Consideration about chinese dialects

[Damin98]

你们好.

我叫文达。我是波兰人。我有一些问题.

 

1) 方言

a) Which dialect has the most combinations of Initials and finals (syllables)?

b) Which dialect has the least homonyms/homophones (where the role of tones in distinguishing words would be the smallest)?

c) Which dialect has the least tones and in which dialect using tones is the least important in mutual understanding?

d) Which dialect is the easiest to learn for 老外

2) 普通话 ≠ 官话

a) Could be 普通话 created on the basis of a different dialect than 官话, more "friendly" to 老外 (relatively small number of homonyms, less number of tones and less emphasis on their correct speaking)

3) 粤语

a) Is it possible to "maim" Cantonese tones and limit them to 3

(growing, falling, high / flat)?

2.5 Cantonese tones ⇒ as growing

1,3,6 Cantonese tones ⇒ as flat / high

4 Cantonese tone ⇒ as falling

b) Is it enough to know English in 香港 and 澳门 ?

c) Is it worth learning 粤语 and other 方言?

4) 梦想 (theoretically)

If we collect different syllables from different dialects, e.g. from seven major dialects: 官话 粤语 客家 话 吴语 闽语 赣 语 晋 语 湘 语

can we create a 普通话 v.2, in which practically there would be no tones (reduced to, for example, 2 or tonic accent) and no homonyms/homophones (because they would be more syllables), which would really (for me) reduce the difficulty of the language.

I hope for interesting answers and discussion.  谢谢

[Publius]

Can't answer all your questions. Just want to point out some obvious misconceptions.

 

17 hours ago, Damin98 said:

b) Which dialect has the least homonyms/homophones (where the role of tones in distinguishing words would be the smallest)?

In Chinese, tones are phonemic. 'Mā' and 'mǎ' are a minimal pair, the same way 'bat' and 'pat' are a minimal pair.

You don't call German words  schon /ʃoːn/ and schön /ʃøːn/ homophones just because they look similar. When typing German on a keyboard that does not have umlaut letters, ä, ö, ü are replaced by ae, oe, ue. On a keyboard that does not support pinyin tone marks mā, má, mǎ, mà, you use tone numbers ma1, ma2, ma3, ma4.

And phonetically, the fact that a German coast guard can't tell the difference between 'think' and 'sink' doesn't make them homophones in English.

 

17 hours ago, Damin98 said:

a) Which dialect has the most combinations of Initials and finals (syllables)?

I don't have the figures but I'd venture to say the differences aren't significant. Because dialects with the least initials tend to have the most finals, and vice versa. (https://baike.baidu.com/item/%E5%A3%B0%E6%AF%8D)

The reason is kinda obvious: they came from the same source. They basically inherited the same amount of distinct words to keep distinctive. Some developed a complex tonal system. Some simplified the syllable structure but added extra syllables to existing words. It's worth noting that southern dialects in which consonant codas (-p, -t, -k, -m, -n, -ng) are preserved tend to be more monosyllabic in general. You lose distinction here, you make up for it elsewhere. This is also how linguists believe tones arose in the first place (https://en.wikipedia.org/wiki/Tone_(linguistics)#Origin).

 

17 hours ago, Damin98 said:

d) Which dialect is the easiest to learn for 老外

Without doubt Standard Mandarin/Putonghua. Not because it's laowai-friendly, but because the shear amount of textbooks and learning materials.

 

17 hours ago, Damin98 said:

2) 普通话 ≠ 官话

a) Could be 普通话 created on the basis of a different dialect than 官话, more "friendly" to 老外 (relatively small number of homonyms, less number of tones and less emphasis on their correct speaking)

I'm not sure I understand your question.

官话 = Mandarin. See here for its geographic distribution: https://en.wikipedia.org/wiki/Mandarin_Chinese#Geographic_distribution_and_dialects

普通话 = Standard Mandarin in the PRC. “普通话是以北京语音为标准音，以北方话（官话）为基础方言，以典范的现代白话文著作为语法规范的现代标准汉语。”

In short, 普通话 is a sub-dialect of 官话. It's a national standard codifying the lingua franca of the Han Chinese of the early to mid 20th century. The reason why Mandarin is historical. Whether it's friendly to non-Chinese is a non-factor. (Would you standardize Polish based on how easy it is for foreigners to learn?)

 

17 hours ago, Damin98 said:

a) Is it possible to "maim" Cantonese tones and limit them to 3

(growing, falling, high / flat)?

Short answer: no.

Native Cantonese speakers can't always tell the difference between mid rising (2) and low rising (5), but I don't think they'll ever miss the other tones, including the entering tones 7, 8, 9, which technically are not distinguished from 1, 3, 6 by pitch contour. High level (1) and mid level (3)? They are as different as mist and mast.

 

17 hours ago, Damin98 said:

If we collect different syllables from different dialects, e.g. from seven major dialects: 官话 粤语 客家 话 吴语 闽语 赣 语 晋 语 湘 语

can we create a 普通话 v.2, in which practically there would be no tones (reduced to, for example, 2 or tonic accent) and no homonyms/homophones (because they would be more syllables), which would really (for me) reduce the difficulty of the language.

Of course you can. But then it would be another language, a conlang like Esperanto (whose creator is, incidentally, Polish). You could even attract like-minded people to form a community. But the rest of the unenlightened world would still be speaking their respective, difficult languages.

[Wippen (inactive)]

50 minutes ago, Publius said:

Can't answer all your questions. Just want to point out some obvious misconceptions.

Appreciate your insightful contribution here.

 

50 minutes ago, Publius said:

And phonetically, the fact that a German coast guard can't tell the difference between 'think' and 'sink' doesn't make them homophones in English.

:-)

And the past tense of 'sink" heard every day on German trains: 'sank' you for traveling with Deutsche Bahn,.

On second thoughts it should be 'sunk' as past tense, oder?

[Bao]

On 3/23/2018 at 1:22 PM, Damin98 said:

 

If we collect different syllables from different dialects, e.g. from seven major dialects: 官话 粤语 客家 话 吴语 闽语 赣 语 晋 语 湘 语

can we create a 普通话 v.2, in which practically there would be no tones (reduced to, for example, 2 or tonic accent) and no homonyms/homophones (because they would be more syllables), which would really (for me) reduce the difficulty of the language.

 

It's an interesting idea but this sounds like it would be an even harder language to learn than modern Mandarin Think of all the new sounds you'd need to learn in order to make up for the loss of tones- For example there are many dialects of Chinese that distinguish between aspirated and unaspirated p, plus b (often transliterated as "p," "b," and "bb") and the same for t (giving "t", "d" and "dd"). This is pretty rare in languages, I only know of Hindi and a few other Indian languages that have this three-way distinction, so I think English and European-language native speakers would have a lot of trouble with this moreso than with tones. In general the 方言 (at least 吴 for sure) tend to use some very difficult sounds.

[Publius]

2 hours ago, Bao said:

Think of all the new sounds you'd need to learn in order to make up for the loss of tones

Good point well made.

 

Middle Chinese has simpler tones: 平/level, 上/rising, 去/departing (the forth "tone" 入/entering/checked is not a tone in the phonetic sense).

At the same time it has a large repertoire of 36 initials:

帮[p]  滂[pʰ]  并(b)  明[m]

非[p̪]  敷[p̪ʰ]  奉[b̪]  微[ɱ]

端[t]  透[tʰ]  定[d]  泥[n]

知[ȶ]  彻[ȶʰ]  澄[ȡ]  娘[ȵ]

......

There's the three-way distinction of stops (p-pʰ-b, t-tʰ-d, k-kʰ-g).

And what the heck is the difference between m and ɱ, n and ȵ (and ɲ)?

Is it easier to learn than Modern Standard Mandarin?

[歐博思]

6 hours ago, Bao said:

I only know of Hindi and a few other Indian languages that have this three-way distinction

Add Korean to this list, and like you said, modern Mandarin tones felt infinitely easier.

[Publius]

Something I'd like to add.

 

Once you understand the origin of tones, you can more easily see them as part of the syllable, rather than additional embellishment that can be dispensed with.

 

From the link above:

The 去, or "departing" tone, arose from the loss of [-s] at the end of words. When we look at Chinese loanwords into neighboring East Asian languages, we find support for this theory. For example, in Korean, the word for "comb", bis, is a loan of the Chinese word bì 篦, which means that when the word "comb" was borrowed into Korean, there was still an [-s] sound at the end of the word that later disappeared from Chinese and gave rise to a departing 去 tone.

 

You wouldn't call /pi/ and /pis/ homophones. Why would you call /pi˥/ and /pi˥˩/ homophones?

 

Another example is the yin-yang tone split in modern Chinese varieties.

Middle Chinese used to have four tone classes: 平 level, 上 rising, 去 departing, 入 entering. Later on in Mandarin the level tone split into two (阴平 dark-level and 阳平 light-level) depending on the voicing of the initial.

 

In general, voiced initial consonants tend to cause the following vowels to sound heavier, acquire a slightly lower pitch. This is a natural, but indistinctive feature of voiced consonants. When the voiced consonants disappear from the language, the voiced-voiceless distinction still needs to be maintained. So the difference in pitch height now becomes distinctive: formerly voiced initial + level tone => 阳平; formerly voiceless initial + level tone => 阴平.

 

For Mandarin, the split only took place in the level tone. For Cantonese, it happened in all four tone classes, thus 平, 上, 去, 入 => 阴平, 阳平, 阴上, 阳上, 阴去, 阳去, 阴入, 阳入 (阴入 later split further into 上阴入 and 下阴入 according to the vowel length). It's interesting to observe that in Cantonese

阴平(1)　阴上(2)　阴去(3)　【上阴入(7)　下阴入(8)】

˥ or ˥˧　　˧˥　　　　˧　　　　　˥　　　　　˧

阳平(4)　阳上(5)　阳去(6)　【阳入(9)】

˨˩ or ˩　　˩˧　　　　˨　　　　　˨

all the "yin" tones are higher than their "yang" counterparts.

 

In this regard, Mandarin is actually the easiest dialect, tone-wise.

 

Well, by now you're probably bored by all the phonetics.

Here's a simple advice. Get over it. Learn the tones. Toneless Chinese, call it what you want, is not Chinese.

 

 

[Wippen (inactive)]

59 minutes ago, Publius said:

Toneless Chinese, call it what you want, is not Chinese

 

This quote is excellent. I am going to keep it and quote you in the future, whenever someone brings up that tones are not relevant. I will say "a native Chinese linguist said. "

Would that be the right attribution?

 

@Publius Really appreciate you give up your time to post and share your evident academic knowledge. What you have written in the past about European languages I really appreciate You have a sound knowledge about linguistics it seems.

[Publius]

6 minutes ago, Tøsen said:

Would that be the right attribution?

Nope. That would be categorically wrong lol. You can freely quote me as "a random guy on the internet" though.

[Wippen (inactive)]

Just now, Publius said:

a random guy on the internet

:-) i am too naive thinking everyone is always telling the truth on the internet.

[ParkeNYU]

In terms of learning resources and opportunities, Mandarin is the easiest Sinolect to learn by far. However, if we were to put that aside, I think the Shanghai dialect of Northern Wu would be the easiest for foreigners to learn despite the three-way distinction between voiced, unvoiced, and aspirated obstruents (e.g. p/ph/b). Why? Although it has one more tone than Standard Mandarin on paper (citation tones), actual speech is rather binary, much like Japanese (and some Korean) pitch accents. In other words, because the presence or absence of initial voicing and a final glottal stop coda (or just the preceding short vowel) already narrows down the number of possible tones through complementary distribution, there are actually only two phonemically distinct tones in practice despite having five in theory. Essentially, isolated syllables can be generally rising or generally falling (the initial consonant will determine whether they are high or low in range), and syllables within clauses just alternate between high and low (again, the pitch height being secondarily influenced by the voicing of the initial consonant).

 

As far as foreigners are concerned, you could just safely tell them that Shanghainese has two tones: high/falling and low/rising (as long as they are very careful with all the vowels and consonants). The changes in actual recorded pitch beyond these two can be considered allotonic.

 

@Publius Of course, the language from which Chinese characters themselves were wrought was a toneless language known simply as The Chinese language with an 'old' prefix only in retrospect.

 

@Damin98

 

I think the others have answered most of your questions quite well (especially regarding the varying weights and complexities of initials, finals, and tones based on region), but I'd like to add my own two fēn regarding your later thoughts.

 

Although eliminating tones altogether would require either a massive increase in homophones or a return to Old Chinese phonological influence, there is a way to reduce them considerably. If you were to have a full inventory of voiced and unvoiced initials, as well as the full inventory of nasal and stop codas, you could reduce Chinese to three contours: flat, rising, and falling (or any three tonal properties you want). To further reduce this to two (more like a pitch accent system), you could play with changes in vowel length. For example, you could have a system whereby the long stressed syllable is tone 1, the short stressed syllable is tone 2, and the short unstressed syllable is tone 3. This would put it in line with Japanese, where you have both binary pitch accent and vowel length. Even in Middle Chinese, the already simple tone system could be divided into two groups: 平(even) and 仄(oblique), the latter comprising 上去入.

 

A sample system where 平(even)/仄(oblique) is distinguished by stress:

 

平 byang (short, unstressed)

上 ZRYONG (short, stressed)

去 KYUU (long, stressed)

入 NJYEP (stopped, stressed)

 

A sample system where 平(even)/仄(oblique) is distinguished by length:

 

平 BYAANG (long, stressed)

上 ZRYONG (short, stressed)

去 kyu (short, unstressed)

入 njyep (stopped, unstressed)