Module talk:ltc-pron

蝗
Something seems to be going wrong at 蝗. ? —suzukaze (t・c) 00:58, 6 February 2016 (UTC)
 * Thanks, fixed. Wyang (talk) 05:34, 6 February 2016 (UTC)

嗅
The Mandarin reflex for 嗅 doesn't seem right. — justin(r)leung { (t...) 07:05, 19 October 2016 (UTC)


 * Thanks for picking it up Justin. Fixed. Wyang (talk) 07:57, 19 October 2016 (UTC)

A pleasant surprise when ...
watching this new video. (6:50) (very nicely done by the way) Wyang (talk) 01:56, 1 March 2018 (UTC)
 * Same here. (I saw the video before I saw this.) — justin(r)leung { (t...) 03:03, 1 March 2018 (UTC)
 * Haha, great minds think alike. Wyang (talk) 04:09, 1 March 2018 (UTC)

n
adds an entry to Category:Middle Chinese lemmas with the sortkey of '?'. This seems very weird to me. —Suzukaze-c (talk) 23:19, 22 September 2020 (UTC)

Source/Reference
Hi! It would be great if one could add a link (or reference) to a source (sources) where all the MC and OC readings for all the characters are taken from. --Kshatriya Drum (talk) 14:52, 22 October 2020 (UTC)


 * Agreed. I am wondering where are the primary sources. And the ordering of the Initials and the Rhymes is based on what?
 * 220.100.54.134 02:20, 7 September 2022 (UTC)

Baxter's MC Transcription
Hi all, I've made an addition to `ltc-pron` to generate Baxter's Transcription of Middle Chinese (https://en.wikipedia.org/wiki/Baxter%27s_transcription_for_Middle_Chinese), the most commonly-used transcription of Middle Chinese in academia nowadays (for example, see "A Student's Dictionary of Classical and Medieval Chinese")

You can test it by changing any page with the `zh-pron` template to `sandbox`, which calls the modules that are in my personal sandbox. If you want to look at the code, it's here: Module:User:iwsfutcmd/ltc-pron, Module:User:iwsfutcmd/ltc-pron/data

I've tested it extensively, things look good, but I wanted ot put a note here because it seemed a little rude to just drop it in. I'll give it a day here and on the Discord for comments, and if there are no objections, I'll post it live. Iwsfutcmd (talk) 00:40, 21 April 2023 (UTC)


 * Note by the way that this is separate from the Baxter transcription in the Old Chinese box, which is directly pulled from Baxter & Sagart's data files. This is instead generated by the data in Module:zh/data/ltc-pron/%E5%93%86, etc., like the reconstructions in the Middle Chinese box. The benefit of this is we can now get a Baxter transcription for any character we have pronunciation data for, as opposed to just the ones Baxter & Sagart included in their data files (which is limited to the characters they've reconstructed for Old Chinese), and also it thus ties the Baxter transcription to a specific MC reading of a character.
 * Also, if you happen to notice a discrepancy between the generated Baxter and the Baxter in the Old Chinese box, please inform me here so I can fix it if it's a bug in the module. Iwsfutcmd (talk) 00:49, 21 April 2023 (UTC)
 * Hi! Hope I'm doing this right :)
 * First, Thanks for your work! I was reading Hill (2019)'s Old Chinese book chapter and noticed that 地 is given as dijH and it agrees with the Old Chinese box/B&S's data but the Middle Chinese box has sjijH. Seems like this is also the case for 四. I guess this is an issue of identifying chóngniǔ pairs? Spectraldani (talk) 17:29, 10 May 2023 (UTC)
 * Hmm, good eye, thanks for noticing that! So there's a tricky thing here—both 地 and 四 are marked as having final #15, which I assigned to "-jij". This would imply I should change #15 to "-ij", however, #17 is already "-ij". My first thought was I mixed the two up, but on examining entries like https://en.wiktionary.org/wiki/%E6%9F%85, it seems #17 should definitely be "-ij". My second thought was that perhaps both should be "-ij", but then I came across https://en.wiktionary.org/wiki/%E5%92%BF, which is #15 and *also* "-jij". So ultimately I'm not totally sure what's going on here. If I make #15 '-ij", it'll solve the problem for 四 and 地, but it'll make 咿 incorrect.
 * I'll put a little more research into it... Iwsfutcmd (talk) 09:31, 15 May 2023 (UTC)

Switching from Zhengzhang to Baxter for the default-displayed transcription
Right now, the infobox contains many different reconstructions, as well as Baxter's transcription of Middle Chinese. But when the infobox isn't expanded, it displays just the Zhengzhang reconstruction.

There's been some discussion on the Wiktionary Discord Sinitic channel that we should switch this default-displayed string from the Zhengzhang reconstruction to something more neutral. All of the reconstructions including Zhengzhang are "opinionated", making decisions about how certain sounds are reconstructed as well as trying to determine which sounds represent phonetic representations of single phonemes. But unlike, say, Proto-Indo-European, there isn't broad scholarly consensus on a "correct" reconstruction, so preferring Zhengzhang here implies an endorsement that we may not want to be promoting.

I suggest that instead we use Baxter's Transcription of Middle Chinese instead, because unlike the reconstructions, it does *not* attempt to be opinionated about the details of reconstruction, it instead just represents the phonetic distinctions reflected in contemporary rime dictionaries and tables such as the Qieyun and the Yunjing, using the Latin script. It is a transcription, not a reconstruction.

As an example of this, Zhengzhang merges the finals "嚴" and "凡" (our finals 145 & 146), representing both as /ɨɐm/ as he considered the distinction non-phonemic. Baxter, on the other hand, represents them differently as "jaem" and "jom", as regardless of his personal feelings towards whether the distinction is phonemic or not, they are distinguished in the rime tables & dictionaries and are thus distinguished in Baxter's transcription.

Another way to put this is there is a 1-to-1 correspondence between our internal data representation of Middle Chinese pronunciation and Baxter's Transcription, which is not true for the reconstructions. Thus any of the reconstructions could potentially be derived straight from Baxter, but could not from each other as the translation from our representation to any of the reconstructions is lossy.

One further argument for Baxter's Transcription is it is quickly becoming the standard Latin representation of Middle Chinese within the academic literature. For example, "A Student's Dictionary of Classical and Medieval Chinese" by Paul W. Kroll includes Baxter's Transcription for each reading of each Classical Chinese character.

I have a working prototype of the replacement in my sandbox module for ltc-pron. If anyone would like to test this, simply go to any page that uses zh-pron and change it to Template:User:iwsfutcmd/zh-pron (which points towards my sandbox modules).

Feel free to ask any questions, I'll try to answer them promptly. Iwsfutcmd (talk) 18:11, 7 August 2023 (UTC)


 * . Why hasn't this been done yet? – Wpi (talk) 18:43, 7 August 2023 (UTC).
 * Well I did do it but it got reverted because I didn't have a corresponding discussion on the talk page.
 * As far as why it hasn't been done historically, I only wrote the Baxter-producing module a few months ago. Iwsfutcmd (talk) 19:01, 7 August 2023 (UTC)
 * Also, process question: how many "Support"s should I wait for until I go ahead and push the change? Iwsfutcmd (talk) 19:02, 7 August 2023 (UTC)


 * , but please fix the slash issue. RcAlex36 (talk) 10:34, 8 August 2023 (UTC)
 * Should already be fixed, thanks. Iwsfutcmd (talk) 15:41, 8 August 2023 (UTC)
 * Alright, it's been long enough and I don't have any objections. I'll implement it now. Iwsfutcmd (talk) 17:49, 15 August 2023 (UTC)
 * An afterthought: I think we have to be consistent in displaying the transliteration and therefore also change ltc-l to use Baxter transcription. This seems to already have been done for the template, but we need to convert the manual tr in a couple places over to use Baxter transcription. – Wpi (talk) 17:55, 15 August 2023 (UTC)
 * Can you give me an example of where that might be? Iwsfutcmd (talk) 02:34, 19 August 2023 (UTC)
 * I've already converted the ones in ltc-l (e.g. ), but my work might contain errors, and there's still other templates (e.g. der, see ) which would also need converting. – Wpi (talk) 05:13, 19 August 2023 (UTC)

obsoleting ltc-l
crackpot idea: obsoleting ltc-l by replacing id with the Unicode 平上去入 tone markers: ltc. in how many cases are there readings with the same tone? —Fish bowl (talk) 22:19, 13 September 2023 (UTC)
 * a lot actually lol. nvm? —Fish bowl (talk) 23:04, 14 September 2023 (UTC)