Talk:Turkish language/Archive 2

Archive 1

Archive 2

Archive 3

Archive 4

Longest word

"Çekoslavakyalılaştıramadıklarımızdan" is NOT the "longest" word in Turkish. It is "Mükemmelliyetçileştiricileştiriveremeyeceklerimizdenmişçesine". Is it possible to translate that into English? —The preceding unsigned comment was added by 85.97.68.59 (talk • contribs) 12:26, 24 October 2006.

Posible. I personaly wouldn't dare attempt though. Also that word doesn't seem to make a whole lot of sense to me... Then again I am only a puny tr-3. --Cat out 04:04, 25 October 2006 (UTC)

I don't know is it possible to translate that or not but ı know it makes sense in Turkish. Mükemmel Mükemmelliyet-çi-leş-tir-ici-leş-tiriveremeye (like "yapıvermek")-cek-ler-imiz-den-miş-çesine (like "yapamadıklarımızdanmışçasına")... —The preceding unsigned comment was added by (?)

It translates to: "In the way as if he/she is not one of those whom we couldn't turn into a perfectionistizer easily". "-leştirici" (which would correspond to -izer in English) is used twice to create the meaning "perfectionistizerizer" (somebody who turns a person into "a person who turns people into perfectionists") But I don't think it practically makes sense. By this rationale, the longest word in Turkish would go to infinity since you could keep combining the word with the same suffix 789 times. --(Tylose 19:12, 16 December 2006 (UTC))

language reform turkish under persian and arabic influence

i changed the part saying that turkish lang. went under the influence of arabic and persian as ottomans adopted islam, which is not true. turks converted to islam before ottoman state or empire established. one of the reasons of the influence of persian and arabic over turkish is of course adoption of islam, but it was not the ottomans who did it. Influence of persian and arabic had begun more than 300 years before 1299 (fictional date of starting of ottoman state in western anatolia) around north and east persia. the second point is that it is not the religion itself as the main reason for this language change, turks had been living close to persians for centuries and were migraitng from central asia to persia continously, so even turkish clans, sultans or states did not choosed islam (or not forced to do so) there will be an influence of persian and arabic as the languages of two rich and powerful culture.Ulubay 00:18, 18 December 2006 (UTC)

Actually since Seljuks, Ottomans were under the influence of literary part of Persian lang., however Ottoman were under the influence of the scientific and the tecnical ability of Arabic which was the official language in formal corresponding in Ottoman E. Still there is much more Persian words than Arabic in the daily life, however in formal subjects like law you only hear Arabic words :) It's true that in daily language Arabic words had influences through islam, but not as much as the inf. of Persian had. —The preceding unsigned comment was added by OktayD (talk • contribs) 01:49, 23 January 2007 (UTC).

To make it clear; I have to say that Persian or Arabic had never been the official language for state or beaurocratic communication in Ottomans, it was Turkish even it was heavily loaded with foreign vocabulary. . Ulubay 15:20, 18 February 2007 (UTC)

Turkish not Asio-European????

I find it baffling that Turkish is not Asio-European while other Asio-European languages such as Farsi(Asio-European)sounds much less European than Turkish/Turkche. I can tell by the words(I speak Urdu)that it's not European,but the pronounciation of words like "affiyet olsun"(good appitite) definately sound European regardless of the meaning or vocab. Is there any connection at all between Turkish/Turkche' and the languages of Europe? I know about the Finnish and Hungarian connection,but shouldn't the article highlight that it sounds European when spoken since many people find it sounds European? And what do linguists think of it's European sounding vowals?Does anyone know? 74.98.241.189 21:01, 21 December 2006 (UTC)Nadirali

Language families like Indo-European are defined by the history of the language, not what it sounds like, its similarities to its geographical neighbors (compare areal linguistics), or how its grammar works (compare language typology). Anyway, I am not sure what "European-sounding vowels" are; the vowels of English, Spanish, French, and Norwegian are all very different.... --Macrakis 17:32, 9 January 2007 (UTC)

Turkish when spoken sounds very similar to French. On other occasions, I've heard people speaking Turkish and could have sworn it was Russian or some slavic tongue. Written Turkish does not look french or slavic at all, but spoken Turkish does. Probably because of the abundance of silent letters.

I would strongly dispute that "spoken Turkish sounds like French". (The bit about "silent letters" is irrelevant; there are "silent letters" in many languages, but this doesn't make them sound "very similar" or at all.) The French language's accentuation alone on the last syllable makes for a very different musicality. The two languages have a distinct sameness in their abundant use of "u" (in Turkish, "ü"), but the overall sound of spoken Turkish could not be mistaken for French, the way, say, one could get confused (from a distance or not hearing exactly what is being said) between German, Swiss and Austrian. The Gnome 10:00, 23 June 2007 (UTC)

Diphthongs

The article currently says "There are no diphthongs in Turkish". Is this correct, or simply a confusion between diphthong and digraph or diphthongized vowels (e.g. English bone [bown]? How about ayran, kıyma, öyle, mevt, favl? I thought that, at least in some versions of spoken Turkish, these are pronounced with glides [ajran], etc. Could someone who knows Turkish phonology better than me comment? --Macrakis 22:06, 7 January 2007 (UTC)

Precisely because these are glides, and not vowels, such examples cannot be considered diphthongs, as a diphthong implies two vowels, whereas glides are technically non-vocalic.201.37.71.146 21:25, 10 January 2007 (UTC)

Is there a clear distinction between the two? How is the 'ay' of Turkish 'ayran' different from the 'i' of English 'kite'? Or is the latter not considered a diphthong either? Perhaps you could take a look at the WP articles diphthong and semivowel (which currently reads: "Semivowels ... are non-syllabic vowels that form diphthongs with syllabic vowels") and clarify (with literature references)? --Macrakis 22:42, 10 January 2007 (UTC)

Phonetically, they are certainly normal diphthongs, but phonologically, one thing that shows that the offglide have the phonological status of consonants is their behaviour under morphophonological rules. Like, when a word ends in -"ay", any suffixes you add will take the form appropriate for consonantal words, without an additional bridging consonant (for instance possessive "ay-ı", not "ay-sı"). Fut.Perf. ☼ 22:43, 11 January 2007 (UTC)

number of speakers

Ethnologue has 50 million. The 65 (or 75) figures need attribution. Also, it is questionable to include Turkmen and Azeri as "Turkish". We have the Oghuz languages article to discuss that larger group. The "Turkish" subgroup of Oghuz includes Gaugaz and Meskhetian, but not Azeri or Turkmen. dab (𒁳) 16:12, 10 January 2007 (UTC)

Hello, sorry for my bad English but the source which you showed is 10 years old at least! The figure was discussed here many time! Ethnic and native speakers aren't the same things! For example, an Azerbaijani mother speaks Turkish because she lives for a long time in Turkey. So she accepts the Turkish language and teaches her children also Turkish, not Azeri language. So the childrens native language is Turkish not Azeri language. What do I want to say in addition to this? Ethnic and native language is different and your source is out of date and distrustful. If you think logically, you can find out the real figure of native speakers yourself. 172.178.20.248 16:32, 15 January 2007 (UTC)

Even excluding Turkic languages, It must be 60 million because Republic of Turkey's popp. is about 60~63 million end it's the only ofical language. --Mko 22:41, 22 January 2007 (UTC)

Total speakers 50 million? This is totally unacceptable, where population of Turkey is 72 million (almost all of 72 million speak Turkish, as a native or second language), and 2,5% of Germany's population speak Turkish. This should be immediately changed. Where Ethnologue says so or not. Kaygtr 16:26, 4 February 2007 (UTC)

Ethnologue in 2000: "46,278,000 in Turkey, 90% of the population (1987)."
Ethnologue in 2005: "46,278,000 in Turkey (1987)."
Clearly, the number hasn't been updated at all. If 46,278,000 were truly 90% of the population today, the whole population would equal to 51,420,000 whereas 2005 estimate is 72,600,000. Assuming that there is no significant change in the "%90" (which is very likely), the latest number of speakers can be calculated as 65,340,000. Unless the new ratio was set in a such way so that number of speakers remains the same, the information provided by Ethnologue is senseless. Without counting the speakers outside of Turkey, Turkish has 65 million speakers. Until someone calculates the total number of outside speakers with citing sources, I am changing the number of speakers to "ca. 65 million". (Tylose 18:00, 20 February 2007 (UTC)).

The 65 million figure is unsourced, while the 50 million figure still is. According WP:V, "The threshold for inclusion in Wikipedia is verifiability, not truth." Also, doing your own math is considered original research. Khoikhoi 04:08, 21 February 2007 (UTC)

65 million figure is sourced, only in ratio, not as a number. I'm not sure if I'm doing original research. The information in Ethnologue is also obtained by math and by using the data belonging to 1987. Since the population of Turkey has changed after all those years, I think it's probably time to do the same calculation with new population rather waiting for Ethnologue to do it. I didn't actually understand why you consider Ethnologue's math, which uses data from 1987 as verifiable. (Tylose 23:57, 21 February 2007 (UTC))

The Republic of Turkey Ministry of Culture and Tourism doesn't seem to be the most neutral of sources, plus one must wonder if they are an authority on the numbers of speakers of Turkish. After all, how are they able to determine the number of Turkish speakers worldwide? Ethnologue seems to be more reliable to me. Khoikhoi 04:22, 22 February 2007 (UTC)

Verifying by the reference to Ethnologue source is highly questionable, as any unbiased look at the WP site for that organisation clearly shows. These are just quotes from that site: "Christian linguistic service organisation", "information regarding more esoteric languages is quite dated", "neutrality ... as a scientific institution is sometimes disputed",

"contains its fair share of errors", "classifications do not meet ... own professed criteria for classification", "much of the information is old". User:noyder 11:41 29 February 2007)

The language in daily life

This section has a certain charm, & includes some useful material, but is it appropriate in this article? It really belongs more in a textbook or phrase book in my opinion. NigelG (or Ndsg) | Talk 21:28, 21 January 2007 (UTC)

== Total speakersapprox. 50 million (unsigned comment by someone)

Any comments on my remarks on "The language in daily life"? It doesn't belong in this article unless specific phrases are tied in to specific grammatical features of the language (eg -sin, 3rd person optative). Otherwise it's just a random list of phrases.

The intro to this section states that:

"Several of them feature Arabic verbal nouns together with the Turkish verb et- ("make, do")."

In fact there are only two such examples:

• Affedersiniz
• Teşekkür ederim

Finally, it's hard to see what insight is given by including the word Alo. NigelG (or Ndsg) | Talk 15:50, 23 January 2007 (UTC)

Proposal: List the article as "Good article" candidate

I believe this article can be nominated as a Wikipedia:Good article and listed at Wikipedia:Good article candidates. In my opinion most of the criteria of GA are met:

“ [GAs] should be well written, stable, accurate, referenced, use a neutral point of view, and wherever possible, be illustrated by appropriately tagged images. Good articles may not be as thorough and detailed as our featured articles, but should not omit any major facets of the topic. ”

For more detailed criteria check WP:WIAGA, WP:STYLE and WikiProject Languages/Template. This process is quite useful: (a) the work done so far can be recognised by the wider Wikipedia community, which hasn't necessarily noticed the progress made, (b) in the process of preparing the article for nomination and through the remarks of the reviewers weaknesses of the article can be noticed and the article can be significantly improved, (c) it is also a big step to raise the article to Wikipedia:Featured article. Checking upon the criteria mentioned, I would like to mention some possible additions to the article:

• There is no photo. The famous Image:Ataturk teaching the Latin alphabet to the people of Sivas is a good addition to the Language reform section. Also a good idea is photo with something written in Turkish (a photo with your own camera is just fine).
• The lead is too short for a lenghty article as this one. See WP:LEAD.
• Per WikiProject Languages/Template, a more general section on the history of Turkish can be added with Language reform as a subsection.
• Though some of this aspects are covered in other articles linked to this one, the above information would not be redundant in this article as A perfect Wikipedia article "[...] is nearly self-contained; it includes essential information and terminology, and is comprehensible by itself, without requiring significant reading of other articles."

I hope that these remarks are helpful. --Michkalas 17:39, 29 January 2007 (UTC)

Even though I share your optimism about the article, I don't think it's anywhere close to be nominated for GA status yet. The most disturbing deficiency, for me, is the lack of a history section. Another serious problem is that there are almost no inline citations. I also do not feel like the article is complete in terms of linguistics (I know this because even most fundamental informations like the Turkish noun declension system or the vowel harmony were missing until these were recently added by me, and I'm just an amateur in this field). If I'm not wrong, there is not even a single sentence in the article about conjugation, verb tenses, moods etc. For many similar things, the article is in serious need of a professional linguist's attention. I will try to do my best to implement your suggestions. Thanks for caring about this article. Atilim Gunes Baydin 18:23, 18 February 2007 (UTC)

Number is low

Turkey=57 million native speaker balkans=1 million middle east and cyprus:1 million diaspora:4 million total:63 million native speaker. these are all ethnic turks there is also 1 million caucasian 1 million arap pomaks lazs georgins 2-3 million kurd and zaza who speak Turkish as mother tongue so 4-5 million total:67-68 million native speakers azeri and turkmen can be thouhgt as Turkish so iran azeri:20 million azerbaijan azeri:8 million azeri in Russia and georgia:1 million azeri in Iraq:1.5 million Turkmen iran:2 million Turkmenistan:4.5 million Turkmen central asia:1 million ttl:38 million ethnic turkish turkey:57 million ethnic turkish plus 4 million other ethnic groups: 99 million turkish spekaers. —The preceding unsigned comment was added by 193.140.194.101 (talk) 10:26, 17 March 2007 (UTC).

"Spoken in" list

Hi, could Kaygtr please explain his reasoning behind changing

"Turkey, Bulgaria, Cyprus, Greece, Republic of Macedonia, Kosovo, Romania, and by immigrant communities in Germany, France, The Netherlands, Austria, United States, Belgium, Switzerland, and other countries of the Turkish diaspora"

into

"Turkey, Cyprus, Azerbaijan, Bulgaria, Germany, Greece, Kosovo, Macedonia, Romania, Serbia, Austria, Uzbekistan
and countries of the Turkish diaspora"?

The list on the top gives the traditional Turkish speaking communities (which were part of the Ottoman Empire) in the order of decreasing population (please see the very well-referenced numbers on Turkish diaspora article), and the countries with minorities established by recent immigration (otside the Ottoman Empire) again with decreasing population of speakers. I do not understand how the countries are ordered in the second list. It's also very bad that it does not include, say The Netherlands and France, when it lists Austria (each have a larger Turkish speaking population than Austria). It lists Azerbaijan (18,000 spekears) where it doesn't include France (370,000), The Netherlands (270,000), Belgium (110,000), United States (117,000). Why are you reverting into this very artibrary and nonsense list without giving a reason? Atilim Gunes Baydin 16:36, 18 March 2007 (UTC)

The purpose is clear, and you have answered it. Why are you reverting into this very artibrary and nonsense list without giving a reason?

There was no reason in your edit.

Also, you removed Northern Cyprus from official language list without explaining any reason.

p.s. The first one does not add Uzbekistan to the list. However, in the article of Turkish Diaspora, number of ethnic Turks in Uzbekistan, Tajikistan, Kazakhstan and Kyrgyzstan is given as 197,000. Can you explain? And as Azerbaijan, Nakhichevan should be noted. Kaygtr 18:31, 18 March 2007 (UTC)

I'm still not able to understand why you reverted the whole thing instead of adding any missing entries to the list. You could simply add Northern Cyprus and Azerbaijan. Please also note that it is me who did all of the referencing work in the Turkish diaspora article and it took many days, so I'm very much aware of the numbers there. At least after noticing the numbers for France, The Netherlands, and Belgium, I hope you can agree with merging these two versions together. I really do not get why you've done a complete revert instead of an addition. Regards, Atilim Gunes Baydin 20:07, 18 March 2007 (UTC)

Nominated as good article

Hi, I just listed the article as a good article candidate, after implementing the changes sugested by User:Michkalas as much as I can. Let's see how the review will proceed. I thank Michkalas for making very constructive suggestions recently and in the past, and also causing me to feel an urge for improving this article. Regards, Atilim Gunes Baydin 04:56, 19 March 2007 (UTC)

some thoughts

Atılım, I think there should be a section titled agglunitation, in my opinion it should be section 4.1, all the current contents of vocabulary should be moved there. in vocabulary section we should have information about say having this many percent the same vocabulary with this other Turkic language X, and the examples are: ... we should also tell how many loaned words from each languages there are (not just in the history section). We can talk about current Turkish's similarity with say Yunus Emre's Turkish especially compared to Ottoman Turkish. We should also mention that many words were created by Turk Dil Kurumu to replace those loan words, like yargıç.

deniz 07:52, 19 March 2007 (UTC)

Please see vocabulary sections of these featured articles:

Bengali_language#Vocabulary, Russian_language#Vocabulary, Swedish_language#Vocabulary, Taiwanese_(linguistics)#Vocabulary, Tamil_language#Vocabulary

The other pages don't have vocabulary sections

good articles:

English_language#Vocabulary, Esperanto#Vocabulary, Nahuatl_language#Vocabulary, Scanian_(linguistics)#Vocabulary

Here is the list of good and featured articles

Should I go ahead and do these changes here and on the vocabulary page? deniz 23:06, 20 March 2007 (UTC)

Hi Deniz, I think we should keep the two word derivation examples in the vocabulary section, because it is without doubt the defining characteristic of the Turkish lexis. If you move these to a new section, nothing much remains for the vocabulary (Do you propose to delete the section altogether? If so, I don't agree with that either, see Wikipedia:WikiProject Languages/Template). Almost all the tables in the grammar section pertains to agglutination due to the nature of Turkish. I don't think it would be good to collect these all into a new section dedicated to agglutination, because such a section would be just a collection of examples and the existing sections will be stripped of content. You could think of the current situation as describing agglutination as it occurs in different parts of speech (current subsections of nouns, verbs, etc.). Having a look to what's been done on other good and featured articles is a very good idea, I try to do the same thing to get ideas. I completely agree that there is much room for improvement in the grammar section in general, and my latest contributions and cleanup almost excluded that part. If you mean to mention numbers or percentages, these would be good additions (I think I've seen some information on the number of loandwords from each language in modern Turkish somewhere, I'll try to find that reference). And I think there is already enough (possibly, more than enough) mention of Türk Dil Kurumu and the new words issue. Regards, Atilim Gunes Baydin 00:48, 21 March 2007 (UTC)

Atilim, do we have "sessiz unsuz benzesmesi" on the article? Also, this might be what we are looking for: [1]. Arabic is most common(6463 words), then comes French (4974 words), Persian (1374 words) then Italian (632) ,... Here is the first 12 words from French: abajur abaküs abandone aberasyon abis abiye

ablatif abone abonman Aborjin abrakadabra absent. Likewise we can get others as well.I checked the Turkish page of this tr:Türkiye Türkçesi, they are listed there as well, and according to that site, ~14% of the words in Turkish are loanwords. deniz^T_C 13:02, 21 March 2007 (UTC)

Currently the only information we have in the article about "ünsüz benzeşmesi" is the following in the nouns and adjectives section:

"The initial consonant of the suffixes for the ablative and locative cases can also vary depending on the last consonant of the noun being voiced or unvoiced, such as having hava ("air") + -da (locative suffix) = havada ("in the air"), but, ağaç ("tree") + -da (locative suffix) = ağaçta ("on the tree"), instead of ağaçda."

The source you provided for the count of current loanwords is wonderful. I think we can incorporate that information into the article, with a nice pie graph. I don't trust much any information on the Turkish Vikipedi, because the contributors there don't seem yet to care about referencing or reliable sources, unfortunately. Atilim Gunes Baydin 17:04, 21 March 2007 (UTC)

I've just made the addition. Atilim Gunes Baydin 18:56, 21 March 2007 (UTC)

On diphthongs

Atılım, you have worked a lot on this article and I hope it becomes GA. Reading again the article I noticed one inconsistency. "There are no diphthongs in Turkish and when two vowels come together, which occurs rarely and only with loanwords, each vowel retains its individual sound." (section: Vowels) vs. "[Turkish] features eight fundamental vowel sounds and a host of diphthongs based thereupon." I do not know which of the two is true, but they can't be both true at the same time. There are is also another comment on diphthongs above in this talk page. --Michkalas 10:08, 21 March 2007 (UTC)

Thanks for pointing out, I've deleted the second sentence for the time being for consistency within the article. As I understand, the people thinking that there are diphthongs in Turkish are talking about situations where Turkish letter y (/j/) follows a vowel. I will try to find a good reference for this diphthongs issue and update the information in the article accordingly. Atilim Gunes Baydin 15:54, 21 March 2007 (UTC)

Congratulations

Thanks a lot Atilim for your hard work deniz^T_C

Thanks. Honestly, I was expecting at least a short review and a list of suggestions by the GA reviewer as usual. Cheers, Atilim Gunes Baydin 23:31, 21 March 2007 (UTC)

Congratulations!!! Baristarim 19:23, 22 March 2007 (UTC)

You know what the next step is. ;-) Khoikhoi 01:57, 23 March 2007 (UTC)

Why don't we have any A class articles? deniz^T_C 02:05, 23 March 2007 (UTC)

Advancing this article to FA

This is possible. Most (if not all) the information are here. Maybe two important things for FA is a considerably expanded lead and more references, especially to English language bibliography. Good and -more or less- updated sources in English and widely available in an academic library near you are: The chapter on Turkish language in the 1998 or 2006 edition of Eva Csato and Lars Johanson's "The Turkic Languages" (good for the grammatical description). Also Jacklin Kornfilt's Turkish is considerable more updated than Lewis. These should be available in any university linguistics library in Istanbul, Izmir or Ankara (and outside Turkey, of course). But, for a FA, compatibility with Wikipedia criteria on how an article should look like (structure, style etc) is crucial. Many useful comments for this purpose can be obtained through Wikipedia:Peer review. We can also leave a note at Wikipedia talk:WikiProject Languages to get more and more specialised comments. In any case, criteria for FA are becoming more and more stricter, so be patient as it will take time. You can check Bengali language, probably the most recently promoted FA article with a modern language as its subject. Compare it with this article and also chech the comments when it was a FA candidate. Anyway, in my opinion, a peer review should be the next step forward.--Michkalas 14:27, 23 March 2007 (UTC)

Thank you very much for the professional advice! I have access to all Nordic libraries through my university and I'll be looking for these sources. Most of them seem to be available already in my city. We currently list Johanson's Discoveries on the Turkic linguistic map (also with a pdf link), a very nice short report on the state of Turkic linguistics, for those who are interested. The book should be even better. I agree with the other points you've made. Eυχαριστώ, Atilim Gunes Baydin 17:27, 23 March 2007 (UTC)

True, with a little work it can easily become FA, also because it is not a controversial or political article or anything.. Baristarim 23:00, 23 March 2007 (UTC)

One suggestion, I think History and Writing System should be placed immeditately after Classification. The Bengali language article would seem to confirm this as a better stucture. --A.Garnet 23:03, 23 March 2007 (UTC)

There's a set layout at Wikipedia:WikiProject Languages/Template.--Domitius 23:06, 23 March 2007 (UTC)

Stress

Article currently says:

Stress is usually on the last syllable, with the exception of some suffix combinations, and words like masa ('masa). Also, in the use of proper names, the stress is transferred to the syllable before the last (e.g. İstánbul), although there are exceptions to this (e.g. Ánkara).

I'm far from an expert, and I haven't been in Turkey for some years now, but I think stress and vowel length need to be discussed more fully. First of all, what does "words like masa" mean? "Like" masa in what way? From the synchronic Turkish point of view (forgetting about etymology), stress is unpredictable (though it is mostly commonly on the last syllable). This includes many common words: 'radyo, e'fendim, is'kele, 'kolkola, lo'kanta, even 'vişne, the example of vowel harmony in the previous sentence! I am not sure what is meant by "some suffix combinations". Some suffixes have stress patterns attached to them, e.g. present tense -'*yor, negative '*-me.

As for vowel length, the current article only mentions it in connection with yumuşak ge. But again there are many words with long vowels, e.g. sa:de, ka:fir, kahveha:ne, ka:nu:n, etc. Obviously in an article of this length, we can't go into the details.... --Macrakis 22:48, 23 March 2007 (UTC)

It says "usually" though, not always. I think for example that the stress moves to the syllable before the syllable "me" when one makes a verb negative. As for vowel length, many of those long vowels are actually in loan words from Arabic.--Domitius 23:04, 23 March 2007 (UTC)

Yes, I mentioned the '*-me case, and yes, the long vowels all come (as far as I know) from Arabic and Persian, but so what? They are active, assimilated parts of the modern vocabulary. Unless the language reformers have been working overtime, I think you still go to the kahvehane and order your sade with long vowels.... --Macrakis 17:01, 24 March 2007 (UTC)

unless you were referring to Sade :) We can add that 'many loanwords are exceptions to the stress is on te last syllable rule', that will cover all the words Macrakis listed above, except 'kolkola', which is in fact 'kol kola' (arm to arm). In that case, in my opinion, we should also note whether most of loanwords are exceptions to 'stress is on the last syllable' rule.deniz^T_C 21:42, 24 March 2007 (UTC)

Macrakis, please check that section on the main article, again, I made some changes. What do you think? deniz^T_C 04:15, 17 April 2007 (UTC)

Number

EU sources are reliable, but Ethnologue is reliable too. I think it's fine to include both sources. See also Turks in Germany. Khoikhoi 01:27, 28 March 2007 (UTC)

Yes, it is reliable, but for 1987.. Since it is given as the lower end of the spectrum, then it would be better to have source from a couple of years back for the lower end of the spectrum, rather from two decades ago.. If there is a better source, it supercedes it, and the way it is given it makes it look like it refers to an estimate of today. It really doesn't make much sense. The range given goes 50 percent up from the lower range - that is not a range, we might as well write we have no friggin idea how many people speak it!! lool. I am taking it out, if it will be included, then mention it after the recent EU survey, and give the 1987 date.. Come on, what is the big deal? First at Kurds in Turkey, now here? Is the EU and private survey agencies engaged in a vast conspiracy theory to screw the Kurds? That's just the way it is: CIA and Ethnologue take a backseat to surveys from 2006 and 2007, I am sorry but that is the case.. The ethnicity survey in the article was an extremely quality, vast and comprehensive survey done by some of the best universities in Turkey, and their surveys include the precise methodology that they have employed in gathering that information - what is the source for the CIA figure? It is some random rounded figure 80-20. No academic would take that seriously against precise figures of 13,xx and 15,xx from a survey from two weeks ago.. Baristarim 01:35, 28 March 2007 (UTC)

Ok, I've found another source, it's dated 2000. It says, "Turkish has close to 50 million native speakers, most of them in Turkey, but there are sizeable indigenous communities also in Cyprus ..." Khoikhoi 01:39, 28 March 2007 (UTC)

Listen, that source you cited most probably used the Ethnologue figure, can you take a look at its bibliography and check it please? That Eurostat survey was one of the first in its kind as far as Turkey is concerned, as well. It is used in the Languages of the European Union - if it is good enough for there, it must be good enough for here as well. Nobody is placing outdated Ethnologue figures in that article, since it is pretty much assumed that a EU 2006 survey by Eurostat supercedes an Ethnologue figure from 1987.. That's all I am saying. If we would like to have the best encyclopedia possible, we have to use the best sources out there. I really doubt that Eurostat is out there to get the Kurds, you know - It clearly says 93 percent of TR population, 8 percent of Bulgaria's population + overwhelming majority of the Turkish diaspora in the EU speak Turkish as a native language. 50 million for all TR speakers (which would mean 45 for Turkey), would mean 65 percent of Turkey's population - if there is anyone in Eurostat who is getting their survey results wrong by 28 percent, I have nothing more to say, but if not, that's the only figure we must be using, right? Baristarim 01:53, 28 March 2007 (UTC)

The other survey in the Kurds in Turkey article (interestingly done by Turks) gives a slightly lower percentage for native Turkish speakers in Turkey, we can use that I think for the lower end number.. Baristarim 01:56, 28 March 2007 (UTC)

I see no harm in keeping multiple sources and giving a range, but using a reference from 1987 (reporting 46M native speakers in Turkey) really does not make much sense (taking native Turkish speakers as 80%, the lowest estimate, with the current population gives 57.5 M native speakers). The source from 2000 (Language and Nationalism in Europe) seems like a social study and a secondary source for this matter. As I understand, the reference introduced by Baristarim reports 93% native speakers (67.5 M) and I changed the 74+M into this figure. Am I right? Atilim Gunes Baydin 02:23, 28 March 2007 (UTC)

As a note, please do not make blind reverts to versions prior to my last edit (I fixed a broken reference and another's format). Could Khoikhoi please add his reference on top of this version, if he decides to revert? Thanks, Atilim Gunes Baydin 02:27, 28 March 2007 (UTC)

Not quite - it says 93 percent for Turkey, but it also says 8 percent for Bulgaria, mentions Greece and Cyprus + the Turkish diaspora in Europe, I added them up from the report and got ~73,5+ mil. I know that at first sight it would seem a bit odd to have 74mil for 93 percent, but the catch is that the report also includes the number of native Turkish speakers in Europe.. Baristarim 02:39, 28 March 2007 (UTC)

Thanks for fixing the reference. Sorry about earlier reverts btw. Baristarim 02:44, 28 March 2007 (UTC)

Oh I'm very sorry, I suddenly forgot about people outside Turkey. It might be good to state this in detail in the geographical distribution section (something like 67.5 M in Turkey (93%), 74M world total). It would also prevent future editors from making the same mistake. Atilim Gunes Baydin 02:46, 28 March 2007 (UTC)

Sure.. This article will reach FA eventually, I am confident :) Baristarim 02:54, 28 March 2007 (UTC)

Some trimming of the article would help. Artaxiad 04:32, 28 March 2007 (UTC)

Ok, I've checked the book, and could find no evidence that it uses Ethnologue as a source. Saying that the book "most probably used the Ethnologue figure" is original research. Since it appears to be a third-party, published work, I think it meets WP:RS. I'm going to restore it based on that. If you can prove that it indeed got the figure from Ethnologue, it can be removed. Khoikhoi 05:45, 1 April 2007 (UTC)

Khoikhoi, what's your personal impression / guess regarding the reason for this discrepancy of about 20 million people between the two figures? Atilim Gunes Baydin 15:08, 1 April 2007 (UTC)

Even the Eurostat figures are new in their domain: such statistics were not done before. I am sorry, but considering the Eurostat's reliability, logistic and academic organization as well as its budget, I am sure it has the capacity to reach a much better number than any single author: In fact, that survey was done to serve as a reference to other individual authors and encyclopedias. The discrepancy is way too big (33 percent) for it to be considered a normal "range". It is seriously making the article look bad. Eurostat figures from 2006 are reliable, sound and are used in many articles in Wikipedia like Languages of the European Union. What is our goal as a Wikipedian? To have the best and most accurate encyclopedia as possible, and it is just common sense that Eurostat statistics are some of the best and most reliable figures out there, am I wrong? Giving an individual author equal weight also violates "Undue weight". Eurostat's survey organization is clearly set out and scientifically explained, which is not the case for all these authors + I really doubt that Eurostat has been bribed by Turkey or anything. Baristarim 16:40, 1 April 2007 (UTC)

Favoring one source over another when they both meet WP:RS violates WP:V. Khoikhoi 18:07, 1 April 2007 (UTC)

But it also violates Undue weight, giving equal coverage to a 2000 book by an individual author as much as 2006 figures from Eurostat, a huge, scientific, impartial and serious organization whose statistics form the basis of many books in real life, and articles such as Languages of the European Union in Wikipedia. Surely you must see that a source who is 40 percent off than 2006 Eurostat figures doesn't have the same stature and is undue weight? Baristarim 18:45, 1 April 2007 (UTC)

Anyways, no biggie. The latest version seems ok and gives a reasonable range. Baristarim 18:48, 1 April 2007 (UTC)

The "Language and Nationalism in Europe" book [2] is not a language atlas (it discusses the influence of language on the national identities in Europe), and does not primarily pertain to the place it is cited here (total number of Turkish speakers). A new reference I've found today, "Languages of the World" [3] by Katzner, is a recent language atlas documenting language families of the world with the total number of speakers, reports 60M native speakers in Turkey (also reports it as 90% of the total population), and specifically notes 3M of the Turkish diaspora. Giving a 50 million figure as the world total of Turkish speakers, doesn't comply with this and other current language atlases I've checked today, contradicts the fact that Turkey's population is about 72M (the lowest estimate of 80% Turkish speakers gives 72 * 0.8 = 57.6M, and there are about 5M Turkish speakers outside Turkey, thoroughly referenced on Turkish diaspora article), and makes the "Language and Nationalism in Europe" book unreliable as a reference for this subject, in my view. Atilim Gunes Baydin 18:57, 1 April 2007 (UTC)

Çekoslovakyalılaştıramadıklarımızdan mıydınız?

I often wondered as I was growing up why people kept repeating the claim that 'Çekoslovakyalılaştıramadıklarımızdan mıydınız?' is the longest word in Turkish. Obviously it is made of two words, because what follows the 'mı' needs to be separated. You can easily modify it, though, to make 'Çekoslovakyalılaştıramadıklarımızdanmışsınız', which means, "reportedly you are one of those whom we have not been able to make Czechoslovakian." --InfoCan 19:59, 30 March 2007 (UTC)

heh, not the longest word. "ademimerkeziyetçilik", "kuyruksallayangiller" are the longest acc. to TDK. But u can build up a word as long as u want by putting more suffixes.

Word order & "verbals"

As well as stating that Turkish word order is SOV, it would be useful to point out the fundamental principle that modifiers precede the modified. This is why relative clauses precede their subjects, as in the example about the fashion reporter N. S.

Incidentally, I too am puzzled by gazetemiz yazarlarından: I would have expected gazetemizin yazarlarından. The table simply says gazetemiz(in) without any further explanation—which will only confuse the reader.

Relative clauses The usual English term for "verbals" is participles. In Turkish there are two types, only one of which (the more straightforward type!) is dealt with in the article. The other type, sometimes called a "relative participle", ought to be mentioned as well, because it is rather more difficult for speakers of European languages to master. I'm referring to constructions such as doğduğum yere döneceğim (I shall return to the place where I was born) or kesemediğin eli öp (kiss the hand [which] you cannot cut off). Some discussion of this construction is essential.

I'm afraid I don't know the details of NS's death; but unless she suffered an unusual death "lost her life" isn't quite right. My suggested translation:

Our reporter/correspondent NS, who brought fashion to the pages of Turkish newspapers, has died.

--NigelG (or Ndsg) | Talk 22:11, 16 April 2007 (UTC)

Participles again Perhaps the clearest way of explaining these would be to take two very simple contrasting examples. For example, you could discuss the difference between:

bizi gören adam ("the us-seeing man — the man who saw us")

gördüğümüz adam ("the our-seen man — the man [whom] we saw")

--NigelG (or Ndsg) | Talk 08:25, 17 April 2007 (UTC)

How about:

"One of our newspaper's correspondents, N. S.,

who brought fashion to the pages of Turkish newspapers, has died."

Gazetemiz yazarlarından is correct, gazetemizin yazarlarından is also correct.

The first one is like "it's our newspaper, s/he is one of the correspondents", and latter one is like "s/he is a correspondent of our newspaper"

The first one is related to "gazete yazarı" (newspaper correspondent), second one is related to "gazetenin yazarı" ([this] newspaper's correspondent)

I hope I could make it somewhat clear, there is not much difference between them. deniz^T_C 08:55, 17 April 2007 (UTC)

I can certainly see what is meant by gazete yazarı; but in this case I'd expect gazete yazarlarımızdan—wouldn't you? But if you, as a native speaker I presume, feel that gazetemiz yazarlarından sounds OK, I'll have to accept it! --NigelG (or Ndsg) | Talk 10:20, 17 April 2007 (UTC)

gazete yazarlarımızdan is a third possibility here (and it is derived from gazete yazarı as well, but by applying "-miz" to "yazarı", not "gazete"), though it's clearly different from the two mentioned above. The newspaper is not 'ours' in this case.deniz^T_C 18:46, 17 April 2007 (UTC)

I'm still fascinated by gazetemiz yazarlarından. Do other native speakers agree that it sounds right? If so, would someone else please try to explain the grammar to me? I'd be most grateful. If it is OK, then it should be possible to say gazetemiz yazarı—which sounds completely ungrammatical to me. I don't expect languages to be completely logical, but they ought to be consistent! --NigelG (or Ndsg) | Talk 23:06, 17 April 2007 (UTC)

I made google searches now, "gazetemiz yazarı" is more popular than "gazetemizin yazarı" and "gazete yazarımız", 16100 vs 2620 vs 71, respectively. Anyway, we can change the sentence deniz^T_C 01:22, 18 April 2007 (UTC)

I think, s/he doesn't have to be a correspondent, if that means on-the-scene reporter. What is the term that encompasses correspondents and commentators? journalists? deniz^T_C 09:04, 17 April 2007 (UTC)

In the UK we talk about "feature writers", which would be quite appropriate for journalists who write regular, but not necessarily daily, articles on topics such as fashion. --NigelG (or Ndsg) | Talk 10:20, 17 April 2007 (UTC)

Instrumental case??

I don't think that -le is usually considered a case, is it? (-le hali diye bir şey var mı yani?). Maybe I'm out of date! In any case, it can only confuse the reader if you say:

"which could be treated as an instrumental case"

"Well, is it or isn't it?" the reader is going to ask. This sounds a bit like WP:OR. --NigelG (or Ndsg) | Talk 08:46, 17 April 2007 (UTC)

It might be OR. The term "-le hali" might not exist, but we have this instrumental case out there, a grammatical case, that matches this. Do we have the term "-den hali" (ablative case)? deniz^T_C 09:16, 17 April 2007 (UTC)

Could you please indent your replies (using :: after this comment of mine)? It makes it much easier to follow the discussion.

I don't know what you mean by "out there", unless you mean that the term exists (for example in Russian grammar). Well, there are 15 cases in Finnish ...

Yes, -den hali & -de hali are accepted terms in Turkish grammar: try google. -le hali, however, is not used to the best of my knowledge. --NigelG (or Ndsg) | Talk 10:34, 17 April 2007 (UTC)

I think, "-le" might be considered a 'name to adverb' suffix, as well. I am no professional linguist, we might need a professional opinion. The google results are mostly from eksisozluk.com replicas and some forums. What they do might be similar to what we are doing with "-le" here. There is a case called ablative case and it matches "-den hali" in Turkish. Likewise, we can for instance say that Turkish has abessive case ("-siz") as well. We need scholarly researches. deniz^T_C 18:21, 17 April 2007 (UTC)

WP doesn't decide on matters like this, it reports what WP:Reliable sources say. In this case, G.L. Lewis's Turkish Grammar treats it is a postposition. After all, it can follow the genitive case suffix in constructions like kiminle. --Macrakis 19:05, 17 April 2007 (UTC)

I remember seeing a mention of the Turkish instrumental case in some English sources. I believe the existence (or, non-existence) of a -le hali in Turkish grammar books is not a good argument for the matter. Also see Chuvash language, Kazakh language, and many others from the Turkic family, which are also reported to have an instrumental case. In any case, it's just a matter of checking a few authoritative sources and reporting it here. Note: I'm the one who wrote the part on Turkish noun cases in the article, sometime around last year, and it was just a quick addition of some missing fundamental information. I had less concern about absolute correctness and reliability given the overall status of the article at the time. Atilim Gunes Baydin 23:14, 23 April 2007 (UTC)

IPA in stress examples

I've made the examples consistent by using the informal method already used in "Istánbul" and "Ánkara". But this could be confusing, since everything else in this section is in IPA! These examples should probably all be given phonemically in IPA (eg /i'stambul/, /'ankara/, etc).

I've also removed the irrelevant wikilinks to English language in the translations of the examples. --NigelG (or Ndsg) | Talk 09:33, 17 April 2007 (UTC)

FAC submission

I'm sorry to say this, but the FAC will not be taken very seriously unless someone gives a much better & more complete nomination than "Very Beatiful [sic] article". If the editors who have worked—and are still working—hard on this article want to nominate it, they should explain why it deserves to be a FA, what improvements have been made to it, how it meets the criteria, etc. --NigelG (or Ndsg) | Talk 10:51, 17 April 2007 (UTC)

I've elaborated the nomination for now. I was not involved in this article however, so perhaps those who were could give a better explanation on the FAC page. --A.Garnet 12:16, 17 April 2007 (UTC)

That's a very useful first step: thanks. Has this article ever had a peer review? If so, the fact ought to be mentioned; if not, perhaps it should have one. I've made a few suggestions here recently, but haven't had time to look critically at the whole article.

I don't think this article had a peer review after becoming a GA. We should ask for one. Thanks Uannis for nomination. I think we should put the nomination on hold for now. At this state, the article might become an FA, but let's have a stronger case. deniz^T_C 17:45, 17 April 2007 (UTC)

My comments have been from the point of view of a foreign student of the language, which is why I've concentrated on features such as participles, which are quite different from their equivalents in W European languages. Remember that this article is not just aimed at Turks: in fact, most people who refer to it will not be fluent speakers of the language, but will be interested in it from a general linguistic point of view. --NigelG (or Ndsg) | Talk 17:10, 17 April 2007 (UTC)

Hi, I think that the FA nomination was somewhat premature. I regret lacking the time to take care of the article (until a few more weeks), but I'm very happy to see new users getting involved. I also liked Nigel's comprehensive comments on the article's flaws. They are most welcome and will be very useful for improvements. Kind regards, Atilim Gunes Baydin 23:06, 23 April 2007 (UTC)

Allophones of L

Although both [l] and [ɫ] appear in the table of consonants, there is no mention of them in the text. It should be made clear that the "dark" [ɫ] occurs with back vowels and the "clear" [l] with front vowels. In phonemic transcriptions /l/ would be in principle be sufficient for both—just as /k/ could be used for both [k] and [c]. --NigelG (or Ndsg) | Talk 18:08, 17 April 2007 (UTC)

Gazetemiz yazarı (continued)

You're quite right: there are lots of examples on google! Perhaps this is a special case—a sort of journalistic jargon (similar in a way to restaurant jargon such as ızgara köfte, where you might expect *ızgara köftesi). After all, the "normal" izafet construction would be gazetemizin yazarı, just like babamın arkadaşı, not *babam arkadaşı.

There are similar examples of jargon in English journalism. For example, it's common to hear on the radio an introduction like the following: "This report from our correspondent AB" (no verb at all: in normal spoken English you'd have to say "This report is ..." or "We now bring you this report ...").

Would you agree with my "jargon" explanation? (Others, please join in the fun!). --NigelG (or Ndsg) | Talk 09:18, 18 April 2007 (UTC)

I think I may have another explanation. If we say gazetemiz = Cumhuriyet gazetesi, then gazetemiz yazarlarından is another way of saying Cumhuriyet gazetesi yazarlarından. I can accept that! --NigelG (or Ndsg) | Talk 18:13, 18 April 2007 (UTC)

Both gazetemiz yazarı and gazetemizin yazarı are acceptable forms of noun completion, the former is called a belirtisiz isim tamlaması, and the latter a belirtili isim tamlaması in Turkish grammar. There are also other types (see [4] for a quick overview). You are quite right with your suspicion of jargon here: the preference of the belirtisiz type constitutes a journalistic jargon. Atilim Gunes Baydin 11:06, 24 April 2007 (UTC)

Phonetic spelling

The final sentence of the lead section states:

"Turkish is characterized by vowel harmony, agglutination, lack of grammatical gender, and a highly phonetic spelling."

Spelling is not really a feature of the language. If you want to mention it, I suggest writing a separate sentence, such as "The modern orthography is highly phonetic". --NigelG (or Ndsg) | Talk 10:40, 18 April 2007 (UTC)

Writing system: phonetic transcription

"Following International Phonetic Alphabet conventions on phonetic transcription, angle brackets < > here are used to enclose written letters, and brackets [ ] are used to enclose symbols that represent the sounds."

In fact most—but not all—of the transcriptions in this section are between /slashes/, representing phonemic transcriptions. Please use one or the other consistently. --NigelG (or Ndsg) | Talk 10:57, 18 April 2007 (UTC)

And another point:

"The <ı> represents /ɨ/, a sound which does not exist in English."

Elsewhere the symbol [ɯ] is used for this sound. Please choose one or the other (I think [ɯ] is better, but I don't know which one is preferred by Turkish phoneticians). --NigelG (or Ndsg) | Talk 11:12, 18 April 2007 (UTC)

Style The last paragraph of this section reads awkwardly, chiefly owing to the slightly odd vocabulary used (critical ... atop ... multiplexed ... vital ... conflated ...serviceable ...sports). Some copyediting needed here. --NigelG (or Ndsg) | Talk 11:24, 18 April 2007 (UTC)

Words "döğmek" and "söğmek" are officially changed to "dövmek" and "sövmek". (They have been written in the latter way for a long time)

Language reform and Modern Turkish

The 3rd paragraph of this section needs rewriting. For one thing, the phrase "A significant amount of" is an example of the "weasel words" which should be avoided in Wikipedia. But a more important point is that the example (dert/ağrı) is a poor choice, because TDK did not invent or encourage the use of ağrı. It should be possible to find a better example from the List of replaced loanwords in Turkish‎. The final sentence, with its interesting comparison with Germanic/Romance words in English (pig/pork?), needs an example from Turkish. Citations should also be given wherever possible. --NigelG (or Ndsg) | Talk 10:02, 20 April 2007 (UTC)

Summaries don't need extra footnotes

I removed the redundant footnotes used in the lead and the infobox. They're both supposed to be summaries of the entire article, quick references rather detailed surveys in their own right complete with their own separate set of notes. Also, please don't repeat footnotes three times in the same sentence. It just breaks up the text for very little gin. Figures are extremely easy to check in online sources since they can simply be entered as a search. Unlike prose, you don't need to read entire texts to verify them. Same thing is pretty much true for print sources, since a cursory glance at a page will quickly reveal any figures.

Peter ^Isotalo 17:14, 21 April 2007 (UTC)

The reason for the over referencing of the introduction paragraph and the figures in the infobox was that it's very common for these (especially the population figures) to get challenged, almost on a monthly basis. I agree that it looks more clean without the footnotes in the introduction, but they were helping to somewhat stabilize the article against such edit wars. I fear, if the population numbers stay without a reference in the infobox and the intro, some user will eventually come and change just the number there, without noticing and caring for the existing reference below the page. I saw this happening quite a few times in the past. Hej då! Atilim Gunes Baydin 22:52, 23 April 2007 (UTC)

History section

Should we add poems (first quatrain or beyit) from each period, like Alp Er Tunga Destanı or a poem by Yunus Emre, then a poem from the classical Ottoman Turkish literature, and a folk song from the same time period (Pir Sultan Abdal, Karacaoğlan, etc) and following these, a modern Turkish poem (Nazım Hikmet or Orhan Veli, etc)? The English translations of some of them are already on other Wikipedia articles. The problem would be IPA (and maybe transliterations of some, also it would be better to have the modern Turkish translations for comparisons). If we do this, what should we do with the oral ones (Turkish folk songs from Ottoman times)? deniz^T_C 10:15, 24 April 2007 (UTC)

It's a nice idea, but the article is already getting too long. In fact, it wouldn't be a bad idea to shorten the existing Aşık Veysel example. The examples you suggest should properly be in Turkish literature (& some of them probably are already there). --NigelG (or Ndsg) | Talk 10:33, 24 April 2007 (UTC)

All the wikilinked ones above, except Karacaoglan, have poems we can use. Can you do the IPA part? We have but one problem with the "Pir Sultan Abdal" one, should we use Ottoman alphabet there? Also, certainly Kaşgarlı Mahmut was not using Latin alphabet. deniz^T_C 10:50, 24 April 2007 (UTC)

I think we don't need the modern ones. deniz^T_C 10:50, 24 April 2007 (UTC)

I don't think my answer was clear. I do not think such examples belong in this article. When I said "It's a nice idea, but ..." I meant No! But it's up to you, of course. --NigelG (or Ndsg) | Talk 11:06, 24 April 2007 (UTC)

I think it will be fine. It will be a good comparison. i am going to go ahead and do the edit (just one poem), please check it. You can revert. If we don't need IPA, we can remove it. deniz^T_C 11:17, 24 April 2007 (UTC)

I don't think, as well, that the article needs more than one example text and a collection of excerpts from literature of the Ottoman or older periods, per other FA-class language articles. Also keep in mind that Ottoman Turkish has its separate article and it would make sense to make these additions to improve that and Turkish literature. Please also note that, by definition, we should keep the focus on modern Turkish in this article. Atilim Gunes Baydin 11:43, 24 April 2007 (UTC)

Possible compromise You could substitute a poem of Yunus Emre for the Aşık Veysel. YE is a poet of world stature, which is not the case with AV (excellent though his singing may be). Perhaps the one with the refrain Acep şu yerde varm'ola Şöyle garip bencileyin. --NigelG (or Ndsg) | Talk 14:55, 24 April 2007 (UTC)

I again want to stress that this article is about current Turkish and as such, I oppose replacing Aşık Veysel with Yunus Emre as the main example in this article. The reason for having Aşık Veysel in the first place was that his work pertains to Modern Turkish (he died in 1973), and his language is a very good representation of the current spoken language (which is what the "example" section is about). The works of Yunus Emre (1238 – 1320) are much closer to Ottoman Turkish. Atilim Gunes Baydin 15:25, 24 April 2007 (UTC)

I think you're right. What's more, all the hard work with the IPA has already been done—so leave it as it is. --NigelG (or Ndsg) | Talk 17:45, 24 April 2007 (UTC)

To-do list

I think if we again use a ";to-do list" approach, where all the issues needing attention are added to a pool (a bulleted list) and the contributors strike the entries through as they make the required changes, it would do much to coordinate individual efforts and measure progress. It also would be very useful for the voters on the FA nomination to clearly see what progress has already been made with addressing the issues raised on the voting page. I saw this approach was of great help with other featured article drives in the past. I've been missing for a few weeks, and I find it very hard to see which of the issues raised here on this talk page are already fixed in the article and which await attention. Would the other editors agree with this working method? I hope they will. Atilim Gunes Baydin 11:28, 24 April 2007 (UTC)

Yannis did that. The basic problem is I believe having too many tables (I removed one of them now, please check Turkish_language#Personal_pronouns, and better the English, if need be). Peter's suggestions are mostly covered, as far as I can see, but we still need to work on them. So basically tables, some sentences that need references, and the section Turkish_language#Language_Reform_and_Modern_Turkish are the things that need immediate attention. deniz^T_C 11:46, 24 April 2007 (UTC)

Having many subsections was also criticized. deniz^T_C 11:52, 24 April 2007 (UTC)

Thanks Deniz! I merged the numbered list you've made here with the existing one. Atilim Gunes Baydin 14:08, 24 April 2007 (UTC)

(We can keep this here or move it to the top of the page, please also note that you can add the to-do list to your watchlist) To-do list moved to the top of this page. Atilim Gunes Baydin 12:40, 25 April 2007 (UTC)

Name of this language in Turkish

I suggest, unless you have strong feelings to the contrary, that the opening words of the article should be:

Turkish (Türkçe or Türkiye Türkçesi, "the Turkish of Turkey") is a Turkic language ...

That at least makes it quite clear from the outset what we're talking about! --NigelG (or Ndsg) | Talk 14:49, 24 April 2007 (UTC)

I won't have anything to the contrary, but the constant repetition of the word Turk (7x in your version above) could be boring / annoying to some readers. Could you please explain why this is needed? Atilim Gunes Baydin 14:58, 24 April 2007 (UTC)

Namecruft in the lead is always dull and seldom appeals to anyone except those who already know the terms. The most common name is fine.

Peter ^Isotalo 15:05, 24 April 2007 (UTC)

OK, the repetition worried me a bit too. I just felt that it ought to be made crystal-clear that Turkish is (more or less) confined to Anatolia. An attempt is made at the end of the paragraph to extend it to Iran & Turkmenistan—to increase the number of speakers? --NigelG (or Ndsg) | Talk 15:18, 24 April 2007 (UTC)

Ikiroid's last fix

Hi, just a quick note regarding your last edit: I was just trying to make it clear that it is the Turkic language family in general that this 1200 year history concerns, and this date is established by the Orkhon inscriptions in the Old Turkic language. I was deliberately bending the "style" a bit in the process. The reason for this is that Old Turkic is not a direct ancestor of today's Anatolian Turkish, and it is possible to get that impression from your version. Happy editing, Atilim Gunes Baydin 21:42, 24 April 2007 (UTC)

OK, sorry about that. By the way, great job. The ikiroid (talk·desk·Advise me) 21:59, 24 April 2007 (UTC)

No need to worry. I'm not sure, we can also keep it your way. Atilim Gunes Baydin 22:03, 24 April 2007 (UTC)

I only changed your edit for aesthetics, feel free to change the paragraph again so that it is closer to what it should be. The ikiroid (talk·desk·Advise me) 22:22, 24 April 2007 (UTC)

I changed it again, I think it looks better this time. Atilim Gunes Baydin 23:22, 24 April 2007 (UTC)

Turkish phonology

I have created a sub-article, Turkish phonology, from the information in this section. I guess we can now trim the section down a little so that it is summary style. I'm afraid to start cutting out information, however—I'd be more comfortable if a more frequent editor did it. The ikiroid (talk·desk·Advise me) 22:21, 24 April 2007 (UTC)

I'm not sure what part can be trimmed down from that section. I surely think that we should have the IPA chart and the vowel / consonant tables in this article, and this leaves only the part on vowel harmony as the candidate. The phonology section, overall, doesn't look very long for my taste. Atilim Gunes Baydin 23:25, 24 April 2007 (UTC)

Please check the classification section deniz^T_C 23:34, 24 April 2007 (UTC)

Please also check the references section, and make sure that the sources listed as printed sources have printed versions, and sources listed as online sources don't. also, we might need to change some citation templates. deniz^T_C 00:32, 25 April 2007 (UTC)

I disagree with Atilim's response. The Sounds section could certainly be summarized; and we don't need the detailed IPA tables in this article, now that the details have been moved to the new Turkish phonology article. Vowel harmony must, of course, remain—but that too could be summarized.

Ikiroid & I both recently worked on an article on the Chinese romanization system Gwoyeu Romatzyh, which greatly benefited from the creation of a new detailed article on spelling, which was suggested to us during peer review.

As you may have noticed, the following note now appears when you edit the main article:

This page is 55 kilobytes long. It may be appropriate to split this article into smaller, more specific articles. See Wikipedia:Article size. --NigelG (or Ndsg) | Talk 12:07, 25 April 2007 (UTC)

Well, the 55 isn't that bad....we cut down International Phonetic Alphabet some time back into numerous subarticles and still ended up with 72K. The phonology section should really stay at or below its current size, with any future additions being made to the subarticle instead of the section. It may help to trim off the fat in some other sections though. The ikiroid (talk·desk·Advise me) 19:14, 25 April 2007 (UTC)

I've had a look at a few other X Language articles, & I see that IPA tables are the norm. OK, they'd better stay, then! I have, however, removed the rather long list of sample words, which can safely be relegated to the specialized article. --NigelG (or Ndsg) | Talk 11:09, 26 April 2007 (UTC)

The main text states: "The sounds [c], [ɟ] and [l] are in complementary distribution with [k], [g] and [ɫ]". In that case, there are three sounds here which are phonemes, not six; the three which are only allophones should be removed from the table of phonemes. I would suggest keeping /k g l/ and removing [c], [ɟ]and [ɫ], which of course may still be mentioned as allophones. If we listed all the possible allophones as "phonemes" in the chart, it would probably, like most languages, contain 50 or 60 members... Jakob37 03:54, 20 May 2007 (UTC)

I don't think that follows. Sounds in complementary distribution (CD) are not necessarily allophones (though they often are). For example, the alveolo-palatals j, x and q in Mandarin Chinese (Pinyin) are in CD to the retroflex sounds zh, sh and ch; yet no one describes these as allophones of the same 3 phonemes. In Turkish the difference between /c/ and /k/ is sometimes phonemic: eg kar /kaɾ/ ("snow") vs. kâr /caɾ/ ("profit"), where the circumflex indicates palatalization. Indeed, it may be no more than a historical accident that both /c/ and /k/ are represented by <k> in the modern orthography. There is an amusing anecdote about Atatürk's dislike of the letter Q in his own name (which would have become Qemal rather than Kemal) leading to rejection of to indicate <c>. In some cases there is a phonemic distinction between /l/ and /ɫ/, too: eg kal /kaɫ/ "stay" and the loanword kalp /kalp/ "heart".

--NigelG (or Ndsg) | Talk 21:25, 22 May 2007 (UTC)

The keys to distinguishing between allophones vs. separate phonemes is context and generalization. While i don't know much about Turkish, it seems that there is a context here with explanatory significance, and the generalization over 3 pairs of phonemes is convincing to me. On Mandarin (where i do know a little more): Agreed, alveo-palatals and retroflexes are not suggested as allophone sets. But this is because velars (g, k and h) are a more likely pairing to the alveo-palatals (thus: g/j, k/x, h/q instead of zh/j, sh/x, ch/q) for reasons pertaining to historical evolution of the sounds (cf. Hartmann 1944 and Hockett 1947, quoted in Norman 1988). --SteveMtl 04:25, 23 May 2007 (UTC)

At this point I ought to come clean & admit what is probably obvious in any case: I'm not a phonetician. The palatalized allophones (if such they are) of /k/ and /g/, and the "dark" version of /l/ were already given in the table when I first came across this article, so I just left them. I knew about the CD in Chinese, so extrapolated from that to Turkish. I think on reflection that Jakob37 is probably right in suggesting that we restrict the sounds to /k g l/ and mention [c], [ɟ]and [ɫ] as allophones.

I'm not sure what the implication of all this is for the IPA transcription of the poem at the end of the article, however. As it stands, it seems to be a narrow phonetic transcription: is that what we want? --NigelG (or Ndsg) | Talk 21:28, 23 May 2007 (UTC)

PS On second thoughts, if [k] and [c] are mere allophones, it's curious that the Ottoman script used two different Arabic letters for them ( ق kaf and ك kef). The same applies to [g] and [ɟ] ([often] غ gayın and گ gef). AFAIK, orthographies rarely distinguish between allophones. I know this argument isn't conclusive, but it does give an indication. --NigelG (or Ndsg) | Talk 08:38, 24 May 2007 (UTC)

There are many curious things about the Ottoman script, and a few curious things about the modern Turkish script. For example, Ottoman distinguishes `ain from hamza, and has several different ways of writing Turkish 'h' (different sounds in Arabic). On the other hand, kaf can represent k, g, n/ŋ, v, and y. Ottoman distinguishes k/c systematically, g/ɟ sometimes, and l/ɫ never. And let's not talk about Ottoman ghain and modern yumuşak ge.... Ottoman encodes the front/back nature of the vowels in some consonants just as Russian distinguishes hard/soft velars, which are probably allophonic. In many of these cases, more than one analysis is possible, but I certainly wouldn't put much weight on either Ottoman or modern orthography as evidence for phonology. --Macrakis 14:19, 24 May 2007 (UTC)

/c/, /ɟ/, /l/ are unquestionably different phonemes from /k/, /ɡ/, and /ɫ/, respectively. /kɑɾ/ is a different word from /cɑɾ/, /ɾol/ does not rhyme with /doɫ/... And you should hear the way people around here say /ɑlo/: the palatalization is as plain as day – it's a completely different sound from what you hear in /oɫɑn/. Geoffrey Lewis also makes it plain that the distribution of these phones is not predictable (although he notes them as /k^y/, etc). Q·L·1968 ☿ 11:47, 5 September 2007 (UTC)

In native Turkish words the difference between /c/ and /k/ etc isn't phonemic. It's only in a small number of loanwords (eg your example of the Persian /caɾ/ "profit" vs. (versus, not ve saire!) the native /kɑɾ/ "snow" that the difference is phonemic. So it's incorrect in general to say that these differences "are unquestionably different phonemes". --NigelG (or Ndsg) | Talk 14:39, 5 September 2007 (UTC)

BTW Where exactly is "around here": Québec? Ankara? --NigelG (or Ndsg) | Talk 19:44, 5 September 2007 (UTC)

First, it's not a small number. We're talking about a huge set of words borrowed from French, such as those ending in -al, -ol (like alkol, metal) not to mention another huge set taken from Arabic and including a great number of personal names (like Kemal, Celal). These items, plus the ubiquitous mobile-phone exclamation alô, ensure that Turkish speakers pronounce a non-allophonically conditioned palatal /l/ probably dozens of times a day. Second, even if the numbers were small (as for /c/ and /ɟ/), a non-predictable distribution and the existence of minimal pairs clinch the argument in favour of phonemic status. By way of comparison, how many actual contrasts can you think of between /ʒ/ and /dʒ/ in English? There are few. If we said that /ʒ/ is mostly found in Romance borrowings, it still wouldn't let us wriggle out of admitting that /ʒ/ was a phoneme. The distribution's unpredictable, and we still have to face one or two pairs like lesion and legion. And third, sorry I forgot to mention – around here is Kadıköy. All the best, Q·L·1968 ☿ 08:38, 11 September 2007 (UTC)

OK: it isn't such a small number! I don't disagree with anything you've said, really. Are you proposing any change in the text of the article? It looks correct to me as it stands. --NigelG (or Ndsg) | Talk 11:23, 11 September 2007 (UTC)

Nope, I'd already fixed it to my satisfaction. I just wanted to explain why! Cheers, Q·L·1968 ☿ 08:17, 12 September 2007 (UTC)

[outdent] I've changed the wording slightly & given a link to some examples in the Writing system section showing the use of the circumflex.--NigelG (or Ndsg) | Talk 11:20, 13 September 2007 (UTC)

PS Also some examples of irregular vowel harmony in loanword suffixes. --NigelG (or Ndsg) | Talk 17:58, 14 September 2007 (UTC)

Dialects map

Hi, I really want to make a nice looking map of Turkish dialects within Turkey, and possibly in the Balkans, but I do not have access to a good source regarding this matter. I had this in my mind for quite a long time, I remember putting the generic map of Turkey to the dialects section as a placeholder for this. I would be very pleased if some of the contributors in possession of such material could share it (via scanning etc.) with me. It could be an already published map based on which I can make a freely-licensed version, or a detailed description of the geographic distribution of the dialects, perhaps. Regards, Atilim Gunes Baydin 12:49, 25 April 2007 (UTC)

Participles

I've added a new section on the important topic of participles. At the same time I removed the example about the Cumhuriyet fashion editor from the Word order section (where it didn't really belong anyway) & added a couple of sentences to that section.

The table of examples of participles may be rather too long: if so, one or two of the examples could go. --NigelG (or Ndsg) | Talk 18:41, 25 April 2007 (UTC)

Alaska

I couldn't help noticing that Alaska is shown on the map as a region "with a significant Turkish-speaking population". Was there an eastwards migration across the Bering Strait that we didn't know about? ;^) --NigelG (or Ndsg) | Talk 09:10, 26 April 2007 (UTC)

It's part of the United States, and so—necessarily—must be shaded in so long as the United States is. Cheers. —Saposcat 09:28, 26 April 2007 (UTC)

By the way, I's stoopid and wasn't notices yer ";^)". —Saposcat 09:49, 26 April 2007 (UTC)

Lead section

I've deleted the following (possibly controversial) sentence from the intro, since no further reference to it appears in the article:

There is a high degree of mutual intelligibility between Turkish and other Oghuz languages, including Azeri, Turkmen, and Qashqai, and if these are counted together as "Turkish," the number of native speakers is close to 106 million.

--NigelG (or Ndsg) | Talk 09:27, 26 April 2007 (UTC)

I've reinserted the mutual intelligibility info, without the possibly controversial "counting together", to the classification section. I think it's interesting and worth mentioning. Atilim Gunes Baydin 12:09, 26 April 2007 (UTC)

Fair enough. Have you got a citation for mutual intelligibility? --NigelG (or Ndsg) | Talk 12:18, 26 April 2007 (UTC)

It's not that rare...while most Germanic languages often change to a high degree from dialect to dialect, such a phenomenon happens in many language groups, like the Slavic, Scandinavian and Iberian languages (excluding Basque). The ikiroid (talk·desk·Advise me) 12:22, 26 April 2007 (UTC)

Who said it's rare? The existing reference from Katzner mentions a mutual intelligibility between all Turkic languages in general, but has nothing specific to the Oghuz branch. I'll see if I can come up with a more specific reference. Atilim Gunes Baydin 13:03, 26 April 2007 (UTC)

Welcome to Europe picture

I think, we should remove it. We can maybe use it on the article Istanbul (I know it contains a lot of images). Atilim, if you are in Istanbul, and regularly crossing the bridge, can you please take a picture of this without all those policemen, I don't remember them being always there. Thanks Deniz^T_C 14:11, 26 April 2007 (UTC)

I like the road sign picture, I think it renders the article more lively. The policemen seemed a bit distracting at first, but they don't disturb me now after getting used to the picture or just because I couldn't find anything better. I like it, but we can of course take it out if a consensus for its removal appears. Perhaps someone from Istanbul can provide us with a better picture. Note: I'm not in Turkey now, and when I'm there, I stay mostly in Ankara. I had the impression that you are living in Istanbul. Atilim Gunes Baydin 14:30, 26 April 2007 (UTC)

Please do not remove it now! I've linked to it from the Vowel harmony section because it illustrates no fewer than three features of VH. And, who knows, someone may want to discuss why "Europe" is Avrupa kıtası ... --NigelG (or Ndsg) | Talk 16:08, 26 April 2007 (UTC)

Wow, I haven't noticed it's actually used for something very useful in the text. Brilliant job! We might as well move the photo to the vowel harmony section. Atilim Gunes Baydin 16:32, 26 April 2007 (UTC)

You don't really need to move it, since there's a link to it. But by all means move it if you think that would improve the balance, visual appeal, etc.

Since I've started confessing my sins (no, I'm not Catholic: why do you ask?), I must admit that when I first saw the picture I started racking my brains trying to remember the location of that obscure village Viyadüğü ... --NigelG (or Ndsg) | Talk 16:47, 26 April 2007 (UTC)

Lewis's Turkish grammar: a confession

I ought to confess that the references I've been giving to TG are in fact to page numbers in my old 1967 (first) edition, which only has 303 pages (the 2nd edition has 328). Since I can't lay my hands on the new edition, I'd be most grateful if some kind soul would check my page numbers & correct them where necessary.

The footnote numbers, & sections of TG with their subjects, are the following:

16: I, 19 (consonant alternation)

18: I, 34 (vowel harmony)

20: XXIV, 21 (further examples: Bayramlaşamadıklarımız)

21: IX, 7 (personal participles) & XVIII, 2 (relative clauses)

22: as 21

23: XV, 2 (word order)

30: I, 23 (pronunciation of ı)

I don't think any of the references will be very far out (some of the early ones are probably OK). Many thanks to anyone who has the patience to do this on my behalf: I suggest you mention the fact here before you undertake this chore, in order to avoid duplicating the work! --NigelG (or Ndsg) | Talk 16:36, 26 April 2007 (UTC)

(BTW I've just noticed the word çalıştay in the Özsoy et al. reference, meaning "workshop". Yet another new term to make me feel old-fashioned!) --NigelG (or Ndsg) | Talk 16:38, 26 April 2007 (UTC)

I think "çalıştay" is a really cool word! I have the impression that it's almost like a shibboleth for one's passion / appreciation for Turkish. People on the "not care" camp just prefer to say "vörkşap", and they are the majority, actually. Atilim Gunes Baydin 18:33, 27 April 2007 (UTC)

Literacy rates

Getting that map of literacy rates in 1927 was an excellent piece of research. To give the claim about the benefits of the new alphabet a sound statistical basis, we now need some figures for the situation 15-20 years later (~1945), by which time the effects would have made themselves felt. Quoting today's figures (85-90%, or whatever it is) wouldn't really make the point, because a similar long-term effect due to general levels of education has probably been observed in Iran as well, without a change of script. --NigelG (or Ndsg) | Talk 17:04, 26 April 2007 (UTC)

Move 'History' section to beginning?

As per Tamil language, Bengali language and Swedish language, all FA. --A.Garnet 15:51, 27 April 2007 (UTC)

I moved 'History' and 'Writing system' sections up. Deniz^T_C 17:13, 27 April 2007 (UTC)

I moved the "Writing system" section back to the end. I believe that the "Sounds", "Grammar", and following sections were pushed too far below. I think the overall flow is less boring this way. I'm also not sure about having the "History" on top, it really shifts the general focus of the article from current Turkish and what it is ("Sounds", "Grammar", and "Vocabulary" sections) to this "Turkic migrations, Persian influence, undoing that influence, changing the alphabet" issue. And honestly, I don't like focusing on that right at the beginning, even before showing what language is like. I think I prefer the section order as is outlined on Wikipedia:WikiProject Languages/Template. I also think that introducing the reader to the subject with "Geographic distribution" (outlining where, and by how many this language is spoken) was be the most fluent and interesting way, and because of this, it should come right after "Classification". Atilim Gunes Baydin 18:22, 27 April 2007 (UTC)

Dialects

This really is listy (after the first paragraph). To say "Cypriot Turkish, spoken by Turkish Cypriots" doesn't really get us very far. That, & the following other dialects, might just as well not be in this article, since it is impossible to verify them or find out anything about them, until someone comes up with some citations.

• Edirne
• Ege
• Güneydoğu
• Orta Anadolu
• Kastamonu
• Doğu

Rumelice & Doğu sound interesting, but again need some more references.

Perhaps, if there are some good sources, there should be a separate article on dialects, with a brief reference here. --NigelG (or Ndsg) | Talk 20:50, 27 April 2007 (UTC)

Language reform and Modern Turkish (2)

I've edited & rewritten this fairly extensively. A few more citations (for example, about the linguistic generation gap and the political dimension) would be useful. Anyone who has a copy of Lewis's Catastrophic Success might be able to find some useful quotations. I no longer have a copy, unfortunately. --NigelG (or Ndsg) | Talk 20:56, 27 April 2007 (UTC)

Featured article!

FA

The article has been promoted to featured article status. My most sincere congratulations to all contributors: NigelG (or Ndsg), Deniz, The Ikiroid, Peter, Free smyrnan, Michkalas, and many others. Atilim Gunes Baydin 23:20, 7 May 2007 (UTC)

Congratulations to you to as well, and to Nigel, Peter, Ikiroid, Free Smyrnan, Yannis, and everyone else. Thanks everyone. Deniz^T_C 23:35, 7 May 2007 (UTC)

Excellent stuff, well done. --A.Garnet 15:45, 8 May 2007 (UTC)

Congrats all round. A good collective effort. --NigelG (or Ndsg) | Talk 22:15, 13 May 2007 (UTC)

PS Has anyone considered putting this article up for WP:TFA? Hello? Anyone still there? ... --NigelG (or Ndsg) | Talk 09:40, 18 May 2007 (UTC)

PPS The 27 May slot is still vacant ... --NigelG (or Ndsg) | Talk 09:53, 18 May 2007 (UTC)

Sorry! I see that Atilim has already submitted the article. --NigelG (or Ndsg) | Talk 13:00, 18 May 2007 (UTC)

Comitative case

Hey, I once read that Turkish has six and a half cases. The mentioned additional "half case" was the Comitative case (-le, -la suffixes in Turkish). Though they're due to the merging of "ile" with the concerned word, they act as a case, even getting a fusing letter (kaynaştırma harfi) and/or changing the ponetics (ie. Baba + ile becomes babayla, instead of Babayle).

I think this "half case" should also be included in the article. I just didn't want to blindly add with no sources, and since I'm not a language expert I wanted to ask for opinions. Regards, Kerem Özcan 16:43, 12 June 2007 (UTC)

Hi Kerem, this same issue has been discussed before (please see under heading "Instrumental case??" about halfway above on this page). The (pseudo-)case you mentioned had been included in the article for a while as an instrumental case, after it was added by me some time ago. The consensus was to not treat this as a case in the article, as this treatment does not seem to be very common in the international literature. Atilim Gunes Baydin 18:10, 12 June 2007 (UTC)

Ah Ok then. Sorry about that btw. I scanned the page for the word "comitative" but couldn't find anything so that's why I put that here. But it's also interesting that this "pseudo"case is both comitative and instrumental at the same time.

Anyway, If the discussion is already over, no need to bring it back. Regards, Kerem Özcan 20:12, 12 June 2007 (UTC)

Well, no one can say the discussion is over. I think it's best to at least mention this view in the paragraph treating the cases, if there are reputable sources that can be cited. Regards, Atilim Gunes Baydin 12:12, 13 June 2007 (UTC)

Actually I just noticed another thing. According to my Turkish grammer book (which is older than 10 years) the genetive suffix (-in) is also not considered as one of the 5 cases of Turkish. It's rather being studied under the Adpositional phrase (isim tamlamaları). So the suffix "-ün" in "Köyün ağacı" shouldn't be the Genetive case in that manner, but rather a possessive Adpositional phrase (Belirtili isim tamlaması).

So does anybody have any sources for the genetive case in Turkish language. Because when I google "ismin halleri", all I get is 5 cases. (Or when I make a googlefight I get this result :) )

To mention here once more, all that I'm saying here can be totally ignorant or outdated (as can my primary school grammer book). Kerem Özcan 13:51, 13 June 2007 (UTC)

In linguistics it's usually denoted as six cases, although we'd learnt "tamlama" as a distinct topic at school. So the -in case should be mentioned also in a "tamlama" section where should include the other cases such as "x'in y'si", "x y'si" and "x y". Also we should compare "the" with "-i". We may contribute an "postpositional phrases" section. Regards.--OktayD 08:55, 15 June 2007 (UTC)

You seem to be using the word "case" in a non-standard way. In linguistics, things like "x'in y'si", "x y'si" and "x y" are "constructions" or "phrases", not "cases". I believe the international linguistic literature on Turkish is unanimous that -in is a case (conventionally called the "genitive") and that "-i" is a possessive suffix. Perhaps there is a different tradition in Turkish school grammar? --Macrakis 13:46, 15 June 2007 (UTC)

I used the word "case" for the suffixes (i.e. -in and -i) not for the phrases. That's why I used such a notation.

Right, Turkish school grammer states five cases, then the -in suffix is introduced under the "phrases" title. Maybe it is because of that the -in suffix could only be used to construct phrases.--OktayD 16:14, 15 June 2007 (UTC)

Vowel harmony vs. reduplication

The new examples illustrating prefixing - sımsıcak and masmavi - are quite interesting. In the note, however, it states that this (not entirely sure what this refers to) is the only case of vowel harmony working backwards. Is this (in sımsıcak, i presume) a case of vowel harmony, or simply reduplication including the vowel ı?--SteveMtl 16:05, 20 June 2007 (UTC)

"This" refers to adjectival/adverbial reduplication in general. (BTW the examples sımsıcak and masmavi have been in the article for some time now.) As you suggest, it's not strictly speaking vowel harmony, since that would imply something like *mısmavi. But since the reduplication preserves the vowel, it's like vowel harmony but more so. --NigelG (or Ndsg) | Talk 19:29, 20 June 2007 (UTC)

I've reworded the footnote: see what you think of it now. --NigelG (or Ndsg) | Talk 21:53, 20 June 2007 (UTC)

Hmmm. Thank you for considering my concerns, but i'll try to be more clear. I'll stick to sımsıcak, to be as precise as possible. The fact that the first two vowels are the same is not sufficient to make a claim for vowel harmony (VH). VH is a process that causes alternation in the form a vowel. (Otherwise, we could argue that in English murder, with two identical vowels, is the result of VH.) For there to be vowel harmony, we need to have evidence that an underlying form has been altered in order to give the new form (i.e. simsıcak → sımsıcak, where i → ı /_Cı: one expression of the VH rule), or that there could be a competing alternative form simsıcak that is somehow less appropriate (blocked) than the attested form sımsıcak. The reduplication in question can be written CVak → CVmCVak (or Xak → XmXak if you prefer). In this case sıcak → sımsıcak. No VH process is necessary, reverse or otherwise, to get the desired, attested result. In other words, i don't see any evidence of VH at all, and i think the reference to it should simply be removed.--SteveMtl 08:32, 21 June 2007 (UTC)

To be honest, I don't feel very strongly about this. What I wanted to stress was that the intensifying prefix is variable, & that part of the variability is that it preserves the first vowel of the base form. The "competing alternative form" you mention might have been CoK- (where K is a variable consonant chosen for euphony): so we might have seen *mosmavi, *somsıcak and *bosbütün rather than the, well, harmonious masmavi, sımsıcak and büsbütün we do in fact see. If you still find the final sentence in the note misleading, by all means delete it. --NigelG (or Ndsg) | Talk 09:49, 21 June 2007 (UTC)

['tyɾkʧe] or ['tuɾkʧe]?

subj

with y, Kerem Özcan 20:34, 21 June 2007 (UTC)

Most widely spoken?

In the intro it is claim it is the most wide-spread Turkic language. However if I go to Turkic languages, the widest geographical language appears to be Kypchak-Nogay. To clarify it should say that Turkish is the most populous of the Turkic languages, rather than wide-spread. Sad mouse 02:57, 23 June 2007 (UTC)

Well it definitely does not say widespread, and though widely can mean geographic area, it also (and firstly) means "commonly." Populous doesn't sound very good at all. I'm happy with widely spoken, which doesn't necessarily mean widely distributed. pschemp | talk 04:26, 23 June 2007 (UTC)

The point is, the context doesn't make it clear which is being spoken about (geographic vs population), so while it is "correct" it is not informative. I'll change it. Sad mouse 04:53, 23 June 2007 (UTC)

I think it does because it says widely spoken, not widely distributed. Just change widely to commonly, your changes didn't make it better.pschemp | talk 05:16, 23 June 2007 (UTC)

Commonly is fine. Sad mouse 07:21, 23 June 2007 (UTC)

... but widely was better. --NigelG (or Ndsg) | Talk 09:15, 23 June 2007 (UTC)

In terms of prose, I agree with you. In terms of precision I disagree. Which is more important in an encyclopedia? Sad mouse 12:33, 23 June 2007 (UTC)

Language reform and modern Turkish

Some clumsy editing has caused the map showing literacy rates to disappear. Part of the image tag survives in the footnote. Could someone please fix this? I can't quite work out what happened. --NigelG (or Ndsg) | Talk 09:18, 23 June 2007 (UTC)

Stresses missing in IPA text

The stress marks " ' "s are missing in the example IPA text. Plus some "a"s, needed to be change to "â"s IPA equivalent. Can somebody with IPA knowledge do that? Otherwise I'll have to learn it :) Kerem Özcan 15:31, 23 June 2007 (UTC)

• Stress See my remark in the To-do list at the top of the page.
• as Which ones? --NigelG (or Ndsg) | Talk 17:47, 23 June 2007 (UTC)

Well I read it and I guess there's just one and that's "selam". Together with "olsun", it doesn't sound alright if I read it as "selam olsun". I feel the urge to read it as "sela:m" (Making the a longer I mean. I think that's how you write it in phonetic alphabet) Kerem Özcan 20:36, 23 June 2007 (UTC)

OK, I agree—I've added the lengthening mark to the IPA. --NigelG (or Ndsg) | Talk 21:19, 23 June 2007 (UTC)--NigelG (or Ndsg) | Talk 21:19, 23 June 2007 (UTC)

Featured article protected?

Featured article is NEVER supposed to be protected. 75.3.112.43 22:59, 23 June 2007 (UTC)

Now it is not. Go forth and edit.  :) --Iamunknown 23:02, 23 June 2007 (UTC)

My mistake. I do apologise - Alison ☺ 23:03, 23 June 2007 (UTC)

A note on citations and references

There has been a misunderstanding that simple footnotes, like "Soucek" in the history section, were mistaken to be improper references. I believe the problem arises when one fails to notice that the footnotes in the "Notes" section point to references given in full detail in the following "References" section. I reproduce my correspondence with User:Behmod below:

"Hi, I would like to point out that the "Soucek" citation you've changed into a "fact" template is a proper citation. Please note that all citations given as notes are pointing to sources given in full in the "References" section (which follows the "Notes" section), and all citations are thoroughly checked during the FA process. The footnote "Soucek", in this case, points to "Soucek, Svat (March 2000). A History of Inner Asia. Cambridge University Press. ISBN 978-0521651691." in the references list. Regards, Atilim Gunes Baydin 00:17, 24 June 2007 (UTC)"

Atilim Gunes Baydin 00:24, 24 June 2007 (UTC)

If someone feels that the citation in question is too general (or whatever), they should at the very least explain why. Claiming that a reference to a book, rather than a specific page, "discredits an FA"[5] is very high-handed, and simply replacing the footnote with a doubt-spreading fact tag only merely degrades reference quality. We need to keep in mind that the mere existence of a page reference doesn't make a fact statement more solid. Most printed works have alphabetic indexes and can easily by searched for any number of facts without the need of a blow-by-blow account of the original author's reading schedule.

But I can't stress enough that this is really something that should be discussed in detail on talkpages rather than fought over in article space. Have a little more respect for the readers.

Peter ^Isotalo 13:35, 24 June 2007 (UTC)

There has been a misunderstanding that simple footnotes, like "Soucek" in the history section, were mistaken to be improper references. I did not notice that they are fairly referenced below the notes in the Reference List. My mistake. I do apologize.--behmod talk 01:01, 25 June 2007 (UTC)

IPA in Sample text

I've reverted the /-ek/ endings to the narrower phonemic /-ec/. They've been like that for several months now. --NigelG (or Ndsg) | Talk 16:30, 2 August 2007 (UTC)

BTW How on earth did all the extra diacritics get into the IPA? Who put them there, & what are they meant to represent? Does anyone know? --NigelG (or Ndsg) | Talk 09:29, 4 August 2007 (UTC)

PS I see that some of them are now given in the Sounds section (eg dental /t/ and/d/). Do we need to make the phonemics quite so narrow? What do others think? --NigelG (or Ndsg) | Talk 09:32, 4 August 2007 (UTC)

They were added by the anonymous user 82.238.140.155 during last month (see [6] and [7]). I was not particularly happy about the narrowing, but I don't have the sufficient knowledge of phonetics to decide whether the addition was an improvement or not. I was actually hoping that you would notice the change and make an expert judgment. Atilim Gunes Baydin 12:12, 4 August 2007 (UTC)

OK, I've undone these changes which, though no doubt accurate, added little to the reader's understanding. If our anonymous contributor feels strongly about this, could s/he please discuss the matter here. --NigelG (or Ndsg) | Talk 16:37, 4 August 2007 (UTC)

Welcome back, 82.238.140.155! --NigelG (or Ndsg) | Talk 18:47, 15 September 2007 (UTC)

Pronunciation of <j> etc

A recent edit to the Writing system section, pointing out that <j> is pronounced "zh", has been reverted. Why? This seemed like a useful addition to the section. It would probably be sufficient to give the IPA transcription (ʒ); but it's certainly worth keeping this piece of information, which won't be obvious to all readers.

In fact it might be useful to include a small table giving examples of the various letters described in this section (eg acı, çarşı, müjde, soğuk [already in the text], ağa), together with their pronunciations in IPA. We could at the same time get rid of the English shall and chill. --NigelG (or Ndsg) | Talk 17:49, 27 August 2007 (UTC)

I've reinstated <j>. Does someone else want to compile the table of examples? --NigelG (or Ndsg) | Talk 19:32, 5 September 2007 (UTC)

Y Done with 3 examples. --NigelG (or Ndsg) | Talk 09:39, 6 September 2007 (UTC)

... now 5 (with the addition of circumflex examples).--NigelG (or Ndsg) | Talk 11:29, 13 September 2007 (UTC)

Trivia

I've reverted a trivia section which was copied over from the Turkish Wikipedia article. I feel the reference to our Czechoslovak friends is sufficient in this article.--NigelG (or Ndsg) | Talk 11:28, 13 September 2007 (UTC)