User:ND381/Wu Expansion

=Basic concept=

I’ve seen some plans for Suzhounese and Wenzhounese implementation into Wiktionary, and Shanghainese people complaining that Wiktionary’s Shanghainese not being the most accurate. This is a page for me to plan out implementing Wugniu Romanisation, possibly completely replacing Wiktionary’s Shanghainese Romanisation.

Hopefully the result should display as such (example word: 人 - 白讀): * Wu * (Shanghai) * Wugniu: gnin6 * Old Wiktionary: nyin (T3) * Sinological IPA: /ȵin¹¹³/ * (Suzhou) * Wugniu: gnin2 * Sinological IPA: /ȵin²²³/ * (Hangzhou) * Wugniu: gnin2 * Sinological IPA (new and old): /ȵin²¹³/ * (Ningbo) * Wugniu: gnin2 * Sinological IPA: /ȵin²⁴/ * (Wenzhou) * Wugniu: gnian2 * Sinological IPA: /ȵaŋ³¹/ And when collapsed: * Wu (Shanghai, Wugniu) gnin6 (Suzhou, Wugniu) gnin2 (Hangzhou, Wugniu) gnin2 (Ningbo, Wugniu) gnin2 (Wenzhou, Wugniu) gnian2

Shanghainese (done!!)
What we have now is functional but not great. The following two tables show how we can get a Wugniu reflex when given a Wiktionary input, and a slight update to the IPA to be more accurate according to the 方言志, as well as what many Shanghainese people say is correct.

A module is present at Module:User:Manishearth/wuu-pron. See more at User:Manishearth/zh-pron-sandbox, Module:User:Manishearth/zh-pron, and Module_talk:wuu-pron.

Full implementation is soon to come; the module has been fully prepared

Romanisation
The tone numbers skip 2-5 in order to facilitate other tone categories that are found in other Wu languages. [*] Note that ɦ + i → y and ɦ + u → w, eg. ɦyn → yun, ɦuən → wen. This applies to all Wu languages unless specified.

IPA
As many have pointed out, the current IPA is a bit funky, to say the least. Sadly, since Wyang has left Wiktionary, we cannot ask where he obtained the IPA readings. Whatever his reasons may be, I believe the following is more accurate and preferred by Shanghainese learners and speakers. All unnecessary diacritics, such as breves and devoicing diacritics, shall be removed.

Implementation
As per Beer parlour/2022/August%23Shanghainese IPA update and Wugniu Romanisation, the following is to be done:
 * The IPA notation is to be replaced (no disagreement)
 * Wugniu will be displayed as the default in the collapsed table, and Wiktionary's transliteration will be kept in the expanded part as a legacy feature (general agreement)
 * A temporary category “w-sh” shall be introduced for Wugniu. This would mean that other modules that rely on the old Wiktionary module will still be functional until they are made to be compatible with the new Wugniu one (general agreement, also see Discord server)
 * Tone notation will be discussed in the Beer Parlour until a consensus is achieved

Suzhounese (done!!)
Here is the scheme for Wugniu’s Romanisation for Suzhounese. [*] Vernacular reading for all

Tones
''Note that this is one of many realisations of Suzhounese tone sandhi and this is the best compremise we have found so far. Each space (␣) represents a new syllable's tone contour, and does not take into account the vowel length of each syllable.''

Right prominent sandhi
Only exists for dark departing: 523 > 51. 笑別人 is realised as shiau523 > 51 + biq2-gnin51 thus resulting in a tone merger : 考兒子 = 靠兒子.

However, this only occurs when the dark departing term is monosyllabic, so 考考兒子 ≠ 靠靠兒子.

Taken from 蘇州方言語音研究 (汪平 編著)

Wenzhounese
Here is the scheme for Wugniu’s Romanisation for Wenzhounese. Note that this scheme contradicts a lot of other publications and is still in the works.

Tones
[*]: Note that because Wenzhounese has a phonemic /j/ distinction, y is used with the /j/ initial instead. Eg. jau → yau

=Furthern expansions=

I might do Hangzhounese, Ningbonese, and Taizhounese romanisations eventually. Yeah that's it

Hangzhounese
We are aware of the alternative classifications of Hangzhounese that put it in the Mandarin category. However due to the lect also using the Wugniu scheme, it shall be covered as a Wu variety until further notice.

Wugniu does not notate New Period Hangzhounese innovations, such as the merger between [ᴇ̃] and [ᴇ]. A future module should hopefully be able to notate both in its IPA. Underlined text indicate possible in-house fill-ins for missing rimes. They are left out of Wugniu's romanisation due to an early New Period shift removing this rime

Tones
3 or more character sandhi could not be found. Taken from 浙江省語言志 and 杭州方言詞典