The Chinese character 上 accompanied by the ruby text "zhâng/", an early form of TCC pronunciation spelling of the character.
The Tangent Constructed Chinese (usually abbreviated as TCC) is a simplified proto-language/conlang hybrid based primarily on Cantonese, Mandarin, Sino-Japanese, Sino-Korean, and Sino-Vietnamese. The title stands for Constructed Chinese that is made by Tangent. The TCC has been used on its website for certain terms, and also in some stories written by Tangent.
Origins
Tangent noted similarities in the pronunciation of Chinese characters in Mandarin, Cantonese, and Sino-Japanese. However, even though Sino-Japanese pronunciations of Chinese characters are said to be "based on Chinese", they don't exactly sound uniformly like either Mandarin or Cantonese. Thus, based on the readings of the mentioned languages, Tangent sought to make a "general pronunciation" that encompasses all of those languages, giving the Sino-Japanese pronunciations "a [non-existent] Chinese to resemble".
Unbeknownst to Tangent, Sino-Japanese can be considered to have been borrowed from an older version of Chinese known as Middle Chinese, which is the ancestor of most modern Chinese varieties. So what Tangent was creating was actually more akin to a proto-language reconstruction, rather than a fictional language. However parts of it still gave it properties of a fictional language, such as the relative little emphasis placed on resolving conflicting information from the source languages, and that TCC has its own vocab different from Standard Chinese.
Description
Phonology of Tangent Constructed Chinese, March 2015 model.
The TCC uses a "top-down" system, similar to Chao's General Chinese, where modern pronunciation takes precedence over actual historical pronunciation. The pronunciations of individual Chinese characters may ideally be converted to that in most Chinese varieties, Sino-Japanese, Sino-Korean, Sino-Vietnamese, and potentially other Chinese borrowings in a predictable manner, using a set of rules, each of which will be language-specific. The orthographic spelling is intended to "look Chinese to an average person", which meant that Eitel's Cantonese Romanization, Wade-Giles, and Pinyin were the main models of which its look is meant to imitate.
As of 2 June 2018, the Basic version of TCC has 33+2 consonants, 7 vowels, and 3 phonemic tones. There are different versions of TCC that are intended to bridge the gap between Middle Chinese and Proto-[given language]. The most well-defined one is probably Modernized TCC (MTCC), which has 6 phonemic tones, no voiced obstruents, exhibits 陽上作去 (provisional), and the vowel ü. TCC's phonology (19 July 2015 update) is summarized below:
Consonants
TCC has a 3-way distinction between tenuis, aspirated, and "voiced" consonants. The "voiced" consonants may in fact be slack-voiced.
| (Bi)labial | Labiodental | Alveolar | Postalveolar | Dorsal | ||||
|---|---|---|---|---|---|---|---|---|
| Plain | Sibilant | Plain | Sibilant | |||||
| Nasal | <m> /m/ | <mh> /ɱ~ʋ/ | <n> /n/ | <ṇ> /n̠/ <gn> /ɲ/ | <ng> /ŋ/ | |||
| Stop | Aspirated | <ph> /pʰ/ | <th> /tʰ/ | <ts> /t͡sʰ/ | <ṭh> /t̠ʰ/ | <ch> /t̠͡ɕʰ/ | <kh> /kʰ/ | |
| Plain | <p> /p/ | <t> /t/ | <ds> /t͡s/ | <ṭ> /t̠/ | <c> /t̠͡ɕ/ | <k> /k/ | ||
| Voiced | <b> /b/ | <d> /d/ | <dz> /d͡z/ | <ḍ> /d̠/ | <j> /d̠͡ʑ/ | <g> /g/ | ||
| Fricative | Voiceless | <f> /f/ | <s> /s/ | <sh> /ɕ/ | <h> /x~h/ | |||
| Voiced | <v> /v/ | <z> /z/ | <zh> /ʑ/ | <gh> /ɣ~ɦ/ | ||||
| Approximant | <w> /w/ | (ʋ) | <l> /l/ | <y> /j/ | ||||
- <mh> /ɱ~ʋ/ is a free variation between a labiodental nasal [ɱ] and approximant [ʋ].
- There were two distinct postalveolar nasals, <ṇ> /n̠/ and <gn> /ɲ/. In later varieties of Chinese, the earlier usually merged with the regular alveolar /n/, while the latter's behavour varied. In Cantonese it was merged with /j/, in Mandarin /ʐ~ɻ/, Hakka /ŋ/, Hokkien /dz/ or /l/, Shanghainese /z/ or /ɲ/, etc.
- The exact articulation of the postalveolar series, aside from <gn>, is not strictly defined. In the earlier versions of TCC, the plain postalveolar series was merged with either the postalveolar sibilants or the plain alveolars. However after consistent suggestions to list them separately, they have been. While the series was previously regarded as alveolopalatal [ɲ cʰ c ɟ] due to them only occuring before [j] and [i], there was evidence from Vietnamese, as well as potentially Mandarin, that worked in favour of putting the series as retroflex [ɳ ʈʰ ʈ ɖ]. With the sibilants, it was Tangent's personal preference to put them as alveolopalatal, after pre-War Cantonese.
- While <y> and <w> may not necessarily be full consonants, TCC has a 3-way medial distinction between <y~i> [j], <w> [w], and <yw~iu> [jw]. Additionally, the last update as of 2 June 2018 features a distinction between <yu> [ju] and <yiu> [jiu], so recording these consonants as separate might be helpful.
Vowels and tones
For the most part, [a] and [ɛ] are in complimentary distribution, with the latter usually only occuring after [j]. [o] only occurs before /ŋ/ and after /j/, and /uŋ/ is phonetically [ʊŋ].
The vowel <â> is given the "unofficial" alternate spelling <aa>.
TCC distinguishes 3 phonemic tones, biēng, zhiấng, and khio (under the November 2017 update). Syllables ending in a stop consonant are classified as having a gnip tone, but that is phonemically not distinguished from the khio tone. The phonemic tones may have different allotones based on the voicing of the initial, e.g. piēng (冰) may be [pjɛŋ˦~pjɛŋ˥] while biēng (平) may be [bjɛŋ˨~bjɛŋ˩].