Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Discussions on the Hokkien (Minnan) language.
Abun
Posts: 115
Joined: Fri Jun 21, 2013 4:15 pm

Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by Abun »

Hello everyone,

I’ve been looking a bit into Medhursts old dictionary, only to realize that although my Taiwanese tends to Chiang-chiu like pronunciation, I have only a very limited knowledge about the sounds in actual Chiang-chiu.
The 15 initials posed no problem, this much I know at least and after all, Medhurst’s transcription doesn’t differ much from later POJ in that point :D However when trying to get through Medhurst’s s system for the finals, I encountered some problems. As most of you probably know, Medhurst’s dictionary is based on the Si̍p-ngó͘ Im and so is his system for the finals, i.e. he named the finals after their respective characters in the Si̍p-ngó͘ Im. He also didn’t seem to have cared about creating a phonetically consistent system but simply tried to find a distinct transcription for each of the finals. While it was rather easy to figure out the POJ for most of these, there were a few which left me kind of baffled, even after reading his explanation and seeing examples, so I wanted to ask your opinions:

Final no. 22 兼 “-ëem/-ëep” clearly stands for POJ “iam,” however both the transcription and Medhurst’s description of the sound indicate that the final sounded rather like [iem] (I’m guessing similar to the way POJ “ian/iat” sound, but with a bilabial in the end. I have never heard of this phenomenon, neither in discussions about the peculiarities of Chiang-chiu variants nor in any account of other variants, ancient or modern. Can anyone confirm that this final is indeed pronounced with an [e] or [ɛ] sound in Chiang-chiu? Or did the sound change over time maybe and is now “iam”?

Final no. 5 嘉 “ay” definitely stands for the open [ɛ] knew existed in Chiang-chiu (rhyming with 假、嫁、下). However, Medhurst lists another final, 伽 (no. 39) which he transcribes in the same way and describes as very similar to 嘉 but slightly distinct (it is of course distinct from POJ “e” (稽 “ey”) as well). 嘉 he describes as the vowel in “care, bear, wear ect.” Since Medhurst was an Englishman, this translates to [ɛ] in my opinion. 伽 he describes as the vowel in “fate, gray, may.” From the perspective of contemporary English, I would transcribe this as [ei, ɛi] or possibly [ɑi], which sounds very different from [ɛ] to me, so I’m wondering whether Medhurst’s mid-19th-century English had a different sound value for these words. The Si̍p-ngó͘ Im lists as rhymes for 伽 only three characters: 莢, 瘸 and a third character which I can’t decipher. 伽 I can only find as “ka,” 莢 I find as “ngeh” (“ngoeh” in Choân-chiu) and 瘸 as “khôe” (“khê” in Choân-chiu). Douglas lists “ngɛh, a pod of grass” (being 莢), but nothing with “ɛ” which would fit 伽 or 瘸. So from todays perspective this final seems to have split and merged with at least two other finals, Chiang-chiu “-ɛ”/Choân-chiu “-oe” and Chiang-chiu “-oe”/Choân-chiu “-e.” Can anybody confirm that?

The situation with the o’s also leaves me a little confused. I seem to remember to once have heard from somebody that traditional Chiang-chiu doesn’t have the open “o͘” (the one with the dot), but I wasn’t sure about that, so I approached the matter with no real expectation as to the existence of this vowel. Indeed, the Si̍p-ngó͘ Im (and therefore Medhurst) distinguish two kinds of o’s: 沽 “oe” (no. 11) and 高 “o” (no. 15). The former rhymes with syllables like 古、湖 and so on, which I would all pronounce with the open “o͘,” while the latter corresponds to the closed one. Medhurst describes the vowel of 沽 as rhyming with English “toe” but “with a full mouth.” The vowel in “toe” (again from contemporary English perspective) is a diphthong to me, something like [oʊ], but a different way of pronouncing the same phoneme doesn’t seem all that odd to me. However, the list includes nasal variations for both of the two o’s: 姑 “ⁿoe” (no. 41) and 扛 “ⁿo” (no. 49). My vocabulary still only contains a very limited number of syllables with a nasalized o, most of which have a nasalized initial (such as the literary readings of 五 and 腦; the only one without a nasal initial that I can think of is the sentence final --honnh), but I had always lived under the impression the nasalized o’s in Minnan usually become open by default, so there would only be open nasal o’s anyway. Does that mean that this is a more recent development and the nasalized open and closed o were still distinct at least in early to mid-19th-century Chiang-chiu? What makes the matter even more confusing to me, is the fact that neither 姑 nor 扛 don’t seem to have any kind of nasalized o in other dictionaries I looked them up in (and actually such a vowel would surprise me especially in the case of 姑 which doesn’t have either a nasalized vowel or a nasal initial or final in any variant of Hokkien I know of, nor in Mandarin).

I actually don’t have any concrete questions about these finals (apart from whether or not “iam” is indeed pronounced as [iem] in Chiang-chiu), they have just surprised me, especially 伽 and the two forms of nasalized o left me wondering whether Chiang-chiu and Choan-chiu were really as phonetically close as I thought (which wasn’t actually that close to begin with, but I had always thought it possible to break the differences down to a few basic rules). So, if any of you has more insight to provide, I’m happy to learn :)
amhoanna
Posts: 912
Joined: Sat Sep 18, 2010 12:43 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by amhoanna »

I can only come at this synchronically. There are dialects either in the south of Ciangciu or in 海陸豐 where even today "our" "open e" corresponds to /ei/, and our "open o" corresponds to /ou/.
Abun
Posts: 115
Joined: Fri Jun 21, 2013 4:15 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by Abun »

amhoanna wrote:I can only come at this synchronically.
Yes, that's one of my problems, too. I believe I know a bit about contemporary Hokkien phonology but that of course doesn't mean the situation was the same when the Si̍p-ngó͘ Im and Medhurst's dictionary were composed. In fact, some of the finals listed there strongly suggest that there were probably two phonemical distinctions which I didn't know existed: the two kinds of nasalized "o" (扛 vs. 姑; /õ/ vs. /õʊ̃/?) and the two kinds of "open e" (嘉 and 伽). Judging by Medhurst's explanations and the fact that he didn't incorporate the distinction in his transcription, I suspect the latter distinction was on the verge of disappearing during the early 1800s, at least in the Batavia region.
The problem is complicated by the fact that Medhurst's comparisons to English pose a second stage for diachronical language change and I can't fully rely on my knowledge of contemporary English to interpret them.
amhoanna wrote:There are dialects either in the south of Ciangciu or in 海陸豐 where even today "our" "open e" corresponds to /ei/, and our "open o" corresponds to /ou/.
Interesting. With this information it's even more obvious that 沽 (/oʊ/?) is in fact the same phoneme as the "open o," just pronounced in a different way. /ei/ may be the final represented in the Si̍p-ngó͘ Im as 伽, seeing as Medhurst describes it with the English words “fate, gray, may.” Do you know whether "our" "open e" is fully replaced with [ei] in those regions or just in some words (in other words, is it "just" another realization of [ɛ] or are they in phonemical opposition)?

P.S.: Just stumbled upon the appendix in Douglas' dictionary where he elaborates on both the orthographical differences with Medhurst and the the phonological differences between the various dialects in the Hok-kiàn region. Douglas confirms that Mh. <ay> corresponds to Dg. <ɛ>. However he doesn't bring up either the second <ay> (伽) or the distinction between <ⁿoe> and <ⁿo> (he identifies the latter one as <ɵ˙ⁿ> in his own orthography but doesn't mention the former). This is peculiar because according to my impression, Douglas seems to have had a very good ear for small differences in pronunciation (although he apparently wasn't as good at identifying certain pronunciations as merely different representations of the same phoneme). Is it maybe also possible that Medhurst, taking the Si̍p-ngó͘ Im as a basis, was a victim of placebo effect and identified distinctions which weren't really there in the contemporary dialect of Batavia? On the other hand, Douglas also doesn't mention the fact that his <ɵ˙> (modern POJ <o͘>) is pronounced as [ou] in certain areas, so maybe his ear wasn't that good after all? Or did he simply never have the opportunity to talk to someone from those areas?
amhoanna
Posts: 912
Joined: Sat Sep 18, 2010 12:43 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by amhoanna »

Yeah, should be safe to say that the two resources were based on different dialects.

The following is based on data I put together a few months ago. I "processed" data from eight dialects:

潮阳棉城 潮州潮安 澄海澄城 詔安城関 海豐県城 龍海石碼 金門金沙 泉州当代字彙

Take for example the 馬 vowel. Only 龍海 and 詔安 actually have [ɛ]. In 詔安 there's free variation with [ie]. All other dialects have [e], in most cases merged with other rymes as well.

Take also the 街 vowel. In urban 泉州 and 金門 it's [oe], merged into other rymes, whereas in all 潮州 dialects (I use this term in the traditional, i.e. SE Asian sense, to exclude 海豐) it's [oi]. In 龍海 it's merged as part of [e], but in 詔安 海豐 it's [ei].

Now let's take the 好 vowel and the 烏 vowel.

The 好 vowel is [o] in 龍海 金門 泉州 海豐, but [ɔ] in 詔安 and all of 潮州.

烏, meanwhile, is [ou] in 詔安 海豐 and all of 潮州, but [ɔ] in 龍海 and points north and east.

Many interesting conclusions can be drawn. From a pan-dialectal POV, there is no /ɔ/ rime; but most dialects have one as a reflex of another rime. 詔安 "preserves" [o] and [ou] and does not have [ɔ].

詔安 is an interesting case. Just in terms of phonology, 詔安 is more Teochew than 海豐 is, while 海豐 is more Hokkien than 詔安.

The Medhurst dialect has a distinction between [ɛ] and [ei], apparently. Modern-day 詔安 has the same. It seems very possible to me that this 詔安-type of dialect was more prevalent in the past, and that the 龍海-type dialects that we know and love are the result of levelling that took place under intense contact with Amoy, Coanciu, etc.

We've discussed the fact that Penang Hokkien is derived from a Ciangciu dialect that is almost a transition to 同安. This dialect was then subject to heavy Coanciu and Teochew influence in the diaspora.
Abun
Posts: 115
Joined: Fri Jun 21, 2013 4:15 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by Abun »

Very interesting, I would be very interested in broad-scale studies there. I have read dialect studies over various Chinese languages, but so far I haven't been able to find one covering different dialects of Hokkien. Have you been doing serious research there or did you note those differences in passing?
amhoanna wrote:The Medhurst dialect has a distinction between [ɛ] and [ei], apparently.
I expressed that unclearly I guess. Medhurst doesn't distinguish [ɛ] and [ei], they're both <ay> in his transcription. However in his introduction, when he explains his transcription, he does so on the basis of the 十五音 rhymes. Since the 十五音 distinguishes [ɛ] and [ei], Medhurst also explains how he percieves their respective pronunciations, but also says the difference is only very slight and ultimately doesn't incorporate it in his transcription. As far as I see it, this leaves us with two possible conclusions: Either the distinction was just barely there in contemporary Melaka Hokkien but on the verge of disappearing, or Medhurst was a victim of placebo effect and tried too hard to identify a distinction which wasn't actually there in the dialect he was observing.
amhoanna
Posts: 912
Joined: Sat Sep 18, 2010 12:43 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by amhoanna »

I carried out an amateur research project in "cross-dialectal comparative Hokkien-Teochew etymo-phonology", for the purpose of creating a cross-dialectal system of kana to be used together with kanji for writing Hoklo.

Conclusion was that the system would be so complex that only linguists and kids would be able to learn it.

不こ我無放棄、最近我れ寫ち款兮閩南字、語理箍仔外兮大众ま卡閒接受〔卡け接受、khah kē ciapsiū〕。後擺当然愛掛調号。
AndrewAndrew
Posts: 174
Joined: Mon Aug 09, 2010 10:26 am

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by AndrewAndrew »

amhoanna wrote:I carried out an amateur research project in "cross-dialectal comparative Hokkien-Teochew etymo-phonology", for the purpose of creating a cross-dialectal system of kana to be used together with kanji for writing Hoklo.

Conclusion was that the system would be so complex that only linguists and kids would be able to learn it.

不こ我無放棄、最近我れ寫ち款兮閩南字、語理箍仔外兮大众ま卡閒接受〔卡け接受、khah kē ciapsiū〕。後擺当然愛掛調号。
It has been tried before for Mandarin, after May 4th movement, but Mandarin phonology is much more complex, than Japanese (which has a very simple phonology), and would require memorising around 400 kana, and Hokkien would be even more complex.
Abun
Posts: 115
Joined: Fri Jun 21, 2013 4:15 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by Abun »

AndrewAndrew wrote:
amhoanna wrote:It has been tried before for Mandarin, after May 4th movement, but Mandarin phonology is much more complex, than Japanese (which has a very simple phonology), and would require memorising around 400 kana, and Hokkien would be even more complex.
And that figure, I believe, already excludes the tone, so if we want a distinct kana for every syllable in every tone that this syllable occurs in, I guess we would need a lot more than 1000 (albeit not 4*400=1600 because some syllables only occur in certain tones). I don't know the figures for Hokkien but it seems to me as well that the number of used syllables will be somewhat higher than in Mandarin (though maybe not as much as one might think because I believe Mandarin has 18 phonemic initials if I'm not counting wrong, compared to Hokkien's 14-15), even without the tones.
amhoanna
Posts: 912
Joined: Sat Sep 18, 2010 12:43 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by amhoanna »

so if we want a distinct kana for every syllable
Not necessary.
in every tone
Also not necessary, although it would be nice.

Most of the high-frequency kanji-less syllables in Hoklo follow a C-V(-ʔ) (consonant followed by vowel and sometimes a glottal stop) syllable structure.

There are a lot of people who write "about" Hoklo, mostly in Mandarin, also in English, Japanese, Siamese, etc. But very few people really write "in" Hoklo across a wide variety of topics. And writing "in" Hoklo, esp. in non-technical usage, is what it takes to realize the shortcomings of the three or four modes that the "market" favors at this point.
Abun
Posts: 115
Joined: Fri Jun 21, 2013 4:15 pm

Re: Medhurst, Si̍p-ngó͘ Im and Chiang-chiu Phonology

Post by Abun »

amhoanna wrote:Most of the high-frequency kanji-less syllables in Hoklo follow a C-V(-ʔ) (consonant followed by vowel and sometimes a glottal stop) syllable structure.
I had never thought about that but now that I think of it, there really are rather few with a (non-glottal-stop) final. Chit and hit come to my mind of course, but most indeed seem to be either open syllables or glottal-stop-final.
amhoanna wrote:There are a lot of people who write "about" Hoklo, mostly in Mandarin, also in English, Japanese, Siamese, etc. But very few people really write "in" Hoklo across a wide variety of topics. And writing "in" Hoklo, esp. in non-technical usage, is what it takes to realize the shortcomings of the three or four modes that the "market" favors at this point.
Very true, and that is pretty much why I feel it may be counterproductive to invent more and more new systems instead of deciding on one existing system, flawed though it may be. Trying to look for 本字 is a slightly different topic if you ask me; you can do that purely out of etymological interest or even to suggest it for a future spelling reform. I can go and research the origin of the Mandarin word which is nowadays written as 這 as well and maybe the standard will be changed if I find an answer (and it is also practical to use). But for the time being, I think it would be best to just decide on one standard. This way, you can finally go and teach Hokkiens how to write their language without requiring the reader to put so much effort into it that many people are just too lazy to even try. This is also, as I mentioned before, why I use the MoE characters whenever there is one, even though from an etymological point of view I am very unhappy with quite a number of them.
Locked