Hong Kong Government Cantonese Romanisation
|This article needs additional citations for verification. (March 2007)|
The Hong Kong Government Cantonese Romanisation (not an official name) is the more or less consistent way for romanising Cantonese proper nouns employed by the Hong Kong Government departments and many non-governmental organisations in Hong Kong. It is not known whether there are strict guidelines for the method circulating in the government, or the method has just established itself and become a common practice over time. The system has been widely used by the Hong Kong Government from the very early days of British rule, and has since gone through some changes between the two World Wars.
Since the method is not standardised, Hong Kong Education and Manpower Bureau has approved a Cantonese Pinyin system for teachers in primary and secondary schools. Besides this, the Linguistic Society of Hong Kong promotes their own Jyutping system. Both schemes are employed by the government to illustrate accurate pronunciation of Cantonese words.
This article illustrates and explains how the proper nouns in Hong Kong are transcribed and romanised, and lists the corresponding pronunciations of the spellings with respect to IPA and Jyutping.
The Hong Kong Government adopts the Eitel/Dyer-Ball system of romanisation, which is based on the spoken Cantonese language. It was first adopted in 1960 to standardise the romanisation of placenames throughout Hong Kong (the standardised placenames were published in the 1960 government publication 'A Gazetteer of Place Names in Hong Kong, Kowloon and the New Territories'. Prior to this 1960 publication, there was no standard, consistent way of romanising placenames in the territory, which, predictably, often led to confusion. Since then, the romanisation system has been extended to the names of local Chinese, which gives romanised Hong Kong Chinese names a distinctive character.
For place names, the type of the place in English (e.g., "Street" and "Road") are often used instead of a romanisation (which would have been "Kai" and "Lo" in the previous example), with just a handful of rare exceptions (for example, the "Fong" in "Lan Kwai Fong", which would have been a "Square" if a translation were used). "Wan" and "Bay", "Tsuen" (or "Chuen") and "Estate" (or "Village"), are, however, equally common. Some places, such as "Un Long", was later renamed as "Yuen Long" according to this standard, with the exception "Un Chau Estate" / "Un Chau Street Estate". Nonetheless, the names "Hong Kong" and "Kowloon" are not transliterated based on this system, as they were already named as such prior to the founding of the colony.
When the romanisations are spoken in an English conversation, they are pronounced in a somewhat anglicised manner. All words are consistently pronounced in tone equivalent to the Yin Ping tone or tone 1. A good everyday example is the broadcast of station names on MTR trains.
Some instant messaging users, having problems typing in Chinese characters, model this rule of romanisation for communication, but they use voiced instead of voiceless unaspirated consonants, such as using 'b', 'd' or 'g' where this system may have used 'p', 't' or 'k'.
It is not a fully standardised system, and many of the phonemes correspond to more than one letter combination or the other way round. All tones are omitted as are distinctions between aspirated and unaspirated stops. The distinctions between the long vowel [a] and the short vowel [ɐ] are omitted like Fat (發, [fat]) and Fat (佛, [fɐt]).
Some of the inconsistencies are due to a distinction that has been lost historically (a distinction between palatal and alveolar sounds, viz. ch versus ts, sh versus s, and j versus z). These consonants are no longer distinguished in present-day speech.
Under the following table, geographical names are used to illustrate. (Biographical names are not used as people have the right to decide how their names be romanised, although the same rule usually applies.)
|pʰ||p||p||Sai Ying Pun||西營盤|
|k||g||k||Tai Kok Tsui||大角嘴|
|kʷ||gw||kw||Cha Kwo Ling||茶果嶺|
|m||m||m||Yau Ma Te||油麻地|
|ŋ||ng||ng||Ngau Tau Kok||牛頭角|
|s||s||s||So Kon Po||掃捍埔|
|sh||Shau Kei Wan||筲箕灣|
|w||w||w||Wong Tai Sin||黃大仙|
|tsʰ||c||ch||Heng Fa Chuen||杏花邨|
|ts||Yau Yat Tsuen||又一村|
|ts||z||ts||Tsim Sha Tsui||尖沙嘴|
|-p||-p||-p||Ap Lei Chau||鴨脷洲|
|-t||-t||-t||Tsat Tsz Mui||七姊妹|
|-m||-m||-m||Sham Shui Po||深水埗|
Vowels, diphthongs, and syllabic consonants
|aː||aa||a||Ma Tau Wai||馬頭圍|
|ah||Wah Fu Estate||華富邨|
|ɐ||a||a||Tsz Wan Shan||慈雲山|
|u||Sham Chun River||深圳河|
|ɛː/e||e||e||Che Kung Miu||車公廟|
|iː/ɪ||i||i||Lai Chi Kok||荔枝角|
|ze||Sheung Sze Wan||相思灣|
|ee||Tat Chee Avenue||達之路|
|eo||Nam Cheong Street||南昌街|
|ɵ||eo||u||Shun Lee Estate||順利邨|
|yː||yu||yu||Yu Chau Street||汝州街|
|u||Kau U Fong||九如坊|
|ue||Yung Shue Wan||榕樹灣|
|aːu||aau||au||Shau Kei Wan||筲箕灣|
|ɐu||au||au||Sau Mau Ping||秀茂坪|
|ei||ei||ei||Lei Yue Mun||鯉魚門|
|ay||Kam Hay Court||錦禧苑|
|ai||Shui Hau Sai Ngan Ma||水口四眼馬|
|i||To Li Terrace||桃李台|
|iːu||iu||iu||Siu Sai Wan||小西灣|
|ɔːi||oi||oi||Choi Hung Estate||彩虹邨|
|oy||Choy Yee Bridge||蔡意橋|
|ɵy||eoi||ui||Ma Liu Shui||馬料水|
|ou||ou||o||Tai Mo Shan||大帽山|
|ŋ̩||ng||ng||Ng Fan Chau||五分州|
- ^ The standard pronunciation of 五 is [ŋ̩]. However, a more common pronunciation in Hong Kong is [m̩] and many [ŋ̩] words are merging with it. The only word that was originally pronounced as m̩ is "唔 (not)", and it is not used in place names.
Pronunciation in English
The romanised words are normally pronounced in a somewhat anglicised way, with the following characteristics which are different from what the above discussion on spelling might indicate:
- The letters p, t, k, plus the combinations kw and ts, are normally aspirated as per English; some English speakers in Hong Kong (including radio announcers) may choose to pronounce them unaspirated if the original Cantonese sounds are known to be unaspirated.
- The sound ng is pronounced as in Cantonese; however, because initial /ŋ/ does not occur in English, English speakers usually have difficulty with them. (It is possible for it to be mispronounced /n/.)
- The sound sh is pronounced as English sh (IPA: /ʃ/), despite such a sound being absent from Cantonese.
- The sound ts is to be pronounced as English ts (German z), but in practice might be pronounced as English ch (IPA: /tʃ/); however, because this sound does not normally occur at the initial position in English, English speakers will have difficulty pronouncing the sound. In Canada, ts is usually mispronounced as a simple /s/ or /z/ even among the Chinese.
- The letters p, t, k are pronounced as in English.
Vowels, diphthongs, and consonants
- The letter a is to be pronounced [a] or [ɐ]; however, English speakers pronounce it [ɑː] at the end, and [æ] before consonant.
- The digraph ai is to be pronounced [ai] or [ɐi]; however, English speakers not familiar with the romanisation may pronounce it differently according to English pronunciation rules, [eɪ].
- The digraph au is to be pronounced [au] or [ɐu]; however, English speakers not familiar with the romanisation may pronounce it differently according to English pronunciation rules, [ɔː].
- The letter i is to be pronounced [i], but [ɪ] before k and ng; before a p, t, m and n, English speakers not familiar with the romanisation may pronounce it as [ɪ] as in English.
- The digraph ei is to be pronounced [ei]; however, English speakers not familiar with the romanisation may pronounce it [aɪ].
- The digraph ou is to be pronounced [ou]; however, English speakers not familiar with the romanisation may pronounce it [aʊ] or [uː].
- The digraph iu is to be pronounced [iu]; however, this sound does not exist in English. English speakers mispronounce [juː].
- The letter u is to be pronounced [u], but [ʊ] before k and ng; before a consonant, English speakers not familiar with the romanisation may pronounce it as [ɐ] as in English.
- The digraph eu is to be pronounced [œ]; however, this sound does not exist in English. English speakers mispronounce [juː] or [uː].
- The letter u (when after y) or the digraph ue is pronounced [y] as in Cantonese (it's pronounced from lip-rounded /i/); however, this sound does not exist in English. English speakers mispronounce [juː] or [uː].
- The diphthong ui is to be pronounced a diphthong [ɵy] (e.g., similar to the diphthong denoted by öy in Finnish or ui in Dutch), but [ui] before b, p, m, f, w; however, these sounds do not exist in English. English speakers mispronounce as a succession of two vowels [uː.i].
- The syllabic consonant m and ng are pronounced [m̩] and [ŋ̩] The sound [ŋ̩] do not exist in English. Many Hong Kong locals do not distinguish [m̩] and [ŋ̩]. This results in a phonological shift in Hong Kong Cantonese that sees a merge of [ŋ̩] into [m̩]. In fact, the tone is the only way to distinguish the surnames 伍 [ŋ̩˨˧] and 吳 [ŋ̩˩], but both are written "Ng", and appear as "Wu" in Mandarin. Note that the standard pronunciation of 伍 is [ŋ̩˨˧] with a rising tone, but 吳 is pronounced [ŋ̩˩] with a low tone. In North America, as a single [ŋ] does not exist in English, ng is usual pronounced as [ɪŋ].
- List of common Chinese surnames shows how they are romanised in this scheme.