Foochow Romanized

Learn a Foochow Idiom a Day

呆牛呆馬搦去刣, 呆儂無法惹伊才.
Ngài-ngù ngài-mā niĕh kó̤ tài, ngài-nè̤ng mò̤-huák niā ĭ cài.
You can butcher bad cattle and horses, but you can do nothing about bad men.

In Israel, a Rabbi often tells his young pupils who are introduced to Hebrew letters: “Yes, they dance!” When Foochowese letters dance in front of you, you will know you have studied a concept well.

Alphabet

We begin our lesson with the overview of the Foochowese alphabet. Including two consonantal digraphs (CH and NG) considered as single letters, Foochowese alphabet has 23 letters totally, viz.:

Vowels

A, A̤, E, E̤, I, O, O̤, U and Ṳ are the nine vowels of Foochowese alphabet. Like German umlauts, Foochow Romanized also mutates vowels by placing two dots - but below rather than above. Special attention should be paid to E and O because they are actually diphthongs: E is [ei] and O [ou]. While the other seven are all monophthongs. An English speaker might find E̤ and Ṳ hard to pronounce. E̤ is much like German ö in hören, and Ṳ is the French u in chute or German ü as in früher.

Consonants

The consonantal system of Foochow Romanized is relatively simple, as you might have inferred from the absence of letters f, j, q, r, v, w, x, y and z in the alphabet. The plosive B/P, D/T and G/K, as well as the lateral and nasal L, M and N, can all be treated as their counterparts in English. Note that, however, the pronunciation of CH is not to be confused with that of the English ch. Foochowese CH is nearly like German z as in Zeit. Also, the letter NG might cause difficulty as most Western languages do not use [ŋ] as an initial consonant (neither does Mandarin Chinese). By saying rapidly “singer” and then omitting the first two letters the correct sound is approximated. Although S is sometimes pronounced as [θ] or [f] in casual speech, only [s] is regarded as the standard sound.

The letter NG, when used as a coda (a coda is a final consonant), is just pronounced the same way as in English. In fact, NG is the only nasal final in Fuzhou dialect. Therefore, 心 (heart), 新 (new) and 星 (star) are all pronounced as sĭng.

Characters that fall into the category of entering tones (入声) end with a final stop K or H, like 白 (băh, white) and 血 (háik, blood). Here K and H mark the abrupt termination - or more specifically, the glottal stop [ʔ] - of its preceding vowel. They reason for using two letters to represent one phoneme is that they stem from two distinct origins. In modern Fuzhou dialect, the contrast between the final K and H in ertering tones can still be observable when characters are spoken in connected speech, where the two types of finals may have different effects on sandhi.

Tones

Like all Chinese languages, Fuzhou dialect is a tonal language. In the Book of Eight Tones (戚林八音), Fuzhou dialect is described as having eight tones. But in practical use, only seven tones exist actually, each marked by a certain type of diacritic in Foochow Romanized.

The tonal mark ˘ as used in prmary smooth tone (上平) and secondary entering tone (下入) is sometimes omitted.

Initials and rimes

In order to form a systematic understanding of Foochow Romanized, you need to adopt a traditional study method: regarding a Chinese character as combination of an initial (声母) and a rime (韵母).

Initials

Initials are consonants that are always placed in the initial part of a character. Not every character has an initial, such as 河 (ò̤, river), 红 (è̤ng, red) and 鸭 (ák, duck). Theoretically, these characters are considered as having a null initial. So, taking the null initial into account, Fuzhou dialect has 15 initials.

Rimes

Rimes are vowels with or without a coda. In this regard, all Foochowese rimes can be organized into three categories: 1) rimes without a coda, 2) rimes with the coda H, and 3) rimes with the coda NG or K. But before we go further, I need you to be aware of the concept of “close/open rimes” first. In the following tables you can find some geminate rimes that are separated by a slash: they are called a pair of “close/open rimes”. We’ll talk more about his feature later.

Rimes without a coda

Rimes with the coda H

Rimes with the coda NG or K

Here I expect you to notice some striking characteristics of Foochow Romanized from the four rime tables above.

1. As I mentioned earlier, E and O are diphthongs representing [ei] and [ou]. But when combined with other vowel letters, they become monophthongs [e] and [o]. For example, IE is [ie] and UO is [uo].
2. When I is followed by O, it turns into [y].
3. When E̤ is followed by the coda NG or K, it turns into a diphthong [øy]. Thus E̤NG should be read as [øyŋ].
4. You might have noticed that E̤ṲNG also stands for [øyŋ]. So why are there two forms of transcription for [øyŋ]? Because E̤NG and E̤ṲNG come from different origins: E̤NG is a close rime while E̤ṲNG is the open rime of ṲNG. By the same token, AU and AIU should never be confused although they both represent the sound [au].
5. The past few centuries have witnessed the merger of IU (秋) and IEU (烧) into [iu], as well as UI (辉) and UOI (杯) into [ui]. Note that the open rimes of IU and UI, though they take different forms as EU and OI in Foochow Romanized, still remain their close rime pronunciation, so they are not to be confused with the close rime EU (沟) an OI (催). For example, 就 and 鸟 are respectively pronounced as [tsiu] and [tseu], although they both share the same spelling CEU (就: cêu; 鸟: cĕu). The method to distinguish between them is by tonal marks, and we’ll discuss this in the next secion.
6. AE̤ may look a bit bizarre. It stands for [ɔy].
7. A closing diphthong should not, in principle, precede the coda H. So letter combinations such as AUH, OH, EH, AE̤H are all considered incorrect.
8. If we examine the last two tables carefully, we will find that they actually form a pair of bijective sets (or one-to-one corresponding sets), which we can name as Set NG and Set K. In mathematical terms, we say that for every x-ng in the Set NG there is exactly one y-k in the Set K, and vice versa: ANG <-> AK, IANG <-> IAK, UANG <-> UAK, IENG <-> IEK, IONG <-> IOK, etc. This is an important feature, and will help you understand the Foochowese phonological structure more thoroughly.

Close/Open rimes

Some rimes in these four tables appear in pairs: close rimes are on the left and open ones on the right. Close/open rimes are sometimes called “tense/lax rimes”, but which term we should adopt doesn’t really matter. I need you to know that this linguistic feature is closely related with tonal categories: close rimes only occur with 上平 (primary smooth tone), 上声 (high tone), 下平 (secondary smooth tone) and 下入 (secondary entering tone), while open rimes only occur with 上去 (primary diminishing tone), 上入 (primary entering tone) and 下去 (secondary entering tone). In connected speech, open rimes stay unchanged in the last syllable position of multi-charactered phrases, but they do undergo sandhi in any position rather than the last, in which case these open rimes degenerate into corresponding close ones. However, under no circumstances will a close rime turn into its open counterpart. For instance, 德 (dáik, [taiʔ]) is a 阴入 character, and thus has an open rime [aiʔ]. But when combined with other characters things get a little more complicated. In 德国 (Dáik-guók, Germany), 德 is in the first syllable position, so the open rime [aiʔ] turns into its close [eiʔ], and the phrase is actually pronounced [teiʔ kuoʔ]; whereas in another phrase 道德 (dô̤-dáik, morality), 德 remains its open rimes because it’s in the last syllable position.

“Hang on... If 德国 is pronounced [teiʔ kuoʔ], then why is it written as ‘Dáik-guók’ in Foochow Romanized? Shouldn’t it be ‘Dék-guók’ or something?” Ah, yes, I know I can’t escape the careful examination of my readers. The fact is, in Foochow Romanized we always write 上去, 上入 and 下去 characters in the open rime forms, no matter whether they will undergo the open-to-close changes in real speech. So 福州 (Fuzhou, [huʔ tsiu]) is written as “Hók-ciŭ”, 四川 (Sichuan, [sy tshuoŋ]) is written as “Sé̤ṳ-chuŏng”.

Now here’s the question we raised previously: how do we distinguish 就 and 鸟 when they have the same spelling CEU? The answer is by their tonal marks. “Cêu” has a 阳入 diacritic ^ and thus must be 就 and pronounced [tsiu]; while “cēu” is a 上声 character and can only be 鸟.

Miscellaneous

(This section is not yet written. Coming soon.)

To summarize, let’s read the two poems from the Book of Eight Tones in Fuzhou dialect, which can help you memorize all the initials and rimes I listed above.

柳边求气低，Liū Biĕng Giù Ké Dă̤,
波他曾日时，Pŏ̤ Tă Cèng Nĭk Sì,
莺蒙语出非，Ĕng Mùng Ngṳ̄ Chók Hĭ,
打掌与君知。Dā Ciōng Ṳ̆ Gŭng Dĭ.

春花香，Chŭng Huă Hiŏng,
秋山开，Chiŭ Săng Kăi,
嘉宾欢歌须金杯，Gă Bĭng Huăng Gŏ̤ Sṳ̆ Gĭng Buŏi,
孤灯光辉烧银釭，Gŭ Dĕng Guŏng Hŭi Siĕu Ngṳ̀ng Gŏng,
之东郊，Cĭ Dĕ̤ng Giĕu,
过西桥，Guó Să̤ Giò,
鸡声催初天，Giĕ Siăng Chŏi Chĕ̤ Tiĕng,
奇梅歪遮沟。Gì Muòi Uăi Ciă Gĕu.

Do you have further questions about this brief course? Or do you have any ideas to improve it?
Please email me at gnudoyngATgmailDOTcom. I’d be very glad to hear from you.

Last updated: 2009/10/22