Talk:Japonic languages

Origins
The Origins section notes the following:


 * Japonic languages are related to modern Korean based primarily on near-identical grammar, but there is scarce lexical similarity between the two; supporters of the Buyeo languages theory generally do not include modern Korean as part of that family.

However, as I point out in the Classification section of the Japanese language Talk page, the first half of this statement displays faulty logic, and the second contradicts several other pages, notably the Korean language page, and the Buyeo languages page, among others. The logical flaw lies in that finding the grammars of any two languages to be similar, or even identical, does not prove relatedness but instead only proves similarity. This is not to say that Korean and Japanese are not related -- I personally hold the view that they spring from common ancestry, given my time spent studying the Japanese and Korean langauges and the history of East Asia. But in terms of making a solid point here on this page, one must look into the histories and linguistics of the two languages to say anything definitive about relatedness.

As to the other Wikipedia pages, I am no expert on the history of the Korean peninsula or language, but what I've read outside of Wikipedia does seem to back up the view that Samhan, Silla, Goguryeo, Gojoseon, and Buyeo are all related.

The Japonic languages page itself contains no citation of any verifiable source for the above quoted statement. Given the logical flaw and the contradictory views on other more extensively footnoted pages, I am inclined to think that this statement here is mistaken. Looking back at the page history, I find that Gilgamesh was the one to add this material. I would very much appreciate it if Gilgamesh or anyone else would be so kind as to add a source for this view, not least in that it would make this page and Wikipedia in general that much more useful for research. Without any such source, my personal feeling is that this page should be changed to be more in line with the other related pages. What do others think?

Thank you, Eiríkr Útlendi 23:17, 3 February 2006 (UTC)

Another bit from Origins:
 * As far as lexical studies have shown, the modern living non-Japonic language with the closest lexical similarity to any of the Japonic languages is Uyghur, a Turkic language.


 * In the wake of these theories, some argue that the similarity between all these languages is merely a sprachbund, and that the attested similarities between some or all of these languages are simply the result of their cultures being close geographic neighbors on the Asian mainland over the course of millennia.

I'm no expert on this subject, but certainly at the present time Japanese and Uyghur are nowhere near each other. I think this subject needs at least a little clarification on where/when such contact would have taken place, e.g. migration theories, reference to other languages that might be more likely to have had contact with proto-Japonic speakers, or links to other articles about them. KarlM 10:47, 21 February 2006 (UTC)

Terminology
User:Cibeckwith claims that Japonic languages is a disputed term an proposes Japanese-Ryukyuan languages instead. I have during my academic career never stumbled upon the latter term and it seems IMHO quite unwieldy. Could an expert on this subject please clear up the confusion? --Himasaram 08:01, 18 November 2006 (UTC)


 * Actually, User:Cibeckwith is an expert on this subject. Please correspond with him personally if you have any dissatisfaction with the nomenclature. For the time being, I am going to remove the unsightly "attention needed" template from the main page, because this article is actually very accurate as it stands, although it is rather bare-boned. Ebizur 00:42, 9 February 2007 (UTC)


 * Both are used. I personally prefer Japonic because it is shorter. It seems better to choose less node-based names for language families in case the classification is reorganized in the future or if varieties that were previously regarded as smaller subnodes are moved up to a higher position, the name is still relevant. --Node 13:58, 18 February 2007 (UTC)

Finnish
The language ofthe Finns is related to Japanese last I heard.Tourskin 22:50, 6 May 2007 (UTC)


 * No, this is pseudoscientific nonsense. Please see Pseudoscientific language comparison.--AAikio 08:42, 7 May 2007 (UTC)

Sorry, not trying to insult any one here, thats what my History teacher told me. Well he was wrong on a lot of things. Tourskin 22:13, 12 May 2007 (UTC)


 * Sorry for the edgy comment, didn't take it as an insult... It's just that there are a lot of wild theories of the origin of Japanese around, and in general most of them are not really worth taking into account in a linguistic article. --AAikio 06:26, 13 May 2007 (UTC)


 * See Ural-Altaic languages, a controversial theory. If valid, the languages are very distantly related, probably comparable to English' relationship to Sanskrit or Hittite, or even further. 惑乱 分からん * \)/ (\ (&lt; \) (2 /) /)/ * 14:41, 21 August 2007 (UTC)


 * Some people doing historical linguistics like to go for super-groupings. In which case it's possible for Finnish to be linked to Japanese. Probably of not much significance to most people. Some people say well it makes, for example, Japanese easier to learn for Finns. But obviously more Finns have learned English than Japanese. So again, not sure just how significant any of that is. — Preceding unsigned comment added by 133.7.7.240 (talk • contribs) 10:46, 14 June 2019 (UTC)

Japanoic/Ryukuan language with consonants in coda?
Hi guys, I'm wondering if this is true; I heard that one of the Japonic, I think it's a Ryukuan langauge, which uses consonants other than /n/ in coda. I could've swore I read it in Wikipedia before. Could someone verify this for me? I think it's one of the smaller ones, but finding out if this is true or if I dreamt it would be really useful. I posted this on the category section by mistake, so I've put the same question here hoping it'll get more hits. InnocentOdion 14:36, 30 November 2007 (UTC)


 * The Miyako language of the Miyako Island group and some dialects spoken in the area of Kagoshima Prefecture on Kyushu allow consonants other than /N/ in the coda of a syllable. Ebizur (talk) 16:48, 31 December 2007 (UTC)


 * In modern Japanese, one way of treating the so-called 'doubled consonants' would be in part as a syllable coda. Like, 'yap-pa-ri'. The first [p] being some form of glottal stop consonant (partial closure of the glottis with labialization at the other end of the vocal tract). — Preceding unsigned comment added by 133.7.7.240 (talk • contribs) 10:40, 14 June 2019 (UTC)

About Ainu languages?
Some one can talk about it...why Ainu languages not expressely mentioned.

Ainu languages, Ainu people, this people is native in Japan, the true Japanese.

Marcio Benvenuto de Lima (talk) 07:39, 29 March 2008 (UTC)


 * Simple. They are not Japonic. — kwami (talk) 09:12, 29 March 2008 (UTC)

Ainu languages are, generally viewed as, a language isolate. They are totally unrelated to any of the Japonic languages. There are some radical theories encompassing it into a ural-altaic family, however this is widely rejected. Mr anonymous username (talk) 18:37, 14 October 2020 (UTC)

classification
We have an ongoing edit war across several articles re. Japonic classification, with editors who won't bother to discuss their edits. Both editors are wrong in some respects. Japonic is often identified as Altaic by Altaicists. In fact, I've heard (though don't know myself) that most Altaicists accept Japanese, which would imply that the inclusion of Japanese is only somewhat more uncertain than the validity of Altaic as a whole. However, Japonic has not been demonstrated to be Altaic, and such classifications clearly remain controversial. On the other hand, it clearly is not an isolate: language isolates are languages; as soon as you have a family, you no longer have an isolate. I mean, we wouldn't say that Indo-European is "a language isolate (or possibly related to Uralic)". It is only considered an isolate when treated as a single language. Japanese generally was considered an isolate half a century ago (and AFAIK the Altaic theory is still a minority POV), but recent classifications have generally treated it as a small Japonic family.

Can we actually discuss these changes? The alternatives are that I edit-protect all the articles you are reverting each other on, which may mean that I protect them at a version that you disagree with, or that I block the both of you for disruption. (On further investigation, I might block only one of you, but don't count on it.) kwami (talk) 16:39, 24 October 2009 (UTC)

Protected
I've indefinitely protected the following articles which are involved in this pathetic edit war: Japonic languages, Japanese people, Altaic languages, and Classification of Japanese. Other articles may be added to this list as well if this edit warring does not stop. Please discuss whatever issues you may have and come to some sort of consensus. Once you have done so, we can look at making the article editable again. ··· 日本穣 ? · 投稿  · Talk to Nihonjoe 20:18, 24 October 2009 (UTC)

Remove protection and fix link!
link broken, here's the right one: http://www.ethnologue.com/show_family.asp?subid=1721-16

--Sigmundur (talk) 18:45, 15 November 2009 (UTC)


 * The protection has not been removed as there has been no discussion as far as I can tell. However, I have updated the link as requested. ··· 日本穣 ? · 投稿  · Talk to Nihonjoe 19:05, 15 November 2009 (UTC)


 * The protection on this page is inhibiting constructive editing. It's been over 5 months. Isolated incidents of edit warring, and persistent edit warring by particular users, may be better addressed by blocking, so as not to prevent normal editing of the page by others. I have no part in this edit war. - Zac  Bowling  (user 04:14, 10 May 2010 (UTC)

Tanegashima and Yakushima dialects
These dialects are classified in Satsugū dialect(ja:薩隅方言), not in Amami language(ja:奄美方言). 安室黎 (talk) 08:30, 9 May 2010 (UTC)

Classification as an Altaic language
The classification of the Japonic languages (and Korean) as Altaic languages is hardly an accepted matter. A tiny footnote about "Not always recognized as Altaic languages" is hardly NPOV when most linguists reject this classification and many reject even the existence of an Altaic language family. — Preceding unsigned comment added by 75.175.220.149 (talk) 16:07, 26 February 2012 (UTC)
 * Do you have sources showing this? WhisperToMe (talk) 03:19, 30 March 2012 (UTC)
 * This again? I thought this had been put to bed a long time ago. Taivo - are you out there?  I hereby grant you a permit from the United Nations to keeel this on the spot, whenever it occurs!  :-) Seriously, Whisper, even I as a historian have seen the (RS) statements that the majority of linguists reject the Altaic hypothesis for Japanese, or Korean for that matter. Incredible how this just keeps sprouting back up on Wiki articles like a weed. HammerFilmFan (talk) 01:57, 2 May 2012 (UTC)


 * The complaint seems to have nothing to do with the article. — kwami (talk) 07:58, 2 May 2012 (UTC)

Kaya toponyms?
Why, in the first part of the article, is Japonic given reference to Kaya over all the other major hypotheses such as Altaic, Koguryoic, Austronesian, and all the mishmashes between those? There is not a single piece of evidence proposed by any linguist that relates Japonic to a Kaya language. The Kaya confederacy was founded way after Japonic was already established in the Japanese archipelago. The reference to Beckwith's book is flawed, too, as the page does not mention anything about toponyms in Kaya, nor does it indicate any connection between Japonic and the Kaya language. This part should definitely be removed. Ramentei (talk) 03:07, 10 August 2012 (UTC)


 * On the Classification page you objected because it's *only* proposed by Beckwith; here you claim there is no such proposal. — kwami (talk) 06:40, 10 August 2012 (UTC)

Maybe you should read page 105 of Beckwith's book again. He mentions the Kara (Gaya) in relation to Japonic but doesn't offer any evidence or even imply that the two are related. If there is anyone else or any other writings of Beckwith that proposes a rigorous hypothesis relating Japanese to the Kaya language for its origins that has equal or greater standing with the Altaic and Austronesian theories (possibly even Koguryoic) for which there have been numerous writings and suggested cognates, then please enlighten me as I'm more than interested in new developments regarding the origins of Japonic. I'm just trying to help make the article more accurate, but I will stop editing if that disturbs you. Including a Gaya connection among the most attested origins of Japonic is just ludicrous to me. As I said I will stop editing but I will keep making constructive contributions on the "talk" pages. Hopefully both pages can be changed to something that is more complete, neutral and accurate. Ramentei (talk) 06:55, 10 August 2012 (UTC)

I suggest the Gaya part to be deleted, at least in the map with the (?). What is the justification for it? Also as someone has already pointed it out, Japonic cannot be an isolate. That is a basic linguistic error, and correction is long overdue. Something like "Possibly a branch of Altaic" would be more appropriate. Ramentei (talk) 08:03, 10 August 2012 (UTC)


 * Yeah, "isolate" is an odd word for a family. It's a holdover from when it was considered a single language. You can do this with most isolates: Tarascan, for example, as a small family rather than a single Purhepecha language. If you count Jeju as a separate language, then Korean is no longer an isolate, etc. And of course Ainu was perhaps a family not unlike Japonic, but is still commonly called an isolate (though in that case only one variety survives.) But the classification itself (or lack of one) hasn't changed in any of these cases. I removed the word 'isolate', but in the past this has caused problems with those who vehemently reject the Altaic theory. Perhaps you can think of better wording?
 * He may not imply this on p 105, but he does on p 28. Or at least with "pre-Kara", which he says cannot be demonstrably related to Gaya-era Gaya. Though he did include pre-Kara within Japonic, so yes, Gaya (if descended from pre-Kara) wouldn't count as a nearest relative. It would however still be the clearest connection outside what is generally recognized as Japanese. The toponymic material is widely recognized as being apparently related to Japanese. The main problem seems to be in identifying what the language of that material was, as reflected in B's hedging over whether it was Gaya or not.
 * I can certainly see ranking the proposals per level of acceptance, in which case Altaic is probably at the top. Austronesian was removed some time ago as not being accepted, but as you note, Dravidian is mentioned and I don't think there is any credible proposal for that. We could did Austronesian out of the article history. So maybe s.t. like (1) Altaic, (2) Korean, (3) Koguryoic [these three geographically connected], (4) Austronesian, (5) crackpot (Dravidian, etc.). Nostratic/Eurasiatic could be mentioned under (1), since they just break up Altaic within a larger proposal. — kwami (talk) 16:11, 10 August 2012 (UTC)

I see the rationale behind the "isolate", I guess, as there are still many Japanese people who think that Ryukyuan is a dialect. Can you elaborate on how an "isolate" would cause problems with vehement anti-Altaicists? It would seem that an "isolate" would be a more favourable expression for their case? Can you please refer me to some sources about the Gaya toponymic material? Last I read was that there were 13 toponyms, for which scholar's aren't even sure whether it's a Gaya language or an older language. That is why I was quite strongly opposed to Gaya having a mention, as 13 toponyms of unknown identity surely cannot be described as the front-runner for the relative or ancestor of the Japanese language. Are there more toponyms found and have they been deciphered to have some strong cognates and sound-correspondence to native Japanese words? I don't think B's recent book has much value in formulating linguistic connections as the whole book is more about history rather than about languages. His previous book about Koguryoic, however, contains substantial linguistic material and can be used as a valid source for articles relating to Japanese and other languages, regardless of what I or others think of the content ;) I completely agree with your ranking. Personally I would have Austronesian higher up, and also because the creole hypothesis is still very strong among the Japanese scholarly community, but since it's has a lot of skepticism among scholars outside of Japan and the Korean and Koguryoic hypotheses are both related in some capacity to the Altaic hypothesis, I think your ranking it very reasonable. "Crackpot" might be a bit harsh on Dravidian, Susumu Ono was a very well renowned Scholar on the Japanese language, but unfortunately ventured into comparative linguistics without much knowledge in that field. His works are, nevertheless, the only attempts to seek answers for the apparent lack of cognates in words relating to agriculture among East Asian languages. It was a genuine attempt, but a failed one. I think the Japanese version of the article is very sound, the only quibble being that they give too much weight to Dravidian. Ramentei (talk) 03:31, 30 August 2012 (UTC)


 * It's removing the word "isolate" that has caused problems with anti-Altaicists.
 * I don't know how many toponyms there are. Several of them seem to be transparently Japonic.
 * Yes, as I said, it's "pre-Kara/Gaya" that's the toponymic material, with too little to link it to Kara/Gaya proper. We could change Kara/Gaya to "pre-Kara/Gaya", to clarify that, though the link would probably be a rd. to the same article. — kwami (talk) 04:15, 30 August 2012 (UTC)

I don't want to sound like a broken record here, but I'm still not convinced that the Gaya/Kara hypothesis has any credibility. If, as you say, several of the toponyms are considered to be clearly Japonic, it needs to be stated clearly which toponyms and how it is perceived to be Japonic i.e. how it cognates to OJ vocabulary. There's not enough written about Gaya/Kara's linguistic relationship to Japanese compared to Altaic, so if it or the pre-language is to be stated as a highly possible relative to Japonic, then I strongly think that all the cards need to be laid on the table, in a similar way examples of possible cognates are given for the Korean and Altaic hypotheses. Otherwise I don't see how it has any more credibility than Dravidian. Ramentei (talk) 06:29, 31 August 2012 (UTC)


 * For our purposes, credibility does not depend on whether you accept it, but on whether our sources support it. They do. (Even some of those which dispute the Koguryoic hypothesis acknowledge that the toponyms are Japonic.) We don't need to lay out the evidence for your evaluation. That would violate WP:OR. However, if you wish to expand the coverage, you of course may. — kwami (talk) 09:01, 31 August 2012 (UTC)

I'm not asking you to lay out the evidence for MY evaluation, but for the readers, or at least refer a better source that explicitly gives Kara toponyms. Beckwith gives numerous examples of Koguryoic toponyms that may relate to Japanese but where are the references to the Kara toponyms? Especially on the classification artical, the part about Kara gives no sources for reference at all. If the Altaic and Korean sections can give a substatial list of vaguely similar words, why can't the same to done for Kara, since you (or whoever included the Kara stuff) claim that it is the closest attested language to Japonic? Ramentei (talk) 03:10, 1 September 2012 (UTC)

Also if what you say here is true: "Even some of those which dispute the Koguryoic hypothesis acknowledge that the toponyms are Japonic." then why not list those sources? That would adequately show readers that Kara is a reasonably well-reasoned hypothesis. So far I only see p.105 of B's latest book referred in either article, for which I quite clearly showed that the page does not suggest a Kara-Japonic linguistic connection. And listing the Kara hypothesis is all fine, but statements like "has the most currency" "clearest connections" is completely subjective and unfit for a wiki article, when they are so many competing theories out there. Ramentei (talk) 03:22, 1 September 2012 (UTC)


 * That would require going back to sources I no longer have access to.
 * If pre-Kara is closer to Japanese than Ryukyuan is, then it is obviously closer than Korean or any other language is.
 * I don't understand what you mean by "does not support". Again, it's not generally for us to second-guess our sources; that would be OR. — kwami (talk) 09:03, 1 September 2012 (UTC)

Well, since the assertion that Gaya/Kara is related to Japonic depends on those sources, they must be available to all readers. If they cannot be provided, statements that presume that Gaya/Kara relationship to Japonic is a solid hypothesis should not be made. I actually think the way you have organised the Kara-language article is much better. It explicitly states that it is what Beckwith proposes, rather than taking for granted that it is the strongest hypothesis concerning Japonic relatives.

Comparing it to Ryukyuan is meaningless, since Ryukyuan probably did not exist in Pre-kara times. The divergence and geographic isolation for centuries has given birth to massive differences now, but Old Ryukyuan could well be very similar to Old Japanese. I think it's worth considering that there aren't any other linguists who are writing anything of note regarding Japonic and Kara. On the Kara language page listing all Beckwith is ok as he is the sole source, but on a page dealing with one of biggest and the most controversial topics in the history of linguistics, putting forth a theory supported by only one person as the most well attested and with poor referencing needs to be reconsidered. What you believe or how well you know the sources that you can no longer access is not the point. The "clearest connections" or the theory with the "most currency" needs to be something that the a large portion of the linguistic community agrees on.

As for p.105 in B's latest book, it says “The one area that seems to have escaped the nation building of the Puyo-Koguryo people, as well as the influence of their language, was the realm of former Pyon Han, in the central part of the south coast of the Korean Peninsula. It became known as Kara, or Minama, and never achieved political equality with the other kingdoms of the peninsula. Little is known about Kara, but it was under heavy Japanese influence and at time was a Japanese tributary state, if not an outright colony.”(Beckwith, 2009) That is all it says about Kara on that page, and nothing about its linguistic affiliation to Japonic, so I fail to see how that serves as a reference for the statement: "The clearest connections seem to be with toponyms in southern Korea which may be in Gaya (Kara) or other scarcely attested languages." Ramentei (talk) 08:59, 4 September 2012 (UTC)


 * "they must be available to all readers". No, that's not how things work here. We don't need to republish the sources, just cite them.
 * I've already agreed that other hypotheses are more widely accepted.
 * I think you're looking at a different book. B. clearly shows pre-Kara as being closer to Japanese than Ryukyuan is. — kwami (talk) 16:37, 4 September 2012 (UTC)

I apologise, I missed the p.27-28 of the other book being cited further below. But the part I quoted is definitely from p.105 from "Empires of the Silk Road". And the citation is slightly dubious. It does help that it is adequately backed up later. I agree, they just need to be cited, which I mistakenly thought was not done sufficiently. As for how strong or "clearly shown" B's theories are, I don't think we will ever come to an agreement so I rest my case here. Ramentei (talk) 22:41, 4 September 2012 (UTC)


 * I haven't really read B's stuff (or heard him speak) for some time, and he's changed his position since I have. When he was first proposing this, it was as evidence that Koguryo was the closest relative to Japonic. His position has now shifted to toponymic evidence that Japonic was once spoken in Korea, without claiming that this was in any known language. Much of the criticism of B's earlier work was that the links between these Japonic-looking toponyms and Koguryoic were spurious, and B has evidently accepted this by abandoning that part of his proposal. But I don't recall anyone ever denying that the toponyms do indeed appear to be close to Japonic: they only denied that those toponyms could be Koguryoic. (Not to say that there aren't people who deny both.) — kwami (talk) 02:16, 5 September 2012 (UTC)

Yes, it's not only Beckwith but linguists haven't been speaking much at all about Japonic for some time. B's list of Koguryoic toponyms don't look Japonic at all to me. The cognates he proposes with Japanese words show poor sound correspondence and the two words have significant gap in meaning that chance resemblance is hard to rule out. It is not much better than the Altaic correspondences proposed by the likes of Miller and Robbeets. I know the yapma-yama "cognate" (Beckwith 2007, p.121) has been sharply criticised as many believe the archaic for to be "dama", as shown in some Ryukyuan languages. However, in B's defense, there is no way of conclusively telling whether the "d" or the "y" is archaic in Proto-J, and in OJ voiced consonants don't appear in word initials. But his other examples, for instance, show too many CVC or VC stems, which are not features of Japonic. That is the big problem I have with the Altaic hypothesis as well. In Mandarin Chinese, for example, that doesn't have C endings, the vowels are very complex and can be clearly compared with various dialects to see how a -k or a -t ending developed into a compound vowel. In Japanese, however, the vowel system is so simple and that I don't see how all the C ending stems in Altaic could produce a standard 5 vowel language with such simple phonology, if they are closely related languages. Sorry I've gone off topic a bit here. Ramentei (talk) 10:07, 5 September 2012 (UTC)


 * Japanese could easily have lost final consonants (Austronesian substratum influence??) without affecting the vowels—that happens all the time, & such typological characteristics are not a good argument for relatedness (witness Austronesian itself). This is, however, hard to square with pre-Kara falling within Japonic. The pre-Kara comparisons include grammatical particles. And *y → d is a Yonaguni innovation postdating Chinese loans, which underwent *y → d as well. — kwami (talk) 16:29, 5 September 2012 (UTC)

You are right, Japanese could have lost final consonants or added vowels to final consonants as well, like they do to European loan words of recent times. But as indicated by those recent loans, such radical shifts in pronunciation is more likely due to loaning rather than relatedness (divergence). B also lists Okog "atsin" (poor) as cognate with OJ "ashi" (bad, evil) (p.121), but the "shi" in OJ is a grammar affix for adjectives, so there is no way it cognates with "sin". There are other examples where B proposes cognates with OJ adjectives for Okog words that have "si" without the nasal ending. This clearly violates sound correspondence without adequate explanation for exceptions. Austronesian substratum influence is possible, but as there is no evidence of Austronesian contact around the time of influence from the Korean peninsula, that would mean that a Austronesian language was spoken in Japan for quite some time and right into the Yayoi period. Not completely implausible, but that would raise the argument of which is the super and which is the substratum. I agree that the grammar particles that supposedly cognate with pre-Kara, and some Koguryoic ones as well, are some of the better evidence. Ramentei (talk) 03:40, 6 September 2012 (UTC)

Classification yet again
How come that Korean language is listed as isolate, while Japonic languages are listed as Altaic, at least colour wise? Is that NPOV? The common view is that there are not conclusive evidence of genealogical relations between Japanese and Korean, let alone Japanese and classic Altaic languages: that's the equivalent of isolate (=unknown family relation) until better proofs are found. Moreover, prominent Korean language experts, even sympathetic toward the Altaic theory like Ki-Moon Lee (2011), have not yet found conclusive proves of genealogical links between Japanese and Korean and they are still very much open to the debate. The Concise Encyclopedia of Languages of the World edited by Elseviere (2006) considers Japanese as an isolated language, as well as Korean. 84.222.239.177 (talk) 18:26, 26 September 2012 (UTC)


 * It's due to people editing one article but not the other. — kwami (talk) 20:18, 26 September 2012 (UTC)


 * It looks like an edit war between mules, to say the truth... – 84.222.239.177 (talk) 21:12, 26 September 2012 (UTC)

Language isolate
The coloring for Altaic in the infobox is highly problematic. The Japonic languages are nearly universally rejected as being "Altaic". Should the infobox be colored "Altaic" or "Language isolate"? --Taivo (talk) 10:24, 5 February 2015 (UTC) Language Isolate (generally accepted) Altaic (?) or       Language Isolate/Altaic (?)
 * Language Isolate. Altaic is a dead proposal.  Either the color for "Altaic" needs to be clearly marked "areal" or else these group needs to be colored as Language Isolate.  "Altaic" should not be included in the infobox classification in either case.  --Taivo (talk) 10:28, 5 February 2015 (UTC)
 * Language isolate. I agree with Taivo—as long as the Altaic hypothesis does not have widespread support among historical linguists, we should use the "language isolate" coloring for this page. —Granger (talk · contribs) 02:46, 6 February 2015 (UTC)
 * Language isolate - no conclusive evidence of the Japonic languages being Altaic, or any other classification.  ミーラー強斗武   (StG88ぬ会話) 03:48, 6 February 2015 (UTC)
 * Altaic. Japonic is not an isolate, it's a family.  Also, the colors are not specifically families:  Amerindian, Papuan, Australian, Caucasian and Khoisan are all geographic, and Nilo-Saharan is still at the proposal stage. (Taivo, how would we mark a color as "areal"?) In any case, we can't very well remove the Altaic coloring when the last consensus is that we mention Altaic in the info box of all top clades (with caveat). — kwami (talk) 05:35, 6 February 2015 (UTC)
 * Language Isolate - Colouring should be Language Isolate with linguistic classification being something along the lines of:
 *  Lux ure Σ  09:19, 9 February 2015 (UTC)
 * Wikipedia doesn't add questionable information with "(?)" after it, anywhere, ever. If you ever see someone do that in an article, you should remove it immediately, per WP:Verifiability.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:06, 2 March 2015 (UTC)


 * Language isolate. "Altaic" is a controversial theory.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:05, 2 March 2015 (UTC)

"Altaic ?" in the Infobox?
In the Infoboxes for Turkic languages, Mongolic languages, Tungusic languages, Japonic languages, and Koreanic languages and most, if not all, of their daughters, assigning the color "Altaic (areal)" to the infobox automatically fills in the text "Altaic ?" in the "family" line of the classification at the top node unless it is specifically overridden by inserting another value into the "family" slot. This should be eliminated since Altaic has been generally discredited among historical linguists and only a small minority of linguists still cling to it. --Taivo (talk) 08:17, 6 February 2015 (UTC)


 * Support. "Altaic" as a genetic unit is approaching fringe status in the community of historical linguists.  Fewer and fewer linguists give any support to it except as an areal grouping.  Only a small number of linguists in the Starostin camp still cling to it.  It is, however, virtually dead as a serious genetic unit.  We should stop giving any credence to it by removing the automated "Altaic ?" tag generated by assigning an infobox color for all these families, subclades, and daughters.  --Taivo (talk) 08:17, 6 February 2015 (UTC)


 * Oppose without a centralized discussion and consensus. Our previous consensus was to list "Altaic" (with some sort of caveat) at the highest nodes (Turkic, Mongolic, Tungusic, Korean(ic), Japonic) and the major languages (Turkish, Mongolian, Korean, Japanese) and to omit it for all other languages.  It seems we can now forgo the major languages without much problem.  It is common practice on WP for us to include controversial groupings at the highest levels (the immediate branches of the proposal) and to omit them from lower nodes and individual language articles.  If Altaic truly is now dismissed to the extent that e.g. Caucasian, Australian, and Khoisan are, then we could use the color without ever including the name in the info box, but IMO that would require a demonstration that this is indeed the current state of knowledge.  And we should avoid a separate discussion for each constituent family.  — kwami (talk) 17:46, 6 February 2015 (UTC)
 * Where is the best forum for that centralized discussion? --Taivo (talk) 19:34, 6 February 2015 (UTC)
 * Altaic languages would be the most obvious, but maybe on the WP:lang talk page. — kwami (talk) 20:04, 6 February 2015 (UTC)

Language Isolate (generally accepted) Altaic (?) or       Language Isolate/Altaic (?)
 * Support - Colouring should be Language Isolate with linguistic classification being something along the lines of:
 * Lux ure Σ 09:19, 9 February 2015 (UTC)


 * Note: A centralized RfC has been initiated here. --Taivo (talk) 17:44, 7 February 2015 (UTC)


 * Comment: What is needed to validate the approach being considered by the RfC is a (secondary, reliable) source which explicitly supports the claim that Altaic is regarded as an anachronistic unit in this context.  Trying to establish the same claim through an informal and impressionistic census of the sources conducted by our own editors is too close to the cusp of original research with regard to synthesis.  S n o w  talk 02:23, 19 February 2015 (UTC)


 * "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. (And that was the first book I picked up off my shelf to locate a relevant quote.)  --Taivo (talk) 08:40, 19 February 2015 (UTC)


 * Works for me; truth be told, I commented before I realized that the RfC bots had once again directed me to a stale discussion and that this issue had already been more or less unanimously decided through the related discussion at WP:WikiProject Languages. But I appreciate the response all the same.  Lyle Campbell -- now there's a name I haven't heard in a good long while!  S n o w  talk 18:21, 19 February 2015 (UTC)


 * Concur with Taivo.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  11:17, 20 February 2015 (UTC) PS: To repeat what I said in the previous RfC: 'Wikipedia doesn't add questionable information with "(?)" after it, anywhere, ever. If you ever see someone do that in an article, you should remove it immediately, per WP:Verifiability.'  If we need to include information on a questionable or controversial classification (because the classification and the controversy about it are well-sourced), we do this in main article prose, not by misleading/confusing question mark labels in infoboxes.  This is not a blog or someone's notebook, it's an encyclopedia.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:11, 2 March 2015 (UTC)
 * Support - There's no basis of being isolate language. Besides, there are some proofs that it's connected to Altaic langugages.--Kafkasmurat (talk) 09:25, 23 March 2015 (UTC)
 * Japaneese Wiki made it Altaic (https://ja.wikipedia.org/wiki/%E6%97%A5%E6%9C%AC%E8%AA%9E) Must be mentioned, at least.--Kafkasmurat (talk) 09:28, 23 March 2015 (UTC)
 * Japenese wiki is not a reliable source for English Wiki. It is mentioned--in the text where it belongs.  I doesn't belong in the infobox.  --Taivo (talk) 13:01, 23 March 2015 (UTC)

Reconstructions in the Japonic_languages section
I'm curious if anyone has any more information about the table there.


 * The symbols used are opaque to most readers. This is a basic usability problem.  Even if a user wanted to learn what the symbols mean, there's no obvious way to find out.
 * → Could someone either add an explanatory paragraph, or a link to a page that explains these?


 * A number of the items listed are problematic.
 * cold | *sàmù- -- There is a cluster of "cold"-related terms that all start with sam-, indicating that the final ù in the reconstruction is probably some kind of grammatical ending, and is thus inappropriate for a root form.
 * give | *ata[-]pa-Ci -- Presumably this is postulated based on modern verb ataeru, ancient form atapu. This is generally derived from root ata in the Japanese materials I have to hand.  If that equates to the initial ata- in the reconstruction, what is the final pa-Ci supposed to be?
 * louse | *sìrámí -- Likely a compound of sira ("white") + mi (uncertain, possibly "body"). See also nomi ("flea"), containing that same mi element.
 * warm | *àta-taka- -- The root appears to be atata, not ata + taka. See related ancient verb atatamu (whence modern atatameru, atatamaru), recognized adjective-forming suffix -ka (as in shizuka, komaka, oroka, etc.).

Thanks in advance, ‑‑ Eiríkr Útlendi │Tala við mig 21:23, 14 November 2018 (UTC)

Classification section to much
We could sum up the content, as mostly everything is repeated in the main article "Classification of Japonic languages". What do other editors think about that?--AsadalEditor (talk) 18:00, 18 February 2019 (UTC)
 * I will reduce the amount, if someone disagree, feel free to revert my edit.--AsadalEditor (talk) 13:00, 19 February 2019 (UTC)

Vovin's Reconstructions
Vovin's 1994 reconstructions listed here don't use the generally accepted notation for Proto-Japonic. He uses  for,  for , and has  all over the place from back when he was still friendly toward Whitman's -loss hypothesis. He also favors the lenition hypothesis (since this was back when he was still pro-Altaic), which is decreasing in whatever popularity it had. Finally, he uses  instead of  everywhere for the indeterminate nasal component of voiced consonants.

I don't have anything to say about his reconstructed pitch accent, since I'm not familiar with that particular aspect of Proto-Japonic, but I'd like to update/clean up this list using more recent research by, say, Pellard. LhikJovan (talk) 03:32, 21 May 2019 (UTC)
 * That sounds reasonable. There doesn't seem to be a usable consensus on how to reconstruct the pitch accent.  We could just write accent classes (2.3a), etc, after the word.  Kanguole 09:50, 21 May 2019 (UTC)
 * I was gonna go ahead an make the edit, but I figured it'd be better if I ran it through here first, since I know it still needs some work. I grabbed a bunch of words from Thorpe, Vovin, Pellard (and probably others that I'm forgetting, since I just have a spreadsheet of this stuff) and made this list. stands for unknown  or, while  stands for unknown (?), , or . Finally,  represents an unknown nasal consonant  or :


 * *aka — red, bright
 * *ama-i — rain, sky
 * *aNpura / amura — grease, fat
 * *arata — new
 * *asa — morning
 * *asu / as(U)ita — morrow
 * *awo-i — blue, indigo (plant)
 * *əpə — large, big
 * *iNta-, *iNta-i- — exit, go out
 * *ika — squid, cuttlefish
 * *inu — dog
 * *ipia — house
 * *itu — five
 * *ka — deer
 * *ka-i — fur
 * *kam- — chew
 * *kama-i — turtle
 * *kamu-i — god
 * *kana-i — metal
 * *kapa — river
 * *kapa — bark, skin
 * *kapaNpori — bat (animal)
 * *kə- — come
 * *kəkənə — nine
 * *kəkərə — heart
 * *keNsu — injury
 * *ki-apu — today
 * *ki-asa — this morning
 * *ki-nə-pu — yesterday
 * *kik- — hear, listen
 * *kimo — liver
 * *koNpO — spider
 * *ko-i — yellow
 * *kOpo-i — love
 * *kumO — cloud
 * *kusori — medicine
 * *kutu-i — mouth
 * *ma-i — eye
 * *memeNsu — earthworm
 * *mətə — root, base
 * *mi — three
 * *mi- — see
 * *miNsə — ditch
 * *miNtəri — new plant growth
 * *mit- — fill
 * *mosi — insect, snake
 * *mu — six
 * *mu-i — body, fruit
 * *muna-i — chest, breast
 * *na — name
 * *nana — seven
 * *naNka- — long
 * *nəm- — drink
 * *omi — sea
 * *pa — feather, leaf; blade, tooth
 * *paNpiru — butterfly
 * *paNtu-i — shame
 * *para — belly
 * *para — field
 * *paru-i — needle
 * *pata — twenty
 * *pi — sun, day
 * *pisa / piNsa — knee, leg
 * *poi — fire
 * *puna-i — boat, ship
 * *puta — two
 * *s(U)i — beast, meat
 * *s(U)ita — tongue
 * *sam- — cold, chilly
 * *sirau — white
 * *sO — tens, -ty
 * *su — nest
 * *suNkUi — cedar
 * *suna — sand
 * *susu — soot
 * *tai — hand
 * *tama — ball, jewel
 * *tana-i / sana-i — seed
 * *tat- — stand
 * *təri — bird
 * *təsi — year
 * *təwə — ten
 * *tuko-i — moon
 * *tuma-i — nail
 * *tunO — horn
 * *tut(U)i — land, earth
 * *wo-, *wo-i- — sit
 * *ya — eight
 * *ya — house, shop
 * *yak- — burn, cook
 * *yə — four
 * *yo — night
 * *yUka — floor


 * Maybe it'd be best to exclude ones with *U or *O for clarity's sake. I'm also not sure how to best format the final on many of the nouns. Thirdly, while it's less of a problem, the list is lacking in verbs and adjectives. Fourthly, it's lacking pitch accent. And, finally, none of this is explicitly sourced; all I've really done is invoked some big authors' names even though I really just pulled from my spreadsheet (which probably accidentally grabbed some independent research with it). Anyone should feel free to take this list and improve upon it/tear it apart in whatever way he or she sees fit. LhikJovan (talk) 11:36, 21 May 2019 (UTC)
 * The lack of sourcing will be a problem. Without it, people will just swoop in and make changes and it will be impossible to know whether they are correct. Note that we already have a sourced comparative table for the numerals. It might be useful to categorize the words in some way instead of having a single long list. Kanguole 12:23, 21 May 2019 (UTC)

Branches
I have reverted the assertion of four branches, whose definiteness goes beyond the scholarly consensus: None of the sources talk about four branches. Kanguole 00:15, 10 July 2019 (UTC)
 * The position of Hachijō is described as "a matter of speculation" (Shibatani) and "unclear and deserves more investigation" (Pellard).
 * The proposed evidence for extinct Japonic languages in Korea is fragmentary and cannot be said to be "universally accepted" (as the relationship between Japanese and Ryukyuan is), though quite a few scholars consider it plausible.

Regarding this quotation: "The genetic relationship between Insular Japonic and Peninsular Japonic is universally accepted in modern scholarship except by the majority of scholars in both South and North Korea, where, unfortunately, the general attitude towards the issue remains politically motivated rather than based on scholarship."

In general, it is better to be sparing with quotations, presenting ideas in our own words. In this case, the added attack on Korean scholars is another reason to avoid quoting.

Nor does Vovin's statement justify treating a peninsular branch as an undisputed fact in Wikipedia's neutral voice. Vovin is a participant in this debate, having advanced much of the proposed evidence himself, but even he acknowledges that there is significant scepticism (while suggesting it is political). Other scholars (e.g. Pellard), while favourably disposed to the idea, are less categorical about it. Kanguole 15:59, 21 July 2019 (UTC)


 * I agree with . Apart from Vovin being a participant in the ongoing debate, his statement is in itself contradictory: if a proposal is rejected "by the majority of scholars in both South and North Korea", its acceptance can't be called "universal". The "Oxford Research Encyclopedia" has a concept that differs from WP. By inviting leading scholars to contribute, the former accepts the risk of biased articles, or as we call it here, "POV-pushing". –Austronesier (talk) 16:31, 21 July 2019 (UTC)

Recent deletions from the lead
An IP editor appears to be concerned that the lead claims that Japanese is related to Korean. It does not claim that. Rather it says "Possible genetic relationships with many other language families have been proposed, most systematically with Korean, but none have been conclusively demonstrated." All three parts of this statement are amply supported in the article body, but note the conclusion: none have been conclusively demonstrated.

The other statement summarizes the separate issue of evidence for Japonic on the peninsula in ancient times: "There is also some fragmentary evidence suggesting that Japonic languages may once have been spoken in central and southern parts of the Korean peninsula." This does not claim that such presence is generally accepted: words like fragmentary evidence and may convey the lack of certainty. But it is considered likely by a significant number of scholars, and this is well supported in the article body. Kanguole 22:59, 26 July 2019 (UTC)

Migration, Peninsular Japonic, etc
You ask for the basis of the claim "Most scholars believe". Well, the sources cited on that sentence say: It is anachronistic to refer to Peninsular Japonic at that stage, when it is simply Japonic. Peninsular Japonic refers to evidence of Japonic on the peninsula in the first millennium AD, i.e. language(s) of the descendents of those who did not make the crossing.
 * Serafim: "It is widely thought that Japanese language entered Japan from the Korean peninsula together with the carriers of the Yayoi culture, who also brought with them the tool kit for wet-rice agriculture"
 * Vovin: "It is also universally accepted today that Japonic speakers probably migrated from the Korean Peninsula to Japanese islands around 700–300 B.C."

I have reverted the demotion of Japanese and Ryukyuan to subsections of an otherwise empty Insular Japonic section. This is unnecessary elaboration, and also places undue weight on the presentation in Vovin (2017).

Finally, you have an unfortunate habit of inserting closely paraphrased material from Vovin (2017) without regard to what is already present in the article, often leading to duplication and/or confusing structure. Kanguole 17:54, 24 October 2019 (UTC)


 * Both your examples of Serafim and Vovin describe the migration of Japonic speakers from the Korean peninsula, but "widely thought" does not mean "Most scholars believe" which is what you wrote. This theory is not mentioned on the Japanese page about Japonic languages. The Japanese page says "Japanese is regarded as an “ isolated language ” for which the existence of a language of the same family has not been proven." On the Japanes page they describe Peninsular and Insular Japanese as two different classifications of the Japonic language family. The Korean peninsula is not the birthplace of the Japonic languages. That is not clearly mentioned in the article.


 * I inserted Vovin's findings partially paraphrased, because it's more trustworthy than your description of referenced material. It is proper to refer to the Japonic languages from the Korean peninsula as "Peninsular Japonic". Because the word "Peninsular" refers to the Korean Peninsula. Japan is not a peninsula. So it clarifies the languages that are referred to.


 * Vovin also said "It appears that combined Insular and Peninsular Japonic once was a much more varied language family than it is today and that at its height it was spoken both on the Korean Peninsula and on the Japanese islands. It is also universally accepted today that Japonic speakers probably migrated from the Korean Peninsula to Japanese islands around 700–300 B.C. and eventually assimilated the local aboriginal languages. Roughly one thousand years later, the Peninsular Japonic languages shared the same fate, being gradually assimilated by the Koreanic languages."


 * The "Japonic speakers" who migrated from the Korean Peninsula 700–300 B.C. did not speak what is now known as the Japanese language. Classical Japanese, Early Middle Japanese and Modern Japanese (part of the Insular Japonic languages) evolved in the Japanese archipelago.


 * You wrote "was brought to northern Kyushu from the Korean peninsula around 700–300 BC by wet-rice farmers of the Yayoi culture".


 * If they were of the "Yayoi culture" then they were the "Yayoi people". You removed the term "Yayoi people". A culture cannot cross the sea. The people who migrated were the "Yayoi people".


 * You wrote "and spread throughout the Japanese archipelago, replacing indigenous languages", but according to Vovin the aboriginal languages were eventually assimilated. Assimilation is different than replacing.


 * I added information about the different Japonic languages that existed on the Korean peninsula according to Vovin: "Silla Japonic language Koguryǒ, Karak, and Paekche, and the Japonic language of Chejudo." You removed that information.


 * The article says "There is fragmentary placename evidence that now-extinct Japonic languages were still spoken in central and southern parts of the Korean peninsula several centuries later."


 * But the term "Japonic languages" refers to the whole Japonic language family. This is specifically about the Peninsular Japonic languages that became extinct on the Korean peninsula around the 9th century. You removed the information that the Japonic languages were gradually assimilated by the Korean languages in the Korean peninsula.(Artanisen (talk) 18:55, 24 October 2019 (UTC))


 * "Most scholars believe" is about halfway between: "It is widely thought..." and "It is [...] universally accepted...". –Austronesier (talk) 19:25, 24 October 2019 (UTC)


 * What other Wikipedia pages say is irrelevant – it is reliable sources that matter.
 * This article says nothing about the birthplace of the Japonic languages. Nor should it, as there is no scholarly consensus to report.
 * In the Vovin passage you quote above, he talks of Japonic speakers migrating to Japan, and of the division between Insular and Peninsular Japonic as being after the migration. Of course, the descendents of the Japonic speech that migrated are the Insular Japonic languages.
 * The text doesn't say the Yayoi culture migrated; it says the wet-rice farmers of the Yayoi culture did. (like Serafim's "carriers of the Yayoi culture")
 * The "Peninsular Japonic" section introduces a bulleted list of different pieces of evidence. Apparently not realizing what the items were, you added in the middle of that list an item According to geography and the number of attestations the following Peninsular Japonic languages were found: the Silla Japonic language Koguryǒ, Karak, and Paekche, and the Japonic language of Chejudo. based on Vovin's on the basis of geography and in descending order according to the number of attestations of glosses, those are the Silla Japonic language Koguryǒ, Karak, and Paekche, and the Japonic language of Chejudo.  (even reproducing Vovin's missing punctuation)  Did you not realize that was what the whole section was about, but in greater detail?
 * Everyone knows that the Korean peninsula became uniformly Korean-speaking, and the former presence of Japonic languages on the peninsula is widely accepted, but there is no consensus on exactly where they were spoken, or when they disappeared. Kanguole 21:03, 24 October 2019 (UTC)

Material on Jōmon, typology
There are some problems with the recently added material and references: Kanguole 20:16, 23 April 2021 (UTC)
 * Chaubey and van Driem (2020) advances the theory that the Yayoi were para-Austroasiatic, but switched to the Japonic language of the Jōmon. As far as I've seen, van Driem is the only one advancing this idea (based on the distribution of Haplogroup O-P31). The Miller theory mentioned is rather different, and is obsolete.
 * Elmer (2019) is an MA thesis, so not a suitable source, but it doesn't seem to say what has been ascribed to it. The idea that Tōhoku was once Ainu-speaking is widely accepted, but the mechanics of typological influence is another matter.
 * The quote from Yurayong and Szeto (2020) is not quite accurate: the original is "our results indirectly speak in favour of a “Paleo-Asiatic” origin of the Japonic and Koreanic languages". This doesn't seem to mean much more than "non-Altaic".
 * I have again reverted the addition of Chaubey and van Driem (2020). Neither of them have worked on Japonic languages: Chaubey is a geneticist and van Driem specializes in Himalayan languages. They rely on Miller (1971) for the theory that the Jōmon spoke Japonic, and that obsolete theory is already covered. Kanguole 11:51, 11 December 2021 (UTC)

Glottolog classification
Including the tree from Glottolog seems hard to justify. The authors haven't worked on Japonic, but have somehow synthesized a classification from other works. As with most of their pages, they give a list of references, but no detail on how they were used. In this case, Pellard's thesis is cited for the subclassification, but it doesn't match the given tree: the algorithmically-generated phylogenetic trees on pp273–276 all have a bifurcate branching, with Hachijo grouped with Japanese. Tsugaru and Old Japanese aren't mentioned. Kanguole 09:09, 22 July 2023 (UTC)

Japanesic?
The idea of a "Japanesic" branch seems to be a particular creation of Glottolog, who have done no work on Japonic languages, and does not appear elsewhere in the literature. Adding it to the infobox seems to be undue weight, and does not reflect the article. (Not to mention that the name is a hideous creation.) Kanguole 04:33, 24 September 2023 (UTC)


 * I agree. Although the hypothesis that the Kagoshima and Tsugaru dialects are divergent enough to be considered distinct languages from the rest of Japanese is worth a brief mention in the Japanese dialects article, I don't think it's worth putting "Japanesic" in the infobox. As you said, the name is not frequently used in Japonic linguistics and it's essentially synonymous with Japanese anyway, even if it's split into multiple languages based on mutual intelligibility. ~Cherri of Arctic Circle System (talk) 20:53, 27 September 2023 (UTC)
 * I should also note that Glottolog's splitting of Kagoshima and Tsugaru into separate languages within their database does not constitute a dispute of their subclassification within the broader Kyūshū or Tōhoku dialect groups respectively. They are displayed outside of those groups in the database as of now as, due to the limitations of their website's UI, divergent dialects cannot be listed as both distinct languages and as parts of a larger dialect group in which the other dialects are considered part of the main language. I asked about this on the project's GitHub page a while ago and one of the authors clarified this and said there were no plans to change it, which is unfortunate. ~Cherri of Arctic Circle System (talk) 21:02, 27 September 2023 (UTC)