Using Memory Palaces for Chinese Vocab and tones

[imron]

5 hours ago, fairlieflyer said:

It does however bring into focus the issue just how to learn tones as a single package with each piece of vocab

You can't learn the tones as a single package until you've internalised the tones.

 

You need to focus on hearing and treating the different tones as different sounds.  I've found it's best to do this without trying to ascribe meaning to the sounds.  Just listen to the sounds over and over again and be mindful of the differences until they begin to naturally sound different.

 

At that point, you should have little trouble remember the tone and sound as a single package.

[anonymoose]

On ‎2‎/‎6‎/‎2019 at 12:28 AM, imron said:

The real problem is that you likely have an incorrect mental model of the language where you see pronunciation as 'sound + tone', when actually you need to see it as 'sound', with different tones being different sounds.

 

I used to subscribe to this opinion also. However, from my observation, I don't think, even for native speakers, that tone is on an equal footing to initials and finals as far as sounds go.

 

Firstly, natives do make mistakes with tones occasionally, and when they realise, then correct themselves. This is only anecdotal, and I have no way of quantifying this, but I've noticed it more than once (and I'm sure there have been many more occasions when I haven't noticed this). Take (as a completely random example) the word jīngjì. It is conceivable (to me, at any rate), that a native speaker might accidentally or mistakenly say jǐngjì, but would very unlikely accidentally say qīngjì or jīngpì.

 

Secondly, there seems to be more variation in the tone part of pronunciation than initials/finals in different areas speaking Mandarin (and to be clear, I'm not talking about accents, where a certain sound is replaced by another sound uniformly). I'm sure many people have noticed that certain people say yīnwèi for 因为, which is the dictionary pronunciation, yet many people alternatively say yīnwéi. Similarly for 时期 which is pronounced shíqī on the mainland and shíqí in Taiwan. The dictionary pronunciation of 坊 is fāng, yet many (most in my experience) pronounce it as făng. 混血儿 is usually pronounced as hùnxŭer even though 血 has no official reading as xŭe. 比较 is officially pronounced bĭjiào, yet frequently pronounced as bíjiăo. Are there similar examples where the initial or final of a particular syllable in a word is changed depending on region or habit? For example, saying pĭjiào instead of bĭjiào. I can't think of any.

 

Thirdly, and this is admittedly a tentative argument as I don't have enough knowledge of the actual situation, but it seems to me that in the various dialects related  to Mandarin (in other words, such as 北方话，四川话 etc.), there is more variation in the tones than there is in the initials/finals. I'm not sure if all tones are replaced in a uniform fashion, in which case it could be argued that this is just a variation in accent, or whether tones are changed selectively, in which case it is more dialectal.

 

Fourthly, tone sandhi refers to a change in tone of a word depending on the tones of the surrounding words. Why does this only happen with tones and not initials/finals? This is even more the case in some dialects, such as Shanghainese,  where sandhi essentially change the tone of every single syllable in a word from their tones in isolation. (To be fair, in Shanghainese, initials do vary is some cases. 常常, for example, is pronounced as săngzāng. But these examples are few, and insignificant compared to tone sandhi which apply to every single word.)

 

And finally, even for native speakers of one dialect (or other tonal languages such as Vietnamese, Thai, etc.), they have similar difficulties with tones when learning other dialects, as do non-native speakers of Chinese. So it's not as simple as saying some people are programmed to hear tone as an integral part of the pronunciation, whereas others aren't.

 

To be clear,  I'm not saying that native speakers are conscious of these things, and certainly aren't thinking about tone numbers when speaking. But I'm a bit sceptical of the argument that tone is as integral a part of the pronunciation of a sound as are initials and finals. It's a nice argument made by competent learners to show that they've reached the native-like stage where tones are subconscious. But is there any evidence that, for native speakers, tone and initial/final are an equal part of any sound?

[markhavemann]

31 minutes ago, anonymoose said:

natives do make mistakes with tones occasionally, and when they realise, then correct themselves

I would argue that English speakers make mistakes with vowels and consonants too, sometimes things that should be unvoiced come out voiced, sometimes the other way around, and sometimes as another sound altogether. I correct my own speech, and I hear other people correcting themselves too. 

 

31 minutes ago, anonymoose said:

there seems to be more variation in the tone part of pronunciation than initials/finals in different areas speaking Mandarin

In English there is relatively little variation in consonants between different accents and the vowels tend to be the sounds that change the most. In American and British English vowels are totally different. Maybe the most obvious differences are in words like "glass" but even in words like "hot" or "jaw" there is a big variation. Does that make vowels less part of the word than consonants? (also, we can still fully understand each other despite this, which I would say is comparable to your Taiwan/mainland example). 

 

Another interesting thing is that in Sichuan, most people can't tell the difference between L and N, and they will use them in place of each other all the time and not even know they are doing it. They just figure out from context which is which and they have huge difficulties with these two sounds in English. Even though an entire province is messing up these 2 consonants, you wouldn't say that it's less of a part of Chinese words than the other consonants are. 

 

31 minutes ago, anonymoose said:

Fourthly, tone sandhi refers to a change in tone of a word depending on the tones of the surrounding words. Why does this only happen with tones and not initials/finals?

Two third tones in a row could be a bit of a mouthful when you are speaking quickly. We have similar changes in English where the plural s will be voiced or unvoiced based on the sound that precedes it (/z/ if the sound is voiced as in "dogs" and /s/ if the sound is unvoiced as in "cats"). I can't think of an example where this kind of consistent change happens to vowels. 

 

I do see where you are coming from and why you would say this, and I think it's a good argument. I think tones can seem less important because there are only 4 so if you make a mistake there are only 3 other things that it could be confused with and context can clarify most of the time, so it seems like they aren't really part of the word in the same way. But, just because you can get something wrong and get away with it (like Sichuanese with L and N, and us with tones) doesn't make any less a part of the word.  

[roddy]

1 hour ago, anonymoose said:

Why does this only happen with tones and not initials/finals?

Wouldn't that be assimilation? To what extent that happens in Chinese I don't know, and searches are hampered by all the cultural assimilation results, but I'd imagine that in rapid speech you see similar features.

[anonymoose]

6 hours ago, markhavemann said:

I would argue that English speakers make mistakes with vowels and consonants too, sometimes things that should be unvoiced come out voiced, sometimes the other way around, and sometimes as another sound altogether.

 

Sure, I acknowledge that. This is an enunciation mistake, which can happen in any language. I'm referring to situations where an incorrect tone may be said due to delayed or vague recollection of the correct tone.

 

6 hours ago, markhavemann said:

In English there is relatively little variation in consonants between different accents and the vowels tend to be the sounds that change the most. In American and British English vowels are totally different. Maybe the most obvious differences are in words like "glass" but even in words like "hot" or "jaw" there is a big variation. Does that make vowels less part of the word than consonants?

 

The situation with accents is different, but even so, what you've illustrated is that there is a hierarchy in significance even between vowels and consonants. So it is certainly not a stretch of the imagination to think there may also be a difference in significance between these and tones.

 

Nevertheless, I'm not arguing about their relative "importance", but rather that initials/finals and tones have distinct identities in a syllable, even amongst native speakers.

 

Another piece of evidence to this would be the zhuyin fuhao used in Taiwan. This set of symbols was designed specifically to represent Chinese sounds. If ā and á are as unrelated to a Chinese speaker as are ā and ē, why are they not all represented as unique symbols? In fact, ā, á, ă and à are all represented by the same symbol, with only an additional tone mark to distinguish them.

 

Perhaps a counter example would be the Yi writing system, where most syllables do have a unique symbol, even for syllables which are identical save for the tone. But even in this example, there are two tones which use common symbols with the addition of a diacritic to distinguish them. For example, the syllables it, ix, i and ip (where the t, x and p represent tones in standard Yi romanisation) are written as ꀀ, ꀁ, ꀂ and ꀃ.

 

7 hours ago, markhavemann said:

Two third tones in a row could be a bit of a mouthful when you are speaking quickly.

 

Personally, I am not completely convinced by this either, although I cannot explain why else the two third-tone sandhi phenomenon would have evolved. In any case, other dialects where tone sandhi are much more pervasive, such as Shanghainese, demonstrate that tone is much more subject to variability than initials/finals for the sake of facilitating enunciation.

 

Interestingly, other languages, like Japanese for example, do show changes in consonants in certain situations, like 時々 "tokidoki" and 人々 "hitobito".

 

Anyway, to get back to my main point: I don't know whether using memory palaces or brute force drilling is more efficient to learn tones. I just don't think memory palaces should be dismissed outright just because they "artificially" separate the tone from the rest of the sound. In the end, the objective is the same - to commit the sounds so indelibly to memory that they can be recalled without having to think consciously.

[fairlieflyer]

Firstly I have to say I'm very glad I posted here. I now know what I really want to achieve with my Chinese. Imrons articles at ChinesetheHardway is especially relevant for me at this post-HSK3 crossroads.- so, thank you everyone.

 

In case others find this post and find themselves on a similar 'good vocab but bad tonal' cul-de-sac and are looking to find a way out,  here's my plan which you may consider... 

 

My focus for 2019 will be Speaking, Listening and Reading.  (I am going to park Writing in the passive learning category this year. No more Skritter, hooray)

This will require more of:

• Repetitive practice of sentences passively and actively using native audio listening to tones and overall flow.
• Talk, listen and read more in real-time (various media and people)
• Find and upload Vocab that is relevant to my learning versus solely HSK lists. (To SRS and sentence/audio this)
• Drill down on more common cognates/homophones so I hear the difference and usage
• Focus on consecutive days learning versus irregular cram days
• Use Textbooks (eg NPCR) based approach to push me through those slump days
• to sit HSK 4

Less of:

• Solely HSK vocab SRS lists
• Less using HSK exams as the reason to study 
• Fewer Mnemonics/Memory Palaces and only when required, ie not for all
• Less cramming and more time for repetition in multiple formats to hear and understand usage.

For what its worth, these are useful tools I will start using/continue with:

• Modern Mandarin Chinese Grammar / NPCR series of texts
• Will look at Glossika, audio  Anki audio shared files, FIM audio, lrts.me, Youtube, local Chinese newspapers, Chinese TV etc to find the audio+text files I need to practice
• Chinese Test Analyser to pull out relevant words I need for reading vocab
• Continue with Anki SRS for core lists and to keep full datasets turning over
• Pleco for everything
• Find a better language partner that can help me focus on the relevant areas

Thank you all for your kind pointers and tips from obvious hard-won experience. I'm very glad I went through this review exercise this week. In some ways, I think it feels I'm starting my learning again and has shades of T.S Eliot's poem I learnt at school, "We shall not cease from exploration, and the end of all our exploring will be to arrive where we started and know the place for the first time. " Very academic! ?

 

Will be a busy year!  加油!

[Publius]

On 2/7/2019 at 8:21 PM, anonymoose said:

But I'm a bit sceptical of the argument that tone is as integral a part of the pronunciation of a sound as are initials and finals. It's a nice argument made by competent learners to show that they've reached the native-like stage where tones are subconscious. But is there any evidence that, for native speakers, tone and initial/final are an equal part of any sound?

No, it's not just a way to show off. It has sound basis in linguistics.

 

For one thing, Chinese tones are phonemic, meaning they form the smallest units of sound that help differentiate units of meaning (morphemes). The /θ/ sound in standard English is a phoneme. 'Thin' and 'tin' are two different words distinguished by one phoneme. They are a minimal pair, as are 'thin' and 'sin', 'thin' and 'fin'. English /θ/ is a difficult phoneme for many foreign learners. They often mispronounce /θ/ and manage to get away with it. But that doesn't make it less phonemic. In Chinese, 妈 and 马 are two different words. Need I say more?

 

Minimal pairs are not confined to consonants and vowels. Word stress in English is also phonemic. 'Insight' and 'incite' are distinguished in speech only by the placement of stress. Stress also differentiates 'record' the noun from 'record' the verb. The initial stress in the noun is a type of derivational suprafix. That suprafix is just as important as any prefix or suffix. Misplaced stress can cause difficulty in verbal communication. Yet when you ask a native speaker, "Which syllable is stressed in the word 'electricity', the 1st, 2nd, 3rd, 4th, or 5th?" they will have to sound out the word before giving you the answer. Because they have never learned the stress as a separate attribute from the pronunciation.

 

Now let's turn to the origin of tones. It is generally believed that Old Chinese did not have tones, or at least not phonemic tones. Tones arose as Old Chinese lost the final consonants /ʔ/ and /s/. Words previously ended with a glottal stop became 上声 (rising tone); words previously ended with -s (a derivational suffix) became 去声 (departing tone); words that did not lose anything remained the same but to contrast with 上声 and 去声 was called 平声 (level tone). Interestingly enough, in Classical Chinese, a bunch of nouns can be used as verbs when pronounced in 去声 (e.g. 衣 yī, 'clothes'; yì, 'to clothe'). The function of a derivational suffix got carried on by a derivational suprafix (tone).

 

Later on, Middle Chinese 平声 split into 阴平 and 阳平 of Modern Chinese to compensate for the loss of voiced initials. In general, voiced initial consonants tend to cause the following vowels to have a lower pitch. This side effect of voicing can be observed in other non-tonal languages (for example, go to Google Translate to hear how these four syllables ka kha ga gha are pronounced in Hindi). When the voicing was lost, the previously undistinctive difference in pitch became distinctive (phonemic tones). 阴平 (1st tone) is higher than 阳平 (2nd tone) in standard Mandarin for a reason. More interesting is Cantonese, where not only 平声 split in two, 上声, 去声 and 入声 (syllables ending in -p -t -k) also split, and all the Yin tones occupy the upper half, leaving all the Yang tones in the lower half. Yang is called Yang because it is masculine, it is heavier. It is "heavier" because there was (maybe still is) a voiced initial.

 

Tones keep distinct words distinct in much the same way consonants /θ/ and /ð/ keep the English words 'cloth' and 'clothe', 'bath' and 'bathe', 'breath' and 'breathe' apart. Shanghainese has only two phonemically contrastive tones, but it has a large repertoire of consonants and vowels. It maintains a three-way distinction of initial stops (p/pʰ/b, t/tʰ/d, k/kʰ/g). It can afford some lack of distinctiveness in tones. Mandarin has only four tones. It has to make up for the loss in other ways, i.e. neutral tone, erhua, and extensive use of disyllabic words (Cantonese, with 6-9 tones, retains a lot more monosyllabic words). In conclusion: yes, tones are as integral a part of the pronunciation as initials and finals. They are in fact vestiges of lost initials and finals, and carry out the same functions of those initials and finals.

 

The rest of your argument is not worth refuting.

 

For example, your second point shows how little you know about the phenomenon called 文白异读 (literary and colloquial readings of Chinese characters). Many characters have variant readings, which more often than not involve different initials/finals. When I was in school, 呆板 was pronounced áibǎn (a staple in pinyin exams), but now only dāibǎn is listed in dictionary. When my teacher was in school, 白璧微瑕 must be pronounced as bóbìwēixiá. In both cases the plebs won. The official pronunciation of 坊 is on its way out. Ditto 血. Ditto 较. You just happened to select the examples that support your view while ignoring the mounting evidence to the contrary, e.g. 这 zhè vs zhèi, 那 nà vs nèi, 剥 bō vs bāo, 壳 qiào vs ké, 癌 yán vs ái; and Cantonese 聽 ting1 vs teng1, 啄 doek3 vs doeng1; and Beijingese 碎 cèi (which is not even a valid syllable in standard Mandarin), 知了 jìniǎor.

 

And your final point ("even for native speakers of one dialect, they have similar difficulties with tones when learning other dialects"), same can be said of native English speakers. How easy is it to do convincing accents for a sustained period of time, for example, Kiwi English or Black English, even if you are told all the sound change rules? Not so easy, I bet. What does it prove? It proves that tones are just as integral a part of the pronunciation as are consonants and vowels. You have to learn the whole package and train your muscles to produce the exact same sound, not just some modulation based on a separate body of knowledge retrieved from a library or a palace.

 

This is even more ridiculous.

15 hours ago, anonymoose said:

Nevertheless, I'm not arguing about their relative "importance", but rather that initials/finals and tones have distinct identities in a syllable, even amongst native speakers.

 

Another piece of evidence to this would be the zhuyin fuhao used in Taiwan. This set of symbols was designed specifically to represent Chinese sounds. If ā and á are as unrelated to a Chinese speaker as are ā and ē, why are they not all represented as unique symbols? In fact, ā, á, ă and à are all represented by the same symbol, with only an additional tone mark to distinguish them.

What defines a distinct entity in a language is the sound, not the letter, not the spelling. In English, 'lead' (verb) and 'lead' (metal) have very different vowels, but share the same spelling. The reason why Zhuyin opted for a standalone tone mark is because a standard typewriter back then had only so many keys and there were no computers! Pinyin does use unique symbols. Look how easy it is to type tone marks even today on a computer! Plus what difference does it make? No matter how you write it, 馬, ㄇㄚˇ, mǎ, mǎ, ma3, ma³, they all represent the same syllable /ma˨˩˦/. And the tone is part of that syllable. Just consider this: If there were no Pinyin, no Bopomofo, no any transliteration system, only Hanzi, would that make tones more integral or less integral part of a syllable? And what if Chinese was a language without a writing system?

[艾墨本]

@Publius Thank you for the insightful post. I learned a lot. This is likely one of my favorite posts on this forum, now.

[markhavemann]

I don't think that I can really add anything after Publius brilliant post. I wonder if you could share some resources or more in depth articles about the development of tones? That' something I would love to read/hear more about. 

 

It seems like anonymoose met some pretty strong criticism/opposing views but I hope it won't discourage him or anyone voicing their opinions in the future because this was maybe the most interesting discussion I've seen on the forum so far.

@fairlieflyer That looks like a good plan, please post and let everyone know how it's going. As an aside, if the goal of writing HSK exams helps to motivate you to study then don't be afraid to use that as your reason for opening your books. Study for the stake of study is a wonderful ideal but having tangible milestones to work towards is useful and motivating. (Also, I found the HSK 4 vocabulary was a really good collection of words to know).

[imron]

15 hours ago, anonymoose said:

It's a nice argument made by competent learners to show that they've reached the native-like stage where tones are subconscious

I've been making the same argument since I was a beginner when I noticed no Chinese person could tell me the tone for a character without first going through all the tones and then picking the right one.  They had a different mental model of how the language worked, and I decided to use that one instead of the one I'd previously been using.

 

I also think discussing which part of the sound plays the most important role in distinguishing the sound is academic because all parts do play a role.  Different tones might be a similar sound, but they are still seen as different.  Likewise there are similar consonants such as 'g' and 'k' or 'd' and 't', and, like the tones, you could replace such similar sounds with each other in your speech and still be mostly understood, but that doesn't mean that there's no need to learn how to pronouce 'k' or 't', or that those sounds are not important and don't play a role in distinguishing separate sounds.

 

6 hours ago, anonymoose said:

I just don't think memory palaces should be dismissed outright just because they "artificially" separate the tone from the rest of the sound.

That's not my main reason for dismissing memory palaces.  It was a side comment on the memory palace OP was building.

 

My main reason for dismissing memory palaces is that I think using the Chinese language well isn't a function of memory but rather a function of usage so it's more effective to focus on using the language rather than on remembering large amounts of vocabulary, but even just looking at memorisation, firstly they require a multistep process to get what you are after, and secondly I think they introduce to much interference - either with meaning or things using an English based pronunciation, that they are less beneficial for getting a good grounding in the language.

 

6 hours ago, anonymoose said:

In the end, the objective is the same - to commit the sounds so indelibly to memory that they can be recalled without having to think consciously.

On this we can agree.

[imron]

5 hours ago, fairlieflyer said:

Focus on consecutive days learning versus irregular cram days

In my opinion, this will be the most significant driver of progress.

 

5 hours ago, fairlieflyer said:

Use Textbooks (eg NPCR) based approach to push me through those slump days

This is also a good idea - not necessarily a text book, but just having a pile ready-at-hand learning material that you can jump in to without needing to think 'what to study'.

 

Just sit down with it, and work your way through it, and stop when you're done.

 

This is what I do with reading - I always have the next book I want to read ready and on hand before I finish whatever I'm currently reading.  That way when I finish, I don't need to think about what to read next because it's already there and ready to go.

[anonymoose]

22 hours ago, Publius said:

For example, your second point shows how little you know about the phenomenon called 文白异读 (literary and colloquial readings of Chinese characters).

 

According to the page that you linked to, "For a given Chinese variety, colloquial readings typically reflect native phonology, while literary readings typically originate from other Chinese varieties, typically more prestigious varieties." All this suggests is that the discrepancy in readings arises from borrowing one pronunciation from effectively another language. This tells us nothing about the relationship between the tones and initials/finals in Mandarin.

 

22 hours ago, Publius said:

You just happened to select the examples that support your view while ignoring the mounting evidence to the contrary, e.g. 这 zhè vs zhèi, 那 nà vs nèi, 剥 bō vs bāo, 壳 qiào vs ké, 癌 yán vs ái

 

As far as I understand it, the zhèi and nèi pronunciations come from a contraction of zhèyī (这一) and nàyī (那一), but I accept your other examples.

 

22 hours ago, Publius said:

And your final point ("even for native speakers of one dialect, they have similar difficulties with tones when learning other dialects"), same can be said of native English speakers. How easy is it to do convincing accents for a sustained period of time, for example, Kiwi English or Black English, even if you are told all the sound change rules? Not so easy, I bet.

 

Not that difficult.

 

22 hours ago, Publius said:

The reason why Zhuyin opted for a standalone tone mark is because a standard typewriter back then had only so many keys and there were no computers!

 

Is that the reason? I did a search, and I cannot find a single example of a Zhuyin typewriter. If you can, please provide a source, as I would be interested to see.

 

21 hours ago, markhavemann said:

It seems like anonymoose met some pretty strong criticism/opposing views but I hope it won't discourage him or anyone voicing their opinions in the future because this was maybe the most interesting discussion I've seen on the forum so far.

 

I'm fine with that. It's always interesting to hear others' points of view.

 

19 hours ago, imron said:

I also think discussing which part of the sound plays the most important role in distinguishing the sound is academic because all parts do play a role.  Different tones might be a similar sound, but they are still seen as different. 

 

I agree with this, and as I stated before, I wasn't arguing about the importance of tones relative to initials/finals, but rather about the distinctness of their identity and how this may lead to them being treated differently. Nevertheless, I suppose on some level, this boils down to the same thing. A piece of evidence to support this is that tones are quite happily sacrificed in singing to make way for the tune, yet finals cannot be arbitrarily manipulated to accommodate the rhyme. As markhavemann said,

 

On ‎2‎/‎7‎/‎2019 at 1:13 PM, markhavemann said:

tones can seem less important because there are only 4 so if you make a mistake there are only 3 other things that it could be confused with and context can clarify most of the time

 

[markhavemann]

2 hours ago, anonymoose said:

A piece of evidence to support this is that tones are quite happily sacrificed in singing to make way for the tune

English songs also sacrifice a lot of the language and can be VERY difficult to hear clearly, (Pearl Jam's Yellowledbetter is a great example). When I've asked Chinese people about using music to learn Chinese they almost always tell me not to because "Chinese music can be really hard to understand". 

 

You still haven't accounted for the way that vowels and consonants can be sacrificed in the same way in English without losing too much of the language.

 

In my classes I have over 100 university students learning English and they make a whole lot of mistakes in their pronunciation: voiced TH becomes Z,  voiceless TH becomes S, V becomes W, short i sounds become long i sounds, the diphthong EI becomes a short E, extra vowel sounds get tacked onto the end of syllables, almost none of them can pronounce / ʒ / and most of them have problems pronouncing R correctly too. Spoon, moon and own, become Spoowen, moowen and owen, and this list is the tip of the iceberg. 

 

The better a student's English, the fewer of these mistakes they make. Students with lower levels make more and more of them and become increasingly difficult to understand. Interestingly, the cut off point for where other students can't actually understand their classmates is much higher than where I stop understanding them. Since my English is much better, I have a better chance of predicting what they are really trying to say and looking through their mistakes. However, there are times when context can't clarify and too many mistake at the same time take them too far away from what I'm able to guess, or they make a mistake that, in a given context, makes it seem like they have said a word that fits in that context but actually isn't what they wanted to say. 

 

The point here is that you quite a large margin for error in a language, especially when talking to native speakers. As long as within a given context there are not too many other words that can be confused for what you just messed up, and you don't mess up on too many levels at once, then native speakers can still get it. Still, I've had misunderstandings because of using tones incorrectly. But usually as long as you don't mess up too much at once, you can get by.

 

I think what you are ignoring is that the rest of Chinese sounds are quite limited and there are only like 800 possible initial/final combinations. For English speakers these sounds are also relatively easy to get something of a mastery over and rule out most of the mistakes that can be made. Native speakers have very good Chinese (unsurprisingly) and are usually able to figure out what you are saying even if you ignore tones altogether. This is because they represent a smaller (but not separate or less important) part of what makes each word mean what it does, and context helps a lot. (just like when a student say "Ze doga is on ze mata", I can figure out they are trying to say the "the dog is on the mat", because they have gotten the majority of it right). 

 

Even in Chinese you can get away with messing up initials and finals, as long as you don't mess up too many at once. In the Sichuan accent CH is almost always pronounced as a C, SH as S, ZH as Z, CH as C, sometimes H becomes F. The vowels change too, one example is pinyin E sound (as in 可) something like the first vowel sound in "organise". I can hardly understand when the older people here speak because they accents are so strong, but native Mandarin speakers are more or less used to it in a week or two and can get past all of this, as long as there aren't too many all at once, just like they can get past our tone mistakes when we talk. 

 

 

[imron]

1 hour ago, markhavemann said:

almost none of them can pronounce / ʒ /

Quick tip to fix this - tell them to pronounce it as /ʃ/.  They'll all be able to pronounce /ʃ/ correctly enough and the only difference between /ʒ / and /ʃ/ is that one is voiced and one is not. 

 

When pronounced in a word /ʃ/ will often pick up some voice from the previous or following sounds and it will sound almost correct.

 

The only two IPA symbols I know are ʒ and ʃ and the only reason I learned them was to figure out how to solve this problem.

[markhavemann]

6 minutes ago, imron said:

Quick tip to fix this - tell them to pronounce it as /ʃ/.  They'll all be able to pronounce /ʃ/ correctly enough and the only difference between /ʒ / and /ʃ/ is that one is voiced and one is not. 

Good advice! It seems kind of amazing that students need a teacher to point this out, and shows how we don't naturally link voiced and unvoiced sounds in our minds.

 

Interestingly doing this only takes them 90% of the way, which made me realise that the sound they are making for SH isn't quite the same as the English SH sound. Chinese SH and X seem to be sort of on either side of the English SH on the /ʃ/ spectrum (if that makes any sense) and my students mostly seem to use one of those for English SH. It's basically imperceptible when it's voiceless, but a little more obvious when voiced. 

 

Still though, it's a good approach and gets them close enough to the real thing that they can play around a bit and find the sweet spot.

[anonymoose]

8 hours ago, markhavemann said:

English songs also sacrifice a lot of the language and can be VERY difficult to hear clearly

 

Precisely. If you start sacrificing consonants and vowels, understanding very quickly becomes impeded.

 

As for the rest of your post, you're talking about accent, which is an entirely different kettle of fish.

[markhavemann]

3 minutes ago, anonymoose said:

Precisely. If you start sacrificing consonants and vowels, understanding very quickly becomes impeded.

Most of the point of my post was that sacrificing consonants and vowels DOESN'T degrade communication very quickly, rather that if you sacrifice too much, or too many different things at once, only then it becomes a problem (regardless of which parts/sounds you are sacrificing on).

 

6 minutes ago, anonymoose said:

As for the rest of your post, you're talking about accent, which is an entirely different kettle of fish.

Unless I've misunderstood, the debate is whether or note tones fall in the same realm as vowels and consonants and in a way their relative importance in a word. Your argument is that you can remove tones to some degree without communication breaking down and they are therefore not the same. To test this, it seems reasonable to consider what happens if you start removing or changing consonants and vowels in the same was as you say it's ok to do with tones. Surely accents and cases where language hasn't been learned fully are a good place to start?

[Publius]

On 2/8/2019 at 10:11 AM, markhavemann said:

I wonder if you could share some resources or more in depth articles about the development of tones? That' something I would love to read/hear more about. 

I think these Wikipedia articles are a good starting point:

https://en.wikipedia.org/wiki/Tone_(linguistics)#Origin

https://en.wikipedia.org/wiki/Four_tones_(Middle_Chinese)

 

 

11 hours ago, anonymoose said:

According to the page that you linked to, "For a given Chinese variety, colloquial readings typically reflect native phonology, while literary readings typically originate from other Chinese varieties, typically more prestigious varieties." All this suggests is that the discrepancy in readings arises from borrowing one pronunciation from effectively another language. This tells us nothing about the relationship between the tones and initials/finals in Mandarin.

You are quoting out of context. That part is not about Mandarin. Just read the whole thing. And don't forget all varieties of Chinese are genetically linked. They are descended from the same ancestor but over time have developed their own phonological systems, the discussion of which, by the way, you precluded while laying out your claim that there is more variation in the tone part of Mandarin than in the initial/final part. I pointed out that these seemingly random variations have historical reasons, they reflect a competition between the old and the new, the elite and the mass, and initials/finals are just as changeable as tones.

 

11 hours ago, anonymoose said:

Not that difficult.

There are also videos of Chinese people doing accents, for example, this guy. The fact that they made a video for it is proof that it is not an easy task.

 

11 hours ago, anonymoose said:

I did a search, and I cannot find a single example of a Zhuyin typewriter. If you can, please provide a source, as I would be interested to see.

Okay, I made that up. But the point still is, how you write a language doesn't affect the language, and most certainly is not a complete and truthful representation of all the features in that language. Arabic is written without vowels. English gives no clue as to where the stress is. Similar for Japanese pitch accent. Those are all important features that distinguish one word from another. There are a number of ways to indicate tones. We can draw a dot at four corners as the ancients did; we can create a new set of letters as in Bopomofo and IPA, or co-opt the existing ones as in the romanizatoin of Miao and Zhuang; we can modify the roman letters a la Vietnamese, or we can just put a diacritic on the main vowel as in Pinyin; we can use numbers to indicate the tone classes as in Wade-Giles and Yutping, or use them to describe the actual tone contour. But none of these changes the nature of tones

 

11 hours ago, anonymoose said:

as I stated before, I wasn't arguing about the importance of tones relative to initials/finals, but rather about the distinctness of their identity and how this may lead to them being treated differently

That's not a new observation. Linguists have a term for things like tone and stress that cannot be nicely divided into units in sequence. They are called suprasegmentals. There are more things in language than consonants and vowels. Some of them are just as important.

 

11 hours ago, anonymoose said:

Nevertheless, I suppose on some level, this boils down to the same thing.

Are you arguing about the relative importance of tones or not? Make up your mind.

 

11 hours ago, anonymoose said:

A piece of evidence to support this is that tones are quite happily sacrificed in singing to make way for the tune

Consonants and vowels are also happily sacrificed in English poetry. We've all seen ever written as e'er, or dimmed as dimm'd (-ed must be a full syllable at the time, and remains so in adjectives like wicked, beloved, learned).

 

11 hours ago, anonymoose said:

yet finals cannot be arbitrarily manipulated to accommodate the rhyme

I'm not that sure. Weird things can happen when reciting classical poetry. For instance the 斜 in 遠上寒山石徑斜，白雲生處有人家 is pronounced xiá. It is not in any dictionary, and most certainly not the sound of the Tang dynasty when the poem was written (from its current 陽平 tone, we know the initial must be voiced). The 奢 in 柳永『望海潮』 also has an arbitrary reading shā. There must be more examples but I can't think of any off the top of my head.

Since you mentioned rhyme, do you know in regulated verse, 平上去 tones of the same final (e.g. 媽、馬、罵) cannot rhyme with each other?

[anonymoose]

6 minutes ago, markhavemann said:

To test this, it seems reasonable to consider what happens if you start removing or changing consonants and vowels in the same was as you say it's ok to do with tones. Surely accents and cases where language hasn't been learned fully are a good place to start?

 

No, because accent is a systematic replacement of one sound with another every time it occurs in a language. This is predictable and can be adapted to.

 

Perhaps a fairer comparison would be, say, remove all the initials, whilst maintaining the tone and final, and compare this to toneless Chinese and see which is easier to understand.

 

 

[markhavemann]

2 minutes ago, anonymoose said:

Perhaps a fairer comparison would be, say, remove all the initials, whilst maintaining the tone and final, and compare this to toneless Chinese and see which is easier to understand.

An interesting point. I'm going to think on it because while it's a good argument and I don't have a response right now, it still doesn't feel right and goes against my own experience with Chinese.

 

I wonder if @Publius or someone else might have a better response against that than I do right now.