Talk:Uralic languages/Archive 4

Archive 1

Archive 2

Archive 3

Archive 4

Archive 5

POV tag over removal of Finno-Ugric and Samoyedic as branches of Uralic

Seems some editors appear to be continuing to push a minority POV by removing of any mention of Finno-Ugric and Samoyedic as branches of Uralic, as seen here, here and here, so a POV tag has been placed.

The fact that Finno-Ugric and Samoyedic are recognised by ISO 639-5 as branches of the Uralic language family points to the international consensus around that classification. For anything to be enshrined into this standard, there first needs to be sufficient level of agreement between linguists. ISO 639-5 has undergone an extensive peer review process before being approved, (and continues to be regularly reviewed and subject to change[1]), therefore we can view that classification contained within it as the consensus view.

So what is the basis for removing any reference to Finno-Ugric and Samoyedic as discussed here?

• Kwamikagami alleges consensus was established in a prior discussion[2] but failed to provide a link to that discussion when requested multiple times. So we can attribute zero weight here.
• Ethnologue is published by SIL International, a Christian not-for-profit organisation created to facilitate the translation of portions of the Bible into numerous minority languages, we know nothing about the identity of their authors or their peer review process, so at best they represent a minority viewpoint.
• A handful of academic authors who oppose Finno-Ugric/Samoyedic classification have been cited, but again they represent a minority viewpoint at best.

So how do we attribute weight to these minority viewpoints? Clearly, these opposing viewpoints are of insufficient weight or controversy as to prompt ISO to remove the Finno-Ugric/Samoyedic classification from the standard. Since ISO continues to publish them and ISO are considered the mainstream, removal of them constitutes giving minority POV that Finno-Ugric/Samoyedic are obsolete branches of Uralic undue weight. --Nug (talk) 20:59, 24 September 2013 (UTC)

Simple, Nug: if you want to change an existing consensus, then you need to work out a new consensus. You may be correct, in which case you should be successful. But pushing your POV by edit-warring, falsifying references, sneaking it in through maps, and other chicanery is not likely to be successful.

BTW, SIL set up both the ISO codes and Ethnologue, and it's disingenuous to claim they're a RS when they agree with you but a minority view when they disagree with you. But that's the degree of integrity we've come to expect of you. — kwami (talk) 23:25, 24 September 2013 (UTC)

What "existing consensus", you still haven't provided any link. All we have is evidence of you recent bulldozing of your POV here. SIL is the registration authority for ISO 639-3, which deals with individual language codes, they have no authority over ISO 639-5 which deals with language families and groupings. --Nug (talk) 23:42, 24 September 2013 (UTC)

All ISO-5 demonstrates is that the clade is or once was common in the literature. It says nothing about whether it is generally accepted or still current. There are several obsolete or rejected clades that have ISO codes. — kwami (talk) 00:47, 25 September 2013 (UTC)

All obsolete or rejected groupings are actively removed as this change list demonstrates[3]. I don't see FU listed. --Nug (talk) 01:15, 25 September 2013 (UTC)

"Some" doesn't mean "all". auf is defined as a subgroup of awd; either the subgrouping or awd is therefore obsolete. aus, cau, and paa are obsolete, unless they're not supposed to be language families. There's no consensus that ccn is valid. Or hok. Or pqe. Or ssa. Or tut. fox is obsolete. As is kdo. And khi. And pqw. Given that, it's hardly surprising they retain fiu, and that's not evidence that it's still generally accepted. — kwami (talk) 01:31, 25 September 2013 (UTC)

Yes. I find this idea that ISO 639-3 is some kind of an authority on language subclassification absurd. And at any rate, we've established that FU is not generally considered "obsolete", so no such argument could oust it here anyway.

Bear in mind that "everyone accepts X" and "nobody accepts X" are not negations of each other. The negation of the former is merely "some people do not accept X". The existence of any mainstream disagreement on the validity of FU is sufficient to demonstrate a lack of scientific consensus. "Consensus" here means that no serious disagreement exists. It does not mean "the most popular view".

I also find strange the idea that not listing everything in terms of FU/Samoyedic is "pushing a specific POV". If we acknowledge that scientific consensus is not currently available (though I realise reaching editor consensus on this point seems to be a work in progress), we must do ourselves the work of putting together a NPOV. The one we've had until now is stating things in terms of the 9 basic divisions, and then discussing the existing views on subgrouping later. Seems unproblematic to me. Consider Afro-Asiatic languages, where Omotic is explicitly listed as a branch, despite the existence of a reasonably popular (if outdated) view that it is a branch of Cushitic. Consider Indo-European languages, where the "Indo-Hittite" theory has not lead to hiding the main branches under a separate "Narrow IE" term. Consider Turkic languages, where despite the generally accepted primary dichotomy between Oghur and Common Turkic, the infobox still lists subdivisions down to the level of the 4 main groups of CT. --Trɔpʏliʊm • blah 15:06, 25 September 2013 (UTC)

May I ask you to explain what you mean by "mainstream disagreement"? Moreover, the concept "everyone accepts X" is a very unlikely situation that will never happen (e.g., even the Uralic family itself is questioned by some linguists). About ISO: it is of course not an authority on language subclassification, nobody said that. But the fact that FU could become and still is in the classification is indicative of the phenomenon that the term and concept is widely used. Presenting a flat structure for this language family is indeed pushing a particular POV, since it is one of the minority views on classification of Uralic languages, while the majority view very likely still contains the FU/Samoyedic dichotomy. KœrteFa {ταλκ} 15:29, 25 September 2013 (UTC)

Where there is significant disagreement about classification, it's reasonable to be agnostic and not enshrine any of them in the info boxes. That's not a POV claim that intermediate nodes don't exist; indeed, most of our sources have some sort of branching, but they may group Hungarian with Samoyed or Khanty with Finnic. So we list Ugric above Hungarian in the Hungarian article, but don't list it as a primary division here. This is a common approach in many language families: list questionable nodes like Eastern Malayo-Polynesian where they're immediately relevant, but not everywhere. — kwami (talk) 21:35, 25 September 2013 (UTC)

"Where there is significant disagreement", the problem is that this disagreement is not that significant but some people here insist on giving WP:UNDUE weight to it. --Nug (talk) 10:07, 26 September 2013 (UTC)

If some secondary sources pick up on a disagreement, the disagreement is significant. This means that the binary-tree model cannot simply be considered the consensus classification and the other ones ignored. The way the classification is currently handled in the article is the most neutral way I can think of. --JorisvS (talk) 10:14, 26 September 2013 (UTC)

But as it stands the minority flat comb model is given primacy at the expense of the majority binary model. --Nug (talk) 10:40, 26 September 2013 (UTC)

No it is not. The flat "model" isn't a model at all, but an admission that we don't really know the branching structure. We're doing the same thing here. We're not giving primacy to anyone, we're just saying the intermediate nodes are not certain. — kwami (talk) 10:46, 26 September 2013 (UTC)

Sure it is, you even quote Kortmann below where he states that most handbooks don't support the competing "comb-like model". --Nug (talk) 19:42, 26 September 2013 (UTC)

What existing consensus? Where's the link to the discussion? If anything we have Marcantonio who questions the validity of Uralic itself denying FU is a node. That's not consensus. Redei, Decsy and Collinder all accept the validity of FU, and as far as I know Vajda doesn't question it. It is entirely appropriate to address that certain authors question the node. It's entirely improper to delete the term as if it has been declared pseudo-science. This is part of a wider trend of arbitrary changes on linguistic pages following one editors personal surmises. Until we have some strong evidence I think such unilateral edits should be summarily reverted. μηδείς (talk) 23:54, 24 September 2013 (UTC)

The consensus is demonstrated by the several years of stability after the classification was changed in all of our Uralic articles. If you wish to change that, you need a new consensus. — kwami (talk) 00:47, 25 September 2013 (UTC)

You implied a discussion had taken place when you stated ""It was a surprise to me too, but they made a convincing case.""[4], where is the link to this "convincing case" to be found? Also your "consensus is demonstrated by the several years of stability" is just BS in light of this. --Nug (talk) 01:15, 25 September 2013 (UTC)

Don't play stupid. We've been over this. If you don't want to take the trouble to establish consensus, fine. But don't waste our time with your BS. — kwami (talk) 01:33, 25 September 2013 (UTC)

Can we stop arguing over what past editor consensus was and focus on building current consensus? Thanks. --Trɔpʏliʊm • blah 15:06, 25 September 2013 (UTC)

Yep, I agree with that. Even if there was a past consensus, it would be totally irrelevant, since now we obviously do not have a consensus about, for example, omitting mentioning FU. KœrteFa {ταλκ} 15:33, 25 September 2013 (UTC)

We don't omit mentioning it. It's listed in the traditional classification. We even have an article on it. — kwami (talk) 21:15, 25 September 2013 (UTC)

"Traditional" does not mean obsolete. --Nug (talk) 10:07, 26 September 2013 (UTC)

No-one said it does. 10:46, 26 September 2013 (UTC)

A couple remarks on the debate in recent secondary sources:

Introduction: The Finno-Ugric/Uralic Language Family

The Uralic language family, consisting of 20–40 languages spoken in eastern Europe and western Siberia, is traditionally described in terms of a binary family tree, starting from the first binary split of Proto-Uralic into Proto-Finno-Ugric and Proto-Samoyedic. However, this model is not unanimously accepted. In the alternative, more “bush-like” models, Samoyedic is simply one of the three or more main branches, which means that the terms Uralic and Finno-Ugric can be used as synonyms. This practice is adopted also in this chapter.

The Finno-Ugric/Uralic language family, irrespectively of the structure of the postulated family tree, consists of six main branches ...

—The Handbook of Language Contact. Raymond Hickey, 2010

Most handbooks depart from a Stammbaum with binary splits, beginning with the split of Proto-Uralic into Proto-Finno-Ugric and Proto-Samoyed. The competing comb-like model, in which Samoyed is merely one of the main branches, would imply that the terms Finno-Ugric and Uralic could be used as synonyms.

—The Languages and Linguistics of Europe: A Comprehensive Guide. Bernd Kortmann, ‎Johan Van Der Auwera, 2011

— kwami (talk) 22:19, 25 September 2013 (UTC)

Regarding your first source, mainstream viewpoints are never unanimously accepted, there of course exists minority viewpoints. Your second source confirms that most handbooks depart or start with the binary split of Proto-Uralic into Proto-Finno-Ugric and Proto-Samoyed. Here are a sample of sources from 2013 that indicate the level of acceptance of the binary split:

• "Finno-Ugric: Largest branch of the Uralic language family", Routledge Dictionary of Language and Linguistics, Hadumod Bussmann, ‎Kerstin Kazzazi, ‎Gregory Trauth - 2013

• "Of the approximately 20 million speakers of the Uralic languages, virtually all, save for the tiny group of peoples called the Samoyeds, speak one of the Finno- Ugric languages.", The Languages of the World, Kenneth Katzner - 2013

• "when Samoyedic is included with Finno-Ugric, the term Uralic is used for the family", Historical Linguistics: An Introduction, Winfred P. Lehmann - 2013

• "Hungarian belongs to a small subgroup of the Finno-Ugric branch of Uralic" Handbook of Orthography and Literacy, R. Malatesha Joshi, ‎P.G. Aaron - 2013

• "The Finno-Ugric languages and the Samoyed languages constitute the Uralic language family.", Finnish: An Essential Grammar, Fred Karlsson - 2013

• "It is known that the total population speaking Finno-Ugric languages (a branch the Uralic language family) is approximately 25 million", Current Multilingualism: A New Linguistic Dispensation, David Singleton, ‎Joshua A. Fishman, ‎Larissa Aronin - 2013

• "Finnish (Uralic family, Finno-Ugric branch, Finnic sub-branch; Finland)", Semantics: From meaning to text - Volume 2, Igor A. Mel’čuk, ‎David Beck, ‎Alain Polguère - 2013

• "The first division of the parent language, into Proto-Samoyedic and Proto-Finno-Ugric", Compendium of the World's Languages, George L. Campbell, ‎Gareth King - 2013

• "Hungarian (Uralic, Finno-Ugric [Hungary])", page 306, Generative Phonology, Iggy Roca - 2013

--Nug (talk) 10:07, 26 September 2013 (UTC)

Now you're falsifying dates. Lovely. — kwami (talk) 10:56, 26 September 2013 (UTC)

Given that this article would be construed as falling under Eastern European topics, it may well be subject to discretionary sanctions via WP:AE, so I would suggest easing up on these baseless accusations. --Nug (talk) 11:22, 26 September 2013 (UTC)

.... — Lfdder (talk) 11:31, 26 September 2013 (UTC)

lol, I was beginning to think 2013 must've been a wonderful year in linguistics (though I'm inclined to think Nug was just careless and did not intentionally falsify the dates). — Lfdder (talk) 10:59, 26 September 2013 (UTC)

Shrug. So I set the google book search to filter for 2013, what's your point? --Nug (talk) 11:22, 26 September 2013 (UTC)

Most of the books you list above were not actually written or even published in 2013. Katzner (2002, 3rd edition), Karlsson (2008, 2nd edition), Lehmann (1993, 3rd edition), are just a few of the errors in dates that you list above, Nug. Perhaps you don't know how to do a proper search in Google Books? If not, then how can we trust anything else you write here? --Taivo (talk) 15:54, 26 September 2013 (UTC)

Have you adopted Kwamikagami's apparent demagogic arguments that border on personal attacks too? Katzner (2013, revised edition)[5], Karlson (2013)[6], Lehmann (2013, revised edition)[7]. These are revised editions, according to google. --Nug (talk) 19:42, 26 September 2013 (UTC)

Pointing out that you don't know what you're talking about isn't demagoguery, it's relevant to what you have to contribute to this discussion. BTW, the three volumes you just cited were published in 2002 (classification unchanged since 1975), 1999, and 1992, as you can verify by actually looking at them. — kwami (talk) 19:55, 26 September 2013 (UTC)

Taivo claimed these were published in 2002, 2008 and 1993, you claim they were published in 2002, 1999 and 1992, if you can't agree on publication dates then your criticism is a bit rich. These books I cite have different ISBN numbers to the earlier editions you cite, so they definitely are revised editions. Are you sure the book preview you see in Google actually come from the latest edition? --Nug (talk) 20:16, 26 September 2013 (UTC)

The ISBNs are those of the ebooks, released this year. The content is unchanged. Gbooks has scans of the originals. [8] [9] [10] — Lfdder (talk) 20:40, 26 September 2013 (UTC)

There are books from the 19th century reprinted in 2013. That doesn't mean they we're written in the 19th century. — kwami (talk) 20:44, 26 September 2013 (UTC)

Regardless of dating issues, none of the literature here shows anything new I think. Most of it is tertiary or unrelated literature that shows that FU maintains a wide recognition out among linguists who aren't specifically studying Uralic subgrouping. We already agreed on this, I believe. A couple of these contribute the point that doubts at least exist. I believe we agreed on this as well. If this is right, the task is to discuss how to determine what views here are reliable, relevant, and how much weight they deserve.

A phenomenon that was noted in this discussion a couple headlines ago is that new ideas take time to propagate into traditional encyclopedias. Hence for keeping our article up to date, we ought to rely mainly on recent secondary sources. (Pass-by mentions in books on Finnish semantics or whatever are only tangentially relevant at best.) The problem, though, is a lack of any such clear sources! It's been 15 years since Abondolo 1998, and whose treatment of subclassification was cursory anyway (Abondolo doesn't review the general picture, as much as simply references a study from Viitso, that he has distanced himself from later, see Viitso 2000). Slightly later is Salminen 2001, though I imagine some of you will not like the fact that he cites much of his own critique from Salminen 2002. (NB the release order being the inverse of the writing order here.) And that's still a decade old. Several studies on the topic have appeared since then. I still hold that this means our best bet is to digress into the primary sources and present an overview of the current competing ideas.

BTW, note that this is not only the case with FU: many suggestions such as Janhunen's "Finno-Khantic" or Viitso's "Ugric-Permic" are not very widely supported. If you look beyond the much better penetration of the concept "FU" into unrelated studies (which is irrelevant), there is entirely comparable widespread-but-undefended support for e.g. the traditional Ugric/Finno-Permic division. Yet there does not seem to be a problem about e.g. not referencing to Finnish as a "Finno-Permic" language. --Trɔpʏliʊm • blah 23:24, 26 September 2013 (UTC)

Agreed. (Katzner, for example, besides being a popular rather than academic account, retains numerous families that went out in the 1970s.) I'd like to see how 2ary sources treat such ideas, but until then we're stuck with the traditional classification and 1ary sources. I was convinced by the last discussion that the best approach is to be agnostic and not favor any one of them, though I've added a couple immediately superior nodes with question marks (Ugric, Fenno-Saamic) when they are widely retained even in those recent sources. — kwami (talk) 01:44, 27 September 2013 (UTC)

In case there's still a question about my dates for the three books I cited, I got them from the Amazon.com pages where the latest edition of each of the books is for sale. None of them were published in 2013 and all of them are second editions or later so the original text is about 10 years (or more) earlier than that. --Taivo (talk) 03:35, 27 September 2013 (UTC)

This recent book Languages of the World: An Introduction by Asya Pereltsvaig, Cambridge University Press 2012[11] treats Finno-Ugric languages as a distinct language group in chapter 3.1 and in chapter 11.2 discusses its relationship to the Uralic macro family in these terms, on page 209:

"It has been proposed that Finno-Ugric languages constitute a part of a larger family know as Uralic. In addition to the Finno-Ugric branch, this proposed Uralic macro family would also include Samoyedic languages'"

and page 210:

"While Uralic itself has not been conclusively proven to be a valid language family, attempts have been made to relate this putative macro family to other languages or language families"

It seems evident, at least with this handbook, that as of 2012 the binary model is still being discussed. As Kortmann states, most handbooks use that binary model. When writers like Hickey state that Uralic is "traditionally described in terms of a binary family tree", "traditional" in this sense means orthodox or commonly accepted, i.e. mainstream. Therefore undue weight is given in presenting what Kortmann calls the "comb-like model". --Nug (talk) 20:25, 28 September 2013 (UTC)

Nug, none of your sources above are written by specialists specifically about the classification of the Uralic language family, they are all general overviews and not specifically focused on Uralic. The real question is what are the Uralic specialists saying in sources written specifically about Uralic subgrouping? One specialist source is always worth ten general surveys. --Taivo (talk) 20:58, 28 September 2013 (UTC)

Toivo, Asya Pereltsvaig is a linguist from Stanford University, are you saying we can't trust her to understand and summarise the current state of the Uralic language family classification? Her book has an extensive list of references. Surely she would irreparably damage her reputation to get it so wrong as you imply. Do you accept Kortmann or Hickey as "specialists", they both state that the binary model is found in most handbooks and is the traditional (i.e. generally accepted) view, which is reflected by the fact that it has found its way into an international standard. Why don't you put in a change request to the relevant ISO authority and have Finno-Ugric dropped from ISO 639-5 if it is no longer applicable as the mainstream view? --Nug (talk) 21:33, 28 September 2013 (UTC)

Nog, you obviously didn't understand anything that I wrote. I will try again. ALL of the authors you cite above are well-trained linguists (I assume), but that does not make any of them specialists in the Uralic language family. I am a well-trained linguist from the University of Kansas with a PhD in the subject, but I am not a specialist in Uralic. I even speak a Uralic language (Hungarian) along with English, but that still does not make me a specialist in Uralic. Just because someone has a PhD in linguistics (which I assume you do not), doesn't make them a specialist in every language family under the sun. Pereltsvaig's book is not about Uralic. It is about all the language families of the world. Thus, it is no better as a source for Uralic than any of the other non-specialist works you cite above. I am a specialist in Native American languages and Pereltsvaig is no more a specialist on those languages than on any other cited in the work. And Joseph Greenberg and Merritt Ruhlen are also linguists from Stanford University and their work on language classification is universally panned by linguists. Again, you simply don't know how to critically evaluate your sources. And neither Kortmann's nor Hickey's works are specialist works, but works on a broader topic, so we don't know their expertise in Uralic studies. Indeed, here we see exactly that Perelstvaig is not a specialist in Uralic studies, but in Slavic studies. A PhD in linguistics does not make one a specialist in every language family in the world. What do the Uralicists have to say to other Uralicists in specialist publications? --Taivo (talk) 22:35, 28 September 2013 (UTC)

Toivo, you don't seem to understand Wikipedia policy with respect to WP:WEIGHT: "If a viewpoint is in the majority, then it should be easy to substantiate it with reference to commonly accepted reference texts" It is clear that the commonly accepted view is that Finno-ugric is a sub-branch of the Uralic family, as the published handbooks of non-specialist linguists show and found in reference texts like EB which was written by Robert Harms, Professor of Linguistics, University of Texas at Austin. Are you suggesting Professor Harms isn't a specialist in Uralic languages? That some specialist linguists dispute the inclusion of FU within Uralic family only represents a minority POV that has not gained common acceptance. --Nug (talk) 23:53, 28 September 2013 (UTC)

And so we prominently cover Finno-Ugric. However, WP:RS etc. speak of specialists in the field, which is Taivo's point. — kwami (talk) 00:57, 29 September 2013 (UTC)

We are talking about the binary model of the Uralic language group being the mainstream viewpoint, as confirmed by its representation in the reference work EB in an article on Uralic languages authored by a specialist in Uralic languages Professor Robert Harms. --Nug (talk) 01:52, 29 September 2013 (UTC)

No, Nug, we are talking about the weight of contemporary specialist opinion versus non-specialists simply copying what they have read in the past. There was a time when the binary version of Uralic was the commonly accepted viewpoint, but our point is that the contemporary viewpoint of specialists has changed. What are Uralicists saying today? Non-specialists don't count in that discussion. That's the real meaning of WP:WEIGHT. It's not about number of non-specialists repeating old theories, it's about the number of specialists writing about the field in modern scholarship. I ask you again, what are today's Uralicists writing to each other in specialized linguistic publications? EB doesn't count. If Harms has written something recently in a specialist publication, then we can weight his opinion along with other Uralicists writing on the subject in contemporary works. And, please, are you completely incapable of correctly copying a five-letter user name? If you can't handle that simple task... --Taivo (talk) 05:09, 29 September 2013 (UTC)

While I understand Taivo's point of view, I agree with Nug that we should build primarily on handbooks. However, I have some questions, in order to better understand Taivo's approach: (1) what should be the *operational* way to decide who are the *specialists* of Uralic studies? If somebody once published a paper on Uralic studies, does it make him a specialist? Or does he need at least 10 papers on the subject? Or does he need some very recent work? Assuming we have somehow clarified this, then: (2) what should be the *operational* way to decide what is the current *mainstream* view among them on Uralic classification? And how many specialists do we need to decide what the current mainstream view is? Is 5 researchers enough? Or 50? And how many specialists do we need to decide that some concept is *not* mainstream, i.e., to conclude that the FU/Samoyedic dichotomy is not mainstream anymore? Isn't it a danger that we are doing original research on the way? KœrteFa {ταλκ} 14:54, 29 September 2013 (UTC)

I would simply rank anyone releasing original work in peer-reviewed reliable sources as a specialist. No need to complicate this any further, since we are not going to be taking any one specialist's word over the others'. — But, this does not particularly help for solving the general view / specialist view disagreement here. After a couple years of looking, I've had to conclude that nobody has ever made a detailed argument for the unity of Finno-Ugric as a distinct subfamily. It has simply been copied as a starting assumption from researcher to researcher for the last 100 years. The best I have located is papers saying "something like this could be assumed" (Sammallahti 1988 has "a sketch" for some soundlaws defining FU; most of them were recently essentially rendered obsolete by Aikio 2012, though again in the absense of secondary sources confirming this, we don't get to claim more than simple disagreement) or "oh hey but Samoyedic is definitely very distinct" (e.g. everyone making an argument from the lexicon). If we had any actual primary source arguing in favor of FU, there would be no problem here: every possible secondary source would happily cite it, and we'd cite them to establish this as consensus. And of course, this analysis of mine is definitely OR, so we're forced to somehow work around these facts.

"Handbooks" are difficult to take at a fixed value. Many articles in these are indeed solid secondary sources, but some others — particularly the general introductions — can be tertiary encyclopedic overviews, while some other articles will be essentially primary sources that synthesize an overview from disparate previous studies. And since Uralic subgrouping has been in little-debated limbo for ages, the handbook treatments are all either superficial tertiary overviews ("as is traditionally known blah blah"), or primary analyses that all stress being rather tentative in their conclusions. True secondary sources on the topic are vexingly few in number. --Trɔpʏliʊm • blah 20:47, 29 September 2013 (UTC)

Indeed it is vexing. Specialist "Uralicists" must either be incredibly secretive or amazingly insular not to have communicated their supposed consensus view that the binary model of Uralic languages has been depreciated to the wider world. After all, it took three years to promulgate ISO 639-5, with the draft being approved via vote in 2005 and the final standard approved by vote in 2008. Did these people not see this glaring error of including the binary model into the standard, an error that appears so evident to our own in-house Wiki-linguists, or did they just simply forgot to vote? Maybe the cat ate the ballot paper? Why didn't Asya Pereltsvaig's faculty colleagues alert poor Asya of her glaring error in her 2012 handbook, doesn't Cambridge University Press enforce any kind of review process? What can one do in these lamentable circumstances?

No no no, nobody is saying that the traditional binary model being outdated would be the current scientific consensus. Or even a majority view. We're, or I am anyway, saying that there is no current consensus among specialists (perhaps to the point of no majority view existing either, but we don't exactly have a poll available on that). You may be getting this confused with the claim that we had a WP editor consensus on the topic before this discussion. That in turn may have been in place for a couple years, but not quite even since 2008.

(Also yes Uralistics can be a somewhat insular field, I think. That'll happen naturally when one quarter of the literature is in Russian, another quarter is in Finnish, and another quarter is in Hungarian.)--Trɔpʏliʊm • blah 23:05, 29 September 2013 (UTC)

Luckily Wikipedia policy does offer a solution, per WP:WEIGHT: "If a viewpoint is in the majority, then it should be easy to substantiate it with reference to commonly accepted reference texts". Which means that, in essence, Wikipedia articles are intended to be a conservative reflection of generally accepted viewpoints, not a venue for original research where Taivo would have us sift through and critically evaluate sources. It's up to these so called "Uralicists" to promote their favoured model so as to become mainstream, not us. --Nug (talk) 21:18, 29 September 2013 (UTC)

That we'd prefer to have a clear majority view to present is not an excuse to ignore minority viewpoints, or to prefer tertiary or outright non-sources to secondary sources. While these are not common, the articles out there that do review the literature out there (e.g. Salminen 2001 or Janhunen 2009) make clear the point that classification is an open question, and that the traditional family tree cannot be treated as plain fact. --Trɔpʏliʊm • blah 23:05, 29 September 2013 (UTC)

What is the change that you want to make to the article? — Lfdder (talk) 23:08, 29 September 2013 (UTC)

Tropylium has nailed the issue precisely. We need to be seeking what the consensus is among Uralicists. It's not WP:OR to put more weight on the opinion of Uralicists than on generalists. OR would be gathering a list of cognates from the Uralic languages and deciding for ourselves whether the evidence supports a binary or comblike classification scheme. Finding out what Uralicists actually say is not OR; that's a proper application of WP:WEIGHT. WEIGHT doesn't mean you count everyone's vote equally; it means that you weigh the "votes" with one specialist's opinion being far more important than someone writing a tertiary summary for an encyclopedia or handbook of the world's languages. --Taivo (talk) 23:29, 29 September 2013 (UTC)

None of Nug's list above constitutes a Uralic handbook, they are all non-specialist sources and most are tertiary at best. Here's an example of a Uralicist source pulled from my shelf: Angela Marcantonio, 2002, The Uralic Language Family: Facts, Myths and Statistics, Publications of the Philological Society 35 (Blackwell). Page 2 shows the usual binary tree with the caption "The Uralic language family tree as usually reported in textbooks". Then on page 60 it shows "Hãkkinen's 'bush-model' diagram (1983:384)" with a non-binary comblike structure. Then on page 62 it shows "Hajdú's circle diagram (1975a:37; after Sinor (ed.) 1988: XV)" with a series of overlapping circles, but no binary split between them. Then also on page 62 is "Salminen's 'ball-diagram' (1999:20) with a series of circles, but no binary separation between Samoyed and the other eight circles. Then on page 63 is "Kulonen's 'isogloss-model' diagram (1995:50)" with no binary division either. On page 63 is "Viitso's family tree diagram (1997b:223)" which is the first alternative to the traditional grouping that actually considers Finno-Ugric to be a node. Then again on page 65 is "Pusztay's language chain diagram (1997:13)" which divides Uralic into three groups--Balto-Finnic; Permian/Cheremis; and Ugric/Samoyed/Mordvin. These are just the graphic respresentations that she includes, there are more alternative theories, most of which do not feature a binary division of Uralic, in the text. Marcantonio herself doesn't accept the unity of Uralic, but it's not because she doubts the connection of Samoyed, it's because she doubts the connection of Ugric, specifically Hungarian. So even in that regard, she doesn't accept the old binary division of Uralic. So this is the kind of information we need in order to deal with Uralic--the data from specialists in specialist publications, not generalists who simply accept the binary division of Uralic "because it's there". Specialists on the whole have apparently rejected the binary division. That's where the weight is--a comblike family structure, not a binary one. But Tropylium nails the problem on the head--most of the specialists write in Russian, Finnish, or Hungarian, three languages that few English speakers learn. --Taivo (talk) 23:58, 29 September 2013 (UTC)

So let me get this straight, you reject the evaluations found in general handbooks by named non-specialist linguists, and instead reckon an evaluation by anonymous Wikipedians would somehow be more reliable? You say you are a well-trained linguist from the University of Kansas with a PhD in the subject but not a specialist, yet other well-trained non-specialist linguists with PhDs from other American universities, such as our apparently misguided Asya Pereltsvaig with her apparent fundamental error that was not picked up by either her peers or reviewers at Cambridge University Press don't seem to be up to the task. When you claim "That's where the weight is--a comblike family structure", your non-specialist status becomes somewhat apparent though, since it appears that this comblike model is an old model that has since been superseded by later scholarship. How could nine language groups simultaneously evolve from proto-Uralic anyway? I suspect Asya would be rolling her eyes. However you make a good point about what the specialists write in Russian, Finnish, or Hungarian, so it would be interesting to compare the Wikipedia article in the respective languages, on the premise that editors in those wikis would be able to access these specialist sources more easily.

• Finnish: binary FU/S structure given prominence, including a map remarkably similar to the one Kwamigagami rejected as "fraudulent" and you rejected was too complicated. How ironic. Competing models are given lesser prominence with the comblike model dated 1999 along with a later trinary model of 2002 and a newer binary model of 2007.
• Hungarian: binary FU/S structure given prominence, no competing models discussed.
• Russian: binary FU/S structure given prominence (Info box: самодийская ветвь, финно-угорская ветвь), no competing models discussed.

So it seems this article is tilted towards Finnish scholarship. BTW, how did ISO 639-5's binary model get approved in 2008 anyway, given the compelling evidence of more weightier models you say existed? --Nug (talk) 11:15, 30 September 2013 (UTC)

Are you really that incapable of understanding what I wrote, Nug? I was making no evaluation whatsoever. I was quoting a specialist work, Nug, written about Uralic for Uralicists. You have quoted zero Uralic specialists. Do you actually understand what a specialist in Uralic is? A specialist in Uralic is someone who does most of their linguistic research in and about Uralic languages. Your current darling, Asya Pereltsvaig, is not one. She's a specialist in Slavic languages, not Uralic languages. Her book is not about Uralic languages, but about all languages of the world. The author I quoted, Angela Marcantonio, is a Uralicist and her book is entirely devoted to the Uralic languages. See the difference, Nug? Marcantonio quotes multiple other Uralicists and the results of their research into Uralic languages. And, Nug, apparently you wonder how a proto-language can develop into nine subbranches. Ever hear of Indo-European? That is precisely the current view of Indo-Europeanists--indeed, it is the standard view--that Proto-Indo-European broke up into its constituent families without identifiable intermediate steps. And you seem to think that quoting anonymous Wikipedia editors takes precedence over quoting Uralicists. You have a rather tiresome and ill-informed point, Nug, that you keep trying to force home with the weight of "editors at Cambridge Press" and the administrators of ISO 639-5 over the quoted opinions of actual Uralicists. --Taivo (talk) 14:58, 30 September 2013 (UTC)

Nug, you also seem to have a very skewed understanding of what the ISO 639-5 codes actually are. They have only a vague relation to linguistic reality. Their primary function is to assist librarians in managing and maintaining their collections and making them maximally accessible to scholars. They have secondary functions assisting funding agencies in terms of managing their monies and assisting scholars in terms of filing and locating research materials. They are not scientific statements and if you actually read the Wikipedia article on ISO 639-5 you would know that. They include invalid groups which are still useful for library purposes, geographical groupings, and non-genetic linguistic groupings (e.g., Creoles). They are tools in the researcher's and librarian's tool kit, not volumes of canon law. --Taivo (talk) 15:09, 30 September 2013 (UTC)

Well, back from my trip and I see that this discussion hasn't progress much beyond where I left it. As another editor earlier pointed out, Marcantonio questions the validity of Uralic itself and pointed to Edward Vajda's critical review of Marcantonio's book here. You claim "neither Kortmann's nor Hickey's works are specialist works, but works on a broader topic, so we don't know their expertise in Uralic studies", but Kortmann and Hickey are just editors, in fact those two chapters on Uralics were written by Johanna Laakso, a full professor in Finno-Ugric studies at the University of Vienna. Your insistence on entire works devoted to Uralic written by Uralicists is unreasonable, considering that Marcantonio's book was published back in 2002, and there exists more recent handbooks that include chapters on Uralic written by a bona fide Uralicist, such as the chapters by Laakso in Hinckley's 2010 and Kortmann's 2011 handbooks, both which you unreasonably reject. --Nug (talk) 05:59, 6 October 2013 (UTC)

Then you need to learn how to properly cite works, Nug. When you are relying on individual chapters or articles with individual authorship, then you don't cite the entire work, you cite just the chapter you think is relevant with the appropriate author's name. That is standard citation practice in science. I never insisted on the citation of entire books, that is your confusion about what constitutes a scholarly work. I insisted on the citation of relevant works by Uralicists. These general handbooks were not relevant since they are too broad. But if you want to cite individual chapters, each written by a Uralicist, that is appropriate. An individual chapter in a handbook, or a journal article, written by a Uralic specialist, and focused on Uralic classification, is an "entire work". And the review of Marcantonio's work isn't really relevant since my citations from it were not from her conclusions chapter, but from the chapter where she reviews the state of the art in Uralic classification. You'll note that all my citations from her work are her own citations to other classificatory works by Uralicists. Now, if you want to cite those two chapters you mentioned, then please quote the exact relevant comments from those chapters so that we can see precisely what those two specialists say about the subgrouping of Uralic. Just a title doesn't tell us anything. --Taivo (talk) 14:25, 6 October 2013 (UTC)

And, Nug, you apparently didn't even read Vajda's review. Note his conclusion: "Despite my rejection of M's central thesis that Uralic is not a genetic family, I still accept much of the skepticism contained in her ground-breaking study. I also urge both Uralicists and non-Uralicists alike to take this book seriously. M succeeds admirably in shedding doubt on many widespread, yet apparently indefensible assumptions about Uralic languages." In other words, this review was hardly "critical" in the sense you intend. All scholarly reviews are "critical" because they take a careful view of the work under review. Even the most glowing scholarly review of any work is "critical" in that sense. And you will note that at no point does Vajda criticize her "non-use" of Finno-Ugric or her conclusions about the complex nature of Uralic historical studies. Vajda himself states throughout the article that the traditional Indo-European model doesn't work for Uralic. Indeed, Vajda's review has extremely little, if anything, that you can use to justify your position that a binary split of Uralic is the mainstream position among Uralicists. --Taivo (talk) 14:41, 6 October 2013 (UTC)

Actually it was Kwamikagami who originally cited Hickey and Kortmann handbook[12], I'm surprised that Kwamikagami apparently does not know the basics of citing sources or that you did not pick up on his omission, and that I had to point that out to you. Then again this whole issue began with a lie that there existed a previous consensus discussion: "It was a surprise to me too, but they made a convincing case."[13]. Laakso is clearly saying that most handbooks use the traditional binary model but it is not unanimously accepted as there exists a competing brush or comb-like model. Any mainstream view isn't unanimously accepted, there of course exists competing minority views. We don't go giving minority views more or equal weight than the traditional mainstream view. The Finnish article, IMHO, gives better weighting with competing models given lesser prominence. --Nug (talk) 19:54, 6 October 2013 (UTC)

Just saying "Laakso says X" isn't sufficient in this discussion. What is Laakso's exact quote (with page number)? And don't go blaming Kwami for your own mistakes. You're talking to me right now, not Kwami. And what "handbooks" is Laakso referring to? The same generalist, non-specialist handbooks that you tried to claim had any weight in this discussion? And other Wikipedias don't matter here. --Taivo (talk) 20:43, 6 October 2013 (UTC)

Okay, Kwamikagami cites these handbooks, along with the relevant quotes (I already linked the diff in my previous comment), you wrongly assumed it was written by Hickey and Kortmann, so how was it a mistake to point out your error and inform you that Hickey and Kortmann are in fact editors and author of the quote cited by Kwamikagami was in fact Johanna Laakso? As to what "handbooks" Laakso was referring to, obviously the ones relevant to a Uralicist like her, but not necessarily relevant to you. --Nug (talk) 11:23, 7 October 2013 (UTC)

You keep trying to divert this discussion, Nug, into accusations without actually engaging in either providing useful references or quality discussion. You keep trying to blame others. Focus on the issue. If, by "already linked the diff" you mean this, then neither of the quotes there is a ringing endorsement of the binary model. Both quotes talk about the binary approach as "traditional", but then talk more favorably about more recent research leading to a comb model. A "traditional" approach is not exactly a valid linguistic argument. Indeed, in the quote from the Hickey handbook (from Laasko? You're still being slippery about your referencing), the author specifically says that she is not going to use the binary model, but the comb model and use "Finno-Ugric" and "Uralic" (both including Samoyedic) as synonyms. Hardly a ringing endorsement of your POV. Indeed, the syntax of the first sentence of the second quote also indicates that the binary stammbaum is quickly becoming a thing of the past as well. --Taivo (talk) 16:05, 7 October 2013 (UTC)

I'm not "blaming" anyone. Your attempted argumentum ad auctoritatem: "I am a well-trained linguist from the University of Kansas with a PhD in the subject", was somewhat undermined when you erroneously dismissed Kortmann's and Hickey's handbooks as non-specialist works. I would prefer to believe that was a simple schoolboy error but the more cynical observer may think it an attempt to misrepresent these sources to support your POV, given your claimed qualifications. Nor am I being "slippery" about referencing of Hickey, I clearly stated that Laasko authored the chapter when I brought attention to your error. Amazon suddenly doesn't work for you anymore that you cannot verify that Laasko indeed authored the chapter in Hickey? Well of course you can access Hickey otherwise you wouldn't have known that she was not going to use the binary model in her chapter of the Hickey book, so why continue with this silliness? But no, Laasko does not "talk more favorably about more recent research leading to a comb model", nor does she endorse one model over the other. Laasko clearly states which camp the majority of the literature falls and mentions an alternate view. That Laasko then adopts what she calls the alternate "brush-like" model out of convenience "irrespective of the structure of the postulated family tree" doesn't negate her assertion that most handbooks use the "traditional" approach or does it represent an endorsement of one structure over another.

It appears that the latest research confirms the so-called "traditional" view of the binary split of the Uralic family tree into Finno-Ugric and Samoyed clades. In the Journal of Evolutionary Biology, the paper Cultural and climatic changes shape the evolutionary history of the Uralic languages by T Honkola, O Vesakoski, K Korhonen, J Lehtinen, K Syrjänen and N Wahlberg (Volume 26, Issue 6, pages 1244–1253, June 2013). In the results section:

"The Proto-Uralic phase diverged ca. 5300 YBP [HPD (= 95% highest probability density): 7503–3330 YBP] with the separation of the Finno-Ugric and Samoyed clades (Fig. 2). This divergence was followed by the split of Finno-Ugric ca. 3900 YBP (HPD: 5371–2695 YBP) into the Ugric and Finno-Permian clades. The date obtained for the divergence of the Ugric clade into the Hungarian and the Ob-Ugric groups was ca. 3300 YBP (HPD: 4895–1690 YBP)."

In the discussion:

"Our phylogenetic hypothesis follows the traditional view of the Uralic language divergences in that the first split occurs between Samoyed and Finno-Ugric languages and is followed by further divergences of the Finno-Ugric group leading eventually to smaller groups such as Finnic, Saami, Permian and Ob-Ugric (e.g. Korhonen, 1981). Thus, our results do not support the highly polytomous views of the Uralic language tree (Hakkinen, 1983; Salminen, 1999)."

"Various views about the degree of polytomous branching in Uralic languages have been suggested, varying from highly polytomous (Hakkinen, 1983; Salminen, 1999) to strictly binary (Korhonen, 1981) trees, along with different intermediate forms (Kulonen, 2002; Michalove, 2002). This ambiguous nature of the intermediate branchings can also be seen in our results: they show the lowest posterior probability values proposing an uncertain pattern of binary branchings."

In the conclusion:

"Our divergence time estimates are largely comparable with the divergence time estimates suggested by earlier linguistic studies. Some differences, however, do exist: our result suggest more ancient divergence for Ob-Ugric, and more recent divergence for the Finno-Ugric, Finnic and Saami unities. For the shape of the Uralic language phylogeny, our results support the hypothesis of polytomous branching of the Finno-Ugric clade."

So in other words, the most recent research confirms binary branching of the Uralic language tree into Finno-Ugric and Samoyed clades, while the evidence indicates Finno-Ugric clade itself has a more polytomous or comb-like structure. So the POV presented in this article is that of a highly polytomous structure as proposed by Salminen, which is clearly a minority view, and is not supported by the latest research. --Nug (talk) 17:02, 8 October 2013 (UTC)

You clearly have a problem understanding linguistic scholarship and specialization, Nug. I have clearly stated that I am not a Uralicist, so your attack on me for not knowing about these two handbooks (which are irrelevant for my particular linguistic specialization) is unfounded and rather sophomoric. Grow up a bit and accept the fact that you've been loose and slippery in your references to scholarship which you don't seem to have a firm handle on anyway. You're not a linguist, so when a linguist gives you some advice on how to reference scholarship, take it like a man and follow it to avoid further problems with scholars in the field. That said, do you actually understand what you just posted here? Apparently not. This is not an endorsement of the binary model. Didn't you read the statement, "This ambiguous nature of the intermediate branchings can also be seen in our results". In other words, their research does not unambiguously support a binary branching of Uralic. But you also don't understand what they are doing in this paper. They are using computer modeling rather than strict adherence to standard historical linguistic methodology. This is still a controversial methodology within historical linguistics. And it is still just one voice in the discussion, not the terminus ad quem. So there is still a very real split in the contemporary Uralicist community, with most recent voices supporting a comb model of some sort and a few still supporting the binary model. You've simply shown that there is still some debate on the issue. --Taivo (talk) 18:39, 8 October 2013 (UTC)

I found this article and have read it in detail. First, this methodology is not an accepted methodology for determining linguistic subgroupings. It is simply a modern variation of the widely discredited and generally ignored glottochronology of the middle of the last century. Such time estimates based on "cognate lists" are highly suspect among historical linguists. The methodology that is universally accepted among historical linguists is that of "shared innovations". Shared innovations are far more reliable a measure of linguistic relationships than simple word counts and computer models. So this article is simply not a strong argument for following the binary model of Uralic. The fact that it was not published in a linguistic journal is also rather informative. This computer methodology is simply not widely accepted within linguistics. When it is based on shared innovations, it can be an interesting adjunct to traditional historical linguistics, as the work being done in Indo-European. But when it is simply based on lexicon and not on shared innovations, then it is suspect. So a close examination of their methodology shows that the authors use a highly suspect measure of relationship, one that has often been used as a preliminary measure of relationship in the early stages of studying a language family, but one that is always supplanted, and sometimes contradicted, by more complete studies of shared innovations. The article is little more than an updated version of lexicostatistic studies done in the 1950s and 60s and subsequently superseded by more sophisticated studies of shared innovations. In other words, your "most recent" article is nothing more than a rehashing of work done in the middle of the last century. --Taivo (talk) 18:59, 8 October 2013 (UTC)

Your continued persistance in maintaining that I had somehow been "slippery" with referencing Laakso's work in Hickey and Kortmann handbook strikes at the heart of your own credibility. Don't you see that? If you cannot be honest about something that easily verified by examining this talk page history, how can anyone trust you to be honest in your appraisal of the above paper? In any case, your criticism of the paper is in essence WP:OR, until someone can cite a published paper that supports your POV

The chapter PROTO-URALIC — What, where, and when?[14] by Juha Janhunen published in Mémoires de la Société Finno-Ougrienne for The Quasquicentennial of the Finno-Ugrian Society in 2009 directly addresses this debate regarding the structure of the Uralic family tree. I quote the first paragraphs in entirety:

"The traditional picture of Proto-Uralic we have today was created by several generations of comparative linguists, starting with M. A. Castrén (see Janhunen, forthcoming) and ending with the synthetic surveys of Pekka Sammallahti (1988) and Daniel Abondolo (1998). In this framework, Uralic is understood as a regular language family whose members represent the divergent, and basically binary, branches and sub-branches of an originally uniform protolanguage. Proto-Uralic was a fully developed natural language that was chronologically far beyond the glottogonic stage. Its structural properties, as far as they can be reconstructed, may therefore be assumed to have been similar to those attested in modern natural languages. The deepest dividing line within the family is traditionally assumed to exist between Finno-Ugric in the west and Samoyedic in the east. For various reasons, subsequent (Post-Proto-Uralic) diversification has been more profound, or is better preserved, within the Finno-Ugric branch, which is today represented by as many as seven major sub-branches, including (from west to east:) Saamic, Finnic, Mordvinic, Mariic, Permic, Mansic (incl. Hungarian), and Khantic.

In practice, all adherents of the traditional framework have always been conscious of certain problems and limitations that call for minor modifications to the approach. For instance, it is generally acknowledged that the protolanguage was not strictly uniform but dialectally diversified, like any natural language. Also, the branching of the language family need not always have taken place in a binary way, and, in any case, there are isoglosses that cross branch boundaries, including even the boundary between Samoyedic and its immediate western neighbours (Khantic, Mansic, and Permic). Even so, many Uralists agree that the classic family-tree model still remains the best for describing internal family relationships. As a possible modification of the binary family-tree, the fuzzy ‘bush’ model has been proposed by Kaisa Häkkinen (1984), later followed by the linear ‘comb’ or ‘rake’ model of Tapani Salminen (2002). However, even these modified models recognise the validity of Uralic as a language family, as well as the relevance of the comparative method as a diachronic tool.A more relevant question is how much effort should be devoted to arguing against paradigms that are based on an insufficient understanding of the discipline. The situation is analogous to that in the natural sciences, where the theory of evolution is being challenged by religious fundamentalists propagating unscientific ‘alternative’ ‘models’, such as ‘creationism’ and ‘intelligent design’."

He further states:

"One might think that such new points of view have ‘endangered’ the existence of Proto-Uralic as a valid diachronic entity. This is not the case, however, for the principles and methods of comparative linguistics, created during the 19th century, are solid enough to make any major ‘revolution’ in the discipline impossible."

Addressing your pet alternate model, Janhunen states:

"Those who advocate the ‘comb’ or ‘rake’ model would say that Proto-Uralic was more or less immediately divided into the synchronically attested major branches, ranging geographically from Samoyedic in the east to Saamic in the west. The principal problem with this model is that it presupposes an extremely sudden and explosive break-up of Proto-Uralic along a rather narrow east-west trajectory extending from Siberia to the Baltic Sea. This is equal to propagating a very broad homeland for Proto-Uralic, for, technically, the homeland would have comprised the whole area where Proto-Uralic would still have been spoken as a uniform language before the individual branches started differentiating, which would have happened only after the initial explosion. Such a high speed of expansion is, however, unlikely."

And in summing up:

"In any case, an unbiased look at the Uralic comparative corpus would seem to reveal a rather systematically westward-branching family-tree, with the division between Samoyedic and Finno-Ugric lying at the foot. The basic dichotomy of the language family is particularly difficult to refute (cf. also Michalove 2002), a situation that has not been altered by fresh additions to the corpus of Finno-Ugric-Samoyedic lexical comparisons (Aikio 2002, 2006)."

Thus in a nutshell, Juha Janhunen is essentally saying that attempting to refute the solidily established binary split of the Uralic language family into Finno-Ugric and Samoyedic branches with alternate models is akin to religious fundamentalists attempting to refute evolution with alternate theories of "intelligent design". And this religous zeal certainly appears evident here. --Nug (talk) 23:53, 8 October 2013 (UTC)

You don't know what WP:OR is, Nug. Stating to you that historical linguists pretty much turn a jaundiced eye on glottochronology and lexicostatistics as being more important for subgrouping than shared innovations is simply stating exactly what historical linguists do and believe. Just because you know nothing whatsoever about the topic doesn't make my comment WP:OR, it just makes your ignorance of the topic more glaring. "All the basic assumptions of glottochronology have been challenged, and it is rejected by most linguists" (Lyle Campbell & Mauricio J. Mixco, A Glossary of Historical Linguistics, 2007, University of Utah Press, pg. 71). "Shared innovation is the only generally accepted criterion for subgrouping" (ibid., pg. 183). The entire article you cite, therefore, is based on a methodology that is "rejected by most linguists" and fails to use the only methodology that is generally accepted. And apparently your religious zeal in pushing the binary tree is only matched by your inability to understand Janhunen's actual position and how that relates to your opposition to issues related to the comb model. Half of Janhunen's statements above have nothing to do with a binary split and only are directed at establishing Uralic as a viable genetic unit. No one is disagreeing with that. You would help yourself immensely if you actually read the paragraphs you were posting and included only those parts that are relevant to the binary versus comb discussion here. Additionally, you keep finding single sources to prop up your POV. I provided you with references to a half dozen sources for non-binary models. Even Janhunen clearly states: "Also, the branching of the language family need not always have taken place in a binary way, and, in any case, there are isoglosses that cross branch boundaries, including even the boundary between Samoyedic and its immediate western neighbours (Khantic, Mansic, and Permic)". The issue has never been whether or not some linguists still accept the binary model, the issue is that you want to ignore the many modern Uralicists who reject the binary model in favor of a comb model. You have completely failed to show that the binary model is more popular among contemporary Uralicists than the comb model. At best you have shown that the two are equally popular. Finally, when you underline material in a quote that is not underlined in the original without noting that the underlining is your own, then you are being false with the source material. Just another indication that you have a difficult time with actual scholarship. --Taivo (talk) 03:16, 9 October 2013 (UTC)

You are grasping at straws. Given that the sources consistently assert that the binary model has traditionally been held as the dominant view, with Laakso confirming that most handbooks continue to present that view, the onus is on you to show that the comb model has become more popular among contemporary Uralicists. Certainly Janhunen is not one of them. "Intelligent design" also implies a comb-like model where every single species appears synchronically everywhere, whereas evolution is by nature more a diachronic process. So it is not surprising that Janhunen would allude to those as "religious fundamentalists" who refute the traditional model even though "the basic dichotomy of the language family is particularly difficult to refute". --Nug (talk) 03:44, 9 October 2013 (UTC)

Since your memory is only a few minutes long, Nug, here are the references to non-binary models among contemporary Uralicists that I posted above: "Angela Marcantonio, 2002, The Uralic Language Family: Facts, Myths and Statistics, Publications of the Philological Society 35 (Blackwell). Page 2 shows the usual binary tree with the caption "The Uralic language family tree as usually reported in textbooks". Then on page 60 it shows "Hãkkinen's 'bush-model' diagram (1983:384)" with a non-binary comblike structure. Then on page 62 it shows "Hajdú's circle diagram (1975a:37; after Sinor (ed.) 1988: XV)" with a series of overlapping circles, but no binary split between them. Then also on page 62 is "Salminen's 'ball-diagram' (1999:20) with a series of circles, but no binary separation between Samoyed and the other eight circles. Then on page 63 is "Kulonen's 'isogloss-model' diagram (1995:50)" with no binary division either. On page 63 is "Viitso's family tree diagram (1997b:223)" which is the first alternative to the traditional grouping that actually considers Finno-Ugric to be a node. Then again on page 65 is "Pusztay's language chain diagram (1997:13)" which divides Uralic into three groups--Balto-Finnic; Permian/Cheremis; and Ugric/Samoyed/Mordvin." And there is a difference between "traditionally shown as binary" and "contemporary models". I'm really amazed that even for someone who knows nothing about linguistics, you continue to ignore the contemporary scholarship and instead keep pushing the line of "traditionally". "Traditionally" is not "contemporary". Traditional Uralic scholarship was more monolithic. But we have to reflect contemporary Uralic scholarship and that scholarship is divided. And I laugh at your pathetic attempt to compare linguistics with biology and religion. Again, it simply shows that you don't have any real solid understanding of the issues involved. Janhunen's comments about the comb model and your pathetic analogy to "intelligent design" also show that you conveniently ignore the Indo-European family tree which shows the exact comb-like model that you think is improbable for Uralic. Indeed, other language families that show comb-like trees include the Algonquian languages and the Uto-Aztecan languages. --Taivo (talk) 04:16, 9 October 2013 (UTC)

I just realised that Kansas is part of the American Bible belt, you are not a closet creationist are you? That would explain a lot. It was Janhunen who invoked the comparison with religious fundamentalists propagating "intelligent design" in challenge to that traditional 19th Century theory called "Evolution", not me. Janhunen also addresses why the comb-like structure may be more probable for the Indo-European family tree but not necessarily for the Uralic tree:

"The break-up of Proto-Uralic may be compared with that of other ‘old’ language families, notably Indo-European. For Indo-European it is normally assumed that the protolanguage was dissolved by a simultaneous formation of several parallel descendant branches (cf., e.g., Anthony 1995: 557, fig. 1). In this case, the possibility of a non-binary division is supported by the fact that the break-up seems to have taken place in a radial manner, with the different primary branches advancing in different directions from the original core area. The diffusion of Proto-Uralic, by contrast, seems to have taken place in a linear manner, with a gradual and repetitive advance in one direction. The difference may not be so radical, however, for Indo-Europeanists have always looked for signs of a chronological hierarchy between the branches, and several actual or potential groupings have been discovered, including Balto-Slavonic and Italo-Celtic. Even more substantially, there are serious reasons to assume that the division between Hittite and the rest of Indo-European (proper) is more fundamental than any other branching within the family, resembling the division of Uralic into Samoyedic and Finno-Ugric (proper)."

You claim the flat comb model as presented in this article is accepted by most Uralicists, but present a single handbook by Marcantonio (who disputes the very existance of the Uralic language family) which references Hãkkinen's 'bush-model', "Hajdú's circle diagram, Salminen's 'ball-diagram', Kulonen's 'isogloss-model' diagram, Pusztay's language chain diagram. Listing half a dozen competing models does not demonstrate the general acceptance of the comb model. You cannot gloss over the fact that Laakso explicitly states most handbooks support the traditional view, even Marcantonio concedes the general acceptance of the traditional view on page 2 with the diagram caption: "The Uralic language family tree as usually reported in textbooks". "Usually" means "mostly", and therefore undue weight is given to the comb model. --Nug (talk) 06:20, 9 October 2013 (UTC)

Again, Nug, you fail to see the point. "Traditionally" handbooks (although you fail to cite any) did, indeed, list only the binary model. But "traditionally" means "the past". Some Uralicists do, indeed, still use the binary model and have their good reasons for it. I have never said otherwise. But the point here is that among contemporary Uralicists ("contemporary" means "right now"), there are a significant number of scholars who do not accept the binary model and have their good reasons for it. That's what you are blinded to--you think that "traditionally" has undue weight over "contemporary". And, Nug, "textbooks" are not specialist Uralic sources where Uralicists write for other Uralicists. "Textbooks" are not "reliable sources" over and above specialist sources. I have posted references to quite a sufficient number of contemporary Uralicists who use a comb model to show you that they are of at least equal importance to contemporary Uralicists who still use a binary branching model. If I could cite only a single Uralicist who uses the comb model, then you would have a point about weight. But I have not. I have cited at least a half dozen contemporary Uralicists who do reject the traditional binary model. I'm not going to continue to show you the error of your ways, but it's clear you have not established a consensus. --Taivo (talk) 10:41, 9 October 2013 (UTC)

I recommend for Nug to look into the concept of confirmation bias. Finding occasional sources that speak in favor of Finno-Ugric does not establish a consensus, and attempting to claim that they do is cherrypicking your sources.

And even as for what we can gather from Janhunen's article here, note that him commenting on the "difficulty of refuting" the FU/Smy dichotomy can actually be taken in two ways — it can mean "still unrefuted", but it can just as well mean "refutable but it takes some work"! So let's see the works he refers to: Michalove 2002, as already stated in our article as well, actually establishes that Finno-Permic, Hungarian, Ob-Ugric and Samoyedic are according to even just the traditional data all lexically distinct from each other. To quote from the abstract:

The 123 bisyllabic forms listed in Sammallahti (1988) comprise a useful corpus of essentially uncontroversial Uralic forms, whose distribution suggests a division into four primary branches of Uralic: Finno-Permic, Ob-Ugric, Hungarian, and Samoyed.

And the situation in light of newer research into Samoyedic etymology is covered by Aikio 2002:

It can be concluded that the small number of cognate lexemes between Samoyedic and the branches of Finno-Ugric has probably resulted both from research history and from extensive replacement of Uralic vocabulary in Samoyedic. A deeper genetic dichotomy between Samoyedic and Finno-Ugric is a less likely explanation, when taking into account that the morphological and phonological isoglosses supporting such a classification are almost non-existent and that there are also innovations common to Samoyedic and Ugric

Hence the conclusion we apparently need to draw is that Janhunen does consider FU poorly supported, despite its traditional standing? Methodologically, note that it is not the case that anyone actually needs to falsify FU; the burden of proof is on the pro-FU side, and this not having been fulfilled very well ought to suffice for not taking it as a fact. But the unreasonable request that FU must be "proved poorly supported" is regardless actually both doable and done.

Incidentally, let's again make the stances clear: this is not a dichotomic traditional vs. comb model discussion, and viewing it as that may be confusing. We use a largely comb-like organization in the article not to support such a division as being the actual reality, but because it is the maximally agnostic starting position for more detailed discussion. The current best-argued-for (not to be confused with "most widely supported") model is the one Aikio hints at in the previous passage: placing the binary split between Ugric+Samoyedic and Finno-Permic. Namely, this is argued for in detail in Häkkinen 2009 (already cited) — a single article, yes, but this already makes it better sourced than the FU/Smy split, for which I reiterate a detailed argument appears to never have been made. Note also Janhunen's closing notes in his article:

After a couple of decades of intensive, and certainly useful, disputes and revisions of the paradigm, several crucial issues concerning time and place in Uralic studies are now approaching a new level of consensus, especially among younger scholars, as summarised by Jaakko Häkkinen (2009), with whose conclusions the present author largely agrees.

It may be too early to say if his "largely" and "new consensus" actually include the newest tree proposal, and not just simply other facets of PU reconstruction Häkkinen covers, given how his own admittedly tentative tree proposal is something else entirely (than any previous model). But the general gist should be clear: the views on early Uralic history are currently in flux with a new model emerging, and tradition should not be adhered to just for tradition's sake. --Trɔpʏliʊm • blah 15:09, 9 October 2013 (UTC)

Could we add one or two of those trees? The Ugric+Samoyed one would be interesting. (Is that the whole of Ugric?) — kwami (talk) 23:37, 9 October 2013 (UTC)

@Kwamigagami, check out how the Finnish Wikipedia article[15] provides balance: the traditional binary FU/S structure is given prominence, including a map remarkably similar to the one you baselessly rejected as "fraudulent". Competing models are given lesser prominence with the comblike model dated 1999 along with a later trinary model of 2002 and a newer binary model of 2007. I would wager that any Finnish Wikipedian would be better informed of the state of Uralic studies than your average American bible-belt linguist.

@Taivo, You got this need to established consensus back to front. Community wide consensus has been established with regard to what constitutes the majority viewpoint and is documented as the Wikipedia policy WP:WEIGHT. This community consensus overrules any local consensus that Kwamigagami may pretend exists here. In this case, per community consensus, majority viewpoints can be easily discovered by reference to commonly accepted reference texts, for example, such as the definition in Oxford English Dictionary, the description in Encyclopedia Britannica authored by Uralicist Robert Harms, as well as the majority textbooks and handbooks (according to Uralicists) which are used to educate and inform most of today's students and linguists. In fact, such is the preponderance of material that defines the Uralic family in terms of a FU/S binary split they even designated a standard ISO code. The burden of proof is upon you to show otherwise. If only associate-professors from bible-belt universities are sufficiently qualified to interpret obscure papers to determine the majority view, then it fails the "it should be easy to substantiate it with reference to commonly accepted reference texts" test. Sorry, you may not like it, but this is Wikipedia.

@Tropylium, Janhunen appears to be referring to the particular aspect of time and place, not tree structure, in his comment about Häkkinen's work. To be clear of the context of your comments I've quoted the relevant paragraph in full:

"In any case, an unbiased look at the Uralic comparative corpus would seem to reveal a rather systematically westward-branching family-tree, with the division between Samoyedic and Finno-Ugric lying at the foot. The basic dichotomy of the language family is particularly difficult to refute (cf. also Michalove 2002), a situation that has not been altered by fresh additions to the corpus of Finno-Ugric-Samoyedic lexical comparisons (Aikio 2002, 2006). There is also motivation to postulate a succession of several lower-level protolanguages, which maybe termed Finno-Ugric, Finno-Khantic, Finno-Permic, Finno-Volgaic, Finno-Saamic, and Finno-Mordvinic. The entities that were separated from these protolanguages are Mansic (Mansi and Hungarian), Khantic, Permic, Mariic, Saamic, and Mordvinic, respectively (cf. the table below). It goes without saying that there are many details in this system that may require revision. For instance, the status of Khantic vs. Mansic remains controversial, and it is still too early to completely reject the possibility of a common ‘Ugric’ protolanguage for all these entities (cf. Honti 1998). Also, the mutual ordering of the three westernmost branches, Finnic, Saamic, and Mordvinic, is open to alternative interpretations. Even so, the basic structure of the family-tree seems to be solid."

What Janhunen appears to be saying is that while many internal details may require revision and remain controversial, the basic Finno-Ugric/Samoyedic dichotomy remains solid and recent research hasn't altered that. I've never advocated the full traditional tree, just the initial binary split into Samoyedic and Finno-Ugric branches as opposed to the comb-model that is currently presented with undue weight but somewhat disingenuously attributed as "agnostic" and sourced to Salminen[16], who is the proponent of the comb model. Even Salminen acknowledges the wide acceptance given to the binary classification by specialists in Uralic studies while dismissing it merely on the basis of its traditionality (from his 2002 paper Problems in the taxonomy of the Uralic languages in the light of modern comparative studies):

"It is true that binary classification has acquired the status of received wisdom in Uralic studies, which makes many specialists reluctant to criticize, let alone abandon it, but insofar as the sole merit of tradition is its traditionality, there appears little need and no justification to keep to it."

Janhunen, a specialist in Samoyedic studies, makes clear that he is not defending the binary model for the sake of tradition, but on the basis of solid evidence as he states in In Hot Pursuit of Language in Prehistory:

"However, the idea of a non-binary division of Uralic contradicts the lexical evidence, which clearly supports the status of Samoyedic as a separate primary branch (Michalove 2002). This conclusion can be disregarded only by assuming ad hoc that Samoyedic has been lexically exceptionally innovative, as compared with all other branches of Uralic. This, on the other hand, is in contradiction with the fact that Samoyedic, located in the eastern periphery of the Uralic family, is both phonologically and morphologically relatively conservative, just as the Finnic branch (Balto-Finnic) is in the west."

So when you say "and tradition should not be adhered to just for tradition's sake" you are just basically repeating Salminen's minority POV at the expense of the body of mainstream literature that support the binary model. --Nug (talk) 11:34, 10 October 2013 (UTC)

Yes, Janhunen's own view certainly appears to be that the basic Finno-Ugric/Samoyedic dichotomy remains solid. All the primary studies he references still fail to show support for the division. I'm OK with treating his paper as a primary source on subgrouping (this is the current state of our article), but hardly a secondary one. E.g. Michalove's analysis only indicates Samoyedic to be a primary branch: it's everything else grouping together as Finno-Ugric that is not supported.

As for "tradition for tradition's sake", my point (as well as Salminen's, though we've reached it independently) is that there is no separate "mainstream literature" treating this particular question of subgrouping. As we've seen, most of the "body" in its favor consists of assertions in favor of the model in tangentially related works or tertiary literature, without hard evidence. Once these are weeded out, the critical voices end up being not a minority anymore. We can count Janhunen as a defender, but who else? --Trɔpʏliʊm • blah 16:14, 10 October 2013 (UTC)

Jahnunen himself unwittingly (perhaps) shows that the traditional division of Uralic into a binary structure has less to do with actual shared innovations and sound comparative work than with the historical progress of understanding the individual languages of the Uralic group: "...it is obvious that a definitive understanding of proto-Uralic will only be possible after binary comparison has been made between proto-Samoyedic and proto-Finno-Ugric. HOwever, as there still seem to be considerable taxonomic and reconstructional problems to be solved for the eastern branches of Finno-Ugric, a simplified but very useful approximation of proto-Uralic can be obtained in the meantime by comparing proto-Samoyedic with proto-Baltic-Fennic" (461) (Juha Janhunen, "Samoyedic" The Uralic Languages 1998, Routledge, 457-479). In other words, Janhunen's "strong" binary position is the result of comparing Finno-Baltic with Samoyedic and muddling through with the stuff in between. This reflects the history of Finno-Ugric studies which have focused primarily on the Finno-Baltic and Samoyedic endpoints and very little in the middle. Indeed, most of Janhunen's argument focuses on geographic/migration arguments and not on the actual historical methodology of identifying shared innovations.

(Side note here: Samoyedic is actually chronically under-researched, particularly in terms of fieldwork. Janhunen's comparision of Proto-Samoyedic and "Proto-Finno-Permic" (essentially the same as Proto-Finno-Samic which has been reconstructed since the late 1890s I think, and in turn also essentially the same as Janhunen's Proto-Uralic) only works at all due to the handy facts that P-Smy is a fairly old and fairly cleanly reconstructible entity; and that as geographically peripheral, it retains several important archaisms like 2nd syllable vowels. Permic and the three Ugric branches have been subject to plenty of research, but they are also heavily eroded or reshuffled phonetically. It's a wide open question if they have any extra evidence to contribute that would require adjusting the Janhunen-originated standard framework of Proto-Uralic, or if determining tons of conditional soundlaws and inter-family Wanderworts would suffice to get them "in line". --Trɔpʏliʊm • blah 18:30, 11 October 2013 (UTC))

Abondolo's introduction the volume The Uralic Languages (1998, Routledge) is also devoid of a discussion of shared innovations. At no point does either Abondolo or Janhunen write about "these are the shared innovations of Finno-Ugric that set it apart from Samoyedic". Since Nug is not a linguist and is ignorant of theory, I'm not surprised that he has not pointed out any real discussion of subgrouping other than "that's the way we've always done it". But based on reading the history of Uralic studies in various handbooks, I'm not surprised that the history of subgrouping has consisted of "1) Here are the Finno-Baltic languages, 2) Here is Hungarian, 3) Here are the Samoyedic languages, 4) Here are a bunch of things we don't know much about in the middle, 5) Let's group the stuff in the west we understand against the thing in the east that is very different and ignore the stuff in the middle for now." Indeed, Janhunen's reconstruction of 1981 is based solely on 1) Finno-Permian + Samoyed. It ignores "Ugric" entirely. In other words, Janhunen's view of a binary Uralic is based on his reconstruction, which is binary in its selection of languages to compare. It's a circular argument. It is not nearly as systematic as Nug thinks it is. Even Janhunen admits this. Marcantonio has a great quote about this (4): "...the literature frequently fails to differentiate between scientific evidence...and opinions and interpretations....As a result, a reader may be forgiven for becoming confused as to exactly what is opinion and what constitutes scientific evidence in the field."

Kwami, here are summaries of the groupings found in various non-binary approaches:

• Häkkinen (1983): 1) Balto-Finnic; 2) Lapp; 3) Mordvin; 4) Cheremis; 5) Zyrian + Votyak; 6) Hungarian + Vogul + Ostyak; 7) Samoyed
• Hajdú (1975): a mapping of isoglosses. He uses the nodes: 1) Finnish; 2) Lapp; 3) Cheremis; 4) Mordvin; 5) Permian; 6) Samoyed; 7) Ob-Ugric; 8) Hungarian. Samoyed shares all isoglosses with Ob-Ugric, while Hungarian is distantly separated.
• Salminen (1999) (ball diagram): 1) Lapp; 2) Balto-Finnic; 3) Mordvin; 4) Cheremis; 5) Permian; 6) Hungarian; 7) Vogul; 8) Ostyak; 9) Samoyed
• Kulonen (1995): a mapping of isoglosses. She uses the following nodes: 1) Lapp; 2) Balto-Finnic; 3) Mordvin; 4) Cheremis; 5) Zyrian; 6) Votyak; 7) Vogul + Ostyak; 8) Hungarian; 9) Samoyed. One isogloss line separates "Finno-Ugric" from Samoyed, but Samoyed shares an isogloss with 5 and 7
• Pusztay (1997): 1) Western Uralic (Balto-Finnic); 2) Central Uralic (Permian + Cheremis); 3) Eastern Uralic (Ugric + Samoyed + Mordvin)

--Taivo (talk) 18:58, 10 October 2013 (UTC)

…and the recent tree per Häkkinen (2013), which could be described largely trinary: 1a) West Uralic (= Samic + Finnic + Mordvinic), 1b) Mari, 1c) Permic, altogether Finno-Permic; 2a) Hungarian, 2b) Ob-Ugric, 2c) Samoyedic, altogether East Uralic.

He only explicitly argues for the West Uralic and East Uralic nodes though — Finno-Permic and Ob-Ugric are carried over from previous literature, which leaves me iffy on presenting this as a tree. There's even the curious consequence that if FP turns out not to hold up (the evidence thus far is even weaker than for Finno-Ugric), this leaves open the option that it was either or both of Mari or Permic that split off 1st. --Trɔpʏliʊm • blah 18:53, 11 October 2013 (UTC)

I don't know how much credibility can be placed in Marcantonio's comment: "...the literature frequently fails to differentiate between scientific evidence...and opinions and interpretations..." given the criticism by Janne Saarikivi (J. Linguistics 40 (2004)):

"The book is divided into ten chapters, each dealing with various issues relating to the historical study of Uralic. It is written in a thoroughly provocative style and includes various accusations against the ‘traditional views’ of the Uralists, which are said to be based on outdated research and a careless use of the comparative method. Regrettably, however, the book under review is itself plagued by fuzzy and irrelevant arguments, misinterpretations of research material and scholarly history, and obscure use of certain concepts. Self-evidently, this state of affairs seriously weakens the credibility of the author’s radical thesis."

"What is worse, it seems that the author is not aware of the distinction between methodologically incorrect comparisons and the limitations of knowledge that are embedded in the comparative method."

As for Abondolo's book The Uralic Languages, the presence or absence any discussion of shared innovations is irrelevant to the observation:

"Some of the internal divisions of Uralic are not entirely clear, but there is close to universal agreement that the primary chronological break was between Samoyedic on the one hand and Finno-Ugric on the other."

Obviously this "universal agreement" isn't unanimous, several alternative models have been presented by five writers over nearly the previous 40 years. Compared to the 5, how many Uralists are there in total, 15, 50, 500? --Nug (talk) 20:43, 11 October 2013 (UTC)

If you actually knew anything about the field of Uralic, Nug, then you wouldn't even need to ask that question. The fact is that in contemporary (= "now") work on Uralic historical linguistics there are few people supporting the binary model with evidence. It's clear that there has not been a thorough reconstruction of Uralic to date. There is work on Finno-Baltic and work comparing Finno-Baltic to Samoyedic (which has a shallow time-depth and less internal variation compared to the other branches), but there is evidently little work being done to actually bring the other branches of Uralic into any kind of order. Abondolo's comment was written in the early 1990s, before much of the non-binary work that is being cited here. At the time of his writing, there may very well have been great agreement, but that is clearly no longer true because of the number of works since that have cast doubt on the "traditional" (= "the past") binary division. Indeed, some of the binary proposals today put Ugric together with Samoyedic rather than with "Finno-Permic". You still haven't provided us with references of Uralic research to counter the non-binary references that we have presented you with. The one source from someone other than Janhunen uses an invalid methodology, so doesn't count. Do you have any other contemporary references to actual work being done in Uralic historical grouping (not generic "everyone thinks this" comments that are pretty worthless in this discussion)? --Taivo (talk) 22:11, 11 October 2013 (UTC)

"Traditional" does not necessarily mean "the past", it can also mean "orthodox" or "standard". You got a source that explicitly states "there are few people supporting the binary model"? Abondolo explicitly states "there is close to universal agreement" and the book was published in 1998, was it not? That's hardly "early 1990s" and post dates the authors you quote above except for Salminen, and Häkkinen with the 2013 paper cited by Tropylium. That's two. It is a fair question, to get a handle on "few", "many" and "most", so how many Uralicists are there? --Nug (talk) 22:52, 11 October 2013 (UTC)

You clearly don't know anything about the publication process, Nug, just as you don't seem to know anything about historical linguistics in general or Uralic historical linguistics in particular. The Routledge series of which The Uralic Languages is a part has typically taken 5-6 years from the time the papers are written and submitted for publication to the time that the book actually appears, so Abondolo's paper was, indeed, written sometime in the early '90s. This Routledge series has a fairly long turnaround time and always has. They announce the book and it takes about 5-6 years to appear. You just don't seem to understand the situation, Nug. Please cite some contemporary articles in Uralic historical linguistics, besides the Janhunen item you've already cited, that actually discusses the binary method besides making broad comments without any evidence to support it. The most recent statement of the "Finno-Ugric/Samoyedic" binary split was Janhunen's, which was based solely on comparing Finno-Baltic with Samoyedic. Of course, comparing two groups will yield a binary split. So where are your citations, Nug? We've given you citations for recent work that rearranges Uralic without using the "traditional" (= past) binary split. You're not really providing any counter-evidence. You're just reiterating the tired "but that's the way it's always been done". --Taivo (talk) 02:50, 12 October 2013 (UTC)

5-6 years to appear? That's demonstrable BS. Abondolo's book cites 1997 papers by Salminen, Hajdú and Raag. So we have close to universal agreement within the profession at least as of 1997, but you imply a near complete collapse in agreement in 15 years to the point "there are few people supporting the binary model" today, based on what? Exceptional claims require exceptional sources. So where is this secondary source that asserts such a collapse occurred? All you have given us is a handful of papers hypothesising alternative models, without any indication of the level of acceptance for each of these models. You don't even seem to know the total number Uralicists, so how can you say "few people"? --Nug (talk) 03:38, 12 October 2013 (UTC)

Abondolo's "close to universal agreement" is the sum over all Uralists, the vast majority of who have not studied the question of subgrouping in any way. The situation looks very different if you dig up the people who have (to repeat for the Nth time). It's among this group, the actual experts on this matter, where the situation appears to have shifted to the point that there is no current consensus classification. I have no doubt that most researchers who don't need to care about if "Finno-Ugric" is a valid group still happily continue to believe it is, but this is irrelevant for the sake of determining what the scientific consensus on the matter is. --Trɔpʏliʊm • blah 16:18, 12 October 2013 (UTC)

While the vast majority of all Uralists may not have published a paper regarding the question of subgrouping, they nevertheless would have read the papers and be aware of the arguments. That they haven't abandoned the grouping of "Finno-Ugric" means they find the arguments presented thus far unconvincing. --Nug (talk) 18:50, 12 October 2013 (UTC)

• The attack on Nug above as "not a linguist" is shameful, and I won't dignify it with further comment.
• Janhunen's and Vajda's reviews show that what Janhunen calls the "revolutionary" standpoint is unproven and far from consensus, if not unprovable. That there are only some 200 Uralic (FU and Samoyed) cognates, while there are on the order of 1000 Finno-Ugric cognates is an insuperable factual barrier for denying a fundamental split absent other evidence. There is no other evidence. The revolutionaries argue not based on actual evidence of borrowing, but of the imagined possibility of borrowing. The proper course here is simple. Give the traditional classification and it's criticisms, and the alternative theories and their criticisms. Deleting the FU/S node as if it were baseless is a disservice to our readers. If we are here to give a full account, giving all non-crackpot theories and [[WP:ATTRIBUTE}}ing them is the course, not coming down on the side of fashion as if it were physics. μηδείς (talk) 04:57, 12 October 2013 (UTC)

Medeis, the number of identified cognate sets is never a valid argument for subgrouping especially when the "weak" set is based on a still poorly documented group of languages with a very shallow time depth. The "factual barrier" is certainly not insuperable if the Samoyedic languages had been given the linguistic attention that Finno-Baltic has been given over the last two centuries. But that "1000 F-U cognates" is also a spurious number given Janhunen's own comments about his evidence. Janhunen admits that his cognate sets generally exclude "Ugric" from consideration and focus almost entirely on Finno-Permic. So that number ("1000") is questionable as to its relevance to the traditional binary tree for Uralic. Your passionate defense of the traditional model is noted, but misplaced. Indeed, historical linguists place far more emphasis on shared innovations as the sole reliable argument for subgrouping. Cognate counts often serve as signposts during the initial stages of building a tree model of a language family, but always move on to shared innovations as the final indicator. And your assumption that we want to "delete" the traditional classification from the article is false. Indeed, as far as I can tell the article as it stands is acceptable to the majority of participants in this discussion. What is unacceptable is Nug's attempt to remove mention of alternate contemporary proposals or to reduce them to footnotes. --Taivo (talk) 07:54, 12 October 2013 (UTC)

More BS. I've never attempted to "remove mention of alternate contemporary proposals or to reduce them to footnotes". I always argued for a more balanced article where the standard and alternate models are given due weight and opposed the removal of FU as discussed here and here as I have previously pointed out to you. --Nug (talk) 09:51, 12 October 2013 (UTC)

Taivo, you are arguing the facts regarding the cognate sets when what is relevant is presenting the positions of experts in the field, not our view of what is true. The article (and other relevant articles) should include the standard FU/S hypothesis with attribution and the more recent criticism with attribution. We are not here to make our own scholarly arguments, we are here to describe the scholarly arguments, plural. μηδείς (talk) 17:02, 12 October 2013 (UTC)

@Μηδείσ: FWIW your "1000 FU cognates" and "200 Uralic" are not commeasurable numbers (where are you getting the former from anyway?). Even if we want to pretend that the Samoyedic data is on level with the rest of the family, "about 200 inherited items in Samoyedic" is the number of the reliable etymologies that remains after spurious and messy comparisions have been weeded out. Sammallahti 1988, which is thus far recognized as making good headway in connecting Permic and Ugric to Janhunen's reconstruction of PU using only the best-behaving data available, lists no more than 83 lexemes found in all three of Smy+Ug+FP, and 267 lexemes found in Ug+FP but not Smy. I don't know about you but 350 reliable Finno-Ugric etymologies does not sound like a major difference anymore to me, especially when you consider there are 20-30 languages that count as Finno-Ugric, versus 6-10 Samoyedic.

If you want the "raw" counts from the literature (yes I'm veering OR), the UEW includes about 500 entries with alleged Samoyedic reflexes. This is low, but by no means aberrantly low: compare with 700 entries for Hungarian, of which some 150 are Ugric only; or 650 entries for Mari, of which at least 200 are Finno-Permic only.

More importantly, if you think the "revolutionary" argument is somehow based on assuming borrowings, you have read neither our article nor the preceding discussion. The Finno-Ugric group is doubted because on the level of verifiable isoglosses, the evidence amounts to largely speculative sound changes and 1-2 derivational innovations. The East Uralic group is proposed because it's supported by half a dozen fairly clear sound changes (some lexical isoglosses as well, though I think they are almost entirely not published yet).

Anyway I think all of us in this discussion agree that due weight for each theory is needed, we just disagree how much weight is due. Preferring one particular subgrouping for stuff like infoboxes or the basic list of Uralic languages is what I gather you're after, but this seems like too much to me: these should be at a level of "just the facts ma'am" which amounts to the 9 branches that are 100% universally accepted. This should not be taken as an endorsement for a flat comb-model family tree, as I believe I've shown above with precedents from other language family articles (Afrasian, IE, Turkic)?

Our subgrouping section, then, does start by presenting the traditional model first and then discussing alternatives, exactly like Nug's example of the Finnish Wikipedia "doing it right". Does someone find this to be insufficient presentation?

@Nug: Equally well, I could claim that some ill-defined "other Uralists" not having criticized the new proposals either means that they find the arguments presented thus far convincing. Absense of discussion is a very difficult point to build on. I'd argue that my interpretation is somewhat more likely than yours, since the also classificatory "Uralic isn't a valid family" proposals from the last 20 years have faced strong criticism. But silence is impossible to cite so this is pointless anyway for the sake of building the article. We still need explicit citations. --Trɔpʏliʊm • blah 19:22, 12 October 2013 (UTC)

It's bad enough you can't spell my name correctly, Tropylium, but why are you arguing the facts? This is not a journal or colloquium. All that matters is our reporting what the scholarly arguments are, not what we as OR specialists think about the evidence. μηδείς (talk) 19:32, 12 October 2013 (UTC)

Medeis, we are not putting OR in the article. But what has happened here is that Nug presented quotes along the lines of "everyone believes this" and expected us to virtually remove all mention of other points of view. We have presented multiple sources to show that the "everyone believes this" comments, even though they were from otherwise reputable scholars, are simply false. When there is a contradiction between otherwise reliable sources, then the Wikipedia editors who have an advanced knowledge of the field must step in and discuss the relative weight and the relative validity of the arguments. While the majority of the time Wikipedia can be the encyclopedia that anyone can edit, there are times when the scholars in a field must make more subtle decisions and arguments about which sources are more accurate in their statements than others and how to shape the article so that readers are not being led down a path that may not be correct. --Taivo (talk) 19:44, 12 October 2013 (UTC)

@Taivo, why do you persist in this lie that I "expected us to virtually remove all mention of other points of view", when in fact the issue has been the removal of all mention of FU as I have pointed out to you just five minutes ago? Are you really that clueless that you don't understand this kind of dishonesty undermines your credibility in the eyes of those observing this talk page? The problem in your "Wikipedia editors who have an advanced knowledge of the field must step in" approach is that you are unwittingly imposing your own POV. You made clear you favour shared innovations over lexicostatistics citing Lyle Campbell, and thus back Salminen's hypothesis while rejecting others. However scholars like Georgiy Starostin are critical of aspects of Lyle Campbell's reasoning and support lexicostatistics as a valid methodology[17]. It is not for us to back who we think is the winner, but merely to give an account of the viewpoints with due weight.

@Trɔpʏliʊm, let's face it, even Häkkinen, Salminen, Kulonen and Pusztay are unconvinced by the rival models presented by each other, otherwise why would they present models of their own, while in the meantime we have Honkola, Vesakoski, Korhonen, Lehtinen, Syrjänen and Wahlberg in their recent 2013 paper and Janhunen all affirming the initial FS/S binary split. So that's 4 against and 7 supporting the initial FU/S split, and 1 supporting and 11 against the models of Häkkinen, Salminen, Kulonen and Pusztay. But it is the acceptance of professional Uralicists that determines whether a hypothesis becomes the mainstream view. As of 1997 sources tell us almost universal agreement existed, and as Janhunen tells us, "the principles and methods of comparative linguistics, created during the 19th century, are solid enough to make any major ‘revolution’ in the discipline impossible". --Nug (talk) 19:59, 12 October 2013 (UTC)

Again, Nug, your lack of linguistic training shows itself. Starostin's methodology is not widely accepted by historical linguists and Campbell's is. Indeed, Campbell's definition is just that--a definition of what the vast majority of historical linguists do. I could post dozens of other citations about the matter if you actually persist. Shared innovations are the primary and definitive measure of subgrouping, not lexicostatistics. Starostin's view is actually WP:FRINGE when you get right down to it. And you will find no evidence here that the linguists involved in this discussion want to remove any mention of the "traditional" binary tree. Zero. It is a part of the history of the family. But even your "universal agreement" source is simply wrong about the facts. Models that questioned the traditional approach have regularly appeared through the history of the subgrouping of the family. And you simply don't understand Janhunen's quote. --Taivo (talk) 23:00, 12 October 2013 (UTC)

I'm debating the facts here in reply to other people who attempt to do the same. I dunno what this unsourced claim of "1000 Finno-Ugric cognates and no other evidence" was aiming for anyway.

As for the Honkola & co. paper, it's from a team made of three biologists, one anthropologist, two general linguists, and zero Uralists. At best you can count this for two expert opinions, and that's quite stretching it (on a quick check, the two linguists involved actually seem to be grad students with no other publications under their belt). Also, isn't the disagreement between Häkkinen/Salminen/Kulonen/etc. perfect evidence that current consensus doesn't exist, as me and others have been arguing for all along?

I don't even know what to say if you are seriously proposing that we could "cite" Janhunen or anyone for a claim that a particular result is too solid to possibly falsify. --Trɔpʏliʊm • blah 02:46, 13 October 2013 (UTC)

Yes, that Honkola et al. paper really isn't good linguistics at all. It's just glottochronology with fancier graphics. --Taivo (talk) 04:20, 13 October 2013 (UTC)

The question if Finno-ugric should be concidered a separete branch has been dealt also before in the User talk:Ante Aikio#'Finno-Ugric' language family is 'controversial'? in 2008. In that discussion page entry Ante Aikio, who is a professional uralic linguist and currently professor of Saami languages in University of Oulu, Finland (but unfortunately no longer edits wikipedia) gave his view on the topic ("The quote you cited is indeed quite accurate", that is: "The term Finno-Ugric is somewhat controversial today") and gave also two references ([18], [19]) to articles by Tapani Salminen, who also is a professional uralic linguist spesialised in samoyed languages and currently works in the University of Helsinki. After 2008 there may of course be also new research, so what is said in these artricles is not of course by no means the final world. For example in 2012 finnish uralic linguist Jaakko Häkkinen (here User:Jaakko Häkkinen, the last edit from him here seems to be from 26.12.2012) has argued here (page 94) that "the Samoyed branch no longer seems to have split first and directly from Proto-Uralic, but instead split from an East Uralic dialect together with the Ugric branches." This of course is only one view, but it has to be noted what Juha Janhunen in an article cited also here in the above discussion says here (p. 75): "At the same time, the author wishes to stress that many of the interpretations presented above are no longer as controversial or ‘counter-revolutionary’ as they would have been only a few years ago. After a couple of decades of intensive, and certainly useful, disputes and revisions of the paradigm, several crucial issues concerning time and place in Uralic studies are now approaching a new level of consensus, especially among younger scholars, as summarised by Jaakko Häkkinen (2009), with whose conclusions the present author largely agrees." And further, the paper by Häkkinen (2009) is here, but unfortunately only in Finnish. But in any case, it just the same paper, where Häkkinen presented the conclusion that I cited above. And of which (as a summary on current research) one of the leading expert in samoyed studies, Juha Janhunen says that Häkkinens paper summarises "a new level of consensus" with whose conclusions he "largely agrees". And in the article Proto-Uralic homeland hypotheses here written by Häkkinen he says the same giving this reference (unfortunately again only in Finnish). - Here he of course summarises his own work, but as we can see from the statement of Juha Janhunen, one of the leading experts in the world on these subjects, not even Häkkinen is just anybody, but someone wityh whose general summary of the development of uralistics Janhunen "largely agrees". And so there at least can not be said to be a consensus that finno-ugric is a separate branch, but on the contary the consensus seems to be at least heading towards the opposte, even though the discussions are still going on and research is being carried out, on how the family tree of the uralic languages finally should look like.

But finally also a disclaimer: I myself am lay person, and not a linguist, but as a native speaker of finnish I am interested of uralic linguistics and try to follow the latest developments. But in my interpretations I may be also wrong. - But at least, when Jaakko Häkkinen still used to edit the finnish language wikipdia, he usually was pleased with my edits on the topics, when I asked him. But now, when he (just like Ante Aikio) no longer edits here, we will have to cope by ourselves and cannot have anybody to check our edits. (In finnish languge wikipedia we still do have also two uralic linguists as freuquent users, who have not given up editing wikipedia, so there may be some hope, but I doubt, if they will have any opportunity to take part in these discussions.). --Urjanhai (talk) 16:57, 15 October 2013 (UTC)

Thank you, Urjanhai, for those very helpful comments about some of the current work in Finnish. --Taivo (talk) 18:45, 15 October 2013 (UTC)

Thanks for your input Urjanhai. Too bad Aikio and Häkkinen don't edit here any more, but I guess they came to the view that "Wikipedia editor" is something they wouldn't want on their curriculum vitae, so they understandably left the project. When Juha Janhunen says that Häkkinen's paper summarises "a new level of consensus" with whose conclusions he "largely agrees", what does Jaakko Häkkinen's 2009 paper actually discuss? The concluding section titled "Uusi kuva hahmottuu" (A new image takes shape) gives us a clue:

"In this review I have argued, on a scientific basis, that the proto-Uralic language area seems to have been established between the Volga River and the Ural Mountains, around the bend of the Kama river basin. Proto-Uralic seems to have split around the beginning of the Northern Bronze Age approximately around 2000 BC. Exactly the same conclusion has recently been reached from different starting points by Petri Kallio (2006), a result which this study confirms."

Häkkinen doesn't appear to be discussing language classification at all, and the "a new level of consensus" with whose conclusions Juha Janhunen "largely agrees" is about time and place. --Nug (talk) 09:33, 16 October 2013 (UTC)

One of Häkkinen's main arguments for relocating PU is based on classification: if Samoyedic is placed in an East Uralic group instead, its eastern location can no more be a strong argument for locating Proto-Uralic east of the Urals. (I need to check if he actually says as much, but the converse also applies: if the other arguments point to a more western origin, then it's unlikely that Ugric and Samoyedic would have dispersed to Siberia independently.) His treatise of the arguments is brief, though; the full treatise of the current evidence is actually to be found in his Master's thesis. I expect further studies on the East Uralic concept to come forth over the next couple of years.

Also FWIW lest anyone is lamenting the absense of editors who have first-hand understanding of the debate: although not a professional, I have some expertise in Uralic linguistics, and e.g. keep in regular contact with both Aikio and Häkkinen. I gather they've decreased their editing not for PR reasons, but mainly due to disillusionment with certain Wikipedia policies - the valuing of neutrality over truth would be the big one, when non-facts such as the existence of "Finno-Ugric" are required to be propped up as long as there are citations claiming as much. :) --Trɔpʏliʊm • blah 15:33, 16 October 2013 (UTC)

Indeed, Tropylium. Wikipedia makes it very hard for scholars to participate here because it so highly values contributions from anyone with a computer and a quote. --Taivo (talk) 16:38, 16 October 2013 (UTC)

@Tropylium, it is probably just as well that Aikio and Häkkinen no longer contribute here, as it could be construed as a conflict of interest in that they may be seen to be promoting their hypotheses at the expense of other experts, like Janhunen, who don't edit here. But then I suppose we have Aikio and Häkkinen's disciples to contend with. I wasn't thinking PR reasons but more one of prestige and reputation. Unlike Encyclopedia Britannica where being invited as a named contributor to write about some topic, like Sir John Lyons on the topic oflinguistics[20], is seen as recognition of one's knowledge, Wikipedia entails having to wrestle in the mud with the rising tide of the unwashed masses. Hardly an edifying spectacle, and for what? All educational institutions, except for maybe some primary schools, treats Wikipedia with caution because it is so relentlessly edited by everyone. All your hard won edits will be eventually eroded and washed away like a sand castle. It's a bit like Sisyphus eternally attempting to roll a rock to the top of the hill only for it to roll back down. I guess taking ownership of articles is one way, but that is expressly forbidden by policy WP:OWN. --Nug (talk) 21:58, 18 October 2013 (UTC)

Strange population table

Relative numbers of speakers of Uralic languages Hungarian   56% Finnish   20% Estonian   4.2% Erzya   2.8% Moksha   2.5% Mari   2% Udmurt   1.9% Komi   1.6% Other   8.9%

There is a very strange population table in the text (section Uralic_languages#Classification). It claims that 8.9% of all speakers of Uralic languages - i.e., a couple million people at least - speak "other" languages than the 8 largest ones listed in the table. But all these "other" languages - Veps, Saami, Khanty and Mansi, Samoyedic - are spoken by comparatively tiny groups. They hardly would have 200,000 speakers together, which is a far cry from the implied 2 million! Any source for the table? -- Vmenkov (talk) 17:27, 1 October 2013 (UTC)

Rechecking, the table was added in a single edit with no sources, based on the previous pie chart that didn't show any exact numbers and only distinguished 5 entries: Hungarian/Finnish/Estonian/"Mordvin"/others. I say we scrap this and make a new one from consistent data (Etnologue maybe?) --Trɔpʏliʊm • blah 18:57, 11 October 2013 (UTC)

I see; so the original "other" sector in the pie chart included all the Permian languages and Mari, which would make the 8.9% number fairly reasonable. Anyway, I hope you or somebody else with time on his hands will redo the chart! -- Vmenkov (talk) 16:03, 12 October 2013 (UTC)

The display color

The display color for Uralic lime green is very bright. I suggest using a darker shade of green. Speling12345 (talk) 5:18, 19 December 2013 (UTC)