with the first tone, otherwise, the fourth tone is used. Abbreviations are usually pronounced letter by letter if it is not too long.

7. an American who has virtually native fluency in Mandarin

When Chinese shopkeepers say "DVD" to me, somehow it always sounds something like

di1 wei1 di4

(where the "V" becomes "wei" of course).  But can't be sure.  When reciting the alphabet, it always sounds to me like the syllables are first tone, except where they need to render the letter with two syllables to fit the Chinese phonology.  Like "ai fu" for "F".  In that case, to me it always sounds like the first syllable is first tone and second is neutral.  So "F" to me sounds like:

ai1 fu (neutral)

Or "X":

ai1 ke si (both neutral).

Just hunches.

8. a native of Harbin

I do not add tones to the two words, and neither do many people.

===

From this testimony, we may conclude that there is no standard for adding tones to letters used in Sinitic languages, and that often tones are simply not added at all.  It would seem that the manner in which tones are added to letters in Sinitic speech varies in accordance with the habits and preferences of individuals.

On a related note, today on the train ride from Hualien to Taichung, the announcer read the names of the successive stations along the route in four languages:  Mandarin, Taiwanese, Japanese, and English.  Of course, when she read out the names of the stations in Mandarin and Taiwanese, her tones were correct for those languages.  Her Japanese didn't sound very good to me, though tones were not an issue there.  But the way she spoke the tones in English drove me nuts.

I do not recall her every syllable, but the overall pattern was clear.  Regardless of their original tones in Mandarin and Taiwanese, the announcer basically would pronounce the first syllable of all the names in the first tone, and the second syllable either in the third tone or in the fourth tone.  It was particularly annoying that she would do this repeatedly and arbitrarily for several names in a row.  Thus it went something like this:  1-4, 1-4, 1-4, 1-4, 1-3, 1-3, 1-3, 1-4, 1-4, 1-4, 1-4.  Huālián became Huāliàn (and she said her English 4th tones very deliberately and emphatically), which really grated on my ears, since I perceive 2nd (rising) and 4th (falling) tones as opposites.  In the same vein, Táizhōng became Tāizhòng, Zhānghuà became Zhānghuǎ, and so forth.

I have no idea what prompted her to do this.  It seems to me that Chinese announcers would be well advised either to retain the original MSM tones for names when speaking English, or, if that feels unnatural to them, then switch to all first tones.  Better yet, use neutral tones, though I must admit that English speakers often accent certain syllables of Chinese names in a manner that may be perceived as a fourth tone.  Still and all, it sounded horrible to me when the announcer repeatedly said "Huāliàn", with a very strong 4th tone on the second syllable.

[Thanks to Fangyi Cheng, David Moser, Jing Wen, Tom Bishop, Yixue Yang, Cynthia Ning, Zheng-sheng Zhang, and Rebecca Shuang Fu]



44 Comments

  1. Bathrobe said,

    April 27, 2016 @ 5:19 pm

    Wiedenhof's list is useful as a reference, but I've found it doesn't work nationwide. In Beijing that's how people pronounce the letters of the alphabet, but if you have to spell your name (your foreign name) over the phone to a southern speaker of Mandarin, it's an exercise in frustration to get them to understand these letter names. For instance, in my experience the áimu ēn distinction is meaningless to southern speakers. ár doesn't make much impression either, since southerners don't indulge in erhua. It's usually easier to resort to 'L for London" etc., or just send an SMS.

  2. Tom said,

    April 27, 2016 @ 6:44 pm

    I thought the letter B was almost never pronounced with a first tone, because it would then be understood as bī 屄 ("vagina").

  3. AntC said,

    April 27, 2016 @ 7:29 pm

    @VHM re 3. native of Sichuan
    she claims that she pronounces "W" in the third tone, but that is impossible since "W" has three syllables

    Is it possible that rather than saying "double-u" the pronunciation is "wei" or "wo" or "wu", or "w"+schwa or some such single neutral vowel? (Sorry I can't do the IPA.)

  4. A-gu said,

    April 28, 2016 @ 12:43 am

    I'm sure that letters carry tones in Taiwan, almost invariably.

    First tone seems to be the default option. You can hear this clearly in the formulation used when specifying letters when spelling things out:

    ABC的C ("The C in ABC")
    All three Roman letters are first tone here.

    One exception might be:
    XYZ的X ("The X in XYZ")
    Where the tones for X would be spread over 3 syllables [e]4 [khə]3 [sʐ]1
    Still, Y and Z would both still be first tone

    Or F, [ef]4 [u]3

    Also consider common compounds used locally, even if root is Taiwanese, but always written with the roman alphabet, like:
    A菜 (A in first tone)
    a common vegetable with a delicious juicy stem
    or A錢, A also first tone) (to steal money, normally via corruption)

    That that these are Taiwanese terms using Roman loan characters seems to bolster the case that the letters have tonal qualities in Sinitic languages.

  5. A-gu said,

    April 28, 2016 @ 12:52 am

    Also Victor, your recollection of tonal changes to "Anglicize" Taiwan place names is highly accurate.

  6. Victor Mair said,

    April 28, 2016 @ 1:12 am

    Late addition from a native Beijinger:

    I believe it varies in different cases–WTO is often pronounced as dábuliu-tì-ōu, whereas papi sounds rather normal, like pāpi.

    That said, it is very likely that people speaking different topolects (i.e. poor Mandarin) will pronounce them in unfathomable ways.

  7. Terry Hunt said,

    April 28, 2016 @ 9:15 am

    Speaking from a mostly monoglot English position: I have sometimes noticed live announcers in UK contexts (railway stations, supermarkets and the like) adopt similarly repetitive and unnatural "sing-song" tones, both in general sentences and in lists of places or products. Perhaps the same impulse was operating here.

  8. Bob Ladd said,

    April 28, 2016 @ 9:43 am

    Seems to me the most obvious explanation for the way the announcer anglicised the Taiwan place names is that the names were all being treated as weak-strong (i.e. greater stress on the second syllable) and that the announcer had very good English intonation. As far as I can see, weak-strong is absolutely the default stress pattern for Chinese two-syllable place names in English (Hong Kong, Beijing, Taiwan, Shanghai, etc. etc.), and in English citation form that is going to come out sounding like 1-4 (or maybe 2-4). In English list intonation the 1-3 sequence that the announcer also used is certainly possible, though by no means the only possibility. Either way, the choice of 1-4 and 1-3 strikes me as an indication of a good command of English, and it's only because VHM was listening "in Chinese" that it sounded weird and arbitrary to him. I'd be willing to bet that a monolingual English speaker would have thought it sounded pretty natural.

  9. Victor Mair said,

    April 28, 2016 @ 10:07 am

    @Bob Ladd

    Nice point!

  10. Bathrobe said,

    April 28, 2016 @ 3:53 pm

    For me, "Shanghai" and "Hong Kong" are equally natural as strong-weak.

  11. Bob Ladd said,

    April 28, 2016 @ 4:29 pm

    @Bathrobe: Agreed, a few of these names may shift – it's a long-term tendency in English for stress to move earlier (General American CIGarette; Southern American INsurance; British BROchure; etc. etc. etc.) Also, all the Chinese names under discussion are subject to "iambic reversal" (a.k.a. the thirteen men rule) when they're used attributively: in phrases like Hong Kong harbour or Shanghai dialect [sorry, VHM!] or Peking Duck or Hainan Island the first syllable is almost always going to feel more prominent than the second, and this presumably feeds the tendency to shift the stress in the citation form. But as a broad generalisation, I think it's accurate to say that most two-syllable Chinese place names are weak-strong in English.

  12. Michael Watts said,

    April 28, 2016 @ 5:45 pm

    she claims that she pronounces "W" in the third tone, but that is impossible since "W" has three syllables; how can you spread a third tone over three syllables? how can you spread any tone over three syllables?

    Well, it's trivial to spread the high tone over three syllables; that would just be three syllables all with high tone. I don't know how well that works in practice.

    In the specific case of third-tone W, it's still not a crazy idea, since W is three syllables and 3rd is a three-point (high-low-high, basically) tone. The overall tone sequence of 大不留, \\/, isn't so far off from a third-tone \/. It just depends on what you consider a "unit" for purposes of analysis.

  13. Victor Mair said,

    April 28, 2016 @ 10:33 pm

    From a native speaker of Cantonese and Mandarin:

    I personally do not add any tones to the letters…

  14. Michael Watts said,

    April 28, 2016 @ 11:13 pm

    I find it harder to understand the people saying they don't pronounce the letters with tones. Tone isn't a feature that can be absent from speech, unless it's whispered.

  15. Bob Ladd said,

    April 29, 2016 @ 1:55 am

    @Michael Watts: I wondered about that too. It might actually tell us something interesting about the nature of tone to have a clearer explanation from one of these native speakers about what they mean when they say they "don't add tones".

  16. Michael Watts said,

    April 29, 2016 @ 2:56 am

    It's probably worth reporting the conversation I just had:

    me: 路就读 lù 对吧 "路 is read lù [fourth tone], right?"
    me: 可以说完全没有声调的 lu? "can [you] say a completely toneless lu?"
    not me: 没有声调那不就是第一声吗?? "toneless – isn't that just first tone??"

    where "not me" is a shanghainese woman in her mid-30s. I don't know how widespread that view of what it means to be "toneless" is, but this has implications for the reports of "I don't use tones to read 'WTO'".

  17. Alon Lischinsky said,

    April 29, 2016 @ 5:27 am

    we may conclude that there is no standard for adding tones to letters used in Sinitic languages, and that often tones are simply not added at all

    Not really. We can conclude that speakers report inconsistent practices, but from these reports to the patterns in actual performance there's a vast gulf. Introspection is a very poor guide to one's linguistic practices.

  18. Neil Dolinger said,

    April 29, 2016 @ 5:33 am

    In my visits to factories in China, in some regions I hear factory managers who pronounce the letter 'p' something like "pxi", where the 'x' is the pinyin x (tip of tongue placed against the back of the bottom row of teeth). I have always wondered where that pronunciation came from, since it doesn't mimic any standard MSM syllable. I should add that these are people who generally do not otherwise speak English, and are using the letter in abbreviations for plastics, e.g. "pxi pxi" = "PP" = polypropylene, "pxi bi ti" = "PBT", etc. I don't think they do the same thing with the b's — 'b' becoming "bji".

  19. Victor Mair said,

    April 29, 2016 @ 8:18 am

    The people who say they don't use tones on letters are native speakers. Many of them have taught Chinese and are teachers of Chinese or Chinese linguists. They know what tones are. They are not ninnies.

  20. Victor Mair said,

    April 29, 2016 @ 8:20 am

    To say that a syllable has "no tone" is not the same thing as saying that it is a neutral tone.

  21. Michael Watts said,

    April 29, 2016 @ 8:29 am

    Does saying that a syllable has "no tone" have any meaning? How can vocalized speech avoid having tone?

  22. Victor Mair said,

    April 29, 2016 @ 8:32 am

    Does English have tones? Does German have tones? Russian?

  23. Matt_M said,

    April 29, 2016 @ 9:06 am

    Does English have tones? Does German have tones? Russian?

    I was wondering what pronouncing Roman letters with "no tone" could mean as well. Obviously English does not have phonemic tones — but each syllable an English speaker pronounces has some intonational contour attached to it. The same must be true of Chinese speakers pronouncing "B" or "J" etc. So I guess the question is: what kind of intonational contour do these syllables have? Is it high, low, rising, falling, or level? How does it differ from the intonational contour of the standard set of tones?

    Or do Chinese speakers import English intonational patterns when speaking them (which would be quite an impressive feat, in my opinion, although of course not impossible)?

  24. JS said,

    April 29, 2016 @ 9:46 am

    Chinese languages have phonemic tone… that's that, for now.

    Talking about whether or not Chinese speakers "add tones" to such terms is an odd way to put it. The question is to what degree, and for which speakers, such borrowings have been phonologically nativized. If they're nativized, the syllables carry tone; if not, they're pronounced with an English-style contour — no different from the situation with borrowings from English in general.

    Also, what Alon Lischinsky said — native speakers' reflections on their linguistic behavior are of dubious value; we need data. This is especially true in the special cases where non-nativized pronunciations still wind up phonotactically exceptional in certain ways: re: the interesting remarks above on "W", try listening here, or anywhere, really. I'll avoid coloring perceptions by offering a transcription…

  25. Michael Watts said,

    April 29, 2016 @ 6:00 pm

    Of course English has tones. Tones are not a feature of the English lexicon, but they are a feature of the English language, and in one notable case are required. As a native English speaker, it is still not possible for me to pronounce the three syllables of W without adding tones to them, but those tones will be determined not as a fact about "W" but by some subset of a number of factors, including but not limited to:

    – The syntactic role "W" plays in the sentence.

    – On a smaller scale, whether "W" is part of a particular subphrase of the sentence, such as "WTO".

    – My current mood.

    – Whether I am asking a (broadly construed) yes/no question. (It's not a coincidence that uptalk is so frequently indicated in text with question marks.)

    – My own attitude toward what I'm saying, for example if I'm reporting a quote or attributing a view to someone.

    – My attitude toward the person I'm speaking to, such as if I want to imply that they are stupid.

    All of those things are properly conveyed by tone when speaking English.

  26. Michael Watts said,

    April 29, 2016 @ 6:12 pm

    I mean, we just had a conversation about a Chinese train announcer altering the lexical tone of Chinese words when pronouncing them in English. It's not possible to preserve the Chinese tones of a Chinese word when using it in natural English, because English has its own requirements for the tone.

  27. Victor Mair said,

    April 29, 2016 @ 8:42 pm

    There's a stark difference between the function of phonemic tone and intonation, also between phonemic tone and accent.

    On phonemic tone, see the comment by JS above.

    http://www.phon.ox.ac.uk/jcoleman/PROSODY1.html

    "Languages such as English do not have phonemic tone, although they use intonation for functions such as emphasis and attitude."

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

    This post is about phonemic tone, not about intonation.

  28. JS said,

    April 29, 2016 @ 8:47 pm

    Real examples of phenomena in English resembling phonemic tone are few: borrowing pinyin numbers, uh1 (= ummm…), huh2 (= ?), huh4 (= ok I see) are different; aw3 (= cute baby) not the same as aw2 (= you gon' get it), etc. Word stress is of course a thing, as is intonation. Which is also a thing in Chinese; Chao Yuan-ren talked about "ripples on waves".

  29. JS said,

    April 29, 2016 @ 8:49 pm

    That is to say, what Prof. Mair said…

  30. Michael Watts said,

    April 29, 2016 @ 8:56 pm

    English contains one word, "I don't know" which features phonemic tone. Famously, it has no features at all except for the three-tone sequence.

    This post is about pronunciation; it is clear that english letters do not have lexical tones attached to them. But that doesn't make it any more possible to pronounce them without a tone.

    Saying "this person does not add tones to english letter names when reading them in Chinese" tells us nothing about "this person"'s pronunciation of the letter names (which was the question!); we already know that they are pronounced with some tone. I have asked if it has any meaning at all. Does it?

  31. JS said,

    April 29, 2016 @ 9:14 pm

    See Matt_M's comment for intonational contour (what you're talking about) vs. phonemic tone, where tone is a phoneme ("toneme" if you like) and thus can figure in minimal contrasts. The Mandarin borrowing of "T" is pronounced ti1, meaning it is a homophone of ti1 'kick' and contrasts with the words ti2 'lift', ti4 'replace', etc.; the contours it can take on in connected speech are delimited by membership in this phonemic category. Whereas in English the contour joined to "T" /ti/ can vary dramatically.

  32. Michael Watts said,

    April 29, 2016 @ 9:21 pm

    JS, I am aware of all of that (except the claim that the letter T is pronounced with a standard lexicalized tone in Mandarin; this seems dubious on the basis of the post). It doesn't address my question.

  33. JS said,

    April 29, 2016 @ 9:56 pm

    To me, "this person does not add tones" to WTO, etc., when pronouncing them in Chinese means that they "import English intonational patterns" or some approximation thereof rather than nativizing them to Chinese phonological norms, as Matt_M and then I said way upthread. Although I've begun to sense your question is largely rhetorical, if of inscrutable (to me) rhetorical intent.

  34. Michael Watts said,

    April 29, 2016 @ 11:03 pm

    I don't think that's likely for multiple reasons:

    – Chinese people generally do not know what is and isn't a valid English intonational pattern. I observed a news announcer once who had clearly studied that exact topic, because her (English) narration featured valid intonational contours — except that they were bizarrely unmoored from the sentences themselves, as if she'd lifted the entire, recognizably English, contour from a different sentence and applied to it the one she was actually saying. When someone whose job involves trying to produce natural-sounding English and who has clearly made a specific study of English intonation still doesn't understand what's appropriate, I find it unlikely that someone else of which neither of those things are true is able to use English intonation.

    Analogously, even Chinese speakers who have studied English for a long time and are quite conversational in it are often unable to end syllables with -s, instead adding on a full extra syllable such as 思 to carry the sound. Chinese intonational patterns are a part of the phonology just as constraints on what is and isn't a legal syllable are. For most people, it is not possible to produce any speech without nativizing the phonology.

    – As you have already stated (and I agree), a letter name pronounced in English may appear with any of a wide variety of intonational patterns, depending on the sentence and speaker. The citation form of an English letter name, as with all citation forms in English, uses a falling tone, which is perfectly normal Mandarin phonology. But we can conclude that the citation form is definitely not what's being referred to here.

    To me, saying "I don't add tones to these words" makes as much sense as the (very common) claim "I don't speak with an accent". We know that (a) a majority of people in the world are happy to make that second claim, and (b) every one of them is mistaken, and in fact the semantics of that claim are malformed. I'd like to know what it is that people mean when they say "I don't add tones", if anything, and where the analogy to "I don't speak with an accent" goes wrong. And, if someone does make the claim "I don't pronounce 'W' with any tone", are there any conclusions we could draw from that about their pronunciation of 'W'?

  35. Victor Mair said,

    April 30, 2016 @ 6:10 am

    You're simply not listening to what other well-informed commenters are saying.

  36. Michael Watts said,

    April 30, 2016 @ 7:56 am

    OK, enlighten me. Here's what I've read:

    My personal experience is that Chinese do not add tones to letters in a consistent fashion, and many do not add tones at all. (VHM)

    To say that a syllable has "no tone" is not the same thing as saying that it is a neutral tone. (VHM)

    I wondered about that ["what do people mean when they say they don't pronounce letters with tones?"] too. It might actually tell us something interesting about the nature of tone to have a clearer explanation from one of these native speakers about what they mean when they say they "don't add tones". (Bob Ladd)

    I was wondering what pronouncing Roman letters with "no tone" could mean as well. Obviously English does not have phonemic tones — but each syllable an English speaker pronounces has some intonational contour attached to it. The same must be true of Chinese speakers pronouncing "B" or "J" etc. So I guess the question is: what kind of intonational contour do these syllables have? Is it high, low, rising, falling, or level? How does it differ from the intonational contour of the standard set of tones?

    Or do Chinese speakers import English intonational patterns when speaking them (which would be quite an impressive feat, in my opinion, although of course not impossible)? (Matt_M)

    Talking about whether or not Chinese speakers "add tones" to such terms is an odd way to put it. The question is to what degree, and for which speakers, such borrowings have been phonologically nativized. If they're nativized, the syllables carry tone; if not, they're pronounced with an English-style contour — no different from the situation with borrowings from English in general.

    Also, what Alon Lischinsky said — native speakers' reflections on their linguistic behavior are of dubious value; we need data. (JS)

    To me, "this person does not add tones" to WTO, etc., when pronouncing them in Chinese means that they "import English intonational patterns" or some approximation thereof rather than nativizing them to Chinese phonological norms, as Matt_M and then I said way upthread. (JS)

    So, Bob Ladd, Matt_M, myself, and JS have all expressed confusion over what it would mean for a syllable to be pronounced with no tone. JS rejected the terminology completely in favor of something else. I have asked the same question that Matt_M asked: what is the intonational contour associated with these "no tone" syllables, and if it is an English contour, how did a Chinese person know how to produce it? For that, I got referred back to Matt_M's comment. Matt_M and I have both expressed surprise that a Chinese person would be able to produce an English intonational contour. JS has provided the assumption that "no tone" in fact means "English intonational contour", but JS also rejected the idea that it made sense to talk about whether a syllable was pronounced with "added tone". VHM, who has repeatedly endorsed the idea of syllables with "no tone", and distinguished it from syllables using clitic tone 轻声, has never provided any interpretation of what it means.

    I can see a few possibilities for interpreting the claim of "no tone":

    1. People claiming this are making a mistake. (Alon Lischinsky, JS, and myself have all mentioned this possibility.)
    2. "No tone" refers to first tone. (Only I have mentioned this; I find it a pretty surprising idea myself.)
    3. "No tone" refers to clitic tone, a tone determined by the surrounding syllables. (Specifically disavowed by VHM.)
    4. A syllable with "no tone" is pronounced with any arbitrary tone, chosen according to the speaker's whim at the time of utterance from the inventory of possible Chinese tones. (This more or less mirrors how most English intonational contours are chosen, although the actual intonational contour produced will not be an English contour.)
    5. A syllable with "no tone" is pronounced using the speaker's impression of what English-speaking people sound like. (This is still likely to be shoehorned into Chinese phonology when a Chinese person produces it, though.)
    6. A syllable with "no tone" is pronounced using a genuine English intonational contour. (How? Which one?)

    JS, and so far as I can see only JS, has suggested that "no tone" should be interpreted as one of (5) or (6); I don't know which. Is this correct?

    Obviously, it is not possible to add "phonemic tone" to a syllable of an utterance, because phonemic tone is just a fact about a language's phonology and vocabulary. That applies equally to the people saying "I pronounce the D of DNA with first tone", though. It doesn't make any sense as a response to the question "when someone vocalizes 'WTO', what does that sound like?" In analyzing a sample of produced speech, the phenomenon of phonemic tone does not exist. Intonational contours do exist at that level; some classification system can then be used to assign contours to various equivalence classes ("phonemes"). I do not understand the relevance of the distinction between "phonemic tone" and "intonational contour" as VHM and JS are trying to point it out to me.

    There was an LL post long ago concerning American news reporters saying the African name "Gbagbo". As they were unable to start a syllable with "gb", they had to use other strategies. One was to insert an epenthetic vowel between the initial g and its b, producing something like "guh-bagbo". Vowels are phonemic in English, and the presence or absence of a vowel is likewise phonemic (consider the distinction between "fraction" and "for action"). But I don't see that it makes sense to ask whether the newscaster saying "guh-bagbo" was inserting a "phonemic vowel" or merely a "vowel"?

  37. Victor Mair said,

    April 30, 2016 @ 10:20 am

    You're following the same unproductive pattern of behavior that you have displayed in previous Language Log discussions — repeating assertions of your own devising without taking seriously evidence adduced by others. You keep insisting that English has tones and that letters of the alphabet when introduced in Chinese absolutely must have tones, even when many native speakers — some of whom are language teachers and linguists — state that they do not always have Sinitic tones.

    You seem to have missed the testimony of a very perceptive Hunanese scholar who even made a clear distinction between adding tones to letters when he speaks his Hunanese topolect and not adding them when he speaks Mandarin. This individual is thoughtful and intelligent; he is able to make such a clear distinction between when he adds tones to letters and when he doesn't. Why do you presume to know more about his use and nonuse of tones than he does himself?

    The tones we are talking about in this post are limited to those that belong to a specific repertoire for a given tonal language — 4 plus neutral tone for Mandarin. The "tones" (i.e., intonations) you're talking about could be all over the place and are essentially infinite in number. You're arguing with yourself. You are not listening to what this post is all about.

    If you cannot contribute anything new and useful to the conversation, I must ask you to cease and desist.

  38. Michael Watts said,

    April 30, 2016 @ 11:11 pm

    This individual is thoughtful and intelligent; he is able to make such a clear distinction between when he adds tones to letters and when he doesn't. Why do you presume to know more about his use and nonuse of tones than he does himself?

    Why do you presume to know more about a Sichuanese woman's use and nonuse of tones than she does?

    I discount the reports of "I don't use tone in pronouncing foreign letters" for the same reason I give no weight to the opinion of a native Japanese speaker saying "the concept of 未来 in Japanese is more ineffable than the concept of 'the future' in English":

    We can conclude that speakers report inconsistent practices, but from these reports to the patterns in actual performance there's a vast gulf. Introspection is a very poor guide to one's linguistic practices.

    Also, what Alon Lischinsky said — native speakers' reflections on their linguistic behavior are of dubious value; we need data.

    I've witnessed an English teacher instructing a class of Chinese students that "affect" and "effect" are pronounced differently. Being a teacher was not somehow protective against this mistake. Introspection is not a reliable tool in linguistic analysis — the fact that someone claims to make a distinction is not proof that they do make that distinction, nor that they are able to do so. The linguistic literature is not lacking in self-assessments that are easily shown to be wrong.

    And as I stated originally, "tone is not a feature that can be absent from speech". It is therefore not possible to interpret that Hunanese professor's claim directly, since the literal interpretation doesn't work. He would have to explain what he meant in other terms.

    It is possible for a syllable to feature a pitch contour that is not particularly close to any of the four standard Mandarin tones, although if you characterize them as "start high, stay high", "start high, transition low", "start low, transition high", and "start low, stay low" (I know this isn't a standard view of 3rd tone, but it's a reasonable description of how that tone occurs when it is not followed by either another 3rd tone or a neutral tone) it would be difficult to avoid matching one of them. Using a more standard description of 3rd tone as "high-low-high", a pitch contour that did not resemble a Mandarin tone would be pretty constrained; it would have to either take the form "start low, stay low" or vary back and forth quite a lot. It is my understanding that the 3rd tone can only realize all three points of its contour when produced as a citation form or when followed by a syllable with neutral tone, which would make a contour with a lot of variation over a single syllable somewhat implausible.

  39. Michael Watts said,

    April 30, 2016 @ 11:17 pm

    (It occurs to me that Japanese is unlikely to use the simplified character 来.)

  40. Victor Mair said,

    May 1, 2016 @ 8:18 am

    You're still unable to distinguish between tone and intonation.

    You're still not addressing the issues raised in the original post.

    You're not being brief.

    You're not being relevant.

    You continue to talk in circles.

    You are incapable of seeing that there are differences in the habits and preferences of individuals in their use or nonuse of Sinitic tones when pronouncing letters of the alphabet and in their ability to describe those habits and preferences. You unceasingly asseverate that there is a single "tonal" (by your own definition) straitjacket to which all Sinitic speakers subscribe in the face of clear evidence that some speakers add Mandarin / Cantonese / Hunanese, etc. tones to the letters and some do not.

    You're not contributing anything new and useful to the conversation.

    ***Please read the comments policy at the top of our home page.***

    The next self-indulgent monolog posted by you will be deleted.

  41. Bob Ladd said,

    May 2, 2016 @ 1:23 am

    The core question behind the broken-down communication in this thread is the following: what does it mean, in objective phonetic terms, for letter names to be pronounced in a Chinese sentence "without tone" or "with tone"? I think that is a legitimate question, and I would certainly be curious to know the answer. I'm perfectly prepared to believe that the very perceptive Hunanese scholar is referring to some objectively describable difference in behaviour when he says that he adds tones in Hunanese and not in MSM, but I'd like to know what that difference is.

  42. Victor Mair said,

    May 2, 2016 @ 6:17 pm

    Ch. lái — trad. 來 / simpl. 来 ("come")

    Jp. rai / ku(ru) — 来 ("come")

  43. Victor Mair said,

    May 2, 2016 @ 9:12 pm

    @Bob Ladd

    Thanks for your good suggestions.

    What I can tell you about the language of the Hunanese speaker is this. The topolect in his hometown belongs to the Yueyang division in the "New Xiang" subgroup of Xiang Sinitic. According to the Chinese Wiki page for "Xiang Sinitic" (https://zh.wikipedia.org/wiki/湘语), there are 7 tones, including the entering tone in several of the subgroups (see also this English Wikipedia page for Xiang: https://en.wikipedia.org/wiki/Xiang_Chinese). The tones and other phonological features among the subgroups of Xiang vary somewhat, but one thing is certain, they are not identical with the tones of Modern Standard Mandarin (MSM) — 4 + neutral, no entering tone.

    I suspect that the Hunanese speaker may automatically add Yueyang tones to English letters when he is speaking his Mother Tongue because it is natural for him to do so; he doesn't have to think about it. However, when he speaks MSM, it is his second language, and the 4 +1 tones of Mandarin are quite different and unnatural to him in the sense that they are secondarily acquired, so he would need to make a conscious effort to add them to English letters when speaking MSM, but he would have no principled means for doing so. Speaking MSM itself with correct tones requires enough of an effort for him. Consequently, when he comes to an English letter or acronym in MSM, he simply forgoes the the tones all together.

  44. JS said,

    May 3, 2016 @ 1:22 am

    "With aspiration" means a native English speaker pronounces the "t"s in "tortilla" [tʰ] or so; "without aspiration" means she pronounces them [t] or so, rejecting (in one respect) nativization to English phonology due to some more or less sophisticated knowledge of Spanish.

    The respondents to VHM's post all have relatively sophisticated knowledge of English and so may make the equivalent decision to retain English-style intonation (among other features) when saying "WTO" (rejecting da3biu1ti1ou1), "DVD" (rejecting di1wei1di1), "Virginia" (rejecting fu2ji2ni2ya4, etc.), "penicillin" (rejecting pen2ni2xi1lin2), etc.

    Chinese has phonemic tone so nativized borrowings feature phonemic tone so… that's all. This was an amazing thread though.

RSS feed for comments on this post