Talk:Altaic languages

From Wikipedia, the free encyclopedia
Jump to: navigation, search
          This article is of interest to the following WikiProjects:
WikiProject Languages (Rated C-class, Top-importance)
WikiProject icon This article is within the scope of WikiProject Languages, a collaborative effort to improve the coverage of standardized, informative and easy-to-use resources about languages on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Top  This article has been rated as Top-importance on the project's importance scale.
WikiProject Central Asia (Rated C-class, High-importance)
WikiProject icon Altaic languages is part of WikiProject Central Asia, a project to improve all Central Asia-related articles. This includes but is not limited to Afghanistan, Kazakhstan, Kyrgyzstan, Mongolia, Tajikistan, Tibet, Turkmenistan, Uzbekistan, Xinjiang and Central Asian portions of Iran, Pakistan and Russia, region-specific topics, and anything else related to Central Asia. If you would like to help improve this and other Central Asia-related articles, please join the project. All interested editors are welcome.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.
WikiProject Turkey (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Turkey, a collaborative effort to improve the coverage of Turkey and related topics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Iran (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Iran, an attempt to build a comprehensive and detailed guide to articles related to Iran on Wikipedia. If you would like to participate, please join the project where you can contribute to the discussions and help with our open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Korea (Rated C-class, Mid-importance)
WikiProject icon This article is within the scope of WikiProject Korea, a collaborative effort to build and improve articles related to Korea. All interested editors are invited to join the project and contribute to the discussion.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject East Asia (Rated C-class, High-importance)
WikiProject icon This article is within the scope of WikiProject East Asia, a collaborative effort to improve the coverage of East Asia on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 High  This article has been rated as High-importance on the project's importance scale.

Citation style[edit]

The style of citing sources used in this article is inconsistent. It needs to be regularized. Otherwise, a "citation style" tag could be placed on this article and sooner or later will be. It looks like this:

(This is of course only an example of the tag. It is not an actual tag placed on an article.)

There are several citation styles accepted on Wikipedia (Wikipedia:Citing sources). Wikipedia recommends only that one of the authorized citation styles be followed and that it be used consistently throughout the article.

There are two major citation styles, footnotes and author-date referencing.

Each style has advantages and disadvantages. In some academic fields, author-date referencing is now preferred. Many linguists prefer it (e.g. the late Winfred P. Lehmann, for instance in his Theoretical Bases of Indo-European Linguistics).

This article currently has 10 footnote references and about 81 author-date references. (The latter figure could be raised if all mentions of a work and its date are included, but it seems advisable to restrict the category to items that would normally receive a note number if footnoting is used.)

In view of the prevalence of author-date referencing in this article, I propose switching the article from its current mixed status to author-date referencing. With this in mind, I am changing the 10 footnote citations to author-date citations.


Since there is no such thing as an "Altaic people", I merged that article here. Cleaned it up a bit, but it needs more. kwami (talk) 00:42, 27 January 2009 (UTC)

Micro-Altaic is dead (or at least in hibernation)[edit]

I am editing the intro section to bring it in line with the current views of Altaicists, as opposed to those they held fifty years ago. The relevant considerations are enunciated by Stefan Georg and his collaborators in their 1999 article[1] (see especially pages 73-74). This is a particularly important article in that Georg is a major opponent of the Altaic hypothesis, whereas his co-authors (Manaster Ramer, Michalove, Sidwell) are major supporters of it. It can thus be taken as representing a responsible consensus. I quote this passage in full to avert any suspicion I am editing the text:

Equally misleading to the non-specialist is the claim (e.g. Comrie 1992, Lyovin 1997) that the traditional Altaic theory connects only the western languages (Turkic, Mongolic and Tungusic), and that some combination of Korean, Japanese and Ainu has been marginally associated with the western languages by some scholars. In fact, the status of Korean, Japanese and Ainu differs greatly in Altaic studies. While it is true that the oldest literature on Altaic dealt only with Turkic, Mongolic and Tungusic, the fact is that Korean has been an integral part of the Altaic theory for most of the postwar period. Indeed, practically all scholars who have accepted the relationship among the Turkic, Mongolic and Tungusic languages since Poppe (1960) have also included Korean in their definition of Altaic (although in practice many Altaicists have not worked with Korean in any depth until recently). On the other hand, the relationship with Japanese was worked out more recently and still, while accepted by many Altaic scholars (such as Miller, Starostin and Vovin) to be Altaic, is not granted by others (e.g. Tekin and Baskakov).

Elsewhere, the authors emphasize that the Altaic hypothesis today concerns "the Turkic, Mongolic, Tungusic, Korean and, in most recent versions, Japanese languages".[2]

Another point: I have used the term "anti-Altaicist" here rather freely. My justification for this is that Stefan Georg likes the term, as you can see from his comment on this talk page above, under "Anti" (last comment in the section).

It is important that both Altaicists and anti-Altaicists feel that their positions are correctly represented, as well as neutrals. This has been my aim in re-writing this section. Regards to all. VikSol 11:07, 29 January 2009 (UTC)

I took Ainu out of the info box.
The lang fam nav box (at the btm of the page) used to have turk / mong / tung listed separately. We could choose either that or Altaic, as we have now, but there's a third choice: list the individual families (as we do now for japonic) and add Altaic as a 'perhaps also', like Tyrsenian. What'd'ya think? kwami (talk) 11:50, 29 January 2009 (UTC)

Honestly, I find the language families template to be a space hog. I think the default should be switched back to "hide" rather than "show", so people can open it up if they want to explore the information it contains, but otherwise be left to concentrate on whatever they're looking into. This said, it is an eye-opener, since it squeezes into a very small space information that is usually too spread out to take in at a glance.

My personal views aside, I don't think there is enough of a consensus at the present time either for or against the Altaic family to either accept or reject it. The problem is to translate this fact into a tabular medium.

I like listing T M MT separately, with "perhaps Altaic" added at the end of the section, providing a convenient link.

By the way, Korean needs to be added to the Eurasian languages. Also, what about Nivkh? I can think of a few other suggestions. Regards, VikSol 13:59, 29 January 2009 (UTC)

Language isolates have been left out. They'd double the size of the template. Unless you think Nivkh should be considered a small family? kwami (talk) 19:13, 29 January 2009 (UTC)

Taxonomically, it makes no difference whether a language family is represented by one language or many. E.g. the Eskimo-Aleut family is composed of two branches, the Eskimo family and the single Aleut language, but each is an equal branch. Likewise, Eskimo is subdivided into the Yupik language family and the single Inuit language, but Yupik and Inuit have equal status. VikSol 21:46, 29 January 2009 (UTC)

Sure, but it's impractical to double the template, and it's also impractical in many cases to distinguish isolates from unclassified languages. Besides, we'd get into arguments with people who don't understand the concept of a family of one. kwami (talk) 10:48, 30 January 2009 (UTC)
I have a copy of Georg's article. It is very good. But I'm confused by your talk section title "Micro-Altaic is dead" considering that I read nothing in the Georg article that could be construed as implying that "Micro-Altaic is dead". --Stacey Doljack Borsody (talk) 20:33, 29 January 2009 (UTC)

Georg and his co-authors state that nearly everybody who believes in Altaic today assigns Korean to it, which suffices to make it Macro-Altaic, and most of them even include Japanese. On page 75, first paragraph: "practically all scholars who have accepted the relationship among the Turkic, Mongolic and Tungusic languages since Poppe (1960) have also included Korean in their definition of Altaic". Anti-Altaicists by definition don't believe in Altaic at all. Skeptics do not support either Micro-Altaic or Macro-Altaic; they simply think the arguments for and against Altaic are about equally strong. This leaves nobody who supports Micro-Altaic. In this sense Micro-Altaic is dead. This doesn't mean that Altaicists reject kinship between Turkic, Mongolic, and Tungusic; it simply means they extend this grouping to Korean and, usually, Japanese (I wish they would start to say Japonic). VikSol 21:46, 29 January 2009 (UTC)

Thanks. That clarifies it better. It didn't seem clear from the first quote. I like that you're updating the article using the information from Georg. It seemed like there was the answers to a number of questions raised in past talk entries sitting right there... --Stacey Doljack Borsody (talk) 17:12, 30 January 2009 (UTC)

Turkic is an iranian-turkic creole language and oghuz turkic is an iranized turkico-iranian.[edit]

Turkic is an iranian-turkic creole language and oghuz turkic is an iranized turkico-iranian.

In the site

they give iranian etymologies to turkic numbers. gi=>eki tse=>uthse tshorts=>tört pandj=>bish atshish=>alti and so on

Non oghuz turkic languages have rather an irano-altaic conjugation endings. kor-gen-men=see-past suffixe-first person(likely borrowed from iranic)ending.

But in oghuz turkic it became gor-d-um=see-iranian past suffixe d-iranian first person ending.

if you look to these maps below,you could easily see that central asia was inhabitated by iranian speaking populations(saka,chorasmians,dahae,margians,bactrians,soghds..)and of course these tribes did not disappear but merged with turkic newcomers as proven by genetic tests and also by the presence of a caucasoid phenotype and caucasoid phenotype influences amongst central asian turks.

john L.Drake —Preceding unsigned comment added by (talk) 17:19, 20 November 2009 (UTC)

NO, YOU'RE WRONG ABOUT NON-OGHUZ TURKIC DIALECTS. In non-Oghuz Turkic, The word "körgenmen" means "I have seen" NOT "I saw". Modern Oghuz dialects have no Present Perfect Tense but old Oghuz had; the word "göryenben" means "I have seen" in old Anatolian Turkish. If you want to say "I saw" in non-Oghuz Turkic, you must say "kördüm" instead of "körgenmen".
Some Examples;


kel-gen-men => i've come => gel-yen-min (Old Oghuz)
yığla-gan-sın => you've cried => ağla-yan-sın (O.O.)
tut-gan => he/she/it has held => tut-yan (O.O.)
bas-gan-mız => we've stepped/pressed => bas-yan-ız (O.O.)
tab-gan-sız => you(plural) have found => tap-yan-sız (O.O.)
ket-gen-der => they've gone = git-yen-ler (O.O.)
tüshün-dü-m => i understood = düshün-dü-m
söyle-di-n => you said => söyle-di-n
bashta-dı => he/she/it began = bashla-dı
böl-dü-k => we divided => böl-dü-k
al-dı-nız => you(plural) took/got => al-dı-nız
ur-du-lar => they struck/hitted => ur-du-lar —Preceding unsigned comment added by (talk) 17:55, 1 March 2011 (UTC)

There's been a lot of mutual influence between Turkic and Iranian. But AFAIK Oghuz has never been claimed to be a creole; in most respects, it is clearly Turkic. kwami (talk) 19:47, 20 November 2009 (UTC)
Agreed. There's plenty of influence, but the Oghuz languages are clearly Turkic. (Taivo (talk) 19:54, 20 November 2009 (UTC))

I have for a long time asked my self, how correct is really this altaic language group that is mostly hypotatical and that the altaic language group is including the turkic language group. I am looking at some facts; the geograpgy the turkic peoples live today is the same geography the scythian peoples lived. The scythian language was an iranian language. Since it was covering such a large place it is very open for influences from others, which make them change more than the southern iranians, the parthians. The area that is called central asia today was called Turan by the Persian long time ago and this is where the Turks/Turkics get their name from. The Turanian ideology that was created in the beginning of the 20th century was going a little too far by trying to include the Uralic and the Darian groups as well. And now this Altaic theory that doesn't convince me when I compare grammar or vocabulary from the different languages that it is supposed to group together. I think the correct way to look and the correct hypotetis to trying out is this; the Turkic or it could be renamed to the Turanian language group should be part of the Indo-European language group, with Turkish, Azerbaijani, Turkmenish, Uzbeki, Kirghizi and Kazakhistani as undergroups. I think the Turkic/Turani is closer to IE than belonging together with Chinese, Korean, Japanese, Uralic or Darian. The Turkic/Turani language group is closer to Iranian than any of the others mentioned above. BUt since it has been open for influence and change for such a long time, it could be included in IE as a own group instead of becoming an undergroup of the Indo-Iranian. Conclusion; the Turkic should not be part of Altaic, but of instead become part of Indo-European as Turkic or Turanian. —Preceding unsigned comment added by (talk) 23:47, 21 November 2009 (UTC)

Just one minor note, please "AVOID" using any term that sounds, smells, or looks like the non-existence term "Turani language". Xashaiar (talk) 01:26, 22 November 2009 (UTC)

Chinese is sino-tibetan not altaic.

Of course kipchak Turkic and at a lesser extent oghuz turkic are agglutinative altaic languages(with mongolic and tunguzic)but with a strong iranic superstratum(or substratum)especially for oghuz turkic.

Kipchak turkic for "I see" is körgenmen(kör+gen+men)with kör=see,gen=past suffixe,men=I. But oghuz turkic for "I see" is gördüm(gör+d+üm) with:


d=past suffixe(along with "t" due to consonant harmony rule)=same as with persian which have either "d" or "t" as past suffixe

üm=1.person ending(along with "im,ım" due to the vowel harmony rule)=same with persian which have "aem" as 1.person ending.

also pir/bir 1 could be connected with indo-european per meaning lone as in english first.

sekiz could have be connected to hekiz>hekt indoeuropean 8.

Perhaps original Turkic numbers are the ones that express decades nowadays.

1=on 2=yirm 3=ot 4=kirk 5=el(one hand or 5 fingers) 6=alt(under) 7=yet(indo-european/indo-iranian borrowing) 8=hek(indo-european/indo-iranian borrowing) 9=tok 10=yiz —Preceding unsigned comment added by (talk) 11:06, 23 November 2009 (UTC)

Not sure why I bother responding. It is clear that the above poster knows nothing of Turkic languages otherwise they'd not confuse readers by claiming that there's only one past tense and then make comparisons with the Turkic languages' two past tenses. --Stacey Doljack Borsody (talk) 06:15, 14 April 2010 (UTC)

im from Turkey. I am a Turkish person. You do not know Turkish language. Your informations are wrong. Your grammer is broken, words are wrongs bla bla.. You dont know Turkish. — Preceding unsigned comment added by (talk) 13:20, 30 July 2011 (UTC)

Japonic languages[edit]

This article claims that Altaic language family includes Japonic languages. While this is one theory, the Japonic language family generally considered a language isolate. On the wikipedia page for Japonic languages, for example, it is identified as an isolate rather than a member of the Altaic family. While the article does use qualifying language, I don't think it's clear enough.

Korean too. In the description, it says that the Altaic family includes the Korean language isolate. That doesn't make sense. If it's an isolate, it doesn't belong to a language family. I don't think that sentence was well phrased. (talk) 10:40, 15 September 2011 (UTC)

Since Altaic is unproven and not widely accepted as a language family, then Korean is still a "language isolate", even when some few linguists want to include it in an unproven and not widely accepted proposed language family. --Taivo (talk) 13:24, 15 September 2011 (UTC)

iV ~ æøy[edit]

Consonant table footnotes 11 & 12 talk of an /iV/ environment. Should I assume this actually means *æ *ø *y? --Trɔpʏliʊmblah 23:08, 3 June 2010 (UTC)

The map in the infobox[edit]

The map in the infobox should be reviewed. In European Russia and Ukraina where are the Tatars, Crimean Tatars, Chuvashs, Bashkirs, as well as smaller communities of Turkic speakers in Caucasus ? Where are Turkic speakers of Iran and Turkish speakers of ex Ottoman teritory in mideast and east Europe ? Where are Turkic and Mongolian speakers of Afganistan ? Nedim Ardoğa (talk) 10:06, 23 June 2010 (UTC)

Short comment on recent edits[edit]

I just looked at the current stage of this page and got the impression that it has noticeably worsened since December 2010. Some unsourced data was rightfully deleted, as was some sourced. But I am not gonna enter into the discussion - I still don't feel inclined to make major contributions to this article. However, I tried to combine the passages on the critique of EDAL of the old and new version. Citing the available discussion as the article used to do was basically the right thing, so I reinstalled that sentence. But focusing on Vovin's claim that EDAL falsified data is not justified - Vovin makes this claim, if I remember correctly, about a particular reading of a phonological reconstruction of a word from the Huayi Yiyu, and shows that the authors of EDAL were inconsistent in reading it, alleging something such as intention. But when I read it, I couldn't help but wonder whether we were talking about wishful thinking on the part of Starostin et al. In any case, Vovin's argument does not sum up to alleging that the authors of EDAL conspired to falsify their data. On the other hand, his pointing out that they ignored data that they must have known is much more prominent in his article and a much more sustained critique that deserves to be cited here. The other point about falsification has not been central to any part of the discussion at all and, if not reinforced by additional evidence, will probably not reflect on the way how anybody might approach this dictionary (which for other reasons is so shaky that only alert expert users who have the necessary knowledge to perceive the innumerable mistakes may dare to have a look at it, as pointed out by Georg). G Purevdorj (talk) 12:07, 21 April 2011 (UTC)

Postulated Urheimat[edit]

The section titled "Postulated Urheimat" does not discuss or postulate an Urheimat. It should be re-titled. Furthermore I would be interested if there is a Postulated Urheimat, like the Atlas Mountains maybe? — Preceding unsigned comment added by (talk) 16:19, 21 September 2011 (UTC)

Japanese and Korean are NOT Altaic languages !!![edit]

East Asian languages such as Japanese, Korean and Ainu are not related to Altaic languages at all. The closest relative of the Altaic languages are probably Uralic language, and not Japanese/Korean.

East Asian languages :

  • Korean
    • Korean hana tul set net tasôt yôsôt ilgop yôdôl ahop yôl
    • Sino-Korean il i sam sa o yuk ch'il p'al ku sip
  • Ainu
    • Proto-Ainu+ sine- tu:- de- i:ne- aski i:hdan- adehdan- tupedhdan- sinepehdan- hdan-
    • Ainu+ shine tu re ine ashikne iwan arawan tupesan shinepe-san wan
    • Kuril shiné do:bechi re:-bichi ine-p ash'kine-p iwam-pe aruwam-pe dobisam-pe shinibesam-pe wam-pe
    • Sakhalin sine-h tu-h re-h i:ne-h asne-h iwan-pe arawan-pe tupesan-pe sinepisan-pe wan-pe
  • Japanese
    • Old+ pitö puta mi yö itu mu nana ya könönö töwo

native Japanese hitotsu futatsu mittsu yottsu itsutsu muttsu nanatsu yattsu kokonotsu to:

    • Sino-Japanese ichi ni san shi go roku shichi hachi ku juu
    • Okinawan tichi ta:chi mi:chi yu:chi ichichi mu:chi nanachi ya:chi kukunuchi tu:

VS Altaic languages :

    • Turkish bir iki üç dört bes, alti yedi sekiz dokuz on
    • Tatar ber ike öch dürt bish alti jide sigez tugiz un
    • Mongolian (Khalkha) nig xoyor guraB döröB taB dzorghaa doloo naym yös araB
    • Manchu emu zhuwe ilan duin sunzha ninggun nadan zhakûn uyun zhuwan

— Preceding unsigned comment added by (talk) 18:10, 13 March 2012 (UTC)

Basically, as this theory has not been generally accepted (yet) you are correct, but the fact is that this article talks about the theory. My knowledge? NOTHING. Hill Crest's WikiLaser (Boom). (talk) 00:13, 26 March 2012 (UTC)

Given the more thorough non-existence of all of Altaic, the whereabouts of Korean and Japonic within this theory should be not much to worry about :-). G Purevdorj (talk) 06:50, 27 March 2012 (UTC)

Lexical comparisons in Altaic[edit]

The Comparative Pronouns Table by Blazek should be eliminated.

1) the comparisons between Altaic and Tibeto-Birman or Sino-Tibetan are well beyond the scope of this article, and are nothing more than simple speculation. 2) The comparisons are an original contribution by Václav Blažek and are not yet accepted by the majority of linguists. 3) The table implies that there is an accepted reconstruction of the Altaic pronouns (there is none generally accepted). 4) Wikipedia should provide useful, reliable information, or at least information accepted by the consensus of scientists, not learned guesses. 5) Encyclopedic content must be verifiable (see under). — Preceding unsigned comment added by (talk) 10:48, 10 March 2012 (UTC)

Why eliminate that table and not the rest?
I don't know what the ST forms are doing there, but they are a good reminder than many of these correlations could be coincidental. — kwami (talk) 10:58, 10 March 2012 (UTC)

the opening lead needs strong "not widely accepted" type of statement[edit]

Reading it, I could get the impression that this is a strong contender as far as language-science goes. An expert linguistic source should be cited that makes it clear this is - while perhaps not a fringe theory - at the least a borderline one. 01:38, 28 May 2012 (UTC)

history repeats itself[edit]

I see that the same "talking past one another" arguments seem to occur in pro vs. anti-Altaic as everywhere else between lumpers and splitters:

According to Roy Andrew Miller (1996: 98-99), the Clauson–Doerfer critique of Altaic relies exclusively on lexicon, whereas the fundamental evidence for Altaic consists in verbal morphology. Lars Johanson (2010: 15-17) suggests that a resolution of the Altaic dispute may yet come from the examination of verbal morphology and calls for a muting of the polemic. In his view, "The dark age of pro and contra slogans, unfair polemics, and humiliations is not yet completely over and done with, but there seems to be some hope for a more constructive discussion" (ib. 17).

When considering distant relationships, splitters habitually seem to view only lexicon as worthy of consideration while lumpers seem to habitually look more at morphology. Not surprising then that mutual hostility results -- the most heated polemics always seem to arise in cases where there's no scientific way to settle a dispute. Exact same stuff being marshalled on the two sides of Dene-Yeniseian, etc. Benwing (talk) 04:39, 4 September 2012 (UTC)

Altaicists and critics of Altaic[edit]

Those lists are a bit confusing. For example Lars Johanson is listed among the Altaicists, however he seems to be agnostic to the problem or at most a proponent of an alternative hypothesis. Judging from his works he simply proposed a method to evaluate pros and cons of the theory (cognates and copy in altaic verb derivation 1999), remaining neutral. He wrote the Altaic part of the "Concise Encyclopedia of Languages of the World" by Elsevier[3] and he seems to be at most neutral, though he is quite critical to a genetic relation between Turkic-Mongol-Tungusic languages, where he seems to favour contact processes. In fact the work cited as an evidence of his support to the Altaic theory (Transeurasian verbal morphology in a comparative perspective: genealogy, contact, chance 2010) clearly states: "defining “Transeurasian” as a group of geographically adjacent languages that share a significant amount of linguistic properties, we do not need to presuppose genealogical relationship. Most of the authors contributing to this volume would not unequivocally subscribe to the hypothesis that the Transeurasian languages are genealogically related."[4] Moreover, four of those Altaicists belong to the same group, Dybo, Mudrak, Sarostin x2, is that adequate? (talk) 05:28, 26 September 2012 (UTC)

I may add that KM Lee is a mild proponent of Macro-Altaic: he believes that an Altaic family exists, he is a proponent of a genetic link between Korean and Tungusic, he is fairly sure about a link between Korean and Japanese, however he acknowledges (2011) that:
  • "common linguistic features does not in any way constitute proof of genetic affinity, but they are suggestive" (about Altaic, Korean and Japanese)
  • "there is no general agreement on the genetic relationships of either Japanese or Korean" (citing Samuel Martin 1991).
  • "In our view, the prospects for comparative work between Korean and Tungusic appear to be somewhat better" [than Korean and Japanese] and "there are, to be sure, matches between Korean and Japanese for which correspondences are not to be found in Altaic or everywhere else."
  • so he concludes "it is more likely than not that Korean is related to Japanese, though at the present stage of our knowledge it is impossible to say just how distant such relationship, if it exists, might be. What we do know is that the task of proving the relationship remains as yet very much incomplete."
So it seems better to me to list him as a proponent of a "Turkic–Mongolic–Tungusic–Korean and possibly Japanese" family rather than simply a Macro-Altaic family, since for Macro-Altaic proponents the inclusion of Japanese is central, more so than Korean.
Also, why is Street listed among Altaicists, while Greenberg and Patrie are among the alternative ones? Are not their ideas somewhat similar? (talk) 17:01, 26 September 2012 (UTC)
S. Robert Ramsey, co-author of KM Lee 2011, is really unsure about a genetic Korean-Japanese relation: "if we the knew for a fact that they were not [genetically related], then we would interpret the findings above as corroborative. If we knew for a fact that they were related, then we would be more impressed by the slimness of the foregoing evidence than by the seeming minimal contrast"(2004). This is the same point of view of Lee, listed among the "Altaicists". (talk) 22:34, 26 September 2012 (UTC)
I agree. In the academic world, Korean and Japonic are generally NOT included in this language family. The article is misleading. --Lysozym (talk) 05:48, 27 December 2012 (UTC)

What evidence and objections are there to Altaic as a family?[edit]

The article repeatedly states that Altaic is controversial and that it is not widely accepted and so on. At the same time, it lists a number of languages that are commonly held to belong to the family, and lists regular sound correspondences and cognates. But what it doesn't say is why. Why are those correspondences valid? How many words adhere to them? Which words do not? Which objections have been raised? Is it a lack of conclusive evidence, a sparsity of undisputable cognates, lack of regular sound correspondences? What evidence is there for a relationship? Is there a lack of evidence? CodeCat (talk) 02:43, 15 March 2013 (UTC)

Correcting Article to Reflect Leading View of Korean and Japanese as Only Hypothetically Part of Altaic[edit]

Of course within any comparative linguistics specialty, the experts there who believe their own work will conclude that this is the proven standard. The question is, what does the broader comparative linguistic community think of their work?

And the answer is quite clear that Japanese and Korean are almost universally not accepted as well-established members of Altaic.

Oxford Dictionary of English: "Altaic - denoting or belonging to a phylum or superfamily of languages which includes the Turkic, Mongolian, Tungusic, and Manchu languages."

Ethnologue —




I am making a few minor changes to the article to make this reality clear to readers, while leaving the Korean/Japanese columns of the "reconstructions" untouched in context of "greater Altaic" claims. — Preceding unsigned comment added by (talk) 01:52, 5 April 2013 (UTC)

Those are not reliable sources. We've been over this several times, and this was the last consensus. Should be discussed if you wish to change it. — kwami (talk) 01:59, 5 April 2013 (UTC)
Wikipedia articles must report first not what one or more Wikipedia editors consider "correct", but what the most standard, mainstream academic treatment of a subject is. Then alternate academic views should definitely also be presented. There are believers in UFOs who will be telling me in another article that the New York Times saying there are no UFOs is "not a reliable source" while their favorite new book by a UFO-ologist is. And they've all discussed that and it's the truth. The linguists writing books on Nostratic are convinced they have proven that Nostratic exists -- and they have in some ways more evidence than for Japanese being part of Altaic. Linguistics is famously argumentative, and Ethnologue, which has no agenda whatsoever to support or not support any particular language family theory, is the most authoritative neutral source today.
I am making no attempt to alter the charts of "reconstructions of proto-Altaic" (or perhaps areal feature comparisons) in the article, only to make sure that readers who knew nothing of this subject, leave the article knowing that most sources not written by Altaicists deem Japanese and Korean to be isolates, which is a verifiable statement of fact.
Do you have any higher-level language reference source, dealing with all language families — not a work by partisan Altaicisist specialists — which recognizes "Macro-Altaic" as standard and accepted by the broader comparative linguistics field?
You need to discuss this first. Please read WP:BOLD. If you don't, you can be blocked for edit warring.
BTW, *all* of Altaic is hypothetical, not just Korean and Japanese. That was part of the last discussion. — kwami (talk) 03:37, 5 April 2013 (UTC)
I've just discussed it. Where's your general reference source?
You have no "consensus" of even Wikipedia editors, just look above: "I agree. In the academic world, Korean and Japonic are generally NOT included in this language family. The article is misleading. --Lysozym (talk) 05:48, 27 December 2012 (UTC)"
That's the whole problem, misleading the readers to believe that something is broadly accepted which is not broadly accepted (whether it's true or not). You have references to support the claim "Altaicists are convinced they are right". Do you have a reference to support the claim that anybody else believes them? Post it, now, or it is you who are trying to vandalize the article with personal opinions and personal agendas. — Preceding unsigned comment added by (talk) 03:49, 5 April 2013 (UTC)
If you don't like the existing consensus, you can try to change it, but personal attacks are not a good way of doing that. And in case a conspiracy theory is next, that doesn't have a good track record either. Present a rational argument and try to convince people. You might want to invite some of the people who participated in this the last time. — kwami (talk) 04:02, 5 April 2013 (UTC)

I apologize for any exasperation. But the "consensus" of the broad comparative linguistics field today is against "Macro-Altaic". And this is such plain fact, supported by the verifiable references that WIkipedia demands, that it's frustrating to have to keep mentioning it. And the "consensus" in this Talk section, is of you refusing to listen to everybody pointing that fact out to you.

You cannot personally own a Wikipedia article. You've gotten it into a distorted state that you like, and now you are hiding behind "not changing things" without "consensus" — meaning your permission. Start a blog on Macro-Altaic to convey your personal views.

I have found a WIkipedia article presenting clearly misleading information to the public. I must correct this. You have the opportunity again now to demonstrate to me that I am wrong on the facts — not wrong on "consensus", not wrong according to your vague assurances — by giving authoritative general linguistics reference sources which present "Macro-Altaic" as the standard view today. Links please. — Preceding unsigned comment added by (talk) 04:20, 5 April 2013 (UTC)

You might want to read the past discussions so you know what you're talking about. It's not for me to convince you, but for you to convince us: You're the one trying to change consensus. — kwami (talk) 04:28, 5 April 2013 (UTC)
Ethnologue is the consensus of linguistics today, which you, and a few other partisan Altaicists editing this article, are trying to replace with a distortion of reality. You and your pals are claiming the sky isn't blue. You say the article can't change unless I convince the people claiming that the sky isn't blue, that it's blue.
I have never once said that Korean and Japanese "are not part of Altaic". I have never said I want the article not to mention anything about them being part of Altaic. I want both sides reported to readers clearly: Altaicists strongly believe they have proven a connection and here's their evidence. The broader linguistic community is not convinced. You're trying to silence and erase the mainstream view in linguistics today. You're trying to own an article and prevent another editor from presenting an important and referenced true fact about the subject, because you (and maybe a few other biased editors) personally don't believe it. I'm going to try to undo your undo. And again until the article gets locked. And if it gets locked in the state blatantly lying to readers, then I will have to file a case with Wikipedia's dispute resolution process. They will see the authoritative status that Ethnologue holds in linguistics today. They will see dictionary definitions. They will see your refusal and inability to present contrary evidence.
Look again in the History versions at the changes I made to this article. All I do is let readers know the reality of what non-Altaicists think about Macro-Altaic. — Preceding unsigned comment added by (talk) 04:56, 5 April 2013 (UTC)
OK, well no response from you. I'm sorry to be confrontational but need to do this in pursuit of Truth, Justice, and the Wikipedia Way. I've also now noticed that you minimized an entire Altaic conference (Stanford 1990) which supported Unger's alternate definition of Altaic as being Japanese-Korean-Tungusic. Strange nomenclature indeed, but not something to be hidden from public knowledge to advance a competing opinion. And you only mention the long-standing Uralic-Altaic hypothesis as a passing error. I'll put that more clearly and prominently in the intro. I'll also be adding a few other additional references.
If I am successful in making all these edit without them being reversed, I will then monitor the article over the long term, and file a dispute case if any attempt to dictatorially suppress non-Macro-Altaic points of view is made again. — Preceding unsigned comment added by (talk) 06:16, 5 April 2013 (UTC)
Wow. You're not going to last long in a social environment like WP if you're this antisocial. — kwami (talk) 07:24, 5 April 2013 (UTC)
I couldn't easily revert just your edit warring, so I reverted everything, as it's not worth my time to baby-sit your edits. You'll end up being blocked if you think this is the way to promote WP:TRUTH. — kwami (talk) 07:26, 5 April 2013 (UTC)

Kwami, can you not see that this is not about your social environment. It's about you trying to impose a blatantly unreal non-fact on a Wikipedia article. Korean and Japanese are simply not generally accepted as members of Altaic by the field of linguistics at large. You're trying to insist that the article say that they are generally accepted.

I keep asking you for some evidence of your claim, but you have none. Because there is none. You keep coming back with "community" issues and threatening me with ostracism. Wikipedia articles are not about serving "you", they are about serving the reading public, with objective reporting on the state of affairs of a given subject. — Preceding unsigned comment added by (talk) 02:31, 6 April 2013 (UTC)

You keep talking about the "truth" but that does not count: WP:VNT. Saying I don't own the article seems a bit like the pot calling the kettle black, doesn't it? You still haven't provided reliable sources that back up your edits. Ethnologue is not a reliable source on this subject, as Kwami has already noted. If the original version has unsourced material, you are free to remove it, but any new material that you add must be sourced. Reverting unsourced edits does not require a source, because neither Kwami nor me have added new material, we have just removed yours and reinstated what was already there. Please provide a reliable source for your material or you will continue to be reverted and eventually blocked. CodeCat (talk) 03:15, 6 April 2013 (UTC)
Reverting back to the last stable version before the current editing. Altaic is not "widely accepted" in any sense. Altaicists constitute a minority of historical linguists and have for decades. Any attempt to try to imply that some version of Altaic is widely accepted or anything more than a minority view is contrary to linguistic fact. If you can build a consensus for some other wording that doesn't imply that Altaic is more than a minority hypothesis then it can be added, but rather than editing on the article and being constantly reverted, I suggest you propose changes here first. --Taivo (talk) 04:06, 6 April 2013 (UTC)
  • Britannica doesn't include Japanese and Korean. The Concise Encyclopedia of Languages of the World states that the inclusion of those languages in the group is controversial. The article seems to be giving undue weight to that controversial point of view. For example, the infobox states that Japonic and Koreanic are "generally included". If those languages appear in the infobox, it should say that they are generally not included. Warden (talk) 09:01, 6 April 2013 (UTC)
You don't understand the point. It may very well be that including Japonic and Korean is more controversial than Altaic as a whole, but the way you are editing the article makes the Altaic hypothesis seem to be widely accepted and only adding Japonic and Korean makes it controversial. That's the issue that you are ignoring. You have to make the entire proposal, whether including Japonic/Korean or not, clearly controversial and not widely accepted. --Taivo (talk) 09:43, 6 April 2013 (UTC)
  • I have not edited the article. The issue here is the status of Japanese and Korean, as indicated by the section title. If you have some other point to make please start a separate section for it. Warden (talk) 10:46, 6 April 2013 (UTC)
If I remember correctly, it was already mentioned by Georg et al. that encyclopedias are more on the traditional side when reflecting the Altaic controversy by not including Korean and Japonic. The active supporters of Altaic (Moscow school, Robbeets) are all Macro-Altaists, and both somewhat historically informed (SIAC, PIAC) and utterly clueless (WAFL) Altaistic conferences are open to contributions on Japonic and Korean. For the anti-Altaicists, the scope doesn't matter too much as contacts between all of these adjacent languages are certainly worth close investigation. So what's the fuss about? G Purevdorj (talk) 11:12, 6 April 2013 (UTC)
I agree that this is much ado about nothing, as there are no grounds to stake a definitive claim to either inclusion or exclusion, so all relevant POV in RS should be presented in accordance with WP:DUE.

Wikipedia definitely needs to clearly distinguish for readers between academic theories that are "standard", "less accepted", "controversial", etc. And this is also the basis for the order and prominence given to the different theories in the article.

Obviously some kind of references are needed to establish this. Not just editors' opinions. If this Macro-Altaic is now widely accept as the "standard", is there some citation that can be given? The only reference currently is "Georg et al. 1999" -- but this is just a journal article by a researcher proposing this, not a higher-level "peer review" of what language families are broadly accepted today in comparative linguistics.

"Macro-Altaic" is not the "standard" since all versions are Altaic are widely rejected by the majority of historical linguists. That's the point here. --Taivo (talk) 04:59, 8 April 2013 (UTC)

Comment from ANI discussion[edit]

Dictionaries and encyclopedias are not RS for this unsettled question. Comparative linguistics is highly speculative, being largely based on statistical comparisons as opposed to historical data. Even where similarities characterized as "typological" (i.e., syntactic) are recognized, lending credibility to a possible connection, if phonological and semantic correlaries are scant, some will discount any connection outright.
Oftentimes positions on this are politically motivated. Nationalists in various countries see any drawing of a connection as a dilution of their pedigree or a threat to their independence, for example.
In the future this topic will become more interdisciplinary. For example, I believe that there is little debate among archaeologists and anthropologists regarding the influx of Tungusic peoples into the northern part of the Korean peninsula. That would seem to provide ample room for a linguistic connection on some level.--Ubikwit  連絡 見学/迷惑 18:57, 6 April 2013 (UTC)
Actually linguistic reconstruction and linguistic relationships are not, and never will be, interdisciplinary. Linguistic knowledge is not genetic, it is not archeological, it is not cultural. Altaic is a dying hypothesis, whether you include Japonic and Korean or not. --Taivo (talk) 23:29, 6 April 2013 (UTC)
Linguistic knowledge itself is not, but if a historical linguist concludes that there must have been contacts between two languages before a certain time and archeologists, historians etc. can provide evidence to the contrary, it does matter, especially if the linguistic evidence in question is based on a rather small number of lexical items. G Purevdorj (talk) 03:41, 7 April 2013 (UTC)
As far as I know, linking archeological evidence to linguistic evidence is itself a somewhat controversial practice. CodeCat (talk) 13:13, 7 April 2013 (UTC)
Cultures evolve through interaction and cross-fertilization. The existence of loan words adopted in conjunction with some practice or product of material culture and the like are a simple demonstration of that, and one doesn't have to go to prehistoric periods to observe the phenomenon. Of course, that would be simple lexical evidence, of which there is little, I gather.
I would also imagine that population genetics will shed a little light on ancient migrations. For example, there is a substantial presence of the haplotype prevalent in Tibet in Japan. What we know about the history of people(s) that speak potentially related languages can help assess whether there is an association, the degree of association, etc.--Ubikwit  連絡 見学/迷惑 13:44, 7 April 2013 (UTC)
Haplotype, shamaplotype. There is no direct connection between language and DNA. Languages are learned behaviors, not inherited behaviors and always have been. Genetic data can be no more than secondary, circumstantial, minor support for a linguistic relationship argument. If the argument cannot be made entirely with linguistic data, then genetics is worthless. Same with archeological/cultural data--there is no direct connection between learned linguistic behavior and learned cultural behavior. The Pueblos of New Mexico and Arizona practice nearly identical cultural behaviors yet speak languages in four unrelated language families. --Taivo (talk) 19:00, 7 April 2013 (UTC)

────────────────────────────────────────────────────────────────────────────────────────────────────People carry more than their DNA with them when they migrate, they carry their language and cultural practices. This is a question related to transmission, adaptation, transformation, etc. If there is archaeological evidence of cultural transmission, then there is a higher probability of the presence of correlary confluences in linguistic factors as well. Hypothetically speaking, of course. You made an analogy about the Pueblos, but did not address the point with regard to whether or not there is any evidence of cultural transmission between those peoples, or if they were isolated, self-sustaining communities. I'm not familiar with their languages and cultures, so this is not simply a rhetorical point. Maybe they did manage to cross-fertilize culturally and maintain linguistic distinction. That is not necessarily the case in general, and would probably be a statistical anomaly.--Ubikwit  連絡 見学/迷惑 19:24, 7 April 2013 (UTC)

Look at the Pygmies of Africa as well as the Pueblos. They are so genetically distinct from their neighbors that they constitute one of the four earliest genetic lines of humans. Yet when the Bantus invaded central Africa a couple of millennia ago, the Pygmies completely abandoned whatever languages they originally spoke and learned the languages of their "conquerors". There is no trace of any "Pygmy" language, they all speak either Bantu or Ubangi languages closely related to or identical with the languages of their genetically distinct Bantu neighbors. Of course, the Pueblos borrowed culture from each other, that's the point--that culture is no more tied to DNA than language is. Their languages are in unrelated families while their cultures are nearly identical, meaning that culture was easily transferred across language boundaries. The languages spoken by the Pygmies were easily transferred across genetic boundaries. When looking at cultural and linguistic spread, even in preliterate times, it is almost never the case that underlying populations ceased to exist--they simply merged into the incoming population, learned the new language, and adopted the new culture. Therefore any attempt to tie genetic evidence with linguistic evidence is doomed to failure. It is marginally circumstantial at best. --Taivo (talk) 19:52, 7 April 2013 (UTC)
Just to add a couple more examples of genetically distinct, but linguistically uniform, groups--the Negrito groups of the Philippines, who all speak Austronesian languages related to their neighbors, and the Aslian groups of the Malay Peninsula, who all speak Austro-Asiatic languages related to their neighbors. --Taivo (talk) 02:22, 8 April 2013 (UTC)
Granted, that it is undoubtedly the case that there are examples of both models, especially in pre-literate societies.
However, a very modern example is illustrative of the dynamic. That example being Japan after the introduction of Western technology. There are two distinct modes of lexical representation related to the adaptation of Western technology:
  1. Neologisms rendered in Sino-Japanese compounds
  2. Loan words rendered in an approximation of their pronunciation in the language of the country from which the concept/object was introduced
In the first mode, they have deployed resources in the native lexicon to render the imported concept/object as a neologism. In the second mode, they have adopted words from a foreign language in conjunction with adopting the corresponding cultural practice/object.
Of course, modern examples are illustrative only with respect to the lexical aspect. The adaptation of syntactical elements would seem to be the crux of the matter, and that is something that will likely remain somewhat obscure.--Ubikwit  連絡 見学/迷惑 10:47, 8 April 2013 (UTC)
Your example has absolutely nothing whatsoever to do with linguistic reconstruction and the issue of Altaic. Are you actually a linguist? If you were, then you would know that your example means nothing for determining linguistic relationships between language families. --Taivo (talk) 15:33, 8 April 2013 (UTC)
I have studied some linguistics, but am not in academia at present, though I work as a translator. A recent focus of mine has been on the development of the Japanese writing system, so I am familiar with the full scope of current scholarship on that topic, which is, albeit, only peripherally related to the present article.
The issue at hand, however, seems not exclusively related to "reconstruction" or "relationships between language families", but whether specific languages are associated with a specific language family.
As mention by another editor above, there are Korean linguists that support a "genetic" connection between Tungusic and Korean, and given what I know about the history of the region, I would definitely support that stance.--Ubikwit  連絡 見学/迷惑 16:40, 8 April 2013 (UTC)
If you are not a linguist, and don't have a focus on the science of historical linguistics, then how can you support any stance that relies on linguistic data? This issue is 100% about linguistic reconstruction and relationships between language families. There is no other issue here whatsoever. If you don't understand the science that we're talking about, then I'm sorry but your point-of-view, in your own words, is "only peripherally related". But even using "peripherally" is to give more credit to "writing systems" than they are due. All the history we are talking about here is pre-literate and Japanese writing has absolutely nothing to contribute. --Taivo (talk) 18:17, 8 April 2013 (UTC)
I suppose it depends on your definition of linguist, if you are challenging my competence to address this issue at all. I have studied linguistics at the undergraduate level and have experience translating two of the languages at issue.
From your comment above, it seems to me that there are two parallel and interrelated issues, the first being whether there is such a language family called Altaic (which it appears you oppose), and second whether Korean and Japanese are included in that language family.
The issue of the origin of writing (both Korean and Japanese) is that it relates to the emergence from pre-literate to literate.
I have briefly looked at the sources, and am highly skeptical of the attempts to refute the existence of the Altaic language family, and to exclude Korean and Japanese from that language family. Perhaps the parameters for this article--until there is a definite consensus in academia--should be defined with a little more leeway.--Ubikwit  連絡 見学/迷惑 18:34, 8 April 2013 (UTC)
"Translating languages" is irrelevant for a discussion of historical linguistics and determining the possible interrelationships between languages, as is a history of writing in those languages, since the period of time when all this occurred was thousands of years before the adoption of writing in any of these languages. The majority of historical linguists do not accept Altaic as a valid linguistic family whether Korean and Japonic are included or not. Of course most of the references listed in the article are the minority which support Altaic. Listing all the sources which oppose such a grouping would be impractical. --Taivo (talk) 20:01, 8 April 2013 (UTC)

This is a forum-like discussion of editors' personal linguistic theories, which is not supposed to occur in Talk pages. If any of this is relevant to specific published sources that an editor proposes using for the article, please give those sources. Otherwise this section "Comment from ANI discussion" should be deleted.

No, this discussion is/was on-topic and useful, and to the point. Altaic is dying fast as a theory, even its original inventors have abandoned it for the most part, and we need to make sure that it is not given much credence any more in the language articles. HammerFilmFan (talk) 16:20, 7 February 2015 (UTC)

Describing the controversy, among other things[edit]

I'm not a linguist (my knowledge of the subject mainly comes from popular science articles, and Wikipedia), so I'm in no position to judge what the current scholarly consensus is. However, I might be able to suggest a better wording that gives a more balanced description of the range of theories. Would something like this be reasonable?

Altaic is a proposed language family that includes some or all of the Turkic, Mongolic, Tungusic, Koreanic, and Japonic languages. Several versions of the theory exist, which include or exclude different sub-families. In its original form (also known as "micro-Altaic"), the Altaic family comprises the Turkic, Mongolic, and Tungusic language families. An alternative version of the theory, known as "macro-Altaic" includes all the sub-families (Turkic, Mongolic, Tungusic, Koreanic, and Japonic). Unger (1990) proposes another grouping, consisting only of Japonic, Korean and Tungusic. The existance of an Altaic family (in whatever form) is itself controversial, and is not accepted by the majority of linguists, who argue that the similarities between the subfamilies are a result of areal interaction between the language groups concerned rather than common descent.

I think that looks reasonably balanced (once proper references are added). A few points that may need considering, though:

  • Regardles of how we rewrite it, the current first line of the lede probably needs to be changed, as it oversimplifies the cited source. (Current text: Altaic ... includes the Turkic, Mongolic, Tungusic, Koreanic, and Japonic languages. Cited paper: Altaic ... comprising the Turkic, Mongolic, Tungusic, Korean and, in most recent versions, Japanese languages).
  • I'm not sure the best way to describe the non-acceptance of the theory by the majority of linguists. Is this a fringe theory, or is it more like "35% accept it, 55% reject it, 10% haven't made up their mind)? Is it rejected just due to lack of evidence, or is there positive evidence against it? We will have to be careful about how we word it, to avoid sounding unduly accepting/dismissive of one side or the other. (In any case, this needs a better discussion in the main body of the article, including perhaps a brief description of any alternative theories. At the moment there is nothing really explaning why the majority reject it).
  • If Unger is the sole person to support a "Japonic, Korean and Tungusic" grouping, is it giving him undue weight to mention this in the lede?
  • Ideally, we should state whether current concensus favours the micro or macro version of the theory (or neither), but given the previous arguments, it this may be too complex an issue for the lede.

Iapetus (talk) 21:23, 9 July 2013 (UTC)

That all seems reasonable. No, it's not fringe; the debate is over whether the numerous lexical and grammatical commonalities are genetic or areal. That's often a very difficult question to answer. Starostin, for example, accepts Altaic, but using the same kind of analysis has recently come to the conclusion that Khoisan is bogus.
I've never heard of Altaic as Tungusic–Koreanic–Japonic; perhaps that could be relegated to 'other' or a footnote.
From previous discussions, it seems that those who accept Altaic today accept the macro version. — kwami (talk) 21:51, 9 July 2013 (UTC)
Majority, minority ... many linguists is simply careful not to support a theory that is controversial and that they don't know enough about. Most of the folks at WAFL take the opposite stance and accept Altaic without knowing the least about the discussion. A similar stance is taken in Mongolia where wishful thinking dictates that Poppe must have been right, with no regard for Starostin's considerable contributions or any evidence to the contrary. The discussion is not there to be decided, but to be closed. Then there are areal or language family specialists such as Johanson or myself who just sit around and wait to see some more evidence, usually with a somewhat pessimistic stance, or the subset of typologists that really care about whether language families exist before deciding on their sample. And ultimately there’s the small set of experts that actually take part in the discussion. Unger is an oddity: I’ve read the tiny summary article quoted in the article and listened to a much more recent paper of his in 2011, but he never bothered to present any evidence. Now I don’t know whether he made any other contributions, but if that is not the case, Unger’s position is as good as the personal opinion of any Wikipedian and should probably be erased from the article entirely. At any rate, I deleted it from the lead. G Purevdorj (talk) 02:37, 10 July 2013 (UTC)
I'm going to add some general comments here. Kwami, Taivo and CodeCat are all editors who are knowledgeable about historical linguistics and have contributed to a wide variety of articles. I would add myself to this category although obviously this is not an unbiased opinion. I don't recognize G Purevdorj but he/she seems to have a good deal of specialist knowledge about this area. User Wardog aka Iapetus appears to be a careful editor who (in his own view) is not a linguistics scholar but from my perspective has a good understanding of the way that scholarly controversies work in general and the proper way to write a Wikipedia article to convey these controversies.
There is a general problem with languages, dialects and language families in that many people have popular opinions and *think* they are competent to judge the scholarly reality when they're not. Unfortunately these popular viewpoints are often held very passionately because they underlie strongly-held views on ethnic identity, political legitimacy, etc. etc., and this passion means that many Wikipedia editors are more than willing to edit-war to get their way. In my opinion it's a minor miracle that articles on certain languages (e.g. Serbo-Croatian) have actually managed, for the most part, to stay in a state that does correctly reflect the scholarly consensus despite the enormous opposing ethnopolitical pressures. Kudos to Kwami and Taivo for their willingness to continue engaging these battles month after month, year after year, long after I would have thrown up my hands in disgust. (As an example, a few months ago I was engaged in a frustrating and demoralizing battle with a Silesian nationalist WP editor who fiercely insisted that Silesian is a separate language, rather than a Polish dialect with strong Czech influence, despite the almost total lack of scholarly sources favoring this position. This and a couple of other likeminded editors had managed to distort all the relevant articles (e.g. Slavic languages, Polish language, the misnamed Silesian language, Dialects of Polish, etc.) in favor of their viewpoint; in some cases these distortions had persisted for years. I eventually gave up monitoring these articles, and I see that some of the Silesian nationalist viewpoint has crept back in.)
In the case of this article, I haven't read through the relevant sources enough to be able to comment definitively, but it seems clear that (a) "micro-Altaic", and likewise "Ural-Altaic", are old ideas that have been sanctified through repetition in the popular sources but aren't accepted any more (to the extent that they ever were); "macro-Altaic" does have significant, but clearly minority, scholarly support.
Overall, I'm somewhat sympathetic to the idea, but it's very hard to evaluate how well-supported the grouping actually is given the "evidence" currently presented in the article. Tables like the vowel table in section 4.2.2 make me feel somewhat skeptical when I see e.g. that the outcome of Proto-Altaic (CaC)u (fifth line) is given as /a/, /o/, /u/ in Proto-Mongolic and /a/, /ə/, /o/, /u/ in Middle Korean. Either the context in which these putative outcomes occur needs to be identified (or at the very least, the dominant outcome(s) should be specially indicated), or the table should be deleted. Likewise for tables of vocabulary; e.g. in table 4.4.2, where /boːjn/, /moŋa-n/, /mje-k/ and /nəmpV/ ("neck", third line) are all claimed to be cognate, and to descend from a putative proto-form /móːjno/, what is the evidence for this? Are there solid rules that can be demonstrated showing how all these forms can be derived? And if so, how well-supported are these rules? As an example of what I'd like to see, take a look at a couple of tables I've created: "Possible derivation of some verbal forms" in Old Irish#Allomorphy, and the big table of the derivation of English "one" ... "seven" and "mother", "heart", "hear" in Old English#Sound changes. The latter table clearly shows, for example, how the PIE forms *kʷetwó:r, *pénkʷe, *septḿ, *h₂ḱousyónom yield the extremely different-looking modern words "four", "five", "seven", "hear". Other tables of varying sizes that I've created, which might be useful for reference, can be found e.g. in Old English phonology, Middle English phonology, Phonological history of French and History of the Slavic languages#Nasalization. Granted, some of these tables may be hard to read or overly large, but they show the sort of information that needs to be presented in order to properly demonstrate cognacy in languages like English, Irish and French, which all have notably complex sound changes, especially in their vowels -- and if macro-Altaic is correct, similarly complex changes must be involved in order to link the sets of words given in section 4.4.
There are also lots of other things that need to be documented -- e.g. why are reconstructed Old Chinese and Proto-Tibeto-Burman pronouns given in section 4.4.1, when Sino-Tibetan is a completely different family from any of the macro-Altaic families, and I don't know of any even remotely scholarly claims linking them?
Benwing (talk) 02:38, 31 July 2013 (UTC)

Proto-Altaic peoples didn't live in Eastern Europe[edit]

it's a Pan-Turkist propaganda. Wikipedia:Identifying reliable sources, [5], Talk:Paleolithic_Continuity_Theory#Horse_terminology Cantspans (talk) 09:58, 9 October 2014 (UTC)

(stirring the pot, but ..) Summary section?[edit]

Shouldn't there be a brief summary section at the end of this article that states that this proposed theory is pretty much discarded now (generally) by the majority of linguists? I think it would help for anyone investigating the subject, just so they walk away with the academic opinion that this is more or less a dead end. HammerFilmFan (talk) 05:58, 27 December 2014 (UTC)

Korean or Koreanic[edit]

There is no consensus on the question. This may be due to the small number of responses to the RFC. It may be necessary to redo this RFC and publicise it to a wider selection of editors. AlbinoFerret 17:51, 3 April 2015 (UTC)

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Why User:Kwamikagami wants to use the term 'Koreanic'? The Koreanic languages are just a proposal languages family. Majority of linguists regard Korean as a language isolate. -- (talk) 05:36, 6 February 2015 (UTC)

Because the extinct relatives of Korean are necessarily included in any language-family proposal. "Korean" excluded those extinct relatives, but "Koreanic" includes them. --JorisvS (talk) 09:27, 6 February 2015 (UTC)
Relationship between ancient languages which were spoken in Korea and 'contemporary' Korean is unconfirmed. It's still researching by linguistic and the Korean studies scholars. -- (talk) 11:36, 6 February 2015 (UTC)
  • Koreanic. Even if the ancient languages are excluded, some linguists are starting to recognize Jegu as a separate (but obviously closely related) language to Korean. Two languages make a language family, thus "Koreanic". --Taivo (talk) 15:20, 6 February 2015 (UTC)

Our article on Koreanic explains what JorisvS and Taivo said. It is therefore the more useful link. — kwami (talk) 17:00, 6 February 2015 (UTC)

  • Korean. Only Some linguists starting to recognise Jeju as a separate language. Others still recongnise jeju as a Korean dialect, especially South Korean linguists. -- (talk) 18:59, 6 February 2015 (UTC)
You're missing the point. Regardless of who recognizes what, our article on Koreanic languages covers the putative branch of Altaic that includes Korean. — kwami (talk) 20:02, 6 February 2015 (UTC)
  • Use "Language isolate" in infobox, cover the controversial classification in article text, as per normal.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  02:49, 25 February 2015 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Removing Altaic as the highest-level genetic link[edit]

There is an RfC [6] concerning whether to eliminate the automatic inclusion of "Altaic ?" as the highest classificatory node in the language infobox. --Taivo (talk) 19:32, 6 February 2015 (UTC)

The RfC has been moved to here. --Taivo (talk) 17:41, 7 February 2015 (UTC)


I have problems with the way the article's lead is written. The first line reads, "Altaic /ælˈteɪɨk/ is a proposed, but widely discredited, language family of central Eurasia." Yet we then are told, "The Altaic language families share numerous characteristics. The debate is over the origin of their similarities. One camp, often called the "Altaicists", views these similarities as arising from common descent from a proto-Altaic language spoken several thousand years ago. The other camp, often called the "anti-Altaicists", views these similarities as arising from areal interaction between the language groups concerned." See the problem? It presents the two "camps" as though they were about equally credible and avoids taking sides, thus contradicting the article's opening. It seems that adding the words "but widely discredited" was intended as an easy way to change the lead to make it clear that Altaic is now regarded as discredited, but more extensive changes will have to be made if conveying accurately that Altaic is now seen that way is the objective. FreeKnowledgeCreator (talk) 05:28, 8 February 2015 (UTC)

Good point. Altaic is, indeed, widely discredited as a genetic unit and its support is withering away. So the "balance" of the older lead needs to be adjusted to reflect the contemporary state of affairs. --Taivo (talk) 05:54, 8 February 2015 (UTC)
@Taivo: You added the phrase "widely discredited" in a recent revision of this page. Can you provide any reliable third-party sources to support these claims? Jarble (talk) 03:48, 1 March 2015 (UTC)
Here --Taivo (talk) 07:10, 1 March 2015 (UTC)
  • The status of Altaic theory is mentioned in the article on Britannica. Quoting the relevant section:

The majority of scholars today consider a genetic relationship between these languages to have been proved and hence regard the Altaic group as a language family, basing this conclusion not only on similarities in vocabulary and language structure but on well-established systematic sound correspondences as well. Nonetheless, some scholars continue to regard the relationship as a hypothesis yet to be proved, while yet others believe genetic relationship to be indemonstrable, given the available evidence. A small number of scholars reject the hypothesis, attributing similarities rather to borrowings and areal convergence.

I think that the widely discredited part needs to be toned down a bit in accordance with the latest research. (Britannica article is from 2013). --Ivan Štambuk (talk) 13:42, 1 March 2015 (UTC)

Britannica is not a reliable source for such information--it is a tertiary source. A linguistics source, such as the Campbell and Mixco one, is a secondary source and thus far more reliable. Britannica's comment that "the majority of scholars today consider a genetic have been proved" is so far from the truth as to be ludicrous. The majority of linguists think the exact opposite. I don't know who they got to write the Britannica article, but it is either 1) someone who isn't a linguist, or 2) one of the tiny number of historical linguists who are trying to push Altaic. Either way, that statement is utterly false. The Britannica article is not "the latest research". It is a tertiary source of dubious origin.
  • "While 'Altaic' is repeated in encyclopedias and handbooks most specialists in these languages no longer believe that the three traditional supposed Altaic groups, Turkic, Mongolian and Tungusic, are related." Lyle Campbell & Mauricio J. Mixco, A Glossary of Historical Linguistics (2007, University of Utah Press), pg. 7.
  • "Although apparently genetically separate from each other, Turkic and Mongolic are entities so intimately interconnected that it will never be possible to understand the one with the other." Claus Schönig, "Turko-Mongolic Relations," The Mongolic Languages (2003, Routledge), pg. 418.
  • András Róna-Tas ("The Reconstruction of Proto-Turkic and the Genetic Question," The Turkic Languages [1998, Routledge], pp. 67-80) is less definitive in his comments, but in no sense does he support Britannica's appraisal of a "majority".
  • Tore Janson, The History of Languages, An Introduction (2012, Oxford) doesn't even mention Altaic in the chapter on "The large language groups", although every firmly established large group is mentioned. (Implying, of course, against Britannica, that Altaic is not firmly established.)
  • P.H. Matthews, The Concise Oxford Dictionary of Linguistics (2007, second edition, Oxford) calls Altaic a "proposed family of languages", not an established one (such as Indo-European, which he calls simply "[A] family of languages").
  • "When cognates proved not to be valid, Altaic was abandoned, and the received view now is that Turkic, Mongolian, and Tungusic are unrelated." Johanna Nichols, Linguistic Diversity in Space and Time (1992, Chicago), pg. 4.
  • "Careful examination indicates that the established families, Turkic, Mongolian, and Tungusic, form a linguistic area (called Altaic)...Sufficient criteria have not been given that would justify talking of a genetic relationship here." R.M.W. Dixon, The Rise and Fall of Languages (1997, Cambridge), pg. 32.
  • Asya Pereltsvaig, Languages of the World, An Introduction (2012, Cambridge) has a good discussion of the Altaic hypothesis (pp. 211-216) and concludes "this selection of features does not provide good evidence for common descent" and "we can observe convergence rather than divergence between Turkic and Mongolic languages--a pattern than is easily explainable by borrowing and diffusion rather than common descent".
The ease with which I have found these comments and sources plainly demonstrates that Britannica is simply wrong on this point. (And to historical linguists, the names Campbell, Dixon, and Nichols carry a great deal of weight.) Britannica is never a reliable source for linguistics when it is contradicted by secondary linguistic sources (which are preferred by Wikipedia's reliable source standards). This isn't the first time that I've had to argue against Britannica's bad linguistics. --Taivo (talk) 14:10, 1 March 2015 (UTC)

The problems with your citations:

  • They are obsolete (Nichols 1992)
  • They do not explicitly support or refute Altaic (Schönig 2003, Janson 2002)
  • They do not describe the general status of acceptance of Altaic theory (as opposed to the personal opinion of the author). The Britannica article does just that. The Britannica article author is a scholar which you can easily check if you click on his name which is linked in the article. Tertiary source are OK to provide a bird's eye view on the status of a topic. Quoting from Wikipedia:No_original_research#Primary.2C_secondary_and_tertiary_sources:

    Policy: Reliable tertiary sources can be helpful in providing broad summaries of topics that involve many primary and secondary sources, and may be helpful in evaluating due weight, especially when primary or secondary sources contradict each other.

The problem with the wording widely disredited is that it's a stronger statement that your sources indicate. Many provide a safe margin (Schönig 2003: "apparently genetically separate"), and none indicate that it has been abandoned altogether. At best it remains proposed and "unproven", in a sense that there is a significant group of historical linguists that don't accept it. Anyways, I did a bit more research:

  • Altaic is a widely, though not universally, accepted language family.. [7] - Concise Encyclopedia of Languages of the World, E. K. Brown, Sarah Ogilvie, Elsevier, 2009
  • The article on Altaic languages in Encyclopedia of Language and Linguistics 2006 (a top-notch reference work I'm sure you agree) authored by L Johanson simply calls it: A common designation for the typologically related languages of the Turkic, Mongolic, and Tungusic families is ‘Altaic languages’; according to some scholars, this designation also includes Korean and Japanese. It then goes on to describe pros and cons of the theory. As for its status it says: There is no consensus as to whether the relatedness is proven, still unproven, or impossible.. Nowhere does it call it discredited or long-abandoned.

To sum it up: while it is not established, it is far from discredited. I suggest that the qualifier widely discredited be removed and replaced with something less strong, e.g. proposed or hypothetical. --Ivan Štambuk (talk) 15:49, 1 March 2015 (UTC)

Sorry, Ivan, but you are wrong and have mischaracterized my sources. Nichols is far from obsolete since there has been no widely accepted work done on Altaic since then. And even one of your sources, the Encyclopedia of Language of Linguistics, calls Turkic, Mongolic, and Tungusic "typologically related" and not genetically related. The fact is simply clear--other than a single overblown comment in Britannica that is not based on any actual fact, Altaic is overwhelmingly rejected as a genetic unit in the sources. At best, the sources state that there are a few specialists who accept it. And, yes, the facts of the matter are that Altaic has been "widely discredited". There have been no major supportive works published in decades. --Taivo (talk) 21:38, 1 March 2015 (UTC)
And I looked up Binnick's resume. He's not a historical linguist, he's a theoretical syntactician. He wrote a theoretical grammar of Mongolian. That is apparently why he was chosen to write about Altaic. Nichols, Dixon, and Campbell are historical linguists. All of them state explicitly that Altaic is not a genetic entity. All of the most reliable historical linguists state that Altaic does not exist. Binnick's statement in Britannica, that a majority of scholars consider Altaic to have been proved is utterly false. Statements by actual historical linguists prove otherwise. --Taivo (talk) 03:53, 2 March 2015 (UTC)
Just as a general issue, Taivo is correct that linguistic sources are preferable to general works of reference such as encyclopedias. FreeKnowledgeCreator (talk) 07:22, 2 March 2015 (UTC)

I was browsing Wikipedia and came across this discussion. Perhaps Altaic could be called a typological grouping in the lead, with clarification that this does not mean it's a genetic grouping, and description of how so-called Altaic languages came to share typological features (language contact, Sprachbund, whatever) — similar to the concept of Standard Average European, which is a typological grouping, not a genetic classification, since it contains Indo-European languages from several branches (Romance, Germanic, Slavic) that do not share a unique common ancestor. Just a thought, from an uninformed bystander. — Eru·tuon 17:42, 13 March 2015 (UTC)

I have problems also "widely discredited" is a very strong statement, and we usually avoid such statements in the lede. The lede is supposed to summarize the article, and the article, appropriately, gives the various views. They cannot be condensed into a single phrase with the necessary qualification. Some less wording can be used "often questioned" or "of disputable status" etc. DGG ( talk ) 19:12, 13 March 2015 (UTC)

The truth, DGG, is that Altaic has been widely discredited among historical linguists. It's not really debated anymore at all. It is, as Erutuon states above, almost universally considered to be a typological grouping among historical linguists. But viewing Altaic as a genetic unit is virtually a fringe position at this time. Erutuon's wording "a typological grouping" could be used rather than "widely discredited", but to imply that there is any creditable debate among specialists as to its genetic status is to misinform the reader. --Taivo (talk) 20:20, 13 March 2015 (UTC)
Yeah, Altaic's considered a dead fish now in most Historian circles, based on the dearth of support for it in the wider majority of linguistic scholars. (talk) 20:48, 13 March 2015 (UTC)

A few comments:

  • The Britannia article makes a distinction between Micro-Altaic (Turkic, Tungusic, Mongolic) and Macro-Altaic (Turkic, Tungusic, Mongolic, Koreanic, Japonic, and others), even though it doesn't explicitly use those terms, and states that the former is 'accepted by a majority of scholars' while the latter is still 'speculative.'
  • The Concise Encyclopedia of Languages of the World makes the exact same distinction, and again states that the former is widely accepted while the latter not so much.
  • Encyclopedia of Language and Linguistics does not make a stand either way.
  • Against the above sources, Micro-Altaic, as described in this very talk section, is *not* the default Altaicist position. The Altaicists listed under the pro-Altaic section, especially the ones that are still in the field, are practically all proponents of Macro-Altaic, not Micro-Altaic (Georg 1999), which has been rebranded by a subset of them as Transeurasian (Robbeets 2014).
  • Where, then, did the idea come re:Britannia and CELW that Micro-Altaic is widely accepted, but Macro-Altaic not? And who exactly is an authority with regards to "the majority of linguists?"
  • As with divisive topics in general, both sides are prone to writing as though their own views are correct, and to marginalize those who disagree. This is especially the case for this issue, where scholars are prone to declaring the premature death of theories and to "decide" for everyone else (Johanson 2010).
  • The standard WP policy with respect to such divisive topics is to not take a stand, but to instead present both sides. To this end, I do agree that "widely rejected" is too strong of a description, as it gives off the impression that only fringe scholars support Altaic/Transeurasian. I do not think that is a fair description of the situation today, and further, those who reject Altaic/Transeurasian as a *genetic* family still recognize its value as an areal-typological family. The header, which emphasizes that Altaic is a "proposed, but widely discredited, language family" ignores this second usage.
  • To this end, my personal opinion is that instead of making the article about a "widely discredited language family" as though we are talking about the history of a fringe theory instead of an area that is still producing a great deal of research in the form of Transeurasian studies, etc., greater emphasis ought to be given to the areal-typological connections. That is to say, rework the header such that it presents the viability of Altaic as an areal-typological family and then comment briefly about the controversy surrounding whether it is a genetic one. That is of greater value to the reader than a firm stand on whether Altaic is rejected as a genetic language family. Wikipedia is not the place to fight that battle.

Lathdrinor (talk) 23:19, 4 May 2015 (UTC)

Altaic is a dead issue in historical linguistic circles except as a typological construct, not as a genetic unit. Encyclopedias are often late to the party and don't reflect the most up-to-date opinions on the issue. Today, Altaic, whether Micro or Macro, is a dead issue as a genetic construct and is, indeed, now fringe. That's not to say that it didn't used to be more popular, especially when it was first proposed back in the '50s and expanded in the '60s. But just because you can find older sources that support it as a genetic unit doesn't make it currently the scientific consensus by the vast majority of specialists. Sure you can find references, but all your references are from general works that are not up-to-date. I've provided multiple sources from specialists in the field. Those are far more reliable than Encyclopedia Britannica (and always have been). The author of the Britannica article isn't even a historical linguist, he's a theoretical syntactician who wrote a theoretical syntax of Mongolian. Typical Britannica--find a non-specialist to write the article. That's not a reliable opinion from a historical linguist. Altaic is, indeed, "widely discredited" by specialists in the field. And each of their "votes" counts for ten of those of general linguists who are behind the times. --Taivo (talk) 03:04, 5 May 2015 (UTC)

I am aware that the theory is controversial, and I am familiar with the arguments against it. There are plenty of linguists supporting the Altaic theory, and there is a lot of ongoing research (Robbeets, the "Moscow school", Seoul National University). The article already says that it is a proposed theory. It has been proposed and supported, and for that reason it is relevant. No need to shoot it down in the first paragraph in my opinion.

I do not believe that consensus has been reached on the talk page for adding this bit at all. It's just you dismissing different users opposing inclusion of this in the lede. So I don't understand why the starting point should be to include a controversial statement, and when more than 6 users including myself oppose adding it, we are accused of "edit warring", whatever that means. (talk) 10:37, 1 October 2015 (UTC)

Altaic is not "controversial" as a genetic unit and only a tiny minority of contemporary linguists deny its rejected state. There is not "plenty of research" being conducted on the genetic unity of Altaic. There are only a tiny number of marginal linguists fighting against the tide in Moscow and Seoul. That does not constitute "plenty of research". All the major and most influential linguists in the fields of Turkic, Mongolic, Tungusic, and historical linguistics in general have rejected it. It's not controversial at all among the leading specialists in the field. That makes it "widely rejected" and the references I provided above demonstrate that completely. Some linguists even make the equivalent of that assertion concerning a summary of the field. For example, Nichols, as far back as 1992 (ref above) stated: "Altaic was abandoned, and the received view now is that Turkic, Mongolian, and Tungusic are unrelated". That's not a statement of "controversy". That's a statement of received fact. It is not against Wikipedia policy to state the simple fact that Altaic is, indeed, "widely discredited" when the most reliable sources themselves make that statement. "Altaic was abandoned" is pretty clear. On another page I likened the tiny number of linguists who are still trying to demonstrate Altaic as a genetic unit to the last passengers on the Titanic clinging to deck chairs. There is no reason whatsoever to change that metaphor. Sadly, many editors are unable to distinguish between plentiful research on the typological relationship between these languages (which is almost universally accepted) and the "widely discredited" genetic relationship that has virtually no historical research being conducted. --Taivo (talk) 14:07, 1 October 2015 (UTC)

Taivo, I'm sorry but there still is a mismatch between your assertion and the actual sources of the article though. The lead sentence, including its "widely discredited" judgment, is ostensibly sourced to the Georg et al (1999) paper. This paper, however, explicitly argues that Altaic "continues to be a viable proposal, despite various published claims that it is no longer accepted". The present state of the article is thus a case of blatant source falsification and needs to be fixed, pronto, one way or another. If you wish to have something along the lines of that "discredited" verdict included there, you need a reliable source demonstrating explicitly that the situation of the field has changed decisively in the anti-Altaic direction after Georg et al's state-of-the-art report was published. So far, I'm afraid that what we have here is only your loudly and frequently repeated assertions of your personal conviction that the Altaic position has a fringe status in current research, against several high quality sources that explicitly say that it does not. Fut.Perf. 21:19, 1 October 2015 (UTC)

I have added appropriate references (most postdate George) to the "widely discredited" statement. Your point that the (previously) only reference to the sentence seemed to contradict the comment was a valid one. The added references should correct that contradiction. The Georg article, however, is hardly the state of the art in Turkic and Mongolic studies, where the scholarship continues to virtually unanimously discredit Altaic. If the specialists in two of the three core "Altaic" groups discount Altaic, then scholars in peripheral areas can hardly be given more weight. That would be like trying to "build" Indo-European without the support of Latin, Greek, Sanskrit, Germanic, and Slavic scholars in support. --Taivo (talk) 21:53, 1 October 2015 (UTC)
Taivo, the sources you have presented do not, in fact, support the phrasing "widely discredited." I don't think anyone here has been suggesting that Altaic is actually a widely accepted language group, or that the article should say anything of the sort. The sources you've presented show, I think, that most scholars do not accept Altaic as a valid genetic group. But "widely discredited" seems to me much stronger than that, especially the use of "discredited." Perhaps there is another wording that could satisfy everybody? "proposed, but now mostly rejected," perhaps? john k (talk) 23:16, 1 October 2015 (UTC)
"Proposed" is unnecessary. The fact that we have an article on a discredited idea automatically means that it was "proposed" at one time. I would accept "mostly rejected", since that means, in essence, the same thing as "widely discredited". The sources, while not using the very words "widely discredited" are crystal clear that modern historical linguists, as well as specialists in both Turkic and Mongolic languages, have largely rejected Altaic as a genetic concept. So I'd be happy with "mostly rejected". --Taivo (talk) 01:17, 2 October 2015 (UTC)
I am also fine with "mostly rejected". To the pedant in me, it's not quite the same as "widely discredited", but is an acceptable compromise that maintains an accurate representation of the (modern, per Taivo) sources fitting for the lede.--William Thweatt TalkContribs 03:00, 2 October 2015 (UTC)
"Mostly rejected" is fine with me as well.--Ymblanter (talk) 07:32, 2 October 2015 (UTC)

Neither is fine with me. The status of the theory is not "conclusively disproved". Rather, it could not be conclusively proved so far, and the research is ongoing. This is similar to the status of other proposed language families, Na-Dene-Yeniseian, Na-Dene itself (whether it includes Haida or not), Afro-Asiatic in its broadest sense (whether it includes Omotic or not), and so on. This is sufficiently summed up by saying "proposed" in the first sentence. "Widely discredited" is a loaded statement, clearly being pushed by someone with an agenda, not reflecting a "neutral point of view". (talk) 07:49, 2 October 2015 (UTC)

"Telling General Linguists about Altaic" (1999), written i.a. by one of the staunches critics of the Altaic theory, Georg, sums up the status of it pretty well, and certainly more objectively. (talk) 07:56, 2 October 2015 (UTC)

You obviously didn't even look at the article. Georg is only one of four authors and the other three authors are either staunch supporters of Altaic or are "lumpers" in general while specializing in other language families. It is impossible to tell which parts were written by Georg and which parts were written by the three other authors. --Taivo (talk) 19:17, 2 October 2015 (UTC)
Taivo, the combined authorship of that paper makes it all the more valuable for our purposes here. The sentence I cited above ("continues to be a viable proposal") is right at the top of the article's summary, so it is obviously jointly representative of the opinions of all co-authors. (Disclaimer: I remember reading the whole article years ago, but don't have access to the full content right now.) The very fact that authors from both sides of the debate wrote such an article together proves one thing: that all of them are still taking each other's positions seriously. That is the very opposite of "discredited". Something similar goes for the Johanson article you wanted to discount by saying that Johanson himself starts off by calling Altaic merely a typological grouping. The very fact that this article was written by somebody who doesn't subscribe to genetic unity gives all the more weight to the fact that he then goes on to present the genetic question as entirely open. What he says is that there is no consensus either way, which is dramatically different from your claim that one position is the near-universal consensus and the other fringe. As for the Nichols book, it may be worth noting that the book is not otherwise centrally concerned with the Altaic question and that the remark about Altaic in the introduction references only one previous authority, a conference report by Unger (1990). That author himself later noted that he felt Nichols was misrepresenting his position [8] and that the panel he was reporting on did not express any opinion along the lines claimed by Nichols, that "the received view now is that Turkic, Mongolian and Tungusic are unrelated". Fut.Perf. 06:44, 3 October 2015 (UTC)

As an uninvolved noncombatant, I recommend that someone propose two alternative wordings and ask for votes on each. From a skim of the commentary, one possibility is something like "Altaic is a typologically related group of languages that some linguists claim are also genetically related." In any event, please stop the edit warring. It's gross. Lfstevens (talk) 18:56, 2 October 2015 (UTC)

I have read the article, and it is out of order for you to suggest I "didn't even look at it". The first three pages sum up the status of Altaic in a way that both sides of the argument could agree on, which is very much the opposite of what you are doing. You have defaced a decent enough article with a controversial and biased statement in the first sentence. Multiple users objected to it, or tried to revert it, or both, you can't be bothered to listen. I'm directing this at Taivo BTW, not Lfstevens. Your efforts at mediation are noble, but do not address the real problem of someone pushing a biased view, and blocking anyone with a different view with determination worth of a better cause. Is this really what Wikipedia has become? (talk) 21:58, 2 October 2015 (UTC)

While claiming that "multiple users" objected to the wording, you conveniently ignored the fact that multiple user supported the wording (and still do). You also claim that my edit was "defacing". I warn you to observe WP:AGF. And the fact that you didn't even know the authorship of the article you were referencing is usually pretty good evidence that you haven't actually looked at it. Since you did apparently look at it, you seem to have failed to notice its actual authorship. I also point out to you that the article was published 16 years ago (and written 18 years ago) while the majority of my sources are from after that date. While Georg et al. may have been "state of the art" in 1999, it can hardly be considered such when there are multiple reliable sources after that date that have rejected Altaic. --Taivo (talk) 00:07, 3 October 2015 (UTC)
It is absolutely inapproppriate for to say that only one user supports the current wording whereas their own edits were reverted by three different users. In fact, since they claimed yesterday they are new to Wikipedia, and today they started wiki-advocacy, they are most likely a sock of a blocked user (no idea which one).--Ymblanter (talk) 02:44, 3 October 2015 (UTC)

You don't have any idea about much at all. Wiki-advocacy? If anything I've expressed how utterly unimpressed I am by all this, and certainly not "advocated" anything. "[T]hey are most likely a sock of a blocked user" - is that your take on "assumption of good faith"? (talk) 10:47, 3 October 2015 (UTC)

This is all quite bizarre. "Widely discredited" doesn't even make sense, it's like saying something is widely disproven or a couple is widely married. Second Taivo selectively quotes Nichols, Dixon, and Campbell who are well known for their skepticism of historical linguistics in general, with Dixon being excoriated in Bowern and Koch for his rejection of genetic linguistics as it applies to the Australian languages; Campbell, who is notorious for advocating that all long range classifications be "shouted down"; and Nichols who proposes replacing genetic linguistics on a long-range basis with a sort of typological numerology.
The b-/m- suppletion pattern alone in the nominative and oblique first person pronouns is a unique shared innovation of the group. Plenty of historical linguists recent and current support the theory. The lead should indeed point out that a vocal minority questions or challenges the validity of the family. But under no circumstances should we be calling the theory "widely discredited". μηδείς (talk) 01:41, 3 October 2015 (UTC)
Medeis, a single pronominal pattern is not convincing evidence for genetic relationship unless you're willing to accept Greenberg's "Amerind" as a valid genetic unit as well. Pronominal patterns can be borrowed in a Sprachbund just as phonological patterns and lexicon can be borrowed. And your claim that skeptics are a "minority" is simply laughable. Even the Georg et al. article doesn't make that unsubstantiated claim. At best, at the very best, Altaic has some support that is more than fringe, but it is hardly proven and as long as a significant body of reliable scholars (whether "noteworthy" skeptics or not) discount it as a genetic unit (its typological validity is not in question), then it should not be treated as a valid node. Ethnologue, Glottolog, and Linguasphere have all abandoned it as a top-level node. And while I'm willing to compromise on the "widely discredited" wording, it is still technically true--a wide range of reliable scholars have discredited the genetic unity of Altaic. --Taivo (talk) 08:19, 3 October 2015 (UTC)

@Taivo: "I warn you to observe WP:AGF" - do you want to warn yourself as well while you're at it? "You didn't even know the authorship of the article you were referencing" - that's hardly "assumption of good faith", is it? My point exactly was that it was written by both proponents and opponents of Altaic, and at least attempts to present a view that is acceptable to both. Which is the exact opposite of what you are doing here. You are pushing one view over and over again, with arguments like "my sources are better than yours". I don't think they are, and there clearly is no consensus on this page for inclusion of this phrase. Anyone apart from you can see this.

I have already expressed a willingness to compromise. You, however, don't seem to understand that reliable sources that were written after your only source contradict that source's assertions. Yes, my sources are, indeed, better than yours in this respect. And the influence of the Altaic camp is declining in general linguistics as evidenced by the most modern classifications of the world's languages in Glottolog (2015), which has never used Altaic as a top-level genetic grouping; Ethnologue (18th ed, 2015), which used to use Altaic, but has dropped it in the most recent editions; and Linguasphere (1999/2000), which never used it. --Taivo (talk) 11:30, 3 October 2015 (UTC)

I agree that the case against Altaic is strong, and supported by a formidable number of sources old and new (Doerfer, Georg, Vovin). Altaicists clearly still have an awful lot of work ahead of them (hence "proposed", and "not being widely accepted as a greater language family" already in the lede). But as Fut.Perf. has pointed out, "my only source" is a good basis for a consensus solution to this debate, for the simple reason that it was written by both proponents and opponents of the theory, and the summary attempts to present its status in a way that is acceptable to both sides. Which is exactly what we are trying to do here, right?

Setting aside the debate on whether I've read it or not, or whether I know who's written it, would you care to look at the first three pages and see if you would be willing to take it into consideration when we try to work out a consensus here? The article can be accessed here: (talk) 12:08, 3 October 2015 (UTC)

The problem with relying solely upon a 20-year-old appraisal primarily written by proponents of Altaic is that it ignores the fact that support for Altaic during the subsequent 20 years has not increased. (I would argue that it has shrunk, although I'm sure you would not agree with that appraisal.) I have read the entire article and not just the first three pages. While it attempts to paint a neutral picture, and comes closer to succeeding than most that attempt the same synthesis, there are still clear hints of personal attacks on skeptics that dismiss their entire body of work rather than describing the legitimate methodological reasons why they are skeptics in the first place. The arguments of major scholars are thereby summarized with the throwaway line "they are always skeptics" without saying why Altaic does not meet their high standards for evidence. It's certainly not a perfect article although it at least tries to be neutral. Propose a lead sentence. I'm sure that it won't be acceptable on first draft, but we can see if it's got enough to work into shape. --Taivo (talk) 15:18, 3 October 2015 (UTC)

"I'm sure that it won't be acceptable on first draft, but we can see if it's got enough to work into shape." - I hope you don't mind me asking, who the fuck do you think you are? Is that you making an effort to come to an agreement, by patronising someone? That was a rhetorical question BTW, I think I'm done with this. I do appreciate the insight this sorry experience gave me into how WP is written, at least now I know to stay away from it. Over and out. (talk) 16:15, 4 October 2015 (UTC)

The wording "widely seen as discredited" is too strong and inappropriate for an encyclopedia.--Jack Upland (talk) 03:48, 3 December 2015 (UTC)
It's not "too strong" when it's accurate. "While 'Altaic' is repeated in encyclopedias and handbooks most specialists in these languages no longer believe that the three traditional supposed Altaic groups, Turkic, Mongolian and Tungusic, are related." Lyle Campbell & Mauricio J. Mixco, A Glossary of Historical Linguistics (2007, University of Utah Press), pg. 7. --Taivo (talk) 04:00, 3 December 2015 (UTC)

Citation Style section 2[edit]

I want to comment on the revision made by Taivolinguist on 1 June 2013 (at 11:16). I mention this to you because you seem to be active in writing for the article, and in maintaining your stance, to the letter. If you, sir or madam, think it is proper citation style to use German (the original language), you ought to put the Title in the bibliography, in German. That is the right place when you have a sequence of Latin that is (as I believe) cluttered. Really, proper citation style is not so clearly agreed upon on Wikipedia that it does not require application of the principles from the editors. I don't believe you read the section the editors wrote on citation, at the beginning of the talk page. Or, perhaps you did. In any case, they are right.

I really believe that my "edit" would make the article clearer, and that the German words of the title are quite translatable. The article is in English, right? You can translate that title without losing any subtleties of the German. Einführung doesn't have a particular connotation of German dictatorship, nor is there any other reason for multiplying words in the article. Let the bibliography be in German, and the article in English, then. Let us follow the course that the editors outlined across the whole of the article; and put original titles in a separate bibliography. That way, people who are interested in learning about Altaic can read and understand the main article, and those who want to check the learned sources of that article can read it, and then check it. That is the reason I think we should follow this course of action. I do not blame you for wanting to show what you know, but knowledge of Latin and German is hardly essential to expressing yourself, or the consensus of editors, on the subject of Altaic. Dsnow75 [[User Talk: Dsnow75|Talk]] (talk) 21:10, 11 September 2015 (UTC)

I now have noticed that the citation style that the editors advocated has generally already been made uniform. So I do not mean to edit for the sake of editing; it's already been done. I have still continued to straighten out the article, though, as I wish you will have noticed. Dsnow75 [[User Talk: Dsnow75|Talk]] (talk) 19:06, 14 September 2015 (UTC)

I don't know what the hell you are going on and on about. Perhaps you should add a link to the edit (from two years ago) that you object to. Do you actually think that I remember every one of thousands of edits that I make every year? --Taivo (talk) 19:19, 14 September 2015 (UTC)

I have read your note. Notice the date of June 1, 2013 at 11:16 AM, look up the history of the main article, and you will find what I am talking about. Yours sincerely, Dsnow75 [[User Talk: Dsnow75|Talk]] (talk) 05:11, 25 September 2015 (UTC)

Place a link to the edit here. Asking other editors to scroll back through the history means that no one is going to look at it and they will ignore you. --Taivo (talk) 06:38, 25 September 2015 (UTC)


Many of the proponents of Macro-Altaic have recently turned to calling the family Transeurasian, eg, so as to disassociate it with the Altai mountains, which the family was initially believed to have come from, but is now no longer popularly believed to have come from. Should this be included in the article? Lathdrinor (talk) 00:44, 9 October 2015 (UTC)

Dixon, Nichols and Campbell?[edit]

Is that the best we can do? The widely discredited claim in the lead is synthesis, based on a combination of those cherry-picked sources. Greenberg, Starostin, and numerous other noted linguists, including Poppe hold Altaic as well demonstrated, simply by looking at the unique suppletive 1st person singualr paradigm. Dixon and Nichols are known for their hostility to the comparative method and claims that it just doesn't work except among Indo-Europeans. In his old field of Australian linguistics, he's widely excoriated. Campbell is notorious for calling upon any long-range theories to be shouted down. Vovin is the one to look to for a respected scholar who switched to the anti-altaicist camp, after Starostin's flawed dictionary was published. But there are plenty of others who hold to macroaltaic as uncontroversial. μηδείς (talk) 05:17, 3 December 2015 (UTC)

I return your question right back at you: "Is that the best you can do?" Greenberg, Starostin, and Poppe are all dead. Almost all of my sources debunking Altaic (above) are still alive. That alone should demonstrate the problem you have with trying to claim that Altaic as a genetic unit is a living hypothesis. If not 100% dead, it is on its dying legs. It is, indeed, "widely discredited" by living specialists in current work. --Taivo (talk) 08:03, 3 December 2015 (UTC)
When I listed the above bibliography, I included just those works on my shelf. But there are things which I don't own, of course:
  • Alexander Vovin, "Northeastern and Central Asia: “Altaic” linguistic history," The Encyclopedia of Global Human Migration, Ed. Immanuel Ness (2013, Blackwell Publishing Ltd.). Pages 1-7.
  • Alexander Vovin. (2005) "The end of the Altaic controversy," Central Asiatic Journal 49(1), 71–132.
--Taivo (talk) 15:15, 3 December 2015 (UTC)
The biological death of leading Altaic proponents is not equivalent to the death of their supporters and students. The Transeurasian school of Martine Robbeets and Lars Johanson is still quite alive, as is Anna Dybo, albeit I do agree that they have much less "clout" than the aforementioned dead Altaicists. Also, they appear to now include works from sub-theorists such as James M. Unger, who continues to argue for his thesis of Tungusic-Japonic-Koreanic in "Shared Grammaticalization: With Special Focus on the Transeurasian Languages" (2013), and John Whitman, who to my knowledge has never retracted his support. These are all new works, so can hardly be disregarded as the writings of dead Altaicists. Lathdrinor (talk) 03:27, 9 December 2015 (UTC)
But Tungusic-Japonic-Koreanic is not Altaic. And, as you have said, the influence of the "living generation" has declined to the point of near non-existence. The great majority of historical linguists, and all the influential ones, have placed it in the dustbin of failed theories. This is not to disparage the linguists who proposed it in the beginning. Many highly respected linguists in the 50s and 60s were lumping things together left and right. But today's historical linguists are far more cautious and demand a more substantial level of evidence. And just because there are still people struggling to prove something that has been rejected by the majority of historical linguists doesn't change the truth of "widely discredited". And a great many of the "new works" may use the term "Transeurasian" or "Altaic", but if you read the abstracts closely enough you will see that "areal" is a prominent term alongside "genetic". --Taivo (talk) 03:49, 9 December 2015 (UTC)

The inclusion of Turkic, Mongolic and Tungus should be reinstated.[edit]

I think doubt of the family came about with the inclusion of Korean and Japanese, but Turkic, Mongolic and Tungus are established relatives based on their morphology and syntax. We should look through more recent sources establishing their relationships. There are also some sources out there (at least one book) arguing the Uralic language family does not exist, but I think most sources affirm it.-NadirAli نادر علی (talk) 04:22, 26 August 2016 (UTC)

The doubt about the existence of Altaic is based on more than just the inclusion of Japonic and Korean. It is not widely accepted even when it's just Turkic, Mongolic, and Tungusic. --Taivo (talk) 05:32, 26 August 2016 (UTC)