Happy new year 2019…and enjoy our new books!


Sorry for the last weeks of silence, I have been rather busy lately. I am having more projects going on, and (because of that) I also wanted to finish a project I have been working on for many months already.

I have therefore decided to publish a provisional version of the text, in the hope that it will be useful in the following months, when I won’t be able to update it as often as I would like to:

EDIT (20 JAN 2019): For those of you who are more comfortable reading in your native language, I have placed some links to automatic translations by Google Translate. They might work especially well for the texts of A Game of Clans & A Clash of Chiefs.

Don’t forget to check out the maps included in the supplementary materials: I have added Y-DNA, mtDNA, and ADMIXTURE data using GIS software. The PCA graphics are also important to follow the main text.

NOTE. Right now the files are only in my server. I will try to upload them to Academia.edu and Research Gate when I have time, I have uploaded them to Academia.edu and ResearchGate, in case the websites are too slow.

I would have preferred to wait for a thorough revision of the section on archaeology and the linguistic sections on Uralic, but I doubt I will have time when the reviews come, so it was either now or maybe next December…

I say so in the introduction, but it is evident that certain aspects of the book are tentative to say the least: the farther back we go from Late Proto-Indo-European, the less clear are many aspects. Also, linguistically I am not convinced about Eurasiatic or Nostratic, although they do have a certain interest when we try to offer a comprehensive view of the past, including ethnolinguistic identities.

I cannot be an expert in everything, and these books cover a lot. I am bound to publish many corrections as new information appears and more reviews are sent. For example, just days ago (before SNP calls of Wang et al. 2018 were published) some paragraphs implied that AME might have expanded Nostratic from the Middle East. Now it does not seem so, and I changed them just before uploading the text. That’s how tentative certain routes are, and how much all of this may change. And that only if we accept a Nostratic phylum…

NOTE. Since the first book I wrote was the linguistic one, and I have spent the last months updating the archaeology + genetics part, now many of you will probably understand 1) why I am so convinced about certain language relationships and 2) how I used many posts to clarify certain ideas and receive comments. Many posts offer probably a good timeline of what I worked with, and when.


I did not add this section to the books, because they are still not ready for print, but I think this is due somewhere now. It is impossible to reference all who have directly or indirectly contributed to this, so this is a list of those I feel have played an important role.

I am indebted to the following people (which does not mean that they share my views, obviously):

First and foremost, to Fernando López-Menchero, for having the patience to review with detail many parts on Indo-European linguistics, knowing that I won’t accept many of his comments anyway. The additional information he offers is invaluable, but I didn’t want to turn this into a huge linguistic encyclopaedia with unending discussions of tiny details of each reconstructed word. I think it is already too big as it is.

I would not have thought about doing this if it were not for the interest of Wekwos (Xavier Delamarre) in publishing a full book about the Indo-European demic diffusion model (in the second half of 2017, I think). It was them who suggested that I extended the content, when all I had done until then was write an essay and draw some maps in my free time between depositing the PhD thesis and defending it.

Sadly, as much as I would like to publish a book with a professional publisher, I don’t think ancient DNA lends itself for the traditional format, so my requests (mainly to have free licenses and being able to review the text at will, as new genetic papers are published) were logically not acceptable. Also, the main aim of all volumes, especially the linguistic one, is the teaching of essentials of Late Proto-Indo-European and related languages, and this objective would be thwarted by selling each volume for $50-70 and only in printed format. I prefer a wider distribution.

At first I didn’t think much of this proposal, because I do not benefit from this kind of publications in my scientific field, but with time my interest in writing a whole, comprehensive book on the subject grew to the point where it was already an ongoing project, probably by the start of 2018.

I would not have been in contact with Wekwos if it were not for user Camulogène Rix at Anthrogenica, so thanks for that and for the interest in this work.

I would not have thought of writing this either if not for the spontaneous support (with an unexpected phone call!) of a professor of the Complutense University of Madrid, Ángel Gómez Moreno, who is interested in this subject – as is his wife, a professor of Classics more closely associated to Indo-European studies, and who helped me with a search for Indo-Europeanists.

EDIT (1 JAN 2019): I remembered that Karin Bojs sent me her book after reading the demic diffusion model. I may have also thought about writing a whole book back then, but mid-2017 is probably too early for the project.

Professor Kortlandt is still to review the text, but he contributed to both previous essays in some very interesting ways, so I hope he can help me improve the parts on Uralic, and maybe alternative accounts of expansion for Balto-Slavic, depending on the time depth that he would consider warranted according to the Temematic hypothesis.

The maps are evidently (for those who are interested in genetics) in part the result of the effort of the late Jean Manco: As you can see from the maps including Y-DNA and mtDNA samples, I have benefitted from her way of organising data and publishing it. Similarly, the work of Iain McDonald in assessing the potential migration routes of R1b and R1a in Europe with the help of detailed maps was behind my idea for the first maps, and consequently behind these, too.

I should thank all people responsible for the release of free datasets to work with, including the Reich and Jena labs, the Veeramah Lab, and also researchers from the Max Planck Institute or the Mainz Palaeogenetics group, who didn’t mind to share with me datasets to work with.

Readers of this blog with interesting comments have also been essential for the improvement of the texts. You can probably see some of your many contributions there. I may not answer many comments, because I am always busy (and sometimes I just don’t have anything interesting to say), but I try to read all of them.

EDIT (1 JAN 2019) I think I should mention at least Chetan, Egg, or Robert George; but then I would leave out old europe, Sgr Ganesh, or Tileman Ehlen; and if I include them I would leave out others…

Users of other sites, like Anthrogenica, whose particular points of view and deep knowledge of some very specific aspects are sometimes very useful. In particular, user Anglesqueville helped me to fix some issues with the merging of datasets to obtain the PCAs and ADMIXTURE, and prepared some individual samples to merge them.

Even without posting anything, Google Analytics keeps sending me messages about increasing user fidelity (returning users), and stats haven’t really changed (which probably means more people are reading old posts), so thank you for that.

I hope you enjoy the books.

Happy new year!

Sahara’s rather pale-green and discontinuous Sahelo-Sudanian steppe corridor, and the R1b – Afroasiatic connection


Interesting new paper (behind paywall) Megalakes in the Sahara? A Review, by Quade et al. (2018).

Abstract (emphasis mine):

The Sahara was wetter and greener during multiple interglacial periods of the Quaternary, when some have suggested it featured very large (mega) lakes, ranging in surface area from 30,000 to 350,000 km2. In this paper, we review the physical and biological evidence for these large lakes, especially during the African Humid Period (AHP) 11–5 ka. Megalake systems from around the world provide a checklist of diagnostic features, such as multiple well-defined shoreline benches, wave-rounded beach gravels where coarse material is present, landscape smoothing by lacustrine sediment, large-scale deltaic deposits, and in places, tufas encrusting shorelines. Our survey reveals no clear evidence of these features in the Sahara, except in the Chad basin. Hydrologic modeling of the proposed megalakes requires mean annual rainfall ≥1.2 m/yr and a northward displacement of tropical rainfall belts by ≥1000 km. Such a profound displacement is not supported by other paleo-climate proxies and comprehensive climate models, challenging the existence of megalakes in the Sahara. Rather than megalakes, isolated wetlands and small lakes are more consistent with the Sahelo-Sudanian paleoenvironment that prevailed in the Sahara during the AHP. A pale-green and discontinuously wet Sahara is the likelier context for human migrations out of Africa during the late Quaternary.

The whole review is an interesting read, but here are some relevant excerpts:

Various researchers have suggested that megalakes coevally covered portions of the Sahara during the AHP and previous periods, such as paleolakes Chad, Darfur, Fezzan, Ahnet-Mouydir, and Chotts (Fig. 2, Table 2). These proposed paleolakes range in size by an order of magnitude in surface area from the Caspian Sea–scale paleo-Lake Chad at 350,000 km2 to Lake Chotts at 30,000 km2. At their maximum, megalakes would have covered ~ 10% of the central and western Sahara, similar to the coverage by megalakes Victoria, Malawi, and Tanganyika in the equatorial tropics of the African Rift today. This observation alone should raise questions of the existence of megalakes in the Sahara, and especially if they developed coevally. Megalakes, because of their significant depth and area, generate large waves that become powerful modifiers of the land surface and leave conspicuous and extensive traces in the geologic record.

ETOPO1 digital elevation model (1 arc-minute; Amante and Eakins, 2009) of proposed megalakes in the Sahara Desert during the late Quaternary. Colors denote Köppen-Geiger climate zones: blue, Aw, Af, Am (tropical); light tan, Bwk, BSh, BSk, Csa, Csb, Cwb, Cfa, Cfb (temperate); red-brown, Bwh (arid, hot desert and steppe climate). Lake area at proposed megalake high stands and present Lake Victoria are in blue, and contributing catchment areas are shown as thin solid black lines. The main tributaries of Lake Chad are denoted by blue lines (from west to east: the Komadougou-Yobe, Logone, and Chari Rivers; source: Global Runoff Data Center, Koblenz, Germany). Rainfall isohyets (50, 200, 800, 1200, and 1600) are marked in dashed gray-scale lines. Physical parameters of each basin are shown in white boxes: Abt, total basin area; AW, lake area; Vw, lake volume; and aW= AW/Abt. Black dots mark the location of the paleohydrological records from Lezine et al. (2011), also compiled in Supplementary Table S5.

Lakes, megalakes, and wetlands

Active ground-water discharge systems abound in the Sahara today, although they were much more widespread in the AHP. They range from isolated springs and wet ground in many oases scattered across the Sahara (e.g., Haynes et al., 1989) to wetlands and small lakes (Kröpelin et al., 2008). Ground water feeding these systems is dominated by fossil AHP-age and older water (e.g., Edmunds and Wright 1979; Sonntag et al., 1980), although recently recharged water (<50 yr) has been locally identified in Saharan ground water (e.g., Sultan et al., 2000; Maduapuchi et al., 2006).

Megalake Chad

In our view, Lake Chad is the only former megalake in the Sahara firmly documented by sedimentologic and geomorphic evidence. Mega-Lake Chad is thought to have covered ~ 345,000 km2, stretching for nearly 8° (10–18°N) of latitude (Ghienne et al., 2002) (Fig. 2). The presence of paleo- Lake Chad was at one point challenged, but several—and in our view very robust—lines of evidence have been presented to support its development during the AHP. These include: (1) clear paleo-shorelines at various elevations, visible on the ground (Abafoni et al., 2014) and in radar and satellite images (Schuster et al., 2005; Drake and Bristow, 2006; Bouchette et al., 2010); (2) sand spits and shoreline berms (Thiemeyer, 2000; Abafoni et al., 2014); and (3) evaporites and aquatic fauna such as fresh-water mollusks and diatoms in basin deposits (e.g., Servant, 1973; Servant and Servant, 1983). Age determinations for all but the Holocene history of mega- Lake Chad are sparse, but there is evidence for Mio-Pliocene lake (s) (Lebatard et al., 2010) and major expansion of paleo- Lake Chad during the AHP (LeBlanc et al., 2006; Schuster et al., 2005; Abafoni et al., 2014; summarized in Armitage et al., 2015) up to the basin overflow level at ~ 329m asl.

Insights from hydrologic mass balance of megalakes

Graph of mean annual rainfall (mm/yr) versus aw (area lake/area basin, AW/AL); their modeled relationship using our Sahelo-Sudanian hydrologic model for the different lake basins are shown as solid colored lines. Superimposed on this (dashed lines) are the aw values for individual megalake basins and the mean annual rainfall required to sustain them. Mean annual paleo-rainfall estimates of 200– 400 mm/yr during the AHP from fossil pollen and mollusk evidence is shown as a tan box. The intersection of this box with the solid colored lines describes the resulting aw for Saharan paleolakes on the y-axis. The low predicted values for aw suggest that very large lakes would not form under Sahelo-Sudanian conditions where sustained by purely local rainfall and runoff. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Using these conservative conditions (i.e., erring in the direction that will support megalake formation), our hydrologic models for the two biggest central Saharan megalakes (Darfur and Fezzan) require minimum annual average rainfall amounts of ~ 1.1 m/yr to balance moisture losses from their respective basins (Supplementary Table S1). Lake Chad required a similar amount (~1 m/yr; Supplementary Table S1) during the AHP according to our calculations, but this is plausible, because even today the southern third of the Chad basin receives ≥1.2 m/yr (Fig. 2) and experiences a climate similar to Lake Victoria. A modest 5° shift in the rainfall belt would bring this moist zone northward to cover a much larger portion of the Chad basin, which spans N13° ±7°. Estimated rainfall rates for Darfur and Fezzan are slightly less than the average of ~ 1.3 m/yr for the Lake Victoria basin, because of the lower aw values, that is, smaller areas of Saharan megalakes compared with their respective drainage basins (Fig. 15).

Estimates of paleo-rainfall during the AHP

Here major contradictions develop between the model outcomes and paleo-vegetation evidence, because our Sahelo-Sudanian hydrologic model predicts wetter conditions and therefore more tropical vegetation assemblages than found around Lake Victoria today. In fact, none of the very wet rainfall scenarios required by all our model runs can be reconciled with the relatively dry conditions implied by the fossil plant and animal evidence. In short, megalakes cannot be produced in Sahelo-Sudanian conditions past or present; to form, they require a tropical or subtropical setting, and major displacements of the African monsoon or extra-desert moisture sources.

Change in mean annual precipitation over northern Africa between mid-Holocene (6 ka) and pre-industrial conditions in PMIP3 models (affiliations are provided in Supplementary Table S4). Lakes Victoria and Chad outlined in blue. (a) Ensemble mean change in mean annual precipitation and positions of the African summer (July–September) ensemble mean ITCZ during mid-Holocene (solid red line) and pre-industrial conditions (solid blue line). (b) Zonal average of change in mean annual precipitation over land (20°W–30°E) for the ensemble mean (thick black) and individual models are listed on right). The range of minimal estimated change in mean annual precipitation required to sustain steppe is shown in shaded green (Jolly et al., 1998).


If not megalakes, what size lakes, marshes, discharging springs, and flowing rivers in the Sahara were sustainable in Sahelo-Sudanian climatic conditions? For lakes and perennial rivers to be created and sustained, net rainfall in the basin has to exceed loss to evapotranspiration, evaporation, and infiltration, yielding runoff that then supplies a local lake or river. Our hydrologic models (see Supplementary Material) and empirical observations (Gash et al., 1991; Monteith, 1991) for the Sahel suggest that this limit is in the 200–300 mm/yr range, meaning that most of the Sahara during the AHP was probably too dry to support very large lakes or perennial rivers by means of local runoff. This does not preclude creation of local wetlands supplied by ground-water recharge focused from a very large recharge area or forced to the surface by hydrologic barriers such as faults, nor megalakes like Chad supplied by moisture from the subtropics and tropics outside the Sahel. But it does raise a key question concerning the size of paleolakes, if not megalakes, in the Sahara during the AHP. Our analysis suggests that Sahelo-Sudanian climate could perhaps support a paleolake approximately ≤5000 km2 in area in the Darfur basin and ≤10,000–20,000 km2 in the Fezzan basin. These are more than an order of magnitude smaller than the megalakes envisioned for these basins, but they are still sizable, and if enclosed in a single body of water, should have been large enough to generate clear shorelines (Enzel et al., 2015, 2017). On the other hand, if surface water was dispersed across a series of shallow and extensive but partly disconnected wetlands, as also implied by previous research (e.g., Pachur and Hoelzmann, 1991), then shorelines may not have developed.

One of the underdeveloped ideas of my Indo-European demic diffusion model was that R1b-V88 had migrated through South Italy to Northern Africa, and from it using the Sahara Green Corridor to the south, from where the “upside-down” view of Bender (2007) could have occurred, i.e. Afroasiatic expanding westwards within the Green Sahara, precisely at this time, and from a homeland near the Megalake Chad region (see here).

Whether or not R1b-V88 brought the ‘original’ lineage that expanded Afroasiatic languages may be contended, but after D’Atanasio et al. (2018) it seems that only two lineages, E-M2 and R1b-V88, fit the ‘star-like’ structure suggesting an appropriate haplogroup expansion and necessary regional distribution that could explain the spread of Afroasiatic languages within a reasonable time frame.

Palaeolithic migrations

This review shows that the hypothesized Green Sahara corridor full of megalakes that some proposed had fully connected Africa from west to east was actually a strip of Sahelo-Sudanian steppe spread to the north of its current distribution, including the Chad megalake, East Africa and Arabia, apart from other discontinuous local wetlands further to the north in Africa. This greenish belt would have probably allowed for the initial spread of early Afroasiatic proto-languages only through the southern part of the current Sahara Desert. This and the R1b-V88 haplogroup distribution in Central and North Africa (with a prevalence among Chadic speakers probably due to later bottlenecks), and the Near East, leaves still fewer possibilities for an expansion of Afroasiatic from anywhere else.

If my proposal turns out to be correct, this Afroasiatic-like language would be the one suggested by some in the vocabulary of Old European and North European local groups (viz. Kroonen for the Agricultural Substrate Hypothesis), and not Anatolian farmer ancestry or haplogroup G2, which would have been rather confined to Southern Europe, mainly south of the Loess line, where incoming Middle East farmers encountered the main difficulties spreading agriculture and herding, and where they eventually admixed with local hunter-gatherers.

NOTE. If related to attested languages before the Roman expansion, Tyrsenian would be a good candidate for a descendant of the language of Anatolian farmers, given the more recent expansion of Anatolian ancestry to the Tuscan region (even if already influenced by Iran farmer ancestry), which reinforces its direct connection to the Aegean.

The fiercest opposition to this R1b-V88 – Afroasiatic connection may come from:

  • Traditional Hamito-Semitic scholars, who try to look for any parent language almost invariably in or around the Near East – the typical “here it was first attested, ergo here must be the origin, too”-assumption (coupled with the cradle of civilization memes) akin to the original reasons behind Anatolian or Out-of-India hypotheses; and of course
  • autochthonous continuity theories based on modern subclades, of (mainly Semitic) peoples of haplogroup E or J, who will root for either one or the other as the Afroasiatic source no matter what. As we have seen with the R1a – Indo-European hypothesis (see here for its history), this is never the right way to look at prehistoric migrations, though.

I proposed that it was R1a-M417 the lineage marking an expansion of Indo-Uralic from the east near Lake Baikal, then obviously connected to Yukaghir and Altaic languages marked by R1a-M17, and that haplogroup R could then be the source of a hypothetic Nostratic expansion (where R2 could mark the Dravidian expansion), with upper clades being maybe responsible for Borean.

Simple Nostratic tree by Bomhard (2008)

However, recent studies have shown early expansions of R1b-297 to East Europe (Mathieson et al. 2017 & 2018), and of R1b-M73 to East Eurasia probably up to Siberia, and possibly reaching the Pacific (Jeong et al. 2018). Also, the Steppe Eneolithic and Caucasus Eneolithic clusters seen in Wang et al. (2018) would be able to explain the WHG – EHG – ANE ancestry cline seen in Mesolithic and Neolithic Eurasia without a need for westward migrations.

Dravidian is now after Narasimhan et al. (2018) and Damgaard et al. (Science 2018) more and more likely to be linked to the expansion of the Indus Valley civilization and haplogroup J, in turn strongly linked to Iranian farmer ancestry, thus giving support to an Elamo-Dravidian group stemming from Iran Neolithic.

NOTE. This Dravidian-IVC and Iran connection has been supported for years by knowledgeable bloggers and commenters alike, see e.g. one of Razib Khan’s posts on the subject. This rather early support for what is obvious today is probably behind the reactionary views by some nationalist Hindus, who probably saw in this a potential reason for a strengthened Indo-Aryan/Dravidian divide adding to the religious patchwork that is modern India.

I am not in a good position to judge Nostratic, and I don’t think Glottochronology, Swadesh lists, or any statistical methods applied to a bunch of words are of any use, here or anywhere. The work of pioneers like Illich-Svitych or Starostin, on the other hand, seem to me solid attempts to obtain a faithful reconstruction, if rather outdated today.

NOTE. I am still struggling to learn more about Uralic and Indo-Uralic; not because it is more difficult than Indo-European, but because – in comparison to PIE comparative grammar – material about them is scarce, and the few available sources are sometimes contradictory. My knowledge of Afroasiatic is limited to Semitic (Arabic and Akkadian), and the field is not much more developed here than for Uralic…

Spread of Y-haplogroup R1b(xM269) in Eurasia, according to Jeong et al. (2018).

If one wanted to support a Nostratic proto-language, though, and not being able to take into account genome-wide autosomal admixture, the only haplogroup right now which can connect the expansion of all its branches is R1b-M343:

  • R1b-L278 expanded from Asia to Europe through the Iranian Plateau, since early subclades are found in Iran and the Caucasus region, thus supporting the separation of Elamo-Dravidian and Kartvelian branches;
  • From the Danube or another European region ‘near’ the Villabruna 1 sample (of haplogroup R1b-L754):
    • R1b-V88 expanding everywhere in Europe, and especially the branch expanding to the south into Africa, may be linked to the initial Afroasiatic expansion through the Pale-Green Sahara corridor (and even a hypothetic expansion with E-M2 subclades and/or from the Middle East would also leave open the influence of V88 and previous R1b subclades from the Middle East in the emergence of the language);
    • R1b-297 subclades expanding to the east may be linked to Eurasiatic, giving rise to both Indo-Uralic (M269) and Macro- or Micro-Altaic (M73) expansions.

This is shameless, simplistic speculation, of course, but not more than the Nostratic hypothesis, and it has the main advantage of offering ‘small and late’ language expansions relative to other proposals spanning thousands (or even tens of thousands) of years more of language separation. On the other hand, that would leave Borean out of the question, unless the initial expansion of R1b subclades happened from a community close to lake Baikal (and Mal’ta) that was also at the origin of the other supposedly related Borean branches, whether linked to haplogroup R or to any other…

NOTE. If Afroasiatic and Indo-Uralic (or Eurasiatic) are not genetically related, my previous simplistic model, R1b-Afroasiatic vs. R1a-Eurasiatic, may still be supported, with R1a-M17 potentially marking the latest meaningful westward population expansion from which EHG ancestry might have developed (see here). Without detailed works on Nostratic comparative grammar and dialectalization, and especially without a lot more Palaeolithic and Mesolithic samples, all this will remain highly speculative, like proposals of the 2000s about Y-DNA-haplogroup – language relationships.


Kortlandt: West Indo-Europeans along the Danube, Germanic and Balto-Slavic share a Corded Ware substrate


New paper (behind paywall) The Expansion of the Indo-European Languages, by Frederik Kortlandt, JIES (2018) 46(1 & 2):219-231.


When considering the way the Indo-Europeans took to the west, it is important to realize that mountains, forests and marshlands were prohibitive impediments. Moreover, people need fresh water, all the more so when traveling with horses. The natural way from the Russian steppe to the west is therefore along the northern bank of the river Danube. This leads to the hypothesis that the western Indo-Europeans represent successive waves of migration along the Danube and its tributaries. The Celts evidently followed the Danube all the way to southern Germany. The ancestors of the Italic tribes, including the Veneti, may have followed the river Sava towards northern Italy. The ancestors of Germanic speakers apparently moved into Moravia and Bohemia and followed the Elbe into Saxony. A part of the Veneti may have followed them into Moravia and moved along the Oder through the Moravian Gate into Silesia. The hypothetical speakers of Temematic probably moved through Slovakia along the river Orava into western Galicia. The ancestors of speakers of Balkan languages crossed the lower Danube and moved to the south. This scenario is in agreement with the generally accepted view of the earliest relations between these branches of Indo-European.

The western Indo-European vocabulary in Baltic and Slavic is the result of an Indo-European substratum which contained an older non-Indo-European layer and was part of the Corded Ware horizon. The numbers show that a considerable part of the vocabulary was borrowed after the split between Baltic and Slavic, which came about when their speakers moved westwards north and south of the Pripet marshes. These events are older than the westward movement of the Slavs which brought them into contact with Temematic speakers. One may conjecture that the Venedi occupied the Oder basin and then expanded eastwards over the larger part of present-day Poland before the western Balts came down the river Niemen and moved onwards to the lower Vistula. We may then identify the Venedic expansion with the spread of the Corded Ware horizon and the westward migration of the Balts and the Slavs with their integration into the larger cultural complex. The theory that the Venedi separated from the Veneti in the upper Sava region and moved through Moravia and Silesia to the Baltic Sea explains the “im Namenmaterial auffällige Übereinstimmung zwischen dem Baltikum und den Gebieten um den Nordteil der Adria” (Udolph 1981: 61). The Balts probably moved in two stages because the differences between West and East Baltic are considerable.

Instead of reinterpreting his views in light of the recent genetic finds, Kortlandt tries to mix in this paper his own old theories (see his paper Baltic, Slavic, Germanic) with the recent interpretations of genetic papers, using also dubious secondary sources – e.g. Iversen and Kroonen (2017) or Klejn (2017) [see here, and here] – which, in my opinion, creates a potentially dangerous circular reasoning.

For example, even though he criticizes the general stance of recent genetic papers with regard to Proto-Indo-European dialectalization and expansion as too early, and he supports the Danube expansion route, he nevertheless follows their interpretations in accepting that Corded Ware was Indo-European (following the newest model proposed by Anthony):

The [Yamnaya] penetrated central and northern Europe from the lower Danube through the Carpathian basin, not from the east. The Carpathian basis was evidently the cradle of the Corded Ware cultures, where the descendants of the Yamnaya mixed with the local early farmers before proceeding to the north. The development has a clear parallel in the Middle Ages, when the Hungarians mixed with the local Slavic populations in the same territory (cf. Kushniarevich & al. 2015).

He still follows his good old Indo-Slavonic group in the east, but at the same time maintains Kallio’s view that there were no early Uralic loanwords in Balto-Slavic, and also Kallio’s (and the general) view that there were close contacts with PIE and Pre-Proto-Indo-Iranian…

NOTE. The latest paper on Eurasian migrations by Damgaard et al. (Nature 2018), which shows mainly Proto-Iranians dominating over East Europe after the Early Bronze Age, have left still fewer space for a Proto-Balto-Slavic group emerging from the east.

Also, he asserts the following, which is a rather weird interpretation of events:

It appears that the Corded Ware horizon spread to southern Scandinavia (cf. Iversen & Kroonen 2017) but not to the Baltic region during the Neolithic.

“However, we also find indications of genetic impact from exogenous populations during the Neolithic, most likely from northern Eurasia and the Pontic Steppe. These influences are distinct from the Anatolian-farmer-related gene flow found in Central Europe during this period.”

It follows that the Indo-Europeans did not reach the Baltic region before the Late Neolithic. The influx of non-local people from northern Eurasia may be identified with the expansion of the Finno-Ugrians, who came into contact with the Indo-Europeans as a result of the eastward expansion of the latter in the fourth millennium. This was long before the split between Balto-Slavic and Indo-Iranian.

In the Late Neolithic there was “a further population movement into the regions surrounding the Baltic Sea” that was “accompanied by the first evidence of extensive animal husbandry in the Eastern Baltic”, which “suggests import of the new economy by an incoming steppe-like population independent of the agricultural societies that were already established to the south and west of the Baltic Sea.” (Mittnik & al. 2018). These may have been the ancestors of Balto-Slavic speakers. At a later stage, the Corded Ware horizon spread eastward, giving rise to farming ancestry in Eastern Baltic individuals and to a female gene-flow from the Eastern Baltic into Central Europe (ibidem).

Late Copper Age migrations in Asia ca. 2800-2300 BC.

He is a strong Indo-Uralic supporter, and supports a parallel Indo-European – Uralic development in Eastern Europe, and (as you can read) he misunderstands the description of population movements in the Baltic region, and thus misplaces Finno-Ugric speakers as Eurasian migrants arriving in the Baltic from the east during the Late Neolithic, before the Corded Ware expansion, which is not what the cited papers implied.

NOTE. Such an identification of westward Neolithic migrations with Uralic speakers is furthermore to be rejected following the most recent paper on Fennoscandian samples.

He had previously asserted that the substrate common to Germanic and Balto-Slavic is Indo-European with non-Indo-European substrate influence, so I guess that Corded Ware influencing as a substrate both Germanic and Balto-Slavic is the best way he could put everything together, if one assumes the widespread interpretations of genetic papers:

Thus, I think that the western Indo-European vocabulary in Baltic and Slavic is the result of an Indo-European substratum which contained an older non-Indo-European layer and was part of the Corded Ware horizon. The numbers show that a considerable part of the vocabulary was borrowed after the split between Baltic and Slavic, (…)

NOTE. It is very likely that this paper was sent in late 2017. That’s the main problem with traditional publications including the most recent genetic investigation: by the time something gets eventually published, the text is already outdated.

I obviously share his opinion on precedence of disciplines in Indo-European studies:

The methodological point to be emphasized here is that the linguistic evidence takes precedence over archaeological and genetic data, which give no information about the languages spoken and can only support the linguistic evidence. The relative chronology of developments must be established on the basis of the comparative method and internal reconstruction. The location of a reconstructed language can only be established on the basis of lexical and onomastic material. On the other hand, archaeological or genetic data may supply the corresponding absolute chronology. It is therefore incorrect to attribute cultural influences in southern Scandinavia and the Baltic region in the third millennium to Germanic or Baltic speakers because these languages did not yet exist. While the Italo-Celtic branch may have separated from its Indo-European neighbors in the first half of the third millennium, Proto-Balto-Slavic and Proto-Indo-Iranian can be dated to the second millennium and Proto-Germanic to the end of the first millennium BC (cf. Kortlandt 2010: 173f., 197f., 249f.). The Indo-Europeans who moved to southern Scandinavia as part of the Corded Ware horizon were not the ancestors of Germanic speakers, who lived farther to the south, but belonged to an unknown branch that was eventually replaced by Germanic.

I hope we can see more and more anthropological papers like this, using traditional linguistics coupled with archaeology and the most recent genetic investigations.

EDIT (4 JUL 2018): Some errors corrected.


Pre-Germanic born out of a Proto-Finnic substrate in Scandinavia


A commenter, Old Europe, drew my attention to the Uralic (Finnic-Saamic) substrate in Germanic proposed by Schrijver in Chapter V. Origins of Language Contact and the Origins of the Germanic Languages, Routledge (2014).

I wanted to share here some interesting excerpts (emphasis mine):

NOTE. I have avoided many detailed linguistic discussions. You should read the whole chapter to check them out.

The origins of the Germanic subfamily of Indo-European cannot be understood without acknowledging its interactions with a language group that has been its long-time neighbour: the Finnic subgroup of the Uralic language family. Indo-European and Uralic are linked to one another in two ways: they are probably related to one another in deep time — how deep is impossible to say3 — and Indo-European has been a constant source from which words were borrowed into Uralic languages, from the fourth millennium BC up to the present day.4 The section of the Uralic family that has always remained in close proximity to the Indo-European dialects which eventually turned into Germanic is Finnic. I use the term Finnic with a slightly idiosyncratic meaning : it covers the Finno-Saamic protolanguage and both of its children, Saami and Balto-Finnic.(…)

Schrijver (2014). The Finnic family tree (simplified)

Linguistically, the relationship between Indo-European and Uralic has always been asymmetrical. While hundreds of loanwords flowed into Uralic languages from Indo-European languages such as Germanic, Balto-Slavic, Iranian, and Proto-Indo-European itself, hardly any Uralic loanwords have entered the Indo-European languages (apart from a few relatively late dialectal loans into e.g. Russian and the Scandinavian languages). This strongly suggests that Uralic speakers have always been more receptive to ideas coming from Indo-European–speaking areas than the other way around. This inequality probably began when farming and the entire way of life that accompanies it reached Uralic-speaking territory via Indo-European–speaking territory, so that Uralic speakers, who traditionally were hunter-gatherers of the mixed and evergreen forest zone of northeastern Europe and gradually switched to an existence as sedentary farmers, were more likely to pick up ideas and the words that go with them from Indo-European than from anywhere else.

Farming requires a different mind-set from a hunter-gatherer existence. Farmers are generally sedentary, model the landscape, and have an agricultural calendar to determine their actions. Hunter-gatherers of the northern forest zone are generally nomadic, and rather than themselves modelling the natural environment they are modelled by it: their calendar depends on when and where a particular natural resource is available.(…)

All of this is no doubt a simplification of the thousands of years of associations between speakers of Uralic and speakers of Indo-European, but the loanword evidence strongly suggests that by and large relations between the two groups were highly unequal. The single direction in which loanwords flowed, and the mass of loanwords involved, can be compared with the relation between Latin and the vernacular languages in the Roman Empire, almost all of which disappeared in favour of Latin. It is therefore certain that groups of Uralic speakers switched to Indo-European. The question is whether we can trace those groups and, more particularly, whether Finnic speakers switching to Indo-European were involved in creating the Indo-European dialect we now know as Germanic.

Convergence of Finnic and Germanic

What both have in common is that the sound structures of Finnic and Germanic, which started from very different beginnings, apparently came to resemble one another significantly. If that is what we observe, we must conclude that both languages converged as a result of contact.

During the approximately five to six millennia that separate Proto-Uralic from Modern Finnish, there was only one episode during which the consonantal system underwent a dramatic overhaul. This episode separates the Finno-Saamic protolanguage, which is phonologically extremely conservative, from the Balto-Finnic protolanguage, which is very innovative.


By the time Finno-Saamic developed into Balto-Finnic, the consonant system was very different:


In Balto-Finnic, the entire palatal series has been lost, apart from j, and the contrast between dentals and alveolars has disappeared: out of three different s-sounds only one remains. The fricatives ð and γ have been lost, and so has the velar nasal ŋ. The only increase has been in the number of long (geminate) consonants by the appearance of ss, mm, nn, and ll. The loss of separate alveolar and palatal series and the disappearance of ŋ could be conceived as convergences towards Proto-Germanic, which lacked such consonants. This is not obvious for the loss of the voiced fricatives γ, ð, which Proto-Germanic did possess. However, this way of comparing Balto-Finnic and Germanic is flawed in an important respect: what we are doing is assessing convergence by comparing the dynamic development from Finno-Saamic to Balto-Finnic to the static system of Proto-Germanic, as if Proto-Germanic is not itself the result of a set of changes to the ancestral Pre-Germanic consonantal system. If we wish to find out whether there was convergence and which language converged on which, what we should do, therefore, is to compare the dynamic development of Finno-Saamic to Balto-Finnic to the dynamic development of Pre-Germanic to Proto-Germanic, because only that procedure will allow us to state whether Balto-Finnic moved towards Proto-Germanic, or Proto-Germanic moved towards Balto-Finnic, or both moved towards a third language. The Pre-Germanic consonantal system can be reconstructed as follows: 7


The slashes in the second and third rows indicate the uncertainty about the Proto-Indo-European nature of the sounds involved. (…)

What resulted was the following Proto-Germanic consonant system:


We are now in a better position to answer the question whether Proto-Germanic and Balto-Finnic have converged. Three striking developments affected both languages:

  • Both languages lost the palatalized series of consonants (apart from j), which in both languages became non-palatalized.
  • Both languages developed an extensive set of long (geminate) consonants; Pre-Germanic had none, while Finno-Saamic already had a few.
  • Both languages developed an h.

These similarities between the languages are considerable.

The idea that perhaps both languages moved towards a lost third language, whose speakers may have been assimilated to both Balto-Finnic and Germanic, provides a fuller explanation but suffers from the drawback that it shifts the full burden of the explanation to a mysterious ‘language X’ that is called upon only in order to explain the developments in Proto-Germanic and Balto-Finnic. That comes dangerously close to circular reasoning.

Verner’s Law in Pre-Germanic

As we have seen in the preceding section, Verner’s law is a sound change that affected originally voiceless consonants, so *p , t , k , kj , kw, s of the Pre-Germanic system. These normally became the Proto-Germanic voiceless fricatives *f, θ, h, h, hw, s, respectively. But if *p, t, k etc. were preceded by an originally unstressed syllable, Verner’s law intervened and they were turned into voiced consonants. Those voiced consonants merged with the series *bh, dh, gh of the Pre-Germanic system and therefore subsequently underwent all changes that the latter did, turning out as *b/v , *d/ð , g/γ in the Proto-Germanic system (that is, v, ð, γ after a vowel and b, d, g in all other environments in the word). When *s was affected by Verner’s Law, a new phoneme *z arose. In a diagram:


While it is very common in the history of European languages for stress to influence the development of vowels, it only very rarely affected consonants in this part of the world. Verner’s law is a striking exception. It resembles a development which, on a much larger scale, affected Finno-Saamic: consonant gradation.(…)

In all Finno-Saamic languages, rhythmic gradation has become phonemic and fossilized. The connection between rhythmic gradation and Verner’s law is relatively straightforward: both processes involve changing a voiceless consonant after an unstressed syllable. (…)

We can therefore repeat for Proto-Uralic the argument that persuaded us earlier that gradation in Saami and Balto-Finnic must go back to the common Finno-Saamic protolanguage: the similarity of the gradation rules in Nganasan to those in Finno-Saamic is so specific and so detailed, and the phenomenon of gradation so rare in the languages of the world, that gradation must be reconstructed for the Uralic protolanguage.

Verner’s law turns all voiceless obstruents (Pre-Germanic *p, t, k, kj, kw, s) into voiced obstruents (ultimately Proto-Germanic *b/v , d/ð, g/γ, g/γ, gw, z) after a Pre-Germanic unstressed syllable. Rhythmic gradation turns all voiceless obstruents after an unstressed syllable into weak-grade consonants, which means that *p, t, k, s become Finnic *b/v , d/ð , g/γ, z. This is striking. Given the geographical proximity of Balto-Finnic and Germanic and given the rare occurrence of stress-related consonant changes in European languages, it would be unreasonable to think that Verner’s law and rhythmic gradation have nothing to do with one another.

It is very hard to accept, however, that gradation is the result of copying Verner’s law into Finnic. First of all, Verner’s law, which might account for rhythmic gradation, in no way accounts for syllabic gradation in Finnic. And, second, gradation can be shown to be an inherited feature of Finnic which goes all the way back to Proto-Uralic. Once one acknowledges that Verner’s law and gradation are causally linked and that gradation cannot be explained as a result of copying Verner’s law into Finnic, there remains only one possibility: Verner’s law is a copy of Finnic rhythmic gradation into Germanic. That means that we have finally managed to find what we were looking for all along: a Finnic sound feature in Germanic that betrays that Finnic speakers shifted to Germanic and spoke Germanic with a Finnic accent. The consequence of this idea is dramatic: since Verner’s law affected all of Germanic, all of Germanic has a Finnic accent.

Late Chalcolithic migrations ca. 2600-2250 BC.

On the basis of this evidence for Finnic speakers shifting to Germanic, it is possible to ascribe other, less specifically Finnic traits in Germanic to the same source. The most obvious trait is the fixation of the main stress on the initial syllable of the word. Initial stress is inherited in Finno-Saamic but was adopted in Germanic only after the operation of Verner’s law, quite probably under Finnic influence. The consonantal changes described in section V.3.1 can be attributed to Finnic with less confidence. The best case can be made for the development of geminate (double) consonants in Germanic, which did not inherit any of them, while Finno-Saamic inherited *pp, tt, kk, cc and took their presence as a cue to develop other geminates such as *nn and *ll . Possibly geminates developed so easily in Proto-Germanic because Finnic speakers (who switched to Germanic) were familiar with them. Other consonantal changes, such as the loss of the palatalized series in both Germanic and Balto-Finnic and the elimination of the different s- and c-phonemes, might have occurred for the same reason: if Balto-Finnic had undergone them earlier than Germanic, which we do not know, they could have constituted part of the Balto-Finnic accent in Germanic. An alternative take on those changes starts from the observation that they all constitute simplifications of an older, richer system of consonants. While simplifications can be and often are caused by language shift if the new speakers lacked certain phonemes in their original language, simplifications do not require an explanation by shift: languages are capable of simplifying a complex system all by themselves. Yet the similarities between the simplifications in Germanic and in Balto-Finnic are so obvious that one would not want to ascribe their co-occurrence to accidental circumstances.

Grimm’s Law in Proto-Germanic (speculative)

Voiceless lenis pronunciation of b, d, g is typical of the majority of German and Scandinavian dialects, so may well have been inherited from Proto-Germanic. Voiceless lenis is also the pronunciation that has been assumed to underlie the weak grades of Finno-Saamic single *p, t, k. If Proto-Germanic *b, d, g were indeed voiceless lenis, the single most striking result of the Germanic consonant shift is that it eliminated the phonological difference between voiced and voiceless consonants that Germanic had inherited from Proto-Indo-European (…) Since neither Finno-Saamic nor Balto-Finnic possessed a phonological difference between voiced and voiceless obstruents, its loss in Proto-Germanic can be regarded as yet another example of a Finnic feature in Germanic.


It is clear that this account of the first Germanic consonant shift as yet another example of Finnic influence is to some degree speculative. The point I am making is not that the Germanic consonant shift must be explained on the basis of Finnic influence, like Verner’s law and word-initial stress, only that it can be explained in this way, just like other features of the Germanic sound system discussed earlier, such as the loss of palatalized consonants and the rise of geminates.

A consequence of this account of the origins of the Proto-Germanic consonantal system is that the transition from Pre-Germanic to Proto-Germanic was entirely directed by Finnic. Or, to put it in less subtle words: Indo-European consonants became Germanic consonants when they were pronounced by Finnic speakers.

Post-Bell-Beaker Europe, after ca. 2200 BC.

The vocalic system, on the other hand, presented less difficulties for both, Indo-European and Uralic speakers, since it was quite similar.

Schrijver goes on to postulate certain asymmetric differences in loans, especially with regard to Proto-Germanic, Balto-Finnic, Proto-Saamic, Proto-Baltic, and later contacts, including a potential non-Uralic, non-IE substrate language to justify some of these, which may in turn be connected with Kroonen’s agricultural substrate hypothesis of Proto-Germanic, and thus also with the other surviving Scandinavian Neolithic cultures before the eventual simplification of the cultural landscape during the Bronze Age.

Conclusion on the origin of Germanic

The Finnic-Germanic contact situation has turned out to be of a canonical type. To Finnic speakers, people who spoke prehistoric Germanic and its ancestor, Pre-Germanic, must have been role models. Why they were remains unclear. In the best traditions of Uralic–Indo-European contacts, Finnic speakers adopted masses of loanwords from (Pre-)Germanic. Some Finnic speakers even went a crucial step further and became bilingual: they spoke Pre-Germanic according to the possibilities offered by the Finnic sound system, which meant they spoke with a strong accent. The accent expressed itself as radical changes in the Pre-Germanic consonantal system and no changes in the Pre-Germanic vowel system. This speech variety became very successful and turned an Indo-European dialect into what we now know as Germanic. Bilingual speakers became monolingual speakers of Germanic.

What we do not know is for how long Finnic-Germanic bilingualism persisted. It is possible that it lasted for some time because both partners grew more alike even with respect to features whose origin we cannot assign to either of them (loss of palatalized consonants): this suggests, perhaps, that both languages became more similar because generally they were housed in the same brain. What we can say with more confidence is that the bilingual situation ultimately favoured Germanic over Finnic: loanwords continued to flow in one direction only, from Germanic to Finnic, hence it is clear that Germanic speakers remained role models.

This is as far as the linguistic evidence can take us for the moment.

Based on archaeology and genetics, I think we can say that the close North-West Indo-European – Proto-Finnic interaction in Scandinavia lasted for hundreds of years, during the time when a unifying Nordic culture and language developed from Bell Beaker maritime elites dominating over Corded Ware groups.

As we know, Uralic languages were in close contact with Middle PIE, and also later with Proto-Indo-Iranian. This Pre-Germanic development in Scandinavia is therefore another hint at the identification of a rather early Proto-Finnic spoken in the Baltic area – potentially then by Battle Axe groups – , and thus the general identification of Uralic expansion with the different Corded Ware groups.

NOTE. The ‘common’ loss of certain palatals, which Schrijver interprets as a change of Pre-Germanic from the inherited Proto-Indo-European, may in fact not be such – in the opinion of bitectalists, including us, and especially taking the North-West Indo-European reconstruction and the Corded Ware substrate hypothesis into account – , so this effect would be a rather unidirectional shift from Finnic to Germanic. On the other hand, certain palatalization trends which some have described for Germanic could in fact be explained precisely by this bidirectional influence.


Proto-Indo-European homeland south of the Caucasus?

User Camulogène Rix at Anthrogenica posted an interesting excerpt of Reich’s new book in a thread on ancient DNA studies in the news (emphasis mine):

Ancient DNA available from this time in Anatolia shows no evidence of steppe ancestry similar to that in the Yamnaya (although the evidence here is circumstantial as no ancient DNA from the Hittites themselves has yet been published). This suggests to me that the most likely location of the population that first spoke an Indo-European language was south of the Caucasus Mountains, perhaps in present-day Iran or Armenia, because ancient DNA from people who lived there matches what we would expect for a source population both for the Yamnaya and for ancient Anatolians. If this scenario is right the population sent one branch up into the steppe-mixing with steppe hunter-gatherers in a one-to-one ratio to become the Yamnaya as described earlier- and another to Anatolia to found the ancestors of people there who spoke languages such as Hittite.

The thread has since logically become a trolling hell, and it seems not to be working right for hours now.

Reich’s proposal based on ancestral components to explain the formation of a people and language is a continuation of their emphasis on ancestry to explain cultures and languages. It seems quite interesting to see this happen again, given their current trend to surreptitiously modify their previous ‘Yamnaya ancestry’ concept and Yamnaya millennia-long R1a-R1b community (that supposedly explains a Yamna -> Corded Ware -> Bell Beaker migration) to a more general ‘steppe people’ sharing a ‘steppe ancestry’ who spoke a ‘steppe language’.

Interesting arrows of dispersal of steppe ancestry, from Yamna -> Corded Ware -> Bell Beaker, from David Reich’s new book (yes, from 2018, number one bestseller in Amazon.com).

This new idea based on ancestral components suffers thus from the same essential methodological problems, which equate it – yet again – to pure speculation:

  1. It is a conclusion based on the genomic analysis of few individuals from distant regions and different periods, and – maybe more disturbingly – on the lack of steppe ancestry in the few samples at hand.
  2. Wait, what? Steppe ancestry? So they are trying to derive potential genetic connections among specific prehistoric cultures with a poorly depicted genetic sketch, based on previous flawed concepts (instead of on anthropological disciplines), which seems a rather long stretch for any scientist, whether they are content with seeing themselves as barbaric scientific conquerors of academic disciplines or not. In other words, statistics is also science (in fact, the main one to assert anything in almost any scientific field), and you cannot overcome essential errors (design, sampling, hypothesis testing) merely by using a priori correct statistical methods. Results obtained this way constitute a statistical fallacy.

  3. Even if the sampling and hypothesis testing were fine, to derive anthropological models from genomic investigation is completely wrong. Ancestral component ≠ population.
  4. To include not only potential migrations, but also languages spoken by these potential migrants? It’s sad that we have a need to repeat it, but if ancestral component ≠ population, how could ancestral component = language?

The Proto-Indo-European-speaking community

This is what we know about the formation of a Proto-Indo-European community (i.e. a community speaking a reconstructible Proto-Indo-European language) in the Pontic-Caspian steppe, which is based on linguistic reconstruction and guesstimates, tracing archaeological cultures backwards from cultures known to have spoken ancient (proto-)languages, and helping both disciplines with anthropological models (for which ancient genomics is only helping select certain details) of migration or – rarely – cultural diffusion:

NOTE. The following dates are obviously simplified. Read here a more detailed linguistic assessment based on phonology.

Most likely Pre-Proto-Anatolian migration with Suvorovo-Novodanilovka chiefs in the North Pontic steppe and the Balkans.
  • ca. 5000 BC. Early Proto-Indo-European (or Indo-Uralic) spoken probably during the formation and development of a loose Early Khvalynsk – Sredni Stog I cultural-historical community over the Pontic-Caspian steppe region, whose indigenous population probably had mainly Caucasus hunter-gatherer ancestry.
  • ca. 4500 BC. Khvalynsk probably speaking Middle Proto-Indo-European expands, most likely including Suvorovo-Novodanilovka chiefs into the North Pontic steppe, and probably expanding R1b-M269 lineages for the first time.
  • ca. 4000 BC. Separated communities develop, including North Pontic cultures probably gradually dominated by R1a-Z645 (potentially speaking Proto-Uralic); and Khvalynsk (and Repin) cultures probably dominated by R1b-L23 lineages, most likely developing a Late Proto-Indo-European already separated from Proto-Anatolian.
  • ca. 3500 BC. A Proto-Corded Ware population dominated by R1a-Z645 expands to the north, and slightly later an early Yamna community develops from Late Khvalynsk and Repin, expanding to the west of the Don River, and to the east into Afanasevo. This is most likely the period of reduction of variability and expansion of subclades of R1a-Z645 and R1b-L23 that we expect to see with more samples.
  • ca. 3000 BC. Expansion of Corded Ware migrants in northern Europe, and Yamna migrants along the Danube and into the Balkans, with further reduction and expansion of certain subclades.
  • ca. 2500 BC. Expansion of Bell Beaker migrants dominated by R1b-L51 subclades in Europe, and late Corded Ware migrants in east Yamna expanding R1a-Z93 subclades.

All these events are compatible with language reconstruction in mainstream European schools since at least the 1980s, supported by traditional archaeological research of the past 20 years, and is being confirmed with Genomics.

For those willingly lost in a myriad of new dreams boosted by the shallow comment contained in David Reich’s paragraph on CHG ancestry, even he does not doubt that the origin of Late Proto-Indo-European lies in Yamna, to the north of the Caucasus, based on Anthony’s (2007) account:

Both images from the book, posted by Twitter user Jasper at https://twitter.com/jaspergregory.

NOTE: By the way, David Anthony, one of the main sources of information for Reich’s group, never considered Corded Ware to have received Yamna migrants, and althought he changed his model due to the conclusions of the 2015 papers, he has recently changed his model again to adapt it to the inconsistencies found in phylogeography.

CHG ancestry and PIE homeland south of the Caucasus

As for the potential origins of CHG ancestry in early Proto-Indo-European speakers, I already stated clearly my opinion quite recently. They may be attributed to:

Just to be clear, an expansion of Proto-Anatolian to the south, through the Caucasus, cannot be discarded today. It will remain a possibility until Maykop and more Balkan Chalcolithic and Anatolian-speaking samples are published.

However, an original Early Proto-Indo-European community south of the Caucasus seems to me highly unlikely, based on anthropological data, which should drive any conclusion. From what I could read, here are the rather simplistic arguments used:

  • Gimbutas and Maykop: Maykop was thought to be (in Gimbutas’ times) a rather late archaeological culture, directly connected to a Transcaucasian Copper Age culture ca. 2400-2300 BC. It has been demonstrated in recent years that this culture is substantially older, and even then language guesstimates for a Late PIE / Proto-Anatolian would not fit a migration to the north. While our ignorance may certainly be used to derive far-fetched conclusions about potential migrations from and to it, using Gimbutas (or any archaeological theory until the 1990s) today does not make any sense. Still less if we think that she favoured a steppe homeland.

NOTE. It seems that the Reich Lab may have already access to Maykop samples, so this suggested Proto-Indo-European – Maykop connection may have some real foundation. Regardless, we already know that intense contacts happened, so there will be no surprise (unless Y-DNA shows some sort of direct continuity from one to the other).

  • Gamkrelidze & Ivanov: they argued for an Armenian homeland (and are thus at the origin of yet another autochthonous continuity theory), but they did so to support their glottalic theory, i.e. merely to support what they saw as favouring their linguistic model (with Armenian being the most archaic dialect). The glottalic theory is supported today – as far as I know – mainly by Kortlandt, Jagodziński, or (Nostraticist) Bomhard, but even they most likely would not need to argue for an Armenian homeland. In fact, their support of a Graeco-Aryan group (also supported by Gamkrelidze & Ivanov) would be against this, at least in archaeological terms.
  • Colin Renfrew and the Anatolian homeland: This conceptual umbrella of language spreading with farming everywhere has changed so much and so many times in the past 20 years, with so many glottochronological and archaeological estimates circulating, that you can support anything by now using them. Mostly used today for abstract models of long-lasting language contacts, cultural diffusion, and constellation analogies. Anyway, he strives to keep up-to-date information to revise the model, that much is certain:
  • Glottochronology, phylogenetic trees, Swadesh list analysis, statistical estimates, psychics, pyramid power, and healing crystals: no, please, no.
Science Magazine
“A first line of evidence comes from linguistic analysis based on quantitative lexical data, which returned a tree compatible with the Anatolian hypothesis

In principle, unlike many other recent autochthonous continuity theories, I doubt there can be much racial-based opposition anywhere in the world to an origin of Proto-Indo-European in the Middle East, where the oldest civilizations appeared – apart, obviously, from modern Northeast and Northwest Caucasian, Kartvelian, or Semitic speakers, who may in turn have to revisit their autochthonous continuity theories radically…

Nevertheless, it is obvious that prehistoric (and many historic) migrations are signalled by the reduction in variability and expansion of certain Y-DNA haplogroups, and not just by ancestral components. That is generally accepted, although the reasons for this almost universal phenomenon are not always clear.

In fact, Proto-Anatolian and Common Anatolian speakers need not share any ancestral component, PCA cluster, or any other statistical parameter related to steppe populations, not even the same Y-DNA haplogroups, given that approximately three thousand years might have passed between their split from an Indo-Hittite community and the first attested Anatolian-speaking communities…We must carefully follow their tracks from Anatolia ca. 1500 BC to the steppe ca. 4500 BC, otherwise we risk creating another mess like the Corded Ware one.

In my opinion, the substantial contribution of EHG ancestry and R1a-M417 lineages to the Pontic-Caspian steppe (probably ca. 6500 BC) from Central or East Eurasia is the most recent sizeable genomic event in the region, and thus the best candidate for the community that expanded a language ancestral to Proto-Indo-European – whether you call it Pre-Proto-Indo-European, Pre-Indo-Uralic, or Eurasiatic, depending on your preferences.

An early (and substantial) contribution of CHG ancestry in Khvalynsk relative to North Pontic cultures, if it is found with new samples, may actually be a further proof of the Caucasian substrate of Proto-Indo-European proposed by Kortlandt (or Bomhard) as contributing to the differentiation of Middle PIE from Uralic. Genomics could thus help support, again, traditional disciplines in accepting or rejecting academic controversial theories.


In the case of an Early PIE (or Indo-Uralic) homeland, genomic data is scarce. But all traditional anthropological disciplines point to the Pontic-Caspian steppe, so we should stick to it, regardless of the informal suggestion written by a renown geneticist in one paragraph of a book conceived as an introduction to the field.

It seems we are not learning much from the hundreds of peer-reviewed, statistically (superficially, at least) sound genetic papers whose anthropological conclusions have been proven wrong by now. A lot of people should be spending their time learning about the complex, endless methods at hand in this kind of research – not just bioinformatics – , instead of fruitlessly speculating about wild unsubstantiated proposals.

As a final note, I would like to remind some in the discussion, who seem to dismiss the identification of CHG with Proto-Indo-European by supporting a “R1a-R1b” community for PIE, of their previous commitment to ancestral components in identifying peoples and languages, and thus their support to Reich’s (and his group’s) fundamental premises.

You cannot have it both ways. At least David Reich is being consistent.


Statistical methods fashionable again in Linguistics: Reconstructing Proto-Australian dialects

Reconstructing remote relationships – Proto-Australian noun class prefixation, by Mark Harvey & Robert Mailhammer, Diachronica (2017) 34(4): 470–515


Evaluation of hypotheses on genetic relationships depends on two factors: database size and criteria on correspondence quality. For hypotheses on remote relationships, databases are often small. Therefore, detailed consideration of criteria on correspondence quality is important. Hypotheses on remote relationships commonly involve greater geographical and temporal ranges. Consequently, we propose that there are two factors which are likely to play a greater role in comparing hypotheses of chance, contact and inheritance for remote relationships: (i) spatial distribution of corresponding forms; and (ii) language specific unpredictability in related paradigms. Concentrated spatial distributions disfavour hypotheses of chance, and discontinuous distributions disfavour contact hypotheses, whereas hypotheses of inheritance may accommodate both. Higher levels of language-specific unpredictability favour remote over recent transmission. We consider a remote relationship hypothesis, the Proto-Australian hypothesis. We take noun class prefixation as a test dataset for evaluating this hypothesis against these two criteria, and we show that inheritance is favoured over chance and contact.

I was redirected to this work by my wife – who discovered it reading BBC News – , suspicious of its potential glottochronological content. However, I must say – speaking from my absolute ignorance of the main language family investigated – , that it seemed in general an interesting read, with some thorough discussion and attention to detail.

The statistical analyses, however, seem to disrupt the content, and – in my opinion – do not help support its conclusions.

Map of Non-Pama-Nyungan languages.

Computer Science and Linguistics

We are evidently on alert to tackle dubious research, because of the revival of pseudoscientific methods in linguistic investigation, promoted (yet again) by Nature.

It seems that journals with the highest impact factor, in their search for groundbreaking conclusions supported by any methods involving numbers, are setting a still lower level of standards for academic disciplines.

NOTE. If you think about it – if glottochronology has survived the disgrace it fell into in the 2000s, to come back again now to the top of the publishing industry… How can we expect the “Yamnaya ancestry” concept to be overcome? I guess we will still see certain Eastern Europeans in 2030 arguing for elevated steppe ancestry here and there to support the conclusions of the 2015 papers, no matter what…

I am sure that worse times lie ahead for traditional comparative grammar. For example, it seems that there will be more publications on Proto-Indo-European using novel computer methods: a group led by Janhunen and Pyysalo, from the Department of Languages at the University of Helsinki, promises – under an ever-growing bubble of mistery (or so it seems from their Twitter and Facebook accounts) – a machine-implemented reconstruction (with the generative etymological PIE lexicon project) that will once and for all solve all our previous ‘inconsistencies’…

Spoiler alert for their publications: whether they select to go on mainly with computer-implemented methods, or they use them to support more traditional results, their conclusions will confirm (surprise!) their authors’ previous reactionary theses, such as a renewed support for the traditional monolaryngealism, and a rejection of Kortlandt’s or Kloekhorst’s (i.e. the Leiden School’s) theories on Proto-Indo-European phonology, and thus a PIE relationship to Proto-Uralic, probably stressing yet again an independent origin for both proto-languages.

See also:

R1b-V88 migration through Southern Italy into Green Sahara corridor, and the Afroasiatic connection


Open access article The peopling of the last Green Sahara revealed by high-coverage resequencing of trans-Saharan patrilineages, by D’Atanasio, Trombetta, Bonito, et al., Genome Biology (2018) 19:20.


Little is known about the peopling of the Sahara during the Holocene climatic optimum, when the desert was replaced by a fertile environment.

In order to investigate the role of the last Green Sahara in the peopling of Africa, we deep-sequence the whole non-repetitive portion of the Y chromosome in 104 males selected as representative of haplogroups which are currently found to the north and to the south of the Sahara. We identify 5,966 mutations, from which we extract 142 informative markers then genotyped in about 8,000 subjects from 145 African, Eurasian and African American populations. We find that the coalescence age of the trans-Saharan haplogroups dates back to the last Green Sahara, while most northern African or sub-Saharan clades expanded locally in the subsequent arid phase.

Our findings suggest that the Green Sahara promoted human movements and demographic expansions, possibly linked to the adoption of pastoralism. Comparing our results with previously reported genome-wide data, we also find evidence for a sex-biased sub-Saharan contribution to northern Africans, suggesting that historical events such as the trans-Saharan slave trade mainly contributed to the mtDNA and autosomal gene pool, whereas the northern African paternal gene pool was mainly shaped by more ancient events.

Maximum parsimony Y chromosome tree and dating of the four trans-Saharan haplogroups. a Phylogenetic relations among the 150 samples analysed here. Each haplogroup is labelled in a different colour. The four Y sequences from ancient samples are marked by the dagger symbol. b Phylogenetic tree of the four trans-Saharan haplogroups, aligned to the timeline (at the bottom). At the tip of each lineage, the ethno-geographic affiliation of the corresponding sample is represented by a circle, coloured according to the legend (bottom left). The last Green Sahara period is highlighted by a green belt in the background

Also, interesting excerpts:

The fertile environment established in the Green Sahara probably promoted demographic expansions and rapid dispersals of the human groups, as suggested by the great homogeneity in the material culture of the early Holocene Saharan populations [62]. Our data for all the four trans-Saharan haplogroups are consistent with this scenario, since we found several multifurcated topologies, which can be considered as phylogenetic footprints of demographic expansions. The multifurcated structure of the E-M2 is suggestive of a first demographic expansion, which occurred about 10.5 kya, at the beginning of the last Green Sahara (Fig. 2; Additional file 2: Figure S4). After this initial expansion, we found that most of the trans-Saharan lineages within A3-M13, E-M2 and R-V88 radiated in a narrow time interval at 8–7 kya, suggestive of population expansions that may have occurred in the same time (Fig. 2; Additional file 2: Figures S3, S4 and S6). Interestingly, during roughly the same period, the Saharan populations adopted pastoralism, probably as an adaptive strategy against a short arid period [1, 62, 63]. So, the exploitation of pastoralism resources and the reestablishment of wetter conditions could have triggered the simultaneous population expansions observed here. R-V88 also shows signals of a further and more recent (~ 5.5 kya) Saharan demographic expansion which involved the R-V1589 internal clade. We observed similar demographic patterns in all the other haplogroups in about the same period and in different geographic areas (A3-M13/V3, E-M2/V3862 and E-M78/V32 in the Horn of Africa, E-M2/M191 in the central Sahel/central Africa), in line with the hypothesis that the start of the desertification may have caused massive economic, demographic and social changes [1].

Finally, the onset of the arid conditions at the end of the last African humid period was more abrupt in the eastern Sahara compared to the central Sahara, where an extensive hydrogeological network buffered the climatic changes, which were not complete before ~ 4 kya [6, 62, 64]. Consistent with these local climatic differences, we observed slight differences among the four trans-Saharan haplogroups. Indeed, we found that the contact between northern and sub-Saharan Africa went on until ~ 4.5 kya in the central Sahara, where we mainly found the internal lineages of E-M2 and R-V88 (Additional file 2: Figures S4 and S6). In the eastern Sahara, we found a sharper and more ancient (> 5 kya) differentiation between the people from northern Africa (and, more generally, from the Mediterranean area) and the groups from the eastern sub-Saharan regions (mainly from the Horn of Africa), as testified by the distribution and the coalescence ages of the A3-M13 and E-M78 lineages (Additional file 2: Figures S3 and S5).

Time estimates and frequency maps of the four trans-Saharan haplogroups and major sub-clades. a Time estimates of the four trans-Saharan clades and their main internal lineages. To the left of the timeline, the time windows of the main climatic/historical African events are reported in different colours (legend in the upper left). b Frequency maps of the main trans-Saharan clades and sub-clades. For each map, the relative frequencies (percentages) are reported to the right

R-V88 has been observed at high frequencies in the central Sahel (northern Cameroon, northern Nigeria, Chad and Niger) and it has also been reported at low frequencies in northwestern Africa [37]. Outside the African continent, two rare R-V88 sub-lineages (R-M18 and R-V35) have been observed in Near East and southern Europe (particularly in Sardinia)[30, 37, 38, 39]. Because of its ethno-geographic distribution in the central Sahel, R-V88 has been linked to the spread of the Chadic branch of the Afroasiatic linguistic family [37, 40].

(…) the R-V88 lineages date back to 7.85 kya and its main internal branch (branch 233) forms a “star-like” topology (“Star-like” index = 0.55), suggestive of a demographic expansion. More specifically, 18 out of the 21 sequenced chromosomes belong to branch 233, which includes eight sister clades, five of which are represented by a single subject. The coalescence age of this sub-branch dates back to 5.73 kya, during the last Green Sahara period. Interestingly, the subjects included in the “star-like” structure come from northern Africa or central Sahel, tracing a trans-Saharan axis. It is worth noting that even the three lineages outside the main multifurcation (branches 230, 231 and 232) are sister lineages without any nested sub-structure. The peculiar topology of the R-V88 sequenced samples suggests that the diffusion of this haplogroup was quite rapid and possibly triggered by the Saharan favourable climate (Fig. 2b).

One of the theories I proposed in the Indo-European demic diffusion model since the first edition – based mainly on phylogeography – is that R1b-V88 lineages had probably crossed the Mediterranean through southern Italy into a Green Sahara region, and distributed from there throuh important green corridors, humid areas between megalakes. Even though this new study – like the rest of them – is based solely on modern samples, and as such is quite prone to error in assessing ancient distributions – as we have seen in Europe -, it seems that a southern Italian route (probably through Sicily) for R1b-V88 and a late expansion through Green Sahara is more and more likely.

If we accept that the migration of R1b-V88 lineages is the last great expansion through a Green Sahara, then this expansion is a potential candidate for the initial Afroasiatic expansion – whereas older haplogroup expansions would represent languages different than Afroasiatic, and more recent haplogroup expansions would represent subsequent expansions of Afroasiatic dialects, like Semitic, Hamitic, Cushitic, or Chadic – as I explained in an older post.

In absolutely shameless speculative terms, then – as is today common in Genetic studies, by the way, so let’s all have some fun here – instead of some sort of R1b/Eurasiatic continuity in Europe, as some autochthonous continuists would like, this could mean that there would be an old Afroasiatic – R1b connection. That would imply:

NOTE. Regarding the contribution of CHG ancestry in the Pontic-Caspian steppe cultures, it is usually explained as caused by exogamy, or by absorption of a previous population (as in the Indo-Iranian case), although a contribution of communities of mainly J subclades to the formation of Neolithic steppe cultures cannot be ruled out. As for some autochthonous continuists’ belief in some sort of mythical mixed steppe people with mixed haplogroups and mixed language, well…

Simple Nostratic tree by Bomhard (2008)

The Pre-Indo-European linguistic situation, before the formation of Neolithic steppe cultures, seems like pure speculation, because a) language macro-families (with the exception of Afroasiatic) are highly speculative, b) sound anthropological models are lacking for them, and c) migrations inferred from haplogroup distributions of modern populations are often incorrect:

  • Haplogroup R could then be argued to be the source of Nostratic, and earlier subclades the source of Starostin’s Borean, given the distribution of its subclades in Asia and the timing of their migrations.
  • But of course one could also argue that, given the comparatively late population expansions that Genomics is showing, supporting Western European linguistic schools – where Russian Nostraticists tend to date languages further back in timeR1b (and not R) expansion could be the marker of Nostratic languages, due to its most likely southern path (and their old subclades found in Iran and the Caucasus), which would be more in line with the wet dreams of Europeans proposing R1b autochthonous continuity theories. I like this option far less because of that, but it cannot be ruled out.

If you have read this blog before, you know I profoundly dislike lexicostatistical and glottochronological methods, and I don’t like mass comparisons either. Whereas these methods pretend to apply mathematics to big (raw) data where there is almost no knowledge of what one is doing, comparative grammar applies complex reasoning where there is a lot of partially processed data.

But, it is always fun to ask “what if they were right?” and follow from there…

See also: