The 7 tones - Recordings?

[amhoanna]

Cool, Ah-bin!

Vaskimies--

Since U're no stranger to tone sandhi, Banlam (Minnan)-type sandhi will be a snap. The basic rule is:

a) The last syllable in a "sandhi group", or "phrase" -- I think the pros might have a better word for this -- takes its citation tone.

b) All other syllables in the group take their "running", or "sandhi'd", tone.

The biggest challenge as a learner is to get a sense for when a sandhi group ends. As a rule, the last syllable of a noun will end its sandhi group -- when it's not being used as an adjective.

Banlam tone sandhi is "sandhi lite" compared to what exists in Eastern Min. Shanghainese tone sandhi works differently, and seems pretty complicated.

I'm curious though, do the tones really vary that much from dialect to dialect, for example Penang to Amoy, that no recording would work for every dialect (or at least most of them)? I know that words are going to take on different tones in different dialects and I'm guessing they also sandhi differently depending on dialect as well, but I'm talking about the tones themselves. Is á really pronounced that different in Xiamen than it is in for example Taiwan that I would need separate recordings for both dialects?

The short answer is that U may need different recordings for each dialect. It depends which ones U're interested in.

Amoy and "Mainstream Taiwanese" (and some Ciangciu dialects, inc. what I've heard of Cebu Hokkien) happen to be "tonally similar". The Amoy dialect is actually mirrored on Taiwan by "the Amoy dialect of Taiwan" -- the dialect of the river-port districts of Taipak (Taibei). Most music was recorded in this dialect up till pretty recently. Twenty years ago, young Taiwanese people generally spoke Mainstream Taiwanese, but sang in Amoy Taiwanese (shades of South Vietnam). These "twin" dialects are a product of a certain era of sea-borne trade and migration which coincided with the rise of Amoy as a big-time port, and the commercialization of northern Formosa. I think lots of older people speak this dialect in Singapore too -- somebody correct me if I'm mistaken. So, it's the most "urban" of all Hoklo dialects. Very few people under 30 are speaking it nowadays, though.

Contour-wise, Coanciu-type dialects, some Ciangciu-type dialects and Penang (generally considered a Ciangciu-type dialect) will differ wildly from Amoy/Taiwan. Just search on this forum for posts on the tones of Bagansiapiapi Hokkien (Sumatra)!! By far the biggest differences between dialects lie in vocab, though. And there's not really a "standard dialect" at this point.

[amhoanna]

To clarify: by "twin dialects", I meant the Amoy dialect of Amoy and the Amoy dialect of Taiwan.

[AndrewAndrew]

The classic tone wheel works for Amoy Hokkien, but not necessarily for other variants. E.g. for Penang Hokkien (which incidentally shares "distinct in sandhi but not in citation form" issue with Chiangchiu, not Choanchiu - see Douglas, who mentions the issue - in some variants of Choanchiu, there is no Yinqu/Yangqu distinction at all), the wheel is simply 1->7->3->1, with two one-way branches 2->1 and 5->7, and an isolated 4<->8 line.

[Ah-bin]

I did ask the linguist, and the answer was that it was possible that some of the sandhied tones preserved the original tone contours better then the unsandhied tones (and he shook his head disparagingly at the terms "running" and "standing"). So I suppose if that were the case you could say the citation tone contours diverged more from the originals than the sandhied tones.

In which case, my closed mindedness was not completely justified, and I apologise for it!

[amhoanna]

This means he posits a "proto-Hoklo" or proto-what-have-you language that had tones, but not sandhi?

Isn't it possible that proto-Hoklo or Proto-Min developed tone sandhi at the same time that it developed tones?

I'm out of my range here... I'm sure much has already been established concerning tonogenesis and Old Chinese.

Also there's the "problem" of alternative theories to Stammbaum. I saw a dense and possibly very sophisticated article in a journal at the book mall last week that identified three "theories" or "patterns" of relative linguistic evolution, one of them being Stammbaum, another being 触 something, and a third I can't remember. (It was written in Mandarin.)

[Ah-bin]

Ah, this person was an expert on tone, but not on Hoklo tone!

It's unlikely that the tones in Hokkien developed on their own, if that were the case, there wouldn't be any correspondences between the tone classes in other Sinitic languages or the old Sinitic loans in Vietnamese.

[amhoanna]

Since accurate info on Penang Hokkien tone contours is kind of hard to come by, I decided to tackle the issue with a Kedah native as informant. This is what came of it: http://banlam.tawa.asia/2012/10/beima-s ... kkien.html

Just now, I found Ah-bin's contour chart, buried in a random thread: http://www.chineselanguage.org/forums/v ... +21#p35264
... for those that don't mind the 3-level "tonemic" approach.

Regarding 33 vs. 44, my informant said he used to think it was 33, but now he thinks it's 44.

He also crafted a chart himself: http://sphotos-h.ak.fbcdn.net/hphotos-a ... 4305_n.jpg

[Ah-bin]

Thanks so much Amhoanna. It's nice to have confirmation of these things. I think the Wikipedia page should also reflect the reality of the situation, i.e. that Northern Malaysian Hokkien has a tone system that is vastly simplified compared to other varieties.

Ah-long has also noted a higher tone 55 in borrowed words such as the final particle me• (this would be your friend's 高陰入 but it is not always a and a few colloquial onomatapoeic words, but basically (if you do what Cantonese textbooks do, and base the number of tones on the sounds rather than the traditional tone categories, then you have only four basic tones:

A mid level tone (1 - long and 8- short)
A mid rising tone (5)
A high rising/falling tone (depends on the speaker, 2 on the chart)
A low falling tone (3, and 4 - long and 7- short)

I feel I've missed something here....

As for the Sandhied versions. I would say,

3 ONLY becomes 1, never 2
4 only becomes 8, never 2 and the distinction in other Hokkien dialects between k,t,p endings and h ending no longer exists. I have had one person tell me that only people who speak Hokkien poorly say khe•h-lâng 客儂 as "Khé•h+lâng" whereas it should be "Khé•+lâng" (as in Taiwanese, with a slightly different vowel and sandhi) but I don't believe I have ever heard a native speaker of Penang Hokkien (even the one who told me) drop the glottal stop in a word like this unconsciously. The only places I have heard it dropped are chiá(h)-lát 食力 and lāu-joá(h) 鬧熱 which ends up sounding like lāu-joa instead.

Sometimes I also wonder whether people are distinguishing -h and -k at all.

[SimL]

Sometimes I also wonder whether people are distinguishing -h and -k at all.

In final position, I certainly distinguish them:

- ah4 (duck) vs. ak4 (to water plants)
- kah4 (to teach / instruct) vs. kak4 (horn, e.g. of a goat or cow)
- sek8 (ripe) vs. seh8 (to go riding around in a car)

Is this old-fashioned?

[Ah-bin]

Not at all Sim, I suppose I was thinking of how often people with no POJ (or other system) write -k for the -h ending, But this probably has more to do with Malay spelling than not making a distinction.

As for the change in tone on a words such as kah and bah, I think that is probably old-fashioned; i.e. used in Malaysia more by older native Hokkien speakers (from China) than younger native speakers of Penang-style Hokkien. The Wikipedia article should reflect this, I think.