Talk:Proto-Indo-European homeland/Archive 1

This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

The references are a mess. There are a lot of links to nonexistent Harvard-style references. Were these copied and pasted here from another article? If so, could someone bring the refs themselves here too? —

Under the graph of the haplogroups the "Y-" had to be added, otherwise the normal reader had to guess weather X or Y-DNA is meant. HJJHolm (talk) 14:24, 18 September 2010 (UTC)

I propose the merger of Paleolithic Continuity Theory into this page, probably reduced to a paragraph or two in the "dissenting views section", as I feel it lacks the notability to warrant an article ofits own.

Boynamedsue (talk) 17:22, 5 March 2011 (UTC)

States that there is no theory that the majority of linguists support. The Pontic-Caspian theory is in fact well-supported, has been accepted by the majority of linguists for quite some time, and is taken as the consensus theory in the main article on Proto-Indo-European. Winter Maiden (talk) 04:48, 4 August 2011 (UTC)

An image used in this article, File:Haplogroup F (Y-DNA).jpg, has been nominated for speedy deletion at Wikimedia Commons for the following reason: Copyright violations What should I do? Don't panic; deletions can take a little longer at Commons than they do on Wikipedia. This gives you an opportunity to contest the deletion (although please review Commons guidelines before doing so). The best way to contest this form of deletion is by posting on the image talk page. If the image is non-free then you may need to upload it to Wikipedia (Commons does not allow fair use) If the image isn't freely licensed and there is no fair use rationale then it cannot be uploaded or used. If the image has already been deleted you may want to try Commons Undeletion Request This notification is provided by a Bot -- talk ) 22:10, 17 August 2011 (UTC)

Searches for this word all seem to reference the same sentence, apparently derived from a particular scientific paper on genetics. The original reference is no longer extant, but it seems to be this one-

http://www.sciencemag.org/content/290/5494/1155.abstract

The word does not seem to be in general usage. So far as I can see it means lepto- "slender, light" dolicho "long" morphous "shaped". So, tall, slim people. I doubt that anyone coming to a wiki page for information would understand this word. It is contrasted with "robust". Wouldn't "gracile" be sufficient?82.71.30.178 (talk) 08:51, 12 January 2012 (UTC)

There is a distinction between an isogloss, which belongs under Wave model, and phylogenetic branching, which belongs under Tree model. An isogloss does not define a branch point, it defines a wave based on a single feature. A branch is defined by comparing multiple phonological, lexical and morphological features. In 19th and early 20th century linguistics, centum and satem branches were defined. They were never really satisfactory, being based mainly on only one feature. Other features indicated a different division was warranted. Finally more progress was made on Hittite and Tocharian was discovered. These did not conform to any such division. Some items were centum, some satem. So, the phylogenetic division was abandoned. There are no centum languages or satem languages. The distinction was recognized as an isogloss, a single feature that spread like a wave over a range that did not correspond to language boundaries. Dialects might be defined on it, but not languages. Our article here does not recognize the difference. This is understandable, as all books before a certain date speak of the centum languages and satem languages. Single isogloses do not define languages, they define dialects and dialect continua. There is too much editorial conjecture as the editor tries hard to put it all together, but unsuccessfuly. We need to stick to the sources on this, and some of them have to be sufficiently up to date to give us an accurate picture of the current state of affairs.Dave (talk) 05:09, 2 February 2012 (UTC)

My understanding is that PIE started out with an initial k (or some sort of hard c) in a great number of words which in certain daughter branches evolved into s via something like ch as in bach, then sh. This process is called satemization. Those branches that split off the earliest, including Celtic and Tocharian, experienced no satemization. Satemization was strongest in the branches that remained within the more centrally located dialect continuum for the a longer period of time. Or to put it another way, the initial k was a feature of early PIE while satemization was a feature of late PIE. In the Celto-Italo-Germano-Balto-(Slavic)-Indo-Iranian continuum, Celtic is the least satemized (not at all), Indo-Iranian is totally satemized, and the branches in between are partially satemized, for example ch or kh in Germanic, sh in Baltic. The point is, it's not just the words for hundred. Zyxwv99 (talk) 14:52, 2 February 2012 (UTC)

You've come thus far, maybe you will be willing to go a little further. I never said it was only two words. First of all, the whole change is only one character, or feature of the language. If a phylogeny is to have any hope of being credibly valid, it must have many of all sorts, phonological, lexical and morphological. Second, there is a distinction between a genetic branch and an isogloss. They aren't the same at all. The branch is a feature of the Tree model, which you need to read forthwith (your friends, such as I, can tell you these things). The isogloss is a feature of the Wave model (which I have only just started revising), and never the twain shall meet. The tree model describes branches in the subsequent evolution of a hypothetical proto-language. There have to be definite discrete languages with definite discrete borders. The isogloss is a line marking the furthest extent of a single feature. The feature does not have to belong to one language. The isogloss can cross language borders. In the big picture, isoglosses have a home in explaining dialects and dialect continua. The original goal of the wave modelists was to replace the tree. The did not think there were any languages as such, only dialect continua. Now, on the one hand you speak of a centum-satem branch, which belong to the tree model. In the same breath you speak of the centum-satem isogloss, which is not the same thing at all but belongs to the wave model. One final thing. How do we know it is an isogloss and not a branch? Well, the neo-grammarians were very strict about their trees. You had to find a linguistic rule, such as Grimm's Law, or Verner's Law, or many other laws, which would create a shared innovation and apply universally to the branch and distinguish it from any other. The problem with centum-satem is, it was not a law. It did not apply universally. The questioning started with Tocharian, which was in the eastern range but was a centum language. Then it went on to Hittite, which has some features of one and some of the other. Moreover, this satemization of which you speak was imperfectly applied to its languages. The whole concept of satemization belongs to the wave model and not the tree model. A law is universal. For example, no one speaks of a "Grimmization". Now, the there is in fact a partial Grimmization along the Rhine creating a dialect continuum. This is one of the problems with the tree model. Nevertheless it is such a good explanation of basic differences in discrete languages that the wave model has not been able to replace it. So, what we end up with is a mixture. The branch into satem and centum has been gone for at least 30 years, probably more. Sturteyvant and Bloomfield opposed it, among others. I don't have the time right now but if you Google books on "Hittite centum" you will find some snippet views that substantiate what I am saying. Don't feel bad - I was brought up on the centum-satem split and it was somewhat of a shock to me to discover 15 years ago that it was gone now. I heard it from the lips of Calvert Watkins, who explained that he was finally convinced by his best student, Craig Melchert, whom not long after I met at a wine party at his house. Watkins was hosting some Russians. I was taking a course. I don't want to drop names on you only to demonstrate the division really is gone. Our editors can't be blamed for not keeping up with everything; after all they are mainly students and besides no one can read everything or be expert in everything. I am shortly going over this article (slowly) but your bringing this up probably underlines the need for a reference there and a little of the history. That maybe should be under Centum Satem Isogloss. Gosh there is so much to do. Well this took some time I could have spent on the article but it probably was worth it as all the editors now need to be convinced and multiple articles changed. The Centum Satem Isogloss article looks all right at first glsnce but I should check it. Later.Dave (talk) 15:50, 5 February 2012 (UTC)

"The Proto-Indo-European

Going to rewrite. Vague and misleading. An urheimat does not explain the origin of a language; it is a geographical location, not a linguistic fact. The language does not explain the origin of the people. Postulating a location does not establish any identity, only a location. What the editor probably means is, which archaeological cultures might be associated with the speakers, but typically it is the other way around, they want to see how far the culture matches an identity inferred from the language.Dave (talk) 16:18, 5 February 2012 (UTC)

I think "Neolithic farmers, possibly coming from Anatolia and settled there, developing pastoral nomadism." should read "Neolithic farmers, possibly coming from Anatolia had settled there, developing pastoral nomadism". A mistransfer I often make myself, but without Cavalli-Sforza 2006 in hand, I can't fix this.--Wetman (talk) 20:56, 22 April 2012 (UTC)

Since January this article has been mostly a discussion of the homelands of the various branches of the Indo-European family rather than, as the title suggests, the Proto-Indo-European homeland. One option to fix this would be to rename the article to something like "Indo-European urheimats", but I think such accepting wide scope makes the article too unwieldy. The original change was by a now blocked

Forget that – looks like the sock's removal of content from

t•c

14:20, 8 December 2012 (UTC)

This article makes the point that early contacts with Uralic, especially early Indo-Iranian loanwords in Uralic, present difficulties for the Anatolian hypothesis (and similar proposals such as the Armenian hypothesis, as well as the fringe Out-of-India proposal) that are not usually paid respect to by archaeologists. While the argument of very early Indo-European loanwords in Uralic, or even a possible original "Indo–Uralic" relationship, is open to doubt, the presence of early Indo-Iranian loanwords in Uralic seems not to be easily discounted, and the best argument for a northern origin of Indo-Iranian, which supports the Kurgan hypothesis but presents considerable problems for the main competitors (as well as Out of India, of course).

I might note that a northern location of Proto-Indo-Iranian (somewhere north of the Caspian, consistent with an identification with the Poltavka/Sintashta/Andronovo context) also shifts the balance in favour of (especially Central/Eastern) Europe and against Asia even more in the context of the locations of the individual branches. Only Anatolian, Tocharian, Armenian and the fragmentarily attested Phrygian are based in Asia, whereas Germanic, Celtic, Italic, Baltic, Slavic, Albanian, Greek and Indo-Iranian as well as various other (extinct and fragmentarily attested in antiquity) branches of unclear affiliation are based in Europe, in the sense that there is general agreement that the relative proto-languages are placed in Europe via general and – except perhaps Indo-Iranian – overwhelming consensus. (Possible evidence for origins in Europe of Phrygian, Armenian and Tocharian is even completely neglected here.) That's a landslide in favour of Europe, and it is difficult to argue that the greatest diversity of Indo-European is found in Asia instead, even within the wiggle room that changes in the tree such as acknowledging Balto-Slavic as a single branch afford. (Unless, of course, one manages to successfully reduce all the European branches to possibly as few as two, but how would Proto-Italo-Celto-Germano-Balto-Slavic and Proto-Balkan-IE – which would, however, have to include Armenian and Phrygian – or Proto-Graeco-Aryan – which might well include Armenian or Phygian, or both – differ from PIE? These large-scale proposals are difficult to defend. Similarly, a Proto-Satem proposal would not affect the European predominance, either. Where the deepest divisions in PIE are found is controversial anyway, hence my throwback to an agnostic comb-type classification which treats all obvious branches as primary.)

The GeoCurrents article also points out that even if Anatolian was the first branch to split off (which is by no means a given, nor would it mean that the rest of IE is monophyletic, i. e., forms a single branch, at all), it would say nothing about the placement of the IE Urheimat; Anatolian could equally be an early off-shoot, as the Kurgan hypothesis envisions. Also, there are some common but unjustified implicit assumptions, such as that Anatolian was always spoken in Anatolia, or that there is linguistic continuity in the region of the Urheimat. --Florian Blaschke (talk) 18:00, 3 January 2013 (UTC)

That would be an interesting read!

I myself prefer Popper, and a good dose of humility when addressing questions larger than any individual or even several.

People are asked to provide sources for all kinds of things, can we objectively define and quantify the academic credibility of something. Sounds a bit vague... — Preceding unsigned comment added by 2605:6000:1800:6D:A4B5:319D:3B9C:C25D (talk) 07:58, 1 July 2014 (UTC)

See

WP:RS for what we have relevant to Wikipedia article-writing. AnonMoos (talk

) 10:01, 1 July 2014 (UTC)

Also, the next sentence in the article ("There is no single archaeological pattern that might correspond to a migration on an appropriate geographic scale throughout Europe.") ignores the fact that recent reconstructions have explored the possibility that much of the early expansion of Indo-European was based as much on "elite recruitment" as on pure military conquest as such (see The Horse, the Wheel, and Language etc.)... -- AnonMoos (talk) 18:59, 9 December 2012 (UTC)

I agree. The lead should reflect what the body of the article says, i.e. that the Kurgan hypothesis has majority support at the moment (from linguists and archaeologists etc – it's an interdisciplinary question, after all) with the Anatolian hypothesis as a runner up.

I interpret that second sentence as saying there's no archaeological evidence for a Europe-wide migration and so backing up Anthony. It is a big vague though. And technically incorrect, since the spread of farming fits the bill.

t•c

21:21, 9 December 2012 (UTC)

The sentence is about linguists, and I'm not sure that archaeologists are as solidly behind it as linguists (if they are, then it would seem to be a change from the 1980s, though I know much less about archaeology than linguistics). AnonMoos (talk) 08:01, 10 December 2012 (UTC)

Why do you write here, if admittedly having no deeper overview??? — Preceding unsigned comment added by 195.4.79.26 (talk) 07:04, 11 March 2014 (UTC)

What does that mean? My having a closer connection to academic linguistics than to archaeology doesn't disqualify me from offering suggestions to improve this article. Not everybody can be an expert equally on everything... AnonMoos (talk) 10:06, 1 July 2014 (UTC)

Was renamed so, because the former title "Glottochronology" was consistent only with the upper part. Moreover the two entries did not convey a minimum overview about the involved problems. HJJHolm (talk) 14:53, 18 September 2010 (UTC)

The now noted "Baltic hypothesis" does not exist. It has been a misunderstanding by J. Mallory of the views of the late Wolfgang P. Schmid, who always and only maintained particular rich connections of the Balic languages to the other Indo-European ones, never a Baltic homeland. Schmid had always been very angry about this misinterpretation. If there are no serious sources for this alternative, it must be cancelled a.s.a.p. HJJHolm (talk) 08:12, 11 March 2014 (UTC)

Ah, OK. But the hypothesis does exist, even if it has been out of fashion for quite some time: as I'm sure you know, it was customary in the early 20th century, especially in German-speaking Europe, to identify the Corded Ware culture with the Proto-Indo-Europeans, hence a Northern European Urheimat which appealed to nationalistic fervour and white supremacist sentiments. That said, the evidence of (especially upper-class) individuals with light hair and eyes in Indo-European-speaking ancient Southern Europe and Asia is suggestive of a Baltic Urheimat. (The high conservativity of the Lithuanian and Old Prussian languages especially has been adduced as another argument, but I don't think it is particularly convincing: it's probably due to the long relative geographic isolation of the speakers of Baltic languages. After all, Iceland isn't the Germanic or even North Germanic Urheimat either.)

People who don't realise that the Proto-Indo-Europeans were called "Aryans" at the time, based on now discredited etymologies, won't understand the connection with Nazi racial ideology: German nationalists regarded light-haired Germanic-speaking Europeans as the purest descendants of the Proto-Indo-Europeans and hence as the "master race", claiming that the Proto-Indo-Europeans, imagined as fair-haired and warlike, had spread their allegedly superior civilisation throughout Eurasia (with the pre-existing swarthier people lumped together as "Semites"). However, this historical baggage, the fact that the Baltic Urheimat hypothesis was abused for the purposes of white supremacist ideology, culminating in the Nazi genocides, is in itself not a reason to reject the hypothesis.

However, Allentoft et al. 2015 (cited, among others, in Blond#Asia) appears to imply that the Southern European and Asiatic branches too can be derived from the Corded Ware, instead of the Yamna, effectively resurrecting the Baltic hypothesis. ADNA evidence (Haak et al. 2015) suggests that the people of the Yamna culture were not particularly fair and that the assimilation of Northern European hunter-gatherers, especially in the Baltic region, resulted in the Corded Ware and the association of Indo-Europeans with light pigmentation. However, it is entirely possible, even plausible (per Klingenschmitt 2005), that the eventual breakup of the final stage of the Indo-European proto-language only happened in the early third millennium, rather than the fourth or fifth, and it is entirely compatible with Gimbutas's original Kurgan hypothesis as well, as her third and final posited wave of migration out of the steppes happened in the early third millennium and involved the Corded Ware.

White supremacists may like this narrative (although one wonders why the Northern European hunter-gatherers needed the help of the swarthy steppe people, and took over their culture and language, if they were so superior, and why there was nothing comparable to the relatively complex and advanced cultures of 4th millennium BC West Asia and Egypt in Northern Europe at the same time, so I can't see this model playing into a white supremacist narrative from a more level-headed point of view, but then, ideologues don't really care for facts and logic), but again, that's not in itself a valid reason to reject it. --Florian Blaschke (talk) 16:46, 28 June 2016 (UTC)

@Joe Roe: here's another source: Asya Pereltsvaig, Languages of the World: An Introduction, p.23 ff. Joshua Jonathan -Let's talk! 15:19, 8 July 2016 (UTC)

Yes, that one is good.·maunus · snunɐɯ· 07:44, 9 July 2016 (UTC)

And Mallory (2013), Twenty-first century clouds over Indo-European homelands. It deals with the steppe hypothesis, Near Eastern proposals, and the Anatolian hypothesis. So, I'm not convinced yet that the Armenian hypothesis should be put in the fringe-section. Joshua Jonathan -Let's talk! 11:52, 11 July 2016 (UTC)

Is the "Armenian hypothesis" fringe, given the R1a-diversification 'in the vicinity of present-day Iran' (Underhill et al. (2014/2015), and the connection between Caucasus and Iran (Lazaridis 2016)? Joshua Jonathan -Let's talk! 19:38, 6 July 2016 (UTC)

Yes.·maunus · snunɐɯ· 19:44, 6 July 2016 (UTC)

Haak et al. (2015)

"The Armenian plateau hypothesis gains in plausibility by the fact that we have discovered evidence of admixture in the ancestry of Yamnaya steppe pastoralists, including gene flow from a population of Near Eastern ancestry for which Armenians today appear to be a reasonable surrogate (SI4, SI7, SI9)."

Joshua Jonathan -Let's talk! 20:31, 6 July 2016 (UTC)

I wouldn't call it "fringe" because it was proposed by legitimate scholars, but it doesn't have mainstream support either. I think using a single line in a single primary source to argue otherwise is

WP:OR

. The overwhelming majority of recent review/secondary sources on IE origins put it down to (at the most) two possibilities: the steppe or Anatolia.

And my thinking in reorganising the article into mainstream vs. historical/fringe was precisely so that we can focus on that current debate, without cluttering readers' heads with a dozen other putative homelands. Joe Roe (talk) 21:57, 6 July 2016 (UTC)

@Joe Roe: it's not a single line, it's a whole section. Can you access the source? I don't remember if it's publicly accessible; anyway, I could go through the article to see if they mention more on it. My point is, that by putting the Armenian hypothesis together with OIT in one section, you clearly put it in the category of lunic errors. I do object to that; the recent publications on the origins of R1a-diversification and the role of north-western Iran in the spread of agriculture-related genes, suggests that the Armenian hypothesis is not that weird after all, as reflected in the comment by Haak et al. (2015). Haak and his colleagues are definitely mainstream; Haak himself is top of the bill. So, we do have to do soemthing with this info. As it is now, I've listed four hypotheses, with Haak et al. (2015) as a reference; maybe we could create a separate section on 'minority-views', apart from 'fringe'? Though, that sounds arbitrary, doesn't it? have you got further ideas? Best regards, Joshua Jonathan -Let's talk! 06:36, 7 July 2016 (UTC)

PS: this article is on the language, of course; in that respect, the Armenian hypothesis is not mainstream. Haak et al. (2015) is about genetics; that's a different cup of tea, though they explicitly write:

"While our results do not settle the debate about the location of the proto-Indo European homeland, they increase the plausibility of some hypotheses and decrease the plausibility of others as follows:"

Okay, the Armenian hypothesis as a subsection of the Steppe-theory, on the genetic ('hardware') origins of the steppe people? Joshua Jonathan -Let's talk! 06:39, 7 July 2016 (UTC)

• A couple of articles that provide indirect support for a homeland scenario that is otherwise not among the top-three supported scenarios do nothing to change the status of the field. We should not even refer to these papers until someone else discusses them in secondary sources. ·maunus · snunɐɯ· 06:54, 7 July 2016 (UTC)

Disagree. Genetics is not at the same speed as linguistics. By the time the seconadry sources pick-up on the most recent genetic research, the research is way further already. The journal papers are the relevant sources here. Look at Lazaridis et al. (2016); it was even published as a pre-print, and is yet highly relevant in this regard. and Mallory (1997) is not exactly up-to-date in this regard; in doesn't even come close in reflecting the recent discussions. Joshua Jonathan -Let's talk! 07:47, 7 July 2016 (UTC)

That is irrelevant because genetic evidence is not evidence for linguistic heimat proposals untill someone writes about it in relation to linguistics. They are not relevant sources, because they ar enot about language. Non-linguistic data is circumstantial to linguistic heimat proposals. And as a general policy we do not include primary sources untill they are referred to in secondary literature. Especially not if their claims and conclusoins are contrary to established mainstream views. This is exactly the same for genetics studies.·maunus · snunɐɯ· 08:44, 7 July 2016 (UTC)

• I think it is fine to include the Armenian hypothesis, and it is fine to call it a minority view instead of a fringe view, but we should still include it following all of the normal policies for fringe content namely avoiding to give it undue weight and avoiding to use primary sources as support. Following these policies I would not want to mention the studies you refer to here untill they receive some sort commentary from scholars who are not themselves proponents of the Armenian hypothesis or authors of the studies. This is just simple measures of caution that should be followed in all areas of science writing on wikipedia.·maunus · snunɐɯ· 08:55, 7 July 2016 (UTC)

"Minority view" sounds better. Haak et al. (2015) is cited 108 times already; some homework to do, shifting them through... Joshua Jonathan -Let's talk! 10:35, 7 July 2016 (UTC)

NB: isn't Haak et al. (2015) a secondary source on the homeland hypotheses, in its discussion of these hypotheses? Joshua Jonathan -Let's talk! 13:06, 7 July 2016 (UTC)

Agreed. Regardless of whether genetics moves faster than linguistics,

WP:PSTS still applies. We need to be careful on basing too much on Haak et al. (@Joshua Jonathan

: I do have access to it, but thanks for asking), when it's:

• A single, primary research paper, per
  WP:NOR
• Published very recently; the way Indo-European studies goes, and genetics for that matter, it could well be considered completely invalid or superseded in a couple of years (cf. the Atkinson and Wade lab's papers on linguistic phylogeny)
• Only covers one aspect of the problem, the genetics of IE spread in Europe, and not the linguistics, archaeology, Asia, etc.

And fortunately we don't have to rely on it, because a number of secondary reviews of the literature have been published very recently (Anthony & Ringe, Pereltsvaig & Lewis, and apparently another Annual Review by Wade & Atkinson is forthcoming), in addition to old staples like Mallory & Adams' encylopaedia and David Anthony's book.

I understand that "fringe" is strong word though, how about we rename the section "Minority and historical hypotheses" and then move the Armenian hypothesis back there? Joe Roe (talk) 13:15, 7 July 2016 (UTC)

Fine with me then, unfortunately ;). But we really have to find another source for a list of mainstream theories. NB: Allentoft et al. (2015) seem to argue against the Armenian hypothesis. It's funny, reading those different papers I get the impression that there's a lot of debate going on in-between the lines. I'd really like to know what those researchers discuss informally, at the coffee-machine so to speak. Joshua Jonathan -Let's talk! 13:30, 7 July 2016 (UTC)

I would suggest some general textbooks on Indo-European - Beekes, Clackson, Mallory & Adams would be good places to find which theories have received any attention in mainstream scholarship.·maunus · snunɐɯ· 07:41, 9 July 2016 (UTC)

Okay, I read some more about this theory; 3rd millennium BCE sounds unlikely indeed. But it doesn't change the questions, or probabilities, which are posed by the studies I mentioned at the start of this thread: the Caucasus/near Eastern genetic contribution to the Yamna-culture. If Underhill et al. (2014) are right in their geographical and time-estimates, it should be quite easy then to postulate some archaeological correlates. Joshua Jonathan -Let's talk! 15:30, 8 July 2016 (UTC)

• I have rephrased the Haak et al quote as a summary instead of as a direct quote, which I think confers undue weight to the study itself, by taking their own word for its significance.·maunus · snunɐɯ· 06:46, 15 July 2016 (UTC)

There is almost no mention at all of the actual linugistic evidence, or the linguistic methods for establishing homelands. We also for example need to consider the distribution and subgrouping of branches as evidence. We should for example probably mention that the Indo-Hittite proposal if accepted based on linguistic data would make the steppe vs. Anatolian discussion moot, since PIE would be steppe but pre-PIE (i.e. proto-indo-hittite) would have been in Anatolia. reading over the article there is a serious undue weight given to genetic relatively to linguistic evidence - the only evidence described in any detail is genetic. This is not a tenable approach.·maunus · snunɐɯ· 07:48, 16 July 2016 (UTC)

Indeed. Coincidentally, I was reading that Antony publication again this morning, and I also figurd that the articles hould start with linguistics, and the linguistic tree. From there, it's 'easy' to explain what theoretical issues are at stake, and to show how they are solved, or not solved, by the various theories. Joshua Jonathan -Let's talk! 15:07, 16 July 2016 (UTC)

The Armenina hypothesis is a subset of near-eastern origin hypotheses. Soviet archaeology seems to also have proposed more southern origins, including the Haraffa culture, and Uruk (Meopotamia) influences. Another subset is Iranian influences; there's more on it, which I will add. The near east altogether between 4,000-3,000 BCE is very complicated, with a multitide of regional cultures which were interacting. Underhill (2014) (origins of rR1a diversification) and Lazaridis (Iranian-Caucasus intraction) are relevant in this regard. Lazaridis et al. (2016) also detected R1b, which is associated with Indo-European, in a Kura-Araxes sample. Looks like they're now pondering over it. Joshua Jonathan -Let's talk! 07:04, 15 July 2016 (UTC)

I am not sure if Armenia qualifies as near east (Anatolia definitely is Near East though). Other than that I think it is ok to have a catalogue of serious scholarly hypotheses that have been made and then failed to get any traction. But the Polar hypothesis that you had a subsection prepared for is not a serious hypothesis and we would have to include a lot of old pseudoscientific hypotheses if that were to be included. Also I would recommend extreme caution with genetic studies that are not specifically linked to the linguistic literature.·maunus · snunɐɯ· 07:11, 15 July 2016 (UTC)

I hardly know what the Polar hypothesis is; wasn't it a Nazi-idea? The info was already in the article, I guess. Joshua Jonathan -Let's talk! 07:25, 15 July 2016 (UTC)

Ok, no it was in fact a veda based idea from the early 20th century. But noone seems to have referred to it since. Glad you won't miss it.·maunus · snunɐɯ· 08:28, 15 July 2016 (UTC)

I put the Polar Hypothesis in with the intention of expanding it down the line. I don't agree that anything is "too fringe" to include – crazy stuff like that is part of the history of IE studies, and if we don't cover that in this article, where else? Joe Roe (talk) 08:44, 15 July 2016 (UTC)

Because of

WP:FRINGE. Not everything needs to be covered - and some things deserve at most a sentence. I dont think the polar hypothesis deserves even that - but if you can show that it has gotten attention in 3rd party RS then perhaps. This is an encyclopedia not an academic freak show.·maunus · snunɐɯ·

08:59, 15 July 2016 (UTC)

There's no question of presenting it alongside the mainstream hypotheses or giving it any weight as a scientific explanation at all, but I think if you look at any historical review of the Indo-European origins problem, late 19th/early 20th century looniness looms large, if only because it played such a big role in marginalising the subject in the later 20th century. Anyway, this is all kind of moot until I (or someone) gets around to writing about it. Joe Roe (talk) 11:05, 15 July 2016 (UTC)

If so, then perhaps a section titled "19th century looniness" would be justified. I dont think the polar theory would deserve mention in such a section either.·maunus · snunɐɯ· 11:16, 15 July 2016 (UTC)

Nice header ;) Regarding "Near eastern": it's the term Mallory used. I don't know how 'correct' those theories are, but at least they've been proposed and discussed; also, the Yamna-culture didn't exist in a vacuum, and these theories seem to be related to these complex interactions. They also show that searching for an even older 'homeland' may be useless on its own, but points to these complex interactions. Joshua Jonathan -Let's talk! 14:59, 15 July 2016 (UTC)

Yes, indeed. Haak et al. (2015) and Lazaridis et al. (2016) trace a genetic influence of Caucasus/north-western Iran on the Steppe-people. The Steppe-people were in contact with the Caucasus (which was in contact with more southern cultures) and the Levant; north-western Iran was in contact with Mesopotamia. Both groups migrated, carrying with them different languages (I suppose), but with partly the same culture (kurgans?) and technology, the Steppe-people north and then east (after mixing with northern Europeans), the Iranians (Kura-Araxes?) east. "The metal and the language" might be a nice paraphrase of Anthony. Joshua Jonathan -Let's talk! 06:03, 17 July 2016 (UTC)

We appear to be pulling the structure of this article in three different directions. I recently reorganised the existing L1 headings for each individual hypothesis into mainstream and non-mainstream sections. Since then, the individual have been hypotheses have been promoted back to L1 headings, but we retained a "Hypotheses" section with duplicate lists of the hypotheses under L2 mainstream and non-mainstream headings, and then some of the L1 hypotheses section were moved to L2s under "other theries", and there's a long-titled L1 section called "Pre-Proto-Indo-European and Near-eastern origins or influences" which I think started out about the Armenian hypothesis, but there's another section on the Armenian hypothesis under "other theories", and it has a L2 under it on the Anatolian hypothesis, which has its own L1 section... do you see where I'm going with this?

I'm not saying it has to be my way (though I did quite like my way), but please let's stop shifting things around all the time and agree on a stable structure we can use to improve this article. I'd propose three principles for structuring the article:

1. One hypothesis, or one closely related group of hypotheses (e.g. "kurgan" and "revised steppe") = one section/subsection.
2. Per consensus in the move discussion above, this article isn't shouldn't just be list of hypotheses, it should also cover broader issues like the theoretical context and history of the IE homeland concept; so individual hypotheses shouldn't be the highest level structure.
3. We need to clearly convey and give
   due weight
   to the fact that there are only 2-3 hypotheses with a significant amount of scientific support, but dozens that are fringe, defunct or were dreamt up by some obscure academic and never heard from again.

And if we can agree on those, I'd suggest the logical way of structuring the article is to bunch together hypotheses into either "mainstream" and "other" sections or into thematic sections which state level of the academic credibility up-front. Joe Roe (talk) 21:59, 15 July 2016 (UTC)

Sorry; I'm trying to figure out how the Caucasian context fits in. A historical section is missing at the moment, and should be added. The point is indeed the Armenian hypothesis, and the Caucasus/Iranian/Anatolian context. On its own, the Armenian hypothesis seems to have little support, but it is being discussed in the context of gentics, and it does seem to be related to a line of thought which favors a more southern "homeland." So, where to fit this in? I'm goi g to self-revert some parts, to see again how that looks. Joshua Jonathan -Let's talk! 04:06, 16 July 2016 (UTC)

What's also missing are serious criticisms of the Steppe-theory; see Mallory (2013) and this blog. Joshua Jonathan -Let's talk! 04:27, 16 July 2016 (UTC)

Here's a blog by Pereltsvaig on the homeland. Joshua Jonathan -Let's talk! 04:58, 16 July 2016 (UTC)

A question regarding the Steppe-theory that's on my mind: R1a-Z93 is also found in East-Asia (Underhill 2014), apparently in the territory of the Afanasievo culture - which is supposed to be founded by a steppe-off-shoot, which are R1a-Z282. Hmm... there's a big question-sign there! Or is it the Tocharian-area? But in that case, there's also a big question-sign, pointing to Indo-European languages south of the steppe. Joshua Jonathan -Let's talk! 05:26, 16 July 2016 (UTC)

• Youre gonna need this one: Chang, W., Cathcart, C., Hall, D., & Garrett, A. (2015). Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis. Language, 91(1), 194-244.

And this one: David W. Anthony and Don Ringe. 2015. The Indo-European Homeland from Linguistic and Archaeological Perspectives Annual Review of Linguistics. Vol. 1: 199-219 ·maunus · snunɐɯ· 07:20, 16 July 2016 (UTC)(Volume publication date January 2015)

Thanks. Joshua Jonathan -Let's talk! 15:07, 16 July 2016 (UTC)

• I don't completely understand why you want to group the Armenian proposal with the Iran, Sogdiana and Mesopotamian connections - but exclude the Anatolian hypothesis. The Anatolian hypothesis is basically a development of Gamkrelidze's and Ivanovs Armenian/glottalic theory - and they all share affinities with the indo-hittite proposal and more or less point out overlapping areas. I alsso dont understand where the "West Asian" label comes from, it is not one I have ever heard - and again geographically speaking Anatolia would fall under the label and Armenia would not.·maunus · snunɐɯ· 17:20, 16 July 2016 (UTC)

I wouldn't mind grouping the Anatolian hypothesis together with the Armenian hypothesis and all the other hypotheses which situate some sort of 'homeland' somewhere between Anatolia and the Caspian Sea. The label seemed to cover this area somewhat; that's why I used it. Regarding the Anatolian hypothesis and the Armenian hypothesis, could you tell more? Renfrew was familiair with Gamkrelidze's and Ivanov's work? To me, it seems that all those theories point to the interactions between various cultures, and the broad range of 'pre-proto-IE' influences. Joshua Jonathan -Let's talk! 19:12, 16 July 2016 (UTC)

Well, the possibility of an Anatolian homeland dates to the early 20th century when Hittite was discovered to be IE - Sturtevants Indo-Hittite was probably the earliest version of it, Gamkredlize and Ivanov's glottalic theory put the homeland in Anatolia and Southern Caucasus. Renfrew of course was familiar with these hypotheses when he realized that Peter Bellwood's idea that language's advance with the spread of faming could be seen as supporting the Anatolian hypothesis since we know that farming entered Europe from Anatolia. Anthon and Ringe mention the conceptual conneciton between Renfrew's and Gamkrelidze's theories. Sure all the theoreies point to those connections and interrelations, but they point in different directions and have very different time frames. ·maunus · snunɐɯ· 20:50, 16 July 2016 (UTC)

You said, maybe jokingly, that the Anatolian hypothesis also situates the homeland in Anatolia, just like the Armenian hypothesis. But it might make sense, indeed, to group those two theories together, together with the Iranian/Caspian influences, in a section "Pre-Proto-Indo-European." Not as a 'homeland', but as regios from which the various influences on PPIE came. This section could open with the Indo-Hittite hypothesis, since the Anatolian languages may be a "cousin", and not a "daughter," of PIE (Anthony 2007). Haak (2015)'s commetn on the Armenian hypothesis, and Lazaridis (2016)'s findings about the interaction between Caucasus and Iran also fit in there, just like the fact that the Yamna people seem to have been a mix of Caucasus/Iran and EGH; which language did those EGH people bring? and what language was brought by the Caucasian/Iranians? What influences are there from Mesopotamia, Anatolia and Semitic languages? As I said, not a 'PPIE-homeland', but a melting-pot of PPIE influences, which gave us PIE. Joshua Jonathan -Let's talk! 04:14, 18 July 2016 (UTC)

The possibility that Hittite is a cousin not a daughter of PIE is the Indo-Hittite proposal. I think it is very important that we stick to sources, and preferably only mainstream reviews, in determining the relaitons and relative standings between different hypothesis. This is why I find it particularly problematic to focus so much on genetic literature which is often not in actual dialogue or accord with the different linguistic proposals. I don't think we should add genetics here unless we have sources specifically about homeland hypotheses that discuss them. When we do we are very likely to stray into problematic synthesis.·maunus · snunɐɯ· 07:15, 18 July 2016 (UTC)

I removed all of the recently added External Links, none of which were really appropriate per

"none of these enjoy a wide acceptance, or are considered to be fringe theories" ... the "or" is a weak choice. I am not familiar with is subject matter. Would "and" or "nor" be the appropriate synonym? — Preceding unsigned comment added by 98.169.215.11 (talk) 05:12, 5 August 2016 (UTC)

I think it means that none now have wide support, while some are fringe... AnonMoos (talk) 02:13, 6 August 2016 (UTC)

Substituting either would fundamentally change the meaning. What's wrong with "or"? Joe Roe (talk) 09:14, 6 August 2016 (UTC)

I removed some material cited to Quiles' and Lopez-Menchero's Grammar of Modern Indo-European. Neither of the two authors are linguists (Quiles is an economist and Lopez-Menchero and engineer) and the book is selfpublished (the indoeueopean associatoin is an advocacy group attempting to revive PIE as a spoken language). We should not use their work as a source here.·maunus · snunɐɯ· 07:34, 18 July 2016 (UTC)

@Maunus: thanks; I hadn't noticed that they are self-published. Joshua Jonathan -Let's talk! 07:35, 18 July 2016 (UTC)

NB: what about Kortlandt? Joshua Jonathan -Let's talk! 07:40, 18 July 2016 (UTC)

Kortlandt is a respected linguist at the University of Leiden. Some of his proposals are non-mainstream, but his published articles are RS.·maunus · snunɐɯ· 07:45, 18 July 2016 (UTC)

So, RS can be found on this. Joshua Jonathan -Let's talk! 07:55, 18 July 2016 (UTC)

Of course, I have already mentioned a number of important secondary sources that you ought to consult on these issues to make sure that the summary is in line with the way it is described by specialists in the field, Beekes, Clackson and Anthony - Anthony and Ringe's Annual Review paper, and Mallory's introduction type works (the encyclopedia).·maunus · snunɐɯ· 09:02, 18 July 2016 (UTC)

• I also dont think Bomhard is a reliable source. The stuff on his academia.edu profile is also selfpublished. And he is apparently a nostraticist which is a fringe POV - especially in relation to pre-PIE.·maunus · snunɐɯ· 09:05, 18 July 2016 (UTC)

He's perfectly academic (see Allan R. Bomhard), but of course doesn't represent the academic quasi-consensus... AnonMoos (talk) 02:50, 6 August 2016 (UTC)

I don't know what that is supposed to mean. An academic is someone who has an academic appointment and publishes in peer reviewed journals. He does neither. Someone who has a PhD but no academic appointment and who self-publishes on the internet about fringe theories that attracts only negative comments from professional academics is called a "CR ackpot." maunus

I've removed him. Joshua Jonathan -Let's talk! 06:10, 10 August 2016 (UTC)

More or less the same topic, with the same info. Joshua Jonathan -Let's talk! 15:05, 15 July 2016 (UTC)

• Support, per above. Joshua Jonathan -Let's talk! 15:05, 15 July 2016 (UTC)

• Support. There is perhaps room for a separate Proto-Indo-Europeans article that brings together info on the language, society, culture, genetics, etc. of the people who spoke PIE, rather than just where they lived, but at the moment that article just duplicates this one. Better to simplify things and cut down on the redundancy in this topic area. Joe Roe (talk) 15:16, 15 July 2016 (UTC)

• Alternatively, after moving the current content of
  Proto-Indo-European and make that a disambig page pointing to Proto-Indo-European language, Proto-Indo-European homeland, Proto-Indo-European mythology, etc. Joe Roe (talk
  ) 15:19, 15 July 2016 (UTC)

• Oppose. The people, their culture, their language, their religion, and their homeland are all notable topics with enough literature about them to merit their own articles. Overlap happens, and should. The "Proto-Indo-Europeans" article must discuss their culture, as it already does somewhat. And, as is already the case, it must have a section about the homeland theories. A section about the language and another about religion should be added. The people themselves are the main topic, and those sections should be introductory vehicles, pointing readers to the more in-depth articles about main topics related to these people. Genetics can be brought up in some or all of those articles, I suppose; but what more natural place for a discussion of genetics is there than the article about the people themselves? Subsuming the people's article under one about the homeland seems all wrong to me. SamEV (talk) 17:05, 15 July 2016 (UTC)

• I disagree that the people are the main topic. They would be if the Proto-Indo-Europeans were actually a real people, but they aren't – they're a hypothetical construct. The main article is and should be
  Proto-Indo-European into a dab page. Joe Roe (talk
  ) 21:20, 15 July 2016 (UTC)

• Oppose Per SamEV. The literature on the homeland hypotheses and on the people are separate.·maunus · snunɐɯ· 18:02, 15 July 2016 (UTC)

• Oppose. If these articles and subjects were shorter and simpler, then I could see a merge. The PIE homeland would be a paragraph or two. But all of this is very involved and has been discussed and researched by so many people, and the simple discussion of the PIE people is very dense and extensive. I don't see any merit in a merge. Kortoso (talk) 22:41, 15 July 2016 (UTC)

• Support per
  Indigenous Aryans arguments (being considered for deletion), and Proto-Indo-European society --Cornellier (talk
  ) 22:27, 23 July 2016 (UTC)

• Oppose This is about a peoples, not their homeland. This would be like proposing merging Pakistanis into Pakistan. Doesn't work.--NadirAli نادر علی (talk) 20:37, 9 August 2016 (UTC)

• Oppose There is more to the peoples than just the homeland; for example, archaeological cultures associated with the people. However, I would support a merge of
  WP:CONTENTFORK right there. 128.84.125.154 (talk
  ) 20:40, 26 August 2016 (UTC)

• Reduce
  WP:REFBLOAT
• Sometimes refs are just given in bare text like "according to According to Lazaridis et al." then what? the user is supposed to scroll down into the sources to find out to what this refers? There's about a couple of dozen of these and they should be properly referenced.
• Too much use of
  WP:PRIMARY
  e.g. a ref to "The Eurasian Heartland: A continental perspective on Y-chromosome diversity, Proceedings of the National Academy of Sciences of the USA".
• There is a mix of references and sources. There are sources listed that are not in the references and vice versa. "The Horse The Wheel And Language" is mentioned in three different places. I don't think this helps the article.
• Explanatory notes. If the information is too long and awkward to read in the article body then it should be moved to another page. This page is a high-level introduction to the topic, since most of the content is covered in more detail on other pages like Kurgan hypothesis, Historical linguistics, Proto-Indo-European language, Proto-Indo-Europeans, Indo-European migrations, etc. For example, the discussion of the nitty-gritty of pre-proto-IE is out of scope for this page, and the info should be at Indo-Uralic languages.
• There are references that contain notes and notes that contain references, it's just a convoluted way to present the informaiton.

--Cornellier (talk) 15:30, 11 August 2016 (UTC)

I strongly agree with your third point especially. While the article has been greatly improved recently (thanks mainly to

synthesizing individual findings from different fields, so we need to wait for that to happen in the secondary literature. And as I've said before, we have plenty of secondary source material in the form of books and review papers to work with. Joe Roe (talk

) 15:43, 11 August 2016 (UTC)

Hi there. Some responses:

• refbloat: I've seen much worse, due to the strong opinions of some editors who believe that the Indo-European languages did not originate at the steppes. it may be use usefull not to remove all of them, but to keep strong 'proves' of some statements which may be offensive to some editors.
• "According to": All the sentences with "According to" have references at the end of the line, as is the custom at Wikipedia. I've deliberately used the phrase "according to," c.q. attributed those specific bits of info, to make clear that these are the conclusions/opinions of specific researchers.
• "Primary sources": there's a lot of relevant research going on in the field of genetics, with highly relevant results on the Indo-European migrations. Authors of these articles include David Anthony, who explicitly uses them to underscore his archaeological research. No way you can ignore him. Another relevant researcher is David Reich; also no way you can ignore him. I also object against the use of the term "primary sources," since it suggests a lack of distance; they are plain
  WP:RS
  , published in "Peer-reviewed journals."

Anyway, I've used them because I've read a lot of them lately, due to the repeated discussions on the supposedly 'debunking' of the socalled "Aryan Invasion Theory," and the misuse of genetic research in that context. This also shows the relevance of this kind of publications.

• "mix of references and sources": that's indeed the case. I've copied sources I've used elsewhere, and I've shortened the references for the lead and the first alineas. This makes for a much easier way of editing, where one can focus ont he text, instead of having to find out again and again what is text, and what is reference. Also, named references like ":1" are not exacrtly helpfull... Most helpfull wpuold be to convert all refernces to sfn, since this is a very clear and simple way to refer to the same source at several places.

The Horse, The Wheel and Language is listed at two places: the sources section, and the further reading section. That's not a "mix of references and sources," but the difference between sources and recommended reading.

Joshua Jonathan -Let's talk! 16:40, 11 August 2016 (UTC)

Any work that is used as a source should not be in the further reading section.·maunus · snunɐɯ· 18:48, 11 August 2016 (UTC)

Per

WP:PSTS, research papers are primary sources, and should be used sparingly and carefully ("Do not analyze, evaluate, interpret, or synthesize material found in a primary source yourself; instead, refer to reliable secondary sources that do so. Do not base an entire article on primary sources, and be cautious about basing large passages on them."). It doesn't matter that they're peer reviewed or authored by trustworthy researchers, because the concern is not that they're not reliable; it's that we aren't in the business of interpreting or synthesizing them ourselves without reference to secondary sources. That's a core policy, not really up for debate. Joe Roe (talk

) 16:52, 11 August 2016 (UTC)

It's a serious problem with virtually any article about genetics, which are full of cherry picked data from sources, including prominent amateur sources. Doug Weller talk 17:59, 11 August 2016 (UTC)

Don't emit the preceding sentences:

"Unless restricted by another policy, primary sources that have been reputably published may be used in Wikipedia, but only with care, because it is easy to misuse them.[4] Any interpretation of primary source material requires a reliable secondary source for that interpretation. A primary source may only be used on Wikipedia to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the primary source but without further, specialized knowledge."

Using "primary" sources is not the same as interpreting them. And note "reputably"; Haak et al. (2015) for example was published in Nature. Keyser et al. (2009) may be shortened, though; seems less relevant here. And the "root populations" could also be shortened. Joshua Jonathan -Let's talk! 18:30, 11 August 2016 (UTC)

I think it generally is the same. Secondary sources are always preferable, and it is not generally possible to assess whether primary sources are apt for inclusion and how untill they have been referred to in secodary sources. ·maunus · snunɐɯ· 18:46, 11 August 2016 (UTC)

So to summarise (please just comment if I got it wrong),

• WP:REFBLOAT
  : remove some but not all
• Ref style. JJ likes it. A small minority of articles use it. I think it's ugly, and less usable, especially on a phone. Why mention it in two places? And I don't think ease of editing trumps ease of reading. Also JJ, agree that named references like ":1" but I always give those human-readable names like name="horse-wheel-lang". But anyway I've got other fish to fry....
• WP:PRIMARY
  : I think we all agree there's work to do.
• Explanatory notes: I'd like to see 'em gone, but it's better than it was, thanks for that. --Cornellier (talk) 23:09, 13 August 2016 (UTC)

Two places, because of "Further reading": when a lot of sources are being used, the most relevant can be recommanded in a "Further reading" section. Joshua Jonathan -Let's talk! 05:22, 14 August 2016 (UTC)

Sorry I wasn't clear; let me clarify. With conventional referencing you mouse on the ref and you see immediately what the ref points to. References with names is how it's done in most articles, and how it's implemented in the editing tools. With sfn-style in this article:

1. Mouse on the ref, see "Jones 2015", which is unhelpful.
2. Click on the ref number, go to the refs section and see "Jones 2015" again.
3. Click on "Jones 2015", and finally see the actual full referenced text. (At least that's what's supposed to happen, currently Jones 2015 doesn't link to anything, which is another downside of the system, it's harder to QA)

It's just not as good usability, especially on a phone.--Cornellier (talk) 17:29, 14 August 2016 (UTC)

@Cornellier: ah, yes that's a good point. Though my computer gives me one step less; from the reference it goes straight to the source.

WP:PRIMARY

As an example of why we shouldn't jump on new genetics papers too quickly, here's a recent paper critical of Haak et al. etc. Joe Roe (talk) 13:50, 27 August 2016 (UTC)

I can't access the publication, so how do you assess the reliability of that particular publication? Anyway, if it's relevant, you should add the info. Joshua Jonathan -Let's talk! 14:55, 27 August 2016 (UTC)

Better to remove Haak et al, untill its status is established.·maunus · snunɐɯ· 18:48, 27 August 2016 (UTC)

No. Definitely

WP:RS; 134citations in just one year. At best, you can the opinions from that paper, if they are relevant. Joshua Jonathan -Let's talk!

21:01, 27 August 2016 (UTC)

After a night's sleep, my brain functions less categorically; we might as well remove all three specificgeneticresearches from this specific Wiki-article,butleave the Pereltsvaig quote. Joshua Jonathan -Let's talk! 04:53, 28 August 2016 (UTC)

It is not a question of whether it is an RS, it is a question of the amount of weight it should be assigned. Not all reliable sources merit inclusion - this is q eustion of editorial decisions weighing sources against eachother by using secondary and tertiary sources to establish relative weight. For any primary source this is determined by the amount and kind of attention it gets in secondary sources. If high quality secondary sources on a given topic ignores the primary source or gives it only negative commentary it probably does not need weight. This source is so new that the most relevant secondary sources have not included it yet, and if specific commentary by peers is critical that suggests including it is premature. Number of citations is not a relevant guide for either reliability or weight - because weight is topic dependent. ·maunus · snunɐɯ· 09:33, 28 August 2016 (UTC)

I still don't agree. The relevant genetic research is in journals; by ignoring this kind of publications, you're missing relevant info. NB: please show the policy which says "editorial decisions weighing sources against eachother by using secondary and tertiary sources to establish relative weight." In addition: David Anthony has commented on genetic findings; I don't expect you to argue that Anthony is a voice to ignore in this regard. Joshua Jonathan -Let's talk! 20:51, 28 August 2016 (UTC)

Policy is pretty clear on this - try to read

WP:PRIMARY. The weighing of different primary sources and viewpoints against eachother is EXACTLY why we are cautioned to use mostly secondary sources. That is the only we it is possible to write balanced articles that reflect neutrally the relevant body of literature. This is a basic editing principle. Anthony commenting genetics is a secondary source. I am not saying that we ignore genetic findings - but that we only include those that have received commentary from other scholars and has had their relevance relative to the homeland debate assessed. IT is really very basic stuff.·maunus · snunɐɯ·

05:40, 29 August 2016 (UTC)

"I am not saying that we ignore genetic findings" - thanks; that's helpful. I'll read the links. Best regards, Joshua Jonathan -Let's talk! 06:42, 29 August 2016 (UTC)

From

WP:WEIGHT

:

'"Neutrality requires that each article or other page in the mainspace fairly represent all significant viewpoints that have been published by reliable sources, in proportion to the prominence of each viewpoint in the published, reliable sources."

• The Steppe-theory is signifcant viewpoint, published in reliable sources;
• Haak's research elaborates on/refers to this theory, offering significant support for this theory.

Joshua Jonathan -Let's talk! 06:50, 29 August 2016 (UTC)

You are simplifying the issue way too much. Haak et al is a primary source for their own view point - which is that the genetic sequencing of 69 ancient skeleton supports the steppe theory. This particular viewpoint should be given weight according to how many other researchers have accepted it in secondary sources. A good deal of researchers have, so it should definitely have weight - that is not at issue. Combined with the fact that Allentoft et al corroborates their conclusion with 101 other ancient genomes, the evidence is weighty and there are many good secondary sources. The issue here is that the section should be written on the basis of the secondary sources, not on the basis of the primary sources - so we see how secondary sources summarize the two primary sources and then we summarize the different secondary sources (wikipedia is a tertiary source).·maunus · snunɐɯ· 06:57, 29 August 2016 (UTC)

Removal of genetics sections

I have removed the two genetics sections both of which relied almost entitrely on Haak et al and Lazaridis et al. We need to find some good secondary sources that discuss the status of genetic evidence for the homeland question. The discussion by the way is quite a bit older, and should include both the early blood type based studies, the later studies by Cavalli-Sforza and more recent studies based on genomic statistical methods. Lets have a discussion of which sources should be used and how they should be weighted.·maunus · snunɐɯ· 12:12, 28 August 2016 (UTC)

I was in the process of rewriting it but got a conflict with your edit, so I'll put my revise text here in case it's useful later:

See also:

Origins of Yamna culture and Yamna component in European genes

According to genetic studies, individuals from the Yamnaya culture have a mix from eastern European hunter-gatherer[1] and Caucasus hunter-gatherer^[2] ancestry. Iran Chalcolithic people with a Caucasian hunter-gatherer component.^{[3][3][note 1][clarification needed]}

Many geneticists consider Haplogroup R1a to be associated with the origins and spread of the Indo-Europeans.^[4][5][6] R1a1 shows a strong correlation with the distribution of the Indo-European languages in Europe and south Asia, being most prevalent in Poland, Russia, and Ukraine, but also observed in Pakistan, India and central Asia.^[5][6] Two specific subclades dominate, namely R1-Z282 in Eastern-Europe, and R1-Z93 which is found in south-Siberia and northern India.^[5][6] According to Underhill et al. (2014), the initial diversification of R1a took place in the vicinity of Iran,^[7] while Pamjav et al. (2012) think that R1a diversified within the Eurasian Steppes or the Middle East and Caucasus region.^[8]

In 2015, a large-scale ancient DNA study published in Nature^[9] found evidence of a "massive migration" from the Pontic-Caspian steppe to Central Europe that took place about 4,500 years ago. It found that individuals from the Central European Corded Ware culture (3rd millennium BCE) were genetically closely related to individuals from the Yamnaya culture. The authors concluded that this was evidence for a steppe origin of the Indo-European languages.^[10] However, archaeologists have argued that although such a migration might have taken place, it does not necessarily explain the either the distribution of archaeological cultures or the spread of the Indo-European languages.^[11]