Jump to content

Talk:Serbian language: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
Line 159: Line 159:


Given voices /ʒ ʃ tʃ dʒ/ don't exist in Serbian language and Serbian Ж, Ш, Ч, Џ (Ž, Š, Č, Dž) are /ʐ ʂ t͡ʂ d͡ʐ/. [[Special:Contributions/79.101.199.185|79.101.199.185]] ([[User talk:79.101.199.185|talk]]) 00:21, 17 December 2011 (UTC)
Given voices /ʒ ʃ tʃ dʒ/ don't exist in Serbian language and Serbian Ж, Ш, Ч, Џ (Ž, Š, Č, Dž) are /ʐ ʂ t͡ʂ d͡ʐ/. [[Special:Contributions/79.101.199.185|79.101.199.185]] ([[User talk:79.101.199.185|talk]]) 00:21, 17 December 2011 (UTC)

== Serbian: Official language in Croatia ==

It states in the head list of places that use serbian as an official language or that serbian is an official language in croatia, but on the map,croatia is light green, meaning that serbian is a recignized language.

Which one is right???

Revision as of 00:56, 21 December 2011


Kosovo and Metohija is a part of Serbia

Kosovo and Metohija is a part of Serbia and there's no need to mention that Serbian is official in Kosovo. You have mention it is official in Serbia. If you want to separate it, then mention Republika Srpska and other parts of other states!

That is a POV that will not be clear to all readers. Some will understand "Serbia" to include Kosovo, others will not. Regardless, Serbian is official in Kosovo, not (only) because Belgrade says it is, but because Pristina says it is. If we omit Kosovo, the reader will be left wondering if Serbian is recognized as official by Kosovo. Therefore it is clearer for us to explicitly list Kosovo. Perhaps we could word it "Serbia, including Kosovo" if you wish, though I'm sure that will spark an edit war as well. — kwami (talk) 07:45, 13 September 2010 (UTC)[reply]

Serbian language is older than Serbo-Croatian

As you know, Serbo-Croatian is communist construction. It is well known that Croats adopted Serbian as literal language in 1850. by Vienna's agreement, so you can not mention that Serbian is dialect of Serbo-Croatian. Shame!!! And I have gone to page Croatian language. Why then there you hadn't mentioned that Croatian is a dialect of Serbo-Croatian? What are you doing in wikipedia???

The only problem is, that the notion of Serbo-Croatian predates Communist Yugoslavia by a century or thereabouts. --Ivan Štambuk (talk) 15:11, 16 May 2010 (UTC)[reply]
Serbian and Croatian language appears centuries before that. Serbo-Croatian has emerged as a fusion of existing language. As someone already said, why then there you hadn't mentioned that Croatian is a dialect of Serbo-Croatian? --Aca Srbin (talk) 21:13, 25 September 2010 (UTC+1)
We do. — kwami (talk) 19:49, 25 September 2010 (UTC)[reply]
Serbian and Croatian languages were created long before the Serbo-Croatian and silly to say that Serbian is a standardized form of the Shtokavian dialect of Serbo-Croatian! In the articles dedicated to Serbo-Croatian language says that it came in the 19th century, the Serbian and Croatian are much earlier. Practically, it should say that the Serbo-Croatian has emerged as a fusion of these two languages, but no the opposite, as you consistently write. --Aca Srbin (talk) 00:41, 26 September 2010 (UTC+1)
None of them were created. The only things that were created were the standard languages, but this article is not just about the standard language.
There is a single language, or rather a series of dialects, which Serbs call "Serbian" and Croats call "Croatian". We call that language "Serbo-Croatian". If you think it's a fusion, please review your history. This has been covered ad nauseum at Talk:Croatian language, as it's generally Croats who object to an objective description of their language. — kwami (talk) 23:18, 25 September 2010 (UTC)[reply]

It's really not fair!!! I see no reason for discrimination against the Serbian language in Wikipedia. :S I think it's best to write to the Serbian language is one of the South Slavic languages, and then based on Stokavian dialect of Serbo-Croatian. So it is in Croatian language. I do not think that this is true, but obviously to emphasize it here. Practically, I have not deleted the fact that the modified version of Serbo-Croat, but I first say that it is the language, and then the details about it.

Indeed, in Bosnia and Herzegovina, "Croatian", "Bosnian", and "Serbian" are considered to be three names for the same official language. Also, I think this sentence is unnecessary, because the already mentioned a similar story with the Croatian and Bosnian earlier. Only a real crowd.--Aca Srbin (talk) 22:02, 28 September 2010 (UTC+1)

There's continual pressure from Croatian, and to a lesser extent Serbian, editors to pretend that Croatian and Serbian are independent SS langs, no closer to each other than they are to Slovene. That, of course, is false, and given all the attempts to convince people of that falsehood, a good article will make it very clear that they are one and the same language dialectologically, and only differ ethnically, politically, and (barely) in their standard forms. This is much like "Hindi" and Urdu: "Hindi" is just Urdu as spoken by Hindus or written in the devanagari script, or at the official level, a different standard form of the same language as Urdu. We shouldn't pretend that those are independent languages either. — kwami (talk) 23:48, 28 September 2010 (UTC)[reply]

Vocabulary section

Hi, I was just wondering if anyone could edit the section on the vocabulary section, in particular the reference to the word "avlija". This is NOT a turkish word, but a Greek one. It entered Turkish as "avlu" from Greek "AVLE" (the E being the long Heta) and at the time of transmission pronounced as "i" as the "upsilon" after the initial "a" was pronouced as a bilabial fricative "V" as per the great vowel and consonant changes in Greek (see the Hellenestic Greek sound changes). The etymology means "courtyard" and the online etymology dictionary reference is here: http://www.etymonline.com/index.php?term=curtain

or pasted as:

"1300, from O.Fr. cortine "curtain, tapestry, drape, blanket," from L.L. cortina "curtain," but in classical Latin "round vessel, cauldron," from L. cortem (older cohortem) "enclosure, courtyard" (see cohort). The confusion apparently begins in using cortina as a loan-translation for Gk. aulaia ("curtain") in the Vulgate (to render Heb. yeriah in Exodus xxvi:1, etc.) because the Greek word was connected to aule "court," perhaps because the "door" of a Greek house that led out to the courtyard was a hung cloth. The fig. sense in curtain call is from 1884. Curtains "the end" is 1912, originally from stage plays." NB: aule = courtyard. Garden would be Kepos.

The equivalent word in turkish for garden is "Bahçe". Whether or not the term entered into Turkish and transmitted by them or by the Greeks is a different story. What is important is that the word is etymologically Greek, just like many Persian words and Arabic words were transmitted to some European languages by the Turks who absorbed many different vocabularies for their daily use.

Thanks, Etymon. 128.250.254.122 (talk) 06:14, 2 September 2010 (UTC)[reply]

This was discussed at Talk:Serbian language/Archive 1#avlija. The compromise was to put a footnote -- which is still there, see note (15), about the ultimate Greek origin of the word. There are many other Arabic and Persian -- and, in this case, Greek -- loanwords in Turkish, but from Serbian perspective, it does not matter, as Turkish was the mediator and apparent origin. No such user (talk) 06:29, 2 September 2010 (UTC)[reply]


I dont' agree with that perspective at all. Etymologically it is not turkic. You yourself said that "from Serbian perspective, it does not matter, as Turkish was the mediator and apparent origin (sic)". Two things are wrong with this sentence, 1.) that it doesn't matter i.e. if it didn't you should then revert to the correct etymology and 2.) the apparent origin? Are you kidding??? The only thing original is ... well I'm struggling to see anything original that came from the turkic language regarding this word's etymology as the -ia (in Serbian's case -ija) ending is a PIE feminine marker! So what you're saying that the turks were the mediators... have you a source for that? I think Greek peasants were using Avli before the Turks arrived (and no it's not from middle Greek but has been in existence for far far longer even with the sound change!) and if you can prove this happened at the time of the ottomans then by all means say they mediated it but for goodness' sake don't say they were the originators of that word. A footnote will NOT suffice and as a linguist I'm deeply offended... I would edit but I'm relying on good faith of more experienced editors! —Preceding unsigned comment added by 121.214.116.14 (talk) 10:49, 2 September 2010 (UTC)[reply]

Well, I fail to see the big deal, but here you go. All that the text says about the issue is, quote, "some 30–50 years ago avlija (авлија < Turkish avlı[15]) was a common word for courtyard or backyard in Belgrade[...]". Here's a reference for the origin: [1]. So, BCS word is indeed a loan from Turkish word, and it is pretty much irrelevant (esp. for the considerations about Turkish loanwords) who Turks loaned it from. Despite that irrelevance, there's a footnote about the ultimate Greek origin. The suffix "-ija" is common BCS reflex for Turkish final "ı", compare e.g. čaršija < çarşı [2], which is of ultimately Persian origin, but I haven't seen many Iranians protest on the talk page. We also have a lot of Greek loans, especially in church and related terminology, but they were not the focus of the paragraph in question. So I don't see why would we enter into too much detail, except perhaps to select a different example to keep Greek editors happy. No such user (talk) 11:22, 2 September 2010 (UTC)[reply]

As per wikipedia (original research policy and sourcing non-primary sources or rather in this case nationalistic sources) - this pages you quoted is from Hrvstka i.e. a Croatian website. This does not qualify as sourceable material. If you can please find a better source of that transmission (not on a natinoalist website without accredited linguistic/sociologic or statistical data sets) then all I can say is you have to remove the entire word altogether. That way you're not offending linguists. Again, the transmission is not done by the turks given that they themselves utilise the word "Bahçe". You could say it was transmitted through Turkish speakers (this way they could have been Greek, Serbian-Slavic, Bulgarian, but I doubt very much so as Turkish).

Thanks again. —Preceding unsigned comment added by 144.137.0.57 (talk) 11:37, 2 September 2010 (UTC)[reply]

HJP is the only online scientific etymological dictionary of Serbo-Croatian (see translated credentials), so it is the most convenient online source to quote. As a linguist, you certainly know that Serbian and Croatian share a large point of vocabulary, and this word is no exception (though it's rather archaic by now, except partly in Bosnia). Most of the Turkish loanwords did not enter BCS through modern Turkish, but through Ottoman Turkish language, which is characterized (among other things) by a larger number of non-Turkic loans. Avlija is most probably one of those. I certainly expect some WP:AGF on your side: we did not borrow the word directly from Greeks, one indirect proof being that it is most frequently used among Muslim Bosniaks, who generally keep the largest share of Turkisms. No such user (talk) 11:53, 2 September 2010 (UTC)[reply]
See also [3] quote: ‡ Od: I. avlija ‘1. dvor 2. ograda, yid okolu ku}ata’ od tur. avlı, avlu ‘dvor’ < gr~. aìl™ ‘dvor’;] or new Serbian Претраживач страних речи и изразa [авлија, -е ж, ген. мн. авлија [тур. avlı од нгрч. avlī од грч. [4] (limited preview). No such user (talk) 12:05, 2 September 2010 (UTC)[reply]

Again not a valid source. Primary sources ONLY... this is conjecture and a Croatian site. The languages may be akin but Croatian and Serbian were different at that time (only under Tito did they merge to formulate one language). Different in the sense of who followed the Latinate or the Eastern/Greek rite. The fact that the Bosnians retained more foreign words lends credibility to the muli-ethnic nature of their words... Greek, Serbs, Croats and Turks all contributed. The point you are making is circular argumentation and not scholarly. I think you either state implicitly that it was inherited from Greek speech via the medium of Ottoman Turkish borderlessness thoughout the empire or you abrogate all ties to the etymology section. —Preceding unsigned comment added by 144.137.0.57 (talk) 12:18, 2 September 2010 (UTC)[reply]

Jesus Christ. I give up. Do you want me to perhaps perform a belly dance for you? No such user (talk) 13:53, 2 September 2010 (UTC)[reply]
It's borrowed from Ottoman Turkish per numerous etymological dictionaries of Serbo-Croatian (Skok, Škaljić, Gluhak..). Your "protest" is a result of nationalist frustration that there is a significant layer of Turkish borrowings in Serbo-Croatian (at least 8000 recorded lexical units) resulting from 5 centuries of political domination of Ottomans on the Balkans. Grow up. --Ivan Štambuk (talk) 16:54, 2 September 2010 (UTC)[reply]

Again this has nothing to do with nationalistic frustration other than the word's etymological rendering. Using circular argumentation from Serbo-Croatian (sic) sources of 30 years ago is not a credible source, given it's conjecture. Where is the reference that states implicitly the "turks" were the transmitors, and I don't mean a hypothetical rendering of the word "avlija" from "avlu" which is from the Greek speech in the region ("avli"). From a pure linguistic perspective, it's impossible for the term to have been inherited from Turkish given it's pronunciation was "avlu" and if so, it would have been rendered as "avluja" not "avlija". If it is "avli" the transmitors would have been Greek speaking. Non sequitur. You're also forgetting that the trashumant population of the Vlachs (heavily influenced by Greek speech) were prob also speaking Turkish, Greek and Aromanian Vlach and were prob more responsible in the word's transmission. Political dominance of the Ottomans is a separate issue and has nothing to do with the word itself, unless you can prove beyound any reasonable doubt the Ottomans used "avlu" in the Middle Eastern provinces of Syria/Damascus and the rest of the Levant.

Suggested reading: User:Jnc/Astronomer vs Amateur. You were given 3 reliable sources, got informed opinion of 2 educated native speakers, but you chose to believe your own fictive etymology. I doubt that you're an educated linguist at all, judging on your attitude and statements. Good bye. No such user (talk) 06:44, 3 September 2010 (UTC)[reply]

I can assure you that the etymology is anything but "fictive" (do you mean fictitious?). The -ija suffix is not productive in Turkish with words ending in -u! Two native speakers who are no doubt anything but linguists themselves. Serbian and slavic languages aside your "nativeness" does not promulgate your own beliefs of the origins of the word avlija and more importantly on who transmitted it. My point is thus: had the word been transmitted by the Turks the word would have been "avlu" not "avlija" (the Greek form Avli with the -ija suffix so prominent in many Serbo-Croatian words). The term would have also been promoted in the Levant had it been the Turks who transmitted it and yet it seems to be suspiciously retained in the Balkans around the Jirecek line of influence and hence the Greek speaking/influenced parts. My point is your sources suspect it was transmitted and there is no evidence of it! —Preceding unsigned comment added by 121.219.244.84 (talk) 10:06, 3 September 2010 (UTC)[reply]

Then, please do attempt to read properly the word that has been repeated at least 10 times in this discussion. The Turkish word ends in -ı, avlı. The BCS reflex of -ı is systematically -ija. I do not care which form was used in Levant, because it could not possibly influence BCS. I doubt that Ottoman Turkish was a monolythical language without significant regional variation. Now, if you have any reference that it was direct borrowing from Greek to BCS, please put it forward. I've had enough of your speculations. No such user (talk) 10:52, 3 September 2010 (UTC)[reply]
Who are you dude? Some IP frustrated that Turkish language has contributed thousands upon thousands of words to their mother language. This is your reasoning: "From a pure linguistic perspective, it's impossible for the term to have been inherited from Turkish given it's pronunciation was "avlu" and if so, it would have been rendered as "avluja" not "avlija"." - Except that Turkish word was not avlu but avlı as No Such User has stated. Epic fail dude. You were given 3 perfectly credible etymological dictionaries as sources for Turkish etymon, and the only evidence that you gave in response was your personal opinion based on fallacious logic. Dude, read this analysis of Rumelian Turkish dialects spoken at the Balkans and pay special attention to point 1.41. --Ivan Štambuk (talk) 11:39, 3 September 2010 (UTC)[reply]

And yet I am yet to find a record pointing out "avli" as turkish. Here http://books.google.com/books?id=QiGy8n8dKlUC&pg=PA27&dq=the+turkish+word+avli&hl=en&ei=_P6ATP3nDYamvgPYt4CGBA&sa=X&oi=book_result&ct=result&resnum=5&ved=0CD4Q6AEwBA#v=onepage&q=the%20turkish%20word%20avli&f=false

and unless you meant "avli = stocked with game" and not "avlu = courtyard" then you're the one with an EPIC FAIL. —Preceding unsigned comment added by 121.219.244.84 (talk) 14:04, 3 September 2010 (UTC)[reply]

No avlı is (was) an alternative form of the "courtyard" word, attested at least since the 16th century in that spelling. But regardless, either avlu or avlı in Western Balkans Rumelian Ottoman Turkish was rendered as [avli]. Read the link I gave above, page 3 point 1.41. --Ivan Štambuk (talk) 14:22, 3 September 2010 (UTC)[reply]

Number of speakers

I really hate the trend of population inflation per one's favorite ethnic group/city/country, but the Ethnologue data about Serbian are bullshit, pardon my French. (Actually, I'd say that most of Ethnologue data are bullshit, but that's another issue). According to the Serbian census 2002 [5], there are 6,212,838 ethnic Serbs in Serbia without Kosovo, which surpasses Ethnologue data for 1.7 million (and please don't tell me that not all of them declare that their mother tongue is Serbian). Add to that at least 1.5 million of Bosnian Serbs, some 500,000 of Croatian Serbs (200,000 still there+numerous refugees), and quite a few in USA, Germany, Austria and Australia, and you will easily get near the 10,000,000 mark. See [6]. Ethnologue figure simply cannot be true, no matter how one analyses it. No such user (talk) 09:37, 28 September 2010 (UTC)[reply]

You mean 4.5M, not 1.7M.
Yeah, that figure doesn't seem credible. If you subtract all the minority langs E. lists for Serbia, you're left with 6.5M, incl. Kosovo, for 8-9M total.
Another oddity is Montenegro, which has 0.2M Serb out of a pop. of 0.6M, and they don't count Montenegrin as a language.
I have no prob using other sources. Ethnologue is our default because it's impossible to find data on a lot of obscure langs otherwise. Do you have s.t. that gives a citable total? — kwami (talk) 10:34, 28 September 2010 (UTC)[reply]
I found one, although I'm not so sure about reliability (it's UCLA, but not the peet-reviewed stuff). Lots of 11-12 millions figures on the web, but no WP:RS as far as I can tell.
Here's an interesting stuff [7] (scroll to the bottom). Not a WP:RS, but it says: "Ethnologue estimates that there are over 11 million speakers of Serbian worldwide.". Apparently, Ethnologue "fixed" their entry at one point. I'm not sure if we can check what did it say in 2006, using Wayback machine or something? No such user (talk) 11:27, 28 September 2010 (UTC)[reply]
They don't update individual entries; the web version is a mirror of their print version. Their link to the 15th edition is here.[8] Note the 11M figure is from 1981, though. They also count 4M for Bosnian, which is the entire population. I don't think that ref is much good either.
Ah, the UCLA fig is just taken from Ethnologue 15. We should be able to do better than that.
The ELL doesn't give pop figs. Do you have census data from Bosnia? We should be able to cobble s.t. together in the absence of a RS. — kwami (talk) 11:50, 28 September 2010 (UTC)[reply]
Census in Bosnia hasn't been held since 1981 (or maybe 1991); it's to be held in 2011. According to the CIA factbook, Serbs consist 37% of country's 4.6 milion ≅ 1.7 mil, which is probably overestimate; some 1.5 mil. would be more realistic.
There's a Google book search for "Serbian "12 million", of which the Lexicographica seems relevant; only snippet view though. No such user (talk) 12:13, 28 September 2010 (UTC)[reply]
ELL: 6.62M in Serbia sans Kosovo (88%). Bosnia: 37.1% of 4.01M estimate for 2004 = 1.49M (just what you thought). Croatia: 45k, not counting several 100k who've fled. Kosovo: 133k, not counting 300k "mostly" Serbs who've fled. (Are these being counted in the Serbian figure?) Montenegro: 400k (60%). Albania: few (< than other langs @ 50k). Macedonia: ? (E. says 33k). Romania: ? (E. says 27k.) So that would be what, about 9 million? (8.75M if all refugees were counted, maybe 9.25M if not.) — kwami (talk) 11:50, 28 September 2010 (UTC)[reply]
Plus the diaspora in the Western World, but it's difficult to count. See references at Serbs. The catch is, there's quite large diaspora in Western Europe, esp. Germany, Austria and France, though many of those are "guest workers", which occasionally visit, and/or plan to return to Serbia (eventually), and might have one or both citizenships (thus, might or might not be counted twice at Serbian and foreign censuses). For those in USA and Australia, it's fairly safe to assume that they're not counted in Serbian censuses. In total, all of that should be some 0.8-1.5 million, probably closer to the lower figure. No such user (talk) 12:55, 28 September 2010 (UTC)[reply]
Per Eth., 45k in the US, 50k in Canada, 39k Australia, not a huge difference. If Croatian and Kosovan refugees were counted in Serbia, and Euro guest workers as well, then we'd still be at 9M. For Euro, Eth. has 120k in Sweden, 142k Switzerland, but dn mention any in Germany, Austria, or France. Other countries @ < 10k. So maybe a max of 10M, if they weren't counted? Note the 21k Serbian speakers in Turkey (1980) are also called "Bosnians". — kwami (talk) 13:17, 28 September 2010 (UTC)[reply]
Yep, I think 10 M is a realistic estimation. I've just removed Turkey from Serbs, because it's everyone's pet "nation inflation tool". Well, many modern Turks have an origin from the Balkans, ergo they still can be safely counted as "Bosniaks" or "Serbs" or "Albanians" and whatnot. All based on references like this. No such user (talk) 14:51, 28 September 2010 (UTC)[reply]

From the Ethnologue Editor

FYI,

Thank you for contacting the Ethnologue with your comments on Serbian [srp] in Serbia.
 
Based on the census statistics for 2002, we will change the population number for speakers of Serbian. 
Please note that any changes will not be made on the Ethnologue website until the 17th edition is published.

Sincerely,

<name withheld>
Managing Editor
www.ethnologue.com

No such user (talk) 06:45, 29 September 2010 (UTC)[reply]

1RR under WP:ARBMAC?

Should this article be placed under the 1RR restrictions of WP:ARBMAC? There's been a Croatian POV pusher changing "Serbo-Croatian" to "Croatian" the last 24 hours. --Taivo (talk) 17:36, 16 October 2010 (UTC)[reply]

IMO that's not necessary. We can block that editor if need be, but there hasn't been much of a problem with this article. — kwami (talk) 22:51, 16 October 2010 (UTC)[reply]

This article has been destroyed. What is happening ?--KudySk (talk) 14:28, 19 December 2010 (UTC)[reply]

Ambiguous punctuation

The article states:

"Although Serbian language authorities recognize the official status for both scripts in contemporary standard Serbian language for more than half of a century now, due to historical reasons, Cyrillic was made the official script of Serbia's administration by the 2006 Constitution."

Which part of the sentence "due to historical reasons" refer to? Do both scripts have an official status for historical reasons, or was Cyrillic made the official script for historical reasons? (And what kind of historical reasons are these? "Historical reasons" sound very broad to me...)

--Image of me (talk) 06:44, 22 April 2011 (UTC)[reply]

The due to historical reasons is linked with the rest of the sentence, meaning that the authorities recognise both scrypts for more than half century now, but, Cyrillic was made the official script of Serbia's administration by the 2006 Constitution, trying to reinforce it as the traditional scrypt because historically Serbs used Cyrillic, but just recently, since 1945, have been using both. The sentence could possibly be expanded. FkpCascais (talk) 06:58, 22 April 2011 (UTC)[reply]

Kosovo usage in Wikipedia

Throughout Wikipedia (see Albanian language, for example), we list Kosovo as a separate state although in italics with a note as to its disputed status. This is the NPOV way to indicate it so that those readers who expect it to be separate see it separately and those readers who expect it to be not separate see it in italics. This has been discussed and agreed to in multiple places in Wikipedia. --Taivo (talk) 10:23, 4 May 2011 (UTC)[reply]

Serbo-Croatian

This whole issue of Serbian as part of Serbo-Croatian has been hashed out with verifiable, reliable sources over and over again here, at Serbo-Croatian language, at Croatian language, etc. --Taivo (talk) 19:46, 13 May 2011 (UTC)[reply]

It's really only one-Croatian language. No such as "serbian".

Look at the facts on wikipedia about Croatian written documents centuries before so called "serbian". — Preceding unsigned comment added by 94.253.200.160 (talk) 23:02, 22 October 2011 (UTC)[reply]

Mistakes in Serbian IPA

Given voices /ʒ ʃ tʃ dʒ/ don't exist in Serbian language and Serbian Ж, Ш, Ч, Џ (Ž, Š, Č, Dž) are /ʐ ʂ t͡ʂ d͡ʐ/. 79.101.199.185 (talk) 00:21, 17 December 2011 (UTC)[reply]

Serbian: Official language in Croatia

It states in the head list of places that use serbian as an official language or that serbian is an official language in croatia, but on the map,croatia is light green, meaning that serbian is a recignized language.

Which one is right???