too many romanization systems for Taiwanese in textbooks, say critics

Groups favoring the promotion of Taiwanese (also called Hokkien, Hoklo, Minnan, and all sorts of other names) are urging Taiwan’s Ministry of Education to come to a consensus on what phonetic system should be used to teach it, according to an article in the Taipei Times.

The article, however, is a bit confused in its facts, so I’m presenting it for what it’s worth and hoping commenters here can help set matters straight. Certainly, though, the variety of systems — and for some of these I use the word loosely — being used in textbooks is a hindrance to effective teaching of the language.

Some might be surprised to know that Tongyong Pinyin is not Taiwan’s official romanization system for Taiwanese, though its schemes were approved for Mandarin and Hakka. I tend to think of Tongyong for Mandarin as more of a nationalist marketing scheme than a romanization. I know there are more than a few people who hold the same view of Tongyong for Taiwanese.

As long as I’m on the subject, I’d like to remind people that, despite the misinformation that has been spread by some of its supporters, Tongyong is not one system that covers all the languages of Taiwan.

Native-language interest groups yesterday urged the Ministry of Education to speed up the creation of a unified phonetic system for the Hoklo language (commonly known as Taiwanese) to preserve Taiwanese culture.

Lee Shang-hsien (???), director of the Taiwan Pik Hap Cultural Association which promotes the Hoklo language, said that different versions of elementary school Hoklo textbooks used three different phonetic systems.

At the moment, Hoklo text books use either Tongyong pinyin, the Hanyu Romanization system, or the Taiwanese Language Phonetic Alphabet (TLPA).

Lee said that the three systems did not reflect regional Hoklo accents, and added that the Hoklo language could not be preserved and passed on to the next generation without a unified phonetic system.

Language teacher Su Huang-hung (???) said that Hoklo originally had eight tones, but due to the similarity of the second and sixth tones, most phonetic systems regarded Hoklo as a seven-tone language.

Using the Tongyong pinyin system reduces Hoklo to only five tones, Su said.

In addition, Hoklo pronunciation changes according to syntactic context, she added.

However, Tongyong pinyin marks these cases as variant pronunciations, creating a situation in which many students do not know the original pronunciation of the words they study, Su said.

Hanyu pinyin and the TLPA are almost entirely identical, since the TLPA system was derived from Hanyu pinyin by the Ministry of Education many years ago, said Lu Ching-ching (???), a member of the ministry’s Native Language Promotion Committee.

Lu said there were similarities among the three systems, and that all three have advantages and disadvantages.

The committee’s main problem was to reach a consensus and decide on one system or to incorporate the three systems into one, she said.

Discussions are still taking place at the ministry on unifying the phonetic system, Lu added.

Huang Yu-chi (???), director of the Taiwan Association for Students Abroad, said that many overseas Taiwanese return during summer and winter vacations and hope to learn their mother tongue, but the lack of a unified phonetic system makes it difficult for them to do so.

Other representatives said that not only students, but also foreigners and foreign spouses need to learn Hoklo and are in need of a unified system.

source: Culture groups urge ministry to expedite Hoklo writing rules, Taipei Times, February 23, 2006

8 thoughts on “too many romanization systems for Taiwanese in textbooks, say critics

  1. This is scary. They might come up with yet another system. Then again hanyu pinyin is only for mandarin. What happend to the international phonetic alphabet? Taiwan is probably the country with the most systems availiable, but also probably the country with the most diverse minorities that try to write their dialects…

  2. Isn’t there any mention of the ‘church’ orthography?

    I don’t know how accurate it is phonemically, but it seems to be adequate enough and have enough tradition (and publishing history) behind it to make it the front runner.

  3. I agree with the author’s view on Tongyong, but there is one canard that gets brought up time and again when talking about Taiwanese romanization which needs to be explained. One of the complaints against Tongyong mentioned in the article is that it reduces Taiwanese’s seven tones to five.

    This is brought about by a misunderstanding both of the tones in Taiwanese and how Tongyong represents them. Taiwanese in the textbook has seven tones, true. But it only has five tone contours:

    Ranging from high 1 to low 5, the tone contours are as follows (obsolete tone 6 omitted):

    1. 1-1
    2. 1-3
    3. 3-5
    4. 3-4 with a stop (h, k, p, t)
    5. 3-4-3
    7. 3-3
    8. 3-3 with a stop (h, k, p, t)

    So essentially tones 4 and 8 follow the same tone contours as tones 3 and 7 respectively, except that they are shorter and end with a stop consonant. One might not consider them seperate tones at all, because the stop is not part of the tonal contour. Because in the orthography (Tongyong and POJ [church]) these stop consonants are represented by a letter on the end of the word, only five tone distinctions are really needed. Tone 4 can use the same diacritic as tone 3, with h, k, p or t on the end of the consonant, and tone 8 can likewise use the same marks as tone 7, with h, k, p or t on the end.

    So even if there really are seven tones in Taiwanese (hmm, I’m not convinced), then Tongyong simply represents them using a combination of diacritics and orthography (stop consonants).

    This argument is reminiscent of a lot of the specious nonsense bandied around about Hanyu vs Tongyong for Mandarin by people who don’t really know what they’re talking about. Tongyong is a bad system, but not because it is ‘stealing’ Taiwanese tones.

    The second point is to answer Michael Farris’ question about the phonemic accuracy of POJ.

    POJ has a problem phonemically in that it uses ‘ch’, ‘chh’ and ‘s’ to represent two sounds each (much like Tongyong does for Mandarin). To use rough equivalents in (Mandarin) Hanyu Pinyin, ‘ch’ represents both ‘z’ and ‘q’, ‘chh’ would be ‘c’ and ‘j’ and ‘s’ for ‘s’ and ‘x’. Also the ending ‘eng’ in POJ is phonemically inaccurate and in HP would be more like ‘ieng’ (sic). Having said all that, POJ does have some literature behind it and more of a history. Also, the problems I have quoted do not impede comprehension, they simply force the learner to acquire more rules and exceptions than are necessary.

    Personally I can use the POJ but in practice I use my own modified Hanyu Pinyin system that takes account of all the phonemic distinctions of Taiwanese. I doubt I’d do anyone a favour by adding another system into the running though!

  4. Taffy:

    You noted that “POJ has a problem phonemically in that it uses ‘ch’, ‘chh’ and ’s’ to represent two sounds each” (emphasis added). In fact the “two sounds” represented by each POJ letter (e.g. ch) are in complementary distribution. i.e. they belong to the same phoneme. It would not be phonemic to explicitly spell out these sound-pairs (though it would be phonetic).

  5. BTW, the authors of this paper found some sixty (!) Hoklo romanization schemes ca. 1999, and this was no doubt an underestimate. Most of these, of course, have had little or no following. Therefore I consider having only three major competitors to be ground for guarded optimism.

    Why so many schemes? Vanity is one factor, I am sure. But we should also consider the existence of a political atmosphere which cast a long shadow over open public discourse on non-Mandarin/anti-Mandarin advocacy. The infrastructure for discourse just wasn’t that solid, and enthusiasts were not always aware of each other’s work, and when they did, some wisely agreed to develop consensus schemes. One group, for example, now support the proposed Tongyong scheme for Hoklo rather than their own similar versions.

  6. Begging your pardon, a-giâu, you are right. I should have written ‘phonetic’ and not ‘phonemic’. An equivalent situation in Mandarin is nicely illustrated by the use of ‘s’ in Tongyong pinyin representing Hanyu Pinyin ‘x’ and ‘s’. While they may be complementary, they differ phonetically, so for pedagogic ease and clarity I would rather see them separated. Again though, I think POJ probably qualifies as the ‘most popular system’ and so should certainly be considered as a front-runner for an ‘official’, even though I dislike it! Plus it has the benefit of the zh-min-nan Wikipedia behind it. By the way, I assume you are the same a-giâu who lurks around the Taiwanese articles on Wikipedia – if so, please keep it up, it’s nice to see informed people commenting on the subject.

  7. Taffy,

    I agree that it’s misleading to say that there are 7 “tones” in Holo. Technically, they’re “tonemes.” Technically–and this is still a bit of a judgment call–we could also say that traditional tones 7 and 8 are one toneme, and traditional 3 and 4 are one toneme, with different tone contours depending on whether a syllable is unchecked, -p/t/k-checked, or glottal-checked, PLUS its position in the sandhi group.

    Technicalities aside, I’d venture to say that the phonemic/tonemic approach to writing would come more natural to native speakers of Holo, even if it didn’t do much for adult learners like myself. (Not to mention, for example, that the phones represented by POJ /s/ are in free variation even preceding /i/.)

    Like you, though, I would support either one of the front-running romanization schemes for the sake of pushing Holo onto the front page in Taioan and southeastern Asia.

  8. I think what this article refers to as “Hanyu Pinyin” is actually “Church Romanization”. Because TLPA is most definitely based on the church system and not on Hanyu Pinyin.

Leave a Reply

Your email address will not be published. Required fields are marked *