Wikipedia:Reference desk/Archives/Language/2012 October 6

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.

October 6

Hindi/Sanskrit pronunciation

How can I make all the hard sounds in Hindi/Sanskrit? (aspirated/unaspirated, retroflex, palatal) --168.7.239.19 (talk) 03:44, 6 October 2012 (UTC)

"Ph", "th", "kh" are fairly easy if you're a native English speaker, since any [p], [t], [k] in English which is followed by a stressed vowel, and is not preceded by [s], is almost always aspirated. If you can pronounce the [t] of "stop" without the [s] and contrast it with the [t] of "top", then you're on your way with being able to produce the "t"-"th" distinction... AnonMoos (talk) 06:28, 6 October 2012 (UTC)

Retroflex is not hard: just turn the tip of your tongue backward. (The hard part, for us anglophones, is making the dentals truly dental; our alveolar "dentals" sound to Indians like retroflex.) —Tamfang (talk) 07:47, 6 October 2012 (UTC)

Also, "pure" vowels are very hard for English speakers. I find it particularly hard to stop 'ए' sounding like "ay" -- Q Chris (talk) 09:54, 6 October 2012 (UTC)

Just say 'air' without the 'r'. KägeTorä - (影虎) (TALK) 06:15, 9 October 2012 (UTC)

That would work for me (British near-RP), but my wife (Texas-USA) would still pronounce it "ay"! -- Q Chris (talk) 09:13, 9 October 2012 (UTC)

Sure she's not from Canada, eh? :) KägeTorä - (影虎) (TALK) 11:45, 9 October 2012 (UTC)

One thing I did with my kids in China to show the difference between aspirated and unaspirated was to give them little strips of paper, get them to put them in front of their mouths, and say 'p', 't', and 'k' (and a few others), and see who made the paper move the most. KägeTorä - (影虎) (TALK) 13:50, 7 October 2012 (UTC)

When I was an undergrad, I took Sanskrit from a professor who carefully produced, and taught us to produce, the difference between aspirated and unaspirated stops and between dentals and retroflexes. Then when I was a grad student at a different university, I took Sanskrit again from a different professor, who told us on the first day: "Sanskrit has four different kinds of t: [tʰə], [tʰə], [tʰə] and [tʰə]", pronouncing all four identically as slightly aspirated alveolars. I think he genuinely thought he was making the difference. Angr (talk) 11:50, 9 October 2012 (UTC)

γγ γκ γξ γχ

My Shelmderdine Greek book says that when these consonants appear together, the γ becomes a [ŋ]. My instructor, when asked, said he kind of knew that was the case but that he wouldn't really ever do it. So: is Shelmerdine correct, and if so, common? 67.164.156.42 (talk) 04:27, 6 October 2012 (UTC)

Also, γμ. One piece of evidence is that γγ γκ γξ γχ are transliterated as NG NC NX NCH into Latin (αγγελος = ANGELUS etc.). For detail on reconstructing ancient Greek pronunciations, you can look at Vox Graeca by Allen. Wikipedia article is Ancient Greek phonology... -- AnonMoos (talk) 06:08, 6 October 2012 (UTC)

So stigma ought to be stiŋma?! —Tamfang (talk) 07:45, 6 October 2012 (UTC)

The velar nasal sound is sometimes called the "angma" based on αγμα. Also, when a verb with a stem ending in a labial stop comes directly before an "m" ending in ancient Greek, the result is "mm": καλυψω / κεκαλυμμαι. The analogous case with velar-final stems is πραξω / πεπραγμαι. Anyway, orthographic "gn" in ancient Latin was [ŋn] at least some of the time... AnonMoos (talk) 11:46, 6 October 2012 (UTC)

I don't mean so much in Classical times, I mean in modern classical education and scholarship. 67.164.156.42 (talk) 10:22, 6 October 2012 (UTC)

If you mean modern pronunciation, then you don't have to worry about γμ; but for the others, γκ γξ γχ would be very difficult to pronounce if γ is not nasal, and for all of γγ γκ γξ γχ the nasal pronunciation is consistent with Greek loanwords into English (angel, sphinx, elenchus etc.). AnonMoos (talk) 12:03, 6 October 2012 (UTC)

Notice that in English, we write the /ŋ/ with <n> before a velar, as in "bank": we are using a letter to represent a different sound which is similar to its primary sound in one respect (nasality) but different in another (place of articulation). The Greeks appeared to have done something similar and used a letter <γ> whose primary sound was similar in one respect (place of articulation) but different in another (nasality). The Romans also did something similar in some contexts: there is considerable evidence that the <g> in words like "cognosco" was pronounced /ŋ/ rather than /g/. --ColinFine (talk) 13:27, 6 October 2012 (UTC)

Stop clusters

Looking at #γγ γκ γξ γχ, it seems to me that the problem is with these clusters is that they're all velar stops. I tried to pronounce the [ɡ], [k], [x] and [ks] with the [g] of the gamma and couldn't. So I wondered, is it generally the case that stops in the same POA cannot be pronounced together? Is this just a restriction I have due to my native English? (As a total sidenote, how is the order of the IPA symbols in the insert tool on the edit page determined?) 67.164.156.42 (talk) 04:27, 6 October 2012 (UTC)

[gk], [gkʰ] are unlikely clusters, since the adjacent consonants differ only in voicing. However, [gg] is not problematic, and can be found in many languages. AnonMoos (talk) 06:23, 6 October 2012 (UTC)

[gkʰ] occurs in English at syllable junctures (usually morpheme junctures), as in "egg-cup". Given the English derivatives that AnonMoos mentions, I think you are just making your life unnecessarily difficult by trying to pronounce these clusters with /g/. --ColinFine (talk) 13:27, 6 October 2012 (UTC)

Sorry, but in "egg-cup" this is at the boundary between two stems of a compound (where any series of consonants which can occur at the end of English syllable can be followed by any series of consonants which can occur at the beginning of an English syllable), and so does not count as an ordinary (stem-internal) intervocalic consonant cluster... AnonMoos (talk) 18:48, 6 October 2012 (UTC)

If we are talking about Greek here, the rule is that gamma before a velar stands for eng, not for gamma. Those orthographic clusters are pronounced /ŋg/, /ŋk/, /ŋks/, and /ŋx/. (OH, I see this was a subhead, not a main question--I read these threads bottom to top.) μηδείς (talk) 17:04, 6 October 2012 (UTC)

Person through relative clauses

To what extent is grammatical person preserved in verbs across relative clauses? I was going to list two examples, but I can't find them to post. The first is from the Aeneid. Someone is talking to someone else (one of the gods, I think) and however it goes, there's qui clause with a second-person verb inside it. The second is from a poem of Ángel González Muñiz that I can't for the life of me remember the exact words to, so I can't find it. But the gist was that the verb was first plural then a que clause and inside it was also a 1P verb.

It seems to me that relative pronouns, like qui or que or who, function as nominatives in the relative clauses. So the verb should be third, yes? It would sound somewhat strange to me in English to say "You, who eat five oranges a day, should be healthy as an ox." rather than "You, who eats five oranges a day, should be healthy as an ox." So is it that Latin and Spanish do it where English does not, or that I am just wrong and English does it all the time? Or just what is the deal with this? 67.164.156.42 (talk) 04:27, 6 October 2012 (UTC)

English historically used to do it, and still does in certain cases ("I, who am..."), but third person now occurs in many other cases. Jespersen has examples in volume VII of his grammar. The Spectator has the sentence "I am one who live in a continual apprehension", but that would be "lives" in modern English... AnonMoos (talk) 07:07, 6 October 2012 (UTC)

Conservative though I am, I'd expect lives there because its subject is one, not I. —Tamfang (talk) 07:43, 6 October 2012 (UTC)

I agree with Tamfang: I would use "I am one who lives...." because the subject of the relative clause is "who", of which the antecedent is "one". And, contrary to the OP, I find it natural to say "You, who eat five oranges a day,...." because the subject of the relative clause is "who", the antecedent of which is "you". Duoduoduo (talk) 14:24, 7 October 2012 (UTC)

It could be considered "attraction" to a logical first-person subject (apparently possible in the English of 300 years ago, but not today). By the way, the article Attraction (grammar) is strangely narrow and limited... AnonMoos (talk) 19:00, 7 October 2012 (UTC)

A contentious issue of biblical exegesis turns on just this question for Hebrew. In Exodus 3:4, Hebrew: אֶהְיֶה אֲשֶׁר אֶהְיֶה, lit. 'I am/will-be which/who/the-one-who I am/will-be' is traditionally translated "I am who I am" or "I am that I am", but some scholars prefer "I am the one who is". Part of the argument in favour of the latter is that Hebrew does have the rule that the verb in the relative clause agrees with the subject of the main clause, and therefore that this can be analysed as containing a full relative clause: on this interpretation, it is as if it said "I am the one who am", though as Tamfang says this is barely possible in Modern English. The traditional interpretation does not treat it as a relative clause at all.

A paper which discusses (and rejects) this analysis is Albrektson, Bertil (1968). Peter R Ackroyd; Barnabas Lindars (eds.). "On the syntax of אֶהְיֶה אֲשֶׁר אֶהְיֶה". Words and Meanings: Essays presented to David Winton Thomas. Cambridge University Press. ISBN 0 521 07270 0.. --ColinFine (talk) 14:16, 6 October 2012 (UTC)

Metamorphoses

How is Ovid's poem pronounced in English? I feel like it should be said /mɛ.tə.mɔr.'foʊ.sizˌ/ instead of /mɛ.təˈmɔr.fə.siz/. I don't really know why. 67.164.156.42 (talk) 04:27, 6 October 2012 (UTC)

In Latin, the stress is on the penultimate syllable because it is long: /mɛ.ta.mɔr.ˈfoː.seːs/. English usually follows the Latin stress for scholarly words like this, but in this case for whatever reason, the stress is usually shifted back a syllable. My Longman Pronunciation Dictionary gives both but puts the third-syllable pronunciation first. Here's what the OED says: "The alternative pronunciation with stress on the fourth syllable is apparently first mentioned as occurring in British usage by H. W. Fowler Mod. Eng. Usage (1926), who describes it as ‘more regular’ and as ‘still often heard’ (this impression of its former currency appears to be mistaken)." Both these sources are talking about the common noun. I could imagine that classical scholars might decide to stick to the Latin stress when talking about Ovid. Lesgles (talk) 05:25, 6 October 2012 (UTC)

I've only ever heard it pronounced with the stress on the penultimate syllable, as the title of the work, in both scholarly and non-scholarly discussion. But, I think if I were meaning "the plural of metamorphosis" in a natural English sentence, I might well produce metamorphoses with a stress on the third syllable. 86.159.77.170 (talk) 18:56, 7 October 2012 (UTC)

Payment B/G

What does "Payment: B/G" mean? bamse (talk) 19:07, 6 October 2012 (UTC)

Possibly bank guarantee? Difficult to make suggestions without more context.--Shantavira|^{feed me} 20:08, 6 October 2012 (UTC)

Possible. As for context, it is on an offer (i.e. somebody is offering a product and I might buy it), but I don't want to put the offer online here. bamse (talk) 20:30, 6 October 2012 (UTC)

Given the context, a Bank Guarantee seems like the most obvious answer. Commercial offers are usually accompanied with a bank guarantee requirement. _{Hisham1987 (talk) 20:46, 6 October 2012 (UTC)}

Latter part/second half of century

"Latter part" = "second half" in a sentence "In the latter half of the 8th century..." ? bamse (talk) 20:06, 6 October 2012 (UTC)

Correct, although perhaps later than 751, if we're talking about a specific event. I'm thinking more like 780 or 790. StuRat (talk) 20:44, 6 October 2012 (UTC)

Agreed; the year 751 would be best described as "mid 8th century". You could also say "late 8th century" for dates towards the end of the scale. Alansplodge (talk) 23:22, 6 October 2012 (UTC)

Of course, if we are talking about something with a wide range of years, like the reign of a king, it might extend from 751-799, and this could be called "the latter part", although I might specifically call it "the latter half". StuRat (talk) 00:01, 7 October 2012 (UTC)

Thanks. And if I don't know what the author mean when writing "latter part", what should I assume? bamse (talk) 08:04, 7 October 2012 (UTC)

Well, perhaps the last third might be a good middle-of-the-road interpretation. StuRat (talk) 08:08, 7 October 2012 (UTC)

My interpretation is that the author was being deliberately vague, so he didn't mean for us to infer anything specific. Duoduoduo (talk) 14:29, 7 October 2012 (UTC)

If we are going to be very grammatical, latter is comparative, and therefore indicates that there are only two things being compared, and this is indeed confirmed by the sentence, as it indicates the century has been split in two halves. As Duoduoduo points out, the author is (probably) being delibaretly vague, as it might be difficult to date exactly when something happened in the 8th century, and therefore uses vague language. As we (on the ref desk) do not know what event is being discussed in that sentence, we can't know whether what was happening might reasonably require 50 years to complete, or if it's just vague dating. I would agree with Alan and StuRat, though, as a general rule of thumb, understanding it as the last decades of the century might be the most accurate. (And, you will notice that I have used specifically vage language here, as I haven't quantified how many decades you should count.) V85 (talk) 22:37, 7 October 2012 (UTC)

"Boost your computer's start-up time" ?

It's clear from the context that they mean to say they will reduce your computer's start-up time, but it certainly sounds like it means the opposite, to me. Is there any way "boost" can properly be used to mean "reduce" ? I saw this in a TV ad for http://www.cleanMyPC.com (they don't seem to mangle English that way on their web site). StuRat (talk) 20:54, 6 October 2012 (UTC)

It seems a confusion between "boost your computer's performance/efficiency" and "reduce your computer's start-up time". -- Jack of Oz ^[Talk] 21:01, 6 October 2012 (UTC)

People take boost to mean increase in speed, not just increase. μηδείς (talk) 21:07, 6 October 2012 (UTC)

Yes, that's right. If they'd said "boost your computer's start-up speed", rather than "... time", there would have been no issue. They should know that IT geeks are usually scientifically literate, and know that speed and time are in inverse relationship. -- Jack of Oz ^[Talk] 21:15, 6 October 2012 (UTC)

I know you are not taking my pointing out that sloppy thought exists as defending it. soon enought the OED will report a sense of boost which means to better, not just to increase. μηδείς (talk) 21:53, 6 October 2012 (UTC)

That already exists. To boost the performance of something usually means to better it, in some sense. Bettering the performance of computers can include making them work faster. It can also include making them do more things, ideally in no more time than it used to take to do fewer things. -- Jack of Oz ^[Talk] 22:05, 6 October 2012 (UTC)

Soon enough? The 1989 edition says: To hoist; ‘to lift or push from behind (one endeavoring to climb); to push up. (Low)’ Webster. Also fig. To assist over obstacles, to advance the progress of; to support, encourage; to increase (in value, reputation, etc.); to praise up, to extol. I think the original meaning involved assisting a climber to gain height, and both the meanings you suggest are figurative uses of that. I can't see that either of them is particularly new. Marnanel (talk) 02:42, 7 October 2012 (UTC)

I was using "soon enough" ironically. μηδείς (talk) 18:26, 7 October 2012 (UTC)

The same ad promises "Your computer will be 100% faster". I'm not quite sure how to take that. Takes half as long to do the same task ? StuRat (talk) 23:59, 6 October 2012 (UTC)

Or no time at all? HiLo48 (talk) 03:08, 7 October 2012 (UTC)

No, that would be 100% less time. μηδείς (talk) 20:51, 7 October 2012 (UTC)

I don't see any issue with that one. Currently your computer "goes" at a certain "speed". They claim it will go at twice that speed, so, yes, it will take half as long to do the same task. So they claim. -- Jack of Oz ^[Talk] 06:28, 7 October 2012 (UTC)

I think the problem is that "fast" is not quite synonymous with speed. "Fast" and "slow" are qualitative judgements of speed. If asked "How fast are you going ?", a perfectly valid answer is "Very fast", while "What is your current speed ?" demands a numeric response. (In response to the "How fast are you going ?", you could give a numeric response, but, to me, this means "I'm going 100 miles per hour, so you decide for yourself if that's fast or not".) Therefore, "100% faster" or "100% slower" are somewhat meaningless, like saying "100% colder", "100% better", or "100% prettier". StuRat (talk) 21:05, 7 October 2012 (UTC)

Interesting. I see what you're getting at, but I'd cut through it to what I said above. Whatever your computer's current performance speed may be, the claim appears to be that the product will increase it by a factor of 100%. Problem is, I'm sure there's more than one way of defining and then measuring a "computer's speed", and I'm not sure any one product can improve all of those metrics, or even any single one of them, by exactly 100% sight-unseen. That would be my main issue with believing such a claim, but I understand what they're driving at. If the product had claimed to "greatly" or "significantly" improve your computer's speed, would that have posed a problem of comprehension? If not, why does a claim of 100% pose a problem of comprehension?

If a computer's speed is numerically measurable in any meaningful way, then it doesn't matter that an owner might choose to describe it as "fast" or "slow". We can also describe a moving car as "fast" or "slow" in relation to some yardstick (maybe stated, maybe not), but if we're doing meaningful comparisons we always have recourse to measuring it exactly, numerically. The same applies to computers, as long as it's possible to measure their "speed" in a rigorous way. -- Jack of Oz ^[Talk] 21:41, 7 October 2012 (UTC)

Since "fast" is based on perception, it doesn't necessarily correspond with speed. For example, when downloading software, if it asks some questions, does the entire download, then asks a few more questions at the end, allowing me to do other things in between, that would seem much faster, even if it took an hour, than some poorly written download that only takes half an hour, but takes complete control of my computer and asks a question every minute, so I have to stay there and stare at it the whole time. StuRat (talk) 22:13, 7 October 2012 (UTC)

Then you're no longer considering the speed of the computer as such. You're considering how long it might take, and how much of your undivided attention it might take, to complete the process of downloading software onto your computer. These are very different things. -- Jack of Oz ^[Talk] 00:05, 8 October 2012 (UTC)

If your PC is on 110 volts house current, try plugging it into a 220. That should make it go fast. ←Baseball Bugs ^{What's up, Doc?} carrots→ 11:57, 8 October 2012 (UTC)

If by "go" you mean "go up in flames", then yes. StuRat (talk) 22:02, 8 October 2012 (UTC)

Well, it depends on what kind of performance boost you're looking for. ←Baseball Bugs ^{What's up, Doc?} carrots→ 22:17, 8 October 2012 (UTC)

German pronunciation help

I have written a new article, Schwurhand, and need to add a pronunciation for the title. I don't do either IPA or German so can somebody help please? The article was the result of a question on the Humanities Desk; a German editor has called my effort "surprisingly original" which I hope is a compliment! Alansplodge (talk) 23:33, 6 October 2012 (UTC)

German pronunciation: [ˈʃvuːɐ̯hant]. Lesgles (talk) 01:57, 7 October 2012 (UTC)

Vielen Dank. Alansplodge (talk) 10:18, 7 October 2012 (UTC)

Resolved