Ancient Italian DNA: an attempt to distinguish Italics from Tyrrhenians

Ygorcs

Active member
Messages
2,259
Reaction score
812
Points
0
Ethnic group
Multiracial Brazilian
Hi, people! I have been very busy in the last few months and unfortunately had to reduce my posts here to a minimum. Now I'm coming back to discuss with you something that I found pretty intriguing and maybe meaningful as I got access to the recently released and made available Italian aDNA samples. I have been reading a lot being said about "Italics and Etruscans being almost indistinguishable, only the language that survived in their groups differing"... but is that so?

Firstly, I modelled some of the most relevant aDNA population samples from ancient Italy using all Paleolithic, Neolithic and Chalcolithic DNA samples available in Global25 datasheets just to see what it'd give me. I also did the same for some Iberian, French and Balkanic aDNA. Interestingly, an imperfect but noticeable pattern started to emerge: the EEF in Italic-related aDNA is more related to Western Europe (Iberia and France), the EEF in Etruscan-related aDNA is more related to Eastern Europe (the Balkans and Pannonian plain in particular).

In addition to that, there's also something else that I found meaningful: Remedello and North Italian BB were predominately "Western EEF", whereas Sicilian BB (with no steppe ancestry) is wholly "Eastern EEF", and Grotta Continenza (Central Italian) DNA had an almost even split between "Western EEF" (52.6%) and "Eastern EEF" (47.4%)... that suggests that the Italian Peninsula had a subtle genetic cline previously from "Western European-like" in North Italy to "Balkanic-like" in South Italy (not really unlike today, right?).

I also tried several other models with fewer reference populations - and the same pattern still emerged every time. Another thing is that, if you add French BB samples, and compare IA Romans with IA Etruscans, only the former pick a lot of ancestry from those, with the latter "preferring" more CA Central Italian (Grotta Continenza), North Italian BB and Yamnaya ancestry than the Roman sample. If you add several Bell Beaker samples, this thing gets even more interesting: the Etruscan sample picks a steppe admixture right from Kalmykia Yamnaya (~28%), whereas IA Roman picks ancestry only from Bell Beaker samples from Germany, North Italy and France (with or without steppe admixture). In all cases, the Etruscan sample gets much more Grotta Continenza admixture than the Roman one. In general, the non-steppic DNA in "Italic" DNA samples(with the exception of Ardea Latini, which is a sort of middle-ground between them) looks more "northwestern" than the more "southeastern" "Tyrrhenian/Etruscan" DNA samples and the genetic outliers in "Italic" sites (Prenestini and Ardea).

Does it all hint to us a more "local" origin for the Etruscans and other Tyrrhenian peoples in the Central & South Italian peninsula, while the genetically similar Italics arose somewhere not very far, but closer to the Alps region (France-South Germany-Switzerland-North Italy)? Also, would steppe ancestry have arrived in the proto-Tyrrhenian population independently, through a different path than the Bell Beaker-like ancestry that gave birth to the Italics?

Interestingly, though, the Villanovan and the Proto-Villanovan DNA samples come off as different from each other, the latter much more "Etruscan-like", and the former more "Italic-like"(is that maybe an indication that in later times the Villanovan culture was influential enough to absorb Italic peoples in its vicinity even if they didn't really forget their language and cultural identity? We know that was the case in the Roman kingdom era).

Another curious thing that would not be unexlainable from a historical point of view is that the Ardea DNA samples look quite a bit more "Etruscan-like" than those from Bovea Ernica, Rome and Prenestini... We know that Ardea was said to be the capital of the Rutuli, whose identity is not definitely known but may have been a Tyrrhenian people akin to the Etruscans, so are we maybe seeing a lot of extant "Italicized" Rutuli people in this Iron Age Ardea aDNA?Some of the outliers are also intriguing: the Prenestini outlier DNA sample has mostly "Eastern EEF" with a lot of West Asian ancestry (affinities with Anatolia, Caucasus and Levant) and looks not very distant from the overall genetic makeup of the earlier BA Sicilian Beak (with the important addition of steppe ancestry since then, though)... was he perhaps a migrant from South Italy? The other outlier, from Ardea (Latini), is pretty similar to the Prenestini outlier in its combinaton of Steppic + Western EEF + Eastern EEF (and also pretty similar to the Sicilian Beaker if you take the steppic ancestry out), and just like the Prenestini outlier it also has a lot of Anatolian/Caucasian admixture... Maybe another South Italian migrant to Central Italy in the Iron Age?

So, what are your thoughts on this?
 
Last edited:
Some example below: one of the models I've tried and explained above. The 1st graph is complete, the 2nd one focuses on EEF + Steppic ancestry, the 3rd one focuses on distinguishing the Western EEF from the Eastern EEF.

https://imgur.com/a/ahYPovS
 

Attachments

  • Img1.jpg
    Img1.jpg
    22.9 KB · Views: 106
  • Img2.jpg
    Img2.jpg
    22.1 KB · Views: 108
  • Img3.jpg
    Img3.jpg
    20.5 KB · Views: 104
Very interesting. Have you determined which Y lineages were Italic and which Tyhrrhenean?
 
So, what are your thoughts on this?

The samples are still too few to draw conclusions or even simply speculate. The difference is only due to a few WHG points. Not to mention the fact that it is completely impossible to speculate on Y-DNA with only one Etruscan tested.

Then, there is no archaeological evidence that the Italics all came from the same places. For many linguists the separation between the Latino-Faliscan languages and the Osco-Umbrian languages took place outside Italy. In both cases, a Western European origin is hardly ever suggested. Archaeologically the only hypothesis of an origin from France-Switzerland concerns the culture of Golasecca, where inscriptions in lepontic language have been found. And there are no other Tyrrhenian peoples in the central-southern peninsula, except the Etruscans, and the only other example of a "Tyrrhenian" people in Italy are the Rhaetians living in the Alps of northern Italy.
 
The samples are still too few to draw conclusions or even simply speculate. The difference is only due to a few WHG points.

Yes, but as a preliminary and of course completely temporary analysis I think we can try our best. Yes, the difference may only be because of a slightly different genetic makeup in their EEF by the time the Late Neolithic, Copper Age or Early Bronze Age arrived (that is what I considered to distinguish "western" from "eastern" EEF, since we're talking about Iron Age people, so using samples not too far removed from those is what is most useful)... but the thing is that by the time the Etruscan and Italic populations got really formed they had to have come from a chronologically close population, not straight from Early Neolithic ANF and WHG. If an Iron Age people had more WHG:ANF ratio in its EEF, that should be explained by what Bronze Age populations they directly came from.

By the way, IMHO it isn't only WHG that distinguishes "western EEF" from "eastern EEF", there were also millennia of gradual genetic drift setting them apart from each other, not to mention that there had also certainly been some genetic structure in the WHG of Europe before, and Southeastern Europe also started to have some minor CHG/Iranian-related input in Central & South Italy and the Balkans (which did not happen in Western & Central Europe) by the Late Neolithic/Copper Age, further differentiating "western" from "eastern" non-steppe people in Europe.

Not to mention the fact that it is completely impossible to speculate on Y-DNA with only one Etruscan tested.

Yes, that's absolutely certain, but that is also the reason I speculate nothing about Y-DNA lineages here.

Then, there is no archaeological evidence that the Italics all came from the same places. For many linguists the separation between the Latino-Faliscan languages and the Osco-Umbrian languages took place outside Italy. In both cases, a Western European origin is hardly ever suggested. Archaeologically the only hypothesis of an origin from France-Switzerland concerns the culture of Golasecca, where inscriptions in lepontic language have been found.

The analysis I did above also did not pinpoint one specific location for the origin of all Italics. It just points a very broad location associated with Bell Beaker populations and with lands inhabited by people more akin to Chalcolithic/EBA Iberians, French and North Italian than to South Italian and Southeastern/Eastern European people. There is no useful aDNA sample without steppe ancestry from CA/EBA Switzerland, Austria and South Germany, but I'm pretty confident they would also be more "western" than "eastern". I think you're interpreting "western" too literally as meaning Western Europe, but that is not the case. What we see there is a cline from "mostly western" to "mostly eastern", and North Italy and Central Europe are fitting locations for the main source of "Italic EEF", too.

What I found interesting about the consistent results I got is precisely that it does fit the long-time speculation of a lot of linguists and archaeologists that Italics originally lived close to the Proto-Celtic homeland, and that both spread from South-Central Europe in the MLBA/LBA, somewhere around/near the Alps, not that far from the later centers of Hallstatt and La Tène culture. Since Proto-Celtic is widely considered to have spread from that broad region, and Proto-Italo-Celtic is widely accepted as if not an actual common proto-language at least as a long-term sprachbund/dialect continuum, an Italic with a less "eastern EEF" component than the pre-Italic peoples of Italy would make sense. It would also neatly explain the Tyrrhenian vs. Italic difference, in that they would actually have come from different populations and different places. Now, of course it's all speculative, but certainly not implausible and unrelated to previous linguistic and archaeological hypotheses.

And there are no other Tyrrhenian peoples in the central-southern peninsula, except the Etruscans, and the only other example of a "Tyrrhenian" people in Italy are the Rhaetians living in the Alps of northern Italy.

That situation in the Later Iron Age, after the expansion of Italics, Greeks, Etruscans and other peoples in the Italian peninsula, does not necessarily represent the demographic and genetic makeup of the peninsula millennia before in the Copper Age and Early Bronze Age. Much like some assume that, because Rhaetians lived in the Alps of northern Italy, Etruscans must've come originally from the north (and the speakers of Proto-Tyrrhenian might've come from there, indeed, but that is different from talking about the main genetic roots of the Etruscans, isn't it?), what we could've been seeing there is an expansion of Tyrrhenian language northward from Central Italy, which we know for a fact that did start to happen (again perhaps?) in the Iron Age with the Etruscans, but soon Celts and Italics overran their expansion.

Let me also remind you guys, lastly, that the non-steppe/EEF part of the Etruscan and Proto-Villanovan DNA samples would also fit links with Southeastern Europe, and not exclusively a local origin in Central Italy. I wonder where Lemnian fits in all of this (though I think it is more likely that it was just the remnant of an Etruscan colonial settlement, but we don't know for sure).
Ax01575.jpg
 
Last edited:
If we know the aDNA for all of these samples, why can't we speculate (however provisionally) about which Y-lineages were Italic and which ones Tyrrhenian?
 
Hi, people! I have been very busy in the last few months and unfortunately had to reduce my posts here to a minimum. Now I'm coming back to discuss with you something that I found pretty intriguing and maybe meaningful as I got access to the recently released and made available Italian aDNA samples. I have been reading a lot being said about "Italics and Etruscans being almost indistinguishable, only the language that survived in their groups differing"... but is that so?

Firstly, I modelled some of the most relevant aDNA population samples from ancient Italy using all Paleolithic, Neolithic and Chalcolithic DNA samples available in Global25 datasheets just to see what it'd give me. I also did the same for some Iberian, French and Balkanic aDNA. Interestingly, an imperfect but noticeable pattern started to emerge: the EEF in Italic-related aDNA is more related to Western Europe (Iberia and France), the EEF in Etruscan-related aDNA is more related to Eastern Europe (the Balkans and Pannonian plain in particular).

In addition to that, there's also something else that I found meaningful: Remedello and North Italian BB were predominately "Western EEF", whereas Sicilian BB (with no steppe ancestry) is wholly "Eastern EEF", and Grotta Continenza (Central Italian) DNA had an almost even split between "Western EEF" (52.6%) and "Eastern EEF" (47.4%)... that suggests that the Italian Peninsula had a subtle genetic cline previously from "Western European-like" in North Italy to "Balkanic-like" in South Italy (not really unlike today, right?).

I also tried several other models with fewer reference populations - and the same pattern still emerged every time. Another thing is that, if you add French BB samples, and compare IA Romans with IA Etruscans, only the former pick a lot of ancestry from those, with the latter "preferring" more CA Central Italian (Grotta Continenza), North Italian BB and Yamnaya ancestry than the Roman sample. If you add several Bell Beaker samples, this thing gets even more interesting: the Etruscan sample picks a steppe admixture right from Kalmykia Yamnaya (~28%), whereas IA Roman picks ancestry only from Bell Beaker samples from Germany, North Italy and France (with or without steppe admixture). In all cases, the Etruscan sample gets much more Grotta Continenza admixture than the Roman one. In general, the non-steppic DNA in "Italic" DNA samples(with the exception of Ardea Latini, which is a sort of middle-ground between them) looks more "northwestern" than the more "southeastern" "Tyrrhenian/Etruscan" DNA samples and the genetic outliers in "Italic" sites (Prenestini and Ardea).

Does it all hint to us a more "local" origin for the Etruscans and other Tyrrhenian peoples in the Central & South Italian peninsula, while the genetically similar Italics arose somewhere not very far, but closer to the Alps region (France-South Germany-Switzerland-North Italy)? Also, would steppe ancestry have arrived in the proto-Tyrrhenian population independently, through a different path than the Bell Beaker-like ancestry that gave birth to the Italics?

Interestingly, though, the Villanovan and the Proto-Villanovan DNA samples come off as different from each other, the latter much more "Etruscan-like", and the former more "Italic-like"(is that maybe an indication that in later times the Villanovan culture was influential enough to absorb Italic peoples in its vicinity even if they didn't really forget their language and cultural identity? We know that was the case in the Roman kingdom era).

Another curious thing that would not be unexlainable from a historical point of view is that the Ardea DNA samples look quite a bit more "Etruscan-like" than those from Bovea Ernica, Rome and Prenestini... We know that Ardea was said to be the capital of the Rutuli, whose identity is not definitely known but may have been a Tyrrhenian people akin to the Etruscans, so are we maybe seeing a lot of extant "Italicized" Rutuli people in this Iron Age Ardea aDNA?Some of the outliers are also intriguing: the Prenestini outlier DNA sample has mostly "Eastern EEF" with a lot of West Asian ancestry (affinities with Anatolia, Caucasus and Levant) and looks not very distant from the overall genetic makeup of the earlier BA Sicilian Beak (with the important addition of steppe ancestry since then, though)... was he perhaps a migrant from South Italy? The other outlier, from Ardea (Latini), is pretty similar to the Prenestini outlier in its combinaton of Steppic + Western EEF + Eastern EEF (and also pretty similar to the Sicilian Beaker if you take the steppic ancestry out), and just like the Prenestini outlier it also has a lot of Anatolian/Caucasian admixture... Maybe another South Italian migrant to Central Italy in the Iron Age?

So, what are your thoughts on this?


Quite interesting, is there any speculative time scale and or concrete archaeological cultures you would try couple these two differing ancestries with?
 
Yes, but as a preliminary and of course completely temporary analysis I think we can try our best. Yes, the difference may only be because of a slightly different genetic makeup in their EEF by the time the Late Neolithic, Copper Age or Early Bronze Age arrived (that is what I considered to distinguish "western" from "eastern" EEF, since we're talking about Iron Age people, so using samples not too far removed from those is what is most useful)... but the thing is that by the time the Etruscan and Italic populations got really formed they had to have come from a chronologically close population, not straight from Early Neolithic ANF and WHG. If an Iron Age people had more WGH:ANF ratio in its EEF, that should be explained by what Bronze Age populations they directly came from.By the way, IMHO it is only WHG that distinguishes "western EEF" from "eastern EEF", there were also millennia of gradual genetic drift setting them apart from each other, not to mention that there had also certainly been some genetic structure in the WHG of Europe before, and Southeastern Europe also started to have some minor CHG/Iranian-related input in Central & South Italy and the Balkans (which did not happen in Western & Central Europe) by the Late Neolithic/Copper Age, further differentiating "western" from "eastern" non-steppe people in Europe.Yes, that's absolutely certain, but that is also the reason I speculate nothing about Y-DNA lineages here.The analysis I did above also did not pinpoint one specific location for the origin of all Italics. It just points a very broad location associated with Bell Beaker populations and with lands inhabited by people more akin to Chalcolithic/EBA Iberians, French and North Italian than to South Italian and Southeastern/Eastern European people. There is no useful aDNA sample without steppe ancestry from CA/EBA Switzerland, Austria and South Germany, but I'm pretty confident they would also be more "western" than "eastern". I think you're interpreting "western" too literally as meaning Western Europe, but that is not the case. What we see there is a cline from "mostly western" to "mostly eastern", and North Italy and Central Europe are fitting locations for the main source of "Italic EEF", too.

An analysis to play and have fun, okay. To be able to do what you would like to do, we would need such a deep knowledge of the Etruscan and Italic world, that obviously we both don't have.

I think that there is a basic error in your reasoning: trying to make prehistoric Neolithic and Bronze cultures perfectly coincide with ethnic groups that only emerge with the Iron Age. The same prehistoric cultures may have contributed to the formation of more ethnos, and not always in the same way. The idea that there are ethnic differentiations in prehistoric cultures based on similar ancestral components is borrowed from the concept of Iron Age ethnos and is not always applicable. Another basic error is believing that people all belonging to the same ethnic group are all identical replicas and so that differences may always imply different origins. Even among all the Bell Beaker samples analysed so far there is a signifcant variation with WHG that ranges from 28% to 2%.

Coming to your nMonte/G25 analysis, it depends very much on how you did it, whether you used single results or averages. With the averages you have to exclude all those not considered fully native, and so only 4 Latin individuals can be used (R851, R1016, R1021, R435), and the Etruscans that you should use are three (R1015, R474, R473), because obviously you have also to include Villanovan R1015 which is clearly Etruscan. R1015 has the same percentage of WHG as the Latins with the highest WHG. Since this is also the oldest of the Etruscans, it is not suggesting that there were differences between Latins and Etruscans in their non-steppe part. The difference is due to the two more recent Etruscan samples who have 2.4% less WHG than the average of Latins, but also within the Latins the WHG fluctuates, with the higher difference between two Latin samples being 2%. Obviously with a much larger number of ancient samples from Italy, and not only of Latins and Etruscans, the small individual differences could increase significantly, making everything even more confusing and less clear.

With a very basic model using nMonte/G25, the Latins have 62.4% EEF, 11.9% WHG, 25.5% Steppe, while the Etruscans have 63.7% EEF, 10.1% WHG, 26.2% Steppe. Of course, based on the unsubstantiated assumption that the results of nMonte/G25 are accurate. In any case the differences are small, Latins have 1.8% more WHG, 0.7% less Steppe and 1.3% less EEF than Etruscans. Too little to justify very different origins, it is not these numbers that can support this.

Considering that the Protovillanovan sample R1 (EEF 59.5%, WHG 2.6%, Steppe 37.6%) has much less WHG than both Latins and Etruscans, the explanation could be much simpler than one believes. One can't forget the Proto-Villanovan sample is most likely a proto-Italic sample, specifically a Proto-Picene who spoke an Osco-Umbrian language. The real question one should ask is why this Proto-Italic sample is so different from the Latins? Much lower WHG and higher Steppe. The explanation that this sample might be of recent Balkan origin is only a guess.

To understand all the steps, we need a really large number of samples from anyhwere in Italy, especially Bronze age samples. These current samples are not enough to reconstruct everything in detail, which in any case must follow what archaeology has found and suggests. Not to mention that the resurgence of WHG could not be due to different origins but to how this resurgence happened locally and even individually.


That situation in the Later Iron Age, after the expansion of Italics, Greeks, Etruscans and other peoples in the Italian peninsula, does not necessarily represent the demographic and genetic makeup of the peninsula millennia before in the Copper Age and Early Bronze Age. Much like some assume that, because Rhaetians lived in the Alps of northern Italy, Etruscans must've come originally from the north (and the speakers of Proto-Tyrrhenian might've come from there, indeed, but that is different from talking about the main genetic roots of the Etruscans, isn't it?), what we could've been seeing there is an expansion of Tyrrhenian language northward from Central Italy, which we know for a fact that did start to happen (again perhaps?) in the Iron Age with the Etruscans, but soon Celts and Italics overran their expansion. Let me also remind you guys, lastly, that the non-steppe/EEF part of the Etruscan and Proto-Villanovan DNA samples would also fit links with Southeastern Europe, and not exclusively a local origin in Central Italy. I wonder where Lemnian fits in all of this (though I think it is more likely that it was just the remnant of an Etruscan colonial settlement, but we don't know for sure).


There's too much unproven speculation in here. Who determined that the non-steppe/EEF part of the Etruscan would also fit links with Southeastern Europe? nMonte-G25? I don't see a smoking gun, honestly.


What I found interesting about the consistent results I got is precisely that it does fit the long-time speculation of a lot of linguists and archaeologists that Italics originally lived close to the Proto-Celtic homeland, and that both spread from South-Central Europe in the MLBA/LBA, somewhere around/near the Alps, not that far from the later centers of Hallstatt and La Tène culture. Since Proto-Celtic is widely considered to have spread from that broad region, and Proto-Italo-Celtic is widely accepted as if not an actual common proto-language at least as a long-term sprachbund/dialect continuum, an Italic with a less "eastern EEF" component than the pre-Italic peoples of Italy would make sense. It would also neatly explain the Tyrrhenian vs. Italic difference, in that they would actually have come from different populations and different places. Now, of course it's all speculative, but certainly not implausible and unrelated to previous linguistic and archaeological hypotheses.


The Italo-Celtic theory is not defined, it is only reconstructive and is not accepted by all linguists. But let's admit that there is something true in this theory, the Proto-Italo-Celtic homeland is never supposed to be in the West but rather in the East with a time of separation around 2500 BC, more than 1000 years before these Latin samples tested.

I'm also convinced that Bell Beaker may have played a role. But how? Archaeologically it is more attested in Etruria, particularly in northern Etruria, than in Latium vetus.

Even if all this must be taken with a grain of salt, because the distances vary depending on the tool that is used, Latin RMPR435b is indeed similar to Bell_Beaker_ITA:I2478, but Etruscan RMPR1015 is somewhat similar to Bell_Beaker_ITA:I1979 (Etruscan RMPR1015 has more WHG than Bell_Beaker_ITA:I1979) and also to some Bell_Beaker from Hungary and Germany.
acJkS.jpg

This map is outdated. For example, today we know that the area around Felsina (modern-day Bologna) did not become Etruscan thanks to an VII century BC expansion that is based on an old theory questioned by archaeologists. That area has been Etruscan at least since 900 BC. Just as the area to the south exaggerates the Etruscan presence in Campania, which was relegated to a slice of land but since the 900 BC. Other boundaries are also questioned by archaeologists.
 
This map is outdated. For example, today we know that the area around Felsina (modern-day Bologna) did not become Etruscan thanks to an VII century BC expansion that is based on an old theory questioned by archaeologists. That area has been Etruscan at least since 900 BC. Just as the area to the south exaggerates the Etruscan presence in Campania, which was relegated to a slice of land but since the 900 BC. Other boundaries are also questioned by archaeologists.
Can you give the sources and/or elaborate? How do we know borders or political control through archeology without written evidence?
With a very basic model using nMonte/G25, the Latins have 62.4% EEF, 11.9% WHG, 25.5% Steppe, while the Etruscans have 63.7% EEF, 10.1% WHG, 26.2% Steppe. Of course, based on the unsubstantiated assumption that the results of nMonte/G25 are accurate. In any case the differences are small, Latins have 1.8% more WHG, 0.7% less Steppe and 1.3% less EEF than Etruscans. Too little to justify very different origins, it is not these numbers that can support this.

What references did you use there? Using Yamnaya samples Latins have more Steppe than Etruscans.
 

This thread has been viewed 8550 times.

Back
Top