Are The Genetically Steppe Bell Beaker People An Offshoot Of The Corded Ware Culture?

Davidski at Eurogenes has been developing the theory that the Bell Beaker people with steppe ancestry in Europe, and not just the cultures widely assumed to derive from the Bronze Age Northeast European Corded Ware Culture both have the same basic source. 

His argument runs primarily from phylogeny. It has long been known that ancient DNA from the Corded Ware Culture is rich (arguably dominant) in Y-DNA R1a-M417, which is ancestral to all of the Y-DNA R1a clades that are predominant among Eastern and Central European Indo-European language speakers.

It has also long been known that the autosomal genetics determined from ancient DNA of the Yamnaya culture was rich in Y-DNA R1b.

But, here's the catch. The Y-DNA clade of R1b found in the Yamnaya culture is not ancestral to the various clades of Y-DNA R1b-L51 which is predominant in the ancient DNA of Bell Beaker men with significant steppe ancestry (see below). Instead, ancient DNA from Yamnaya culture men was rich in Y-DNA clade R1b-Z2103, which is a sister clade to R1b-L51 which is so common in Western European Bell Beaker people and their descendants.

From Eupedia.

The initial assumption was the there was an unsampled population of Yamnaya people, presumably in the west, where the steppe meets the Balkans, in which there were Yamnaya people with Y-DNA R1b-L51 rather than the sister clade Y-DNA R1b-Z2103. The Yamnaya's patrilineal descendants are described in the chart below and are relatively minor contributors to the modern West Eurasian gene pool.

But, according to Davidski, at some point, ancient DNA from a significant number of men who geographically and archaeologically were Corded Ware Culture men were found with Y-DNA R1b-L51, suggesting that the sister clade to the Yamnaya R1b-Z2103 was found in the forested steppe to the North of the Yamnaya culture proper, and not to the west where the steppe meets the Balkans.

Fierce competition of patrilineal and patrilocal clans within the Corded Ware Culture is what causes Y-DNA R1a-M417 and R1b-L51 to become dominant.

As Davidski summarizes and clarifies in comments on his most recent post:

[The Bell Beaker culture] formed in the Lower Rhine region, because all Bell Beakers show ancestry from Dutch Bell Beakers, and they have hunter/farmer ancestry from the North Sea area. Bell Beakers are derived from the Single Grave culture, and the Single Grave culture came from the east, and probably via Poland.

I hope to update this post with journal article references at some point.

This isn't a complete narrative, but it is much closer than it was and quite different from what seemed like the most plausible hypothesis not very long ago. The paradigm has not caught up and still focused on the related but less directly Yamnaya people as the source of European Indo-Europeans.

The Details

In a July 20, 2021 post he sets forth this hypothesis more fully:

There's been a lot of talk lately about the finding that the peoples associated with the Corded Ware and Yamnaya archeological cultures were close cousins (for instance, see here). As I've already pointed out, this is an interesting discovery, but, at this stage, it's difficult to know what it means exactly.

It might mean that the Yamnayans were the direct predecessors of the Corded Ware people. Or it might just mean that, at some point, the Corded Ware and Yamnaya populations swapped women regularly (that is, they practiced female exogamy with each other).

In any case, I feel that several important facts aren't being taken into account by most of the interested parties. These facts include, in no particular order:

- despite being closely related, the Corded Ware and Yamnaya peoples were highly adapted to very different ecological zones - temperate forests and arid steppes, respectively - and this is surely not something that happened within a few years and probably not even within a couple of generations

- both the Corded Ware and Yamnaya populations expanded widely and rapidly at around the same time, but never got in each others way, probably because they occupied very different ecological niches

- despite sharing the R1b Y-chromosome haplogroup, their paternal origins were quite different, with Corded Ware males rich in R1a-M417 and R1b-L51 and Yamnaya males rich in R1b-Z2103 and I2a-L699

I suppose it's possible that the Corded Ware people were overwhelmingly and directly derived from the Yamnaya population. But right now my view is that, even if they were, then the Yamnaya population that they came from was quite different from the classic, R1b-Z2103-rich Yamnaya that spread rapidly across the steppes.

Indeed, perhaps what we're dealing with here is a very early (proto?) Yamnaya gene pool located somewhere in the border zone between the forests and the steppes, that then split into two main sub-populations, with one of these groups heading north and the other south?

 The Funnelbeaker Culture (4300 BCE to 2800 BCE)

This conclusion seems to follow from the conclusion that the Single Grave Culture has Y-DNA R1b-312 in ancient DNA, as he explained in a January 16, 2019 post. The Single Grave Culture, replaced the Funnelbeaker culture, and was a Copper Age (2800 BCE to 2200 BCE) local variant the Corded Ware Culture (2900 BCE to 2350 BCE).

Single Grave term refers to a series of . . .  communities of the 3rd millennium BCE living in southern Scandinavia, Northern Germany, and the Low Countries that share the practice of single burial, the deceased usually being accompanied by a battle-axe, amber beads, and pottery vessels.

As he explains in an August 27, 2021 post:

R1a-M417 and R1b-L51 are by far the most important Y-chromosome haplogroups in Europe today. More precisely, R1a-M417 dominates in Eastern Europe, while R1b-L51 in Western Europe.

It's been obvious for a while now, at least to me, that both of these Y-haplogroups are closely associated with the men of the Late Neolithic Corded Ware culture (CWC). Indeed, in my mind they're the main genetic signals of its massive expansion, probably from a homeland somewhere north of the Black Sea in what is now Ukraine.

I'm still not exactly sure how the east/west dichotomy between R1a and R1b emerged in Europe, but, thanks to a new paper by Papac et al. at Science Advances, at least now I have a working hypothesis about that. . . .

so even though the CWC was clearly a community of closely related groups, there must have been some competition between its different clans. And since these clans were highly patriarchal and patrilineal, this competition probably led to different paternal lineages dominating different parts of the CWC horizon, with M417 becoming especially common in the east and L51 in the west.

Of course, the expansions of post-Corded Ware groups, such as the M417-rich Slavs in Eastern Europe and L51-rich Celts in Western Europe, were also instrumental in creating Europe's R1a/R1b dichotomy, but obviously these groups were in large part the heirs of the CWC.

Bell Beaker Men With And Without Steppe Ancestry

Bell Beaker men from Iberia in the early few centuries of this culture do not have enhanced steppe ancestry, suggesting that Indo-Europeans from the steppe were converted to the Bell Beaker culture by cultural means at some point fairly early on somewhere in Western Europe, rather than having the culture spread purely by expansion and demic replacement from a common source.

Supersymmetry and String Theory Are In Bad Shape

Woit makes his latest and still convincing case that Supersymmetry and String Theory are in deep trouble in his latest blog post reviewing some recent publications. 

He opens with a quote from the Economist in an article directed at educated laymen:

But, no Susy, no string theory. And, 13 years after the LHC opened, no sparticles have shown up. Even two as-yet-unexplained results announced earlier this year (one from the LHC and one from a smaller machine) offer no evidence directly supporting Susy. Many physicists thus worry they have been on a wild-goose chase… 

Without Susy, string theory thus looks pretty-much dead as a theory of everything. Which, if true, clears the field for non-string theories of everything.

This is right on target. 

Earliest Modern Human Bones Outside of Africa Found

This news is three and half years old, but I didn't blog it at the time, so I'm mentioning it now.

While genetic evidence suggests that almost all modern humans outside Africa are primarily descended from an Out of Africa migration about 50,000 years ago, coinciding with the Upper Paleolithic era, there is hard evidence that there were modern humans outside of African much earlier than then. It could be that the earlier humans were an expansion that failed, or that they were so demographically overwhelming by a later wave of modern humans who were only slightly different genetically from them, that the genetic traces of the earlier waves aren't distinguishable from modern genetic evidence. 

Earliest modern humans out of Africa

Recent paleoanthropological studies have suggested that modern humans migrated from Africa as early as the beginning of the Late Pleistocene, 120,000 years ago. Hershkovitz et al. now suggest that early modern humans were already present outside of Africa more than 55,000 years earlier (see the Perspective by Stringer and Galway-Witham). During excavations of sediments at Mount Carmel, Israel, they found a fossil of a mouth part, a left hemimaxilla, with almost complete dentition.

The sediments contain a series of well-defined hearths and a rich stone-based industry, as well as abundant animal remains. Analysis of the human remains, and dating of the site and the fossil itself, indicate a likely age of at least 177,000 years for the fossil—making it the oldest member of the Homo sapiens clade found outside Africa.

Abstract

To date, the earliest modern human fossils found outside of Africa are dated to around 90,000 to 120,000 years ago at the Levantine sites of Skhul and Qafzeh. A maxilla and associated dentition recently discovered at Misliya Cave, Israel, was dated to 177,000 to 194,000 years ago, suggesting that members of the Homo sapiens clade left Africa earlier than previously thought. This finding changes our view on modern human dispersal and is consistent with recent genetic studies, which have posited the possibility of an earlier dispersal of Homo sapiens around 220,000 years ago. The Misliya maxilla is associated with full-fledged Levallois technology in the Levant, suggesting that the emergence of this technology is linked to the appearance of Homo sapiens in the region, as has been documented in Africa.

Israel Hershkovitz, et al., "The earliest modern humans outside Africa", 359 (6374) Science 456-459 (January 26, 2018). DOI: 10.1126/science.aap8369

Free Parameters

The paper below has a basic problem. 

It looks at how well nine different models fit observed galaxy rotation curves using a Chi-squared test. But it doesn't fully adjust for the fact that some models have more free parameters than others, even though it does used a reduced Chi-square test. It also fails to consider the the dark matter models aren't predictive.

When you compare to different models to see which is a better description of the data, while the Chi-square goodness of fit is important, you are supposed to penalize models with more free fitting parameters, usually using Akaike’s Information Criterion. Why?

In estimating the amount of information lost by a model, AIC deals with the trade-off between the goodness of fit of the model and the simplicity of the model. In other words, AIC deals with both the risk of overfitting and the risk of underfitting.

Another test used for the same purpose that is subtly different is Bayesian Information Criterion which is designed for Bayesian statistical analysis as opposed to frequentist statistical analysis like that in the AIC (other criteria exist too, such as the Hannan-Quinn Information Criterion, but are similar and less often used).

In other words, if you have enough free parameters you can make even a bad model fit the data, but this doesn't necessarily make the model better. Some of it is just smoothing out noise without adding value on a case by case basis.

The money chart is this one (which makes dark matter models, in general look much better than modified gravity models, and makes some of the dark matter models look a bit better than others, although the large uncertainty in their estimation of the modified gravity theories also exaggerates the differences, each of which have fit values within two sigma of the best fits that are below the mean values of the dark matter models, and hence are also not statistically significant differences):

But this breaks down when you see that the dark matter models each have three or for parameters adjusted on a galaxy by galaxy basis to maximize the quality of the fit, while the modified gravity models have just one (which is shared with the dark matter models) in addition to a parameter that is fixed for all galaxies considered:

The paper's comparison of the fits of the dark matter models which have either three or four parameters each, is more fair since it is more comparable (and has a much lower uncertainty as well, which also makes it more valid). The differences between the dark matter models aren't statistically significant. But, the data do weakly disfavor the NFW halo shape that standard sterile cold dark matter theory predicts that dark matter halos should have, as analytical calculations can establish.

The preprint and abstract are as follows: 

We use the galaxy rotation curves in the SPARC database to compare 9 different dark matter and modified gravity models on an equal footing, paying special attention to the stellar mass-to-light ratios. We compare three non-interacting dark matter models, a self interacting DM (SIDM) model, two hadronically interacting DM (HIDM) models, and three modified Newtonian dynamics type models: MOND, Radial Acceleration Relation (RAR) and a maximal-disk model. The models with Gas-DM interactions generate a disky component in the dark matter, which significantly improves the fits to the rotation curves compared to all other models except an ad-hoc Einasto halo; the MOND-type models give significantly worse fits.

Nicolas Loizeau, Glennys R. Farrar, "Galaxy rotation curves disfavor traditional and self-interacting dark matter halos, preferring a disk component or ad-hoc Einasto function" arXiv 2105:00119 (April 30, 2021).

Non-Cosmological Bounds On Neutrino Masses

The Tightest Neutrino Mass Bounds (Cosmology)

While the cosmology bound on the sum of the neutrino masses (87 meV in the most strict limitation published) is subject to all sorts of theoretical limitations, the bounds on the differences between the neutrino masses due to neutrino oscillation measurements are not. If this bound applies (ruling out the inverted hierarchy as well), then the neutrino masses are about:

mv1= 6 ± 3 meV

mv2 = 15.2 ± 3.7 meV

mv3 = 64.7 ± 4.7 meV

with mv1 neutrino mass uncertainty correlated 100% with most of the uncertainty in the other values.

Direct Measurement Bounds

The Particle Data Group lists upper bounds on the electron neutrino mass as < 1.1 eV at 90% confidence, muon neutrino mass < 0.19 MeV at 90% confidence, and tau neutrino < 18.2 MeV at 95% confidence. By itself this would limit the sum of the three neutrino masses to < 18.4 MeV.

Katrin experiment papers this summer have suggested that the electron neutrino mass is really < 0.8 eV and that this limit could fall to < 0.2 eV by the end of its data collection run.

Direct Bounds Combined With Neutrino Oscillation Bounds

But, even in an inverted hierarchy case, we know from neutrino oscillation data there is a maximum spread of 0.1 eV between the three masses. So, this places a combined limit of about 1.2 eV at 90% confidence on muon and tau neutrino masses, individually, and a sum of the three neutrino masses < 3.4 eV. 

If the direct detection limit is set at 800 meV rather than 1.1 eV, as one recent Katrin paper would suggest, the sum of the three masses is limited to < 2.5 eV. 

Katrin, at the end of its run, could bring the limit on the sum of the three neutrino masses to < 0.7 eV, which is still nine times the cosmological bound, but shows some convergence.

The Particle Data Group Direct Bounds For The Two Heavier Neutrino Masses Are Way Too High

As a result, the vastly higher direct detection boundaries from the Particle Data Group for the two heavier neutrino masses aren't very meaningful. 

The direct detection bound on the first neutrino mass is probably at least 90 times too high.

The direct detection bound on the second neutrino mass is at least 158,000 times too high, and probably more like 8.68 million times too high.

The direct detection bound on the third neutrino mass is at least 15.2 million times too high, and probably more like 251 million times too high.

Another LambdaCDM Fail

The LambdaCDM model of cosmology (short for the cold dark matter with cosmological constant model) predicts a characteristic scaling relationship between galactic cluster size and velocity dispersion. Reality doesn't agree. The power law should have an exponent of three in the LambaCDM model. The power law actually differs from that by more than four sigma at 4 ± 0.1.

We investigate a kinematic scaling relation between the baryonic mass and the flat velocity dispersion, i.e. mass-velocity dispersion relation (MVDR), from the brightest cluster galaxies (BCGs) to the galaxy clusters. In our studies, the baryonic mass of BCGs is mainly estimated by photometry. The velocity dispersion profiles are explored with the integrated field unit (IFU) by Mapping Nearby Galaxies at Apache Point Observatory (MaNGA). 

For the first time, we reveal two significant results with 54 MaNGA BCGs: (1) the flat velocity dispersion profiles; (2) a tight empirical relation on the BCG-cluster scale together with cluster samples, i.e., MVDR, 

, with a tiny lognormal intrinsic scatter of 10+2−1% 

log(Mbar/M⊙)=4.1+0.1−0.1log(σlos/kms−1)+1.6+0.3−0.3

This slope is identical to the acceleration relation in galaxy clusters, which is reminiscent of the spiral galaxies, albeit at a larger characteristic acceleration scale. The residuals of the MVDR represent a Gaussian distribution, displaying no correlations with four properties: baryonic mass, scale length, surface density, and redshift. Notably, the MVDR on the BCG-cluster scale provides a strict test, which disfavors the general prediction of the slope of three in the dark matter model.

Yong Tian, Han Cheng, Stacy S. McGaugh, Chung-Ming Ko, Yun-Hsin Hsu "Mass-Velocity Dispersion Relation in MaNGA Brightest Cluster Galaxies" arXiv:2108.08980 (August 20, 2021) (published in 24 The Astrophysical Journal Letters 917).

The introduction to the paper explains:

A kinematic scaling relation is the counterpart of a dynamical scaling relation, which plays a major role in understanding fundamental physics. For examples, Kepler’s law leads to Newtonian dynamics in the Solar system. However, Kepler’s law is no longer applicable in spiral galaxies, well-known as the dark matter (DM) problem. Instead, a different kinematic relation was discovered between the baryonic mass and the flat rotation velocity, called the baryonic Tully-Fisher relation (BTFR, McGaugh et al. 2000; Verheijen 2001; McGaugh 2011; Lelli et al. 2016, 2019). Similarly in elliptical galaxies and galaxy clusters, such a correlation was found between the baryonic mass and the velocity dispersion, called the baryonic Faber-Jackson relation (BFJR, Sanders 2010; Famaey & McGaugh 2012). 

In galaxies, the BTFR and the BFJR can be implied by the radial acceleration relation (RAR) in the low acceleration approximation. Not long ago, McGaugh et al. (2016) have explored a tight correlation between the observed acceleration g(obs) = |∂Φ(obs)/∂r| = (v^2)/r and the baryonic acceleration g(bar) = GM(bar)(< r)/r^2 as 

g(obs) approximately equals the square root of g(bar)g(†) , (1) 

with an acceleration scale g(†) = (1.20±0.02)×10^−10 m s^−2 . 

In addition, the RAR implied v^4 = GM(bar)g(†) for the BTFR  and σ^4 approximately equal to GM(bar)g(†) for the BFJR (Lelli et al. 2017; McGaugh 2020; Milgrom 2020). Particularly, those correlations were clearly described in Modified Newtonian Dynamics (MOND, Milgrom 1983) about four decades ago. Furthermore, MOND has been tested in gravitational lensing effects (Tian et al. 2009; Tian & Ko 2019; Brouwer et al. 2021). However, MOND found a residual missing mass in galaxy clusters (Sanders 2003; Famaey & McGaugh 2012), which questioned the validity of the RAR on the cluster scale. 

Recently, Tian et al. (2020) investigated the existence of a radial acceleration relation on cluster scales using the Cluster Lensing and Supernova survey with Hubble (CLASH, Postman et al. 2012). In the CLASH sample, g(obs) was measured by strong-lensing, weak-lensing shear-and-magnification data (Umetsu et al. 2016), while g(bar) was calculated with X-ray data sets (Donahue et al. 2014) plus the estimated stellar mass from the empirical gas fraction (Chiu et al. 2018). The result is a parallel RAR (CLASH RAR):

g(obs) approximately equal to the square root of g(bar)g(‡) , (2) 

with a larger acceleration scale g(‡) = (2.0±0.1)×10−9 m s−2 than observed in rotating galaxies. The CLASH RAR is a tight correlation with a small intrinsic scatter σint = 15%. Other works also demonstrate a similar acceleration scale in galaxy clusters (Pradyumna et al. 2021). 

While supposing the validity of the CLASH RAR in dynamics, one can derive mass-velocity dispersion relation (MVDR), σ^4 approximately equal to GM(bar)g(‡), on the BCG-cluster scale. Moreover, its implications include a flat velocity dispersion profile in both BCGs and clusters. This similarity is reminiscent of the RAR in galaxies, albeit with a larger acceleration scale g(‡). 

Recently, Tian et al. (2021) have quantified the MVDR with 29 galaxy clusters in the HIghest X-ray FLUx Galaxy Cluster Sample (HIFLUGCS). In their studies, the baryonic mass was dominated by X-ray gas (Zhang et al. 2011) plus the stellar mass estimated on the scaling relation (Giodini et al. 2009). All 29 HIFLUGCS clusters illustrated a flat tail in the line-of-sight (los) velocity dispersion profile. By Bayesian statistics, have obtained the MVDR in galaxy cluster as

log(M(bar)M(*)) = 4.1 ± 0.4 (log σ(los)/km s^−1) + 1.6 +1.0 −1.3 , (3) 

with a small intrinsic scatter of σint = 12 ± 3%. In addition, this intercept implied a consistent acceleration scale g(‡) of the CLASH RAR. Consequently, the success of an MVDR in galaxy clusters raises the same issue in BCGs. 

Although the kinematic scaling relations were widely studied in BCGs such as the Faber-Jackson relation and the fundamental plane (e.g., see Oegerle & Hoessel 1991; Zaritsky et al. 2006; Bernardi et al. 2007; Samir et al. 2020), it has never been clear if there is continuity in a single MVDR from the scale of individual BCGs to clusters of galaxies (Sanders 1994). As for the flat profiles of BCGs, it has been a lack of systematic studies of the velocity dispersion profiles. Besides, there is only one BCG per cluster, and sufficiently sensitive, spatially resolved observations have been rare until now.

The discussion section of the paper explains: 

4. DISCUSSIONS 

For the first time, we reveal a tight empirical kinematic correlation, i.e. MVDR, on the BCG-cluster scale. The MVDR is a counterpart of a dynamical relation rather than a coincidence, which can be derived by the CLASH RAR (Tian et al. 2020). As a new discovery of a strong correlation, the MVDR can provide a crucial test for the dark matter problem. 

4.1. Consistency with the CLASH RAR 

The CLASH RAR derive three implications of the kinematics in BCGs and clusters (Tian et al. 2020, 2021): (1) the flat velocity dispersion profile in BCGs; (2) the flat velocity dispersion profile in galaxy clusters; (3) the MVDR on the BCGcluster scale as σ^4 ∝ GM(bar)g(‡). Initially, the implications in galaxy clusters were first confirmed in Tian et al. (2021) for (2) and partially (3), see Equation (3). Subsequently, our studies examine the rest of (1) and (3) in MaNGA BCGs and validate all the implications of the CLASH RAR. In addition, the MVDR on the BCG-cluster scale is identical to that in HIFLUGCS clusters, see Equation (7) and Equation (3). 

A larger acceleration scale also determines a smaller scale length r(‡) of a flat velocity dispersion profile. In pressure supported systems, Milgrom (1984) defined a scale parameter as r(†) = σ^2 (los)g^−1( † ), for a flat velocity dispersion. When calculating with g(‡) instead, we found r(‡) = (0.4 − 4.1) kpc in MaNGA BCGs. Moreover, this scale is much smaller than a typical R(e) ≈ 30 kpc (e.g., see Tian et al. 2020), which indicates a flat velocity dispersion even in the innermost region of BCGs. 

Besides all the success of our results, one exception in our samples needs to be stressed. BCG ‘8943-9102’ displays a declining profile and contributes the smallest velocity dispersion among the sample. This exception raises an interesting issue on BCGs: whether it appertains to a flat profile inherently. 

4.2. Implications for Dark Matter Problems 

The ΛCDM model implied the slope of three for the MVDR on the BCG-cluster scale, by assuming a constant baryon fraction (e.g., see section 4.2 in Tian et al. 2021). According their studies, the prediction in the ΛCDM model demonstrated a discrepancy for smaller galaxy clusters. Coincidentally, MaNGA BCGs also disfavors the ΛCDM model with the same offset. On the contrary, the BFJR of elliptical galaxies were explained by adopting the abundance matching relation (e.g., see the discussions in Desmond & Wechsler 2017; Navarro et al. 2017). Regardless, it still remains a mystery of such an explanation on the BCG-cluster scale. 

MOND naturally explained the slope of four for the MVDR on the BCG-cluster scale, although a larger acceleration scale g(‡) needs to be explained. One conceivable interpretation is the acceleration scale depending on the depth of the potential well (Zhao & Famaey 2012; Hodson & Zhao 2017). Besides, two characteristic scales could imply an underlying phase transition mechanism behind. Nevertheless, our discoveries provide a strict test for the attempts on the fundamental theory in MOND paradigm.

Deur explains the difference between the scaling relation in galaxies and the scaling relation in galactic clusters as arising from a different shape of the mass distribution in galactic clusters which is closer to point masses, than in galaxies, which is closer to a disk. 

Fissile Rasberries

Physics nerd humor. 

From here (alt-text available at the link).

The Makings Of Hungary

Hungary is one of the most notable examples of a population that experienced elite driven language shift (to the Uralic language of their Magyar conquerors), without experiencing much population genetic (a.k.a. demic) change in the long run. Razib Khan has posted a stunning short history of it.

True wholesale population replacement has been rare in the post-Bronze Age era, when the broad outlines of the genetic makeup of modern Europe were largely established, although the conquering Turks in Anatolia and the Uralic people in Finland both gave rise to more genetic introgression than the Magyars in Hungary. (The non-conquering Jews and Gypsies in medieval Europe roughly contemporaneously with the Magyars in Europe, had even less of a demic impact on the general European population.) 

We know a lot about how this happened because it happened in the historic era recorded by contemporary scribes, and because ancient DNA for Conqueror graves in Hungary rule out alternative hypotheses. 

Linguistic and genetic data show that their Magyar language and people were more closely connected to that of the modern Mansi and Khanty people of Siberia (who combined number under 50,000 speakers today), than to the Finns, Saami, Karelian and Estonian peoples in the vicinity of the Baltic Sea, which split linguistically around 2000 BCE to 1000 BCE (with genetic data favoring a more recent date, around the time of Bronze Age collapse, than linguistic estimates).

But, the Magyar language also borrowed words from Iranian and Turkish peoples (probably in that order), who were also nomadic pastoralists of the Eurasian steppe, the most western extent of which includes Hungary.

Prior to their entry into Europe just before 900 CE, they were subordinate partners in coalitions with Iranian and then Turkish nomadic pastoralists. The Huns, who harried the late Roman Empire, in contrast, were linguistically and ethnically Turkish. Notably:

The first attestation of Magyars refers to their service as mercenaries under the Bulgar Khan in 831 AD. From this date, historians can trace the migration of the Magyars westward in annals of Byzantine historians, until they enter Europe just before 900 AD. Despite their coexistence with Iranian and Turkic people for thousands of years, and a strong cultural imprint of both of these groups on the Magyars, they somehow managed to preserve their ethnolinguistic identity. The same cultural continuity persists down to the present. While the ancient Bulgarians spoke a Turkic language, rather than the Slavic speech dominant there today, the modern Hungarians speak a language descended directly from pastoralist Magyar forebears.

Like most conqueror peoples over time, and certainly all Uralic conqueror peoples. the Uralic contribution to Hungary, in addition to being small, was male dominated with invading men marrying local wives, as shown by comparing uniparental genetic markers (Y-DNA comes from fathers, mtDNA comes from mothers):

[G]raves of elite Magyars, the Conquerors, dated to the period between 1000 and 1200 AD. About 30% of the Y-chromosomal lineages in these graves were East-Eurasian haplogroups, where we see fewer than 5% among modern Hungarians. The mtDNA from these samples shows a substantial East-Asian origin, nearly 40%, as opposed to the 1% in modern Hungarians. . . . Even the Siberian Mansi and Khanty are only 25%-33% East Eurasian in their mtDNA haplogroups, whereas medieval Magyars were 10-20%.

 

The only notable exception to that rule that comes to mind is Japan, where indigenous fisher-gatherer Jomon source Y-DNA is found in a very substantial proportion of the modern Japanese population, while Jomon source mtDNA is much less frequent (despite the fact that there are almost no traces of the Jomon language in the modern Japanese language).

Also notable is the the survival of pagan beliefs to a late date, and even into the modern era, more closely tracks mtDNA than Y-DNA, among Uralic peoples. This lines up with the evidence from psychology and sociology that women tend to be more religious than men.

The demise of the Magyar elite, and the survival of the Magyar language can probably be attributed to a couple of key factors. 

First, the Magyar elite swiftly converted to Western Christianity and allied itself with Western Europe, after a half century period of raiding across Western Europe all of the way to Spain, the boot of the Italian Peninsula, and Belgium, against weak, disorganized, divided and undisciplined local feudal nobles. 

After they converted, they joined the Crusades, defended Europe from the Mongols (first losing in 1241-1242 CE only to win against a second round invasion in the 1280s), and then defended Europe from the Ottoman Muslims. 

Second, the Magyar elite actively participated in these wars and their own internal civil wars, not infrequently suffering catastrophic defeats.

In 1241 and 1242, led by the general Subetai, the Mongols ravaged Hungary. At the Battle of Mohi, nearly the whole Hungarian army was slaughtered, up to 10,000 men. In the year after this defeat, as much as 25% of the population may have died due to the chaos ushered in by Mongol units having free rein over the Pannonian plain. . . .

Between 1396 and 1526, the Hungarians fought the Ottoman Turks for supremacy, ultimately losing. At the Battle of Mohacs, 14,000 Hungarian soldiers died, 1,000 nobles were killed, and 2,000 captured prisoners were executed. Hungary was partitioned between the Habsburgs of Austria and the Ottomans.

The rulers of Transylvania maintained some semblance of Magyar independence in their mountainous domain. Nevertheless, it is notable that the first two ruling Houses of Transylvania’s elective monarchy, the Zápolya and the Báthory, were not descended paternally from Conqueror lineages. Rather, they were Croat and German respectively.

The end result was that in terms of genetic distance, the Hungarians are closer to their European neighbors, bear few traces of the Conquerors, and few genetic traces of Uralic populations (another surprising face is that the Swedes bear genetic similarities to the Welsh people, despite the Welsh people have Celtic linguistic affinities that persist to today and the Swedes speaking a Germanic language).

Group in Philippines Sets Denisovan Ancestry Record

A newly genome tested population in the Philippines with the highest levels of Denisovan ancestry of any modern human population (in addition to Neanderthal admixture) tweaks but does not dramatically alter the current paradigm regarding ancient Denisovan admixture.

It does make H. luzonensis a much stronger fossil candidate for Denisovan remains. But, that find doesn't provide all that much to go on (nor do the sources of ancient Denisovan-like DNA).

Excavations in 2007, 2011 and 2015 at Luzon’s Callao Cave yielded a dozen H. luzonensis fossils at first — seven isolated teeth (five from the same individual), two finger bones, two toe bones and an upper leg bone missing its ends, the scientists say.

Denisovans are an elusive bunch, known mainly from ancient DNA samples and traces of that DNA that the ancient hominids shared when they interbred with Homo sapiens. They left their biggest genetic imprint on people who now live in Southeast Asian islands, nearby Papua New Guinea and Australia. Genetic evidence now shows that a Philippine Negrito ethnic group has inherited the most Denisovan ancestry of all. Indigenous people known as the Ayta Magbukon get around 5 percent of their DNA from Denisovans, a new study finds.

This finding fits an evolutionary scenario in which two or more Stone Age Denisovan populations independently reached various Southeast Asian islands, including the Philippines and a landmass that consisted of what’s now Papua New Guinea, Australia and Tasmania. Exact arrival dates are unknown, but nearly 200,000-year-old stone tools found on the Indonesian island of Sulawesi may have been made by Denisovans (SN: 1/13/16). H. sapiens groups that started arriving around 50,000 years ago or more then interbred with resident Denisovans. . . .

Only a handful of confirmed Denisovan fossils exist. Those consist of a few fragmentary specimens from a Siberian cave where Denisovans lived from around 300,000 to 50,000 years ago (SN: 1/30/19), and a roughly 160,000-year-old partial jaw found on the Tibetan Plateau (SN: 5/1/19).

Fossils from the Philippines initially classed as H. luzonensis, dating to 50,000 years ago or more (SN: 4/10/19), might actually represent Denisovans. . . . 

Geographic ancestry patterns on Southeastern Asian islands and in Australia suggest that this region was settled by a genetically distinct Denisovan population from southern parts of mainland East Asia, Teixeira and his colleagues reported in the May Nature Ecology & Evolution.

From Science News. The articles cited are in this story and one related one are:

M. Larena et al. Philippine Ayta possess the highest level of Denisovan ancestry in the world. Current Biology (August 12, 2021). 

doi: 10.1016/j.cub.2021.07.022.

J.C. Teixeira et al. Widespread Denisovan ancestry in Island Southeast Asia but no evidence of substantial super-archaic hominin admixture. 5 Nature Ecology & Evolution 616 (May 2021). 

doi: 10.1038/s41559-021-01408-0.

F. Détroit et al. A new species of Homo from the Late Pleistocene of the Philippines. 568 Nature 181 (April 11, 2019). 

doi:10.1038/s41586-019-1067-9.