![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 | Archive 2 |
This reads more like an advertisement than an encyclopedia article. Does it really belong here? Should it be deleted? -- April, 7 Mar 2002
Agreed on deleting. -- Rgamble
I disagree. Google found 17,400 hits. The article does need some work though.... -- maveric149
Well, I look at it like this... Effectively the article is about a product that people aren't going to know about unless they see the website. In which case they won't learn anything else here... I'm not sure what else can be said about the site as an article except what it does... and that's an advertisement. Perhaps I'm wrong, but if someone can make a real article out of it, I won't complain. ;) Rgamble ps: Really hope that made sense. Brain's shutting down.
Here is the "article"
The Ethnologue: Languages of the World is a web/print publication of SIL International, which describes itself as an educational charity and a christian volunteer organisation. They claim that it contains statistics for 6,809 languages (2000 edition) which describes the number of speakers, location, dialects, linguistic affiliations, availability of the Bible, etc.
Most likely the copying permissions are not sufficiently liberal to allow use of the data in Wikipedia.
The web version is available at http://www.ethnologue.com/.
This is clearly not an encyclopedia article; the fact that it gets a lot of hits in google doesn't sound compelling to me.
I have two suggestions, to those interested: 1) at whatever time this can be developed into an encyclopedia article -- although what an article about what is effectively a store wouldbe, I don't know -- it can be restored.
2) in the meantime, it certainly does seem appropriate to put the link to this website into other articles, perhaps language, linguistics, missionary organizations, indigenous people.
SIL certainly would deserve its own article, although I am not prepared to write it. Without a doubt, SIL has played a vital role in transforming indigenous languages from oral to written forms, and thus, arguably, plays a role in preserving them. But as a US based missionary group, its influence on indigenous societies and its relationship with other governments is complex; David Stohl has written about this and I know there are others (I think there is a book called God is an American or something like that, check out Cultural Survival and IWGIA. So an article on SIL would be very interesting. In the meantime, the link may be useful to those who study language and linguistics -- why not just put it there? SR
The Ethnologue certainly does deserve an article; it is a vital part of the debate about what to supplant the ISO 639 codes with, and is a very well-known reference among linguists. You almost might as well not have an article about Unicode! As a credible effort to provide an alternative to the Ethnologue, Linguasphere may deserve its own article as well, if anyone can be bothered (I can't at the moment.) Mustafaa 22:20, 8 Apr 2004 (UTC)
It might be an opportune moment to remind folks that Wikipedia:Deletion policy is that deletion is for irredeemable articles; articles which would be useful once repaired should be listed under Wikipedia:Cleanup. In any case, this article is now much improved. I am puzzled that several people have objected to this being an article "about a store" or "about a product". They do sell some stuff on that website (mainly hard copies of stuff available free as soft copy) but most of it (hundreds of megs of data) is freely available for non-commercial use and some of it is freely available for any use.
Their policies give fairly liberal rights to non-commercial users, but commerical users must seek permission for most stuff; this would be incompatible with the GFDL. Still maybe it's worth someone contacting them to ask since they do leave open what permissions might be granted. Also, the SIL codes are available free for everyone, along with their SIL/ISO mapping tables. As the search results for these codes are also made freely available (for display that is, not copying), would it be useful & GFDL acceptable to include links like this "SIL code: ABC" in language articles? (And as an aside, they also offer a few free fonts which might be useful to authors of linguistics articles, plus a huge searchable bibliography which may be useful in researching articles. Securiger 17:08, 17 May 2004 (UTC)
In the current version of the article it states that SIL is a "...service organization which studies lesser-known languages to provide missionaries for their speakers." I believe it is incorrect to say that SIL studies that languages in order to 'provide missionaries'.
SIL educates and provides resources for people (call them linguists or missionaries, or linguistically trained missionaries) who are trained to create writing systems and then translations of biblical texts in languages where there was previously no bible/biblical texts. As a byproduct of this goal of creating bibles in native languages, SIL and its trainees have created the first linguistic documentation of many previously unstudied, or understudied languages. Some of their material (such as the ethnologue) is useful to academic linguists as well.
Point being, I guess, that I want to change this sentence "...study...languages to provide missionaries" to something like: "SIL...studies lesser-known languages primarily to provide the speakers with native language biblical texts."
what do you think?
-JD
Then let us do it... Refdoc 23:49, 26 Jan 2005 (UTC)
looks good, thanks for doing that -JD Hospitalhill
Would it be worth pointing out that the reason why Ethnologue lists more languages than other such catalogues (hence the approving phrase "most comprehensive" in the article) is that they also list dialects without making any distinction? I don't mean that as a criticism of Ethnologue - the distinction between a language and a dialect is complicated (see discussion of dialect) and ignoring it may be the best solution. Nonetheless, the reader should be aware of this. As an example, look at Ethnologue's list of languages of Germany, which includes, in addition to "German, Standard", the following dialects: Bavarian, Allemanic, Kölsch, Mainfränkisch, Luxembourgeois, Pfälzisch, Lower and Upper Saxon, Swabian and Westphalian, which no-one has ever claimed to be separate languages. What is slightly irritating, though (I'm allowed to be POV on the discussion page) is that this is not followed through consistently - the list of languages of Britain does not include such dialects as Yorkshire or Lancashire English. -- Doric Loon 06:23, 7 Apr 2005 (UTC)
I found the paragraph in the article to be very confusing on this point. Also it includes some statments that I think are incorrect. I will list the inaccuracies, and then suggest a revised paragraph...1)ethnologue 15 codes are in lowercase. 2)all ethnologue 15 codes have a corresponding, matching ISO639-3 code. dakota is not an exception. in effect, the fifteen edition of the ethnologue uses the ISO639-3 standard for language identification.
here is the paragraph i propose: "In 1984 the Ethnologue released a three-letter coding system, called SIL code, to identify each language it describes. This set of codes significantly exceeded the scope of previous standards e.g. ISO 639-1 and RFC 3066. In 2002 the Ethnologue was asked to work with the International Standardization Organization (ISO) to integrate its codes into a draft international standard. The Ethnologue now uses this standard, called ISO 639-3." Hospitalhill 03:25, 14 April 2006 (UTC)
227.2 Ethnologue Language Codes Merged with New ISO Code Set
From M. Paul Lewis (Editor_Ethnologue@sil.org) 13 August 2005:
Anyone who has recently visited the Ethnologue site (www.ethnologue.com) will know that the new 15th edition of the Ethnologue has adopted a revised set of 3-letter language codes. The change is described in the "History of the Ethnologue" section of the Overview/Introduction, at:
While SIL has tried to let as many people as possible know about this change, many users are surprised or confused.
Briefly, what has happened is this: The ISO (the International Organization for Standardization, www.iso.org) decided a couple of years ago to expand its existing (1998) standard 3-letter codes for the representation of names of languages (ISO 639-2) beyond the 400 or so languages that it had been covering to include all the known languages in the world. SIL was asked if it would be willing to let the ISO use the Ethnologue codes for this purpose. In order to do that, however, and maintain continuity between the ISO 639-2 and the new expanded ISO 639-3, SIL had to agree to change those codes in its set that didn't line up with the existing ISO 639-2 codes. In addition there were some codes reserved in the new standard for "private use," and if SIL had existing codes within that "codespace" it had to agree to change those as well. In the end, SIL had to change about 400 of its approximately 7300 3-letter language codes to accommodate the merger with the ISO codes.
The new SIL/ISO code set, though still a "draft" (it is expected that it will be fully implemented by the ISO in 2006), was introduced in the 15th edition of Ethnologue, which came online a few months ago. SIL has also agreed to serve as the "administrative authority" for the coding of living languages--supervising the addition of new language codes, merging codes, removing codes, etc. LinguistList will provide the equivalent service for extinct and classical languages that Ethnologue does not include in its inventory.
The new SIL/ISO language codes are represented by lower case letters, and Ethnologue has further adopted the convention of placing them between square brackets (e.g. [abc]). The case distinction can be used to customize the Ethnologue URL for a specific language to return a page with information about that language *either* from the "old" 14th edition or from the "new" 15th edition. The form of the URL is:
where the CCC can be any 3-letter language code. If the three letters are upper case, the data from the 14th edition will be returned with a "banner" indicating that the data has been superseded by data in the 15th edition and indicating any code change that has been made. However, if the CCC is three lower case letters, the system will take that to be a "new" ISO code and return only the 15th edition data.
Full technical details of the new language coding system, with downloadable code tables and other features, can be found at:
The editors of Ethnologue recognize that this transition period is going to require all of the code users to make adjustments and update any databases they have created using the previous coding system, but they believe that in the long run the larger community of users will be better served by having an internationally recognized and agreed-upon system for identifying all of the known languages of the world.
pasted by – ishwar (speak) 08:58, 20 April 2006 (UTC)
Ethnologue Languages of Iran Website: http://www.ethnologue.com/show_country.asp?name=Iran
Here are the emails (I have removed my email address from them), if you have any doubt you can check the authenticity of the emails by emailing the editor Editor_Ethnologue@sil.org). PS: Emails are in reverse order (first email is at the bottom).
+++++++++++++++++++++++++++
From: Ray_Gordon@sil.org on behalf of Editor_Ethnologue@sil.org Sent: 25 May 2006 18:31 To: kiumars
Kiumars, Thank you for bringing this to my attention. I am not able to locate the original source from 1997. You are correct that our published source is too high, although some estimates range up to 40 millions. In line with your calculations we agree that the figure is likely closest to 11,000,000. We will do further research and update our figures for the next edition.
Yours, Ray Gordon Ethnologue, Research
+++++++++++++++++++++++++++++++++++++++++
From: Kiumars Sent: 05/23/2006 06:58 To: Editor_Ethnologue@sil.org
Dear Mr Gordon, Thank you for replying to my email, my problem is that my figures do not tally up with your figures and I am interested to know the sources of your information.
If we add up the population of all the provinces you have mentioned (i.e. East Azerbaijan (3.4 million), West Azerbaijan (2.8 million), Ardebil (1.2 million), and Zanjan (0.94 million), they add up to 8.5 millions, based on the latest Population Estimation by Statistical Centre Of Iran and various US and UN official organisations. http://www.sci.org.ir/english/default.htm
Assuming that all the population of these provinces are Azerbaijani and there is no non-Azerbaijani living in these provinces (which is not the case as we know and other languages like Kurdish and Farsi are spoken widely in these regions), and If we also again very generously assume that 1/4 of Tehran Province's 12 million population (Tehran itself is about 8 million) are Azerbaijanis (that gives us another 3 millions), we end up with a total of 11.5 millions (roughly and very generously estimated).
Your article also mentions "Some Azerbaijani-speaking groups are in Fars Province and other parts of Iran". Lets assume another 0.5 million or even 1 million live in these areas, that makes the Total 12.5 millions.
So comparing this figure (12.5m) with your figure (23.5m) as you can see I am missing 11 million Azerbaijanis in Iran! Where are they then? Where else could they be? Doesn't that worry you?
We are missing half of the Azerbaijanis of Iran if your figures are correct!
Please help me find them.
Kind regards, Kiumars
+++++++++++++++++++++++++++++++++++++++++
From: Ray_Gordon@sil.org On Behalf Of Editor_Ethnologue@sil.org Sent: 22 May 2006 16:43 To: kiumars
Kiumars, The Ethnologue information is of languages not ethnic groups. See other sources for ethnic information.
Best regards, Ray Gordon Ethnologue, Research
+++++++++++++++++++++++++++++++++++++++++
From: Kiumars Sent: 08 May 2006 02:16 To: Editor_Ethnologue@sil.org
Hello, I was searching the web on Iranian ethnic groups for a project I am currently working on and found your site. Some of the information on your site does not tally up with other well known sources may I ask what are the sources you have used? I am referring mainly to the population of the minorities (See CIA factbook on Iran). I would appreciate if you furnish me with contact address or web url to the references.
Regards, Kiumars Kiumars
Kiumars 20:16, 10 July 2006 (UTC)
Mehrdad (17:37, 22 July 2006), your figures do not add up either! Below are two links to an Excel spreadsheet and an html version of it I made from Ethnologue website data and copied the number of people that Ethnologue claim speak a language in a different column (to be able to sum up), and classification of the languages in a different column (to be able to sum up based on classification on sheet 2). As you will see from the spreadsheet the total is 72.4 million against the Iran’s population of 67.5 million according to the current Ethonologue website. So you see that 5 million people are still missing as I said before! How did you make it 69.4 million? Why your sum is 3 million less than what is actually on Ethnologue website?
http://www.zen49535.zen.co.uk/Public/Iran/Wiki/Languages_of_Iran_2html.zip http://www.zen49535.zen.co.uk/Public/Iran/Wiki/Languages_of_Iran_2.zip
Based on the 2005 actual provincial statistics, the population of Iran was 68 million (Which I believe still includes about 2 million Afghan and Iraqi refugees and other migrants).
Let’s look at the figures again, the regions that Asaris are in vast majority are Ardabil, East Azerbaijan and Zanjan.
East Azerbaijan (3.5 millions), Ardebil (1.2 millions), Zanjan (0.97 million), West Azerbaijan (3 million, roughly 50% Azari and %50 Kurds (Approx 1.5 millions Azaris)), Hamadan (1.8 million, 28% Azaris on the border with Zanjan and Markazi (Approx 0.6 million)), Gilan (2.4 million, roughly 10% Azaris (Approx 0.24 million)), Ghazvin (1.2 million, roughly 20% Azaris (Approx 0.24 million)).
Total: 3.5 + 1.2 + 0.97 + 1.5 + 0.6 + 0.24 + 0.24 = 8.25 millions
Mehrdad, you also mentioned “Plus Azerbaijanis are majority in East Azerbaijan West Azerbaijan , Ardabil, Zanjan, Qazvin, Hamadan and are constituting a large part of Gilan, Markazi and Kurdestan provices. Add to this Qashqais , Khorasani Turks, Afshar of Kerman and other scattered Turkic tribes”.
I already catered for East Azerbaijan West Azerbaijan, Ardabil, Zanjan, Qazvin, Hamadan and Gilan above, are you happy with those figures or you disagree with them?
Now let’s look at the remaining regions i.e. Markazi, Kurdestan, Qashqais, Khorasani Turks, Afshar of Kerman, and other scattered Turkic tribes. I need you to tell me what percentage of Markazi (Arak) and Kurdestan are Azaris. As for the rest, let’s assume that Khorasani Turks are Azaris (although Ethnologue does not classify their language as Azeris!), so we add another 400k. I also add another 1.5 million for Quashquis (just for the sake of argument at the moment, despite the fact that Ethnologue says it may be a dialect of Azari and the fact that most Quashguis will kill you if you call them Azari! and also the fact that The Library of Congress Country Studies estimated the total Turkic-speaking population of Fars to be about 500,000 in 1986). Let’s also add 300,000 for Afshars, 5,000 Aynallu, 7,500 Baharlu, 1,000 Moqaddam, 3,500 Nafar 1,000 Pishagchi, 3,000 Qajar, 2,000 Qaragozlu, 130,000 Shahsavani (as per Ethnologue recommendations). Are you happy with this figures? Have I missed anything? As you can see I am using Ethnologue figures to prove that their sums do not add up!
400+1500+300+5+8+1+4+1+3+2+130= 2,354 (2.4 million)
So we add 8.25 + 2.4= 10.65, and so far we have a total of 10.65 millions (I have not deducted anything for the non-Azaris living in Ardabil, Zanjan and East Azerbaijan although I think we should have deducted at least 5% for non-Azaris living there but as we are calculating roughly I am going to please you by not deducting these).
Now we are left with Tehran Province that has 12 million populations (10.5 Urban and 1.5 Rural). I don’t think you find many Azari farmers in the Rural areas of Tehran province but to keep you happy I am going to look at the bigger number i.e. 12 millions. So as you can see even if all the Tehran province population (both Urban & Rural) were Azaris you still would be short of a few million Azaris to make the Ethnologue numbers!
Mehrdad, you also mentioned “most reliable sources agree that up to third of Tehran and most of Karaj population are Azeri ethnics, plus Azerbaijan has been the source of most migrants to all parts of the Iran.
Let us see what we can make of Tehran and Karaj now. The province of Tehran includes 9 districts (Tehran, Shemiranat, Rey, Islamshahr, Shahreyar, Karaj, Savejbolagh, Varamin and Damavand) and 12 cities and 38 towns (major cities are Eslamshahr, Damavand, Firoozkooh, Karaj, Pakdasht, Robot Kari, Ray, Savojbolagh, Shahryar, Shemiranat and Varamin). The great Tehran (that is Tehran and its adjacent town is about 8 million) and Karaj is the second city with almost 1 million population. I think it will be easier and more advantageous for Azari numbers if we look at the big figure rather than trying to looking at these cities one by one, as you know there are not many Azaris in some of these cities (especially in the Rural areas).
Even if we assume that 1/3 of the population of Tehran province are Azaris as you suggested (I do not say that I agree with this assumption, I am just doing a calculation based on what you suggested), that gives us another 4 millions, and adding that to the previous sum it gives us 4 + 10.65 = 14.65 millions and you see that we are still more than 10 millions short to make Ethnologue figures (23.5 millions Azaris and 1.5 millions Qashqa'I = 25 millions).
Mehrdad, you also mentioned “If you travel in Iran you would find Azerbaijanis even remotest areas from Gorgan to Bandar Abbas and Mashhad to Ahvaz”. So just adding up the population of Azerbaijani provinces would not do.
I already catered for Azaris in some of the northern provinces (Gilan 10% and Ghazvin 20% Azaris), aren’t you happy with these percentages? Just tell me what you think they should be and I do the calculation again. But now I need you to tell me how many Azaris live in other parts of Iran. I am sure you are not claiming 10 millions! are you?
I speak Azari and Kurdi and Fasi and I have lived and worked in the west of Iran (From Azerbaijan to Khositan) over 40 years but I have not seen many of these 10 million Azari immigrants you are talking about! There are some that come for short term jobs (mainly Civil Engineering projects) but they go back home after the contract is over! Let’s see how you make up for the missing 10 million! Mehrdad, try again but next time use figures and statistics to support your views not rhetorics.
Mehrdad, you comments are loaded with political rhetorics and clearly state your biased point of view on the subject. Now let us see what we can make sense of some of your rhetorics.
Re: The responsibility of gathering data, specially in non democratic countries like Iran falls on the shoulders of the governments, as simply no one else is allowed to gather unbiased data. However in Iran there has not been any comprehensive census covering the data on ethnic groups.
So, is Iran less democratic than any other country in the region? Is Iran less democratic than Saudi, Kuwait, Egypt, Rep of Azerbaijan, Iraq, Afghanistan, Pakistan, or even Turkey? And are you suggesting that as there are no reliable data (as you put it) then we can make up one ourselves?
By the way if you read Ali’s posts you will see that a snapshot census was carried out in 1991 on ethnic groups. See --Ali doostzadeh 09:33, 12 July 2006 (UTC) where Ali says” Probably the best census I have seen is the following[4]. In this census, 49,588 mothers who gave birth and were issued birth certificates during the Iranian month of Mordad in 1991 were asked about their mother-tongue. In that census Kurds were 10%, Arabs 5%, and Azaris 20.6% of the total which matches calculations based on provincial statistics.”
Re: As for the number of Azerbaijani people in Iran, most reliable sources agree that up to third of Tehran and most of Karaj population are Azeri ethnics.
Who are these reliable sources? When did they carry out a census? How reliable are these people?
By the way, an interesting outcome of my research on this subject has been about the population of the Kurds in Iran. According to Ethnologue figures (see page 2 of the spreadsheet), the population of the Kurds in the Kurdish areas of Iran is 7.6 millions (compared with 8.25 million Azaris calculated above). And if we apply the same percentages for migration we applied for the Azari population then the Kurds are far less represented than the Azaris! (Food for thought!).
Also see
http://en.wikipedia.org/wiki/Talk:Azerbaijani_people
Section "When are we going to sort out this mess? What is the next step?"
--Ali doostzadeh 09:33, 12 July 2006 (UTC) & --Ali doostzadeh 21:47, 12 July 2006 (UTC)
Kiumars 16:12, 26 July 2006 (UTC)
The comments on Scottish Gaelic were removed, I don't know why. Only one dialect is mentioned, the extinct dialect of East Sutherlandshire. The equally extinct Pertshire dialect is also mentioned as the base for Church Gaelic. None of the living dialects (still at least three major dialects) are mentioned. Ethnologue states that the language in spoken in the Hebrides and on Skye. As Skye belong to the Hebridies, Ethnologue has probably confused Na h-Eileanan Siar with the name "The Hebridies". JdeJ 12:29, 26 October 2006 (UTC)
Removing the "citation needed" from the comment about SIL's work in Europe being minimal, based on the lack of data on European languages at
http://www.sil.org/silewp/indexes/countries.asp http://www.ethnologue.com/country_index_bibl.asp
and the minimal amount of references on European languages, compared with other parts of the world, at
http://www.ethnologue.com/country_index_bibl.asp.
Putting such a citation in the text would be rather cumbersome!
-- Drmaik 05:54, 26 October 2006 (UTC)
The sentence The neutrality of Ethnologue as a scientific institution is sometimes disputed, particularly in areas of language classification associated with the Bible and Abrahamic religion. concerns me. I'm guessing it's because some people disagree with the internal classification of Semitic? In that case, why not say so? And is there any evidence that their classification is not neutral or based on the Bible? (I just don't know). Right now it sounds like the Ethnologue proposes we split everything into Semitic, Hamitic and Japhethitic. (It doesn't)
Oct. 8, 2007 The current article says: "The neutrality of Ethnologue as a scientific institution is sometimes disputed, for example in the classification of Semitic languages." This is neither helpful nor constructive, citing no source or detail. Is this about the way Ethnologue speaks of Arabic languages (instead of dialects), or simply about the way the editor follows one particular scheme of classification, or what? There are competing classification schemes for Semitic (three for Ethiopia-Semitic, alone). Being "neutral" in this is tricky. Even specialists in various language fields (e.g. Nilo-Saharan, Austronesian, etc.) can't agree about classifications. If the Ethnologue follows one classification scheme, they are no longer "neutral". Unless somebody can cite something specific, I think this sentence ought to be deleted. Pete unseth 14:00, 8 October 2007 (UTC)
I have added the unreferencedsection-tag to the Linguistics section. Right now it reads:
"Ethnologue sometimes goes against general linguistic community consensus (and the opinion of the majority of the speakers themselves in some cases) as to what constitutes a separate language (as opposed to a dialect). The designation of Irish dialects does not match that generally used by Irish linguists, with an otherwise unattested dialect called Munster-Leinster mentioned by Ethnologue. For Scottish Gaelic only two extinct or moribund dialects are listed; none of the main living dialects are mentioned. Ethnologue also attributes separate language status to "Yinglish", an English vernacular spoken by some Jewish Americans which is to some degree influenced by the Yiddish and Hebrew languages. Some of these classifications do not meet Ethnologue's own professed criteria for classification."
The way it is worded now, it is original research. All claims of this section are masked by weasel words ("general linguistic community consensus", "opinion of the majority of the speakers", "Irish linguists") without giving any citation of any source. The only reference given is to the Ethnologue itself, which hardly serves as a documentation of the criticisms against it. There is no doubt that many people are unhappy with this or that Ethnologue classification, but the claim that these classifications go against linguistic consensus needs to be supported somehow. If there is such a consensus, you will find it documented in some reliable public source, and then you can quote it here. If there is nothing you can quote, then these claims should not be made on Wikipedia. Someone on this talkpage suggested to make this article a place to collect errors in the Ethnologue. This is a good idea - as long as these errors are documented according to Wikipedia's standards, that is with good citations of reliable published sources. Everything else would be original research.
Just to make this clear: I'm not against criticism of the Ethnologue in Wikipedia. The section Statistics is a good example of how this might look like with good supporting evidence. Landroving Linguist ( talk) 13:53, 15 February 2008 (UTC)
I removed the following uncited sentences:
A notable case is the classification of several dialects of Swedish as separate languages with unique language codes. In cases like Scanian, the dialect does not meet the minimum criterion for mutual unintelligibility from Standard Swedish citation needed, though some linguists have nevertheless classified it separately.
As a native speaker of Swedish, I know that such claim would be difficult to find reliable sources for, and back it up. It just isn't true. Kdehl 18:24, 29 April 2007 (UTC)
Therefore I remove the probably obsolete template {{external links}}. Someone cleaned, and the remaining ones are just fine. ... said: Rursus ( bork²) 12:09, 16 March 2009 (UTC)
The follow sentence looks like a heavy understatement: As is inevitable in an enterprise so enormous, the Ethnologue contains some errors Unfortunately, the errors in Ethnologue are so common and so significant that its use as a reliable source could be questioned. The errors are not limited to small languages, even rather large languages are given incorrect descriptions [above comment posted by 62.78.185.120 20:45, 21 Feb 2005 (UTC)]
-Scottish Gaelic: The number of speakers is based on the census in 1971. There is one census every 10th year in the UK; census results from 1981, 1991 and 2001 are availablde and should of course be used. Not having the newest information is not a big deal, but using information that is 35 years old definitely is. Even worse, the only dialect listed for Scottish Gaelic is East Sutherland Gaelic. That dialect is a very marginal and almost extinct one, and all the major dialects are completely ignored.
-Irish Gaelic: Once again, a very old census is cited and the number given is not even close to the real number of Irish speakers. There are three main dialects of Irish: Ulster, Connacht and Munster. Ethnologue suggests two more, Leinster and Donegal. Donegal Irish is the same as Ulster Irish whereas Irish died out in Leinster in the 19th century.
-Welsh: Once more, a 35 year old census.
-Swedish: The spelling of some of the dialects is almost beyond recognition. It claims that Gutniska is a separate language, a claim I haven't seen anywhere else and certainly not in Scandinavian linguistics. The claim that 'proper' Swedish is spoken in Svealand (alone) is unique to Ethnologue and lacks any foundation in reality. Some dialects described are actually the same dialect but with different names.
-Serbo-Croatian: It is said to be a language of Yugoslavia. No country has been called Yugoslavia for some years now. Bosnian, Croatian and Serbian all link to the Serbo-Croatian page and that can be defended on a linguistic basis. What cannot be defended is the information that this language is spoken in "Yugoslavia and Macedonia". What happened to Croatia and Bosnia? The dialects are mixed up. First they are called CHAKAVIAN, KAJKAVIAN, STOKAVIAN, then they are called KAYKAVSKI, CHAKAVSKI, SHTOKAVSKI. Someone not knowing the language might think these are different dialects. The Macedonian minority in Bosnia that Ethnologue has discovered would be unknown both to Bosnians and to Macedonians.
I picked these languages because they are the ones I know best. It would be very strange indeed if only the language I speak are exactly the ones that Ethnologue makes serious mistakes in describing. I would expect to find about the same kinds of errors in other languages as well, although I'm not capable to detect mistakes in languages such as (random pick) Quechua, Romanian, Persian or Japanese.
Ethnologue is a bit like Wikipedia in some sense - a vast ressource which accumulates information from other places and re-publishes it. Some of the information is extremely well researched, some is based on only a few sources. AFAIK once one goes into the depths of the database thse differences in the evidence base become more obvious. And i think this is fair enough.
Refdoc 00:35, 24 Feb 2005 (UTC)
I agree with Refdoc's assessment - except for one crucial point: the Ethnologue only corrects mistakes at multi-year intervals when new editions come out, and is much less responsive. Why not add a new section to this article - Ethnologue errors? With suitable sourcing, it could become a useful resource not only for linguists online but for the Ethnologue editors themselves. - Mustafaa 23:43, 24 Feb 2005 (UTC)
Well, no, if you follow the above links you will see that the words are being used distinctly. -- Doric Loon 05:07, 4 April 2006 (UTC)
I do not know if they are errors, but there are things which annoy me with Ethnologue (otherwise often an useful source, in addition to all the others that is). On one side, they regularly extrapolate the "population" for the main language of any country by substracting from the census the counts for the lesser-used languages: of course this is completely missing out those people which do have two mother languages, for example when both parents are not speaking the same language to their children. Another thing which is surprising are the statistics about immigrants: for example, www.ethnologue.com/show_language.asp?code=fra lists various African countries with counts that seems the ones for the French or Belgian "cooperants" there, but we do not see entries for Germany or Spain, despite the numerous French local population... For the UK, the number of 14,000 could be some estimate for the French-speaking census of the Channel Islands, but it is clearly missing the immigration toward the City (current estimations have 250,000 French residents in the UK, according to an official agency)... I guess similar remarks can be done for the major languages, starting with English, supposedly without native speakers in neither Belgium nor France nor Spain (but with several at St.Pierre and Miquelon or Andorra). -- Antoinel 14:25, 7 April 2006 (UTC)
The article faults previous editions of the Ethnologue for their classification of Khoisan languages, but does not specify what the specific problems are. Since there is a new edition of the Ethnologue now available, I will delete this section. If the new edition is seen to continue the previous fault, then somebody can feel free to insert the complaint again, but they should include some specific details and cite references. Pete unseth ( talk) 12:35, 30 July 2009 (UTC)
The first paragraph says "The Ethnologue contains statistics for 7,358 languages in the 16th edition". I'm wondering what this 7,358 number comes from. In the Introduction on www.ethnologue.com, it says "we tally 6,909 languages which are known to have living speakers who learned them by transmission from parent to child as the primary language of day-to-day communication", i.e. "a person’s “first language” or “mother tongue”". And later in the intro, "the 7,413 languages listed in this edition. These break down as 6,909 living languages, 55 macrolanguages, 28 languages used only as a second language, and 421 recently extinct languages."
If someone arrived at 7358 by using the above figures and excluding macrolanguages (6909 + 28 + 421), the article should probably say so. Since the 7358 is at odds with the most prominent figures put forth by the Ethnologue itself, and directly at odds with the "7413 [unqualified] languages" figure, it is confusing. To my mind, 7413 would be a more intuitive number to use than 7358; but again if someone thinks 7358 is the more useful number to display, they should explain what it refers to.
There is also the issue of comparing apples to oranges between different editions. The article says "up from 6,912 in the 15th edition", but that number refers to living languages. (The total number of languages in the 15th ed. is listed as 7,299 in the intro.)
The word "statistics" could probably be improved on as well. There is at least one number for most languages, but typically there is much more non-numeric information, such as alternate language names and dialect names, location, language family relationships, and so on.
I would propose we replace "The Ethnologue contains statistics for 7,358 languages in the 16th edition" with "The Ethnologue contains listings for 6,909 living languages in the 16th edition". That would clear up the language count confusion in principle, as well as the problem of misleading comparison between editions, and would address the use of the word "statistics". Huttarl ( talk) 20:34, 20 August 2010 (UTC)
According to Ethnologue (16th edition) itself [2]: "The Ethnologue database has been an active research project for more than fifty years. It is probably the most comprehensive listing of information about the currently known languages of the world. Thousands of linguists and other researchers all over the world rely on and have contributed to the Ethnologue database." No more emphasis on missionaries. -- Taivo ( talk) 09:22, 18 October 2012 (UTC)
There has been discussion as to whether Ethnologue is a survey of the world's languages. Not sure why this is difficult. The first sentence says, "The purpose of the Ethnologue is to provide a comprehensive listing of the known living languages of the world" (16th ed p. 7 and 15th ed. p. 7). Earlier editions said, "The Ethnologue tries to bring together the best known information available on the languages of the world" (14th ed. p. vii, 13th ed. p. vii, 12th ed. p. vi). I hope this is adequate to clarify that its purpose is to provide a survey of the world's languages. Pete unseth ( talk) 20:06, 18 October 2012 (UTC)
I am not inclined to get into the political underbelly of Ethnologue, though their historical origins and mission are important to describe. Most all of us who are language scholars agree that religion, among other cultural revolutions, have brought about the destruction of peoples and their languages altogether. Still, the tone here could use a little refining. Also, when I have contacted Ethnologue and complained about, for instance, a language being more endangered than it is represented, Ethnologue editors modify the data and are courteous about it. I've made a few modest edits here. As for SIL in this article, they don't just translate the bible, by the way. They also publish slender primers and chapbooks that describe a given culture's oral history. I am not an SIL affiliate, but I certainly know linguists who are.
KSRolph ( talk) 16:01, 22 March 2013 (UTC)
The Ethnologue was online prior to 2009, I think the 14th version is archived online still. So a change would be needed in the 1st sentence or two of this article. -N — Preceding unsigned comment added by 61.7.176.174 ( talk) 03:23, 18 September 2013 (UTC)
The 13th is archived , but that's not relevant . — kwami ( talk) 19:38, 19 September 2013 (UTC)
This talk page is over 50k and it has comments older than most articles. I'd like to use lowercase Sigmabot to start automatic archival. Chris Troutman ( talk) 01:32, 23 July 2014 (UTC)
In the Overview section, SIL International is described as "(formerly known as the Summer Institute of Linguistics, originally Wycliffe Bible Translators)". This is not quite right. The Summer Institute of Linguistics and Wycliffe Bible Translators were legally incorporated as separate organizations in 1942. Although they work closely together, and have overlapping membership, they were never the same organization, and both exist as legally-separate organizations at this time. Therefore, I think the wording should be something like "(formerly known as the Summer Institute of Linguistics, which is associated with Wycliffe Bible Translators)". Perhaps better, simply omit the reference to Wycliffe entirely, because mentioning that name doesn't add anything in context beyond what is already said in describing SIL, and people can always find out about the relationship of the two organizations by following the link to the article on SIL. As a member of both organizations, and part of the Ethnologue staff, it isn't appropriate for me to make the change, but I submit this for consideration by other editors. AlbertBickford ( talk) 15:24, 27 November 2014 (UTC)
The EGIDS scale, used to rank language vitality in Ethnologue, was revised in the 17th edition to make it easier to apply to sign languages. The revisions are documented in this article: http://www.tandfonline.com/doi/abs/10.1080/01434632.2014.966827. Since I'm the lead author, I won't add it to the WP article on Ethnologue, but if someone else agrees that it is worth citing, the place I'd suggest inserting the reference would be right after the reference to the original Lewis and Simons article on EGIDS. AlbertBickford ( talk) 21:31, 2 December 2014 (UTC)
I adjusted a sentence in the overview to better express the dependency of recent editions of Ethnologue on ISO 639-3. It now reads "The 15th edition of the Ethnologue was the first edition to use this standard, called ISO 639-3, and since then Ethnologue relies on the standard to determine what is a language." Other editors should check this, since I'm a research editor for Ethnologue, to make sure they agree this is from a NPV. AlbertBickford ( talk) 19:44, 22 January 2015 (UTC)
![]() | This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page. |
Archive 1 | Archive 2 |
This reads more like an advertisement than an encyclopedia article. Does it really belong here? Should it be deleted? -- April, 7 Mar 2002
Agreed on deleting. -- Rgamble
I disagree. Google found 17,400 hits. The article does need some work though.... -- maveric149
Well, I look at it like this... Effectively the article is about a product that people aren't going to know about unless they see the website. In which case they won't learn anything else here... I'm not sure what else can be said about the site as an article except what it does... and that's an advertisement. Perhaps I'm wrong, but if someone can make a real article out of it, I won't complain. ;) Rgamble ps: Really hope that made sense. Brain's shutting down.
Here is the "article"
The Ethnologue: Languages of the World is a web/print publication of SIL International, which describes itself as an educational charity and a christian volunteer organisation. They claim that it contains statistics for 6,809 languages (2000 edition) which describes the number of speakers, location, dialects, linguistic affiliations, availability of the Bible, etc.
Most likely the copying permissions are not sufficiently liberal to allow use of the data in Wikipedia.
The web version is available at http://www.ethnologue.com/.
This is clearly not an encyclopedia article; the fact that it gets a lot of hits in google doesn't sound compelling to me.
I have two suggestions, to those interested: 1) at whatever time this can be developed into an encyclopedia article -- although what an article about what is effectively a store wouldbe, I don't know -- it can be restored.
2) in the meantime, it certainly does seem appropriate to put the link to this website into other articles, perhaps language, linguistics, missionary organizations, indigenous people.
SIL certainly would deserve its own article, although I am not prepared to write it. Without a doubt, SIL has played a vital role in transforming indigenous languages from oral to written forms, and thus, arguably, plays a role in preserving them. But as a US based missionary group, its influence on indigenous societies and its relationship with other governments is complex; David Stohl has written about this and I know there are others (I think there is a book called God is an American or something like that, check out Cultural Survival and IWGIA. So an article on SIL would be very interesting. In the meantime, the link may be useful to those who study language and linguistics -- why not just put it there? SR
The Ethnologue certainly does deserve an article; it is a vital part of the debate about what to supplant the ISO 639 codes with, and is a very well-known reference among linguists. You almost might as well not have an article about Unicode! As a credible effort to provide an alternative to the Ethnologue, Linguasphere may deserve its own article as well, if anyone can be bothered (I can't at the moment.) Mustafaa 22:20, 8 Apr 2004 (UTC)
It might be an opportune moment to remind folks that Wikipedia:Deletion policy is that deletion is for irredeemable articles; articles which would be useful once repaired should be listed under Wikipedia:Cleanup. In any case, this article is now much improved. I am puzzled that several people have objected to this being an article "about a store" or "about a product". They do sell some stuff on that website (mainly hard copies of stuff available free as soft copy) but most of it (hundreds of megs of data) is freely available for non-commercial use and some of it is freely available for any use.
Their policies give fairly liberal rights to non-commercial users, but commerical users must seek permission for most stuff; this would be incompatible with the GFDL. Still maybe it's worth someone contacting them to ask since they do leave open what permissions might be granted. Also, the SIL codes are available free for everyone, along with their SIL/ISO mapping tables. As the search results for these codes are also made freely available (for display that is, not copying), would it be useful & GFDL acceptable to include links like this "SIL code: ABC" in language articles? (And as an aside, they also offer a few free fonts which might be useful to authors of linguistics articles, plus a huge searchable bibliography which may be useful in researching articles. Securiger 17:08, 17 May 2004 (UTC)
In the current version of the article it states that SIL is a "...service organization which studies lesser-known languages to provide missionaries for their speakers." I believe it is incorrect to say that SIL studies that languages in order to 'provide missionaries'.
SIL educates and provides resources for people (call them linguists or missionaries, or linguistically trained missionaries) who are trained to create writing systems and then translations of biblical texts in languages where there was previously no bible/biblical texts. As a byproduct of this goal of creating bibles in native languages, SIL and its trainees have created the first linguistic documentation of many previously unstudied, or understudied languages. Some of their material (such as the ethnologue) is useful to academic linguists as well.
Point being, I guess, that I want to change this sentence "...study...languages to provide missionaries" to something like: "SIL...studies lesser-known languages primarily to provide the speakers with native language biblical texts."
what do you think?
-JD
Then let us do it... Refdoc 23:49, 26 Jan 2005 (UTC)
looks good, thanks for doing that -JD Hospitalhill
Would it be worth pointing out that the reason why Ethnologue lists more languages than other such catalogues (hence the approving phrase "most comprehensive" in the article) is that they also list dialects without making any distinction? I don't mean that as a criticism of Ethnologue - the distinction between a language and a dialect is complicated (see discussion of dialect) and ignoring it may be the best solution. Nonetheless, the reader should be aware of this. As an example, look at Ethnologue's list of languages of Germany, which includes, in addition to "German, Standard", the following dialects: Bavarian, Allemanic, Kölsch, Mainfränkisch, Luxembourgeois, Pfälzisch, Lower and Upper Saxon, Swabian and Westphalian, which no-one has ever claimed to be separate languages. What is slightly irritating, though (I'm allowed to be POV on the discussion page) is that this is not followed through consistently - the list of languages of Britain does not include such dialects as Yorkshire or Lancashire English. -- Doric Loon 06:23, 7 Apr 2005 (UTC)
I found the paragraph in the article to be very confusing on this point. Also it includes some statments that I think are incorrect. I will list the inaccuracies, and then suggest a revised paragraph...1)ethnologue 15 codes are in lowercase. 2)all ethnologue 15 codes have a corresponding, matching ISO639-3 code. dakota is not an exception. in effect, the fifteen edition of the ethnologue uses the ISO639-3 standard for language identification.
here is the paragraph i propose: "In 1984 the Ethnologue released a three-letter coding system, called SIL code, to identify each language it describes. This set of codes significantly exceeded the scope of previous standards e.g. ISO 639-1 and RFC 3066. In 2002 the Ethnologue was asked to work with the International Standardization Organization (ISO) to integrate its codes into a draft international standard. The Ethnologue now uses this standard, called ISO 639-3." Hospitalhill 03:25, 14 April 2006 (UTC)
227.2 Ethnologue Language Codes Merged with New ISO Code Set
From M. Paul Lewis (Editor_Ethnologue@sil.org) 13 August 2005:
Anyone who has recently visited the Ethnologue site (www.ethnologue.com) will know that the new 15th edition of the Ethnologue has adopted a revised set of 3-letter language codes. The change is described in the "History of the Ethnologue" section of the Overview/Introduction, at:
While SIL has tried to let as many people as possible know about this change, many users are surprised or confused.
Briefly, what has happened is this: The ISO (the International Organization for Standardization, www.iso.org) decided a couple of years ago to expand its existing (1998) standard 3-letter codes for the representation of names of languages (ISO 639-2) beyond the 400 or so languages that it had been covering to include all the known languages in the world. SIL was asked if it would be willing to let the ISO use the Ethnologue codes for this purpose. In order to do that, however, and maintain continuity between the ISO 639-2 and the new expanded ISO 639-3, SIL had to agree to change those codes in its set that didn't line up with the existing ISO 639-2 codes. In addition there were some codes reserved in the new standard for "private use," and if SIL had existing codes within that "codespace" it had to agree to change those as well. In the end, SIL had to change about 400 of its approximately 7300 3-letter language codes to accommodate the merger with the ISO codes.
The new SIL/ISO code set, though still a "draft" (it is expected that it will be fully implemented by the ISO in 2006), was introduced in the 15th edition of Ethnologue, which came online a few months ago. SIL has also agreed to serve as the "administrative authority" for the coding of living languages--supervising the addition of new language codes, merging codes, removing codes, etc. LinguistList will provide the equivalent service for extinct and classical languages that Ethnologue does not include in its inventory.
The new SIL/ISO language codes are represented by lower case letters, and Ethnologue has further adopted the convention of placing them between square brackets (e.g. [abc]). The case distinction can be used to customize the Ethnologue URL for a specific language to return a page with information about that language *either* from the "old" 14th edition or from the "new" 15th edition. The form of the URL is:
where the CCC can be any 3-letter language code. If the three letters are upper case, the data from the 14th edition will be returned with a "banner" indicating that the data has been superseded by data in the 15th edition and indicating any code change that has been made. However, if the CCC is three lower case letters, the system will take that to be a "new" ISO code and return only the 15th edition data.
Full technical details of the new language coding system, with downloadable code tables and other features, can be found at:
The editors of Ethnologue recognize that this transition period is going to require all of the code users to make adjustments and update any databases they have created using the previous coding system, but they believe that in the long run the larger community of users will be better served by having an internationally recognized and agreed-upon system for identifying all of the known languages of the world.
pasted by – ishwar (speak) 08:58, 20 April 2006 (UTC)
Ethnologue Languages of Iran Website: http://www.ethnologue.com/show_country.asp?name=Iran
Here are the emails (I have removed my email address from them), if you have any doubt you can check the authenticity of the emails by emailing the editor Editor_Ethnologue@sil.org). PS: Emails are in reverse order (first email is at the bottom).
+++++++++++++++++++++++++++
From: Ray_Gordon@sil.org on behalf of Editor_Ethnologue@sil.org Sent: 25 May 2006 18:31 To: kiumars
Kiumars, Thank you for bringing this to my attention. I am not able to locate the original source from 1997. You are correct that our published source is too high, although some estimates range up to 40 millions. In line with your calculations we agree that the figure is likely closest to 11,000,000. We will do further research and update our figures for the next edition.
Yours, Ray Gordon Ethnologue, Research
+++++++++++++++++++++++++++++++++++++++++
From: Kiumars Sent: 05/23/2006 06:58 To: Editor_Ethnologue@sil.org
Dear Mr Gordon, Thank you for replying to my email, my problem is that my figures do not tally up with your figures and I am interested to know the sources of your information.
If we add up the population of all the provinces you have mentioned (i.e. East Azerbaijan (3.4 million), West Azerbaijan (2.8 million), Ardebil (1.2 million), and Zanjan (0.94 million), they add up to 8.5 millions, based on the latest Population Estimation by Statistical Centre Of Iran and various US and UN official organisations. http://www.sci.org.ir/english/default.htm
Assuming that all the population of these provinces are Azerbaijani and there is no non-Azerbaijani living in these provinces (which is not the case as we know and other languages like Kurdish and Farsi are spoken widely in these regions), and If we also again very generously assume that 1/4 of Tehran Province's 12 million population (Tehran itself is about 8 million) are Azerbaijanis (that gives us another 3 millions), we end up with a total of 11.5 millions (roughly and very generously estimated).
Your article also mentions "Some Azerbaijani-speaking groups are in Fars Province and other parts of Iran". Lets assume another 0.5 million or even 1 million live in these areas, that makes the Total 12.5 millions.
So comparing this figure (12.5m) with your figure (23.5m) as you can see I am missing 11 million Azerbaijanis in Iran! Where are they then? Where else could they be? Doesn't that worry you?
We are missing half of the Azerbaijanis of Iran if your figures are correct!
Please help me find them.
Kind regards, Kiumars
+++++++++++++++++++++++++++++++++++++++++
From: Ray_Gordon@sil.org On Behalf Of Editor_Ethnologue@sil.org Sent: 22 May 2006 16:43 To: kiumars
Kiumars, The Ethnologue information is of languages not ethnic groups. See other sources for ethnic information.
Best regards, Ray Gordon Ethnologue, Research
+++++++++++++++++++++++++++++++++++++++++
From: Kiumars Sent: 08 May 2006 02:16 To: Editor_Ethnologue@sil.org
Hello, I was searching the web on Iranian ethnic groups for a project I am currently working on and found your site. Some of the information on your site does not tally up with other well known sources may I ask what are the sources you have used? I am referring mainly to the population of the minorities (See CIA factbook on Iran). I would appreciate if you furnish me with contact address or web url to the references.
Regards, Kiumars Kiumars
Kiumars 20:16, 10 July 2006 (UTC)
Mehrdad (17:37, 22 July 2006), your figures do not add up either! Below are two links to an Excel spreadsheet and an html version of it I made from Ethnologue website data and copied the number of people that Ethnologue claim speak a language in a different column (to be able to sum up), and classification of the languages in a different column (to be able to sum up based on classification on sheet 2). As you will see from the spreadsheet the total is 72.4 million against the Iran’s population of 67.5 million according to the current Ethonologue website. So you see that 5 million people are still missing as I said before! How did you make it 69.4 million? Why your sum is 3 million less than what is actually on Ethnologue website?
http://www.zen49535.zen.co.uk/Public/Iran/Wiki/Languages_of_Iran_2html.zip http://www.zen49535.zen.co.uk/Public/Iran/Wiki/Languages_of_Iran_2.zip
Based on the 2005 actual provincial statistics, the population of Iran was 68 million (Which I believe still includes about 2 million Afghan and Iraqi refugees and other migrants).
Let’s look at the figures again, the regions that Asaris are in vast majority are Ardabil, East Azerbaijan and Zanjan.
East Azerbaijan (3.5 millions), Ardebil (1.2 millions), Zanjan (0.97 million), West Azerbaijan (3 million, roughly 50% Azari and %50 Kurds (Approx 1.5 millions Azaris)), Hamadan (1.8 million, 28% Azaris on the border with Zanjan and Markazi (Approx 0.6 million)), Gilan (2.4 million, roughly 10% Azaris (Approx 0.24 million)), Ghazvin (1.2 million, roughly 20% Azaris (Approx 0.24 million)).
Total: 3.5 + 1.2 + 0.97 + 1.5 + 0.6 + 0.24 + 0.24 = 8.25 millions
Mehrdad, you also mentioned “Plus Azerbaijanis are majority in East Azerbaijan West Azerbaijan , Ardabil, Zanjan, Qazvin, Hamadan and are constituting a large part of Gilan, Markazi and Kurdestan provices. Add to this Qashqais , Khorasani Turks, Afshar of Kerman and other scattered Turkic tribes”.
I already catered for East Azerbaijan West Azerbaijan, Ardabil, Zanjan, Qazvin, Hamadan and Gilan above, are you happy with those figures or you disagree with them?
Now let’s look at the remaining regions i.e. Markazi, Kurdestan, Qashqais, Khorasani Turks, Afshar of Kerman, and other scattered Turkic tribes. I need you to tell me what percentage of Markazi (Arak) and Kurdestan are Azaris. As for the rest, let’s assume that Khorasani Turks are Azaris (although Ethnologue does not classify their language as Azeris!), so we add another 400k. I also add another 1.5 million for Quashquis (just for the sake of argument at the moment, despite the fact that Ethnologue says it may be a dialect of Azari and the fact that most Quashguis will kill you if you call them Azari! and also the fact that The Library of Congress Country Studies estimated the total Turkic-speaking population of Fars to be about 500,000 in 1986). Let’s also add 300,000 for Afshars, 5,000 Aynallu, 7,500 Baharlu, 1,000 Moqaddam, 3,500 Nafar 1,000 Pishagchi, 3,000 Qajar, 2,000 Qaragozlu, 130,000 Shahsavani (as per Ethnologue recommendations). Are you happy with this figures? Have I missed anything? As you can see I am using Ethnologue figures to prove that their sums do not add up!
400+1500+300+5+8+1+4+1+3+2+130= 2,354 (2.4 million)
So we add 8.25 + 2.4= 10.65, and so far we have a total of 10.65 millions (I have not deducted anything for the non-Azaris living in Ardabil, Zanjan and East Azerbaijan although I think we should have deducted at least 5% for non-Azaris living there but as we are calculating roughly I am going to please you by not deducting these).
Now we are left with Tehran Province that has 12 million populations (10.5 Urban and 1.5 Rural). I don’t think you find many Azari farmers in the Rural areas of Tehran province but to keep you happy I am going to look at the bigger number i.e. 12 millions. So as you can see even if all the Tehran province population (both Urban & Rural) were Azaris you still would be short of a few million Azaris to make the Ethnologue numbers!
Mehrdad, you also mentioned “most reliable sources agree that up to third of Tehran and most of Karaj population are Azeri ethnics, plus Azerbaijan has been the source of most migrants to all parts of the Iran.
Let us see what we can make of Tehran and Karaj now. The province of Tehran includes 9 districts (Tehran, Shemiranat, Rey, Islamshahr, Shahreyar, Karaj, Savejbolagh, Varamin and Damavand) and 12 cities and 38 towns (major cities are Eslamshahr, Damavand, Firoozkooh, Karaj, Pakdasht, Robot Kari, Ray, Savojbolagh, Shahryar, Shemiranat and Varamin). The great Tehran (that is Tehran and its adjacent town is about 8 million) and Karaj is the second city with almost 1 million population. I think it will be easier and more advantageous for Azari numbers if we look at the big figure rather than trying to looking at these cities one by one, as you know there are not many Azaris in some of these cities (especially in the Rural areas).
Even if we assume that 1/3 of the population of Tehran province are Azaris as you suggested (I do not say that I agree with this assumption, I am just doing a calculation based on what you suggested), that gives us another 4 millions, and adding that to the previous sum it gives us 4 + 10.65 = 14.65 millions and you see that we are still more than 10 millions short to make Ethnologue figures (23.5 millions Azaris and 1.5 millions Qashqa'I = 25 millions).
Mehrdad, you also mentioned “If you travel in Iran you would find Azerbaijanis even remotest areas from Gorgan to Bandar Abbas and Mashhad to Ahvaz”. So just adding up the population of Azerbaijani provinces would not do.
I already catered for Azaris in some of the northern provinces (Gilan 10% and Ghazvin 20% Azaris), aren’t you happy with these percentages? Just tell me what you think they should be and I do the calculation again. But now I need you to tell me how many Azaris live in other parts of Iran. I am sure you are not claiming 10 millions! are you?
I speak Azari and Kurdi and Fasi and I have lived and worked in the west of Iran (From Azerbaijan to Khositan) over 40 years but I have not seen many of these 10 million Azari immigrants you are talking about! There are some that come for short term jobs (mainly Civil Engineering projects) but they go back home after the contract is over! Let’s see how you make up for the missing 10 million! Mehrdad, try again but next time use figures and statistics to support your views not rhetorics.
Mehrdad, you comments are loaded with political rhetorics and clearly state your biased point of view on the subject. Now let us see what we can make sense of some of your rhetorics.
Re: The responsibility of gathering data, specially in non democratic countries like Iran falls on the shoulders of the governments, as simply no one else is allowed to gather unbiased data. However in Iran there has not been any comprehensive census covering the data on ethnic groups.
So, is Iran less democratic than any other country in the region? Is Iran less democratic than Saudi, Kuwait, Egypt, Rep of Azerbaijan, Iraq, Afghanistan, Pakistan, or even Turkey? And are you suggesting that as there are no reliable data (as you put it) then we can make up one ourselves?
By the way if you read Ali’s posts you will see that a snapshot census was carried out in 1991 on ethnic groups. See --Ali doostzadeh 09:33, 12 July 2006 (UTC) where Ali says” Probably the best census I have seen is the following[4]. In this census, 49,588 mothers who gave birth and were issued birth certificates during the Iranian month of Mordad in 1991 were asked about their mother-tongue. In that census Kurds were 10%, Arabs 5%, and Azaris 20.6% of the total which matches calculations based on provincial statistics.”
Re: As for the number of Azerbaijani people in Iran, most reliable sources agree that up to third of Tehran and most of Karaj population are Azeri ethnics.
Who are these reliable sources? When did they carry out a census? How reliable are these people?
By the way, an interesting outcome of my research on this subject has been about the population of the Kurds in Iran. According to Ethnologue figures (see page 2 of the spreadsheet), the population of the Kurds in the Kurdish areas of Iran is 7.6 millions (compared with 8.25 million Azaris calculated above). And if we apply the same percentages for migration we applied for the Azari population then the Kurds are far less represented than the Azaris! (Food for thought!).
Also see
http://en.wikipedia.org/wiki/Talk:Azerbaijani_people
Section "When are we going to sort out this mess? What is the next step?"
--Ali doostzadeh 09:33, 12 July 2006 (UTC) & --Ali doostzadeh 21:47, 12 July 2006 (UTC)
Kiumars 16:12, 26 July 2006 (UTC)
The comments on Scottish Gaelic were removed, I don't know why. Only one dialect is mentioned, the extinct dialect of East Sutherlandshire. The equally extinct Pertshire dialect is also mentioned as the base for Church Gaelic. None of the living dialects (still at least three major dialects) are mentioned. Ethnologue states that the language in spoken in the Hebrides and on Skye. As Skye belong to the Hebridies, Ethnologue has probably confused Na h-Eileanan Siar with the name "The Hebridies". JdeJ 12:29, 26 October 2006 (UTC)
Removing the "citation needed" from the comment about SIL's work in Europe being minimal, based on the lack of data on European languages at
http://www.sil.org/silewp/indexes/countries.asp http://www.ethnologue.com/country_index_bibl.asp
and the minimal amount of references on European languages, compared with other parts of the world, at
http://www.ethnologue.com/country_index_bibl.asp.
Putting such a citation in the text would be rather cumbersome!
-- Drmaik 05:54, 26 October 2006 (UTC)
The sentence The neutrality of Ethnologue as a scientific institution is sometimes disputed, particularly in areas of language classification associated with the Bible and Abrahamic religion. concerns me. I'm guessing it's because some people disagree with the internal classification of Semitic? In that case, why not say so? And is there any evidence that their classification is not neutral or based on the Bible? (I just don't know). Right now it sounds like the Ethnologue proposes we split everything into Semitic, Hamitic and Japhethitic. (It doesn't)
Oct. 8, 2007 The current article says: "The neutrality of Ethnologue as a scientific institution is sometimes disputed, for example in the classification of Semitic languages." This is neither helpful nor constructive, citing no source or detail. Is this about the way Ethnologue speaks of Arabic languages (instead of dialects), or simply about the way the editor follows one particular scheme of classification, or what? There are competing classification schemes for Semitic (three for Ethiopia-Semitic, alone). Being "neutral" in this is tricky. Even specialists in various language fields (e.g. Nilo-Saharan, Austronesian, etc.) can't agree about classifications. If the Ethnologue follows one classification scheme, they are no longer "neutral". Unless somebody can cite something specific, I think this sentence ought to be deleted. Pete unseth 14:00, 8 October 2007 (UTC)
I have added the unreferencedsection-tag to the Linguistics section. Right now it reads:
"Ethnologue sometimes goes against general linguistic community consensus (and the opinion of the majority of the speakers themselves in some cases) as to what constitutes a separate language (as opposed to a dialect). The designation of Irish dialects does not match that generally used by Irish linguists, with an otherwise unattested dialect called Munster-Leinster mentioned by Ethnologue. For Scottish Gaelic only two extinct or moribund dialects are listed; none of the main living dialects are mentioned. Ethnologue also attributes separate language status to "Yinglish", an English vernacular spoken by some Jewish Americans which is to some degree influenced by the Yiddish and Hebrew languages. Some of these classifications do not meet Ethnologue's own professed criteria for classification."
The way it is worded now, it is original research. All claims of this section are masked by weasel words ("general linguistic community consensus", "opinion of the majority of the speakers", "Irish linguists") without giving any citation of any source. The only reference given is to the Ethnologue itself, which hardly serves as a documentation of the criticisms against it. There is no doubt that many people are unhappy with this or that Ethnologue classification, but the claim that these classifications go against linguistic consensus needs to be supported somehow. If there is such a consensus, you will find it documented in some reliable public source, and then you can quote it here. If there is nothing you can quote, then these claims should not be made on Wikipedia. Someone on this talkpage suggested to make this article a place to collect errors in the Ethnologue. This is a good idea - as long as these errors are documented according to Wikipedia's standards, that is with good citations of reliable published sources. Everything else would be original research.
Just to make this clear: I'm not against criticism of the Ethnologue in Wikipedia. The section Statistics is a good example of how this might look like with good supporting evidence. Landroving Linguist ( talk) 13:53, 15 February 2008 (UTC)
I removed the following uncited sentences:
A notable case is the classification of several dialects of Swedish as separate languages with unique language codes. In cases like Scanian, the dialect does not meet the minimum criterion for mutual unintelligibility from Standard Swedish citation needed, though some linguists have nevertheless classified it separately.
As a native speaker of Swedish, I know that such claim would be difficult to find reliable sources for, and back it up. It just isn't true. Kdehl 18:24, 29 April 2007 (UTC)
Therefore I remove the probably obsolete template {{external links}}. Someone cleaned, and the remaining ones are just fine. ... said: Rursus ( bork²) 12:09, 16 March 2009 (UTC)
The follow sentence looks like a heavy understatement: As is inevitable in an enterprise so enormous, the Ethnologue contains some errors Unfortunately, the errors in Ethnologue are so common and so significant that its use as a reliable source could be questioned. The errors are not limited to small languages, even rather large languages are given incorrect descriptions [above comment posted by 62.78.185.120 20:45, 21 Feb 2005 (UTC)]
-Scottish Gaelic: The number of speakers is based on the census in 1971. There is one census every 10th year in the UK; census results from 1981, 1991 and 2001 are availablde and should of course be used. Not having the newest information is not a big deal, but using information that is 35 years old definitely is. Even worse, the only dialect listed for Scottish Gaelic is East Sutherland Gaelic. That dialect is a very marginal and almost extinct one, and all the major dialects are completely ignored.
-Irish Gaelic: Once again, a very old census is cited and the number given is not even close to the real number of Irish speakers. There are three main dialects of Irish: Ulster, Connacht and Munster. Ethnologue suggests two more, Leinster and Donegal. Donegal Irish is the same as Ulster Irish whereas Irish died out in Leinster in the 19th century.
-Welsh: Once more, a 35 year old census.
-Swedish: The spelling of some of the dialects is almost beyond recognition. It claims that Gutniska is a separate language, a claim I haven't seen anywhere else and certainly not in Scandinavian linguistics. The claim that 'proper' Swedish is spoken in Svealand (alone) is unique to Ethnologue and lacks any foundation in reality. Some dialects described are actually the same dialect but with different names.
-Serbo-Croatian: It is said to be a language of Yugoslavia. No country has been called Yugoslavia for some years now. Bosnian, Croatian and Serbian all link to the Serbo-Croatian page and that can be defended on a linguistic basis. What cannot be defended is the information that this language is spoken in "Yugoslavia and Macedonia". What happened to Croatia and Bosnia? The dialects are mixed up. First they are called CHAKAVIAN, KAJKAVIAN, STOKAVIAN, then they are called KAYKAVSKI, CHAKAVSKI, SHTOKAVSKI. Someone not knowing the language might think these are different dialects. The Macedonian minority in Bosnia that Ethnologue has discovered would be unknown both to Bosnians and to Macedonians.
I picked these languages because they are the ones I know best. It would be very strange indeed if only the language I speak are exactly the ones that Ethnologue makes serious mistakes in describing. I would expect to find about the same kinds of errors in other languages as well, although I'm not capable to detect mistakes in languages such as (random pick) Quechua, Romanian, Persian or Japanese.
Ethnologue is a bit like Wikipedia in some sense - a vast ressource which accumulates information from other places and re-publishes it. Some of the information is extremely well researched, some is based on only a few sources. AFAIK once one goes into the depths of the database thse differences in the evidence base become more obvious. And i think this is fair enough.
Refdoc 00:35, 24 Feb 2005 (UTC)
I agree with Refdoc's assessment - except for one crucial point: the Ethnologue only corrects mistakes at multi-year intervals when new editions come out, and is much less responsive. Why not add a new section to this article - Ethnologue errors? With suitable sourcing, it could become a useful resource not only for linguists online but for the Ethnologue editors themselves. - Mustafaa 23:43, 24 Feb 2005 (UTC)
Well, no, if you follow the above links you will see that the words are being used distinctly. -- Doric Loon 05:07, 4 April 2006 (UTC)
I do not know if they are errors, but there are things which annoy me with Ethnologue (otherwise often an useful source, in addition to all the others that is). On one side, they regularly extrapolate the "population" for the main language of any country by substracting from the census the counts for the lesser-used languages: of course this is completely missing out those people which do have two mother languages, for example when both parents are not speaking the same language to their children. Another thing which is surprising are the statistics about immigrants: for example, www.ethnologue.com/show_language.asp?code=fra lists various African countries with counts that seems the ones for the French or Belgian "cooperants" there, but we do not see entries for Germany or Spain, despite the numerous French local population... For the UK, the number of 14,000 could be some estimate for the French-speaking census of the Channel Islands, but it is clearly missing the immigration toward the City (current estimations have 250,000 French residents in the UK, according to an official agency)... I guess similar remarks can be done for the major languages, starting with English, supposedly without native speakers in neither Belgium nor France nor Spain (but with several at St.Pierre and Miquelon or Andorra). -- Antoinel 14:25, 7 April 2006 (UTC)
The article faults previous editions of the Ethnologue for their classification of Khoisan languages, but does not specify what the specific problems are. Since there is a new edition of the Ethnologue now available, I will delete this section. If the new edition is seen to continue the previous fault, then somebody can feel free to insert the complaint again, but they should include some specific details and cite references. Pete unseth ( talk) 12:35, 30 July 2009 (UTC)
The first paragraph says "The Ethnologue contains statistics for 7,358 languages in the 16th edition". I'm wondering what this 7,358 number comes from. In the Introduction on www.ethnologue.com, it says "we tally 6,909 languages which are known to have living speakers who learned them by transmission from parent to child as the primary language of day-to-day communication", i.e. "a person’s “first language” or “mother tongue”". And later in the intro, "the 7,413 languages listed in this edition. These break down as 6,909 living languages, 55 macrolanguages, 28 languages used only as a second language, and 421 recently extinct languages."
If someone arrived at 7358 by using the above figures and excluding macrolanguages (6909 + 28 + 421), the article should probably say so. Since the 7358 is at odds with the most prominent figures put forth by the Ethnologue itself, and directly at odds with the "7413 [unqualified] languages" figure, it is confusing. To my mind, 7413 would be a more intuitive number to use than 7358; but again if someone thinks 7358 is the more useful number to display, they should explain what it refers to.
There is also the issue of comparing apples to oranges between different editions. The article says "up from 6,912 in the 15th edition", but that number refers to living languages. (The total number of languages in the 15th ed. is listed as 7,299 in the intro.)
The word "statistics" could probably be improved on as well. There is at least one number for most languages, but typically there is much more non-numeric information, such as alternate language names and dialect names, location, language family relationships, and so on.
I would propose we replace "The Ethnologue contains statistics for 7,358 languages in the 16th edition" with "The Ethnologue contains listings for 6,909 living languages in the 16th edition". That would clear up the language count confusion in principle, as well as the problem of misleading comparison between editions, and would address the use of the word "statistics". Huttarl ( talk) 20:34, 20 August 2010 (UTC)
According to Ethnologue (16th edition) itself [2]: "The Ethnologue database has been an active research project for more than fifty years. It is probably the most comprehensive listing of information about the currently known languages of the world. Thousands of linguists and other researchers all over the world rely on and have contributed to the Ethnologue database." No more emphasis on missionaries. -- Taivo ( talk) 09:22, 18 October 2012 (UTC)
There has been discussion as to whether Ethnologue is a survey of the world's languages. Not sure why this is difficult. The first sentence says, "The purpose of the Ethnologue is to provide a comprehensive listing of the known living languages of the world" (16th ed p. 7 and 15th ed. p. 7). Earlier editions said, "The Ethnologue tries to bring together the best known information available on the languages of the world" (14th ed. p. vii, 13th ed. p. vii, 12th ed. p. vi). I hope this is adequate to clarify that its purpose is to provide a survey of the world's languages. Pete unseth ( talk) 20:06, 18 October 2012 (UTC)
I am not inclined to get into the political underbelly of Ethnologue, though their historical origins and mission are important to describe. Most all of us who are language scholars agree that religion, among other cultural revolutions, have brought about the destruction of peoples and their languages altogether. Still, the tone here could use a little refining. Also, when I have contacted Ethnologue and complained about, for instance, a language being more endangered than it is represented, Ethnologue editors modify the data and are courteous about it. I've made a few modest edits here. As for SIL in this article, they don't just translate the bible, by the way. They also publish slender primers and chapbooks that describe a given culture's oral history. I am not an SIL affiliate, but I certainly know linguists who are.
KSRolph ( talk) 16:01, 22 March 2013 (UTC)
The Ethnologue was online prior to 2009, I think the 14th version is archived online still. So a change would be needed in the 1st sentence or two of this article. -N — Preceding unsigned comment added by 61.7.176.174 ( talk) 03:23, 18 September 2013 (UTC)
The 13th is archived , but that's not relevant . — kwami ( talk) 19:38, 19 September 2013 (UTC)
This talk page is over 50k and it has comments older than most articles. I'd like to use lowercase Sigmabot to start automatic archival. Chris Troutman ( talk) 01:32, 23 July 2014 (UTC)
In the Overview section, SIL International is described as "(formerly known as the Summer Institute of Linguistics, originally Wycliffe Bible Translators)". This is not quite right. The Summer Institute of Linguistics and Wycliffe Bible Translators were legally incorporated as separate organizations in 1942. Although they work closely together, and have overlapping membership, they were never the same organization, and both exist as legally-separate organizations at this time. Therefore, I think the wording should be something like "(formerly known as the Summer Institute of Linguistics, which is associated with Wycliffe Bible Translators)". Perhaps better, simply omit the reference to Wycliffe entirely, because mentioning that name doesn't add anything in context beyond what is already said in describing SIL, and people can always find out about the relationship of the two organizations by following the link to the article on SIL. As a member of both organizations, and part of the Ethnologue staff, it isn't appropriate for me to make the change, but I submit this for consideration by other editors. AlbertBickford ( talk) 15:24, 27 November 2014 (UTC)
The EGIDS scale, used to rank language vitality in Ethnologue, was revised in the 17th edition to make it easier to apply to sign languages. The revisions are documented in this article: http://www.tandfonline.com/doi/abs/10.1080/01434632.2014.966827. Since I'm the lead author, I won't add it to the WP article on Ethnologue, but if someone else agrees that it is worth citing, the place I'd suggest inserting the reference would be right after the reference to the original Lewis and Simons article on EGIDS. AlbertBickford ( talk) 21:31, 2 December 2014 (UTC)
I adjusted a sentence in the overview to better express the dependency of recent editions of Ethnologue on ISO 639-3. It now reads "The 15th edition of the Ethnologue was the first edition to use this standard, called ISO 639-3, and since then Ethnologue relies on the standard to determine what is a language." Other editors should check this, since I'm a research editor for Ethnologue, to make sure they agree this is from a NPV. AlbertBickford ( talk) 19:44, 22 January 2015 (UTC)