search this blog

Sunday, July 7, 2019

Open thread: How did steppe ancestry spread into the Biblical-era Levant?


It's likely that at least two of the Philistines from Feldman et al. 2019 harbor relatively recent steppe ancestry. They're labeled ASH067 and ASH068 in the paper. The former individual is a male who belongs to Y-chromosome haplogroup R1, which appears to be R1b-M269 judging by the data from the relevant BAM file.

This is just the second instance of Y-haplogroup R1 from the pre-Crusades Levant, and, of course, neither R1 nor R1b-M269 appear in the Near Eastern ancient DNA record prior to the expansions of the Yamnaya and other closely related pastoralist groups from the steppes and forest steppes of Eastern Europe.

So how did the Yamnaya-related ancestry spread into the Biblical-era Levant? Did it come via Anatolia, the Caucasus and/or the Mediterranean?

To try and answer this question I analyzed separately the genome-wide data for ASH067 and ASH068 with qpAdm, relying on outgroup and reference populations that weren't featured in the qpAdm runs in the Feldman et al. paper. I also limited the analyses to what were in my view the most proximate two- and three-way solutions in terms of chronology and geography.

The models with the best statistical fits, each labeled with their "tail probs", are available in a zip file here. From my experience with qpAdm, I'd say that the most useful models generally show comparably high tail probs but low chisq values and standard errors. Please note also that I discarded all of the models with at least one standard error higher than 0.2 and/or based on less than 100K SNPs.

As far as I can see, these two are among the very best outcomes. Bell_Beaker_FRA are nine samples associated with the Bell Beaker culture (BBC) from what is now France. Interestingly, the BBC population was rich in Y-haplogroup R1b-M269.

Levant_ISR_Ashkelon_IA1_ASH067
Bell_Beaker_FRA 0.116±0.059
GRC_Minoan 0.507±0.111
Levant_ISR_Ashkelon_LBA 0.377±0.117
tail prob 0.530432
chisq 9.018

Levant_ISR_Ashkelon_IA1_ASH068
Bell_Beaker_FRA 0.237±0.044
GRC_Minoan 0.763±0.044
tail prob 0.943265
chisq 4.736

In my opinion, these models basically confirm that both ASH067 and ASH068 harbor Yamnaya-related ancestry. It's heavily diluted and minor, but it's there. Admittedly, even after looking over the qpAdm output several times, I'm still not quite sure how their ancestors acquired this ancestry. But for the time being, Mediterranean Europe appears to be the most plausible proximate source one way or another. Any thoughts about that? Feel free to share them in the comments below.

See also...

Evidence of European ancestry in the Philistines

R1b-M269 in the Bronze Age Levant

Late PIE ground zero now obvious; location of PIE homeland still uncertain, but...

198 comments:

Alexandros said...

Very interesting! What happens if Minoan is replaced with Mycanean? Apparently Minoan provides a better fit and that's why it ended up in your model, but if the European ancestry in Philistines is coming from the Mediterranean, I would suspect a more Mycenaen-like population as a source rather than a Minoan-like. Would be interesting to see what that does to the model.

Davidski said...

@Alexandros

The Mycenaean sample set caused the marker count to dip below 100K, and this definitely had an impact on the models. The tail probs went up and so did the standard errors.

As a result, I combined the Mycenaeans with the Empuries Greeks, because they basically form a single cluster as far as I can see, and with that the marker count went up to over 100K. But then the tail probs dived.

In fact, none of the models with the Mycenaean/Empuries Greek reference pop produced better fits than the models that I included in the zip file.

Matt said...

For a cross check of these models, using their model proportions in dimension 1-3 of the "West Eurasia PCA": https://imgur.com/a/QQmn1J1

ASH068 is an almost exact match between the models and PCA. Exactly identical in PC1 v PC2.

ASH067 is slightly less of a match; the model proportions from the qpAdm are a bit more "northern" shifted and a bit less CHG shifted compared to the dot on the West Eurasia plot. The model for ASH067 actually matches the positions for ASH2-3 (the sample that looked admixed to me) better than the positions for ASH067 (almost exact match).

Does not mean that the qpAdm models are wrong necessarily; they may be better than the West Eurasia PCA for some reason. This is also assuming the same individuals present in each pop for both qpAdm and West Eurasia PCA.

Davidski said...

@Matt

If you get time, please cross check the other qpAdm models for ASH067 with the PCA.

They may well align a lot better than the model with the French Beakers, because there's no real difference between most of the tail probs and chisq values in the zip file.

Actually, some of the qpAdm models for ASH067 don't include any significantly steppe admixed reference pops, so it might be interesting to see if they produce better results.

Erikl86 said...

I would basically copy paste my post on exactly this matter from Anthrogenica (with minor changes):

What is apparent, is that while indeed the Philistines carried Steppe ancestry into the Southern Levant, it didn't last, and the Steppe-like admixture seen separating MBA Levantines from later Levantine populations, already existing in Roman-era Northern Levantines (as detailed in Haber et al. 2019), didn't come from the Sea People and, doesn't appear to have existed in the Southern Levant by the 10th century BC in any meaningful significance.

This is quite substantial, for several reasons:

1. We now know that the vast majority (the one that actually appear to last to Roman-era and contemporary times) of Steppe-admixture most likely reached the Southern Levant after Iron IIA, which is rather late.

2. This puts the dating post Mitanni or Hittites, which leaves either Assyrians from the Neo-Assyrian Empire era, Persians or later Macedonian/Hellenistic Greeks. Because we now know it already existed among Roman-era Levantines.

3. Perhaps it already existed in the Northern Levant during the Iron IIA, as suggested by the previously released Levant MLBA sample from Tel Megiddo, and only later reached the Southern area. That would require Iron IIA samples from Lebanon. It's unlikely that it happen this way though, because we know the Mitanni and Hittite had already affected culturally the Southern Levant by that time, and given the Mitanni didn't rule directly over the Southern Levant, it must have reached via Northern Levantines moving from North to South or trade. So if they already had Steppe-like ancestry, it would spread with that cultural influence.

This, IMO, makes me believe even more that the best candidate for the Steppe ancestry in all of the Levant would be the Neo-Assyrian Empire, not direct Mitanni or Hittite. I also believe the Persian Empire didn't rule for long enough the Levant nor did it experience with movement of populations to the extent of the Neo-Assyrian Empire. So this leaves mostly two candidates - Neo-Assyrians vs. Macedonian/Hellenistic Greeks. The sheer disparity between just how extensive Iran_ChL ancestry among contemporary Levantines, as well the still mysterious Mesopotamian pull we see in many non-Muslim Levantines, compare to the almost total lack of any visible remaining European Mediterranean admixture in modern Levantines, makes me pretty certain the Neo-Assyrian Empire and it's expansion and long rule over the Near East, proceeded by the Babylonians, is what brought most Steppe ancestry to the Levant.

Drago said...

@ Davidski
Interesting, I tend to agree
BB was moving around Mediterranean even, Adriatic, Ionia.
Shame about coverage, but M269 could even be L151
Maybe Rocca's looked into it

@ Erik

''This, IMO, makes me believe even more that the best candidate for the Steppe ancestry in all of the Levant would be the Neo-Assyrian Empire''

That makes sense. It must surely link, somehow, with ''Ashkenazi Z93' ?

zardos said...

@Eric: But which ethnicity if you are right? I dont know of any large scale immigration or deportation of a people with enough steppe ancestry. There is something you probably miss?

zardos said...

Consider the low amount of steppe ancestry in all Levantine population, this would require near replacement of the preceding population.

Erikl86 said...

@Drago yes, and to my own paternal subclade, Q-M378.

@zardos both Mitanni and Hittite which the Assyrians would have absorbed during the Middle Assyrian Empire, centuries before the rise of the Neo-Assyrian Empire. Also early on during the Neo-Assyrian Empire, they subjugated the Medes, an Iranian people.

Matt said...

@Davidski, sure, here's the top 5 highest P models* for ASH067 - https://imgur.com/a/eG1iHm0

They're all slightly "north" of where ASH067 sits on PC1; the highest p model with Yamnaya_BGR is least similar on PC1, but the most similar on PC2 and PC3 (e.g. could interpret as best matching on East-West and CHG affinity, despite least matching on North-South).

For Yamnaya_BGR model, having a slight excess of "north" fits with the worst f4 stats in dscore: f_4(Base, Fit, Rbase, right2) of:

Positive:

- f4(Base, Fit; Yamnaya_BGR, RUS_Karelia_HG): D= 0.001928, Z= 3.94174
- f4(Base, Fit; Yamnaya_BGR, SRB_Iron_Gates_HG): D= 0.001692, Z= 4.24575

E.g. Fit is closer to Karelia and Iron Gates relative to Yamnaya. Shifted to Euro HG relative to non-Euro HG part of Yamnaya_BGR

Negative:
- f4(Base, Fit; Yamnaya_BGR, Barcin_N): D= -0.00083, Z= -2.24417
- f4(Base, Fit; Levant_ISR_Ashkelon_LBA, SRB_Iron_Gates_HG): D= -0.000751, Z= -2.014734

E.g. Fit is to Yamnaya_BGR relative to Barcin_N and simultaneously closer to Ashkelon_LBA relative to Iron_Gates_HG (possibly too much Ashkelon_LBA ancestry relative to Barcin?).

I would guess there are some similarities in the f4(Base,Fit;X,Y) stats in other models, though because BB France, etc will be closer to BarcinN, they won't be as high in absolute terms, but the patterns should be similar.

*Not including the Unetice EBA model as I assumed it would be minimally different from others and I was getting bored writing them out.

Matt said...

Quick try using the lower P two-way models featuring Minoan+KA Velikent & Minoan+Kaman-Kalehoyuk MLBA (only two way models which worked but got lower ps, about 0.41 for Minoan+KK_MLBA and 0.31 for Velikent): https://imgur.com/a/zvPWqOm

In West Eurasia PCA those look Minoan shifted against the position of real ASH067 (and so slightly north shifted).

Worst f4(Base,Fit;X,Y) stats for KK_MLBA seem to be:

Positive:
f4(Base,Fit;GRC_Minoan, Anatolia_Barcin_N): D= 0.000769, Z= 2.566794
f4 (Base, Fit; Anatolia_Kaman-Kalehoyuk_MLBA, GEO_CHG) D= 0.000908, Z=2.075531
Fit too close to Barcin and CHG relative to Kamen_K and Minoan.

Negative:
f4(Base, Fit; Anatolia_Kaman-Kalehoyuk_MLBA, Levant_PPNB): D = -0.000892, Z=-2.27158
f4(Base, Fit; Anatolia_Kaman-Kalehoyuk_MLBA, Barcin_N): D= -0.000696, Z=-2.143
f4(Base, Fit; Anatolia_Kaman-Kalehoyuk_MLBA, Iberia_C): D= -0.000646, Z=-1.979131
Fit too close to Kamen_K relative to Levant_PPNB, Barcin_N, Iberia_C.

To be honest, I can't really interpret those in, the way that the above stats seemed more obvious relations to the PCA positions. I don't know if those would somehow optimise better in a 3-way KK+Ashkelon_LBA+Minoan model, but I think that wouldn't be feasible to run in qpAdm.

Erikl86 said...

@zardos, it doesn't really mean total replacement of the population.

IMO, and this is something which is brewing in my head for some time now, we're looking at a very complex genetic "pendulum" kind of scenario, reflecting the turmoils of the post Bronze Age collapse and the different waves of migration entering into the Levant throughout the Iron Age. I believe at first, especially in the very Late Iron Age (so 8th-6th centuries BCE), Iran_ChL admixture had risen to be more than the 45-50% that was suggested by Haber et al. (2017) for MBA Levantines and which we repeatedly confirm with Global25 and other tools for both contemporary and Roman-era Levantines. I believe Assyrian, then Babylonian, then Persian penetration and rule over the Levant from roughly the late 10th century BCE all the way to the early 4th century BCE, had brought in both Steppe admixture and additional Iran_ChL admixture. This is btw reflected in these following models, both with Levant_ISR_Ashkelon_IA2 samples, and using Canaanite MBA:

[1] "distance%=2.5263"

Levant_LBN_Roman

Levant_ISR_Ashkelon_IA2,82.8
IRN_Hasanlu_IA,8.8
IRN_Hajji_Firuz_IA,8.4


[1] "distance%=1.9723"

Levant_LBN_Roman

Levant_Canaanite_MBA,87.2
IRN_Hasanlu_IA,7
IRN_Hajji_Firuz_IA,5.8


Then, since the 9th century BCE onward, Arab tribes began migrating and settling in the Levant - this is actually documented by the Assyrians and and other Near Eastern Empires of that period. Those Arabs would have more Levant_N than Iran_ChL, and would resemble the EBA Levant samples from Jordan more than the MBA Canaanite samples:

[1] "distance%=3.7053"

Levant_JOR_EBA

Levant_PPNB,67
IRN_Seh_Gabi_C,33

And contemporary Arabs:

[1] "distance%=4.9655"

Saudi

Levant_JOR_EBA,96.8
TZA_Zanzibar_1300BP,3.2 (* post Muslim East SSA admixture)


So when I try to model Roman-era Lebanese samples with both additional Iran_IA, Hellenistic Greek, AND Arab-like Levantine (with 67/33 balance between Levant_N and Iran_ChL), I get the best distance:

[1] "distance%=1.3653"

Levant_LBN_Roman

Levant_ISR_Ashkelon_IA2,39
Levant_JOR_EBA,24
Iberia_Northeast_Empuries2,17.4
IRN_Hasanlu_IA,11.4
IRN_Hajji_Firuz_IA,8.2

Probably later Ghassanid and other Arabian people which kept on migrating and settling in the Levant during late antiquity (aka Byzantine time) further introduced more Levant_N.

It would be best if we could have:
aDNA from Mesopotamians (Assyrians or Babylonians).
aDNA from Arabs.

To actually test if these models really do represent reality.

Because honestly, if this "pendulum" scenario did occur in one way or another, we are left with extremely enigmatic question as to why Samaritans pretty much overlap with Canaanite MBA, while Roman-era Levantines already show more Steppe ancestry, and why Lebanese do not overlap to the same degree. I mean I just have really hard time accept that Samaritans stayed exactly the same (other than extremely drifted due to severe intermarriage) since 1750 BCE, while other Levantines haven't.

Andrzejewski said...

Both Assyrians and Babylonians has the reputation of deporting populations throughout their empires. We know that 40% of Assyrians were R1a on their y-dna

Andrzejewski said...

@ErikI86 I don’t believe in Dr. Elhaik’s theory regarding the Khazar origins of Ashkenazi Jews, but Q-M378 seems suspiciously Hunnic-derived

Erikl86 said...

@Andrzejewski, it isn't really, because I (and all other Jewish Q carriers, except Yemenite Jews) share an additional down streaming subclade Q-L245. This is a Southwest Asian subclade which has nothing to do with East Asia.

Jews share this subclade with Armenians, Western Iranians and Marsh Arabs (as well as Lebanese Q).

Read Grugni et al. (2014):

https://www.researchgate.net/publication/274510291_Phylogenetic_Structure_of_Q-M378_Subclade_Based_On_Full_Y-Chromosome_Sequencing

Andrzejewski said...

Samaritans are descendants of the Kingdom of Israel, with some Assyrians brought in, mostly women. Maybe that was what elevated their Steppe ancestry.

And didn’t Iran_Chl mixing into Levant_N/Natufians is what created Western Semitic peoples?

Erikl86 said...

@Andrzejewski, where can I see haplogroup of ancient Assyrians? I was unaware it was published.

Andrzejewski said...

So it’s probably an Iran_Chl/CHG derived. Because Ancient North Eurasians (ANE) contributed to both CHG and Iran_Chl populations, just like it did to EHG. In fact, Iran_Chl is ~40%-45% ANE so it’s likely when Iran_Chl admixed with Natufians/Levant_N to create Western Semitic people, ANE y-dna Q came with them paternally

Erikl86 said...

@Andrzejewski I think Q-M378 entered the Levant later, and like R1a points to Steppe admixture.

Btw, Q-M346 also exist in Lebanon (alongside Q-M378), which points to Arabic admixture (for example all Yemenite Jewish Q's are Q-M346). Maybe Ghassanids? Or Nabateans? Or Itureans?

Leron said...

An ancient account from Sudas mentions Sardinians in Crete at the time of a king Minos (relating to a myth of Talos). There are also the enigmatic Elymeans of Sicily that are hypothesized to speak an Anatolian language and were considered to have come from Troy.

If we assume the Sherden of the Sea Peoples were Sardinians and the Peleset a Southwestern Anatolian tribe, it seems to reveal a wide Mediterranean confederation spanning the Nuragic people on the western side and West Anatolians on the east. A clue to this might be the mysterious Piyama-Radu, who could have been a chief of such confederation. He was a Luwian but operated way off into the sea, outside the reach of Hittites and their vassals in Asia Minor that he repeatedly harassed. The Hittite king had to plea to the Mycenaeans to capture Piyama-Radu for them, maybe indicating that his base was in Crete or even further west.

epoch said...

@Erikl86

There was also this R1b from pre-BA-collapse Canaan:

http://eurogenes.blogspot.com/2019/04/r1b-m269-in-bronze-age-levant.html



rozenfag said...

AFAIK, there are no published ancient DNA of Assyrians, Babylonians, Sumerians or anyone from Mesopotamia(Besides some fragmentary mtDNA from Mari in Syria).

Andrzejewski said...

I’d love to find out the dna of the Sumerians v. Ubaidians/Halafians.

Andrzejewski said...

Could it be that the Philistines are like Pelasgians, ie the Non-IE speakers of Minoan/Lemnian before the Dorians hellenized them (Pelasgians)? We know that Mycenaean Greeks were only 20% Steppe with the rest mostly EEF with some CHG shifted?

Erikl86 said...

@epoch,

If you re-read Davidski's recount of that Levant MLBA sample (which I've also mentioned), you'll see he's really not certain about either it's paternal subclade nor it's Steppe ancestry.

Judging by the now available several LBA and IA2 Ashkelon samples from Feldman et al., it appears that no significant Steppe ancestry existed in the Levant at that period and at that area. Even though the Philistines themselves did carry Steppe ancestry, it didn't last, so it seems.

Andrzejewski said...

Erik, did you read a finding by Lazaridis that Israel 6000 years ago was swarmed by Anatolia_N farmers with y-dna T and blue eyes light pigmented skin?

https://www.researchgate.net/publication/327118987_Ancient_DNA_from_Chalcolithic_Israel_reveals_the_role_of_population_mixture_in_cultural_transformation

I believe it’s referred to as the “Pequi’in” cluster, and it’s impact lasted longer in the Northern Levant than the Southern one

Andrzejewski said...

Many religious Jews wouldn’t be happy to hear that, but many Canaanites were assimilated into rather than exterminated by the incoming Israelite IA tribes/confederacy. Turns out that 1/3 of Canaanite kings’ names were Hurrian, namely Kura-Araxes in origin. Jebusites were an offshoot of the Mittani Kingdom with some Indo-Aryan elite plus admixture with Hittites. Therefore, the name “Aravna” was Indo-European. It’s theorized that many rituals practiced by the Jebusites formed the basis for the temple worship in Jerusalem. Solomon himself was half-Hittite. I am sure that’s how the Ancient Biblical Jews have acquired their Steppe and EEF admixture.

Mark said...

It has been suggested in theories dating back to 1846 that the Philistines were the same as the "sea people" who ravaged the Mediterranean:

https://en.wikipedia.org/wiki/Philistines#Egyptian_inscriptions

Who were the sea peoples? It always seemed logical to me that since the Urnfield Culture was busy ravaging Europe at the exact same time that they would make and obvious candidate. They expanded to the tip of the Italian peninsula, so the idea of them grabbing some boats from their conquered victims "grand-theft-auto" style and continuing their rampage across the sea is plausible:

https://en.wikipedia.org/wiki/Urnfield_culture#Migrations

Two unique identifiers of the Urnfields: 1. cremation and 2. Y-haplogroup R1.

The Philistine site showed: 1. cremation and 2. R1 Y-haplogroup. Nobody else practiced cremation in the Levant at that time.

Also, if I remember correctly from the earlier archaeological report on the Philistine site the use of cremation lasted only a couple generations which coincides with the recent paper stating the genetic impact of this group not lasting in the Levant.

Now that you have shown that the Philistines have a Bell Beaker genetic affinity that would make perfect sense since the Urnfields were descended from BBC.

So it all matches up.

PF said...

I'm getting flashbacks to analyzing the Anatolian BA samples: specific Steppe seems vague, while what the models want is more CHG. And same as I thought then, that if there is steppe in these samples, the most likely source involves the Balkans.

Recreating the qpAdm from the post with scaled G25:

[1] "distance%=4.4382"

Levant_ISR_Ashkelon_IA1:ASH067

GRC_Minoan_Lassithi,50.2
Levant_ISR_Ashkelon_LBA,45.4
Bell_Beaker_FRA,4.4


[1] "distance%=4.0005"

Levant_ISR_Ashkelon_IA1:ASH068

GRC_Minoan_Lassithi,79.6
Bell_Beaker_FRA,16.2
Levant_ISR_Ashkelon_LBA,4.2

I get slightly better fits using Peloponnese_N_outlier than with Minoans. A bit more CHG, and of course CHG-shifted farmers in Neolithic Greece is already a clue. Bell Beaker drops with this one.

[1] "distance%=4.1175"

Levant_ISR_Ashkelon_IA1:ASH067

GRC_Peloponnese_N_o,60.6
Levant_ISR_Ashkelon_LBA,38.4
Bell_Beaker_FRA,1

Fit very slightly worse for ASH68, but Ashkelon_LBA disappears completely:

[1] "distance%=4.1033"

Levant_ISR_Ashkelon_IA1:ASH068

GRC_Peloponnese_N_o,87
Bell_Beaker_FRA,13

For reference:

[1] "distance%=3.4012"

Peloponnese_N_o

Boncuklu_N,73.8
Levant_N,16.8
Seh_Gabi_LN,4.8
CHG,4.6

Using this same test of basic Neolithic components, ASH67 doesn't get any EHG, and ASH68 gets only 2.2%:

[1] "distance%=5.8416"

Levant_ISR_Ashkelon_IA1:ASH067

Boncuklu_N,58.2
Levant_N,16.8
Seh_Gabi_LN,13.4
CHG,11.6

[1] "distance%=3.465"

Levant_ISR_Ashkelon_IA1:ASH068

Boncuklu_N,79.6
Levant_N,6.4
CHG,6
Seh_Gabi_LN,5.8
EHG,2.2

At least using G25/nMonte, anything similar to Bell_Beaker_France works just the same. For example I tried to find something with the Balkans and here Varna_outlier produces basically the same fits:

[1] "distance%=4.097"

Levant_ISR_Ashkelon_IA1:ASH067

GRC_Peloponnese_N_o,58.8
Levant_ISR_Ashkelon_LBA,38.6
BGR_Varna_En3,2.6


[1] "distance%=4.4067"

Levant_ISR_Ashkelon_IA1:ASH068

GRC_Peloponnese_N_o,87
BGR_Varna_En3,13

Just like for many of the Anatolia_BA samples, I feel we're missing something...

Bob Floy said...

@Davidski
"none of the models with the Mycenaean/Empuries Greek reference pop produced better fits than the models that I included in the zip file."

So then they probably weren't literally Greek, but something close by and similar.

Samuel Andrews said...

Everything doesn't have to be looked at from Steppe-perspective. The biggest news for Philistine DNA some of them have southeast European ancestry which of course includes minor Steppe admix.

Bob Floy said...

@Andre
"Solomon himself was half-Hittite. I am sure that’s how the Ancient Biblical Jews have acquired their Steppe and EEF admixture."

You have an awful lot of faith in the literal truth of the Bible ;]

Andrzejewski said...

@Sam "Everything doesn't have to be looked at from Steppe-perspective. The biggest news for Philistine DNA some of them have southeast European ancestry which of course includes minor Steppe admix."

It matters if we are to judge whether they spoke a Greek-like IE language or a non-IE Minoan or Lemnian Pelasgian one.

So far, the few words/names/toponyms in the Old Testament betray a West Anatolian/Luwian speaking ethnos, such as "Goliath"=Aliathes and "Seren"=tyrannos.

My money is that Philistines are nothing but uprooted refugee Luwian-speaking Trojans.

Andrzejewski said...

@Bob Floy "@Andre
"Solomon himself was half-Hittite. I am sure that’s how the Ancient Biblical Jews have acquired their Steppe and EEF admixture."

You have an awful lot of faith in the literal truth of the Bible ;]"

I have almost ZERO faith in the literal truth of the bible, alas some events may or may not have resonance in history/archeology.

For instance, "Abraham"'s migration from Haran, a Kura-Araxes center, with lots of Book of Genesis customs being actually Hurrian-based, non-Semitic. Other facts were the presence of Hurrians, Hittites and Philistines who are probably mostly ANF/EEF with some Steppe and non-Steppe-CHG.

The bible also talks about tales of deportations of Judeans and Samaritans into Babylon, with some exiles coming back and some deportees from other fringes of the Assyrian/Babylonian empires forcibly moved in. Moreover, Ezra and Nehemiah were talking about some futile attempt to force Jews who married neighboring countries wives (Moabites, Ammonites, etc) to divorce their spouses and get rid of their mixed marriages' offsprings.

I'm aware that the OT (and the NT!) have a clear agenda, and that they were edited centuries after purported events, but I'm attempting to find resonance of historical events to justify how Jews have 20% Yamnaya (would give lots of Stormfront posters a cardiac arrest!) and almost 30% EEF even before migrating into Europe approximately 2,000 years ago and founding the "Ashkenazi" lineages.

Andrzejewski said...

@Bob Floy "So then they probably weren't literally Greek, but something close by and similar."

My bet is on Luwian refugees from the Trojan war, speaking an IE Anatolian language.

Bob Floy said...

@Andre
"I have almost ZERO faith in the literal truth of the bible, alas some events may or may not have resonance in history/archeology."

But you take the Uriah the Hittite story literally? I'm not saying that there can't be some truth to it, but I wouldn't rely on it for figuring out a problem like this, just saying.

"My bet is on Luwian refugees from the Trojan war, speaking an IE Anatolian language"

Maybe not actual Trojan refugees, but I'm inclined to agree that the Philistines/"Sea peoples" were Anatolian speakers from the west coast, or maybe Phrygians. That's the best working theory, I'd say.
There are users here who seem to want to dismiss that, but those users just enjoy being pedantic, and don't seem to have any better ideas.

Drago said...

@ Bob

Sometimes; all one can do is tell what’s probable and what isn’t
As I’ve said before ; Sea Peoples were a constellation of groups
It might be frustrating, but it’s better than bible stories and pet theories you and Andre like to splash around in

Bob Floy said...

@Drago

I can't see what's wrong with discussing what's plausible, also.
A real ethnic group made their way to the shores of modern day Palestine, and they didn't come from nowhere. Maybe they were of the same blood as the group(s) who caused the trouble in Egypt and elsewhere, maybe they weren't, I'm not saying that we know who they were, I'm speculating within reasonable boundaries. Whoever they were, they were real people, not an abstract blob of data. So it's hard to see what's so wrong with naming likely candidates, as long as one isn't dogmatic about it, like some are. You think for some reason that I'm making dogmatic assertions, but I don't have a "pet" theory that I'm not prepared to throw right in the trashcan, if it dosen't fit with new evidence in the future.

FrankN said...

Two notes - both of them rather meant as "food for thought" than as final explanation:

1. Among the seven "Sea Peoples" recorded by Old Egyptian sources, two are often linked to the W. Mediterranean, namely the "Sherden" ~ Sardinians and "Sechel" ~ Sicilians. From both suspected sources, we still lack MLBA aDNA. Archeology suggests significant "orientalisation" of the W. Mediterranean during the MLBA, e.g. pithos (amphora) burials spreading as far as SE Iberian El Argar B. Pithos burials appear to have originated in the Caucasus - one of the earliest attestations is Areni 1 (ARM_CA). They were common (albeit not exclusive) for Kura-Araxes, and are a/o well attested for pre-Hittite (Hattian?) NC Anatolia, and Minoan Crete. As such, if "Sherden" and "Sechel" really refer to Sardinians/ Sicilians, this could culturally and genetically have meant some back-flow. IOW - MLBA Sicily might theoretically have hosted a mix of (French-like) BB, Minoan, and Levantine-Caucasian-derived populations that could potentially qualify as source of Philistine ancestry.

2. While human "steppe" aDNA apparently didn't substantially affect modern Levantine populations, another LBA/early IA entrant did so, for a different species, namely pigs.

https://www.researchgate.net/publication/258254257_Ancient_DNA_and_Population_Turnover_in_Southern_Levantine_Pigs-_Signature_of_the_Sea_Peoples_Migration
"Unexpectedly, wild boars from Israel have the DNA sequences of European wild boars and domestic pigs. To understand how this anomaly evolved, we sequenced DNA from ancient and modern pigs from Israel. Pigs from Late Bronze Age (until ca. 1150 BCE) in Israel shared haplotypes of modern and ancient Near Eastern pigs. European haplotypes became dominant only during the Iron Age (ca. 900 BCE). This raises the possibility that European pigs were brought to the region by the Sea Peoples who migrated to the Levant at that time. Then, a complete genetic turnover took place, most likely because of repeated admixture between local and introduced European domestic pigs that went feral.

Replacement of Near Eastern (ANF/EEF) by European pig aDNA commenced during the MN East of the Rhine. It is well documented for German TRB, from other candidate regions (Hungary, Poland, Baltic States) pig aDNA hasn't yet been analysed. The first Central European MN culture that shifted from cattle to pigs as dominant livestock was GAC, following earlier Pontic examples (Darkveti-Meshoko, Crimea, Varna-Gulmenita), whereby the sparse pig aDNA available from Gulmenita still suggests prevalence of Anatolian/Armenian pigs there.

As such, Levantine early IA pig aDNA points backwards to an ultimately Central European connection that could also have transmitted human "steppe" aDNA (not GAC, still without such ancestry, but CW/BB/Unetice a.t.l.). The route is still unclear. For pigs, a trail via the Carpathian EBA Wietenberg Culture into MBA Armenia and E. Anatolia has been evidenced (Srubnaya as possible intermediary), but a Balkans route to W. Anatolia and ultimately the Levante may equally be considered.

Davidski said...

@All

Are there any good, mainstream papers online about the potential language spoken by the Philistines?

Bob Floy said...

@David

This is interesting.
https://www.academia.edu/26654209/King_Taita_and_His_Palistin_Philistine_State_or_Neo-Hittite_Kingdom

Davidski said...

Did the Neo-Hittite states speak Hittite?

Labayu said...

@Davidski

Regarding the Philistine language:

Philistine Names and Terms Once Again: A Recent Perspective

The etymology of Goliath in the light of Carian pn Wljat/Wliat: a new proposal

Of course the only language we have a record of them speaking is a local Canaanite dialect, but the names and potential loanwords give some clues as to what they may have originally spoke.

Bob Floy said...

@Dave

They spoke various languages descended from Luwian, IIRC.
Lycian and Lydian are the first which come to mind, there are others that I'm less familiar with.

Bob Floy said...

The neo-Hittite states, that is.

Labayu said...

The administrative language of the Neo-Hittite or Syro-Hititte kingdoms was Luwian, but the populations also spoke Northwest Semitic languages.

Philistine sərān has been connected to tyrannos. However, I think it may be a Northwest Semitic cognate of Akkadian šarru (king) and Hebrew śar (prince, chieftain, commander). The Ugaritic cognate is śrn (prince), more often spelled srn. Which fits with the fact that we now know there had been a Kingdom of Philistia just north of Ugarit prior to their arrival on the southern Levantine coast. The Hebrew Bible spells sərān as srn.

Drago said...

@ Bob

Yes these were real people, not imaginary. The implication i understood from your comment (''pedantic'') was as if attention to detail is a bad thing. At this point of time, that's exactly what we need.
Needless to say; the cline linking these people from the Levant to Southern Europe (& beyond) is self evident; along with collateral evidence. Hence I’m confused by some of the reticence to accomodate this

Bob Floy said...

@Drago
"The implication i understood from your comment (''pedantic'') was as if attention to detail is a bad thing."

Of course I'm not implying that attention to detail is a bad thing, good lord. You keep reading my posts wrong. If you look at the language I'm using, it's very non-committal, I'm not saying anything is definitely this or that, I'm throwing out possibilities that I think make fairly good sense. Any one of these(or all of them) could be wrong, I'm not married to any of them. It's not even a sure thing that the Philistines were part of the "sea peoples" phenomenon. I *suspect* that the Philistines were Luwian speakers.

Bob Floy said...

I also *suspect* that there was never a ton of steppe ancestry in Anatolia, my bet is that when we get more samples from the Hittite period we're only going to see small amounts, probably with a little more in the royal Nes samples, if we're lucky enough to get a look at them.

Andrzejewski said...

@Bob Floy Don’t tell me that you agree with Reich and BROAD that PIE came from South of Caspian...

Bob Floy said...

@Andre

Good God no :S

I just don't think that Anatolian was brought by a massive invasion or anything like that, probably a modestly sized elite which for whatever reason was able to extend it's influence across a large area. How they got there in the first place is another question, as is their exact relationship to the rest of the IE family.

No, I definitely don't agree with Reich and Co. on that issue, in fact I think what they're doing is kind of bizarre. Makes me wonder.

Erikl86 said...

@all,

The connection between Northern Palistin or Walistin, the Syro-Hittite, Luwian-speaking kingdom that existed in Northwest Syria in the period of 11th-9th centuries BCE, and the South Levantine Philistia, has never been proven, and is extremely controversial.

If anything, the connection between Philistines and Aegean people is much stronger, for example the early use of Cypro-Minoan script as well as typical Mycenaean-styled artifacts.

When examines one of the archaeological remains from the Syro-Hittite Walistin/Palistin, there is no evidence from these artifacts found at sites like at Tell Tayinat, either pictorial nor philological, to indicate a link to known Aegean civilizations, as opposed to the Philistines.

I don't know if the Philistines were actually Mycenaean per se, but it could be they have been from either Cyprus or Crete, perhaps Minoans with heavy Mycenaean influence.

And in any case, trying to find similarity between Walistin and Philistines is inherently flawed, because the name Philistines is an english translation derives from the Greek translation of the Hebrew (or Canaanite, as Hebrew was a dialect of Canaanite) word "Plishtim" which means "invaders" or "foreigners", while the name Walistin is comes from the Luwian language and was used by the local population itself to name their kingdom.

Davidski said...

The ancient samples from these papers might end up being useful in figuring out where the Philistines came from.

Ancient island hopping in the western Mediterranean

Bob Floy said...

Wow, those Balearic Island samples sure do have a lot of steppe. Mallorca_EBA looks like an Irishman.

Labayu said...

I have to disagree with Erikl86 on a couple points. The archaeological connection between Syro-Hittite Pilistin and the Philistines is actually quite strong.

Philistine Bichrome is basically locally manufactured Late Helladic IIIC. For Cypriot, Cilician, and Levantine finds, Late Helladic IIIC has become the preferred term over Mycenaean IIIC on the basis of it being a distinct Eastern Mediterranean tradition parallel to a regionally specific Mycenaean IIIC tradition.

In fact, something like 90% of the pottery from Ta’yinat Phase N is locally manufactured Late Helladic IIIC, decorated in motifs which first appear in Western Anatolia during the Late Bronze. The closest parallels to Tay’inat Phase N pottery are found in Philistine assemblages. Therefore, the archaeological connection between Northern Levantine Pilistin and Southern Levantine Philistia is unambiguous.

See also: Aegean Style Pottery in Syria and Lebanon during Iron Age I

Regarding the name, Biblical Hebrew for Philistia is plšt, which is pəlāšet with the Masoretic pronunciation. The Masoretic vowel pointing dates to something like the Eighth or Ninth Century CE, so we can’t really assume it represents the original pronunciation. I’m fairly certain there is no known Semitic etymology for plšt. If anyone has read otherwise, I assume it was simply someone’s speculation. Medinet Habu refers to the “the people of the land of plst”. So comparing consonants, we get the following:

Plstn = Luwian
Plšt = Hebrew
Plst = Egytian

For all I know, the final letter in Luwian is grammatical, but then I don’t know much about Luwian.

Bob Floy said...

@Labayu

It would be a really absurd coincidence if there was no connection.

idurar said...

Steppes signals in ancient samples from Mediterranean Islands might also help to explain why Berbers have minor steppes-related ancestry.

Drago said...

Bob
I see; all good.

Grey said...

Samuel Andrews said...
"Everything doesn't have to be looked at from Steppe-perspective."

true but there's also indirect effects like if PIE were making a nuisance of themselves along the northern and western edges of the Black Sea then maybe some of the people in those regions decided to move away?

and if there was a group like that who already made their living with boats maybe they moved away by boat?

Erikl86 said...

@Labayu, I don't disagree, the connection can be found, but it seems like some here have taken this almost as a fact, as if its consensus - it really still isn't. Far from it, while there is some evidence that link the two, it's still not strong enough as other epigraphic and material findings regarding the Philistines and their almost certain Aegean origin.

Visiting the Philistine museum in Ashdod few months back, the overwhelming consensus right now is still of Aegean origin, not Luwian. If anything, the cultural osmosis between Late Helladic Greeks and Anatolians can be responsible for some similarities which can be found between the two.

As for the name... while in the Aleppo inscription they did find the name Palistin, in the Shaizar and Meharde inscriptions the names Wadasatini and Padasatini are preferred.

And as for the difference between the artifacts and architecture between Palistin and Philistia - most of the discoveries at Tell Tayinat indicate a typical Luwian state. For example, the Syro-Hittite inhabitants used predominantly red slipped burnished ware, which is totally different from the Aegean-type pottery used by the early farming inhabitants. And secondly, the names of the kings of Palistin and the kings of the successor state of Pattin are also Hittite, even though there is no evidence of a direct link between Taita and the old Hittite royal house.

If indeed Palistin and Philistine has similar etymological origin, then I would suggest that perhaps some Philistines also settled also among Luwians, also bringing with them more Aegean-like Late Helladic IIIC pottery, and then got assimilated into the local indigenous Luwian population, similar to what happened to the Philistines and the local Canaanites in Philistia, giving birth to a Neo-Hittite state which has retained a similar name.

Perhaps, the origin of Philistines is the etymologically similar Aegean people called Pelasgians.

Labayu said...

@Erikl86

Luwian origin and Aegean origin aren’t mutually exclusive propositions. There is both archaeological and linguistic evidence pointing to an origin in southwestern Anatolia, which is along the eastern coast of the Aegean where Luwian was also spoken. The parallels between the material culture of Syro-Hittite Palistin and Philistia suggest either a common origin for Palistin and Philistia or that the Philistines arrived in the southern Levant by way of the northern Levant. Which interpretation seems more plausible partically depends on what chronology one subscribes to. By the Low Chronology, the Philistine material culture doesn’t appear in the southern Levant until about 1125. Jeff Emanuel’s article that Bob Floy linked addresses some of your objections regarding the Palistin issue.

A couple of bits of relatively recent information worth considering:

1) Philistine lion-headed cups now appear to be Anatolian/Northern Syrian in origin.

2) Aegean-like material culture in Cilicia was more prominent than in Philistia.

Everything I’ve said aside, there is nothing that necessitates a single origin.

Tropichighlander said...

@Leron It's a plausible hypothesis. Giving that material finds in Crete, Cyprus and in the Western Mediterranean, especially in Sardinia, support the idea of a strong and direct connection between the Western Mediterranean islands and both Crete and Cyprus during the late Bronze age that seems to have continued even after the Bronze age collapse.
As for the Philistines, they were a separate group from the Sherden or Shekelesh and their material culture is a mix of Aegean, Cypriot and Anatolian features. Unfortunately the only sea peoples' group whose settlements we know for certain are the Philistines with their well known cities such as Ekron, Gath and Ashkelon, and possibly Dor for the Tjekker. While we don't know any settlement or site belonging to either the Sherden, Shekelesh, Tursha, Ekwesh or Weshesh. If we knew some sites belonging to those peoples then the sea peoples' enigma would have already been unveiled.

andrew said...

The history strongly supports a scenario in which the Philistines are Mycenaean Greek Sea People migrating to the Biblical-era Levant from the greater Aegean Sea, with the Mycenaeans themselves being Indo-European conquerers of Greece who admixed with the local derived proximately from Indo-Europeans in the Balkans not more than a century or two earlier and possible a little as a few decades earlier, who in turn derived from Indo-Europeans from the Pontic-Caspian steppe.

Since we know that the Mycenaean Greeks were Indo-European language speakers and that the Philistines of Ashkelon were probably culturally Mycenaean Greek and probably spoke the same language, it is notable that some of the Philistines were Y-haplogroup R1b-M269 which was common in the Bell Beaker culture (but, if I recall correctly, also not that uncommon among Minoans).

If the Philistines were linguistically Indo-European, and are genetically a mix genetically of Bell Beaker and Minoan people, it would seem that this would be some of the strongest evidence that the Bell Beaker people were indeed Indo-Europeans (the more common view) rather than linguistically Vasconic (the most popular minority view).

Gaska said...


Well, I suppose that uniparental markers will have something to do with this story, and as far as I know neither the Sardinians, nor the Minoans, nor the Mycenaeans, nor the Anatolians are precisely peoples where R1b-M269 abounds. Or we have not yet found them, or maybe we should look elsewhere. Someone has verified the possible coincidences of the mitochondrial lineages of the Philistines?

Surely everyone can think where this lineage abounded in the Bronze Age and had access to the routes of the eastern Mediterranean (Italy, Sicily where we have seen R1b-Df27, Iberia, southern France?). I really do not understand the obsession with steppe ancestry.

andrew said...

@Mark

"the Urnfields were descended from BBC."

Evidence?

Źródełka said...

https://m.scirp.org/papers/69428
The Philistine Inscription 4.5 from Ashkelon (Israel)

Ryan said...

Andrew - Minoans had R1b-269? Source?

Bob Floy said...

@Andrew

Is it opposite day where you are?

Alexandros said...

Very interesting linguistic article you shared. Here is the overall conclusion for everyone. Unfortunately inconclusive.. "Rather than searching for a single Philistine identity, language, and origin, we have tried to show that the Philistines used a variety of scripts and languages for a number of different purposes as part of the construction and ongoing negotiation of their identity/identities, processes which might be described as transcultural and entangled"

Alexandros said...

Some evidence below for Philistines connection with BA Cyprus. Appears to be a news article but was not able to locate the original archaeological report so far.

https://www.haaretz.com/archaeology/philistine-gath-startlingly-alike-cypriot-cities-1.5428043

andrew said...

@Bob Floy

I think my memory is referencing Martinez, "Paleolithic Y-haplogroup heritage predominates in a Cretan highland plateau" (2007), https://www.nature.com/articles/5201769 which was actually based on modern Cretan DNA but contained a strong inference that it was pre-Mycenaean because of the relative paucity, at least at the time, of R1b in the rest of Greece and the predominance of R1a in other Indo-European populations in Central/Eastern Europe and India at the time. From the abstract:

"In the present work, the geographic stratification of the contemporary Cretan Y-chromosome gene pool was assessed by high-resolution haplotyping to investigate the potential imprints of past colonization episodes and the population substructure. . . . Comparisons of Y-haplogroup frequencies among three Cretan populations as well as with published data from additional Mediterranean locations revealed significant differences in the frequency distributions of Y-chromosome haplogroups within the island. The most outstanding differences were observed in haplogroups J2 and R1, with the predominance of haplogroup R lineages in the Lasithi Plateau and of haplogroup J lineages in the more accessible regions of the island. Y-STR-based analyses demonstrated the close affinity that R1a1 chromosomes from the Lasithi Plateau shared with those from the Balkans, but not with those from lowland eastern Crete. In contrast, Cretan R1b microsatellite-defined haplotypes displayed more resemblance to those from Northeast Italy than to those from Turkey and the Balkans."

and from the body text: "Based on the age of the R1b-associated Y-STR variation for the Crete-without-Lasithi-Plateau population, the genetic affinity between R1b haplogroups from North Italy and Crete might be the imprints of an Italian gene flow before the end of the Minoan civilization[.]"

Further discussion along the same lines in King RJ, et al., "Differential Y-chromosome Anatolian influences on the Greek and Cretan Neolithic" (2008 Ann Hum Genet 72:205–214). https://onlinelibrary.wiley.com/doi/full/10.1111/j.1469-1809.2007.00414.

A 2010 paper had also ruled out a Paleolithic origin. https://journals.plos.org/plosbiology/article?id=10.1371/journal.pbio.1000285

Was also considering that the Bronze Age Minoan culture appeared in Crete starting around 3100 B.C.E. and seems to represent a pretty major break with the preceding Neolithic cultures of the region which prevailed from about 6000 B.C.E. until asbout 3200 B.C.E.

I've only seen ancient Minoan mtDNA discussions, discussed at http://forwhattheywereweare.blogspot.com/2013/05/ancient-minoan-mtdna.html although with speculation that the part of Cretean highlands where R1b is common is found is an older layers of the palimpsest than the rest of the population of the island, and might be a pre-Mycenaean refugia.

Entirely possible that I've missed more up to date sources of ancient Y-DNA from Crete. But, also not an idea from nowhere.

andrew said...

Looking further the 2017 paper doesn't have any Minoan Y-DNA R1b of any kind, but it also only had three samples of ancient Minoan Y-DNA, two J2a (both from the Lasithi Plateau) and one G2a (from elsewhere in Crete). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5565772/ These are results that could easily show up in a random sample of three individuals even in a modern population. Also the one Mycenaean was also Y-DNA J2a and the only other ancient Y-DNA sample, from Anatolia, was J1a. So, this ancient DNA doesn't shed any light on where R1a-M269 which is now the plurality haplogroup on the Lasithi Plateau came from, in a population that based on mtDNA otherwise looks like a relict population.

Again, the evidence for it isn't terribly strong, but the evidence against it is also very weak, and there is no indication of lots of trade between the Lasithi Plateau and Italy in the Iron Age or at any time since then.

andrew said...

Maju summarize the results in the first paper I cited as follows:

"A clear issue is that the current inhabitants of the plateau have a distinctive genetic signature in their Y-DNA, quite different from that of other Cretans, with much higher frequencies of R1b and R1a and much much lower frequencies of the most common Cretan lineage: J2a1. However they also almost lack the main mainland Greek haplogroup E1b, what suggests that the recolonization from Peloponnese story is not correct either.

Interestingly Cretan R1b, so important in Lasithi Plateau (almost 50%), is also largely derived from Western Europe (although the other half could be Balcanic), maybe via Italy, and cannot be ancestral to it (almost all the Western variant belongs to a derived subclade common in Italy, Central Europe and France: U152)."

Autosomally, the 2017 paper states: "This analysis shows that all Bronze Age populations from the Aegean and Anatolia derived most (~62–86%) of their ancestry from an Anatolian Neolithic-related population (Table 1). However, they also had a component (~9–32%) of ‘eastern’ (Caucasus/Iran-related) ancestry. It was previously shown that this type of ancestry was introduced into mainland Europe via Bronze Age pastoralists from the Eurasian steppe who were a mix of both eastern European hunter-gathers and populations from the Caucasus and Iran4,6; our results show that it also arrived on its own, at least in the Minoans, without eastern European hunter-gatherer ancestry. This ancestry need not have arrived from regions east of Anatolia, as it was already present during the Neolithic in central Anatolia at Tepecik-Çiftlik17 (Supplementary Information, section 2). The eastern influence in the Bronze Age populations from Greece and southwestern Anatolia is also supported by an analysis of their Y-chromosomes. Four out of five males belonging to Minoans, Mycenaeans, and southwestern Anatolians (Supplementary Information, section 3) belonged to haplogroup J which was rare or non-existent in earlier populations from Greece and western Anatolia which were dominated by Y-chromosome haplogroup G21,2,17. Haplogroup J was present in Caucasus hunter-gatherers3 and a Mesolithic individual from Iran4 and its spread westward may have accompanied the ‘eastern’ genome-wide influence."

The PCA shows only a very weak steppe pull for Mycenaean Greeks relative to Minoans. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5565772/figure/F9/?report=objectonly

andrew said...

One could imagine a detour by Bell Beakers in Crete. But, hard to know.

FrankN said...

Alexandros: I find that Maeir e.a. article shared by Labayu, which is indeed worth reading, not inclusive at all. It shows that it may be futile to search for the "single origin" of Philistines, and endlessly debate whether Minoans or Luwians are a better fit. Instead, we need to reckon with all sorts of influences/ populations from various directions - a diversity becoming apparent already from the Feldman e.a. paper, and illustrated further by Dave's apAdm analyses here.

Dave: Looking at your linked runs, it appears that for ASH_068, BB_Hungary provides even a better fit as "Steppe" source than BB_France (I am not that versed in interpreting qpAdm outputs, so I may err in this respect). It reads as follows:

Levant_ISR_Ashkelon_IA1_ASH068
Bell_Beaker_Hungary 0.268
GRC_Minoan 0.732
tail prob 0.974633
chisq 3.830

Have you tried out BA Hungarian samples (Vatya, Maros)? I mean, during the EBA, the major route for trading Cornish tin to the E. Mediterranean ran through Unetice (along the Elbe) and Vatya/ Maros (along the Danube). Seeing their transit business collapse after the Thera eruption, those Danubians would have had good reasons to "check out" what was going on in the Aegean… Otherwise, the a/m replacement of Anatolian by European pig aDNA in the Transsylvanian Wietenberg Culture may indicate an eastward expansion from the Budapest area already during the early 2nd mBC - Wietenberg is otherwise reknowned for Aegean/ Mykenean influence.

Besides, Maros is timewise and geographically the closest sample to the yet uncovered 2nd mBC Balkans (those "Bulgaria_EBA" samples date to the turn of the 4th to the 3rd mBC and are actually Chalcolithic, not BA - someone should correct the labelling).

andrew said...

The distribution of R1b-U152 ancient DNA is generally speaking in Western and Central European Bell Beakers, with ancient DNA appearing in the Mediterranean starting in the Iron Age and far from the Aegean, which only makes the R1b-U152 on the modern Lasithi Plateau and in the Philistines only more mysterious. https://www.google.com/maps/d/viewer?mid=15zJwudMOPqf20VVXXNXkm4qydzTP3VW0&ll=42.13330000000002%2C3.1082999999999856&z=5 This doesn't look like a population that was spreading by maritime means until very late.

Has any more detailed haplogroup typing within R1b-M269 that has been done for the Philistines?

andrew said...

Lots of Y-DNA R1b-M269 in Philistines is a good fit to a large BB_FRA component or a Bell_Beaker_Hungary 0.268 fraction, especially if the BB admixture was male dominated. I like the narrative suggested by @FrankN

The lack of a good fit the Mycenaeans or other Greeks that cluster with them in the data, and the lack of much steppe ancestry in either ancient DNA from Minoans or Mycenaeans relative to what one would expect with someone who is about 1/4 Bell Beaker (despite the fact that the autosomal data sample size is larger than the Y-DNA data, and the fact that a large sample size is much less important for autosomal data which is more representative of the general population than uniparental markers) doesn't make much sense in either the Luwian or the Mycenaean narratives suggested by the relics, fragmentary writing and Egyptian and Biblical attested historical data, however. It is hard to square that circle.

What narrative could explain the disconnect between the material culture and written materials and Philistine DNA?

I suppose that it is possible that the sea people who gave rise to the Philistines were basically BB folk from Central Europe or the Italian highlands, who underwent elite dominance cultural change upon being integrated into a "sea people" culture, in which members with maritime skills, that BB rank and file folks would have lacked, have outsized prestige and cultural influence. Also, as far as I know, there is no indication that the BB folk had a language that existed in written form as of Bronze Age collapse, so any writing would have necessarily been in someone else's language (or at least someone else's script). If the BB rank and file were illiterate and the minority of maritime leaders and people who were literate were Greek or SW Anatolian, that could also contribute to cultural elite dominance of the non-BB folk in the community. The rank and file BB people may very well have been basically disposal and desperate ill fed refugees as the short statute of the Philistines would suggest by the time of Bronze Age collapse, and much more prone to being in a culturally subordinate position than they were in the EBA and MBA when they were conquering the first farmers of Europe.

Perhaps the ancestors of the modern residents of the Lasithi Plateau were a bit like the Basque in that their Y-DNA over several generations came to be R1b-M269 (and more specifically R1b-U152) dominated, but without the huge autosomal impact, perhaps due to the descendants of the modest number of members of an initial BB male wave (possible long before a Bronze Age collapse wave at their cultural fitness peak when they were expanding) marrying local women generation after generation and thereby diluting the BB autosomal impact while retaining a significant Y-DNA impact. It would be interesting to see how much steppe ancestry there is in modern residents of the Lasithi Plateau.

Drago said...

Andrew

“Paleolithic Y-haplogroup heritage predominates in a Cretan highland plateau" (2007), https://www.nature.com/articles/5201769 which was actually based on modern Cretan DNA but ...”


That’s a modern DNA paper ; so be careful with inferences, some of which are problematic

EastPole said...

@Źródełka

„https://m.scirp.org/papers/69428
The Philistine Inscription 4.5 from Ashkelon (Israel)”

This is why I was interested if there was any Philistine-Phrygian connection. Phrygians are related to Thracians and

“Based on archaeological evidence, some scholars such as Nicholas Hammond and Eugene N. Borza argue that the Phrygians were members of the Lusatian culture that migrated into the southern Balkans during the Late Bronze Age
…..
After the collapse of the Hittite Empire at the beginning of the twelfth century BC, the political vacuum in central-western Anatolia was filled by a wave of Indo-European migrants and "Sea Peoples", including the Phrygians”

https://en.wikipedia.org/wiki/Phrygians

After R1a-Z280 was found at the western edge of Lusatian culture there are speculations that it was a Slavic culture.
There are many words and traditions in Prygian, Thracian and Greek religions which are related to Slavic and also Vedic.

Drago said...

Mark

“the Urnfields were descended from BBC."”

In some shape or from everyone in western and Central Europe has BB descent ; but there’s a 1700 year gap between the two. Not sure how much direct cultural filiation we can see; very little in all likelihood. That’s because BB cultural legacy was fairly continuous in Atlantic Europe; whilst Urnfield expanses from the carpathian basin; where such legacy doesn’t exist

Erikl86 said...

Just few models I ran over nMonte:

[1] "distance%=3.705"

Levant_JOR_EBA

Levant_PPNB,67
IRN_Seh_Gabi_C,32.8
Yamnaya_RUS_Samara,0.2

[1] "distance%=1.8653"

Levant_ISR_Ashkelon_IA2

Levant_PPNB,54.2
IRN_Seh_Gabi_C,45.6
Yamnaya_RUS_Samara,0.2


[1] "distance%=2.2759"

Levant_LBN_Roman

Levant_PPNB,50.8
IRN_Seh_Gabi_C,42.4
Yamnaya_RUS_Samara,6.8


While these models are pretty simplistic, they demonstrate nicely that Steppe ancestry arrived to the Levant before Roman-era, but after Iron IIA.

Bob Floy said...

@andrew

"2007"
"Maju"

Yeah, I gotta be honest. Your case isn't looking too good.

Bob Floy said...

@East Pole

You might have to accept that the Philistines don't have anything to do with Slavs.

EastPole said...

@Bob Floy

“You might have to accept that the Philistines don't have anything to do with Slavs.”


Don’t argue with me. Argue with this article:


https://m.scirp.org/papers/69428

Joshua Lipson said...

@Erikl86

Incidentally, that set of models sort of quells a concern I've had about generalizing from Bronze Age North Levantines to Bronze Age South Levantines and vice versa: the ratios of Levant Neolithic : Iran Chalcolithic are identical.

The only caveat is that there's a chance later Phoenician expansions changed the composition of the Ashkelon population over the course of the Iron Age (resettlement from the north?). Could you model LB Ashkelon individuals using the same source populations?

FrankN said...

One author of this paper
https://m.scirp.org/papers/69428

is Giancarlo T. Tomezzoli, astronomist by education, and long-term staff member of the European Patent Office.

http://independent.academia.edu/giancarlotomezzoli/CurriculumVitae

The man has certainly qualifications, but is clearly an amateur when it comes to ancient scripts, archeology, and linguistics. He has published quite a lot on German WWII military installations, and otherwise identified proto-Slavic in various inscriptions, including - aside from this Philistian one - Venetic, early Roman Carinthia, Minoan Crete, and an Ogham inscription from Scotland.

Bob Floy said...

@East Pole

Yeah, the Philistines spoke fucking Croatian and nobody noticed, sure. That guy is a genius.

Drago said...

@ Bob
There’s another guy out there that claims the Minoans and ancient Gauls were also Slavs :)

AWood said...

Mycenaeans can't also be modeled as BBC_France and Minoan? Just curious. If not, is there a way to determine what population is closest that brought the "steppe" part? BA_Iberian?

Mark said...

Andrew & Drago

The Urnfield culture was one of successive Central European bronze age cultures all overlapping the same basic footprint.

As for as the comment "Urnfield expanses from the carpathian basin" I have no idea what you are referring to. Urnfield sites are located in Bohemia, Germany, Austria and Western Poland.

Yes there is a 1700 year gap between BBC and Urnfield. Well, more like 1200 years (2400-1200 BC). Here is the lineage: Urnfield evolved from the Tumulus Culture, Tumulus culture evolved from the Unetice culture, Unetice Culture evolved from the BBC. All cover the same central European areas. Bell beaker brought the same R1b Y chromosomes that is in Central Europe today so the genetic continuity is obvious.

So when I say "descended from" I am speaking in a genetic sense as you are descend from your great-grandfather. Culturally, the Urnfields had many differences from BBC just like central Europeans are culturally different from their predecessors 1200 previous. But they are genetically descended from them.

Bob Floy said...

@drago

I'm sure there is, and I'm really sick of this kind of thing.
These guys will literally believe anything if it means that their particular ethnic group is the center of the universe, ya know, because that's how things work. Jesus Christ.

AWood said...

@andrew

Is there any evidence these specific R1b men were short or small stature though? How do we know the article that cited that the Philistines were small was not referring to the later Iron Age when local Levantine genes became dominant once more in a short period? I'm no religious person, but perhaps some of these non-Semitic groups that were encountered in the Levant were indeed Biblical "giants" so to speak, if you compare someone who is "gracile" 5"2 and to a heavy boned Bell Beaker M269+ guy who is 6"0, he would appear to be a giant.

Drago said...


@Mark

“Here is the lineage: Urnfield evolved from the Tumulus Culture, Tumulus culture evolved from the Unetice culture, Unetice Culture evolved from the BBC. All cover the same central European areas. Bell beaker brought the same R1b Y chromosomes that is in Central Europe today so the genetic continuity is obvious.”

There are important differences.
Unetice did not descend from BB- it was a rather different culture- but has important admixture from BB & CWC.
Urnfields expanded from Transdanubia; where the urn burial tradition is rooted. Tumulus culture, on the other hand, expanded from Germany / France back east.
These are important differences
As I suggested; if we want to look the most direct continuity in post-BB times- it’s in Iberia; west & south France; & British isles

Samuel Andrews said...

@Matt,

Btw, remember when I said it was news that Phoecians were typical Levantie people. I thought Phoecians were supposed to be of southeast European origin. It was the Philistines who I heard were supposed to be of southeast European origin. Now, it turns out they were of partly southeast European & Anatolian origin, not pure Levantie.

Samuel Andrews said...

@All,

I don't think the existence of Steppe admix in the Philistines is significant. It was already widespread in southeast Europe at this time, some Philistines had SE european ancestry, so by default they had some Steppe ancestry.

This is a plausible source for Steppe admix in the Levant region today instead of Iranian ancestry which is assumed to be the source. That is important. But should not be the focus.

Everything on this blog looks at things from a Steppe angle. It is significant for Europe & central Asia but Not for the Middle East.

Indo European, Steppe impact on the bulk of the Middle East is tiny. Even Persians only have 15% Andronovo ancestry.

Davidski said...

@Samuel Andrews

I don't think the existence of Steppe admix in the Philistines is significant.

It's significant and important for two reasons:

1) it corroborates the theory based on multidisciplinary research that the Philistine population derived, at least in part, from outside of the Levant

2) if their steppe admix came from an Anatolian-speaking population, such as from the Luwians, then this obviously has important implications for the Proto-Indo-European homeland debate, because it would suggest that there was steppe admix including R1b-M269 among Anatolian speakers.

Everything on this blog looks at things from a Steppe angle. It is significant for Europe & central Asia but Not for the Middle East.

Yeah, well, this blog is called the Eurogenes Blog, so please keep that in mind.

Samuel Andrews said...

@Davidski,

Fair enough. My personal focus with these samples was who in southeast Europe and Anatolia contributed to Philistines. Its safe to assume they spoke Indo European so to me this was a non issue. I do see how they can contribute to the debate over where Anatolian languages come from.

Leron said...

Be careful about opening this Pandora’s box, or you might just see a lot of diverse ethnicities claiming European heritage (and certain socialized rights) by having just one drop of Steppe blood.

Davidski said...

@Leron

Be careful about opening this Pandora’s box, or you might just see a lot of diverse ethnicities claiming European heritage (and certain socialized rights) by having just one drop of Steppe blood.

LOL


Drago said...

@ Davidski

“if their steppe admix came from an Anatolian-speaking population, such as from the Luwians, then this obviously has important implications for the Proto-Indo-European homeland debate, because it would suggest that there was steppe admix including R1b-M269 among Anatolian speakers.”

It certainly strengthens it; indeed I think some version of a modified Black Sea homeland always looked solid. But the question is just what impact did steppe migrations really have in Mycenaeans & Anatolian IEs : were they the raison d’etre, or “co -travellers” ?

Somewhat interlinked -if there were 4 or so migrations from the steppe (with distinctive cultural sets, often thousand or so years apart), were they all IE?

I think that’s where the question needs to head . Otherwise we’ll just assume that everyone in Europe spoke IE wherever there’s some steppe or R1-something here or there; and everybody else lived on trees until they somehow managed to then de-IE much of Europe

Ryan said...

@David - do your models favour a western source of steppe ancestry or do they not have a preference? I'd think BB and Luwians would be pretty different.

Any hint of Sardinian too?

Erikl86 said...

@Joshua Lipson

Again, here are all the Levantines from EBA to Roman-era (without the IA1 Philistine-mixed):

[1] "distance%=3.705"

Levant_JOR_EBA

Levant_PPNB,67
IRN_Seh_Gabi_C,32.8
Yamnaya_RUS_Samara,0.2

Apparently, IRN_Hajji_Firuz_C offers better results for later periods, IMO:

[1] "distance%=1.9153"

Levant_Canaanite_MBA

IRN_Hajji_Firuz_C,56.2
Levant_PPNB,43.6
Yamnaya_RUS_Samara,0.2

[1] "distance%=2.6546"

Levant_ISR_Ashkelon_LBA

IRN_Hajji_Firuz_C,61.2
Levant_PPNB,38.2
Yamnaya_RUS_Samara,0.6


[1] "distance%=3.1708"

Levant_ISR_MLBA

IRN_Hajji_Firuz_C,52.4
Levant_PPNB,46.4
Yamnaya_RUS_Samara,1.2


distance%=2.1784"

Levant_ISR_Ashkelon_IA2

IRN_Hajji_Firuz_C,55.4
Levant_PPNB,44.6


[1] "distance%=1.8959"

Levant_LBN_Roman

IRN_Hajji_Firuz_C,55.6
Levant_PPNB,40
Yamnaya_RUS_Samara,4.4


^ One can actually see that somewhere in the MBA, there's a trend of rising Iran_ChL admixture, culminating in "Levant_ISR_Ashkelon_LBA", and then in the MLBA sample from Tel Megiddo there's a Steppe-like "anomaly" (~1.2%), which I don't know how to interpret. It's not surprising if indeed that sample is R1b. It quickly disappears in the IA2 Levantine samples though. My feeling here is that perhaps Mitanni and Hittite admixture did penetrate the Northern Levant after the MBA, bringing small amounts of Steppe, which have managed to diffuse slowly all the way to Northern ancient Israel, but didn't reach the southern parts yet. Or perhaps - just didn't reach the coastal areas of the Levant. For that we need, as I was saying few posts ago, IA2 samples from Lebanon, to see if indeed Steppe admixture was missing in the Northern areas of the Levant at that period of time like we see in Ashkelon (Southern Levant).

By the time of the Roman-era samples though, Steppe ancestry is very much present, not much different from contemporary Lebanese.

Davidski said...

@AWood

Mycenaeans can't also be modeled as BBC_France and Minoan? Just curious.

Nope. Total fail.

https://drive.google.com/open?id=19aijHF4PuTMUJR56_mrRl7SF6PQpsVHl

And splitting up the Minoans into Minoan_Lassithi and Minoan_Odigitria doesn't work either.

https://drive.google.com/open?id=1aITvAJrbFjd4EwiKYOridswGCt-Ain3w

https://drive.google.com/open?id=1Rd7iO-VG2xQyS6LCc1ZjpSYLcCIM9kvP

But for whatever reason, the Sintashta-like MLBA individual from Bulgaria does work when used alongside the Minoan_Lassithi set. So the steppe ancestry in Mycenaeans is unlikely to be a collateral signal coming via Mediterranean maritime routes, but rather something that entered the Balkans from the steppe/forest steppe.

https://drive.google.com/open?id=1eynVmyyBIYZsfIANqLmtpzkbhFtgVgMW

If not, is there a way to determine what population is closest that brought the "steppe" part? BA_Iberian?

It's impossible to say, but it may indeed have been something very western and thus derived from Beakers.

Davidski said...

@Ryan

Do your models favour a western source of steppe ancestry or do they not have a preference? I'd think BB and Luwians would be pretty different.

I think they do favor a more westerly source, but it's difficult to be specific. You have to check out the qpAdm models that I posted and see for yourself.

https://drive.google.com/open?id=1GHd2OsVYxvcu3ZDa_VLfvwakL0t2oa6z

But I'd recommenced keeping an open mind until more Copper, Bronze and Iron Age samples from around the Mediterranean are released, like those from Sicily from that recent preprint.

Cpk said...

What does small steppe admix in Philistines have to do with Anatolian IE? There is still no steppe admix in Anatolia before 2500 BC and cranial shape change begins only after 2000 BC. Even if Anatolian languages came to Anatolia from outside it came with a population with already diluted steppe admix.

Źródełka said...

And what precludes the arrival of a pitcher with inscriptions in pra Slavic in the territory where it was found? Maybe it's your brain that can not grasp it?

Davidski said...

@Drago

Otherwise we’ll just assume that everyone in Europe spoke IE wherever there’s some steppe or R1-something here or there; and everybody else lived on trees until they somehow managed to then de-IE much of Europe.

It's just a matter of working out what sort of ancestry can be shown to realistically link the far flung speakers of Indo-European languages in Europe and Central and South Asia, and then also seeing if this type of ancestry is present in at least most of the attested Indo-European-speaking ancient groups.

As things stand, ancestry from the Eneolithic/Bronze Age steppe appears to fit the bill quite nicely.

Of course, even if this steppe ancestry is the PIE marker, then it shouldn't be surprising that it spilled out one way or another into populations not speaking Indo-European languages. And indeed there are now several examples in the ancient DNA record of such groups, such as the Iberians, Etruscans, Uralics and Turks.

Some of these language families may have come from the steppe or surrounds, but if so then there should be some linguistic and archeological evidence for it, and if there is then that's OK.

epoch said...

@Davidski

Which sample is the Bulgarian MLBA?

Grey said...

"it may indeed have been something very western and thus derived from Beakers"

is the other way round possible i.e. the ancestry in Philistines coming from the original *source* region of western BB?

Davidski said...

@epoch

This one...

I2163, Y-hg R1a-Z93, mtDNA U5a2, 1750-1625 calBCE, Merichleri, Kairyaka necropolis

But it's not the only one that works. Yamnaya Bulgaria does as well.

Drago said...

Davidski

“As things stand, ancestry from the Eneolithic/Bronze Age steppe appears to fit the bill quite nicely.”

of course this can be dissected even more specifically. It could be what links IE speakers is actually a steppe / MNE mix ; with the latter higher in south European IE and former in northern. This is even consistent for Indo-Aryan (who have obvious EEF admixture; but of course we are waiting for Narasimhanms final version and it’s acceptabiltu for South Asia).

Davidski said...

@Grey

is the other way round possible i.e. the ancestry in Philistines coming from the original *source* region of western BB?

I'm not sure what you mean? This type of Beaker ancestry wasn't native to Western Europe. It spread there from the steppe via North-Central Europe well after 3,000 BCE, and then moved east with Atlantic admixture via the Mediterranean during the peak of the Beaker period.

Ric Hern said...

It will be an interesting twist if Hittites turn out to be Bell Beaker related....

Davidski said...

@Drago

It could be what links IE speakers is actually a steppe / MNE mix ; with the latter higher in south European IE and former in northern.

Maybe. Even Afanasievo has some MNE admix, probably via Dereivka/Sredny Stog, so even if it's shown that the Tocharians are a direct offshoot of Afanasievo, then a PIE population of mixed steppe/MNE origin in the North Pontic and surrounds might still be a plausible solution.

I wouldn't have any objections to that if everything fits.

Davidski said...

@Ric

It will be an interesting twist if Hittites turn out to be Bell Beaker related.

I think we're in for a surprise in regards to the Hittites and other Anatolians, but that won't be it.

Grey said...

Davidski
"I'm not sure what you mean?"

just wondering about possible directions of travel

1) region x to levant

or

2) region x to west then island hop back east again to levant

A said...

@ Davidski

Is it possible there’s a mix of Bell Beaker and Sintashta-like ancestry in Mycenaeans? Archaeology indicates a Bell Beaker presence in Greece before 2000 BC:

https://www.academia.edu/5429585/_2013_V._Heyd_Europe_at_the_Dawn_of_the_Bronze_Age._In_Transition_to_the_Bronze_Age._Ed._by_V._Heyd_G._Kulcs%C3%A1r_and_V._Szever%C3%A9nyi._Archaeolingua_Main_Series_30_Budapest_Archaeolingua_p._9-66

Davidski said...

@A

You mean the presence of Bell Beaker populations or just Bell Beakers?

If just Bell Beakers, then they may have been imported goods.

A said...

Bell Beaker pottery and other artefacts such as wristguards and toggles. Quote from the paper:

“The best explanation for the relatively late appearance of these Aegean wristguards, and other Bell Beaker related finds, lies with further Adriatic pottery of the Dalmatian Cetina [culture] that also reached the Peloponnese in the later third millennium BC. Joseph Maran has described their background and context in detail (e.g., 1998), and he is surely right in seeing at work a migratory event, bringing Early Bronze Age Adriatic people incrementally to southern Greece for some decades from the transition of early Helladic II to III. And since early Cetina is one of those syncretistic Bell Beaker cultures of its southeastern periphery as shown above, this best explains the manifestation of these Bell Beaker package elements deep in southeastern Europe.”

V. Heyd, 'Europe at the Dawn of the Bronze Age' (2013)

PF said...

@A

Hmm, I was actually thinking of trying something Adriatic-related for these Ashkelon samples. Ever-so-slightly better fit than Bell_Beaker_France in G25, but still not great. Closer to the percentage split Davidski got with qpAdm though.

[1] "distance%=3.9862"

Levant_ISR_Ashkelon_IA1:ASH068

GRC_Peloponnese_N_o,72.4
HRV_MBA,27.6


@Davidski

Is it possible to get the other Minoan samples (not just Lassithi) into the G25?

FrankN said...

https://www.timesofisrael.com/as-archaeologists-say-theyve-found-king-davids-city-of-refuge-a-debate-begins/

Labayu said...

@FrankN

Yeah, Yossi Garfinkel also found the biblical city of Sha'arayim at Khirbet Qeiyafa. See, the way he knows is that his excavation unearthed two gates at Khirbet Qeiyafa and Sha'arayim means “two gates”. Now one might object, noting that it’s not at all unusual for a city to have two gates, or point out that the excavators didn’t actually find two gates, just one gate and then some large stones that Garfinkel decided to call "the remains of a gate".

I was at a conference where Garfinkel started talking about how he had found the actual serving ware used at Saul’s coronation. There was even a sherd that had lmlk written on it, that is “for the king”. Which is not at all unlike the vast number of jars that have been found with lmlk stamps on them, since such jars were used to pay taxes in olive oil. Israel Finkelstein just sighed, stood up, and walked out of the room.

I imagine Garfinkel’s work is better appreciated by the Ministry of Tourism than by his peers.

JuanRivera said...

The uploaded Sikora et al samples don't appear in non-sponsor G25 nMonte.

Davidski said...

@JuanRivera

The uploaded Sikora et al samples don't appear in non-sponsor G25 nMonte.

They're in the G25 datasheets though, so they should eventually appear in the web runner.

In the meantime, use the R version.

JuanRivera said...

Link for site?

Davidski said...

I re-posted the links to the G25 datasheets in the last blog entry...

Evidence of European ancestry in the Philistines

The links are always the same.

Mark said...

@Drago

“Unetice did not descend from BB- it was a rather different culture- but has important admixture from BB & CWC"

First of all, to say that Unetice did not descend from BB but has an important admixture is a contradiction. I am certain you have an important admixture from your grandfather, because you are descended from him. The same is true of BBC and Unetice.

But there are plenty academic of sources that support my point. How about The concise Oxford Dictionary of Archeology, for example:
“The early phase seems to have developed out of the local Bell Beaker Culture and embraces a series of regional groups including Nitra (western Slovakia), Adlerberg (mid‐Rhine), Straubing (Bavaria), Marschwitz (Oder Basin), and Unterwölbling (Austria).”
https://www.oxfordreference.com/view/10.1093/oi/authority.20110803110638139


Your other point:
“Urnfields expanded from Transdanubia; where the urn burial tradition is rooted. Tumulus culture, on the other hand, expanded from Germany / France back east.”

Once again there are plenty of academic sources that support an origin of Urnfield from the Tumulus culture. How about the famous Marija Gimbutas, for example. Page 298 from Bronze Age cultures in Central and Eastern Europe:
“The available Late Bronze Age material from central Europe, which comes from thousands of sites, indicates an unquestionable Tumulus-Urnfield continuity of culture.”
https://books.google.com/books?id=BvtRdigDtFoC&printsec=frontcover#v=onepage&q&f=false

I find it hard to believe that I have to defend the concept that the successive Bronze age cultures of central Europe represent a successive lineage. But oh well. Central Europeans today still have a very similar genetic makeup as the BBC/CW.

So once again: BBC->Unetice->Tumulus->Urnfield. Then around 1200 BC the Urnfields rampage around Europe, down the boot of Italy, into some boats to attack the Mediterranean palace civilizations only to end up Canaan as the Philistines. Fast forward to the 21st century they dig up the Philistines evidence of cremation (unique to the Levant) and they show R1b Y chromosomes, and our good friend Davidski detects Bell Beaker affinity in them. Is this so complicated?

Drago said...

Mark
We need to understand the difference between meso -regional population-genetic continuity and local - regional cultural shifts ; and the problem is you think you’re self evidently correct when in fact you’re completely missing the point

At a specific level; we cannot say there is “continuity between BB & Unetice”; because they’re different groups; with different organisations; and in fact Unetice represents a rejection of BB norms in Central Europe
Generations of BB family cemeteries end; the BB body orientations end. Completely new burials are opened; with new grave forms, new treatment of the body and new ways of expressing prestige.
This is in fact a complete social rupture in the key zones where Unetice took hold- part of SW Poland, Slovakia and Germany. Genetically they are “very similar genetic makeup” but it’s notable that P312 isn’t the majormlineage as in BB; but we rather see I2, G2, U106; and R1a (in the earlier phase) whereas just outside this key zones, the change was far less abrupt and we can simple state that BB communities continued on more but adapted to new social conditions . Further out still; in the western periphery; there is virtually no change; and local descendants of BB continue on with little change- eg a baroque, country styled BB regional groups in Iberia.

So once again: “BBC-> Unetice -> Tumulus -> Urnfield” is fair ok when for looking for possible Central European in exotic lands (Levant); but for within europe such a sweeping claim hinders rather than helps in understanding the important details of post -Late Neolithic phenomena

zardos said...

@Mark: Dragoner is right with his last comment. From all we can say Urnfield ended the BB tradition in most of Central Europe. New culture, new socio-political system, burial rites, new technologies, changed physical type and patrilineages.
There are refuges like Adlerberg culture, but the Urnfield phenomenon at its core is the rejection of the BB rule.
Certainly not just a peaceful change. I would compare it with the end of the Assyrians, with an alliance of newcomers, neighbours and suppressed getting rid of the BB yoke.

Something similar happened from Hallstatt to La Tene. Elites which base their rule on a cult, knowledge, monopolies and technology get overthrown.
BB = Copper Bronze, Hallstatt/Hittites = iron.

In the case of BB it might have ended pretty ugly for most clans in non-majority parts of Central Europe. I would not see that as real continuity.

JuanRivera said...

Went to model the Swat_IA groups and Ust_Belaya (including the outlier). All the Swat_IA groups have more than 20% Sintashta_MLBA (except the outlier, which has ~15% Sintashta), and two samples belong to Y-haplogroup R1 (one R1a and the other R1b). Ust_Belaya seems to have traces of steppe ancestry (with some models displaying 0.83% Afanasievo and others 0%, indicating a true percentage of ~0.3-~0.6% of Afanasievo), though its inclusion leaves worse fits. When Okunevo is used, the fits of the main groups further worsen, whereas that of the outlier shows the best fits, with Okunevo proportions always above 1% (though its Kolyma1 part makes unclear the percentage of steppe ancestry).

JuanRivera said...

The Okunevo part is adressing Ust_Belaya and Ust_Belaya_o. Though, in the models Shamanka_BA is used (which has steppe ancestry in itself).

Davidski said...

None of the Swat IA samples belong to R1b. That's an R2.

Keep in mind that many, if not most, of the Y-haplogroup assignments in that preprint are wrong.

JuanRivera said...

Thanks for the correction.

Andrzejewski said...

So you’re postulating BB as a ruling elite in lieu of a fairly homogeneous population of CW/SG with Atlantic farmer (TRB?) admixture who pretty much exterminated other farmer and forager populations across Europe?

Andrzejewski said...

Speaking of Beakers, I’ve noticed that the areas in the island of Ireland where Beaker culture did NOT protrude was more or less overlapping with today’s “Gaeltacht” (small area where the native tongue of the people is Irish and not English).

Is there any coincidence?

Bob Floy said...

@Andre

That's because of more or less modern historical events, nothing to do with the bronze age.

Leron said...

The Mycenaeans were only reaching Cyprus and the Ionian coast by the time the Hittite empire fell. The Lycians seem to have been guiding them in raids against Cyprus. The Lycians were also part of the Sea Peoples movement although they may have only penetrated former Hittite dominion of the "Lower Lands", later known as Lycaonia. It was the Luwians or Luwic people who were ahead of the Greeks in reaching eastern territory; expanding as the Cilicians, Mopsos/Muskas dynasty, Neo-"Hittites", and P/Walistin kingdom which could be related to the locality of Parista/Palista that fought against the Hittites as part of the Assuwa league in Asia Minor. I don't buy the Philistines as Mycenaeans. It's sort of like expecting to find the British in Texas while the Spanish were still exploring Mexico.

self-consumer said...

@Andre

I imagine the only coincidence could be an invasion from the south and/or east pushing the resistance westwards and into more defensible terrain. Certainly there would be no connection beyond that.

Davidski said...

@PF

Is it possible to get the other Minoan samples (not just Lassithi) into the G25?

I just put three Minoan_Odigitria samples into the G25 datasheets. But I don't know how reliable they are. Same links as always.

zardos said...

"Speaking of Beakers, I’ve noticed that the areas in the island of Ireland where Beaker culture did NOT protrude was more or less overlapping with today’s “Gaeltacht” (small area where the native tongue of the people is Irish and not English).

Is there any coincidence?"

I agree with self-consumer. Its a coincidence because the areas of refuge stay the same through the ages. Even in modern wars people retreated to the mountains for example (the Swiss Army plan includes that to this day) or better soils being more expensive than bad ones. So if a people lost its position, they will retreat with a pattern. Oftentimes the winners from yesterday were the next losers.

The situation of BB was different in different places, but Central Europe, or better the later Urnfield core territory, was at its fringe and not as completely transformed as other places, like the later Adlerberg territory. After their defeat and with the new cultural phenomenons spreading, even the surviving nests adapted culturally and finally mixed. But again, at first Urnfield seems to have broken their rule. Like La Tene stepped out of the ashes of Hallstatt elite fortresses and monopolised iron production and trade.

Alexandros said...

Interesting yet bold throry. Specifically..
"Then around 1200 BC the Urnfields rampage around Europe, down the boot of Italy, into some boats to attack the Mediterranean palace civilizations only to end up Canaan as the Philistines".

Would you thus suggest that the Dorians who descented and dominated Mycanean Greece and the Philistines who settled in the Levant are one and the same? i.e Urnfields?

Davidski said...

@All

I updated this blog entry and the datasheet.

Global25 workshop 1: that classic West Eurasian plot

If anyone knows how to use the Transform tab in PAST to flip these sorts of plots so that they correlate a little better with geography, please let me know, because I can't figure it out.

zardos said...

BBC in CE ended with Unetice of course and was long gone at the time of Urnfield, but between BB and Urnfield was a series of upheavals which left little of the BB heritage at all.
The break came with Unetice already.

Aram said...

PF

There is a onomastic data supporting West/SouthWest Balkanian origin of Philistinians. Via Crete and Cyprus.

You can try Montenegro BA sample also. If I remember correctly there was a such sample in Allentoft et al. And it was quite western.

I couldn't find it in Global25.

andrew said...

@AWood "Is there any evidence these specific R1b men were short or small stature though"

Good points.

Aram said...

http://polishgenes.blogspot.com/2016/02/pca-of-rise595-rise596-and-rise598.html?m=1

After some googling I found this old post of Davidski about those Montenegrin samples. With PCA.

Ric Hern said...

What does short or tall got to do with it ? I have been to Israel and found people in Ashkelon to be a bit shorter than in other parts of Israel. I am 190cm and felt like a Giant among them. Head and shoulders taller. Are they still R1b ? I am, so I don't see the relevance and any direct correlation between Shortness or Tallness and Y-Haplogroup R1b.

Leron said...

Aram: And did this “West Balkanic” group skip over Greece or did they leaves traces there? Unless you are somehow equating them with Dorians, but it’s really strange they’d teleport all the way to the Levant while Mycenaeans were only trading there.

PF said...

@Davidski

I just put three Minoan_Odigitria samples into the G25 datasheets. But I don't know how reliable they are. Same links as always.

Thanks. They look a tiny bit less CHG-shifted than Lasithi, as in the original paper, but basically the same. Guess it's a very small hint of where the more CHG-shifted farmers came from, though that's obvious anyways.

@All

We really need to get our hands on some Mesopotamian genomes! I think a lot would fall in place after that, with much more accurate modeling instead of using CHG and Iran. It's a shame that entire region is a war-torn hellhole right now. Anyone know if there are any skeletons laying around in some museum that could be sequenced!?

JuanRivera said...

There's one in Penn Museum.

Arza said...

@ Davidski

Shouldn't "Transform -> Evaluate expression -> -1*x" do the trick?

Dita said...

Dardanians were on the border of Israel in 1270BC at the Battle of Kadesh. That's hundred years before Philistines appear. Linguists like Bonfante, and plenty other scholars argued that Philistines were Illyrians.

Simon_W said...

According to Eudoxus of Cnidus who lived in the 4th century BC, the people living in Italy between the Umbri and the Iapygians, that is, in Abruzzo approximately, was called Phelessaioi. AFAIK no other source mentions them. But the ancient Greeks would have pronounced the Ph as aspirated P, so I was always wondering: Couldn't these be related with the Peleset/Philistines?

Dita said...

The river strymon was called Palaestinus in antiquity. Also in Illyria you had the place names "Palaeste" and from it the tribe of the Palaestini. The "-ini" suffix is common to Illyrian tribes.

Some illyrian tribes:

Tergestini from Tergeste
Praenestini of Praeneste
Atestini of Ateste

There are many many more examples of tribes with this suffix, which is an Illyrian feature, not a greek one.

For more see:

"Who were the philistines" by G. Bonfante
"Some early philistine history" by G.A. Wainwright

Andrzejewski said...

Are Albanians descendants of the Illyrians?

Dita said...

Occams razor points to that being the case. Y-dna of Albanians has contuity in west balkans since deep antiquity. Linguistics also points to that at the very least albanian is a sister language to Illyrian, but the probability is most likely that its just its direct descendant:

""Eric Hamp, who has been the most careful and conservative of specialists on Albanian dealing with this issue, has decided that the evidence linking Albanian to Messapic, which in turn has been linked to Illyrian, is sufficient to see Albanian as a descendent of a sister-language to Illyrian, if not a descendent of Illyrian itself (reported in Lezo, 2008)"

Victor Friedman,
Oxford Research Encyclopedia
March 2017"

Matt said...

@Davidski, generally, you can extract the output of re-processing through PAST3's function using the Scores tab (https://imgur.com/a/9qLffXl) then use the Transform->Landmarks function to flip and rotate dimensions (https://imgur.com/a/RiNaOl2) until you get something that looks right.

In this case though, I would say that since reprocessed dimension 2 is what looks unaligned with geography when you reprocessed output as with plot X as PC2 and Y as PC1, then if you reverse the dimensions the reprocessed PC2 loads on before reprocesses, then that will make PAST3 flip the reprocessed PCA. If that is at all comprehensible (I'm struggling to find a better way of wording it). (May be what Arza was suggesting).

Basically multiply the contents for columns for input G25 dimensions 2, 3, 5 and 10 by -1 then feed back into PAST3 and run the PCA function, then you can plot like this: https://imgur.com/a/mVBrQgH

Any given dimension can be basically flipped with no effect on distances, so the direction of each dimension is basically just an arbitrary part of PAST3's PCA algorithm (which annoyingly you can't flip within the software!). Flipping dimensions that you *input* into the software makes whatever arbitrary part of the algorithm in PAST3 that assigns ends as positive/negative do the reverse.

Samuel Andrews said...

@PF,
"@All

We really need to get our hands on some Mesopotamian genomes! I think a lot would fall in place after that, with much more accurate modeling instead of using CHG and Iran. It's a shame that entire region is a war-torn hellhole right now. Anyone know if there are any skeletons laying around in some museum that could be sequenced!?"

True. There's also no DNA samples from Arab Iraqi(s), Arab Syrians. But I bet that CHG, IranNeo, AnatoliaN, Natufian will continue to be perfect sources for explaining Middle Eastern variation. Mesoptamia was probably originally IranN territory.

Andrzejewski said...

Would Semitic people originate in Natufians or Iran Neolithic?

Andrzejewski said...

The Arabs (or pre-Muslim minorities like Iraqi Christians) probably have a higher rate of Anatolia_N admixture and therefore they probably resemble Europeans more

Suyindik said...

@Davidski, Samuel Andrews, PF

Would you consider the genetic structure of the population of the Peqi’in Cave from Chalcolithic Levant to be identical to the non tested Y-dna of Neolithic Mesopotamia?
Harney et al 2018 was mentioning that the population of the Peqi’in Cave were native migrants from Northern Mesopotamia.

Samuel Andrews said...

Afro-Asiatic languages probably come from Natufian or Natufian-heavy people. Why? Because, Afro-Asiatic East Africans have lots of Natufian ancestry. They also have 0% or very little IranN, AnatoliaN ancestry.

Samuel Andrews said...

@Suyindik,
"Would you consider the genetic structure of the population of the Peqi’in Cave from Chalcolithic Levant to be identical to the non tested Y-dna of Neolithic Mesopotamia?
Harney et al 2018 was mentioning that the population of the Peqi’in Cave were native migrants from Northern Mesopotamia."

The Chalcolithic samples from 5th millenium bc Israel had only a little bit of ancestry from from Iran/Iraq. 13% in total. 87% of their ancestry was from Levant Neolithic aka was native to the region.

Bronze age samples from 3rd millenium bc from Levant show more Iran/Iraq ancestry. About 30-40% of their ancestry was from Iran/Iraq and 60-70% was native to Levant.

FrankN said...

Sam: "I bet that CHG, IranNeo, AnatoliaN, Natufian will continue to be perfect sources for explaining Middle Eastern variation. "

Not sure about that. One one hand, there was the Persian Gulf Oasis (the Persian Gulf was only flooded with the post-Younger Dryas global sea level rise from ca. 8.000 BC onwards, contemporary to Doggerland). IMO, that Oasis is the prime candidate when it comes to looking for a Basal Eurasian hideout (to the extent BE really existed and is not just some statistical artefact).
OTOH, the LGM lead to a reversal of Monsoon patterns, with the Eastern Monsoon weakening or even collapsing, turning most of the Ganges plain into an uninhabitable desert. Conversely, the Western Monsoon strengthened substantially, making Baluchistan much more humid than to date (or - say - than during Alexander the Great's time). This implies the possibility of a quite sizeable LGM population along the coast from Baluchistan down all the way to Sri Lanka, which might have taken part in the re-population of Arabia and Mesopotamia (In fact, there seems to be a South Asian signal present in Iran_Neo).

"Mesoptamia was probably originally IranN territory." In fact, for reasons poorly understood so far (possibly climate-related), there aren't any Epi-Paleolithic or Mesolithic sites in Mesopotamia south of Iraqi Kurdistan. The archeological record starts suddenly, and massively, with the Hassuna and Samarra cultures around 6000 BC, i.e. after the 8.2 ky event. So, "originally", Mesopotamia was a kind of "no-man's land". This, of course, makes the question even more interesting from where its colonisers came…

Otherwise, I'd like to extend the "whish list" to the Arabian Peninsula that from the BA onwards has a rich archeology history (ancient irrigation systems, Copper smelting in Oran etc.), w/o much of a Mesolithic/Neolithic record. How come Arabians speaking Semitic, in spite of their yDNA being dominated by "Caucasian" J, and cultural connections for a long time rather going towards Mesopotamia and India than the Levante?

Andrzejewski said...

So Iran_N people (later on to mix with Natufians to form Levant_N on 50:50 basis) originally spoke Elamite, Dravidian or BMAC language?

PF said...

@Andrzejewski

Levant_N has nothing to do with Iran_N. Rather it's a mix between Natufians and Anatolia_N. As Samuel said Iranian ancestry there starts with Levant_Chl (a bit) and increases a lot with the BA samples.

Who knows what language was spoken around ancient Iran, but yes, given what we know about pre-IE South Asia I guess it's possible that it's related to Dravidian.

JuanRivera said...

Some models show Simulated_AASI_NW not only in Iran_Mesolithic/N, but also CHG, West_Siberia_N (at 0.83%) and by extension EHG and EHG-admixed european HGs (from their CHG admixture). Even MA1 and AG3 tend to choose Simulated_AASI_NW over all other East Eurasian populations as ENA admixture source. By the way, Samara_HG not only has more ANE than Sidelkino_HG, but also more CHG (and maybe even some Hotu_HG, which sometimes shows up at 0.83%, which was definitely absent in Sidelkino_HG).

Grey said...

Dita said...
"The river strymon was called Palaestinus in antiquity. Also in Illyria you had the place names "Palaeste" and from it the tribe of the Palaestini. The "-ini" suffix is common to Illyrian tribes."

interesting idea - famously piratical in Roman times - a recent thing or something to do with climate/geography going back a long time which pushed them into maritime activities?

Andrzejewski said...

That’s what Davidski has always said, that Steppe component is very old, dating back to the Mesolithic period. I suspect that Yana (ancestral to CHG) and ANE (ancestral to EHG) are sibling populations.

JuanRivera said...

Ran even more models. It turns out Iran_Mesolithic/N and CHG prefer Natufian to "Basal Eurasian" (approximated by Eth_4500BP) as source of extra non-crown eurasian ancestry. Also, Sidelkino_HG and Ukraine_Mesolithic model better as having some Anatolia_Pinarbasi_HG (alongside AG3, WHG and CHG in the former and Sidelkino_HG, WHG and CHG in the latter). Samara_HG shows some Ukraine_Mesolithic ancestry (alongside ANE, which increases from ~13% to ~20%, CHG and Sidelkino_HG), but oddly trying to model it with WHG instead of Ukraine_Mesolithic shows it as having 0% extra WHG (outside of Sidelkino_HG). Couldn't verify if Samara_HG has extra Anatolia_Pinarbasi_HG (outside of Ukraine_Mesolithic and Sidelkino_HG) because nMonte only allows up to 4 source pops. That means that the prehistoric steppe and forest-steppe had even more complex population histories, with Anatolia_Pinarbasi_HG and possibly Hotu_HG being added to the known ANE+WHG+CHG makeup of the EHG and Ukraine_HG populations of the northern steppe and forest-steppe.

FrankN said...

Andrzej: "So Iran_N people (later on to mix with Natufians to form Levant_N on 50:50 basis) originally spoke Elamite, Dravidian or BMAC language?

Not so fast..

1. The "Iranian" 50/50 signal in the Levante is
a) not Neolithic (see Sam's comment above), but CA/EBA and to some extent even from later periods, and
b) can, for the CA/EBA, be attributed to the archeologically well documented Kura-Araxes expansion. The Levantine variant of KA is known as Khirbet-Kerak ware/culture. Erikl86 above has provided some models based on Hajii Firuz_C, which is near Urmia Lake, may to some extent be considered ancestral to KA, and lies, while preceding KA, within that culture's early range. My guess is that KA proper samples would yield even better modelling results for the CA/EBA Levante..

2. The old Elamite-Dravidian hypothesis can be considered as refuted. Starostin 2002 (link below) has shown that lexically, Elamite shows most parallels to Afro-Asiatic, followed by Uralic (!) and IE. As concerns Elamite-Dravidian parallels, most of them are undiagnostic. E.g. Elam. mak- "to eat" may be linked to Drv. *moq, but equally to Proto-Korean *mək "to eat, drink", Proto-Japanese
* maka-nap "to feed", or Malay makan "meal", maka-nan "to eat" - apparently a pan-Asian paleo root. There also appear to be a number of Elamite parallels to Austro-Asiatic, e.g. Elam ka "fish" vs. AusAs *ka "fish" (c.f. Malay ikan "fish").
Starostin concludes: "At this point, I would probably describe Elamite as a "bridge" between Nostratic and Afroasiatic." I think there is good reason to add Austro-Asiatic, which Starostin didn't consider, here as well. Ultimately, Elamite looks to me very much like a creole for long-distance pan-Asian trade that might only have formed during the CA.
http://starling.rinet.ru/Texts/elam.pdf

3. While I am at it: A paper from the Moscow Linguist School has demonstrated Hurrian substrate in Sumerian. Otherwise, certain Sumerian pictograms apparently go back to PIE, e.g. a fish-shaped pictogram with the sound value PES (in Sumerian, "fish" was ka, as in Elamite and PAusAs), or the eye-shaped IGI pictogram. The general consensus appears to be that Sumerians were early 3rd mBC newcomers, who imposed their language onto resident Hurrian (plus IE?) speakers. Parpola suggests Sumerian to have been an Uralic language, Maykop-derived. Others propose an Altaic origin. The general features of Sumerian (agglutinative, vowel harmony, prohibiting intra-syllabic consonant clusters) certainly align with Uralic or Altaic, but I personally have problems in consolidating such theories with the archeological record.

4. Dravidian: Witzel envisages an Iranian origin of Dravidian. IMO, however, S. India should have been a good enough place to weather the LGM to also have formed its own linguistic phylum. This phylum should at least have left over some language isolates and/or a recognisable substrate, both of which doesn't appear to be the case. As such, my feeling is that Dravidian developed in situ, as AASI phylum.

JuanRivera said...

Yana doesn't seem ancestral to any population, only a sister pop of ANE (though ANE may actually be a mixture of a Yana-like proto-ANE population and something else). When it comes to North Eurasian ancestry, CHG and Iran_Mesolithic/N choose MA1 over AG3, whereas EHG and similar HGs (all ANE-admixed european and west siberian HGs) mostly choose AG3 over MA1 (all ANE outside of their CHG ancestry). It's something similar to how Native Americans prefer MA1 over AG3, whereas Baikal HGs prefer AG3 over MA1.

FrankN said...

Continuing my previous post:

5. BMAC: Essentially, we need to envisage the following population streams:

a) The Pamir Epipaleolithic (possibly even during the LGM), so far unsampled but maybe a genetic component that has NMonte models tending to prefer Sarazm over other sources;
b) In East Central Asia, Mesolithic entrants from China, bringing with them the East Asian Yubetso pressure knapping technology;
c) As above for W. Central Asia, but this time "bullet-shapped" pressure knapping technology originating from the Middle Yenissey and associated with AG3-like ancestry;
d) Neolithisation from the S. Caspian (Sang-e-Chakmak->Jeitun). Goat aDNA shows that the Central Zagros (->Iran_Neo) wasn't yet significantly involved in this process;
e) Eneolithic expansion out of Iran as well visible from the Narasimhan e.a. aDNA;
f) Interaction with the Indus Valley - obvious for the AASI in some late BMAC samples, but possibly already at work much earlier.
Linguistically, this implies that the BMAC language should have born some parallels to Iran-Neo, but both most certainly weren't identical.

Interesting in this respect is Burushaki, which Witzel regards as part of what he calls "Macro-Caucasian" (with Basque at the western end), and that may as such also relate to the Neolithic S. Caspian. However, I don't know enough about that language to make any qualified statements of my own.

FrankN said...

Juan Riveira:
" EHG and similar HGs (all ANE-admixed european and west siberian HGs) mostly choose AG3 over MA1 ".

Of course they do! After the Younger Dryas, the upper/middle Volga basin was re-populated from both the West and the East. There are a number of good publications about it, e.g. by Hartz e.a.. The Siberian input was "bullet-shaped" pressure knapping, first attested from the Middle Yenissey, i.e. AG3. This isn't "preference", this was an actual migration by a source population that we luckily have aDNA from.

JuanRivera said...

I know. There's also some CHG in Sidelkino, with some increase in later steppe HGs (especially in Samara_HG).

JuanRivera said...

I'm also reading about the cultures of "Neolithic" Chukotka and Kamchatka (and of eastern siberia and the russian far east as a whole), to see how cultures change and therefore track the movement of steppe-admixed paleosiberian groups (the presence of steppe ancestry in Ust'-Belaya, albeit at traces [when modeled as Shamanka_EBA+Anzick+USA_Alaska_TrailCreek+Afanasievo, given the present lack of Kolyma1 in the web runner, which otherwise could produce a higher steppe ancestry percentage] puts a lower bound at 4250 BP). The lower bound is close to the proposed age of the Proto-Chukotko-Kamchatkan language (at roughly 4000 years ago).

Jerry said...

@Davidski

I've noticed that Taleb has recently started to become interested in genetic archeology. Do you find his criticism legitimate?

https://twitter.com/nntaleb/status/1148599452053192709

Davidski said...

@Jerry

Do you find his criticism legitimate?

Not based on that tweet, because there are many different ways to analyze genetic data and when they all basically correlate to show one and the same thing then the result can't be questioned.

Genetic maps (PCA) are just one type of analysis and mostly for visuals anyway. There's no point obsessing about how far apart two dots are from each other if you've got a system that generally produces the same results as a wide range of other stats, especially formal stats.

Andrzejewski said...

So what you, @FrankN are saying is thst the Ubaidians and Halafians were Hurrian speakers?

Andrzejewski said...

Then maybe PIEs are not an admixture of CHG (MA1) and EHG (AG3) but a separate, distinct Yana-derived population but the PCS mistakenly puts it at an exact distance between EHG and CHG whereas PIEs May be just Yana + some EEF (15% Anatolian)?

Andrzejewski said...

PCA, not PCS :)

Andrzejewski said...

Narasimhan 2018 debunked and discredited any trace of BMAC (or Botai for that matter) on extant South Asian populations. So the idea that Andronovo acquired their religious ideas and terminology from BMAC had been relegated to the trash bin of history

Andrzejewski said...

Thus the similarities between Chukotko-Kamachkadal language to pie may stem from an influence of the latter on the former, due to “magadan ancestry”?

Alexandros said...

Interesting read below.

Sea Peoples of the Northern Levant? Aegean-Style Pottery from Early Iron Age Tell Tayinat (REVIEW)
American Journal of Archaeology 123(3), 2019

https://www.ajaonline.org/book-review/3895


In conclusion, "... the earliest Iron I occupation at Tell Tayinat most likely consisted of a combination of immigrants from around the eastern Mediterranean (the Aegean, Cyprus, and western Anatolia) and elements of the indigenous population, who may themselves have been immigrants from other settlements around the Amuq Valley, including neighboring Tell Atchana".

Ebizur said...

FrankN wrote,

" Proto-Japanese * maka-nap 'to feed'..."

There is no such word. The word that actually does exist is the following:


まかな・う〔まかなふ〕【賄う】の意味
Meaning of makana-u (Classical Japanese makanaf-u) 【賄う】

出典:デジタル大辞泉(小学館)
Source: Digital Daijisen (Shougakukan)

[動ワ五(ハ四)] [verb with WA five-step (Classical FA four-step) conjugational paradigm]

1 費用・人手などを用意する。ととのえる。「寄付で費用を―・う」
1. To get funds, a work force, etc. ready. To raise. To supply. To cover. "Kifu de hiyou o makana-u" (to cover an expenditure with donations)

2 食事をととのえて出す。「夕食を―・う」
2. To provide; to obtain ingredients, prepare, and serve (a meal). "yuushoku o makana-u" (to provide (a person) with an evening meal, to prepare and serve a dinner, to take care of food for the evening)

3 事を処理する。切り盛りする。「親からの仕送りで―・う」
3. To run, to manage (a household). "oya kara no shiokuri de makanau" (to maintain oneself with remittances from one's parent[s])

4 とりはからう。とりしきる。「疎略なき様に―・ふべし」〈浄・娥歌かるた〉
4. to take care of, to dispose of, to deal with, to see to it. "soryaku naki yau ni makanaf-u beshi" ([You] ought to see to it that good care is taken. / [You] should see that nothing is neglected. - from the Jōruri, Kaoyo Uta Karuta, i.e. "Pretty-faced/Good-looking Poem/Song Card[s]," ca. 1714 CE)

The Chinese character used to write the root makana-, 賄 (Mandarin huì), means "to bribe; bribes; riches, wealth, property, belongings; to give belongings," formed from the semantic radical 貝 "shellfish; shell; money, wealth" and the phonetic marker 有 yǒu "to have." This suggests that the Japanese word was considered by native speakers to be associated primarily with financial matters at the time when the kanji was associated with the kun'yomi makana-.

The verb is sometimes used in relation to the preparation of meals for human consumption as in Example 2 above, but this sense is more strongly apparent in the deverbal noun form, makanafi > makanai "(room and) board, meals; food preparation, food service; waiting (on a diner), waitstaff; food that kitchen staff prepare from ingredients that they happen to have on hand for themselves to eat; management, handling; covering (an expense), paying, taking care of (a bill); making (something) serve a purpose" and the antiquated derived noun makana(f)i-kata "a cook, kitchen staff, personnel in charge of delivery of foodstuffs and preparation and serving of meals (in feudal Japan)." However, there is no clear way to derive the Classical Japanese verb makanaf- "to cover (an expenditure, etc.); to take care of everyday business" from a hypothetical verbal root that might mean "to eat."

FrankN said...

Ebizur: Thx for the correction. I had taken the Japanese example straight from Starostin's text, without further checking.

So, Jap. makana and derivatives seem to semantically resemble German (be-)wirten, Wirt: It originally meant care-taking/taker, still present in constructions such as Wirt-schaft "economy, enterprise", Land-wirt "farmer" (lit. "care-taker of the land"), c.f. Engl. warden. Related to it is Wert "value, worth". In modern German, however, Wirt has taken the meaning of "host, inn owner", i.e. a person providing food, drinks, and/or a place to sleep.

I aggree that such a root shouldn't be directly related to the meaning "to eat".

Suyindik said...

@Samuel Andrews-"The Chalcolithic samples from 5th millenium bc Israel had only a little bit of ancestry from from Iran/Iraq. 13% in total. 87% of their ancestry was from Levant Neolithic aka was native to the region.
Bronze age samples from 3rd millenium bc from Levant show more Iran/Iraq ancestry. About 30-40% of their ancestry was from Iran/Iraq and 60-70% was native to Levant. "



Mesopotamia includes regions from South-Eastern Turkey, Syria, Iraq and Western Iran. Till now there is no study which published ancient Neolithic Y-dna from South-Eastern Turkey, Syria and Iraq.

And the correct ancestry frequencies of the Chalcolithic Peqi'in population is as following:

Levant_N (57%)
Anatolia_N (26%)
Iran_ChL (17%)

So, 43% of their ancestry is from outside of the Southern Levant. This means they are not native to the Southern Levant region(as is described in the paper), but they came from Upper Mesopotamia as written in the paper.

I think the ancestry of 26% Anatolia_N shows their connection with the Neolithic South-Eastern Turkey part of Mesopotamia. And the 17% Iran_ChL shows maybe a migration of Northern Mesopotamia into Chalcolithic Iran and Azerbaijan. And also migrations from Upper Mesopotamia into the Levant started beginning in the Pottery Neolithic period. Thats why we can see the Y-dna of Chalcolithic Levant in PPNB Levant(in minority) but not in the PPNA and the Natufian periods. And the 57% of Levant_N probably comes from the PPNB part of the migration(and the mixing with the local Natufians) that started in that period from Northern Mesopotamia into the Southern Levant.

And it is also very probable that the Natufians were not native to the Southern Levant, because they were probably native to pre Neolithic Northern African regions.

epoch said...

@FrankN

"Of course they do! After the Younger Dryas, the upper/middle Volga basin was re-populated from both the West and the East. There are a number of good publications about it, e.g. by Hartz e.a.. "

Got any links?

Suyindik said...

@Andrzejewski-"So Iran_N people (later on to mix with Natufians to form Levant_N on 50:50 basis) originally spoke Elamite, Dravidian or BMAC language?"

Its very much probably that the Elamite, Dravidian, BMAC, Maykop languages have their origin in Neolithic Mesopotamia just like the languages of the Sumerians, Ubaid, Halaf, Hassuna and Jarmo.

FrankN said...

Epoch:

I have discussed the issue here:
https://adnaera.com/2018/12/10/how-did-chg-get-into-steppe_emba-part-1-lgm-to-early-holocene/

From there, you can also take the links. Scymczak 2002 illustrates the general trail of pressure knapping from E. Asia into Europe, with an East Caspian split into a "southern" (Iran_Neo-> PPN Anatolia-> EEF) and a "northern" route (Kama-Perchora->Butovo->Kunda->Maglemose/Duvensee). Brunet 2012 (plus other publications by him) discusses the Central Asian split into a Western, "bullet-shaped" (MA1/AG3-derived) and an Eastern, Yubetso (Jomon, China) sphere. Hartz e.a. 2010 deals specifically with the early Mesolithic dualism of "Western" and "Siberian" traditions on the Upper/Middle Volga. Further links there discuss details, e.g. whether pressure knapping reached the Yennissey (AG3) rather from the Trans-Baikal, of from the Altai (MA1).

FrankN said...

Andrzej: "So what you, @FrankN are saying is thst the Ubaidians and Halafians were Hurrian speakers?"

See my comment above to Sam. More specifically:

a) While Akkadian is commonly assumed to have been a later entrant, Levantine colonisers speaking some form of (para-)proto-semitic might have been from the outset involved in settling Mesopotamia.

b) I am pretty certain that there was substantial settlement of Mesopotamia from the south, i.e. the Persian Gulf Oasis. The topic is intensively discussed in Rose 2010:
https://www.academia.edu/386944/New_Light_on_Human_Prehistory_in_the_Arabo-Persian_Gulf_Oasis

c) Some presence of populations from the Western Indian Ocean may also be anticipated.

Until we better understand which populations, in which ratios, colonised and neolithicised Mesopotamia, any speculation about who might have spoken which language is premature.

A said...

"And the correct ancestry frequencies of the Chalcolithic Peqi'in population is as following:

Levant_N (57%)
Anatolia_N (26%)
Iran_ChL (17%)"

Levant_N was apparently already 1/3 Anatolian_N, and Iran_ChL also appears to have had a significant amount of Anatolia_N:

"Among first farmers, those of the Levant trace ~2/3 of their ancestry to people related to Natufian hunter-gatherers and ~1/3 to people related to Anatolian farmers"

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5003663/

"We show that there was a west-to-east cline of decreasing Anatolian agriculturalist-related admixture ranging from ~70% in Chalcolithic Anatolia to ~33% in eastern Iran" (p.7)

https://www.biorxiv.org/content/biorxiv/early/2018/03/31/292581.full.pdf


If I'm not mistaken the the Chalcolithic Peqi'in people might have been around 50% Anatolian neolithic farmer.

Open Genomes said...

@David

Gedmatch comparison of ASH068 (European Iron Age 1 Philistine) vs. an unadmixed Samaritan NZ8875603

According to my ASH068 has at least 3% Kubano-Tersk Steppe. So does this Samaritan, who is in Global25, have any Steppe?

Open Genomes said...

@David, that Samaritan is 149533 in Global25.

PF said...

@Suyindik @Samuel

Well, Iran_Chl itself has Levant_N ancestry, so Sam was correct when he said that Levant_Chl is still predominantly derived from Levant_N.

[1] "distance%=3.1608"

Levant_ChL

Levant_PPNB,89.4
IRN_Seh_Gabi_C,9.8
CHG,0.8


1] "distance%=3.4278"

Levant_ChL

Levant_PPNB,92.4
IRN_Seh_Gabi_LN,5.2
CHG,2
Anatolia_Boncuklu_N,0.4


[1] "distance%=3.1358"

IRN_Seh_Gabi_C

IRN_Seh_Gabi_LN,53.2
CHG,17.6
Levant_PPNB,16.8
Anatolia_Boncuklu_N,12.4

Suyindik said...

@PF-"Well, Iran_Chl itself has Levant_N ancestry, so Sam was correct when he said that Levant_Chl is still predominantly derived from Levant_N."

There are Levant samples from the Natufian, PPNA and PPNB. The paternal ancestry of Levant Chalcolithic is found beginning in PPNB, and before this(Natufian, PPNA) the Natufian paternal ancestry is dominant in the region. And it is also very possible that the Natufians themselves were not natives of the Levant, but were natives of earlier periods of Northern Africa. And, archaeological material show that migrations from Northern Mesopotamia into the Levant happened starting from the Pottery Neolithic period until the Chalcolithic period.

So, the Levant_N ancestry within Iran_Chl shows that the ancestors of Levant_Chl who migrated into the Southern parts of the Levant from Northern Mesopotamia beginning from the PPNB period(being major in the Chalcolithic period and then disappearing from the region since the Bronze Age), also migrated afterwards into the regions of Iran_Chl(Western Iran and Azerbaijan).

The Anatolia_N ancestry in Levant_Chl I think is showing a connection with Neolithic South-Eastern Turkey(these populations migrated in some period of time in the Neolithic into Western Turkey and later also into the Aegean Sea region), Iraq and Syria. But for this, I am very curious when a study with Y-dna from Neolithic Mesopotamia will be done(maybe some academicians already have some unpublished data), and if this will show connection to Chalcolithic Levant.

Aram said...

Leron

I also mentioned South West Balkanes, which would be in the neighborhood of Mycenean world. If Philistines are from that region then they definitely didn't skip Greece because they were using Helladic pottery, architecture. As for traces. Did they left genetic traces in Levant? No. But probably if we have more samples from Iron Age Greece we would have different ideas. Samples from South West Anatolia are also.important for the origins of Philistines.

Simon_W said...

Regarding the Phelessaioi of Italy that I mentioned, it's also interesting that the town Ascoli Piceno, on the southern border of the Marche region of Italy, the Roman Asculum, was called "Askoulon" by the ancient Greeks - who have called Ashkelon "Askalon". A mere coincidence?

Well, maybe! Because the name Ashkelon doesn't come from the Philistines, but predates them. According to Wikipedia: "Ashkelon is mentioned in the Egyptian Execration Texts of the 11th dynasty as "Asqanu." In the Amarna letters (c. 1350 BC), there are seven letters to and from Ashkelon's (Ašqaluna) king Yidya, and the Egyptian pharaoh."
So naturally, the Philistines cannot have taken the name Ashkelon from Italy to the Near East. It could be the other way round, in theory.