Talk:Altaic languages

From Wikipedia, the free encyclopedia

This is the talk page for discussing improvements to the Altaic languages article.
Article policies
This article is within the scope of the WikiProject Languages, an attempt at creating a standardized, informative, and easy-to-use resource about languages. If you would like to participate, please visit the project page, where you can join the project and see a list of open tasks.
language Portal
Start This article has been rated as start-Class on the quality scale.
Other languages WikiProject Echo has identified Altaic languages as a foreign language featured article. You may be able to improve this article with information from the Russian or German language Wikipedias.

Contents

[edit] Just A Few Suggestions

  • David, some notes on the consonant/vocalic correspondences don't use the SUP commands and look weird.
  • All notes (especially on phon. correspondences) could be clickable (I can make them clickable if you wish)
  • As for the morphological correspondences, I would use either of the following strategies (for better lucidity):
  • Put the Altaic proto-morpheme into one column, and its meaning into another column
  • You could do the same thing for daughter proto-languages, but that would result in too many columns, hence:
  • Those morphemes which differ in function from that reconstructed for the Proto-Altaic, could use notes (instead of bracketed explanations).
  • The same for the cognate tables. Also, if there are more forms, I would choose only one, moving the rest to the notes. You can make a separate column for comments and explain the superscript numbers therein.
  • Sometimes you could use the < BR > in the tables, too...

--Pet'usek [petr dot hrubis at gmail dot com] 18:35, 9 April 2007 (UTC)

  • This article is very long with far to much abtruse detail. It should be cut back dramatically to work as an encyclopaedia article. Remember that the primary purpose of an encyclopaedia is to impart succinct, informed and accurate information to people who may not anything about the subject. This article reads more like an essay intended to be read by experts and, judging by the discussion, a controversial one at that.

87.80.9.63 (talk) 04:52, 31 January 2008 (UTC)

[edit] Comments

You guys really need to get your facts straight, this is one the reasons why Wikipedia will never become a 100% trusted source of information. For one the theory that Japanese is part of the Altaic family is a theory that died out among linguists in the early 1900s, the most accepted theory is that Japanese is in the Japonic language family and there is NO linguist out there worth his weight that claims Japonic to be a part of Altaic. Please look up the facts before spattering them online, thank you.

This is a purely amateur opinion.

Linguists may not have proven a genetic relationship of Japanese or Korean, but the syntax is so similar I was able to learn it quickly simply by plugging the words into the same place they would be in Japanese (a language I have written 7 books in). If the relationship is only a "sprachbund," I wonder how it is so close a fit structurally speaking. Could we have genetic relationships which involve more than one parent and include some borrowing by contact so that this is not an either/or matter?

In respect to the non/existance of a Altaic language family, an anecdote:

In an Indian grocers in London I heard some men talking at work in the back of the store and asked if they spoke Tamil. They said ""yes," do you know Tamil?" I said, "No, this is the first time I have ever heard it."

Yes, I knew it because Tamil sounded just like Japanese (Niigata prefecture, to be exact). I had heard Tamil was Ural-altraic and what do you know!

robin d. gill "Rise, Ye Sea Slugs!"


I m not professional linguistic either, but so far what I know, similar syntax is not very strong proof for genetic realtonship between languages. Also Tamil is not Ural-Altaic, but Dravidian language. I think this article is still slightly POV. Altought Ural-Altaic hypothesis has strong supporters, majority of the linguistics do not favor this idea. I will do some editing and if someone has objections, please bring them up here.--Kulkuri 09:29, 5 Jul 2004 (UTC)
I strongly disagree with the claim that Altaic is mainly based on similar syntax and typology; it is practically a slander on the many linguists who have put together collections of Altaic roots, and indeed reconstructions, one of which I have linked to below. While I agree that most linguists probably don't favor the idea, I am not at all convinced that that applies to the majority of historical linguists, or of linguists working on these languages. The case against Altaic needs to be made too, but not by attacking claims that Altaicists haven't made. - Mustafaa 20:24, 5 Jul 2004 (UTC)

Thanks for edits and feedback. Especially history of altaic theory is useful. I will do some editing later, perhaps splitting article in two paragraphs (pro and against altaist theory). Right now I will take few weeks holiday from wikipedia and other duties.--Kulkuri 07:22, 7 Jul 2004 (UTC)

That sounds great - I'll be interested to see what you come up with! - Mustafaa 07:29, 7 Jul 2004 (UTC)

[edit] reworked opening paragraph

I have, er, "improved" the opening paragraph. I felt that "controversial" is best used for something whose existence is accepted but whose importance or value is debated, and have replaced it with "putative", which I believe better indicates that some people just don't believe in it. Likewise I have rephrased other points for consistency with this state. I do not claim that the original was insufficiently NPOV, and I do not believe I have increased or decreased the level of scepticism within the article, but rather used terms more consistently conditional.

Also I've reworked the bullet list into a sentence, the better to increase symmetry with the latter items, rather than have three items bulleted and three not.

Sharkford 22:12, 2004 Oct 20 (UTC)

[edit] A recent anon edit on Ural-Altaic languages

...which I rejected as off-topic there, but is more relevant here, and informative, if highly POV. - Mustafaa 04:09, 21 Jan 2005 (UTC)

There are two main schools of thought about the Altaic theory. One is that the proposed constituent language families (Turkic, Mongolic, and Tungusic in the basic theory; with the addition of Korean and Japanese in extended 'Macro' versions) are genetically or 'divergently' related by descent from a common ancestor, 'Proto-Altaic'. The other school rejects this theory (so it is often called the 'Anti-Altaic' school) and argues that the member languages are related by convergence (mainly loan influence). The adherents of the divergence theory reject the criticism of the 'convergence' theorists, but they have not been able to overcome some flaws pointed out by the critics that essentially disprove the 'divergence', or traditional 'Altaic' theory. The continued popularity of 'Altaic' nevertheless exceeds that of the Flat Earth theory. Some adherents of the theory note strong similarities in the pronouns and other elements of the proposed members of the family and argue that the languages may even be related through a larger family such as Nostratic, but this is popular linguistics and wishful thinking that distorts or ignores the scientific principles of historical-comparative linguistics. The 'Altaic' and 'Ural-Altaic' theories are based on typology and loanwords. Both Uralic and 'Altaic' languages do follow the principle of vowel harmony, are agglutinative (stringing suffixes, prefixes or both onto a single root) and lack grammatical gender (see noun case). However, this is not evidence of genetic relationship. Unfortunately, genetic (divergence) theories are much more popular than convergence theories. No scholar of the convergence school has written a thorough monograph on the debate, citing the scattered, rather obscure scientific literature and examples.

Retrieved from "http://en.wikipedia.org/wiki/Ural-Altaic_languages"

[edit] References are a bit out of date, I'm afraid

The citation of Comrie (1981) is out of date and may be misleading. Comrie seemed to follow Greenberg and Ruhlen in believing that Altaic was too narrow a group. In any case, in The Atlas of Languages, rev. ed., (2003), Comrie et al. include Altaic as a language family. See page 38, where Altaic, Uralic, and Indo-European are postulated as part of the larger Eurasiatic group. For more, see pages 46-49, which shows that Comrie et al. clearly accept Altaic but are not sure about the affiliation of Korean and Japanese. This affiliation is strongly advocated by Miller (1995) in Languages and History, an excellent summary of the Altaicist position, showing its basis in phonology and morphology (sounds and words) rather than relying on syntax and loanwords.

Most opponents of the "Altaic hypothesis" apparently are specialists in Indo-European. See Winfred P. Lehmann (1962/1997) p. 86 but compare to p. 87. Lehmann says "such proposed relationships...are not based on evidence obstained through the comparative method..." which suggests he has not read Miller. Roger Lass (1997) actually seems quite upset that anyone would propose "Altaic", but he singles out Greenberg and Ruhlen in Scientific American (1992) and R. Wright in Atlantic Monthly (1991) for scolding. He makes no mention of Miller. However, both Lass and Lehmann refer to Lyle Campbell's article in Philip Baldi (ed.) Linguistic Change and Reconstruction Methodology (1990). I haven't had a chance to read this, because I live in Taiwan and don't have access to a good library; I will, however, try to read it this summer when I am back home (USA).

Taking the "pro" side, Aleksandra Steinbergs in an article in Grady et al. (1996) claims there is "substantial evidence that Korean and Japanese are also members of the Altaic family." (pp. 396-7).

Finally, when I clicked on Roy Miller, there was no article. When I checked the web, there were so many entries under that name that it was difficult to sort for the right one. I propose using his full name, Roy Andrew Miller, which narrows the web search considerably (but still leaves some sorting to do). I will shortly prepare an article on Miller for Wikipedia; the problem is finding out where he is and what he's doing now. Any suggestions? Retired in Hawaii? Teaching in Norway?

As for the discussion of affiliations, I worry about the problem of loanwords coming into play. Both Japanese and Korean borrowed heavily from Chinese, especially nouns and verbs. Anyone who has studied Japanese can see remnants of this in the use of kanji; kana are used to show inflections and postpositions, making the division clear. Korean is not so clear, but as a student of Chinese I often use the loanwords to figure out the meaning of Korean sentences. The fact that Japanese and Korean have very similar patterns (syntax) is also interesting. Postpositions in both languages follow very similar rules and even sound the same. This, of course, may be due to borrowing. Both Korean and Japanese may have had different syntaxes long ago, the way Old English did. Without records, we cannot be sure. This is what makes Miller's work on reconstruction so informative and helpful.

--Monty 02:34, 6 Apr 2005 (UTC)

Thank you for these thought-provoking points! It would be great if you could edit them into the article. - Mustafaa 20:13, 6 Apr 2005 (UTC)
I made a few minor modifications to the "controversy" part of the text. Monty 23:51, 19 Jun 2005 (UTC)Monty

The footnote given after the parenthetical reference regarding Dr. Vovin's change of opinion does not lead the reader anywhere. The statement is accurate, however.

[edit] Updating information

I will tackle the editing shortly. I'm also waiting for a journal article and a recent book on the subject to arrive. Monty--163.28.81.2 08:54, 18 May 2005 (UTC)

[edit] Why are the PIE experts so hard to convince that Altaic family exists?

This is the crux of the problem. Staunchest opponents of the Altaic theory do not know the languages involved. Turkish and Mongolian are so similar that dismissing a genetic relationship is like denying the genetic relationship of identical twins. On the other hand, the same experts lower the bar so low so that the extinct Anatolian languages can be admitted to IE. It is hard not to see the racist undertones in this general approach. Especially when "white" people speak a non-indo-european language there just isn't any interest in their languages and experts quickly annnounce that these languages are language isolates, suggesting that academic interest in such languages would not be a good career move. And about the non-white people's languages, well, frankly, these experts' time is too valuable to waste on such things.

Wikipedia could be most effective in such controversial topics by listing the facts and creating an enviroment where new knowledge is created. Turning Wikipedia an imitation of Encyclopedia Britannica by repeating what experts say is not doing justice to the power of this interactive medium.

By the way "A Board of Trustees" for Wikipedia is really a bad idea.

Wait a minute. What you say about dismissing the relationship of Turkish and Mongolian is very true, and then you go on a tangent and make some icky comments that I feel the need to oppose.

Admitting the extinct Anatolian languages into IE can't seriously be called "lowering the bar" in any context. In fact, Anatolian is providing us with a wealth of new insight into Proto-IE. Hittite and the Indo-European Verb by Jay H. Jasanoff is still blowing my mind. Believe me, the inclusion of Anatolian raised the bar for IE.

Second of all, any educated IEist worth anything understands that Indo-European languages aren't quote-unquote "white languages". Indo-European covers everything from Pashto to Swedish. That's a pretty broad range of human skintones if you ask me! Any true racist simply couldn't be a recognizable IEist nowdays without having a major sense of conflict with known facts. They would inevitably become some nutty maverick off to the side claiming that Sweden is the true homeland of the IE or some crazy thing. And certainly Altaic et alia are not 'non-white languages' unless you can somehow explain what 'white' even means. Again, Altaic-speaking peoples have various physiologies and skintones. Let go of the "us-versus-them" mentality. IEists aren't out to destroy Altaic studies ;) However, perhaps if there is racism, it involves what the general masses are spending their time on. They will tend to gravitate to IE because there is so much more written about it, while Uralic and Altaic gets put to the side. And of course, there are those hardcore racists drawn to IE because of its unfortunate spinoff theories by the NAZIs and the bible-based Japhetic nonsense otherwise long since buried in IE Studies History.

But I'd bet that the obsession over IE instead of Uralic or Altaic is not so much because of any racist intentions on anyone's part so much as the fact that IE languages enjoy a longer written history than most other language groups. Semitic also has a long written history and much is known about that group and Proto-Semitic. Altaic does not have such a long written history.

Thirdly, a board of trustees idea is what I was thinking of too. I like the name :) It seems like a natural evolution of things. And I support your idea that new knowledge could be created. Perhaps people could be organized into 'organized projects' of sorts. Not just these simple "entry projects". It would really band people together towards various causes or interests. It would be a step up from this plain ol' Wikipedia thing ;) --Glengordon01 05:51, 8 July 2006 (UTC)

[edit] Are two separate disputes being had here?

It seems to me that wikipedia is oddly dubious of the Altaic languages, to an extent that is not true for similarly controversial language groups like, say, Nilo-Saharan. For instance, Category:Altaic languages is a member of Category:Proposed language families, which warns us to be careful of using it. Altaic is surely not a proposed language family in the same way as the Nostratic languages, or even the Elamo-Dravidian languages. Altaic has been accepted as a language family for a long time, and is generally listed as such in standard reference works. It seems as though a much more controversial issue - "should Korean and Japanese be included in Altaic?" - is being confused with the question - "should Turkic, Mongolic, and Tungusic languages be combined into Altaic?" While there is dispute about both questions, it seems as though the former theory has never reached the status of being generally accepted; while the latter was generally accepted, but is now being challenged. I think our treatment of both controversies should be to err on the side of the established convention - that is, of an Altaic language family that only includes Turkic, Mongolic, and Tungusic. This should be the baseline position given in the article, and controversies on both sides should be mentioned - is the group not broad enough? Is the group nonexistent? john k 18:14, 10 August 2005 (UTC)

Well said. Personally, I don't see what the fuss is about with Altaic. Even if it's a sprachbund, there is absolutely something left under all this rubble that can be still be called "Altaic". I think you're dead on about the "arguement confusion" and your suggestion is very reasonable to me. --Glengordon01 05:57, 8 July 2006 (UTC)

[edit] Edit on opening paragraph

I am merging the second and third sentences in the opening paragraph

"The relationships among these languages remain a matter of debate among historical linguists, and the existence of Altaic as a family is rejected by many. Some scholars consider the obvious similarity between these languages as genetically inherited, others propose the idea of the Sprachbund."

into

"The relationships among these languages remain a matter of debate among historical linguists, and some scholars consider the obvious similarity between these languages as genetically inherited, others propose the idea of the Sprachbund."

which enables the paragraph to sound more neutral while still strongly stating the doubt on its existence. My reason for this edit was that in the previous version, there was no information on the number of Altaic family proponents, only a mention of that it "is rejected by many". This asymmetry was also incompatible with the statement

"The Altaic theory is supported by many linguists, but many other linguists (eg Doerfer 1963) do not regard Altaic as a valid group"

later in the article. Atilim Gunes Baydin 12:48, 24 November 2005 (UTC)

[edit] More 'accessible' intro paragraph

As a non-linguist, may I suggest that this would be more understandable, without sacrificing meaning:

"There remains a scholarly disagreement among historical linguists as to the nature of the relationships between these languages, and the existence of Altaic as a distinct family of languages is a matter of linguistic debate. Some linguists consider that the obvious similarity between these languages points to their being genetically related, while other linguists consider that such similarities are a result of linguistic convergence and propose the idea of the Sprachbund."

While longer, I think it better explains the underlying concepts, as befits an introduction to an encyclopedia article, bearing in mind that this article will be read by those with no background in linguistics, as well as experts. User:Pedant 19:47, 6 September 2006 (UTC)

[edit] Martin deleted

"In fact, its proponents have put together a large variety of grammatical, lexical, and syntactic regular correspondences between the sub-groups of Altaic (e.g., Ramstedt, Poppe, MARTIN, Starostin)."

Martin did in fact contribute very much to the discussion of the Altaic theory (namely, Korean and Japanese), but he isn't a proponent of the Altaic theory. Nowadays, he isn't even so sure about Japanese and Korean any longer, so he shouldn't be mentioned in a sentence like this.

[edit] content

i dont find the article incredibly useful it has very little information

[edit] Ural-Altaic, IE, Hungarian and Turkish

On the whole, I agree with most of Gordonglen's opinions on the subject - It is important to let go of the "us-versus-them" mentality and consider the issue with objectivity.

The "Altaic" problem has three sides to it:

  • The political perspective
  • The scientific perspective
  • The practical problems

The political perspective is that the "Pan-Turkic" movement (still kept alive in various countries) has its roots in the 19th century and the Ural-Altaic language group. If the Ural-Altaic hypothesis is generally accepted, the idea can be revitalized. ( For example, the common cultural ancestry of all Turkic speaking peoples and Mongolians, and Finnish and Hungarians could be traced to the Hsiang-Nu Emprise of the Asian Steppes. ) On the other hand, the PIE hypothesis could be used to support the Huntington claim of Clash of Civilizations thus: The IE ethnic group could be claimed to have intrinsic values that allows it to prevail, these values are shown in religion, daily life and are progressed through centuries by language; and the superiority of the IE culture can be claimed to be proven by the geography that it has expanded to. This problem arises when the language family is equated with an ethnic group - a speculative approach with dangerous political consequences: See Paleolithic Continuity Theory for an hypothesis claiming that IE *did* originate in Europe. Also see Genetics section under Proto-Indo-Europeans. (Contrary to what may appear to be the case, there is a claim that IE originated in Europe in the Paleolithic period, and it can have political consequences.) Fact is, that language families, apparently a harmless scientific subject, can become tools for right-wing policies when they are equated with ethnic groups, cultural spheres or races.

Both political views are way too much distorted and senseless to be considered seriously in academic circles; but both ideas have a chance gaining popularity - with 'masses general' as Glengordon puts it.

The scientific perspective is that the comparative linguistics is insufficient to explain certain cases in languages. I am also an amateur, however the following example could be good for explaining the dilemma: In modern Turkish, (Altaic family) the singular pronouns are as follows: Ben, sen, o. (3rd being neutral sex.) The Hungarian (Uralic family) singular pronouns are: En, Te, ö. (pronounced "o," same as the Turkish equivalent.) Here is the problem: The 1st and 3rd pronouns may suggest common ancestry. However, the 2nd pronoun could suggests otherwise. Pronouns are very basic word forms and are not usually borrowed. Therefore, the situation does not have sufficient evidence to suggest common ancestry, but borrowing is not a viable explanation as to the existence of common fundamental vocabulary.

On the other hand, the comparative linguistics method has been most succesful with the IE family, in a sense, the best support for comparative linguistics comes from the language family that gave rise to the theory. Imagine an alternate scenario where the theory for linguistic ancestry could have been developed by people speaking Nilo-Saharan family languages: Perhaps they would have considered a different methodology to find common ancestry?

The practical problem with the Altaic language family studies contitutes three issues: First, agreeing with Glengordon, IE enjoyed a much longer period of written texts. Second is the lack of texts in the given time period. The third is a general lack of interest: The IE languages were written with various alphabets, so common roots could be determined only after these different alphabets were transcribed in the modern latin alphabet. This is yet to be done for the Uralic and Altaic families.

For this page, I think it is best to mention that mainstream linguistics consider Uralic and Altaic families to be separate groups, but also to mention the Ural-Altaic hypothesis. It is certainly not a good idea to merge the pages: They should stay separate pages. This is also the framework for various common ancestries suggested for PIE, such as the Afro-Asiatic, Indo-Uralic and Proto-World.

On a side note: Hittite and Luvi *are* Indo-European. No "bar" is involved, in fact, Hittite is *the* language that supported the IE family in the sense that the theoretical basis for laryngels (proposed before Hittite was deciphered) was quote-unquote proven by Hittite. -- User:Sinanozel {timestamp}

Ante Aikio claims on Talk:Ural-Altaic languages#Proto-Altaic, how can anyone logically avoid it? that the Ural-Altaic hypothesis is "dead" and "rejected". I assume you're referring to this? To me, "rejected" implies that the theory is disproven, which it of course isn't. A scientifically objective way of stating the status of Ural-Altaic is: "Ural-Altaic is insufficiently proven to the satisfaction of mainstream linguists". Also, Ural-Altaic may be a dead name but it's hardly a dead theory since it continues to thrive on in the form of a subhypothesis of the larger Nostratic family. So we should really perhaps be mentioning "Nostratic" not "Ural-Altaic" per se.
The dismissal of Uralic-Altaic comparisons is due more to a fundamental denial of Proto-Altaic itself by a vocal group of conservative extremists. Afterall, by denying Proto-Altaic altogether, it slyfully undermines any comparison of Uralic to a now-conveniently "non-existent" protolanguage. A delightfully orwellian tactic :) I've noticed extreme conservatives play mindgames by purposely abusing the term "mass comparison" to vilify anything produced by an opposing camp.
The pronominal systems of Uralic and Altaic are clearly related somehow because of not only the "mi-ti-an" pattern noticed long ago, but an entire set of deictic and interro-relative pronouns too, as well as the presence of a trace pronominal oblique case marker *-n- in both families. IE and Etruscan can even help to confirm all of these same patterns. Allan R. Bomhard has published ample stuff on this but I may be ignorant of other lesser known individuals who have as well. No sensible linguist thinks that it's unlikely that Uralic *mun ~ *minä (Jahunen) would be unrelated to Altaic *bi and its oblique stem *min- (Starostin) because these similarities are part of a larger structural context (namely specific similarities within an entire pronominal system that as a whole cannot likely be due to chance anymore).
If sound correspondences between Uralic and Altaic haven't been well laid out to this day, it is in my view strictly because of linguist politics that hold the field down by constantly propping up a petty war between narrow-minded conservatives and out-to-lunch mavericks. Somewhere in between are the victims of moderateness and honest science. I've already noted, for example, that the Altaic *m/*b alternation is a specific pattern attested in a few languages like Southern Min and the Guarani language where it is particularly associated with "nasal harmony" [1]. In other words, we can see by comparison with real-world languages that *b in Altaic *bi must be from earlier *mi in order to explain the preservation of *m in the oblique *min- where a following *n is coincidentally present. Pre-Altaic nasal harmony is the only rational cause of this pattern based on my observation. That's how internal reconstruction is done. This common sense helps to balance out one-sided accusations by anti-Altaic denialists.
As for your whole apologetic self-described amateur thing, User:Sinanozel, hold your head up high. Never dismiss yourself as an amateur because that implies that there's actually a point when we're not. Modesty is a bad habit :) Not one of us knows everything. And consider: Should education only count if you pay for it? Is blindly following a pre-defined curriculum better than a self-directed education requiring a greater initiative and zest for knowledge? University is not set up to produce society's problem-solvers. A PhD is just a status symbol. While I'd love to live in an utopia, we have to accept that PhDs don't necessarily prove that their holders know what they're talking about. --Glengordon01 02:56, 28 July 2006 (UTC)

[edit] Question

"who now rejects a genetic connection between Korean and Japanese" In the above passage what does the term genetic mean? User:Pedant 17:09, 6 September 2006 (UTC)

That they are descended from a common ancestor. --Ptcamn 18:06, 6 September 2006 (UTC)
Would that passage be equivalent to: "who now denies that the Korean and Japanese languages are derived from a common Language family"  ? I'm trying to wrap my brain around this article. Thanks for your help User:Pedant
Yes, it would. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 11:11, 1 June 2007 (UTC)
Except that "deny" can imply "lie", while "reject" can't. Because we have no reason to suspect that Vovin is dishonest, to use "deny" would not be a good move.
To be pedantic myself: it would be "derived from a common ancestor", "derived from the same protolanguage", or "part of the same language family". David Marjanović 21:01, 23 June 2007 (UTC)

[edit] Altaic Controversy

According to major reference works, everbody agrees on Turkic, Tungus, and Mongolian families. These language families have some common words and they are typologically similar. Whether they are genetically related or not is the subject of ongoing research. The similarities could be result of long-term contact.

All languages are influenced by languages they are in contact with. According to the standards set by linguists, languages that make up a family must show productive-predictive correspondences. The shape of a given word in one language should be predictable from the shape of the corresponding word, or cognate, in another language. Turkic, Tungus, and Mongolian satisfies all these similarities.

I do not agree with the idea that the Anti-Altaic hypothesis necessitates total abandonation of the Altaic one. This Altaic Language article seems to me a propaganda of the Anti-Altaic hypothesis rather than an objective comparison of the two.

The family name "Altaic" is a commonly used terminology to label these languages. Turkic, Tungus, and Mongolian are still Altaic regardless of exact status of Altaic is. The dispute is a minor issue among a small circle of specialists. Furthermore, even these specialists use the term Altaic to label, especially, for these three language groups. e104421 7 September 2006, 00:47 (UCT)

I completely agree with E104421. While I personally haven't seen any compelling evidence for an Altaic connection, I don't believe that the label is useless. I've seen just as little evidence for the Niger-Congo language family and have seen very good arguments against the unity of the Uralic family, yet there's little dispute over whether or not they're useful labels.

Additionally, I agree that the emphasis in this article should be on Turkic, Mongolic, and Tungusic. Starostin's connection of Korean and Japanese to Altaic are much more controversial than previous definitions of the family and should possible get their own section within the article, rather than being lumped with the more established branches. Straughn 14:38, 14 September 2006 (UTC)

Ok, first of all a brief note on the Altaic issue. Having thought about it, I now agree with E104421 that the discussion on the controversy should go to Altaic languages. And while this may be beside the point, I have to ask user Straughn: what are the these supposedly "very good arguments against the unity of the Uralic family"? If you're referring to Marcantonio's monograph, I can only recommend to read the reviews. --AAikio 16:20, 14 September 2006 (UTC)
I agree as well that Altaic controversy need not exist independently of Altaic languages. As to the Uralic stuff, it seemed reasonable and well argued when I read it through, but I certainly didn't agree with Marcantonio's arguments. I merely wanted to point out that even well-established families can come under fire and that Altaic was merely one that got a comparatively large amount of attention. Straughn 17:51, 14 September 2006 (UTC)
I guess you mean the article Altaic hypothesis instead; I agree, it should definitely be merged with Altaic languages. As for Marcantanio, I must say that I found her book unreasonable and poorly argued instead. As has been pointed out by numerous reviewers, her main arguments are methodologically invalid, and often also based on ignorance or gross misrepresentation of Uralic data. For a brief summary of the book's flaws, please see my comments on the Linguist List: [2]. --AAikio 06:56, 15 September 2006 (UTC)

I've made Altaic hypothesis a redirect. As stated on Talk:Altaic hypothesis, that article was nearly copy-paste only. --Pjacobi 11:35, 19 October 2006 (UTC)

348? Some random guy changed it from less than two hundred to over two hundred to finally over three hundred million speakers with no citation or reason why he changed it. Does anyone dispute this? Swatpig 23:09, 23 February 2007 (UTC)

[edit] Cognates table

Hi, where did the cognates table come from? Some of the Korean items seem like gibberish, not Korean. --Kjoonlee 13:32, 11 October 2006 (UTC)

Hi, I think a few initial entries are from the published work cited before the table. I recently added some others from the German version of this article, de:Altaische Sprachen, which is highly complete and comprehensive compared to the English version, and is marked "lesenswert" (recommended for reading). I think no one knowing Korean checked the page before, could you please fix if there are mistakes? One possible reason that these might not make sense could perhaps be that they are Old-Korean and not properly marked. Pangapseumnida! Atilim Gunes Baydin 14:09, 11 October 2006 (UTC)
The only fixes I can do are deletions, because I'm unaware of any not knowledgeable about cognates, and adding non-cognates would be an error. I had a look at the German version, and I can only recognize 5 modern words out of the 18 pairs of Old Korean-Korean that are there. --Kjoonlee 01:32, 12 October 2006 (UTC)

No, no, no. Firstly, find the original source for the cognate table (no source is cited). If it's older than the Altaic Etymological Dictionary from 2003, DELETE IT. That book presents a new reconstruction of Proto-Altaic with radically different vowels. I'll look for a better cognate list in Blažek's paper (after I'll have finished transcribing the phoneme inventory and the correspondence lists from there).

If it survives, change what it looks like. Currently, the Turkish orthography is used for Turkish, one of the common transcriptions is used for Japanese (why modern Japanese instead of Old Japanese?), and a completely idiosyncratic phonetic transcription (probably Starostin's) is used for the rest (including Korean, for which two official transcriptions exist)! What we should do is obvious: use both official orthographies (which exist for all of the modern languages given) and IPA transcriptions.

David Marjanović | david.marjanovic_at_gmx.at | 21:40 CET, 2006/10/29 | major edit 17:53, 2006/10/30

(Regarding the deletion above, there are lots of cognates in Blažek's paper, but only a few selected Japanese, Silla, Koguryo, Kitan and/or Tabgač reflexes are given for each – nothing of encyclopedic value.)

David Marjanović | david.marjanovic_at_gmx.at | 17:50 CET, 2006/11/3

So. I replaced the old cognates table two new tables, based on the EDAL ( = Starostin et al. 2003), one for a large number of numerals and one for other nouns; note that some of the new reconstructions (e. g. "heart") are strikingly different from the old version. I think the article is long enough now. :o) David Marjanović 17:02, 7 April 2007 (UTC)

[edit] Sound correspondences

[sorry for the chaos yesterday – that bizarre computer didn't show me the cognate table because I hadn't refreshed the page, but did show me my own edits!!!]

I'm the one who made today's changes. I put the sentence about Comrie in italics because I don't understand it. Does Comrie think that Turkic, Mongolic and Tungusic are members of Nostratic, but not more closely related to each other than any of them is to any other Nostratic language families?

I'll start adding the sound correspondences listed by Blažek next. Someone, after all, has to put some actual evidence for Altaic into the article! David Marjanović | david.marjanovic_at_gmx.at | 21:40 CET | 2006/10/29

Added correspondences between the plosives.
David Marjanović | david.marjanovic_at_gmx.at | 23:13 CET | 2006/10/29
Finished consonant and vowel tables.
Yesterday I also added the reconstructed phonology of Proto-Altaic. Then someone moved it below the sound correspondences. If nobody protests till Sunday, I'll put it back, above the correspondence tables, because the current order, although more logical scientifically, is worse didactically and stylistically: readers should first see what sounds there presumably were (two short tables) and then how they evolved (two very long tables).
Done. Also renamed "reconstructed morphology" to the much better fitting "morphological correspondences".
David Marjanović | david.marjanovic_at_gmx.at | 17:49 CET | 2006/11/9
Next: Case and number suffixes and pronouns from Blažek's paper.
I hope somebody can check out the prosody table in that paper. Does it mean Starostin et al. reconstruct pitch accent or even tone for Proto-Altaic?
Also please compare my version of the vowel correspondences to the original. Blažek puts a dot under some Proto-Turkic /a/ and /e/ and a breve on some Proto-Korean /a/; those are apparently separate phonemes, but what are they supposed to have been pronounced?
David Marjanović | david.marjanovic_at_gmx.at | 19:25 CET | 2006/11/1
Hi, as someone very much interested in Altaic studies, I'm following your contributions to the article with great excitement. May I suggest you to create a proper Wikipedia account and continue with your work on the article using that? If you haven't noticed yet, it's very simple to get an account, just use the "Create account" link on the top right corner of any Wikipedia page. That way, it would be easier to communicate with you (via your talk page) and recognize your contributions in the edit summaries. Edits from a registered user always get more credence than from an anonymous ip address. There are situations where anonymous edits are not allowed at all, like on pages with "protected" status. Please see Wikipedia:Why create an account? and keep up the good work. Regards, Atilim Gunes Baydin 19:15, 1 November 2006 (UTC)
Thank you – I will register sometime soon to create the article Phylogenetic nomenclature which is sorely missing. So far I don't need an account, IMHO. I don't need any credence at present – I cite my sources, which are usually available online for free and linked to in the references or external links section (as in this case), and I mention my edits on the respective talk pages.
Communication might be even easier via e-mail. That's why I put my address in my "signature".
David Marjanović | david.marjanovic_at_gmx.at | 22:19 CET | 2006/11/1
Entered morphology table and pronoun table. Please check for misunderstandings! Unlike Starostin et al. I put the /tʰ/- forms and the /s/- forms in the same lines because the required sound shift would be no weirder than what happened in Greek (Classical Greek /sy/ "thou" instead of the expected */ty/, without a regular sound shift).
I wonder if anyone who has read the etymological dictionary by Starostin et al. has ever argued against the existence of Proto-Altaic… do we know if there still is a controversy?!?
David Marjanović | david.marjanovic_at_gmx.at | 22:24 CET | 2006/11/1
To answer my own question... yes, Vovin and Georg. I've read Georg's critique. It doesn't seem to make any arguments except that some, perhaps many, etymologies are wrong or doubtful – OK, but that still leaves a lot! – and that Starostin's method of using a part of glottochronology to "prove" that languages are related is entirely unconvincing – a point on which I agree, but which has nothing to do with the rest of the EDAL. Importantly, Georg does not provide an alternative hypothesis about what the branches of Altaic are supposed to be related to; as a biologist, I prefer having a hypothesis over having none. David Marjanović 01:01, 26 April 2007 (UTC)

Just a quick comment: This link to John Whitman is incorrect. The link goes to a different John Whitman (an author) and not the historical/asian linguist at Cornell university. 163.152.180.29 16:28, 17 December 2006 (UTC)

[edit] revert 13.3.2007

I reverted the edit by User:David Marjanović for two reasons. First, I don't think it's a good idea to introduce a less common term such as Buyeo languages to the infobox, especially as that classification is in itself highly controversial; and second, the "Altaic" origin of Japanese is certainly not a very widely accepted view among specialists (see Japanese language classification).--AAikio 11:59, 13 March 2007 (UTC)

"Buyeo languages" is not a common term i agree with you, but Japanese language or at least Old Japanese language and Goguryeo language are very close, you can not refute that hypothesis because there are proofs written in the Hou Han Shu and also Old Korean book written in Classical Chinese characters. According to my knowledge, there are also no real evidence relationship between Japanese language and Ryukyuan languages, which both belong to the Japonic language family. RegardsWhlee 19:09, 13 March 2007 (UTC)

OK, OK. At least Korean is back in again.

However, whoever wrote "Unofficial" into the Altaic languages template (at the bottom of the page) was being childish. This is science, not bureaucracy. There is no such thing as "official". There are only "disproven", "controversial", "improbable" and the like.

The main reason why I put Buyeo (a term I had never seen before; I've only seen "Koreo-Japonic", which is an ugly word, and "Eastern Altaic", which is vaguely mentioned in the EDAL) into the infobox, instead of just restoring Korean, was that it spares me the work of including Goguryeo, Silla, Baekje and whatever else all separately...

That the membership of Japonic in Altaic is not widely accepted (at least among people who haven't read the EDAL :-> ) does not matter here – what matters is that most Altaicists (maybe all now, now that the EDAL has been published) include it.

I'll write "Korean and its relatives" and "Japonic and its relatives" into the infobox, make them "often included", and restore Ainu as "rarely included" – it's not included in the EDAL, but for all I know there are still people who consider it possible that Ainu is Altaic (rather than "Austric" or than just giving up). David Marjanović 23:10, 21 March 2007 (UTC)

Better yet: to avoid confusion, I wrote "and its extinct relatives" behind Korean and Japonic. David Marjanović 23:18, 21 March 2007 (UTC)

[edit] Template

This seems like a better location to get this discussion going... I'd like to propose deleting Template:Altaic languages. It doesn't seem to achieve anything new or informative. It includes the same information and links already presented in Categories and Infoboxes. All it does at the moment is take up space on the page. Is a template really needed? --Stacey Doljack Borsody 15:44, 15 March 2007 (UTC)

I think we should keep that template : Template:Altaic languages, because lots of Koreans "ignore" that pure/native Korean words (i.e. those which are not Sino-Koreans words) of the Korean language are related to the Altaic languages family, i'm also wondering whether Japanese peoples know that pure Japanese vocabulary (Kun-yomi) may also come from the same family. On the other hand, i also think that we should improve that template by being inspired on Template:Romance languages. I would be glad to be helped. Regards. Whlee 07:36, 16 March 2007 (UTC)
I don't understand your reasoning. It doesn't have any basis in webpage and information design. Take a look at a page like Turkish language. You can view an Infobox to the right side that clearly lists language family, with links to Oghuz languages, Turkic languages, and Altaic languages. A reader can quickly gather information regarding the classification of this language and click on the links to see what they are about. This is good presentation of information. The templates at the bottom don't do anything except take up page space and act as a redundant "See also" section. I'm not a big fan of these kinds of templates. Template:Romance languages and Template:Turkic languages are equally useless, but tolerable since the scope of the "family" is small. --Stacey Doljack Borsody 23:28, 18 March 2007 (UTC)
because lots of Koreans "ignore" – you have mixed up French (where ignorer means "not know") and English (where ignoring something is a deliberate act of not recognizing it). David Marjanović 01:03, 22 March 2007 (UTC)
Yes you are right, ignore in that case means do not know (and not refuse to recognize). I made confusion between those terms. Thanks for your comments. Regards.Whlee 08:48, 23 March 2007 (UTC)
A couple of quick comments from me. I study ancient near east languages, especially the semitic ones. They have distinct regional dialects and many modern derivatives. The whole language family is important to my field of study, because sometimes the best way to get a good guess at an ancient word that is ambiguous in context is to look for possible cognate words. What this means is, for my purposes only, the templates at the bottom are more useful than the info boxes. I'm a rare breed, but I'm not alone. I think rare people like me want the info in the templates at the bottom. Most people want the info box, but some of us want the template. I think it is appropriately placed at the bottom, that way people can ignore it unless they are serious. It comes after See also and references. Essentially the template box is just a way of organizing a lot of See also and category type information for the page. Let's just try to make these templates default to hidden form, as they do on the medical pages, what do people think? Alastair Haines 02:30, 11 April 2007 (UTC)

[edit] Japanese connection to Altaic is a joke

The supposed link between Proto-Japonic personal pronouns and Proto-Altaic pronouns is, guess what, also the same for many Chinese dialects (Sino-Tibetan language family)!

In the Wu dialect of Chinese, "I" is a (open schwa) or ala /6?la/ (阿拉); "thou" is /na/ (那); the plural in the Wu dialect is /la/ (伊拉 = they); the alternative plural is /ta?/ (儂達 or 儂塔 = you all); the dative-locative is /li/; and the possessive is /g@?/ (个). I don't think anyone is suggesting that a southern coastal Chinese dialect like Wu is Altaic now? Additionally, the purported Proto-Altaic phonology (including its pitch accent) looks almost identical to the modern Shanghainese phonology. The only thing missing in Shanghainese is the vowel harmony, which frankly Japanese doesn't quite have anyway. This just shows that convergence doesn't imply genetic relationship.

From article: "The many correspondences between Altaic pronouns found by Starostin et al. (2003) could be rather strong evidence for the existence of Proto-Altaic." "strong evidence"? What a load of crap.

--Naus 03:17, 23 May 2007 (UTC)

My sentiments exactly, Naus! The "Japanese and Altaic" controversy is all the convoluted aftermath of a certain linguist's bad joke.

But it should be noted that Wu dialect 伊拉 (i-la, "they") looks like a compound of Tungusic i ("he/she") or Korean i ("one, the one that...") with Turkic -lar ~ -ler (plural suffix) or Japonic -ra (plural/collective suffix). And we really should emphasize the fact that Japanese does not have vowel harmony. Ebizur 01:31, 25 May 2007 (UTC)

People, did you read the article? Firstly, the latest reconstruction of Proto-Altaic lacks vowel harmony. Whether Japanese later developed vowel harmony and then lost it again doesn't matter. Secondly, it's not just the pronouns: it's the pronouns plus lots of other words plus regular sound correspondences plus a considerable amount of grammar.
(Incidentally, I find it interesting that you use Chinese characters as a syllabary.) David Marjanović 17:53, 2 June 2007 (UTC)

[edit] Turkic Numerals

Here are some thoughts ("original research", I know..) that I, as a non-scholar, came across and would like share:

conflictive opinions in favour of either the "Altaic language familiy" or the "Altaic sprachbund" theory seem to be kind of revolving around the controversy about the numerals of the Turkic languages that differ strikingly from its Mongolian and Manchu-Tungusian counterparts (note: Indo-European languages' affinity have been largely established through comparison of ancient written languages, thus old Greek, Latin, Sanskrit, Avestian, and the clear etymologies concerning numerals; comparative linguistics seem to be relying rightfully on this empiricism when dealing with other languages).

But... look, here: the numeral four in Turkic resembles its Mongolian counterpart, whereas the numeral five resembles Indo-Iranian (Saka/Scythian?!). Is this simply coincidence? Original numerals in Proto-Turkic could have been replaced by loans. Then what about the other numerals? What struck me when comparing numerals in Turkic to modern Mandarin (sorry, this is by no means scientific research I am doing, just my observation..), was that the words for one, two and six and seven (with a little imagination and background info) sounded really similar! Consider this: Chinese is a tonal language. Since over the course of history Chinese has changed quite a lot concerning pronunciation, those Turkic numerals that I stated above could be its speakers' interpretation of the Chinese numerals that they heard; thus adapted to Proto-Turkic pronunciation.

Just think of the numerals of sinic background in Korean and Japanese. The only numeral that has common spelling (almost 100%) in these languages (non-tonal, of course) is three. Pronunciation and interpretation of original Chinese varies greatly. So what about Turkic then? Only speculating, but the spelling variants of the word for three at hand could be derived from a common idiomatic rendering of a word which might have sounded like sam/san. Ok, the argument looks weak here.

Some other aspect that I think should be taken into account is this: the numerals 6-9 in Turkic show two types of suffixes (the latter lacking in Tuvan, for example) that seem to indicate plural or could be connected with the tribal organizations of the peoples that spoke Turkic, i.e. originally these might have not been present in the Turkic counting system.

Turkic, to me, seems to be an Altaic language that due to contact of Proto-Turkic tribes with Iranic tribes had evolved to some sort of "Creolic" nature (excluding language characteristics, grammar, etc.). Common vocabulary between Turkic and Mongolian does not necessarily mean these are loans. There also exists common vocabulary between Turkic and Tungusian languages, which Mongolian lacks. It should also be noted that Proto-Japanese originally might have been an Altaic language with vowel harmony, or it might have separated from Altaic at an early stage of that language when there was no vowel harmony.

Thraco-Anatolian Turkish might be a bad choice for research on Altaic. But it highlights something that is remarkable: loanswords can replace (basic) vocabulary, even everyday language can be affected; religious cult, fashion and trends, an affinity for sedentary cultures can prove so influential that the use of foreign vocabulary can substitute original vocabulary of a language, hence the evolvement of Ottoman-Turkish. Salyr Turkmen, for instance is a better choice in my opinion. The lack of older written documents further complicates research. 134.100.32.213 15:37, 28 May 2007 (UTC)

If comparing with Chinese you might use the reconstructed Old Chinese or Middle Chinese numerals. I do not see similarity with the reconstructed Proto-Altaic numerals. --JWB 19:02, 28 May 2007 (UTC)

Well, I meant certain Turkic numerals as opposed to Mongolian and Tungusian, clearly not the reconstructed Proto-Altaic numerals. Thanks anyway. 134.100.32.213 07:22, 30 May 2007 (UTC)

I can't see any resemblance whatsoever between Turkic and Chinese (at any stage) numbers. Could you list actual examples? --JWB 07:44, 30 May 2007 (UTC)
Not just Chinese. If you can, have a look at the reconstructed Proto-Sino-Tibetan numerals. No particular resemblance to Proto-Turkic or for that matter Proto-Altaic.
Also, please read the article again: vowel harmony was most likely not a feature of Proto-Altaic, it developed separately in each branch in different ways.
The "Old Turkic" mentioned in the article is the language of the Orkhon runes. That is fairly old.
It may well be that Turkish is a poorer choice than Turkmen – but is Turkmen a good choice? No: the only good choice, for obvious reasons, is reconstructed Proto-Turkic.
Don't use "a little imagination". Instead, look for regular sound correspondences. If you find any, and get that published, the Wikipedia article will need to be updated. Otherwise not. David Marjanović 17:46, 2 June 2007 (UTC)

Well, I didn't consider changing the article at all. All I wanted to know was if there is actually any theory that deemed certain Turkic numerals as loans from Chinese. Here are the sound correspondences that I found (note: I added the sinic numerals in Japanese and Korean for comparison; of course none of them are exactly identical, just illustrating how the pronunciation can vary):

1 Turkic: bir ~ Chinese: yī ~ Korean: il ~ Japanese: ichi

2 Turkic: eki ~ Chinese: èr ~ Korean: i ~ Japanese: ni

6 Turkic: alty (alt) ~ Chinese: liù ~ Korean: ryuk/yuk ~ Japanese: roku

7 Turkic: yeti (jeti<jet) ~ Chinese: qī ~ Korean: chil ~ Japanese: shichi

The ones in brackets are my own interpretation (in kypchak.-Turkic, but without the suffix > for explanation see above).

By the way are there any theories that claim the Turkic numeral five (besh/pesh) is a loan from Indo-Iranian (possibly Saka)? 134.100.32.213 09:54, 15 June 2007 (UTC)

All the Chinese and xenosinic numerals descend from Middle Chinese, which is later than Proto-Turkic, so you should use Middle Chinese or Old Chinese as basis for comparison with Proto-Turkic.
It looks like your resemblances are as little as the words containing i, e, l for 1, 2, 3. --JWB 16:40, 15 June 2007 (UTC)

[edit] Cognate table

The cognate table is confusing. How come there are two words for "I" and "thou"? No case is given, what's the difference? What does (8) mean in the numeral table? There are a lot of these small things that need fixing... 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 11:23, 1 June 2007 (UTC)

It's a very good question why there are two words for "I" and "thou" each. Normally when such phenomena are reconstructed, we are looking at irregular case differences (as in English: "I" vs "me", "we" vs "us") – but I have only read the preface of Starostin et al. (2003), and there they simply present these forms without explaining them. I can send you the preface as a pdf (warning: it has 270 pages).
"(8)" referred to the footnote 8. The parentheses are there to make clearer that the 8 is superscript. Obviously this didn't work. I just changed that to a superscript "(footnote 8)". David Marjanović 17:38, 2 June 2007 (UTC)
Alright, thanks for answering. I believe you, however sloppy it looks. 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 19:08, 2 June 2007 (UTC)
Found it! EDAL p. 225:

4.2.3.1. Personal pronouns

1 p. *bĭ̀, pl. *ba ~ *bu (obl. *mi-n-, *ma-n- ~ *mu-n-)

For PA we can also reconstruct a stem *ŋa, reflected in some cases as Mong. *na-d-, *na-m-, and preserved in Korean as *nà and in Jpn. as *a-.

2 p. *si, pl. *su (obl. *si-n-, *su-n-)

It seems also possible to reconstruct a second stem *na, preserved in Kor. *nə̀ and Jpn. *ná, and possibly reflected in the PT 2d p. ending *-ŋ (although velarization here is not quite clear).

The relationship within the suppletive pairs *bi - *ŋa and *si - *na is not quite clear; the forms *ŋa and na may have originally been restricted to some oblique cases (cf. the situation in Mongolian).

Not that I knew the situation in Mongolian, but this seems fair enough to me. David Marjanović 17:04, 16 June 2007 (UTC)
Aaahh.. Well done! 惑乱 分からん * \)/ (\ (< \) (2 /) /)/ * 17:02, 20 June 2007 (UTC)

[edit] Three more things

  • Why do people insist that Korean and Japanese are only "sometimes included"? Georg et al. (1999) make clear that, by now, almost all Altaicists include Korean, and that most include Japanese. Give me your e-mail address, and I'll send the pdf to you.
  • Why do people keep deleting the sentence saying that Korean and Japonic do have shared innovations according to Starostin et al. (2003)? That's a fact: Starostin et al. really do say that, and they published several years after Vovin said Korean and Japonic lack shared innovations. I can send the pdf of the introduction to Starostin et al. (2003) to interested parties – warning: it has 270 pages.
  • Precisely why is the "neutrality disputed" tag up there? It may well be justified – I have not been able to get all relevant papers, and it seems nobody who has read them contributes here – but slapping such a tag on an article without explaining why is not going to help anyone. David Marjanović 18:16, 2 June 2007 (UTC)
You should ask User:Naus. This is the diff where the tag was included[3]. It seems he doesn't agree with Korean and Japanese being Altaic languages.[4] And I would be interested in those PDFs. I'd appreciate if you share the files. I can be reached at cydevil@naver.com Cydevil38 02:12, 7 June 2007 (UTC)

I'd like to get a copy of those pdf files also. My email is yerhan @ earthlink dot net (without the spaces and dot = .)

[edit] Disputed reconstructions of single words

Starostin et al. (2003) mention a Proto-Japonic word */nəmpV/ "neck". An anonymous user slapped a "dubious – see talk page" tag on this and commented their edit with "There is no form */nobV/ in Japanese that means "neck." Who comes up with this BS?)". Well, who? Starostin et al., as cited. I don't have the part of the book where they explain how they arrived at this particular reconstruction, but keep in mind that Proto-Japonic is not the same, lexically, as even Old Japanese.

If you put a "see talk page" tag into an article, why don't you write something on the talk page…?

However, I've found a paper that disputes another reconstructed word in that table. I'll fix that in a few minutes. David Marjanović 16:33, 16 June 2007 (UTC)

Someone added this link to starling.net.ru citing two words from Ryukyuan languages as evidence for Proto-Japonic */nəmpV/ "neck". Accordingly I have removed the "dubious" tag. David Marjanović 22:06, 12 July 2007 (UTC)

[edit] It is the words "father" and "mother" that give us the direction

We all know that when comparing with words from different languages, no linguists will use the words "father" and "mother" for almost all the languages are the same in these two words. But the languages below will tell us how unique their father and mother are. It is their uniqueness that gives us the right direction.

father/mother

finnish:isä/ema
nenets:nāće/nema

turkish:aka/ana
yakut:aka/ije

khalkha:eceg/ex
daguar:ecig/eg

manchu:ama/ənijə
evenki:ami/əňin

korean:apoji/omoni

japanese:chichi/haha and otou/okaa

ainu:míci/hápo

nivkh:etk/emk

yukaghir:et'ie/emej

chukchi:ellen/ella
itelman:is'h/lahs'h

inuit:aapa/aana

tlingit:eesh/tlaa
apache:taa/maa

Please explain what your point is. I don't get it. David Marjanović 21:58, 12 July 2007 (UTC)

[edit] Question about A. Vovin

I would like to know what Alexander Vovin's current position on Altaic is since I have no chance to lay my hands on his publication "The End of the Altaic Controversy"(2005). From the sparse information that I have read elsewhere he is now an opponent of the supposed genetic relationship of Japanese and Korean, and their genetic affiliation with Altaic respectively. But does he now contest the very existence of the Altaic language family as consisting of Turkic, Mongolic, Tungusic in favor of the Sprachbund theory?

It would be appreciated if that info was worked into the article or at least discussed on the talk page. Thanks. 134.100.1.101 12:24, 13 July 2007 (UTC)

I have now read a paper by Georg and Vovin where they explain they accept half of Altaic – a Tungusic-Korean-Japonic grouping they term Manchuric. I'll update the article later today. David Marjanović 17:17, 21 September 2007 (UTC)

[edit] Vovin's and Georg's position

My name is Stefan Georg and I have occasionally written on the Altaic matter, both together with Alexander Vovin and independently. I am well acquainted with A. Vovin's position on the matter (and mine, of course :-) and I can confirm that Vovin, especially in his 2005 article, explicitly opposes the view that the so-called Altaic languages form a valid genealogical grouping. I share this position entirely, which extends to saying that, indeed, there is no genealogical relationship between any of the language families involved. It is true that, in earlier publications, we acknowledged some possibilities for what we called "Macro-Tungusic" back then, but we both are no longer prepared to defend this possibility (of course this does not preclude any future progress in the field). On the Starostin et al. dictionary, we both hold (and are ready to defend, like, incidentally, most published reviews of this work) the position that the dictionary, in spite of its size, does not represent any meaningful progress towards the (inevitability of) the acceptance of Altaic as a genealogical grouping, since we find a huge amount of factual and methodological infelicities, paired with a neglect of the state of the art of the individual philologies (and of course the discussion as a whole, as it developed over the last 150 years or so), so that we regard this work as basically failed. Btw., when I sent off my reply to Starostin's reply (on my review of his dictionary), which is mentioned in the article, to Diachronica, Sergej Anatolevich was still alive, news about his untimely demise reached me only later, so this mention may create the impression that I endulged in a quarrel with a man who was no longer able to defend himself, which never was my intention. I, and I can speak for A. Vovin here, too, maintain, though, that the genealogical relationship of "Altaic" has never been sufficiently demonstrated. I am actively involved in investigating the *areal*, i.e. convergence, relationship which *does* hold between at least the continental languages, which I and others regard as the only fruitful way to meaningfully describe the commonalities of these languages.

Stefan Georg Georg-Bonn@t-online.de —Preceding unsigned comment added by 212.79.166.21 (talk) 12:41, 25 September 2007 (UTC)

[edit] The meaning of "Altay"

Someone has written in the caption of our photo of the Altay Mountains that altay means "high (mountain)" in Turkic. This is completely false. Altay means "Gold Mountain" if anything. Ebizur (talk) 23:47, 17 January 2008 (UTC) Altın = Gold, al= red, dağ= mountain tay= colt may be Aldağ (red mountain?), Aladağ( redish mountain) or simply Al+tay=red cold

No, this is completely right, the etymology involves ala "high" and tag/taw/tay "mountain", "gold" would be "altun", and for this, the name is lacking an entire syllable. Of course, etymologizing names is always a risky business and the result can hardly ever be controlled. But "gold" is definitely out here. —Preceding unsigned comment added by 212.79.166.21 (talk) 11:29, 7 February 2008 (UTC)

According to my Turkish dictionary, /ala/ means "speckled, mottled, variegated in color; brightly colored; red, reddish, light brown." Nowhere does it mention /ala/ meaning "high." /ala/ as "high" sounds more like a Manchu-Tungus form, e.g. Manchu /alin/ (mountain), /alarame/ (along a hill), etc. The link that has been provided as a reference for the claim that "Altay" means "High Mountain" in Turkic is a Turkish website that allows one to search a database of Turkish personal names; this is not really an appropriate reference for the claim that is being made on the "Altaic languages" wikipage. Should no one provide a better source for the etymology of Altay as "High Mountain" in Turkic, I will altogether remove the claim of Turkic etymology from the Altaic languages page. Ebizur (talk) 20:34, 11 February 2008 (UTC)
  • Your interpretation is not correct. The word Altay is often used as personal name in Turkish. The site is the official site of Turkish language association. Mentioning the meaning of "Altay" in Turkic languages is informative. Regards. E104421 (talk) 21:06, 11 February 2008 (UTC)
It doesn't matter whether "Altay" is frequently used as a personal name in Turkey. That website is not an adequate reference for the etymology of /altay/ as a toponym, as in "Altay Mountains."
I have just looked at another Turkish-English dictionary, the Langenscheidt Standard Dictionary, and these are the six entries it has between /akzambak/ "Madonna lily" and /alabanda/ "bulwarks; broadside":
al1 1. scarlet, crimson, vermillion, red; 2. chestnut (horse); 3. bright red; 4. rouge; 5. [med.] eryspelas.
al2 deceit, intrigue.
âl, -li 1. family, dynasty; 2. high-born; âli Osman Ottoman Dynasty.
ala 1. colo(u)rful; 2. light brown; 3. s[ee] alabalık; ~ kaz white-cheeked goose.
âlâ first-rate, excellent, very good.
alabalık zo[ology] trout.
Thus, there is nothing in either of the Turkish-English dictionaries I have investigated to support the contention that there exists a Turkic etymon having the morphophonological form /al/ or /ala/ and having the meaning "high" (as in "High Mountain"). I request that you provide evidence from another Turkic language or from a reputable, verifiable source to support the present claim that the name of the Altay Mountains derives from a Turkic expression meaning "High Mountain." Ebizur (talk) 02:00, 12 February 2008 (UTC)
By the way, the editors of the Altay Mountains wiki have taken the position that "Altay" derives from Turkic etyma with the meaning "gold mountain," while the editors of the Alatau wiki have taken the position that "Alatau" derives from Turkic etyma meaning "motley mountain" (i.e. "mottled mountain"). Your contention that "Altay" means yüksek dağ ("high mountain," "lofty mountain," "great mountain") is apparently not generally accepted. All these wikipages need to be provided with scholarly references so that readers may verify the claims of particular etymologies for "Altay" or "Alatau." Ebizur (talk) 02:22, 12 February 2008 (UTC)
  • As a native speaker of Turkish, i can confirm that "altay" means "high mountain". The Turkish Language Association's dictionary is a reliable source. I recommend you to check the Turkish-Turkish dictionaries. You're trying to construct the meaning mathematically, but linguistics is not mathematics. Yes, "al" means "red" but also it means "take". On the other hand, "alt" means "below or under" and "ay" means "the moon". Mathematically, one can deduce that "alt" + "ay" = "altay" meaning "under the moon" :))) , but that's not correct, again. The word "altay" is used as a name, means "high mountain". That's it. How the word came about historically is another question. Yes, all wiki articles should reflect the same information. I'll check. Regards. E104421 (talk) 05:24, 12 February 2008 (UTC)

And as a native speaker of Turkish i ask you where do you get definition of "high" as it is logical to suggest tay as dağ, i ve never heard such usage of "al" if you can explain everyone will be happy. —Preceding unsigned comment added by 88.244.184.162 (talk) 15:30, 1 April 2008 (UTC)

[edit] Controversy

There are clearly a number of controversies about Altaic. I've tried to clarify

  • the validity of the family, and
  • the inclusion of particular languages in the family, notably Japanese and Korean.

It also seems as though opinion differs:

  • Between specialists in Altaic and specialists in individual languages
    it appears that Altaic specialists tend to include Japanese in Altaic, while Japanese specialists tend to exclude Japanese from Altaic
  • Between different linguists
  • Over time

...which presumably underlies contentiousness.

AFAICT, the family is relatively (but not universally) accepted by linguists, but Japanese and Korean are widely not included; there's clearly an ongoing debate on these matters, as evidenced by the references.

Nbarth (email) (talk) 22:55, 13 March 2008 (UTC)

[edit] whats the definition to altaic languages

in chemistry a group of elements has common chemical properties not phisical but as i see Turkic Mongolian comparison is mostly due to similarity of vocabulary and slightly grammar, but those are also similar to Hungarian language, and if the vocabulary is enough then Northwest caucasian languages are in the same group with Latin due to similar vocabulary of basic words(i know there are some hypothesis about abkhazian and hatti language).

anyway my point is if you want to form a group you must show a common ancestor or ancestors hybridizing, deriving etc etc as for IE languages or semitic if not Turkic is itself a group, Mongolian forms another group....

About genetics... hey dont forget Mongolian army, almost all of the women in continental asia were raped by them sure you can also find genetic clues with turkics that suffered from mongolian invasions( the ones who managed flee are in middle east caucasus and balkany will surely show less mongolic properties) —Preceding unsigned comment added by 88.244.184.162 (talk) 15:48, 1 April 2008 (UTC)

Whoever you are, the talk about physical genetics is irrelevant to discussion of linguistic relationship. Linguists use the term "genetic relationship" to talk STRICTLY about language--not about DNA, etc. American English has a genetic relationship with German even though 10% of our population is of Mediterranean and Mesoamerican Indian origin genetically, about 8% is of African origin genetically, etc. But our LANGUAGE has a genetic relationship with northern Europe no matter what our DNA says. So don't start talking about "rape" here. It's irrelevant. (Taivo (talk) 17:31, 1 April 2008 (UTC))

[edit] Phenetic?

While Greenberg's method of multilateral comparison is undoubtedly "controversial", I have removed its characterization as "phenetic". Phenetics is a method used in biological taxonomy and its validity as a descriptor of Greenberg's linguistic method is obscure. VikSol 04:29, 30 April 2008 (UTC)

I've never heard a linguist use the word "phenetic". I thought it was a typo for "phonetic". (Taivo (talk) 06:46, 30 April 2008 (UTC))

[edit] "postulated" to "arrived at"

Taivo, I see what you are getting at with the change of "found" to "postulated", but there is a problem of idiom with "postulating" something "using" something else. I have changed "postulated" to "arrived at", which I hope is neutral enough for you. VikSol 23:34, 1 June 2008 (UTC)

I'm a native speaker of English and I have no problem with "idiom" between "used" and "postulated". Einstein used mathematical formulas to postulate the theory of relativity. Ptolemy used the biblical text to postulate that the Earth was flat. Just because we "use" a tool to study an issue does not mean that we can't "postulate" why something happens when the tool is used. (Taivo (talk) 07:02, 2 June 2008 (UTC))

[edit] "Anti?"

I think that your list of "anti-Altaicists" is very misleading. It implies that "anti-Altaic" is a minority position, when it is, in fact, the majority position among historical linguists. You would need to mention most authors of historical linguistics textbooks (e.g., Lyle Campbell, Terry Crowley, etc.) who are all "anti-Altaic" and cite the controversy in terms of the confusion between areal features and genetic features. Just because they have not written a full-length monograph refuting the Altaic hypothesis does not mean that they are any less authoritative in being "anti-Altaic". It's fine to list prominent Altaicists with their proposals. I actually found the "pro" list quite informative. But leave out the "anti" list. It's not a true picture of the field in historical linguistics. (Taivo (talk) 08:57, 8 June 2008 (UTC))

Taivo, I'm happy to get your feedback. I agree we need to present a balanced picture. I too have been thinking that the nub of the controversy is areal versus genetic interpretations and that it is perhaps time to state this explicitly in the article, since it constitutes the real underlying issue. I think, though, that it is just as important to give people an overview of the principal critics of Altaic as of its advocates: Clauson began the critique, Doerfer has done the most substantive work on it, and Shcherpak and Rona-Tas are often considered the other most important critics of Altaic. I'm simply putting the names in that have come to my attention to this point. If the numbers are not equal, this might be simply because destructive criticism is more economical than constructing a language family, valid or invalid.
The headings "Altaicists" and "Anti-Altaicists" are not meant to take a position. I also thought of "Linguists in favor of Altai / Linguists opposed to Altaic" and "Pro-Altaic / Anti-Altaic". We could also try "Proponents of Altaic / Critics of Altaic" and the like. I would be equally happy with these or any reasonably equivalent set of expressions.
I have provisionally put the Anti-Altaicists back in, because I think the information given is very important.
It's true that not being an Altaic specialist (pro or con) does not mean a linguist may not have a very authoritative position. Also, a linguist may delve into these issues quite deeply without necessarily publishing extensive matter on the subject. I actually hesitated about Helimski, for instance, who has I believe done quite a bit of work on Turkic. I see you have already put him under the Altaicists.
So let me try to find unimpeachably neutral categories here, which would permit us to list as many linguists as relevant, while giving the reader an accurate view of the actual spread of prevailing opinion.
I have put in the material under new headings which hopefully recontextualizes it (or at least begins to) and added the names you suggest.
The lists given are not meant to be definitive. As you can see, they have holes in them, e.g. not all authors have a work cited yet. They will I hope be considerably expanded by interested parties. Regards, VikSol 09:56, 8 June 2008 (UTC)
Much better. I just didn't like the way the other format looked like a "vote". I tweaked it just a bit. I don't think we need a laundry list of everyone who has written a page against Altaic (Campbell, for example) or everyone who has listed it as a "bad idea" (Crowley, for example). I think if we leave it at "Major Critics" then we have the meat of the issue--those who have expended some serious ink (or toner?) on showing that Altaic is not a genetic unit without going through the last 40 years of publications in historical linguistics. Nichols may not have written a specific treatise on Altaic, but her methodology is the "in thing" right now and since her methodology doesn't support Altaic, she needs to be here. (Taivo (talk) 11:53, 8 June 2008 (UTC))