Jump to content

Wikipedia:Reference desk/Archives/Language/2022 May 13

From Wikipedia, the free encyclopedia
Language desk
< May 12 << Apr | May | Jun >> Current desk >
Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is a transcluded archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


May 13[edit]

Lemnos Stele[edit]

I tried to write a program to transliterate the romanized Lemnian in Lemnian language ยงย Lemnos Stele to the original Old Italic script. Can someone check that the correspondences are correct, and/or that the output looks right? โ€Ž

Python 3 code
table = {'a': '๐Œ€', 'e': '๐Œ„', 'h': '๐Œ‡', 'i': '๐Œ‰', 'k': '๐ŒŠ', 'l': '๐Œ‹', 'm': '๐ŒŒ', 'n': '๐Œ', 'p': '๐Œ', 'r': '๐Œ“', 's': '๐Œ‘', 't': '๐Œ•', 'u': '๐Œ', 'v': '๐Œ…', 'ลก': '๐Œ”', 'ฮธ': '๐Œˆ', 'ฯ†': '๐Œ˜', 'ฯ‡': '๐Œ™'}
table = str.maketrans(table)

text = '''A.1. hulaieลก:naฯ†uฮธ:ลกiaลกi
A.2. maraลก:mav
A.3. sialฯ‡veiลก:aviลก
A.4. evisฮธu:ลกerunaiฮธ
A.5. ลกivai
A.6. aker:tavarลกiu
A.7. vanalasial:ลกerunai:murinail
B.1. hulaieลกi:ฯ†ukiasiale:ลกerunaiฮธ:evisฮธu:tuveruna
B.2. rum:haraliu:ลกivai:epteลกiu:arai:tiลก:ฯ†uke
B.3. ลกivai:aviลก:sialฯ‡viลก:maraลกm:aviลก:aumai'''

print(text.translate(table))
Output

A.1. ๐Œ‡๐Œ๐Œ‹๐Œ€๐Œ‰๐Œ„๐Œ”:๐Œ๐Œ€๐Œ˜๐Œ๐Œˆ:๐Œ”๐Œ‰๐Œ€๐Œ”๐Œ‰
A.2. ๐ŒŒ๐Œ€๐Œ“๐Œ€๐Œ”:๐ŒŒ๐Œ€๐Œ…
A.3. ๐Œ‘๐Œ‰๐Œ€๐Œ‹๐Œ™๐Œ…๐Œ„๐Œ‰๐Œ”:๐Œ€๐Œ…๐Œ‰๐Œ”
A.4. ๐Œ„๐Œ…๐Œ‰๐Œ‘๐Œˆ๐Œ:๐Œ”๐Œ„๐Œ“๐Œ๐Œ๐Œ€๐Œ‰๐Œˆ
A.5. ๐Œ”๐Œ‰๐Œ…๐Œ€๐Œ‰
A.6. ๐Œ€๐ŒŠ๐Œ„๐Œ“:๐Œ•๐Œ€๐Œ…๐Œ€๐Œ“๐Œ”๐Œ‰๐Œ
A.7. ๐Œ…๐Œ€๐Œ๐Œ€๐Œ‹๐Œ€๐Œ‘๐Œ‰๐Œ€๐Œ‹:๐Œ”๐Œ„๐Œ“๐Œ๐Œ๐Œ€๐Œ‰:๐ŒŒ๐Œ๐Œ“๐Œ‰๐Œ๐Œ€๐Œ‰๐Œ‹
B.1. ๐Œ‡๐Œ๐Œ‹๐Œ€๐Œ‰๐Œ„๐Œ”๐Œ‰:๐Œ˜๐Œ๐ŒŠ๐Œ‰๐Œ€๐Œ‘๐Œ‰๐Œ€๐Œ‹๐Œ„:๐Œ”๐Œ„๐Œ“๐Œ๐Œ๐Œ€๐Œ‰๐Œˆ:๐Œ„๐Œ…๐Œ‰๐Œ‘๐Œˆ๐Œ:๐Œ•๐Œ๐Œ…๐Œ„๐Œ“๐Œ๐Œ๐Œ€
B.2. ๐Œ“๐Œ๐ŒŒ:๐Œ‡๐Œ€๐Œ“๐Œ€๐Œ‹๐Œ‰๐Œ:๐Œ”๐Œ‰๐Œ…๐Œ€๐Œ‰:๐Œ„๐Œ๐Œ•๐Œ„๐Œ”๐Œ‰๐Œ:๐Œ€๐Œ“๐Œ€๐Œ‰:๐Œ•๐Œ‰๐Œ”:๐Œ˜๐Œ๐ŒŠ๐Œ„
B.3. ๐Œ”๐Œ‰๐Œ…๐Œ€๐Œ‰:๐Œ€๐Œ…๐Œ‰๐Œ”:๐Œ‘๐Œ‰๐Œ€๐Œ‹๐Œ™๐Œ…๐Œ‰๐Œ”:๐ŒŒ๐Œ€๐Œ“๐Œ€๐Œ”๐ŒŒ:๐Œ€๐Œ…๐Œ‰๐Œ”:๐Œ€๐Œ๐ŒŒ๐Œ€๐Œ‰

There are a few things I was not sure about, like whether ๐Œ‘ is right (the actual character looks sideways), and the supposed ๐Œˆ's in part B look like ๐Œ‡'s. 98.170.164.88 (talk) 02:41, 13 May 2022 (UTC)[reply]

Seems best thing would be for you to compare with a scholarly transcription - here are two that don't require a journal subscription: [1] and [2] 70.67.193.176 (talk) 15:27, 16 May 2022 (UTC)[reply]
Second link gives me an error, but the first one looks valuable. Thanks. I'll report back with what I find. 98.170.164.88 (talk) 02:29, 17 May 2022 (UTC)[reply]

Pronunciation respelling[edit]

What does the letter H at the end of the syllables on the pronunciation respelling mean? For example, the Finnish name Takala is respelled TAH-kah-lah and is [หˆtษ‘kษ‘lษ‘] in IPA. How would respelling TA-ka-la then indicated in IPA? --40bus (talk) 11:17, 13 May 2022 (UTC)[reply]

It's because, per Help:Pronunciation respelling key, "a" on its own represents /รฆ/, while "ah" represents /ษ‘ห/. 98.170.164.88 (talk) 11:55, 13 May 2022 (UTC)[reply]
98.170.164.88 -- You ignored the footnote about checked syllables. The third syllable in takala is definitely open, and the second syllable is also unsuitable for a "checked" vowel, since the first vowel of a -VCV- sequence can only be "checked" if its sykllable is stressed ("caterpillar" etc). Of course, all this is irrelevant to Finnish... AnonMoos (talk) 22:40, 13 May 2022 (UTC)[reply]
You're right, I didn't think to check the footnotes. I do think that adding the "h" doesn't hurt at least, and certainly doesn't make it any more ambiguous... but I guess it's not strictly needed if /รฆ/ is impossible in that position. It has the advantage of being consistent at least, if that counts for anything. 98.170.164.88 (talk) 23:36, 13 May 2022 (UTC)[reply]
Tangent alert โ€” to some extent this question (and the responses to it) expose a problem in the very concept of respelling as used in Wikipedia. Unfortunately, Wikipedia editors who like phonology have tried to turn respelling into a rigorous codification of a subset of IPA, with certain letters or combinations of letters corresponding uniquely to IPA phonemes.
That isn't what respelling is for. The point of respelling is so that a competent speaker and reader of English doesn't have to think about phonemes. They just read it the way it's written. Some care needs to be taken to write it so there's only one way to read it, but that's the editor's problem, not the reader's.
With the "respelling key" as it currently stands, there are many combinations that are likely not to be read as intended, when a competent speaker and reader comes across them. --Trovatore (talk) 23:07, 13 May 2022 (UTC)[reply]
Dictionary.com appears to use the same key, giving the pronunciation of the word lie as [ lahy ].[3] This may be how an unwary reader would pronounce this. Several dictionaries have [ lฤซ ]. This may work if the reader is a native English speaker, but it will confuse ESL learners, the ones most in need of finding a word's pronunciation. ย --Lambiam 04:40, 14 May 2022 (UTC)[reply]
I think respellings in Wikipedia are really more for native speakers (or highly competent second-language speakers who have internalized the idiosyncrasies of "regular" English spelling). Second-language speakers are probably more likely than native speakers (or at least more likely than Americans) to know IPA in the first place, and they're less likely than native speakers to be helped by respellings that appeal to other rules of English spelling.
If you're going to make people look up what a pronunciation key means in a table, then that table should just be IPA, which allows for more nuance and precision. The advantage of respelling is that, at least in certain cases, the reader doesn't have to look anything up.
But to make that work, respelling should use more of "regular" English spelling. For example, the so-called "long vowels" /eษช/, /iห/, /aษช/, /oสŠ/, /juห/, followed by a consonant, should be represented by A<consonant>E, E<consonant>E, I<consonant>E, O<consonant>E, U<consonant>E. There aren't really good alternatives for these, especially for /oสŠ/. The "key" has /oสŠ/ as "oh", but "oh" followed by a consonant is rendered as /ษ’/ in regular English spelling. --Trovatore (talk) 20:31, 15 May 2022 (UTC)[reply]