Penang Hokkien

Discussions on the Hokkien (Minnan) language.

Penang Hokkien

Post by Sim » Fri Jan 30, 2004 4:50 pm

I thought I'd take the liberty of starting a new topic on Penang Hokkien.

One of the first points I wanted to post is a relatively minor one, but part 2 of it is slightly interesting.

1) (This one has already been pointed out) I say "ci" instead of "cit" in normal conversation for "one".

I think Niuc pointed this out to me. [ Niuc: I can't find where you pointed this out anymore, do you know where this was? I've looked many times! ]

I do this irrespective of the type of word which follows or the sound which follows (i.e. it is not context-sensitive).

... "ci pua*3" (half), "ci phun5" (bucket), "ci7 ban7" (thousand)
... "ci tui1" (heap), "ci thang2" (pail)
... "ci keng1" (room), "ci khang" (hole), "ci giah8" (an expression meaning "for show")*
... "ci mE*5" (night), "ci ni5" (year), "ci lui2" (flower)

*: "co3 ci7 giah8" (non-sandhi), literally "do a worthwhile-amount", which means (perhaps paradoxically), just doing something for show, without really meaning or wanting to.

2) (This one is more interesting). Instead of either "cit e5" or "ci e5" for "one (thing)", I say "ci le5", i.e. with an "L" inserted.

This is quite a strange irregularity, but I checked with my parents, and:
a) My father says everybody in Penang says this.
b) My mother confirms this by saying that when she first started learning Penang Hokkien, she remembers finding this unusual.

Perhaps Andrew could comment on both points? Is it just something we do only in my family, or have you heard other Penang Hokkien speakers do this?


Penang Hokkien - classifiers

Post by Sim » Fri Jan 30, 2004 5:11 pm

The next note in this thread is about the noun classifiers I use in Penang Hokkien.

I noticed that the use of classifiers can differ within the varieties of Hokkien, and I'm quite interested to know other classifiers (perhaps I forgot some, perhaps there are some I don't know of or don't use), *and* their usage (perhaps there are some listed below which are used differently by other readers of this forum).

My classifier for:
... animals (longish) "bue2"
... animals (roundish) "ciah4"
... books "pun2"
... clothing "tiau5"
... flowers "lui2"
... people "e5"
... plants "chang5"
... rooms "keng1"
... ropes and long objects "tiau5"
... round objects "liap1"
... sheets of paper "tiau*1"
... stick-like objects "ki1"
... vehicles (cars, boats, planes) "teng2"

... general objects "e5"


1. There are two classifiers for animals "bue2" and "ciah4". The former I suppose is the same word as "tail" (my ignorance of characters makes such assertions always a bit doubtful), and I think I would also use "bue2" for more "longish" animals, and "ciah4" for more "roundish" animals. For example, prawns, fishes, snakes, squids would be "bue2", and pigs, dogs, cows, birds, frogs, lizards, snails, crabs would be "ciah4". [ Could someone confirm if "bue2" as classifier is the same word as "tail"? ]

The idea that "bue2" might be for animals without legs and "ciah4" for those with looks ok, but is not quite accurate. For example, snails have no legs, but are still "ciah4". I think "longish" vs. "roundish" captures the distinction best. [ Prawns do have legs, but their legs are so tiny that they don't add much to the overall impression of a prawn, so I don't see this point as being against the "legs vs. no legs"-distinction. ]

One doubt I had was whether "ciah4" has an upper limit as to the size of the animal. Lions, tigers, even cows feel ok as "ciah4", but at the size of elephants and giraffes, I begin to wonder. This could just be my inexperience talking about elephants and giraffes in Hokkien.

Crocodiles make me hesitate between "bue2" and "ciah4", which also makes sense to me, as they are both a bit "longish" and a bit "roundish". [ We only use the Malay word "buaya" for crocodile, we pronounce it "bue3-ia2", so "ci bue bue-ia" sounds a bit strange anyway, because of the double "bue". ]

What classifier do people use for insects? I definitely say "ciah4" for cockroaches, grasshoppers, crickets, but I'm not sure about flies, spiders (not strictly an insect!). It's a bit frustrating that many insects are more "longish" than "roundish". Perhaps a better definition is then: bue2 = longish, no obvious legs, reasonably flexible; ciah4 = all other animals not requiring "bue2". How do worms fit in here? I think I would say "ciah" for worms, which contradicts all the attempted definitions so far.

2. The classifier "tiau5" for ropes and long objects seems to require that these objects be at least a little bit flexible. A rubber or plastic hose for watering plants or washing the car would easily be "tiau5". But a stiff plastic pipe (e.g. under the sink, or for protecting cable or wiring) would already be a bit doubtful for me, even though it bends a little. A stick would be getting a bit too inflexible for "tiau5" (but then, for sticks, there is "ki1"), while a lamp-post, definitely long, is definitely out, because it is not at all flexible.

What classifier would readers use for a bridge? Somehow "tiau5" seems to ring a vague bell here. They're certainly long, and early bridges might have been a bit flexible (thinking of rope bridges across canyons!).

The classifier "tiau5" seems to be shared between long objects and clothes. This makes some sense to me, as shirts and trousers and ties are longish and flexible. [ Are they the same character? ]

3. The classifiers "lui2", "chang5", "ki1" have already been discussed (at length!) in a previous topic.

4. The classifier "e5" seems to be both specifically for people, and also used as a general classifier for nouns not using the other classifiers. I believe this is similar to Mandarin "ge4".

5. I have not included a word like "te3" (a piece of something) as a classifier. In "cit3 te1 pO3" (a piece of cloth), "cit3 te1 ca5" (a piece of wood), "cit3 te1 tu3-bah3 (a piece of pork) etc, it seems to me more to be a sort of "quantifier", describing how much of an uncountable noun. The word "te3" is not used for a particularly specific subset of (uncountable) nouns, it can be used for any such nouns of which one can have a piece.

Similarly, I have not included a word like "tih4" (a drop) as a classifier. In "cit3 tih8 cui2" (a drop of water), "cit3 tih8 bak1" (a drop of ink), "cit3 tih8 hueh4" (a drop of blood) etc, again, it seems to describe how much of an uncountable noun. The word "tih4" is not used for a particularly specific subset of nouns, it can be used for any such nouns of which one can have a drop.

However, it could be argued that this is not a very consistent position (calling "bue2", "ciah4" etc classifiers, and not "te3", "tih4" etc). The concept of countable vs. uncountable nouns is valid for most European languages. With countable nouns, one can (generally) have a plural, and one just says: "one, two, three, etc <noun>(s)". With uncountable nouns, one can (generally) not have a plural, and one must say: "one, two, three, etc pieces, slices, fragments, drops, glasses, etc of <noun>". Some grammarians claim that *all* Chinese nouns are uncountable (pigs, snails, water, ink), and that classifiers are used to specify a "quantity" of the uncountable noun. In such an analysis, "cit ciah tu" (one ('animal of') pig) and "cit tih hue" (one (drop of) blood) are completely equivalent constructions.

This makes some sense when one looks for example at bananas. In my Hokkien, we say "ci keng1" for the whole huge thing which hangs from the banana tree. Each "keng1" can have several (8-20) "se1" attached radially to a central, vertical stem. I believe the English word for "se1" is "hand", one says: "one hand of bananas". Each "se1" then has several (8-16) individual bananas attached to it. For an individual banana I would say "tiau5". In such a situation, one can argue that "kin3 cio1" is just an uncountable noun, indicating the fruit, and that quantities are selected out "keng", "se", "tiau". [ BTW, can anyone confirm if "se1" is the same character as "comb" (for hair). The way the bananas are attached in one "se1" does make it look a bit like the teeth of a comb attached to a comb, and we call a comb a "ca3-se1" (wood comb?). ]

If one accepts the argument of these grammarians, then "te3" (piece) and "tih4" (drop) are also classifiers. And indeed, one can say that "te3" is used for firm but "separable" objects, of which one can take sub-portions to make a piece, and "tih4" is used for liquids, when one is referring to very tiny, separate portions, each with a roundish shape. In this sense, "tiau*1" (which I initially felt to be a proper classifier) fits in the same category as "te3" and "tih4", paper being uncountable in English anyway.

6. Lastly, I've always been curious about the phrase "cit, nO, sa* etc liu2 mi7-sua*3" (non-sandhi) or "cit, nO, sa* etc liu1 mi7-sua*3" (sandhi).

[ "mi7-sua*3" is (I believe) a special type of Hokkien noodle. It's made from rice, and hence white, but it's not the common white rice-noodles, "bi1-hun2". Instead, it's a lot less shiny and less transparent (actually, completely opaque). It's never fried, only ever cooked in soup. It can be had in soup just like any other Chinese noodle, but it is also used in a very specific (and delicious!) dish called "ah8-thui2 mi7-sua*3" (duck drumstick mi-sua*). It has the additional characteristic that it should be eaten quickly after it is cooked, otherwise it swells rapidly, soaking up the soup, and turning the whole dish into an unpleasant soggy muck. ]

Now, "mi7-sua*3" is usually (was usually? I haven't bought or eaten it in years) sold in cardboard boxes, and the noodle was "pre-packaged" in a special way: there were individual "loops" in each cardboard box, say 12-24 "loops" in each box. very neatly packed. Each loop sort of looked like a bow-tie, or an "infinity symbol": two ellipse-shaped bits with a thin "waist" at the centre, tied with a little pink thread.

When my mother was making these for us for lunch, she would ask each one of us: "<X>, lu bueh ci liu O nO liu", "<Y>, you want ci liu O nO liu", in order to find out how hungry we were and how much we wanted to eat. [ Notice that she would say either Hokkien "lu bueh" or (Malaysian) English "you want" at random, but she would only ever use English "O" (=or), never "a3". ]

Anyway, to get back to the original point(!), the "classifier" for "mi7-sua*3" was "liu2". As far as I know, this is not used for any other noun. Now, I've always had a private theory about this (based totally on speculation). The word for "button" (as in for clothing) in Hokkien is "liu1-a2", and old-fashioned Chinese buttons used to look something like "mi7-sua*3". Or rather, the individual loops of "mi7-sua*3" in the box looked a bit like old-fashioned Chinese buttons. These buttons were made of about 6-8 loops around a central axis. I have always suspected that the "liu" of "mi7-sua*3" was related to the "liu" of "liu-a". Can anyone confirm or refute this?


Re: Penang Hokkien

Post by Niuc » Sat Jan 31, 2004 7:15 am


Actually what I noticed is not "one" (一 'cit8') but "this" (這 'cit4'). It's in ... 225&t=1225 (About "bochap").

Interestingly sometimes I hear some people say 'cit8 le5' too, although mostly say 一個 'cit8 e5'. May be the "L" sound is inserted to smoothen the pronunciation.


Re: Penang Hokkien

Post by kaiah » Sat Jan 31, 2004 8:23 am

Hi, everybody,
In our usage, sometimes -t would assimilate the next vowel into L- consonant.
We say:
cit8 le5 (one)
cit4 le5 (this one); hit4 le5 (that one)
sit8 la2 (wing)
pat8 la2 (guava)
thuat4 li1 bu2 脫衣舞 (striptease)


Re: Penang Hokkien

Post by Niuc » Tue Feb 03, 2004 4:05 pm


Your exposition about classifiers used in Penang Hokkien is very impressive. I really enjoy reading & learning from it. It also prompts me to think about their usage in my accent.

Almost all of the classifiers mentioned are identical in mine (with some phonology differences). For me, 個 'e5' is also the general classifier. I would say that if someone doesn't know the classifier for an object, it's safe to use 'e5'. It may sounds funny or awkward but not totally wrong.

For me, 隻 'cia4' is the common classifier for animal and 尾'ber2' is its subset (as 'cia4' itself is a subset of 'e5'). 'ber2' indeed means "tail". While I use both in the same way as you do, I don't think that it's wrong to use 'cia4' for fish, prawn, snake etc. It may sound a bit unfit but still ok. For elephant (象 'chiu*7', 大肥象 'tua7 pui5 chiu*7') and giraffe (長頭鹿 'tng5 thau5 lok8', some say should be 長頸鹿 'tng5 kun2 lok8'), I definitely use 'cia4'. For crocodile (we call it 'buai3-a0') and worms, I think both 'cia4' and 'bue2' can be used.

Beside 條 'tiau5', other classifiers for clothes are 領 'nia2' and 'su1' (what's the character?). I usually use 'nia2'. I also heard of 套 'tho3' but may be it's of Mandarin influence.

For plastic and iron pipes, I am not sure too. May be either 條 'tiau5', 支 'ki1' or 管 'kong2' can be used. Kaiah or others, please help.

For bridge, I heard of 'tiau5' and 座 'co7'. 'co7' is usually used for big construction.

Your 弓 'keng1'/'king1' and 條 'tiau5' for banana are identical with mine. Although I also call "banana" 'kin1 cio1', actually it comes from 弓蕉 'king1 cio1', i.e. banana that resembles a bow. Your 'se1' is 'pi5' in mine. I don't know the character for 'pi5'. Your 'se1' is very likely the same character for "comb" 梳. It's 'sue1' in my accent, 'sue1' tends to mean "to comb" in traditional way. Nowadays we usually say 'lua8' for "to comb", 'luai3-a0' for "comb". I heard some Taiwanese said 'lua3-a0' for comb.

I don't know whether 麵線 'mi7 sua*3' is originated from Hokkien but indeed it's very important in Hokkien tradition. It's a necessity in traditional birthday meal. Our classifier for 'mi7 sua*3' is 結 'kat4' ("knot").

Kaiah, do you also call guava 'nai5 a8 puat8' beside 'pat8 la2'?


Re: Penang Hokkien

Post by kaiah » Thu Feb 05, 2004 4:14 am

Hi, Niuc,
My father call guava as "nio'2 puat8 a2(la2)"(Tainan city accent, 台南市腔), and my mother says "niau2 puat8 a2"(Tainan county accent, 台南縣腔). "Pat8 a2" or "puat8 a2"(菝仔) is popular accent in Taiwan. We usually call guava in Mandarin as "ba1 le4"(芭樂) which is translated from "pat8 la2".

According to limkianhui from Amoy, he said:
na5-a2-put8(Amoy); lap8-pat8-a2/pat8-a2(Ciangciu)


Re: Penang Hokkien

Post by Sim » Thu Feb 05, 2004 8:43 am

Hi Kaiah,

Well, in my variant, we use only the Malay word "jambu", pronounced "jam1 bu3" (with sandhi), so, with no non-native Hokkien sounds, it's a totally integrated word.

I have a comical (and slightly crude) story from my childhod to tell about this word.

As regular readers of this forum may remember, my maternal relatives, speak a more Amoy-type Hokkien, and (like Kaiah) my maternal grandmother's word for it was "na5 put8" (with sandhi). As a child, this struck me as a really strange word, and whenever she said it, I would reply: "na5 pang1-sai2 na5 put8" (as you ****, you fart).

Of course, "put8" is not really the Hokkien word for "fart", but was meant by me to be the imitation of the sound of someone passing gas. As a child, this seemed to be hilariously funny.

[ I have to confess that even now, it still strikes me as funny, which is why I tell it here. Apologies for anyone who might be offended by this posting. I read somewhere that Chinese, in particular Hokkiens, are very down-to-earth people, and bodily functions and sexual matters are talked about quite plainly, or humourously. I have to say, that this matches my experience. ]


Re: Penang Hokkien

Post by Niuc » Fri Feb 06, 2004 6:43 pm

Hi Kaiah,

Thanks a lot for reminding me about the discussion on the other forum. Beside 'nai5 a8 puat8', some of us call it 'ni5 a8 puat8'. 'na', 'nai', 'ni', 'nio', niau' should be all accent variations. Oh now I see that actually Mandarin "ba1 le4"(芭樂) is derived from Taiwanese "pat8 a2"(菝仔). No wonder my friends from Mainland China don't know 芭樂, since they call it 番石榴 (fan1shi2liu2).


In Indonesian we call guava as "jambu batu" or "jambu biji". "Jambu" only is another fruit, it's call "wax jambu" in Florida Agriculture site . In Hokkien we call this fruit 'liam2 bu7'. I saw that in Taiwanese TV it was written as 蓮霧.

Hokkien people are indeed very down-to-earth. Until now I still don't understand why the word you wrote as **** above is a kind of taboo (or at least, vulgar) in English. The word 放屎 'pang3 sai2' is a common word in Hokkien. Of course the term 大便 'tai7 pian3' is more refined and official, but I don't think the former one as being vulgar. Nonetheless there are indeed a lot of vulgar words, mainly sexual terms used in scolding, in "market/rude" Hokkien. This kind of misuse shouldn't be allowed. Yet the correct usage of terms such as 'pang3 sai2' shouldn't be deemed as vulgarity either.

Andrew Yong

Re: Penang Hokkien

Post by Andrew Yong » Sat Feb 07, 2004 1:43 am

Incidentally, how do you write e5 as in goa2 e5, i1 e5, lu2 e5, etc.? Is it 個, 其, 之 or 的?


Re: Penang Hokkien

Post by Niuc » Wed Feb 11, 2004 4:03 pm

Hi Andrew,

I used to think that 'e5' in 'gua2 e5', 'i1 e5' 的 (de5 in Mandarin); while 'e5' in 'cit8 e5' as 個 (ge4 in Mandarin). 廈門方言詞典 uses these two characters. Interestingly, Douglas' 廈英大辭典 suggests that both 'e5' are actually one and derived from 其 ('ki5' literary pronunciation, qi5 in Mandarin). I came to know this because of your question. Thank you.

Are there any other "evidences"/"references" that 其 is the original character for 'e5'?

Andrew Yong

Re: Penang Hokkien

Post by Andrew Yong » Wed Feb 11, 2004 11:46 pm

I dislike the word 的 because there is some evidence to show that it was invented to represent the mutated Beijing pronunciation of 之, and that the two words are originally one and the same.

The problem with 其 is that it is not widely used. I don't know whether it is correct to use or not. In written Mandarin it means 他的, so using it to mean 的 looks strange. My personal preference is 個 because it is used in other dialects like Hakka and Shanghainese. It covers both e5's and you can treat le5 and ge5 as spoken variants of 個.

Re: Penang Hokkien

Post by hong » Thu Feb 12, 2004 2:33 am

see Prof Ang web site on e,he,ge discussion page 4 out of 27 pages now.
Andrew Yong

Re: Penang Hokkien

Post by Andrew Yong » Thu Feb 12, 2004 12:32 pm

I typed a longer reply but it got lost when I was trying to post

Basically he says current usage is 個 for ko3/kO2 and 嘎 for e5, etc. He says the latter also used in Hakka and Shanghainese. Cantonese uses two separate characters.

I don't like all these phonetically inspired characters, 嘎, &#22021;, etc but ko3 and e5 seem to different for both to be the same character, though he doesn't say anything about it. He deals with 的, but not with 其. I suspect the Revd Dr Douglas was looking for a character in the 5th tone that had vaguely the same meaning as 個.

Strangely elsewhere on the site he says we should use 儂 for lang5 (person) to distinguish between jin5 and lang5, and that it is the actual character of lang5. In my dictionary it is listed as [nong2], Shanghainese for 'you'.

Re: Penang Hokkien

Post by Niuc » Fri Feb 13, 2004 4:30 pm

In 我的 'gua2 e5' (my), the word before 的 'e5' retains its tone. While in 一個 'cit8 e5' (one), the word before 個 'e5' shifts its tone. Is this because they are different 'e5'?

個 is 'ko3' in 一個月 'cit8 ko3 ge8' (one month), 兩個月 'nng7 ko3 ge8 (two months), 個人 'ko3 jin5' (individual), etc. But I think that using 個 as 'e5' is quite ok, better than 嘎, &#22021;. I have no idea whether 儂 is the original character for 'lang5', but personally I think that 人 is the character for both 'lang5' and 'jin5'.

Meanwhile, using 的 (literary pronunciation 'tik4') to indicate possesive form 'e5' seems ok, until we know its original character. I don't know whether 之 (l.p. 'ci1') has something to do with this 'e5' but it's interesting to know that in Japanese Hiragana "e" looks like 之.


Re: Penang Hokkien

Post by kaiah » Sat Feb 14, 2004 5:32 am

Hi, everybody,
I know Prof. Ang often uses the word "个"(simplified "個") for "e5" in his writtings, but not "嘎". I cannot see the hanji "e5" in his website but see a blank space. Hakka language also use the same word "个" or 表音字 "介" for their "的/個".

Both "e5" for "的" or "e5" for "個" are the same way in our usage whether original tone or sandhi tone. In our usage, "我" before "我的"(gua2 e5) and "一" before "一個"(cit8 e5) both shift their tones.
e.g. (in tone pitch)
gua53 + e24 => gua44 e24
cit4 + e24=> cit1 e24

Fuzhou language just uses "其" for "的", refer to the articles(pdf file):
福州方言的“其”和“過” ... TUSETQTRTS
福州方言的結構助詞及其相關的句法結構 ... TQTSTQTQTX

article about Hokkien "e5":
閩南方言的結構助詞 ... TQTSTQTQTW
Download "Adobe Reader" from this website to read pdf files:
and "Asian font packs" to read Chinese simplified: ... tpack.html