User:KureYoshitatsu/Simplified Transcription of Middle Chinese

The Simplified Transcription of Middle Chinese (STMP) is a scheme of phonetic transcription for Middle Chinese with a simplified phonology. The foundation of this scheme is Guangyun (廣韻), the Chinese rhyme dictionary compiled from 1007 to 1008 during the Song dynasty. Jiyun (集韻), which is an extension to Guangyun, is also used to complement Guangyun for any missing entries.

Four systems of romanization are used as the main referencing points of transcription: Baxter's transcription for Middle Chinese, Pe̍h-ōe-jī (POJ), Hanyu Pinyin, and the Vietnamese orthography.

With the Middle Chinese reconstructed phonology as the foundation, modifications are made by referring to the phonologies of modern Chinese dialects (mostly Mandarin, Hokkien, Cantonese) and Sino-xenic pronunciations of Chinese loanwords (mainly Sino-Korean and Sino-Vietnamese).

Alphabet
This romanization scheme uses 28 letters, including four letters with circumflex: â, ê, ô, ŷ. indent=1|* A a

Â â

B b

C c

D d

E e

Ê ê

G g

H h

I i

J j

K k

L l

M m

N n

O o

Ô ô

P p

Q q

R r

S s

T t

U u

V v

X x

Y y

Ŷ ŷ

Z z

Initials - 聲母 ⟨Srêng Mux⟩
The table below shows the initials found in Guangyun but with mergers, with the transcription symbols placed in angle brackets ⟨ ⟩, and the IPA values placed in slashes / /.

Rules for initials:

 * Baxter's transcription and POJ are used as the starting point and modifications are made from here. This is because both shares a three-way distinction of tenuis, aspirate and voiced stop consonants.
 * The alveolar stops are from the merger of the dentals, lateral, and retroflex stops in Baxter's transcription. This merger is based on Hokkien phonology.
 * The alveolar sibilants are called "dental sibilants" in Baxter's transcription. The use of ⟨c⟩ and ⟨ch⟩ are based on "ch" and "chh" in POJ respectively. The use of ⟨z⟩ is based on "z" in Hanyu Pinyin but it is voiced here. The diagraph ⟨sz⟩ here symbolizes that ⟨s⟩ cancels the affricate element of ⟨z⟩ and thus making it a pure fricative.
 * The postalveolars are from the merger of the palatals and the retroflex sibilants in Baxter's transcription. This merger is based on Mandarin phonology. The merger between 崇 and 俟 is shown in Karlgren's reconstruction, which is referenced in Baxter's handbook. The merger between 禪 and 船 is based on a controversy referenced in Baxter's handbook, which states that there was a common confusion of these two initials in Late Middle Chinese times. As a result, 崇/俟/禪/船 are all romanized as ⟨dr⟩ here. For example, 士 (/崇止/), 俟 (/俟止/) and 市 (/禪止/) are all spelled as ⟨drŷx⟩, and pronounced /ʒɨ˧˥/.
 * The use of ⟨tr⟩ is based on the way "tr" is used in Vietnamese and from here ⟨thr⟩, ⟨dr⟩ and ⟨sr⟩ are derived. The use of ⟨r⟩ is based on "r" in Hanyu Pinyin but the pronunciation is based on Burmese phonology where the Burmese "r" is pronounced /j/.
 * The velars and laryngeals are the same as those in Baxter's transcription. The use of ⟨q⟩ to represent /g/ here is based on the Gulf Arabic pronunciation of the Arabic letter Qāf (ق). Subsequently, ⟨g⟩ is used to represent /ɣ/ which is based on Vietnamese orthography. The use of ⟨h⟩ to /x/ is based on Hanyu Pinyin.
 * Generally, syllables with 影 are spelt with the vowel letters as the first letter like in Hanyu Pinyin, POJ, Vietnamese orthography and most romanization schemes of Chinese dialects. However, if the syllable has a /j/ or /ɥ/ glide, ⟨e⟩ is added in front. Similarly, ⟨o⟩ is added in front of syllables with the /w/ glide. For example, 恩 (/影痕/) is written as ⟨ân⟩ and pronounced /ʔən˥/; 憂 (/影尤/) is written as ⟨eiu⟩ and pronounced /ʔju˥/; 威 (/影微/) is written as ⟨oui⟩ and pronounced /ʔwi˥/.
 * 云 is treated as an allophone of 匣, and hence placed with the laryngeals. Since most modern dialects reflect the late Middle Chinese merger of 云 and 以 into 喻, 云 is treated as an approximant here.
 * 云 has two possible romanizations: ⟨j⟩ and ⟨v⟩. Whether to assign ⟨j⟩ or ⟨v⟩ depends the modern pronunciations found in Mandarin, Cantonese, Hokkien, Korean and Vietnamese. For example, 有 (/云有/) is written as ⟨jux⟩ and pronounced /ʝu˧˥/, referencing "yǒu" in Mandarin, "jau5" in Cantonese and "유" in Korean; 爲 (/云支/) is written as ⟨vê⟩ and pronounced /ʋe˥/, referencing "wéi" in Mandarin, "wai4" in Cantonese, "uî" in Hokkien, "위" in Korean and "vi" in Vietnamese.

Codas - 韻目 ⟨Vuns Miuk⟩
The syllable codas are arranged based on the 攝 (rhyme class) order in Guangyun, each coda is classified under their respective 等 (four divisions), and also 開 (unrounded) and 合 (rounded). The codas are split into two groups: vocalic codas that end with vowels, and codas that end with nasal and stop consonants. After some phonetic mergers, the transcription symbols are placed in angle brackets ⟨ ⟩, and the IPA values placed in slashes / /.

Rules for the phonology of codas:

 * Traditional rhyme doublets or Chongniu (重紐) are not distinguished in this transcription scheme. This is because a few linguists and philologists had omitted Chongniu distinctions in their reconstructions. Moreover, Chongniu distinctions have been largely lost in most modern Chinese dialects.
 * There are numerous reconstruction schemes for Middle Chinese codas available. Each of these reconstructions were considered, and subsequently mixed and matched. Complicated consonant clusters are simplified, and from there codas with similar reconstructed pronunciations are merged.
 * The vowels ⟨e⟩ and ⟨o⟩ can optionally be more open or near-open. That is, ⟨e⟩ can be more pronounced more like /æ/, while ⟨o⟩ can be more like /ɒ/.

Spelling conventions derived from POJ and Vietnamese orthography:

 * The diphthong /wa/ is written as ⟨oa⟩ just like in POJ and Vietnamese. Based on this pattern, /ja/ is written as ⟨ea⟩. This also applies when consonant endings are added.
 * Similar usage of vowel letters ⟨a⟩/⟨â⟩/⟨e⟩/⟨ê⟩/⟨o⟩/⟨ô⟩ like in Vietnamese. The exception is ⟨â⟩, which is expanded to also include the phonetic value of "ơ" in its usage.
 * The digraph ⟨uy⟩ here is similar to "uy" in Vietnamese. This is espectially so for ⟨uyn⟩ and ⟨uyt⟩, which are based on "uynh" and "uyt". Similarly, ⟨uê⟩ is exactly like "uê" in Vietnamese.
 * The combinations ⟨ie⟩/⟨uo⟩/⟨uye⟩ are similar to "iê"/"uô"/"uyê" in Vietnamese, although the pronunciations are not so similar.
 * The use of ⟨ŷ⟩ here is based the use of "y" in Vietnamese to complement "i". However, ⟨i⟩ and ⟨ŷ⟩ here represent the vowels /i/ and /ɨ/ respectively, unlike in Vietnamese where both "i" and "y" represent /i/.

Other spelling and phonology rules:

 * The diagraph ⟨eo⟩ is based on "eo", the romanization for Korean vowel "ㅓ". However, ⟨eo⟩ here represents /jɔ/ instead of /ʌ/ in Korean.
 * The diagraphs ⟨êi⟩ and ⟨ôu⟩ are based on "ei" and "ou" in Mandarin and Cantonese, but with circumflex added.
 * New rhyme doublets arising from the merger of the palatals sibilants (章/昌/禪/書/船) and the retroflex sibilants (莊/初/崇/生/俟) in Baxter's transcription. Wherever a coda contains a /j/ glide, /j/ will be lost when paired with the traditionally retroflex sibilants and the loss will also be reflected in the transcription. For example, 莊 (/莊陽/) is written as ⟨trang⟩ and pronounced /t͡ʃaŋ˥/, while 章 (/章陽/) is writing as ⟨treang⟩ and pronounced /t͡ʃjaŋ˥/.
 * Consonants with palatal character, namely ⟨r⟩, ⟨y⟩ and ⟨j⟩, when paired up with codas containing the /j/ glide, /j/ will be lost and the loss will also be reflected in the transcription. For example, 如 (/日魚/) is written as ⟨ro⟩ and pronounced /ɲɔ˥/ instead of ⟨reo⟩ and /ɲjɔ˥/; 用 (/以用/) is writing as ⟨yôngs⟩ and pronounced /joŋ˥˩/ instead of ⟨yiôngs⟩ and /joŋ˥˩/; 有 (/云有/) is writing as ⟨jux⟩ and pronounced /ʝu˧˥/ instead of ⟨jiux⟩ and /ʝju˧˥/. Additionally, since ⟨v⟩ is an allophone of ⟨j⟩, similar rule will also be applied. For example, 宇 (/云麌/) is written as ⟨vôx⟩ and pronounced as /vo˧˥/ instead of ⟨jiôx⟩ and /ʝjo˧˥/. This is based on "ú" in Hokkien, "우" in Korean, and "vũ" in Vietnamese.

Rules for the tonal values:

 * It is commonly accepted that the 平 tone is with a level contour, the 上 tone a high rising tone, the 去 tone a falling tone, and the 入 tone a checked tone. Their corresponding tonal values may be reconstructed as 33 /˧/, 35 /˧˥/, 51 /˥˩/, and 33ʔ /˧ˀ/ respectively. Using this as the starting point, slight modifications were made to the tonal values of the 平 and 入 tones.
 * The tonal value of 平 tone is raised from the original mid level (33 /˧/) to the high level (55 /˥/), which is the tonal value of the 1st Tone (the 陰平 "dark level" tone) in Mandarin, Cantonese and Hokkien.
 * The tonal value of 上 tone remains as high rising (35 /˧˥/), which is the tonal value of the 陰上 "dark rising" tone in Cantonese. Additionally, a tone sandhi rule in Mandarin states the when there are two 上 tones in a row, the first 上 tone will be pronounced as 35 /˧˥/ instead of its original low dipping tonal value.
 * The tonal value of 去 tone remains as high falling (51 /˥˩/), which is the tonal value of the 去 tone in Mandarin.
 * The tonal value of 入 is changed from the original mid level constricted (33ʔ /˧ˀ/) to the low falling constricted (31ʔ /˧˩ˀ/), which is the tonal value of the nặng tone in Vietnamese.

Rules for marking the tones:

 * 平 tone has no marking, just like the 陰平 tone in Hokkien as represented in POJ, and also the ngang tone in Vietnamese.
 * 上 tone is marked with a trailing ⟨x⟩, similar to Baxter's transcription where the uppercase version "X" is used.
 * 去 tone is marked with a trailing ⟨s⟩, because it is believed that the 去 tone came from an earlier "-s" suffix on Old Chinese.
 * 入 tone is marked with a trailing ⟨p⟩, ⟨t⟩ or ⟨k⟩ depending on the stop consonant which the syllable final ends with.