Jump to content

Wikipedia:Reference desk/Archives/Language/2012 September 10

From Wikipedia, the free encyclopedia
Language desk
< September 9 << Aug | September | Oct >> September 11 >
Welcome to the Wikipedia Language Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.


September 10

[edit]

Hindi v/w

[edit]

What are the rules for when it's a v and when it's a w? --168.7.236.92 (talk) 03:56, 10 September 2012 (UTC)[reply]

It seems to be a matter of either personal choice, or pronounceability. Our article on the subject has some information, if you have not already seen it. KägeTorä - (影虎) (TALK) 10:46, 10 September 2012 (UTC)[reply]
Our article does say they're conditional allophones and the choice between them depends on context in some situations, but it doesn't say what the contexts are. It just says that native Hindi speakers use [v] in [vrət̪] and [w] in [pəkwɑːn], and that they're in free variation in [əd̪ˈwɛːt̪ ~ əd̪ˈvɛːt̪] but doesn't go into any more detail. Angr (talk) 20:12, 10 September 2012 (UTC)[reply]
In Hindi, the sound represented by व is actually neither /v/ nor /w/ but /ʋ/, the Labiodental approximant (for the OP: a sound that is kinda half way between "w" and "v") and to "Hindi ears" /v/ and /w/ are the same sounds. All three allophones are in free variation in some positions but tend toward /w/ in on-glide positions (i.e. before vowels) and /v/ in other positions, especially before consonants where a /w/-colored pronunciation could sound like a vowel.
Remember, though, that not all languages with allophones have them in free variation (i.e. you can't always "pick" which one you'd like to use). The Korean letter ㄹ, called 리을 (riŭl), may represent either an R or L sound, but which sound it actually makes is governed by its position in a word and the things around it. Kim Jŏng-il's name in Korean, 김정일, would never be read "Kim Jŏng-ir", and the Korean folk song Arirang, written 아리랑, would never be "Alilang". And if you happen to be wondering about the surname "Park", there is actually no ㄹ at all in it; it is written 박 (pak). I think the R was added as a result of the way it sounds to British ears, being a rough homophone of the English word "park".  dalahäst (let's talk!) 04:53, 11 September 2012 (UTC)[reply]
Yup. There's no /r/-like sound in the original versions of the words Juggernaut, nark, Burma or Myanmar either; they all got added by non-rhotic speakers. Angr (talk) 12:21, 11 September 2012 (UTC)[reply]
Or non-hotic, as we must therefore pronounce it. KägeTorä - (影虎) (TALK) 13:53, 12 September 2012 (UTC) [reply]

Russian translation

[edit]
Napoleon with Satan after burning Moscow (19th century)

Would a Russian speaker please provide a transcription of the text in the illustration on the right and an English translation of it, for inclusion on the Commons file description page? Thanks. — Cheers, JackLee talk 15:02, 10 September 2012 (UTC)[reply]

I've transliterated the text as if it were written in modern Russian (it is actually written in pre-reform Russian):
Top:
Наполеон с С... после сожжения Москвы [Napoleon s S... posle sozhzheniya Moskvy] [Napoleon with S... after burning Moscow]
Bottom:
Все треснули теперь предположенны планы; [Vse tresnuli teper' predpolozhenny plany;] [All plans devised earlier have gone for naught;]
Все угрожают мне, мной покорённы страны; [Vse ugrozhayut mne, mnoy pokoryonny strany;] [Everybody threatens me, all the countries conquered by me;]
Всё взяло ход иной — всё стало изменять; [Vsyo vzyalo khod inoy — vsyo stalo izmenyat';] [Everything has gone the other way—everything has begun to change;]
Тиранить всё меня, — пропал я, — что начать! [Tiranit' vsyo menya, — propal ya, — chto nachat'!] [Everything is tormenting me—I'm busted—what should be started!]
Note it's not the best possible translation; I'm only posting it so you'd have something if nobody else replies. Also note that "S..." in the top line stands for "Satan".—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); September 10, 2012; 16:46 (UTC)

It seems that the title "French crow soup" given on the Wikimedia page is actually the title of another Russian cartoon.[1] Alansplodge (talk) 16:48, 10 September 2012 (UTC)[reply]

Ah, thanks! I've renamed the file so it's more accurately named. — Cheers, JackLee talk 17:19, 10 September 2012 (UTC)[reply]
Thanks! Could you provide the Cyrillic transliteration as well? I'll add that to the file description page. Also, what are the inscriptions on the left and right side under the verse? — Cheers, JackLee talk 17:01, 10 September 2012 (UTC)[reply]
You mean original Russian text? I've added it to the text above (please let me know if that's not what you were asking). Once again, it's in modern Russian, not in pre-reform Russian.
I can't make out the text on the left under the verse as it's too faded (perhaps it is an abbreviation of some kind?). The line on the right contains this:
из Разг. Н.... с С..... страница 20 [iz Razg. N.... s S.... stranitsa 20] [from the conv{ersation} of N{apoleon} with S{atan} page 20]
Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); September 10, 2012; 17:28 (UTC)
Yes, that's what I meant. Thanks a lot! By the way, is Разг. Н.... с С..... the name of a book? — Cheers, JackLee talk 07:35, 11 September 2012 (UTC)[reply]
From what I've been able to find, it's the title of the image. By the by, according to this, its authorship is attributed to Ivan Ivanov. According to the same source, the images were published in the Syn Otechestva (Son of the Fatherland) magazine.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); September 11, 2012; 12:09 (UTC)
Isn't 'Ivan Ivanov' (= 'John Johnson', 'Jan Janssen) a conventional way of saying 'Anonymous', like the English Joe Bloggs, Alan Smithee, etc? AlexTiefling (talk) 14:11, 11 September 2012 (UTC)[reply]
Wow, great. I updated the file description page. Please check if I did it correctly. Also, I meant to ask what the last line of the poem ("... I'm busted—what should be started!") means. What is meant by "busted", and "what should be started"? — Cheers, JackLee talk 15:28, 11 September 2012 (UTC)[reply]
I think the file description page should use the original spelling rather than modern spelling, but I'm not confident enough of my ability to read the old letters in this handwriting-like style to do it myself. Angr (talk) 15:35, 11 September 2012 (UTC)[reply]
Yes, I think that would be a good idea. Ezhiki, can you help? — Cheers, JackLee talk 18:17, 11 September 2012 (UTC)[reply]
I'd love to, but I can't seem to get the old Russian letters to work on the computer I'm on :(
As for the last line of the poem, I used "busted" because it's shorter than "I'm done with", "I'm in a desperate situation with no escape", or "they won" and also more literal. As for "what should be started", I have no idea what it means in Russian (it doesn't follow from the context), so I gave a very literal translation (alternatively, you could use "what should I start"). To me it almost looks like a nonsensical phrase added so the line would rhyme, but the truth is I just don't know.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); September 12, 2012; 17:46 (UTC)
Ah, too bad! The letters don't appear in the "Special characters" insertion window? Maybe someone else can help with the Russian transcription. As for the translation of the last line of the verse, perhaps "I'm done for" would be a better translation than "busted". "Busted" sounds very American, and brings to my mind people being arrested by cops! — Cheers, JackLee talk 20:18, 12 September 2012 (UTC)[reply]
It's apparently a problem with my fonts. I can't even copy/paste the yat letter from that article or elsewhere on the Internet without my browser acting all sorts of weird (although I know I was able to do it in the past... go figure). As for the translation, feel free to replace it with whatever sounds more natural. I live in the US, and "busted" is used around me in all kinds of situations, formal and informal (not only when someone's is arrested by cops :)), but I have no idea how it is perceived by non-Americans. Cheers,—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); September 13, 2012; 15:47 (UTC)

It seems that Ivan Ivanoff isn't (in this case) "Joe Bloggs" but a leading lubok artist of the period. "Ivan Alekseevich Ivanov also moved away from his training in 1812, inspired by Terebenev’s example. Ivanov was born in 1779 in Moscow and entered the Academy at Petersburg in 1789. Upon his graduation, Ivanov was awarded a certificate of the first degree and the rank of artist [khudozhnik]... Shortly after Napoleon’s invasion, early in 1813, Ivanov took up a position at the Imperial Public Library (now the National Library in St. Petersburg). It was here, in the capital, sometime in 1812-1813 that he saw Terebenev’s prints and decided to express his own patriotic outrage by the same means."[2] (pp. 10-11). Alansplodge (talk) 14:21, 12 September 2012 (UTC)[reply]

language problem

[edit]

why dont you keep all the languages for all the topics 117.192.53.9 (talk) 15:33, 10 September 2012 (UTC)[reply]

Could you explain what you mean more clearly? Thanks. — Cheers, JackLee talk 15:38, 10 September 2012 (UTC)[reply]
I will guess that the OP is referring to us not having articles in their language (IP geolocates to India) for every topic. This is obviously because they just haven't been written yet. The OP can feel free to make the articles themself, either by translating or by creating new articles with research and relevant references. KägeTorä - (影虎) (TALK) 16:16, 10 September 2012 (UTC)[reply]
The OP perhaps assumes that whenever an article is written in one language arm of Wikipedia, the system should immediately create corresponding articles in all the other languages we host. That would be a nightmare to administer. And then what about subsequent changes? Would they all have to be replicated every time? Imagine an edit war being conducted on not just one article but on 250-odd different-language versions of the same article. And we think we sometimes have a tough time now. The fact is that each language version of a particular topic is not merely a translation of the original (except occasionally), but a completely separate article written by different people, at different times, using perhaps some of the same sources but usually mostly different sources. That's why basic details such as the subject's date of birth can vary wildly between different versions of the same "article". They're not really the same "article" at all, but different articles about the same person or subject. -- ♬ Jack of Oz[your turn] 01:15, 11 September 2012 (UTC)[reply]
Theoretically, this would be possible, if machine translations were of sufficient quality. However, current translation programs are woefully inadequate. When they can translate as well as a human, that's when we can offer all articles in all languages. StuRat (talk) 08:50, 11 September 2012 (UTC)[reply]
And when that day comes, I'll have lots of time to work on Wikipedia articles since my job as a translator will be taken over by a computer and I'll be laid off. Angr (talk) 12:16, 11 September 2012 (UTC)[reply]
Don't worry, angr, it will never happen in our lifetimes, and even if it does, proofreaders will always be needed. Just means you'll need a higher volume to keep your current salary. KägeTorä - (影虎) (TALK) 14:04, 11 September 2012 (UTC)[reply]
Actually, I'm already about 75% proofreader and only about 25% translator. Mostly all I get to do is proofread the work the freelance translators do. Angr (talk) 14:23, 11 September 2012 (UTC)[reply]
I don't think we are all that far off. A decade or two ought to be enough, at least for high quality machine translators between common languages. Obscure languages won't have high quality machine translators until much later, if ever. StuRat (talk) 18:57, 11 September 2012 (UTC)[reply]
People have been thinking it might be a "decade off" since the 1950's, but so far it hasn't been. See AI winter#Machine translation and the ALPAC report of 1966 etc. -- AnonMoos (talk) 21:56, 11 September 2012 (UTC)[reply]
The difference now, besides improved processing speed, is that we have products marketed with voice recognition features. Figuring out what people are actually saying is an important part of both voice recognition and machine translations. The current versions that just look at each word in isolation don't do a good job, so won't do well in the market, once this becomes common knowledge. Therefore, considerable resources will be devoted to solving this particular problem. StuRat (talk) 22:14, 11 September 2012 (UTC)[reply]
Which has no bearing on this discussion since we're talking about written translations, not spoken interpreting. Angr (talk) 22:17, 11 September 2012 (UTC)[reply]
I disagree. In both cases the program needs to examine the grammar to try to figure out what each word means in context: "Find a store where I can buy a TV" versus "Find a place where I can store my TV". StuRat (talk) 22:44, 11 September 2012 (UTC)[reply]
I wonder how a machine translator would deal with a sentence like "It's food". Humans generally interpret that as an abbreviated way of saying "It is food", although in context it could well be supposed to mean "The food belonging to it", because we know from experience that certain notorious people spell the possessive pronoun its as it's. We humans are smart enough to deal with this, but could machines ever be educated to recognise when "it's" is meant to mean an abbreviation and when a possessive pronoun? If the entire sentence was "It's food", how well can a machine read the surrounding context and work out what the writer must have meant? This isn't trivial because, while the two words differ by (only) an apostrophe in English, they're rendered quite differently in other languages. -- ♬ Jack of Oz[your turn] 01:41, 12 September 2012 (UTC)[reply]
This is nothing special, given that the general problem of distinguishing identically-written words with different meanings exists.81.159.108.116 (talk) 02:52, 12 September 2012 (UTC)[reply]
"Did you mean: its food?" [3]  Card Zero  (talk) 10:08, 12 September 2012 (UTC)[reply]
No, I meant: it's food. But your question brings up the other issue, of people omitting the apostrophe when they should be putting it in. Either way, the waters have been well and truly muddied. Once upon a time, it would have been safe to interpret It's food as an abbreviation for It is food, without any reliance on the context, and it would have been safe to interpret Its food as meaning The food belonging to it, again without any reliance on the context. But now, who would ever know? Some people never or rarely use any punctuation, while others are deliberately rewriting the rules to insert apostrophes where they're not required. We absolutely need the context now to work out what either of these statements means. Progress? You tell me. How well can a machine read the context and work out the meaning? What are the instructions it's given? -- ♬ Jack of Oz[your turn] 22:43, 12 September 2012 (UTC)[reply]
A: "What's that?" B: "Its food." A: "What's food?" B: "We have an article on that." KägeTorä - (影虎) (TALK) 06:25, 14 September 2012 (UTC)[reply]
Congratulations, you win the inaugural Who's on First Award, KageTora.  :) -- ♬ Jack of Oz[your turn] 08:31, 14 September 2012 (UTC)[reply]
I agree with Jack. Some of my kids from the primary school I teach part-time at here in Hungary speak to me on Facebook, and some of them only speak Hungarian to me. I only speak a little Hungarian, so I have to rely on Google Translate (which churns out gibberish at the best of times) to work out what they are saying. Besides the machine translation, there are two other problems: 1) many of them use 'hip' spelling that all of their friends use, and 2) there is very little punctuation. This leads to complete incomprehensibility by Google Translate, or worse, terrible misunderstandings. As an example, one kid asked me something. She meant to say "Do you teach English to the slower students?", but Google translated it as something which I interpreted as "Are you the retarded English teacher?" (with some extra letters added). Just to make sure of the meaning, I had to Skype someone to ask before I replied to her. Even my 70-year-old mother uses LOL speak when she sends me an SMS. I always reply in properly spelled, well-punctuated English. Having said that, Google Translate does a very good job translating 'LOL' or 'ROFL' or 'LMAO' into Japanese. Makes you wonder what they are doing in Sun Valley.KägeTorä - (影虎) (TALK) 07:24, 13 September 2012 (UTC)[reply]
I've seen that syndrome before -- native French-speakers posting a message in ungrammatical misspelled French without accents to an English-language forum, and then English-speakers who don't know French running it through Google translate and not getting much of anything... AnonMoos (talk) 15:29, 13 September 2012 (UTC)[reply]

Even humans are not perfect machine translators. Once computers start learning on their own, they may outpace us. At this point, it would require several literary and computing geniuses just to lay out the problem of how to design a good computer translator, not to mention solving it. See Jeff Hawkins' On Intelligence for an excellent speculative work on the issue. This is one of the better books of the century so far. μηδείς (talk) 04:21, 12 September 2012 (UTC) /span>]][reply]