Talk:Uralo-Siberian languages

Uralo-Siberian and Nostratic
This article is blatantly POV and completely misleading; some serious editing would be needed. At present, it is not stated anywhere that Uralo-Siberian is a fringe theory that is supported only by a couple of scholars in the world - even Indo-Uralic and Nostratic, both highly controversial, have more proponents. --AAikio 04:00, 20 August 2006 (UTC) -- Hmm. You're way too literal and suspiciously hostile towards progressive theories. You don't seem to be fully aware that Uralo-Siberian is merely a subset of Nostratic. So to say that Nostratic is more supported than Uralo-Siberian is a little nonsensical. This shows me that you haven't read what you purport to understand. A simple flip-through in one of Allan R. Bomhard's books on Nostratic will show you exactly where he stands on the Uralo-Siberian connection. Hard to miss if you've paid attention. So please. More reading, less hostility. --Glengordon01 07:20, 21 August 2006 (UTC) -- Perhaps I should clarify that in articles such as these, we need to keep focus on what the theory claims, not merely the title of the theory. For example, "Nostratic" and its potential subsets of it have been named a variety of things throughout time and depending on the scope of the researcher. So this is why I say that you are too literal. You take the name "Uralo-Siberian" and seem to easily forget the theory behind that name. --Glengordon01 07:27, 21 August 2006 (UTC) -- The Uralo-Siberian theory claims that there is a specific genetic relationship between Uralic, Yukaghir, Chukotko-Kamchatkan and Eskimo-Aleut. Many variants of the Nostratic theory do not include all these language families, so it is not true that Uralo-Siberian is a subset of Nostratic. And the fact remains that this is article is POV and gives the uninformed reader a distorted picture, unlike for example the article Nostratic languages. This is not a matter of hostility but objectivity. --AAikio 07:56, 21 August 2006 (UTC) -- Oh dear, but Allan R. Bomhard has written a book you never have read called *NOSTRATIC* and the Indo-European Hypothesis (1996) where he even charts out an easy-to-understand tree specifically connecting Uralic, Yukaghir, Chukotko-Kamchatkan and Eskimo-Aleut together. I think you can track it down in Braille format :P

Since Bomhard's work exists to your contradiction, regardless of whether you personally deem him worthy or not, we see that there does exist at least one important variant of the Nostratic theory that considers "Uralo-Siberian" (however we might call it) subset. Trying to hide facts that you find distasteful is a little POV.

PS: Are you really terribly current on Nostratic or related issues? Honestly? Just so that we're on the same page: Vladislav Illich-Svitych died a full forty years ago, even years before the late Janis Joplin performed at Woodstock. In fact, his death was a decade prior to my arrival on the scene as a wee spermatozoid, before which I was possibly an edible mushroom, bottle of alcohol, or even a hot dog. Yes indeed, before Elvis Presley choked on some codeine pills. Before New Wave music. Before "Karma Chameleon"! Sweet feather of almighty Maat, even before Star Trek: TNG! Quite frankly, you may be too far behind to catch up now, hehe :) --Glengordon01 08:18, 22 August 2006 (UTC) -- Just to repeat my point: nothing that you wrote above contradicts my main point that "Many variants of the Nostratic theory do not include all these language families". The rest of your message is too bizarre to comment on. I might participate in this discussion again if there are edits to the article. --AAikio 01:11, 25 August 2006 (UTC) -- Some say bizarre, some say light-hearted. Some are happy, some are cranky. Anyways, until you can specify WHAT variants you're speaking of, your statement is without substance. Variants that you might find from 1840 don't really count nowdays. --Glengordon01 20:23, 2 September 2006 (UTC) --

Update needed
Currently the article focuses almost entirely on Fortescue. Could someone knowledgeable (AAikio?) read Seefloth's more recent work (see external links) and modify the article accordingly?

BTW, it does not make sense to say that Nostratic does not include Uralo-Siberian. Sure, some published versions do not include all four U-S families, but that's because the authors had not investigated those families. AFAIK there is no case where any Nostraticist has said "no, my colleague is wrong, I present evidence that Eskimo-Aleut (or whatever) does not belong to Nostratic". (Such a claim has only been made for Afro-Asiatic... and soon evolved into the claim that Nostratic and Afro-Asiatic are sister-groups, which actually means it has gone full circle.)

It also doesn't make sense to require that U-S must be better supported than Nostratic. Have a look at Indo-European -- everyone agrees on which families are included and which excluded (ignoring here the purely nomenclatural issue on whether to include Anatolian), but there is very little agreement on the internal stucture of IE: Balto-Slavic is universally accepted, Indo-Iranian nearly so, but that, stunningly, is all, after well over a century of serious research. Nostratic is (currently) similar.

David Marjanović 22:33, 15 April 2007 (UTC)


 * Fair point. Sergei Starostin's Borean tree has all four U-S families in Nostratic, but also puts Eskimo-Aleut, ChukKam & Yukaghir in with Nivkh in a Paleosiberian node that's slightly closer to Altaic than to Uralic. I think Fortescue himself used to hold to a Ural-Altaic grouping exclusive of the rest, so it's a brave linguist who would say that the branching order is a done deal. But notice that Fortescue is also pretty adamant that Nivkh isn't U-S, or Nostratic either; instead, he thinks it might be Mosan (or with a Mosan substratum). In other words, there's a large overlap of membership between Fortescue's U-S and Nostratic, but the groupings as such are mutually exclusive alternatives for any version of Nostratic that includes Nivkh.


 * It may be because I'm dog tired (or returning to a linguistics article after things like Germanic mysticism), but I just realised that what I wrote above doesn't make a lot of sense. I think what I meant to say was, there's a large overlap between Paleosiberian and U-S, but you can't include both in the same version of Nostratic if Fortescue's right about Nivkh. Or something. Hell, just concentrate on my second paragraph, ok? Gnostrat 05:06, 16 April 2007 (UTC)


 * Anyway (unless I'm missing your point) I don't think the article is claiming that U-S has to be better supported than Nostratic. At least, when I added the Relationships section last year I was simply summarising Fortescue's own view of the matter: that U-S may just be proveable but Nostratic is too remote. That's mainly because he thinks that substrata and superstrata and borrowings back and forth will, given enough time, accumulate to the point where they overwrite the original inherited elements so that genetic linkages can't be recovered. There's no reason why an opposing view shouldn't also be put to balance things out. Gnostrat 00:15, 16 April 2007 (UTC)


 * So Fortescue says Nivkh is not Nostratic? That is worth mentioning in the Nostratic article.
 * But… Mosan? Mosan? That is… an impressive idea. Can you tell me more about this? David Marjanović 12:40, 28 May 2007 (UTC)


 * Well, this one has a bit of history behind it. At the 1988 Ann Arbor symposium on Language and Prehistory, Sergei Nikolaev and Oleg Mudrak presented a paper entitled 'Gilyak and Chukchi-Kamchatkan as Almosan-Keresiouan languages: Lexical Evidence'. 169 etymologies, of varying plausibility, although they note only one regular sound correspondence (in 6 etymologies). This has to be seen in the context of the broader hypothesis (advanced by Nikolaev and, I understand, supported by Shevoroshkin) that all these families are part of 'Sino-Caucasian'. Robert Austerlitz also reported that he had found resemblances between Nivkh and Nootka, but didn't pursue it because of the geographical implausibility.


 * Fortescue thinks the typological features which tie Chukchi-Kamchatkan to Mosan (stem reduplication, incorporation, prefixing and other stuff) and separate it from the rest of U-S are less fundamental and less archaic than the U-S affinities, but Nivkh is something else with its Mosan-like inclusive/exclusive 1st plural, numeral classifiers, "unusual" vowel harmony, lack of accusative and genitive markers, 1st sing. n-... He doesn't say it's a genetic relationship but it looks closer to it than in the case of Chukchi-Kamchatkan. And he points out the meagre evidence for the Nostratic affiliation of Nivkh. I quote (p. 59):
 * "As regards the Nivkh 1p inclusive pronoun mir/mer, according to Panfilov (1962...) this (like 1st dual megi/meŋ) is based on mi/me 'two' and thus can have nothing to do with 'Nostratic' m-initial 1st person forms (including CK muri 'we'). Attempts to link Nivkh to 'Nostratic' (and hence also US) languages, such as in Bomhard & Kerns (1994), strike me as particularly weak."


 * As to how Mosan 'got into' Siberia, Fortescue sees Mosan traits underlying coastal Na-Dene and thinks the stratum may have followed the coast northward all the way round. Think of an ice-age Beringian 'continent' big enough to hold Na-Dene speakers (possibly 'Aurignacoid' mammoth-hunters from central Asia) and Mosan (or Almosan-Keresiouan?) speakers, temporarily shut out of America proper by the closing ice barrier but retaining links down the Okhotsk coast. The Mosan-Nivkhic substratum in northeast Asia was widespread enough to have left traces in Korean, apparently.


 * For what it's worth, I think it's more likely that Nivkh represents an eastward migration from north-central Asia which absorbed Mosan-like features on the coast. Has anyone noticed that Nivkh has apparently borrowed the IE word for 'wheel' (kwekwlos, cf. Nivkh kulkuls 'wheel', kulkul- 'go round')? Unless IE speakers got as far east as the Amur, we would probably have to assume that proto-Nivkh went east from somewhere adjacent to the Kazakh steppe, or Xinjiang perhaps. Of course, that still doesn't make it Nostratic necessarily. Gnostrat 02:31, 4 June 2007 (UTC)


 * Interesting (just in case I don't repeat myself often enough :o) ).
 * Could kulkul- be inherited instead of borrowed? (This is testable in principle: PIE */k/ is supposed to come from Proto-Nostratic *. If Nivkh k does not correspond to that, we have a borrowing… though it still might be a Wanderwort, and Nivkh has consonant gradation or something.)
 * Shevoroshkin thinks that Almosan, if not "Almosan-Keresiouan", is Dené-Caucasian – to be precise, he thinks it's a branch of Avar-Andi-Tsez within East Caucasian. Boggles the mind, but a Dené-Caucasian affiliation is plausible to me in the light of the cognate sets that Shevoroshkin proposes. This would, incidentally, mean that the first Almosan speakers arrived in North America at a very late date. But then Basque is new to the Iberian peninsula, too – it shares agricultural vocabulary with Basque and Burushaski!
 * The n- 1st-person pronoun fits Nostratic (*/m/-) about as well as Dené-Caucasian (*/ŋ/-), though you're right that Proto-Salishan had */n/- and Proto-Algic.
 * Why should a personal pronoun in a language that (Nostratic-style) distinguishes dual and plural come from "2"? Plus, the -r in the plural and the velar in the dual look Nostratic again.
 * Similarities to Korean are not unexpected based on geographic proximity and shared putative Nostratic membership.
 * Separate inclusive and exclusive "we" are reconstructed for Proto-Nostratic and have survived in several extant languages. The numeral-classifier fusions are weird, though. Accusative and genitive markers can disappear… Clearly, more research is needed. Preferably before Nivkh and the Mosan languages die out. :-|
 * Oh, and… the Aurignacien was a culture, not (necessarily) a genotype. David Marjanović 19:20, 9 June 2007 (UTC)

Morphology again
AAikio deleted the sentence "All of these are considered evidence for Nostratic by others.", claiming it was "too vague" and didn't mention who the "others" are. That's why the link is in there: read the Nostratic languages article, and you'll see. So I put the sentence back in, with somewhat clearer wording.

Now, does anyone have access to Seefloth's work? The Linguistlist posts on it are quite impressive, but too confusing for me to make a Wikipedia article out of them. David Marjanović 12:37, 28 May 2007 (UTC)

Support by genetics
The Uralo-Siberian hypothesis is fairly probable from the view of genetics. We know that a part of Mongoloids bearing Y-chromosomal N3-lineage expanded somewhere from Northern China after the end of the Ice Age (ca. 14 000 BP) and colonized Siberia that was then sparsely inhabited by the remnants of Paleoindians (Q). Today, the N3-lineage is the main male lineage of the Uralic and Chukotko-Kamtchatkan speakers. It is also present in the Eskimo-Aleut group that actually came into being as a mixture of North-East Siberian Paleoindians (Q) with N3-Mongoloids. However, the marked presence of the Paleoindian Q-lineage in Kets, Yukaghirs and Nivkhi reveals that their isolated languages may be descendants of the Old Siberian Paleoindian language. 82.100.61.114 12:04, 1 June 2007 (UTC)


 * That's interesting and deserves further research, but don't confuse genes and languages. They are not always inherited in the same way. David Marjanović 21:03, 3 June 2007 (UTC)

What is Fortescue's real position?
Actually, although previous scholars may have been arguing for a genetic relationship between these languages, Fortescue (1998) does not do so. See the following from his own book: "...although it has not been proven conclusively that the Uralo-Siberian languages constitute a deep genetic stock... the attempt has brought us considerably closer to establishing the reality of an ancient Uralo-Siberian mesh displaying a particular typological profile, out of which most of the existing circumpolar languages of the Arctic arose. Recall... that my conception of a 'mesh' covers any degree of historical relatedness between a group of erstwhile geographically adjacent languages linked by relationships of lexical and/or phonological or structural 'family resemblance'. This ranges from Sprachbunds of unrelated languages, through interlocking chains of languages where the ends are unrelated by where there is considerable overlap and actual language mixing in the core region, to situations where all the ingredient languages are ultimately derived from a single ancestral proto-language but the time depth is simply too great to prove it, and finally to cases of traditional language meshes known exclusively to involve related languages (such as the Northern Athabaskan one)" (Fortescue 1998:230). --149.159.2.216 03:39, 23 September 2006 (UTC)


 * Fortescue qualifies the central thrust of his book in a cautious way, but it remains quite clear: he believes there is a genetic relationship between Uralic, Yukagir, Chukotko-Kamchatkan, and Eskimo-Aleut. He reconstructs the sound system of the proto-language (pages 124-133) and offers a list of 94 lexical items reconstructed to the proto-language (pages 152-159), along with extensive remarks on its morphology (chapter 4) and typology (chapter 3).


 * Typical of Fortescue’s modest but unmistakable claims for the validity of a Uralo-Siberian genetic node is this statement (page 86):


 * I must stress that my use of the term ‘Uralo-Siberian’ is to be understood (for the time being at least) as referring to a mesh rather than to a strict genetic unit for which a proto-language can be reconstructed in detail, although, as will be seen in the following chapters, it does approach the level of reconstructability.


 * Compare this statement (page 219):


 * If the language spoken immediately north of the Sayan mountains at a certain time began to expand outwards, eventually to cover most of Siberia, it would have had its roots in a mesh focused in this area but would have approximated to a unitary proto-language as it expanded via the adjacent spread zone.


 * Note also Fortescue's chapter title, "Who could have spoken Proto-Uralo-Siberian - and where?"


 * Thus, while one may agree or disagree with Fortescue’s thesis, it must be acknowledged that it essentially posits a genetic relationship between the languages he classifies as “Uralo-Siberian”, although he is also sensitive to areal interactions that may have ensued before and after the language spread territorially. VikSol 03:37, 18 October 2007 (UTC)

Dead links
Some of the links seem to be dead. I'm placing the tag after the broken links until anyone finds working ones.Pe t 'usek [ petr dot hrubis at gmail dot com ] 15:51, 1 November 2015 (UTC)

External links modified
Hello fellow Wikipedians,

I have just added archive links to 3 one external links on Uralo-Siberian languages. Please take a moment to review my edit. If necessary, add after the link to keep me from modifying it. Alternatively, you can add to keep me off the page altogether. I made the following changes:
 * Added archive https://web.archive.org/20071025101035/http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0105&L=nostratic&D=1&F=&S=&P=725 to http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0105&L=nostratic&D=1&F=&S=&P=725
 * Added archive https://web.archive.org/20071029075108/http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0106&L=nostratic&D=1&F=&S=&P=193 to http://listserv.linguistlist.org/cgi-bin/wa?A2=ind0106&L=nostratic&D=1&F=&S=&P=193
 * Added archive https://web.archive.org/20120218124335/http://www.nostratic.ru/books/(82)Nivkh.pdf to http://www.nostratic.ru/books/(82)Nivkh.pdf

When you have finished reviewing my changes, please set the checked parameter below to true to let others know.

Cheers.—cyberbot II  Talk to my owner :Online 08:59, 13 February 2016 (UTC)

Presenting a fringe theory as mainstream
There's a heck of a lot of content in this which falls very far outside the scholarly consensus, not to mention some very bad linguistics in the word pairings sections. I can respect that that's what was published, but this article should not treat the subject matter as a given or even realistically accepted by scholarly consensus, especially considering the references to Nostratic that were present. I'd definitely ask some other linguists to help clean this up if possible, a theory with this little support doesn't need this in-depth a wikipedia article justifying its existence. At least not without a significant counterpoint making its place in the world of historical linguistics abundantly clear.

I've added the Fringe Theory tag for now, which is absolutely appropriate until a more neutral POV can be established in the article.

Warrenmck (talk) 03:35, 14 June 2023 (UTC)


 * I came here to bring up a similar point, in that this article needs a solid section detailing the criticisms of the theory as well as why it is still seen as putative. AengusB (talk) 23:33, 19 August 2023 (UTC)
 * "putative"
 * That's putting it very diplomatically. Warrenmck (talk) 21:21, 21 August 2023 (UTC)
 * Yeah, this article presents a disturbing lack of the mainstream view's criticisms of this proposal, a lot of the evidence is presented as undisputed fact. This article definitely needs to be revamped. – Treetoes023 (talk) 19:48, 23 August 2023 (UTC)
 * I've reverted a substantial number of changes added by @ValtteriLahti12. Fortescue should not be the primary source for the data in this article and he should not be treated as inherently credible on this topic considering its wide rejection in linguistics circles. The acceptance of this proposal, at present, is absolutely fringe, and we need to remember that a neutral point of view on this topic necessitates an article that treats the relationship as implausible, not one which presents the long-rangers "findings" as co-valid with the rest of the entire field. Their acceptance (or, more specifically, lack thereof) matters and the recent changes have drastically reduced the factual utility of this article. I'm not trying to accuse the editor of that article of editing in bad faith, I just don't think that the additions which exclusively cite Fortescue and weakly mass-compared cognates is acceptable on Wikipedia.
 * @ValtteriLahti12, if you very strongly object to "fringe theory" I'm open to suggestions, but other than Altaic itself (as rejected as it is it still has serious adherents) and Dené–Caucasian languages I'm aware of no current long-range macro-family proposal with any meaningful degree of support among the wider linguistics community, though that could obviously be wrong. Warrenmck (talk) 21:50, 23 August 2023 (UTC)
 * Hei! Before I type my question I want to make it very clear that I am biased here (as someone who finds the Uralo-Siberian hypothesis very interesting). The families in question aren't my area of expertise and I am doing my best to avoid bias but I want to let you know in case any partiality shines through.
 * I mainly wanted to bring up the "widely rejected" description in the infobox. I agree that it's definitely a fringe theory. Most other linguists whom I've mentioned it to haven't even heard of the theory, and they're skeptical of whether it be true or not. The thing is, their skepticism comes from the theory being new and sounding unbelievable. It's fair to say it is, as this is a broad-reaching connection that is being postulated. My reasoning for questioning whether "widely rejected" should be in the info box is that when I've looked for actual papers on the theory, I've mostly found sources arguing for it or providing proposed evidence. I've been looking for sources critical of the theory and have had trouble finding papers on it. I agree that the theory is fringe and lacks widespread acceptance, and that it is quite speculative. Compared to fringe theories like Altaic, it seems to me that Uralo-Siberian has far fewer linguists saying they outright believe the grouping is inaccurate, and more saying they just don't know. You're an actual historical linguist and I am a university student so I want to make it clear that I'm not trying to contradict your expertise! You say that it's widely rejected, and I recognize that you're actually in a position to know this! My concern is that the current wikipedia page lists the "widely rejected" without sources. As I said, I'm a linguistics student and I'm having difficulty finding sources that are outright critical of the hypothesis. Even though I believe you when you state the theory's lack of support, I worry that others (ie linguistics students, hobby linguists, etc.) would be less likely to acknowledge that you're actually an expert in the field. So because of that, I'm asking whether there are any reputable sources that are specifically critical to the theory that could be added to the page? Isfjella (talk) 21:18, 26 October 2023 (UTC)
 * Hei hei! So a couple of things to address here, but first and foremost I generally want to say I agree with you! I think there's a big problem with the categorization of macrofamily proposals and we should probably standardize them, because right now there's a fairly random approach and I think there's a risk of overstating the amount of eyes that have been on a given theory. Also, while I am a historical linguist I publish in geosciences, so I would argue that as a student you're probably more up to date on your work in many ways!
 * The thing is, their skepticism comes from the theory being new and sounding unbelievable.
 * The people who know the languages and have looked at the theory have a fairly negative impression, as far as I understand it. I can look at something like Déné-Caucasian generally warmly but lack the expertise for a strongly informed opinion.
 * I've mostly found sources arguing for it or providing proposed evidence. I've been looking for sources critical of the theory and have had trouble finding papers on it.
 * So there's been big discussions on this exact issue on the FTN (where there's one ongoing now you may want to join in on) and linguistics wikiproject which is, essentially, academics aren't in the habit of publishing statements that fringe theories are fringe. Rather, they ignore them as outside the scope of their research. For example, pick a fringe topic in physics that doesn't have a huge amount of adherents and try to find credible papers on it, and you may struggle. This is a broader problem for Wikipedia, imo, but it's a serious one in these macrofamily articles, where someone publishes a paper and academics go "nah" and move on, except maybe a small core of researchers.
 * You're most likely to find sources critical of a theory when that theory had signifficant historical support, i.e. Altaic, Nostratic, etc.
 * And on that note;
 * Compared to fringe theories like Altaic
 * I think you should tread with a slight degree of caution here, there are still some serious, credible Altaicists and while I think advancing Altaic as a given is a fringe stance (and I think it's not a real family, to be clear) I do think it's a bit riskier to call that specific family a fringe theory. When I added the PROFRINGE tag to a lot of macrofamily articles I pointed left that one with a NPOV tag instead.
 * it seems to me that Uralo-Siberian has far fewer linguists saying they outright believe the grouping is inaccurate, and more saying they just don't know
 * There is a chance you've done more digging on this topic than anyone else on Wikipedia. Be bold! That said, the reason I elected to roll with "widely rejected" is the use of mass comparison as a method to arrive at the family proposal. Mass Comparison itself is widely rejected in linguistics circles and nothing that comes from it is treated as inherently credible in the absence of other, accepted evidentiary threads. This transitively makes the macrofamily proposals arrived at using mass comparison equally spurious, and this is key, in the absence of other sources of evidence (since mass comparison can highlight relationships, it can just also imagine them, as well).
 * I think perhaps it's best for us to take this to the broader linguistics wikiproject and come up with a standardized set of labels. Warrenmck (talk) 02:59, 27 October 2023 (UTC)

Plausibility and Long-Ranging
Is Fortescue a long-ranger? Is Uralo-Siberian anything like Altaic? I saw Josh(NativLang)'s videos on Proto-World and Altaic. With information provided by him about fringe macrofamilies, I wonder where Uralo-Siberian counts as one... if Fortescue can account and ever accounted for borrowings... if he or any proponent of his hypothesis can do what I call the Blust-Vovin method of gathering six words from Uralic languages like Proto-Uralic or Finnish, Eskaleut or its protolang, and each of the Siberian languages or their protolang(s), each of the six words possessing the same meaning and completely predictable sound correspondences as one of the six words of each of the other languages. Hopefully, they'd guys will remember the requirements to avoid borrowings and unaccounted segments, yet allow identical semantics. Kaden Bayne Vanciel (talk) 06:00, 16 July 2023 (UTC)
 * Fortescue is far from being a long-ranger like Starostin or Bomhard. He is a lumper, but a careful one and is certainly not a Nostraticist. — Sago tree spirit  (talk) 14:24, 28 July 2023 (UTC)
 * My personal editor-turned-linguist opinion is that Fortescue's lexical reconstruction work is sloppy and unreliable; regardless he's been the only one doing anything much actually about Uralo-Siberian in print in the last few decades (and is the one who established the name), so naturally that tends towards his work getting oversized attention. The base hypothesis per se remains at least plausible though and there are several ongoing strands of work on it too; hopefully we get something better out one of these decades.
 * For now more editor attention probably should be put here on other, older work that e.g. shows close similarities in person and number marking in Uralic and Eskimo (unless that rather belongs in the majorly overlapping Eskimo–Uralic languages article); or, if anyone wants to integrate some views from a recent review article, Georg 2023, "Connections between Uralic and other language families" (from the recent Routledge handbook of Uralic ed. Abondolo & Valijärvi) has a couple pages of well-informed comments. Ending quote: "Fortescue's observations are encyclopaedic, and often innovative and inspiring, but the picture arrived at is mostly one of disiecta membra." -- Trɔpʏliʊm • blah 14:27, 30 October 2023 (UTC)

Validity for Uralo-Siberian?
(This could add more to my previous topic. I forgot I even talked about it.)

Several months ago while looking at a video of its uploader creating a fleshed-out version of Vostyach similar to DJP's fleshing out of the Game of Thrones and Dune languages, I said this: "I heard that Uralo-Siberian is controversial, but I never looked into why. Could it be like Proto-World and Altaic? Is Michael Fortescue(and maybe Diego Marani) anything like G.J. Ramstedt, Nicholas Poppe, Joseph Greenberg, Merritt Ruhlen, Oleg Mudrak, Sergei Starostin, and even Anna Dybo? Is Fortescue a long-ranger fishing for resemblances between words of different languages/protolangs where properly-sharpened minds see differences in complex histories?"

A few weeks ago, someone named Sophia Schlier-Hanson replied: "Some long rangers are fringier than others. Ruhlen and Bengtson do nothing BUT their thing— collecting dictionary words that kind of look similar and arbitrarily making up a word that kind of sounds like all of them, a far MORE primitive and amateurish than the approaches 18th century philologists who first noticed pan-IE cognates were using. Greenberg is a decently well respected Africanist who HAS successfully identified some large, very old families on JUST the right side of max time depth for the comparative method, but in his old age he got cocky and got in with Ruhlen and Bengtson for some reason. Starostin overextended the comparative method a few millennia past its usual limit, did all his work with midcentury through ‘80s reconstructions which are now badly dated, and doesn’t do well at filtering out loanwords, but he WAS the first to attempt comparative method reconstructions of several Siberian indigenous language groups and his work is close enough to methodological respectability to be okay for plausible, consistent diachronic conlanging if not actual academic work building on it. Fortescue, about whom you are asking, is an actual Siberianist who from everything I’ve seen uses perfectly solid methodology, iirc has published a couple etymological dictionaries that were first of their kind, and is one of the foremost experts on some of the languages he’s studied."

I thanked her, and she responded: "You're welcome! I actually quite like Fortescue (and Vajda, the other big name in Pacific Rim historical linguistics, a rather niche special interest of mine). Happy researching/ conlanging! :D"

I would look up the Uralo-Siberian article on Wikipedia, and the preface says it is "considered a fringe theory by linguists" and "utilizes mass comparison", though those pieces of information lack citations. Whoever decided to put those in the article... where did they even get those ideas? As in, where is the evidence to prove it? Which well-ranked linguistics(Campbell, Nichols, Georg, the late Vovin, etc.) are against the hypothesis?

I would ask people on Reddit, and they told me about Georg writing a review a few decades ago for "Language Relations Across the Bearing Strait". https://www.jstor.org/stable/30028571 This information could help out, though it might be outdated. Kaden Bayne Vanciel (talk) 02:39, 7 November 2023 (UTC)


 * There was a large discussion about this on both the linguistics wikiproject and the fringe theory noticeboard, but the truth is they are fringe in nature at present. Several of the linguists you mentioned, while broadly considered competent and credible linguists, are nonetheless pretty far off of the deep end in terms of doing some research that the wider linguistics community simply rejects. It's possible more nuance is needed in differentiating some of the people working on it since there's a spectrum of serious linguists to fringe theorists, but at the end of the day it's not really in dispute that Uralo-Siberian is pretty widely rejected and that most of the work on it does utilize mass comparison. If more evidence from more mainstream techniques come out, then we can update it as needed. Warrenmck (talk) 20:08, 7 November 2023 (UTC)
 * If what you say is true (and I have no reason to think otherwise), is it possible to add citations to the "fringe theory" and "mass comparison" statements in the lead section? Rexodus (talk) 17:11, 27 November 2023 (UTC)
 * Would be interesting if some source can be dug up for the accusation of "mass comparison", but this seems clearly false: he works mainly with well-established Uralic etymologies from earlier literature, has done himself the groundwork in getting Chukotko-Kamchatkan and Eskimo lexical reconstruction to good shape, and clearly asserts sound correspondences between them. To the extent his etymologies are poor, they're poor for entirely usual reasons like details of phonology and issues with semantics. E.g. the worst systematic problems noted in a review Holopainen (2022) of his and Vajda's recent book are probably neglecting the Proto-Uralic contrast *ï : *a and the inclusion of some words only reconstructed as far as Proto-Samoyedic. This is by no means mass comparison in the usual sense of adducing disparate modern words, that may not be even themselves related, as representing some member family. -- Trɔpʏliʊm • blah 19:46, 10 December 2023 (UTC)
 * Since no sources have been presented so far for the labels "fringe", "widely rejected" and "mass comparison", I have removed these statements. I have added a citation to the latest review and adjusted the citation of Georg's article to the page that actually deals with Fortescue. –Austronesier (talk) 18:35, 13 January 2024 (UTC)

Merge with Eskimo-Uralic?
Should we merge this article with Eskimo-Uralic? Uralo-Siberian still is a one-scholar hypothesis that builds on the much wider studied Eskimo-Uralic proposal. In its latest version, it only differs from the latter in the inclusion of Yukaghir.

So we could well turn this article with borderline WP:SIGCOV into a short section in Eskimo-Uralic, describing its more extensive scope and whatever is written about it in secondary sources. I haven't seen a secondary source yet that goes into such detail about Fortescue's evidence as the current version of the article does, even after the moderate trimming by User:Warrenmck. Which of course means that we should cut down the presentation of his evidence to a minimum to maintain due weight. Austronesier (talk) 18:31, 13 January 2024 (UTC)