Wikipedia:Reference desk/Archives/Language/2017 April 18

= April 18 =

Symbol for transcribing Chinese
On the left-hand page here, there's (between the two Chinese characters) a strange symbol used in one of the transcriptions (vaguely similar to a 3 or letter z). Does anyone know what it is, and what transcription system it belongs to? Thanks, HenryFlower 15:34, 18 April 2017 (UTC)
 * Looks like an Ezh, but not sure what transliteration system it comes from. --PalaceGuard008 (Talk) 15:49, 18 April 2017 (UTC)
 * Looks more like an altered yogh to me. Deor (talk) 22:47, 18 April 2017 (UTC)
 * In modern Pinyin, 字 = zì. See 字 —Stephen (talk) 22:16, 18 April 2017 (UTC)
 * Thanks -- ezh and yogh both look plausible. Ezh seems a bit closer in terms of the sounds which it's used to represent (though still not very close; on the other hand, I suspect yogh is a more common letter, which might make it more likely. I'm still curious about the system. HenryFlower 07:27, 19 April 2017 (UTC)
 * Looks like Fraktur 𝖟 of the Legge system found in his translation of I Ching. Шурбур (talk) 20:31, 20 April 2017 (UTC)
 * It's most plausible given the original value of the Chinese sound. This transcription might have come from Germans as well.--Lüboslóv Yęzýkin (talk) 17:27, 21 April 2017 (UTC)
 * It looks to me like what Unicode calls Egyptological alef; see Transliteration of Ancient Egyptian. —Tamfang (talk) 05:23, 21 April 2017 (UTC)
 * Thanks Шурбур, that does look like a good bet. Here 𝖟hui represents pinyin cui, while here 𝖟ing in pinyin jing, which are both reasonably similar. Being picky, how sure are you that the symbol which Legge uses is actually a fraktur z? HenryFlower 20:17, 21 April 2017 (UTC)
 * If you mean sources, I don't have any. Here you can see both capital and small letters. Шурбур (talk) 06:33, 22 April 2017 (UTC)
 * Thanks again -- with the upper case also, that does look a reasonable bet. I was hoping that somewhere he'd set out his system, but it doesn't seem to be anywhere in the I Jing book, at least. Incidentally I see he uses it there to represent tne initial sound in modern Qin, so that's quite a range of sounds. HenryFlower 06:47, 22 April 2017 (UTC)
 * Here's the table. Шурбур (talk) 07:02, 22 April 2017 (UTC)
 * Splendid -- cheers. HenryFlower 16:50, 22 April 2017 (UTC)

The transcription system was actually first mentioned in the first volume of The Sacred Books of the East (pp. xlviii-lv, letter #36 ), where the editor, Max Müller, referred to his earlier work, however, there I couldn't find any mention of this particular letter, but we can deduce that it's surely Fraktur German z. It's neither Cyrillic З which looks quite different, nor Ezh which was invented later and also looks slightly different.--Lüboslóv Yęzýkin (talk) 22:16, 23 April 2017 (UTC)
 * Actually I've found the direct mention that this letter is "German" (read Fraktur), though it was initially devised for Hebrew and Arabic.--Lüboslóv Yęzýkin (talk) 23:35, 23 April 2017 (UTC)
 * Thanks, that's very helpful! HenryFlower 07:49, 24 April 2017 (UTC)

Muhammad variants
I've been trying to find some work which would explain if not every but at least most well-known variants of the name in various languages/dialects. Even better if with such a work after just looking at the particular variant one could identify from what language/dialect it may have come. For example, just looking at Mehmet one could deduce that it is a Turkish variant, and looking at Mamadou, that it is a West African variant. We have tables in Muhammad (name) and Muhammad, but it's mainly unsourced and unexplained.

Let's take an example of Turkish Mehmet:

1) Initial /m/ is OK in Turkish. Unchanged.

2) But Turkish knew no long consonants, hence /mm/ > /m/.

3) Turkish knew no /ħ/, so changed to /h/.

4) Final voiced consonants in Turkish tend to become unvoiced, hence /t/.

8) The Turkish variant might have been borrowed from an Arabic dialect (Egyptian? Syrian?) where the name has had the colloquial form [mæħæmmæd], and as Turkish knew no /æ/, hence /æ/ > /e/.

6) Vowels in penultimate syllables in Turkish might tend to be elided (not sure about that), hence we have the final variant Mehmet.

Or let's take Latin Mahometus. Here I haven't got explanations but questions. Why the first vowel is /a/ and the second /o/? Why /-et-/? Is it because it was derived from some language/dialect which had had such a form? What was that? Why in that language/dialect did the name have such a form? And so on. --Lüboslóv Yęzýkin (talk) 19:35, 18 April 2017 (UTC)


 * I would assume that "Mehmet" is probably the result of blending with Arabic aħmad. There were actually much more altered/corrupted forms floating around in the middle ages, such as Mahound... AnonMoos (talk) 20:49, 18 April 2017 (UTC)


 * In Saracens: Islam in the Medieval European Imagination by John Tolan, he lists Mamed (actually Greek), Machmit, Machometus, Machomis, Mahmet, Mahom, Mahomes, Mahomet, Mahons, Mahoumet, Mahound, Mahummet, Malphumet, Mathomus, Maumette. So the Greeks heard the final -d, but in Latin and other western European languages, it almost always became a -t or an -s. (I see a couple of other spellings with a -t in the 12th-century chronicle of William of Tyre: Mahemet and Mahometh.) The name may also be the origin of the word "Baphomet". Adam Bishop (talk) 00:08, 19 April 2017 (UTC)
 * Thanks! I didn't expect such an overwhelming variation already in such early times. It seems to be impossible to explain every variant, they simply could not make up their mind how to write that name! Any proper explanation eludes me. However, it just has struck me that the second /o/ may be a result of the assimilation to the pharyngeal /ħ/. The first /a/ may be a result of a Maghrebi dialect, where short /u/ is often reduced to a schwa which could be heard as /a/ to Europeans. Other letters are beyond the explanation, they're simply idiosyncratic. Though, I look forward for any further inputs, especially I'm interested in the variation within Arabic itself and its dialects, which I know a little but not enough, and in some languages of Islamic nations (Turkish, Persian, Urdu, etc.). Looks like I've found a tabula rasa in onomastic.--Lüboslóv Yęzýkin (talk) 17:23, 21 April 2017 (UTC)

A > E sound change
Ae there any examples of low vowels being raised mid vowels, like e, e̞, or ɛ? I have been told this type of sound change is rare but I was never told why. And if there are sound changes like this, is the reverse more common? Idielive (talk) 23:52, 18 April 2017 (UTC)


 * A contextual change of "a" to "e" under the influence of an "i" vowel or "y" semivowel in the next syllable is common in some Germanic languages. It can be seen in English "men" as the plural of "man", and was the origin of the "ä" spelling in Modern German... AnonMoos (talk) 03:19, 19 April 2017 (UTC)


 * There are quite a few Latin verbs where an a in the root becomes e or ē in the perfect stem. Examples are facio ("I make"), perfect fēcī; iacio ("I throw"), perfect iēcī; stat ("he/she stands"), perfect stetit. When verbs with an a take a prefix, in the present stem the vowel often becomes i, but in the supine (root of the past participle) the a often becomes e. Examples factum ("made", supine of facio), but confectum ("prepared, made up", supine of conficio); iactum ("thrown", supine of iacio) but iniectum ("thrown in", and other derived meanings, supine of inicio). --ColinFine (talk) 21:36, 19 April 2017 (UTC)


 * The change of vowel in compound verbs is probably a relic of a first-syllable-stress era in early Latin and/or pre-Latin (we have something about it at Latin spelling and pronunciation... AnonMoos (talk) 11:23, 21 April 2017 (UTC)


 * Among the most straightforward examples of a > e is of course Middle English long /a:/ in the Great Vowel Shift. Another example is Indo-European /a:/ in Greek, which (in most environments) became /ɛ:/ in classical Attic Greek, and then further raised through /e:/ to /i/ in post-classical Greek. William Labov, in Vol. 1 of his Principles of Linguistic Change, has an illuminating discussion of why vowels in certain phonological constellations tend to rise along the a>i axis. On the other hand, the Latin alternation between a and e in words like facio/feci as mentioned by ColinFine is not really a reflex of an a>e sound change, but an effect of Indo-European ablaut. Fut.Perf. ☼ 21:58, 19 April 2017 (UTC)


 * The stetit formation is an example of reduplication, of which there is quite a lot in Latin. 86.168.123.128 (talk) 09:21, 20 April 2017 (UTC)


 * Reduplication is less common in Latin than in ancient Greek or Sanskrit, and the steti perfect is an obscure example (listed as reduplication in the Gildersleeve and Lodge grammar, but not conforming to a traditional Indo-European pattern of reduplication, as far as I can see)... AnonMoos (talk) 11:18, 21 April 2017 (UTC)


 * Many Turkic languages (Turkish, Kyrgyz, Uzbek, Yakut) changed /æ/ into /e/. In some /æ/ has a limited use, as in Kazakh where it's mostly reserved for Perso-Arabic loanwords, while in Kazakh words it's surfaced to /e/, particularly in suffixes, e.g. the plurality suffix has variants -lar/-ler, while compare it with Tatar -lar/-lär. Some Finno-Ugric languages know this as well, like more archaic Moksha /æ/ corresponds to Erzya /e/; it also happened in Hungarian and in the Permic languages.--Lüboslóv Yęzýkin (talk) 18:06, 21 April 2017 (UTC)
 * I also saw your previous question but didn't answer in time. Again from Turkic: in Tatar and Bashkir the Common Turkic vowels e, ö, o correspond to i, ü, u, and vice versa, i, ü, u,  ï (IPA /ɯ/) to e, ö, o, ë (/ɤ/). E.g., Kazakh ot, Tatar ut for "fire". --Lüboslóv Yęzýkin (talk) 18:22, 21 April 2017 (UTC)