Unified romanization for all Taiwanese native languages possible?

I decided to have a separate discussion from the “Waishen” thread, because I think there’s a lot of bickering there, and I want a thread where people can explore the possibility of having a unified romanization for all languages native to Taiwan together.

I don’t know if we can get a simple and unified system out of this, I really don’t. I just think it’s better for Taiwan, and people who are interest in Taiwan and learning Taiwan’s languages if such a system existed. That’s why I am sharing my progress with everyone, hopefully discussion will create a better system, or even declare such a system impossible.

I’m been compiling a chart of phonology of all Taiwan’s languages. So far, I’ve only completed the task for Holo, Haka, and Mandarin. I’ll fill in Aboriginal languages as we go along. I left many TUR fields blank (with green flood-fill), because I am not certain what should be used. However, I would like to see a complete list, and that list will make deciding easier.

I’ve notice there are notable differences from how I say 4 tones and what PY teaches 4 tones. If I’m close to the Taiwan standard, that means Taiwanese speak 4 tones a bit differently than the Chinese. Namely, Taiwanese speak in a lower pitch with a small range, and the 3rd tone for standard Taiwanese Mandarin is 11, at most 112 or 212…

As for Holo, there are two sources of tonal numbers that I’ve found. They are basically the same, except one has a wider range, 5 to 1, as opposed to 3 to 1. I think that really depends on the people. In any case, tones are relative to each other for a speaker, so as long as a speaker can separate the tones within his/her range, it doesn’t really matter about the absolute values.

The Hakka tones is trickier, but I think it would be relatively simple if we simple separate Si-yen from Hoi-liuk.

My thought is to separate the tones into 13 tones, at most 14 if we add a special tone for Mandarin’s 3rd tone.

33, 23, 11, 31, 12, 22, 24, 41, 44, 42, 14
4ʔ high-stop, 2ʔ low-stop

Revise that, maybe we can further simplify it to 9 tones. It can be mapped to all Taiwanese native Sinitic languages.
TUR, Holo, PY, Hakka
high-high ◌ , ◌ 1 陰平 , ◌̄ 1, ◌ 陰去 (四縣) / ◌̀ 陽平 (海陸)
high-low ◌́ , ◌́ 2 陰上, ◌̀ 4, ◌́ 陰上 陽上 (四縣)
low-low ◌̀ , ◌̀ 3 陰去, ◌̌ 3, ◌̀ 陽平 (四縣) / ◌̊ 陽去 (海陸)
low-stop ◌ʔ , ◌ʔ 4 陰入, , ◌ 陰入 (四縣) / ◌̍ 陽入 (海陸)
low-mid ◌̂ , ◌̂ 5 陽平, , ◌́ 陰上 陽上 (海陸)
mid-high ◌̋ , ◌̋ 9, ◌́ 2, ◌̂ 陰平 (四縣)
mid-low ◌̌ , ◌̌ 6 陽上, , ◌̂ 陰平 (海陸)
mid-mid ◌̄ , ◌̄ 7 陽去, , ◌ 陰去 (海陸)
high-stop ◌̍ʔ , ◌̍ 8 陽入, ◌ 0, ◌̍ 陽入 ( 四縣) N/M 陰入 (海陸)

Here’s the link to my google spreadsheet:
docs.google.com/spreadsheets/d/ … sp=sharing

I haven’t looked in detail yet, but I think it’s folly to take a single Romanization for both Hakka-Taigi-Mandarin, all Chinese languages, and indigenous languages, all Austronesian languages. The difference in consonants, vowels, and syllable-building is so great you might as well work at a unified transcription for Chinese and Arabic. The only realistic option is IPA or a simplified IPA.

My recommendation: One unified system for indigenous languages, and one unified system for Sinitic languages.

[quote=“Hokwongwei”]I haven’t looked in detail yet, but I think it’s folly to take a single Romanization for both Hakka-Taigi-Mandarin, all Chinese languages, and indigenous languages, all Austronesian languages. The difference in consonants, vowels, and syllable-building is so great you might as well work at a unified transcription for Chinese and Arabic. The only realistic option is IPA or a simplified IPA.

My recommendation: One unified system for indigenous languages, and one unified system for Sinitic languages.[/quote]

I’m aiming for a simplified IPA, and swapping out the non-latin symbols for combination of not-used latin alphabets.

Although your suggestion to separate Sinitic from Austronesian makes sense. If when the table is completed things look irreconcilable, that’s the first fall-back position.

What is your design philosophy?

  1. Only latin letters if possible
  2. Avoid diacritics aside from tones as much as possible
  3. As close to IPA as possible
  4. When IPA symbol for a phone is non-latin, use a combination of latin letters to represent it when possible
  5. Stay as close to Tailo as possible, although oo and o in Tailo should probably be changed
  6. If Austronesian languages also lacks tɕ and ʨʰ, represent them with tsi and tshi to simplify the system
  1. Only latin letters if possible
  2. Avoid diacritics aside from tones as much as possible
  3. As close to IPA as possible
  4. When IPA symbol for a phone is non-latin, use a combination of latin letters to represent it when possible
  5. Stay as close to Tailo as possible, although oo and o in Tailo should probably be changed
  6. If Austronesian languages also lacks tɕ and ʨʰ, represent them with tsi and tshi to simplify the system[/quote]

If you’d allow me, these are design techniques, not design philosophy.

I was looking for:
What are the goals? What effects do you want to achieve? What’s wrong with the current situation? How do you envision your design to be used? Why do people want to use your design? What are the impacts it will have?

[quote=“sofun”]

I was looking for:
What are the goals? What effects do you want to achieve? What’s wrong with the current situation? How do you envision your design to be used? Why do people want to use your design? What are the impacts it will have?[/quote]

My goals were described in the first post of this thread as well as in the waishen thread.

forumosa.com/taiwan/viewtopi … l#p1621919

I want a romanization system to not only be useful to non-natives, but also Taiwanese people of all ethnicities. I envision it to allow Holo, Hakka, Mandarin and Aboriginal speakers to be able to effortlessly read and pronounce things in each other’s languages, with hopes of encouraging cross ethnicity understanding, easier to comprehend other native languages, and adopt words and terms from other languages.

Non-native speakers will be able to learn one romanization and apply it to all native languages. Road signs could use this romanization and no longer be bound to Hanji. Instead of Beinan for 卑南 on street signs, we will have a system to spell Puyuma and other aboriginal names. Aboriginals can have their own naming rights back. They can have just their aboriginal names on their ID cards, and never have to worry about people not being able to pronounce it correctly.

At the same time , it will allow native Taiwanese to have a easier time learning foreign languages, since the romanization system they are familiar with is closely bound to IPA.

My ultimate goal is shed Taiwan of it’s Han Chinese centric mindset when it comes to government, naming, and basically all things, and have true equal opportunity for all ethnicities and languages. I’d like to see Taigi, Hakka and every single Aboriginal language listed as an official language of Taiwan and have Taiwan be a truly multicultural nation.

If, and I mean if, except for Mandarin, all the other languages have become non-working, then theoretically, the goals you listed above can still be fulfilled after you’ve come up with a design, right? I mean, one can read the road signs and namesakes, and perhaps sign a few songs.

Don’t get me wrong, I also believe that namesakes and name right are absolutely critical, no doubt about it. But I do not see your goals being set to promote fluency and literacy of each individual language, to become a full-fledged, working language. It seems that your aim is to promote lateral “understanding” only. It has the appearance of equalizing, but does not address the problems currently plaguing the development and survival of each language.

Is this something that perhaps you could put more emphasis on? Or am I missing something?

[quote=“sofun”]If, and I mean if, except for Mandarin, all the other languages have become non-working, then theoretically, the goals you listed above can still be fulfilled after you’ve come up with a design, right? I mean, one can read the road signs and namesakes, and perhaps sign a few songs.

Don’t get me wrong. namesakes are absolutely critical, no doubt about it. But I do not see your goals being set to promote fluency and literacy of each individual language to become full-fledged working languages. It seems that your aim is to promote lateral “understanding” only.

Is this something that perhaps you could put more emphasis on? Or am I missing something?[/quote]

The fluency and literacy of each individual languages is a task that each family, community, society, ethnicity group and the government has to work out. However, making the society less Han Chinese biased can encourage the speaker of each individual languages to learn and teach their own languages. The best way to go is for the government to give each language official status and allow people to take tests, fill government applications in each language. In that case, having a common romanization system also helps.

Tell you what. If there is a symbolic tone to all this effort, the most critical one is for this ROC administration to provide synchronized translation services in broadcasting. These services should be provided in Parliamentary sessions, official announcements, and officially sponsored events.

As of today, the only politician in Taiwan that consistently delivers speeches in none-Mandarin is Lee Teng-Hui.

I don’t know about Hakka, but if no one speaks it in formal settings, in public, in the context of public affairs or politics, then I’m not sure about its survival.

I’d suggest you map out all the phonemes, not just the consonants and vowels.

You want the entire tables to see where the “empty cells” are and why they are empty. Then you know what seems to be two different consonants can actually be merged into one, in the written form. Things like that.

It is a daunting task if you’re tackling multiple languages at the same time. But this is what you choose to do.

[quote=“sofun”]I’d suggest you map out all the phonemes, not just the consonants and vowels.
[/quote]

Agreed. That will be my next step. Although while I can probably list out all the phonemes of Taigi, Hakka and Mandarin, I’ll need serious help with Aboriginal languages…

Maybe it’s a good time to start learning the aboriginal languages.

facebook.com/pages/說賽德克語/427909957322125

Mgeela ini knkela, snpusan mtluhe!
むげえら いに ぐんけら すんぶさん むづるへ

[quote=“sofun”]Mgeela ini knkela, snpusan mtluhe!
むげえら いに ぐんげら すんぶさん むづるへ
[/quote]

Again, kana is terrible at annotating Aboriginal languages as well. How would anyone reading that know how to pronounce Mgeela from むげえら(Muge era), or knkela from ぐんげら (Gun-ge-ra), or snpusan from すんぶさん(Sun bu-san), or mtluhe from むづるへ(Mudzuruhe)…

Heck, you even have げ for both ge and ke

I’m going to depend on these two documents for phonology of Aboriginal languages:

Dr. Ross’s “In Defense of Nuclear Austronesian” 2012
ling.sinica.edu.tw/files/pub … 6_1278.pdf

MOE’s “Wirting system of Aboriginal Languages”
edu.tw/userfiles/url/2012112 … iginal.pdf

[quote=“hansioux”][quote=“sofun”]Mgeela ini knkela, snpusan mtluhe!
むげえら いに ぐんけら すんぶさん むづるへ
[/quote]How would anyone reading that know how to pronounce Mgeela from むげえら(Muge era)?[/quote]

0:11.

There are so many extra syllables in the slowed down version I doubt it would be comprehensible when strung together by a non-native speaker. All this will do is make a bunch of people speaking Seediq with heavy Japanese accents.

It will make a bunch of people speaking Seediq, that’s true.

With youtube virtually for free, the language will take care of its own accent. The language is good.

Using kana for Seediq honestly makes no more sense than using kana for English. But all EFL students have perfect American or British accents when they speak because YouTube, right?