My Taiwanese (Hokkien) adventure. From zero to fluency.

[mandarynski]

Oh sooo pumped up about my Skype chat tomorrow afternoon!

Today, I did the minimal pairs from the Maryknoll book followed by more practise from the Spoken Hokkien SOAS course. I am very pleased I could get the Glossika course. It looks slightly overwhelming at the moment but I couldn't wait to explore the package and listen to a few sentences. I'm really grateful for the IPA transcription. All in all, I am sure it's going to work out for me, I have used very similar methods before with some results. For now, I'm going to concentrate on the tones and sounds ( a couple of consonants are giving me grief! ) and building up a vocabulary base. To wind down, I am going to listen to this lady http://youtu.be/YnhcE7Jxr24

http://youtu.be/YnhcE7Jxr24 especially as, according to my Taiwanese speaking friend, her pronunciation is excellent. I also enjoy following the lyrics KARAOKE style.

Bye for now

[mandarynski]

Day 4.

Well, the Skype chat went better than I expected (by a combination of Hokkien, Mandarin, POJ, pinyin and Hanzi).

Got a bit side-tracked later by a particularly active thread here, regarding the Glossika stuff. I got the fluency course but did not really get into it yet. Tomorrow, I'll build up my list of very simple expressions and continue my tone exercises. Today I practised with the recordings. It's amazing how much you can get away with on public transport here these days. Twenty five years ago people would have thought you were nuts, talking to yourself. Nowadays, they just assume you're on your mobile, talking in some strange language.

[mandarynski]

Day 6.

Getting into the swing of things.

In my eclectic approach, I have been following advice from lots of other members of this forum and other accomplished polyglots with a view to being able to progress rapidly in my core skills. So far I have been concentrating on practising the sounds and tones and building up basic vocabulary. Over the last six days I have had a few opportunities to chat online and got to say a few words and received feedback. I keep doing the daily routine from Taiwanese book 1 and Spoken Hokkien and adding to my base of simple phrases enhancing my vocabulary, using frequency lists and phrasebook sentences. I am also working on my back story, as Benny Lewis likes to call it. Another thing I found useful is how Luca Lampariello (mentioned in a book by Benny Lewis) approaches intonation "like the network that holds a sentence together". I find it very useful in my practice as I go through the phrases rather than isolated words as soon as I develop a reasonable understanding of how the sounds and tones go together. I have searched for some resources on line and found a few interesting websites such as this thread on Forumosa: http://www.forumosa.com/taiwan/viewtopic.php?f=40&t=120290 with links to interesting approach for teaching tones with music and videos of Harvard Taiwanese 101. I also found this online dictionary very useful: http://jimbu.bricksquare.com/dictionary/

Another useful website: http://www.reddit.com/r/ohtaigi/ where I found information about two Anki decks, among other things. I'm going to continue with my plan to get as much exposure to spoken Min Nan as possible and will "officially" start the Glossika course this Tuesday as part of my daily routine.

[mandarynski]

Day 7

Wow! A week already!

Today, I started on the Anki decks. One for Harvard Taiwanese 101 and the other one from here:

轻松学闽南语

Very happy they both come with audio. The other part is doing lots and lots of drills from Taiwanese book 1 by Maryknoll. Came across a blog post earlier on today by a guy learning Thai, doing 90 minutes of pronunciation drills every day. He is a Hero! I could do it but not in one sitting.

Tomorrow is a big day; starting the Glossika S.Min GMS and GSR. First going through the 1000 sentences in book 1 in a single sitting, need to put some time aside for that for sure. Will continue with the other two textbooks for grammar etc.

Hope to be more detailed in this thread in the near future as my experience grows to give you guys more of an idea about the resources that I'm using.

[Kobo-Daishi]

I also found this online dictionary very useful: http://jimbu.bricksq...com/dictionary/

 

 

I found that site a while back, but, thought it wasn't working. Seems I was using it wrong. I would put something into the input box and hit "enter", which just reset the dictionary. So I thought they were still in beta and hadn't got it working yet.

 

Now I know to hit the search button.

 

Anyway, it appears the content is from an old Maryknoll Taiwanese Minnan dictionary that Maryknoll recently put out as an open source spreadsheet. And it appears the TED site has added definitions from the CC-CEDICT as well.

 

But be aware that the Chinese characters might not match up to the Minnan POJ pronunciations as the Chinese is the standard written Chinese (Mandarin) for the Minnan POJ words.

 

Also, there seems to be a bug with their search algorithm.

 

When I enter "hat", I get this.

 

 

But when I enter "帽", I get this.

 

 

The word "帽子" isn't in the first search even though the definition "cap, hat" includes "hat".

 

I specifically looked up "hat" because when I was testing out the dictionary, I was comparing it to a spreadsheet file where I had tried to add Chinese characters matching the POJ pronunciation Romanization found in the Maryknoll spreadsheet.

 

Really difficult when you don't know the language. Gave up after a while.  

 

Come to think of it. The first search, the one for "hat" didn't even return an entry for "hat". The closest thing is "hat worn by officials".

 

Kobo.

[Michaelyus]

@ Kobo-Daishi,

 

Evidently the search on that website seems to match strictly from the beginning of the entry. Needs some work.

[geraldc]

Any progress?

[ParkeNYU]

Mike Campbell is definitely the guy to talk to about the Min-nan language. I've had a few chats with him and he's very knowledgeable and helpful.

 

One day, I too would like to learn 'Taiwanese', but first I am determined to learn Middle Chinese so that I can have a strong foundation. I wish you luck, as the Min languages are the most tonally challenging of all (because of tone sandhi).

[Shelley]

Mandarin has tone sandhi.

[ParkeNYU]

Most of the tone sandhi in Mandarin can be reduced to:

 

1) consecutive third tones

2) 一

3) 不

 

For the Xiamen/Taiwan dialect of Min-nan, however, tone sandhi looks like this:

 

[Lu]

Parke is right. Mandarin has a few tone sandhi, but in Minnanese (Taiwanese), every syllable has sandhi, unless for some reason it doesn't.

[ParkeNYU]

This is why I love Middle Chinese.

 

I hope your next adventure is Shanghainese! Tones will be a breeze.

[Takeshi]

Not sure if this link only magically works for me because I'm in a university, but if anyone wondered for what reasons a syllable wouldn't have tone sandhi, here is the classic source:

 

http://journals.cambridge.org/download.php?file=/2371_4D78A2BF2C656D561B1E87E5C169DAFA_journals__PHO_PHO4_01_S0952675700000798a.pdf&cover=Y&code=2dcbcfb99d53eb3c7005a5003ed21b36

[AMHOANNA]

I'm a semi-pro "Hoklologist", if You will. I started learning Hokkien as a very young adult. Today, many plateaus later, I speak it at a near-native level. I was on the engines looking for a "Grammar of Hokkien" type of PDF in English — since new things pop up sometimes — and happened across this thread and felt I had to jump in and comment, either for the benefit of the OP or whoever else might happen across this thread.

 

Now, a list of things to keep in mind:

 

1. B1 in Taiwanese Hokkien in just a few months is very ambitious. (I wonder if the OP made it?) The language is tough on adult learners. However, there is one class of speedy adult learner in Taiwan at present: Vietnamese people. There was also a class of speedy adult learners in the mid 20th century: native speakers of coastal SE Chinese languages such as Teochew and Hokchiu (Foochow). Coincidental? I think not.

 

1A. The dialects (emphasis on the plural) of Hokkien spoken in M'sia and Singapore are easier for adult learners, for reasons I can get into if anybody's interested. BTW Amoy Hokkien is just like Taiwanese. But the Hokkien spoken in most of the rest of Hokkienland (southern Hokkien, or Fujian in Mandarin) can be pretty different, lexically and phonologically.

 

1B. The connection between Vietnamese and Hokkien may seem cryptic. I started on Vietnamese when I was already conversational in Hokkien. I was surprised to recognize a lot of Hokkien-esque words, structures and semantics in it. Strangely, my Hokkien improved a lot while I was living in Vietnam and never speaking it. I'd say Hokkien is intermediate in many ways between Vietnamese and Mandarin. I came to Hokkien with a strong Mandophone background. The Vietnamese helped "balance out" the Mandarin. Another way to see it is that Vietnamese is structurally very close to Classical Chinese. Another way to see it is that Vietnamese, Hokkien and Classical Chinese are all part of the Mainland SE Asia Sprachbund, whereas Mandarin (at least the kind most of us know) has grown away from it over the centuries.

 

2. Hokkien is way different from both Mandarin and Cantonese. This caught me off guard back in the day. The concept of "the Chinese dialects" is deeply ingrained in the minds of Chinese as well as non-Chinese people. For the most part, the concept fits reality pretty well. If we take the languages of people who identify as Chinese, for the most part they're very close to each other when it comes to grammar and lexicon. Ninety % or more of the "Han" part of China is (was) covered by Chinese languages that use(d) pretty much the same structures and the same words with apparently very similar semantics. Learn one and the rest are child's play. Even "white collar" Cantonese is deeply similar to Mandarin in spite of the surface differences. BUT ... there is that less than 10% of the map that's covered by wildly divergent languages. And Hokkien is part of that. Other big-name players are Teochew and Hoisan.

 

2A. Armed with the assumption that Hokkien and Mandarin were just different-sounding versions of each other, I jumped into my Hokkien learning by "mapping" Mandarin to Hokkien phrase by phrase and trying to pass that off as Hokkien. I was doing the same with Cantonese at the same time. Cantonese speakers found my Cantarin quite acceptable, but Hokkien speakers would just look at me all weird and say (usually in Mandarin) something like, "Wait. You don't speak Hokkien, do You? Why don't You just speak Mandarin?" Sometimes they were pretty hostile, but While mileage would most likely vary. The weirdness went away as my Hokkien became more idiomatic. With Cantonese, though, the "community" allowed me to fake it the whole time till I made it. With Hokkien, I had to "make it" first behind the scenes and resurface as a finished product. 

 

2B. To recap: a good base in Mandarin will help w/ Hokkien (or Japanese or almost any mainland E/SE Asian language), but not nearly as much as You'd probably expect. The grammatical, lexical and semantic "divergence" (historically, probably better understood as a "lack of convergence") between Hokkien and Mandarin goes way beyond what most learners would expect. (However, Hokkien and Cantonese semantics seem to line up very well, so someone w/ a good base in Cantonese has that in their favor.)

 

3. A lot of the difficulty in learning Hokkien springs from the lack of good learning materials. Quantity seems to breed quality over time, but Hokkien learning materials are pretty sparse compared to what You'd find for, say, Thai or Korean or even Cantonese.

 

3A. Beyond this, a lot of newer materials made in the last 30 or 40 years seem to approach Hokkien not on its own terms but rather as "a dialect of Chinese" and therefore some kind of retro offshoot of Mandarin, as if that makes any sense. This is OK for getting to know the aspects of Hokkien that are Mandarin-like, but You end up not seeing the aspects of Hokkien that are not. This is also a problem in newer materials for Cantonese.

 

3B. On another level — and older materials also have this problem somewhat — Hokkien learning materials focus too much on nouns and verbs, and too little on grammar, morphology (which supposedly doesn't exist), grammatical words like kā and hō͘ and sian and kiàn, sentence particles, etc.

 

​3C. With the exception of very old materials, Hokkien learning materials tend to do a poor job of accounting for dialect variation and lower registers. Expressions like "thong sèkài" (THE WHOLE WORLD) or "thōng hó" (BEST) will not be found in any Hokkien textbook. The reasons for this are probably too complex to get into casually. It's not just that they leave out low-frequency usages. A lot of high-frequency usages get left out too. As a rule, the better a word or structure corresponds to something in Mandarin, the more likely it is to show up in textbooks and dictionaries. And vice versa.

 

 

4. Another tough aspect of learning Hokkien is the sociolinguistics of it. The ecology of Hokkien took some big hits starting in the 1960s, when nationalism swept through Southeast Asia. (Not that colonialism was good.) Then, in the 70s and 80s, the Republic of China government and then Singapore legislated to push Hokkien — the majority language in both places — out of middle class public life. In the 90s, officials moved to enforce the speaking of Mandarin in all public contexts (and mass media) in Hokkienland itself (in China) while flooding the cities with non-Hokkien-speaking migrant workers. Today, heritage speakers are often not native speakers, and native speakers even up to the age of 60 often speak Hokkien poorly. Educated urban folks under 50 are most likely to speak Hokkien badly. Unfortunately for the future of Hokkien, uneducated and rural folks (understandably) want to be like those educated urban folks in almost every way. And unfortunately for most polyglot would-be Hoklophones, the Hokkien speakers we tend to have access to are educated urban folks under 50 who mostly speak a shallow, watered-down version of Hokkien and only use it in a narrow range of contexts.

 

4B. This is a challenge in terms of access, but it also helps kill learner motivation.

 

5. I'll close, if this can be called that, by recommending the Maryknoll textbooks, Books 1, 2 & 3. That is basically how Taiwanese Hokkien is spoken today by people whose primary language is Hokkien, not Mandarin. The Harvard text seems quite good as well. The Bodman text is good too, but not Taiwan-oriented.

 

5A. The Maryknoll dictionary, on the other hand, is bookish and very limited. You're better served using the ENGLISH-AMOY dictionaries from 100 years ago, found online here: http://minhakka.ling.sinica.edu.tw/bkg/chong-su-tian.php?gi_gian=eng.

 

5B. The Glossika program is poorly executed. Most of the sentences are not idiomatic (i.e. not how an everyday Hokkien speaker would say it), and a large number of the sentences are flat-out wrong. The lapse is truly bizarre given that Glossika Spanish seemed so professionally done. My guess would be that Campbell put the Hokkien program together himself based on his extensive knowledge of Mandarin, then hired an educated urban native Hokkien speaker with very (apparently, very) limited Hokkien as a consultant. B/c of the ecology, any Hokkien learner (or speaker) may encounter speakers of such corrupted Hokkien in real life at some point. But there is no point in starting out this way. Even today there are still young people in Taiwan — mostly working-class kids — who speak deep, idiomatic Hokkien. So let's not kid ourselves: Mandarized Hokkien is not the new Hokkien.

 

5C. If You like the massive input approach (and I do) and You know either Mandarin or Japanese, You can use these materials put together by a gentleman from outside Amoy: http://hokkienese.com/?p=715

 

Best wishes to everyone in their Hokkien-learning and Hoklological endeavors.

[ParkeNYU]

I was under the impression that Mike Campbell knew Hokkien the best; it's a bit ironic that his Mandarin is purportedly better.

[AMHOANNA]

From what little I've heard of it, Campbell's Mandarin is amazingly good. But his Hokkien just makes You think, "He must speak real good Mandarin." He speaks Hokkien with Mandarin vowels, Mandarin tone contours, Mandarin cadences, Mandarin vocabulary, Mandarin structures and idioms. Coming from a linguist of his caliber, it would seem he's doing it on purpose, whatever that purpose may be.

[davoosh]

@AMHOANNA, thanks, that was a very interesting read.

 

Regarding Mandarin diverging from the SE Asia Sprachbund, do you really think it's diverged so far that Vietnamese can be included and Mandarin left out? I think Mandarin is still firmly rooted within that sprachbund, although it might be a bit strange phonologically (but so was Classical Chinese, based on reconstructions.)

 

Although "white collar" Cantonese may be more similar to Mandarin than Hokkien, could it be that this is because Cantonese has more of a tradition of reading 'baihua' texts using Canto-readings (and thus reinforcing Mandaring-like structures)? Very colloquial, 'non-white-collar' Cantonese can be very different and I've found it often uses Canto-specific colloquial expressions (similar to things like 'thong sèkài' that don't appear often in learning materials. I believe the same thing is happening to Shanghainese, in that it's becoming a lot like Mandarinised-Shanghaihua.

 

I've also read that Min dialects are supposedly the only Chinese languages that don't descend from Middle Chinese (presumably they descend from something earlier)? Perhaps this is another reason why it seems so different from other Chinese languages.

[ParkeNYU]

All major Chinese language branches have two strata of character readings: colloquial (白讀) and literary (文讀). The latter form, in all cases, descends directly from the Late Middle Chinese of the Tang-Song period, including the Min branch. However, the colloquial stratum of Min was indeed thought to predate Middle Chinese altogether, which I find a bit dubious, since many of the colloquial Min readings overlap rather well with those of Early Middle Chinese (Northern/Southern-Sui-Tang).

[AMHOANNA]

@davoosh

 

Thanks. Glad You enjoyed it.

 

Re: the Mainland SEA Sprachbund, Vietnamese is to the Sprachbund what Germany is to the EU!   Sharp minds may differ on who's in and who's out, but Tai-Kadai, Hmong-Mien, and Viet-Muong are clearly core.

 

There's a useful paper out there "left-braining" the Sprachbund and assessing where it ends. The title and author slip my mind, but it's on the web. An esp. insightful paper to read is Blench's "Ethnographic and archaeological correlates for an MSEA linguistic area".

 

«Very colloquial, 'non-white-collar' Cantonese can be very different and I've found it often uses Canto-specific colloquial expressions (similar to things like 'thong sèkài' that don't appear often in learning materials.»

 

Yes. Probably left out unintentionally, subconsciously and for sociolinguistic reasons.

 

«Although "white collar" Cantonese may be more similar to Mandarin than Hokkien, could it be that this is because Cantonese has more of a tradition of reading 'baihua' texts using Canto-readings (and thus reinforcing Mandaring-like structures)?»

 

I'd say this is a secondary factor, b/c the effects on Hokkien vs Cantonese have likely been generally comparable, with a slight "edge" to Cantonese … till the last 50 years or so.

 

What happened is that w/ the advent of all-Mandarin classrooms in Taiwan and "the Malay countries" in the mid 20th cen., the skill of voicing kanji in Hokkien (and other South Chinese languages) was abruptly lost. Not only were the Hokkien readings not taught, but an ideology was introduced where kanji & Mandarin are siamesed to each other. In China & the Philippines this started in the late 20th. Today, the average middle-aged Taiwanese person is not only unable to voice kanji in Hokkien, but unable to grasp the concept. They think kanji belong to Mandarin, w/ the uncontemplated exception of when they're used for Japanese.

 

The silver lining is that the inability to voice kanji in Hokkien insulates Taiwanese and South Seas Hokkien from this path to Mandarization. No such luck in China. Paradoxically, though, middle-aged China Hoklophones probably speak better Hokkien than Taiwanese and South Malaysian Hoklophones of the same vintage. China fell off a cliff starting with the generation now in their 30s, though. Young China Hoklophones can voice Mandarin text msgs in Hoklo. That's better than using straight Mandarin, I guess. The downside is that when they talk, they sound like they're reading off a Mandarin text. I haven't spent much time among them but this seems to be the case.

 

The MAIN factor, though, is the flow of people induced by the lay of the land. The yellow brick road from the Yangtze Valley into the tropics runs along the Xiang River of Hunan into the West-Pearl (西江、珠江) drainage basin. This explains the Sinicism of Cantonese, esp. learned Cantonese, vs the coastal tongues both up and down the coast. So Pearl River Cantonese is kind of a Mandaristic intrusion on the coast. (Now, in this context, “Mandarin” has a broader and more “diachronic” meaning than it did in the last paragraph.) It might be the most Mandarized coastal language south of Ningbo. (Then again, inland valley languages in South China tend to be more Mandarized than coastal ones. Even the seaward dialects of Hakka are less Mandarized than the inland ones...)

 

The coast from east of Hong Kong all the way up to southern Zhejiang has always been last to plug into whatever new grid was being rolled out from "the heartland". This shows up in a million facets, not least linguistic. Again, counter-intuitively, the inland side of what became the coastal provinces have tended to plug in before the coast. At times the coastal tribes even go and plug themselves into overseas things. The Hoklos are also often the last to unplug from old, failing grids. The non-convergence (usually thought of, possibly incorrectly, as divergence) of the Hoklo tongues and the political status of Taiwan can both be understood from this angle, among others.

 

«I believe the same thing is happening to Shanghainese, in that it's becoming a lot like Mandarinised-Shanghaihua.»

 

I believe so.

 

«I've also read that Min dialects are supposedly the only Chinese languages that don't descend from Middle Chinese (presumably they descend from something earlier)?»

 

There may be a grain of truth in this. But the model we've been working with where the South Chinese languages descended and diverged from various past “central Chineses” is probably no match for millennia of raw reality. This ties in to a re-working of the basic paradigms of historical linguistics, going "back" to Indo-European linguistics (the 中原 of modern historical linguistics  ). 

 

Even under the 20th century paradigm, though, there are other Chinese languages that don't “descend from Middle Chinese”. These tend to be isolated languages spoken in the hills of northern Kwongtung, southern Hunan, etc. Piu (標話), for instance. But only the Hoklo tongues have sea views, millions of speakers, and international connections.

 

Now Vietnamese has much the same kind of relationship w/ MC that Hokkien does; and in fact both have something that Teochew doesn't, namely a full set of MC-based literary readings. Interesting to consider in light of the fact that in the late 10th century when the proto-Vietnamese seized on the fall of the Tang to establish their “pattern of independence”, much (even most) of what became the Hoklo and Hakka homelands was still beyond the pale, i.e. there were people there ... but they weren't “citizens”. ;)

 

Most of us were lulled into thinking that Pearl River Cantonese was somehow the end of the line, and everything else must be either more Mandarish than it, or off the grid. The reality is way more interesting.

 

 

@ParkeNYU

 

A worthy inquiry.

 

«All major Chinese language branches have two strata of character readings: colloquial (白讀) and literary (文讀).»

 

This seems to be true for the "wok" cultures of Canton and northern Zhejiang, but by time we get “out” to Hoklo or Vietnamese, we may find ourselves looking at three or more strata, w/ the "internal architecture" increasingly unclear.

 

The kanji 下, for example, maps uncontroversially to four etyma in Hokkien: kē, ē, hē, hā — all much used, non-obscure, and non-interchangeable. The kanji 方 maps uncontroversially to png, hng, & hong, w/ possible ties to yet other etyma.

[ParkeNYU]

Could it be that both the colloquial and literary strata sometimes contain multiple readings each? The Sino-Japanese stratum has four subdivisions of readings (呉音, 漢音, 唐音, and 慣音 or 'customary'), while the native Japanese stratum has no limit.