Middle Chinese: phonology, sources, and historical significance

The traditional analysis of the Chinese syllable, derived from the fanqie method, consists of the initial (consonant) sound (聲母 / 声母, shēngmǔ) and the final sound (韻母 / 韵母, yùnmǔ). Modern linguists further subdivide the final sound into consonants: an optional "middle" glide (韻頭 / 韵头, yùntóu), a main vowel or "nucleus" (韻腹 / 韵母, yùnfù - "core vowel"), and an optional final consonant or "coda" (韻尾 / 韵尾, yùnwěi). Most reconstructions of Middle Chinese include the glides (semivowels) /j/ and /w/ and a /jw/ combination, but many also include the vowel "glides" such as /i/ in a diphthong /ie/. The consonants /j/, /w/, /m/, /n/, /ŋ/, /p/, /t/, and /k/ are widely accepted, sometimes with additional codas such as /wk/ or /wŋ/. Rhyme syllables in Qieyun - as assumed - have the same core vowel and coda but often different inlaws.

Middle Chinese reconstructions by various modern linguists vary. These differences are minor and hardly controversial with respect to consonants; nevertheless, there are more significant differences here than with vowels. The most widely used transcriptions are Li Fang-Kuei's modification of Karlgren's reconstruction and William Baxter's keyboard-writable notation.

Anlaute

The preface to the Yunjing identifies a traditional set of 36 anlauts, each named with a pattern character. An earlier version, comprising 30 anlauts, is known from fragments among the Dunhuang manuscripts. In contrast, identifying the Qieyun's anlauts required a head-scratching analysis of fanqie relationships throughout the dictionary, a task first undertaken by the Cantonese scholar Chen Li in 1842 and refined by others since. This analysis revealed a slightly different set of anlauts from the traditional set. Moreover, most scholars believe that some distinctions among the 36 anlauts were not even in common use at the time of the rhyme tables, but were retained under the influence of earlier dictionaries.

Early Middle Chinese (English: Early Middle Chinese; EMC) had three types of plosives: voiced, voiceless, and voiceless aspirated. There were five series of coronal obstruents with triple distinction between dental (or alveolar), retroflex, and palatal among fricatives and affricates, and a dual dental/retroflex distinction among plosives. The following table shows the anlauts of Early Middle Chinese with their traditional names and nutritional values:

Old Chinese had a simpler system without palatal or retroflex consonants; the more complex system of EMC probably arose from a combination of Old Chinese obstruents followed by /r/ and/or /j/.

Bernhard Karlgren developed the first modern reconstruction of Middle Chinese. The main differences between Karlgren and more recent reconstructions of the anlauts are:

• The rearrangement/loss/cancellation of /ʑ/ and /dʑ/. Karlgren based his reconstruction on Song dynasty rhyme tablets. Due to mixtures/transitions of these two sounds between Early and Late Middle Chinese, the Chinese phonologist who created the rhyme tables could only rely on tradition (oral tradition) to determine the respective (phonetic) values of these two consonants; obviously they were inadvertently switched at some stage.
• Karlgren also assumed that EMC retroflexes were actually palatals because of their tendency to occur simultaneously with front vowels and /j/, but this view is no longer held.
• Karlgren assumed that voiced consonants were actually aspirated. Today, this is assumed only for the LMC, not for the EMC.

Various changes occurred between the time of the Qieyun and the rhyme tablets:

• Palatal sibilants intermingled with retroflex sibilants.
• /ʐ/ merged with /dʐ/ (consequently forming four separate EMC phonemes).
• The palatal nasal /ɲ/ also became retroflex, but became a new phoneme /r/ rather than blending with an existing phoneme.
• The palatal allophone of /ɣ/ (云) merged with /j/ (以) to form a single laryngeal anlaut /j/ (喻).
• A number of new labiodentals arose from labials in certain environments, typically where both centralization and rounding occurred (e.g. /j/ plus a back-tongue vowel (William Baxter's reconstruction), or plus a rounded front-tongue vowel (Chan's reconstruction)). Modern Min dialects, however, retain bilabial affixes in such words, while modern Hakka dialects retain them in some common words.
• Vocal obstruents were aspirated (still present in Wu (speech) varieties).

The following table shows a representative summary of the anlauts of Late Middle Chinese.

The voiced/unvoiced distinction is maintained in modern Wu dialects, but has disappeared from other varieties. In Min dialects, retroflex dentals have merged with dentals, while elsewhere they have merged with retroflex sibilants. In the South, they also merge with the dentic sibilants, but are retained in most Mandarin dialects. The palatal series of modern Mandarin dialects, which arose from a blending of palatal allophones of dental sibilants and velars, is a much more recent development that has nothing to do with the earlier palatal consonants.

Loud

The remainder of a syllable after the initial consonant is the final syllable, which is represented in Qieyun by several equivalent auxiliary fanqies. Each final syllable is assigned to a singular rhyme class, but a rhyme class can contain between one and four final syllables. Pronouns are usually decomposed to consist of an optional vowel (either semivowel, reduced vowel, or a combination thereof), a vowel, an optional final consonant, and a tone. Their reconstruction is much more difficult than for the anlauts due to the combination of multiple phonemes in a singular class.

The generally accepted final consonants are: the semivowels /j/ and /w/, the nasals /m/, /n/ and /ŋ/, and the plosives /p/, /t/ and /k/. Some authors also propose the codas /wŋ/ and /wk/, based on the separate treatment of different rhyme classes in dictionaries. Final vowels with vowel and nasal codas may have one of three tones, namely level tone, rising tone, and falling tone. Final vowels with plosive codas distribute in the same way as the corresponding nasal final vowels, and are described as their "entering tone" counterparts (7th and 8th tones, 90% of which became 4th tones in modern Chinese).

There is less agreement on the inlaut and vowel. It is predominantly thought that "closed" vowels had a rounded glide/w/ or vowel /u/ and that the vowel in "outer" vowels was more open than that in "inner" vowels. The interpretation of "divisions" (English: "divisions") is more controversial. Three classes of qieyun vowels appear exclusively in the respective first, second, and fourth rows of the rhyme charts and have therefore been designated as division I, II, and IV vowels. The remaining vowel sounds have been called Division III vowel sounds because they occur in the third row; however, they may also occur in the second and fourth rows in some vowel sounds. Most linguists agree that the Division III vowel sounds contained an inlaut-/j/ and that the Division I vowel sounds did not possess such an inlaut, but other properties that vary depending on the reconstruction. To account for the many rhyme classes that Qieyun distinguishes, Karlgren proposed 16 vowels and 4 inlaut. Later scholars have proposed numerous variations.

Sounds

The Qieyun classified characters into four components according to their tone: flat tone (平聲 / 平声, píngshēng), rising tone (上聲 / 上声, shǎngshēng), leaving tone (去聲 / 去声, qùshēng), and entering tone (入聲 / 入声, rùshēng). It should be noted, however, that only three of the many tones are phonemic. Open syllables or those ending in a nasal sound distinguish between the first three tones. The "entering tone", on the other hand, occurs only in syllables ending in a plosive (/p/, /t/, or /k/). As a rule, however, the syllables ending in a plosive can be assigned a corresponding ending in a nasal sound, so that one can alternatively regard the tones as phonemic and the distinction between plosives and nasals in the end of the syllable as allophonic.

The entering tone is then characterized by a distinct inflection. It is difficult to determine the exact contours of the other tones. Karlgren interpreted the names literally and assumed a corresponding flat, rising and falling tone. The oldest known description of the tones was found in a Song Dynasty (early 9th century) quotation 元和韻譜 / 元和韵谱, Yuánhé Yùnpǔ - "Onset and End Tables" (no longer extant): "The level tone is sad and stable. The rising tone is shrill and rising. The breaking tone is clear and distant. The entering tone is straight and abrupt."