Image via Razib Khan.
A new ancient DNA study, Fulya Eylem Yediay et al., Ancient genomics support deep divergence between Eastern and Western Mediterranean Indo-European languages, bioRxiv (December 2, 2024), shows a deep rooted divide between the Western Mediterranean in Iberia, Southern France and Italy, and the Eastern Mediterranean in Greece, with a jumble in the Balkans.
The abstract of the preprint is as follows:
The Indo-European languages are among the most widely spoken in the world, yet their early diversification remains contentious. It is widely accepted that the spread of this language family across Europe from the 5th millennium BP correlates with the expansion and diversification of steppe-related genetic ancestry from the onset of the Bronze Age. However, multiple steppe-derived populations co-existed in Europe during this period, and it remains unclear how these populations diverged and which provided the demographic channels for the ancestral forms of the Italic, Celtic, Greek, and Armenian languages.
To investigate the ancestral histories of Indo-European-speaking groups in Southern Europe, we sequenced genomes from 314 ancient individuals from the Mediterranean and surrounding regions, spanning from 5,200 BP to 2,100 BP, and co-analysed these with published genome data. We additionally conducted strontium isotope analyses on 224 of these individuals.
We find a deep east-west divide of steppe ancestry in Southern Europe during the Bronze Age. Specifically, we show that the arrival of steppe ancestry in Spain, France, and Italy was mediated by Bell Beaker (BB) populations of Western Europe, likely contributing to the emergence of the Italic and Celtic languages. In contrast, Armenian and Greek populations acquired steppe ancestry directly from Yamnaya groups of Eastern Europe. These results are consistent with the linguistic Italo-Celtic and Graeco-Armenian hypotheses accounting for the origins of most Mediterranean Indo-European languages of Classical Antiquity. Our findings thus align with specific linguistic divergence models for the Indo-European language family while contradicting others. This underlines the power of ancient DNA in uncovering prehistoric diversifications of human populations and language communities.
While genes do not necessarily match languages, and pots are not people, the correlations between population genetics, language, and material culture tends to be strong, especially in more ancient times.
This analysis places the question of the baseline model into which controversy over the answer to the origins of the now extinct Anatolian languages must be fit on a more sturdy footing.
This new paper is analyzed at Bernard's blog (quoted below in English per Google translate with some obvious translation errors corrected, all images from the new article except as noted):
Bronze Age individuals from Italy are grouped into three distinct clusters. The first is the group linked to the Bell Beaker common to individuals from France and Spain. It includes all the ancient individuals from Corsica and central Italy. A second group is more linked to the [first] farmers of Europe. It includes the ancient individuals from the Olmo site in northern Italy. The third group is linked to the Yamnaya group and includes the ancient individuals from the Adriatic coast.
The Adriatic coast group is probably an early precursor of maritime Magna Graecia colonization. The people with first farmer genetics are probably the ancestors of the Etruscans and kindred non-Indo-European ethnicities of the region. These genetics generally support the Italo-Celtic grouping of Indo-European languages.
Map from Wikipedia
The spread of the Neolithic to the Caucasus and Iran contributed to the diffusion of Anatolian farmer ancestry into this region, which mixed with local hunter-gatherer ancestry from the Caucasus CHG. In addition, the expansion of the Kuro-Araxes culture during the third millennium BC connected the Caucasus with the Levant and Mesopotamia through trade networks. Interaction between Anatolia and the Caucasus increased during the Chalcolithic and Bronze Ages, leading to the expansion of the CHG component into these regions and beyond into the Mediterranean.
In this study, the authors analyzed the genomes of 25 Bronze and Iron Age individuals from Anatolia. They all originate from a genetic mixture between a local Anatolian farmer component, a Caucasian hunter-gatherer component and a small proportion of the Iranian farmer component. In addition, some individuals have a small proportion of Eastern European hunter-gatherer (EHG) ancestry. Both CHG and EHG components are higher in the Iron Age than in the Bronze Age. Thus, in the Iron Age, the authors observed the arrival of a steppe ancestry in central Anatolia identical to that present in the Balkans and Greece, suggesting a migration from the latter region to Anatolia. The authors hypothesize that this migration is linked to the emergence of the Phrygian state in the second millennium BC.
This suggests Indo-European migration into Anatolia from the west via Greece, rather than the East, via the Caucuses.
The ancient individuals from Cyprus suggest that this island and especially the coastal cities were a genetic melting pot during the Bronze Age. They have a genetic profile close to those of ancient individuals from Lebanon and eastern Anatolia. However, one individual is genetically close to the early farmers of Anatolia, another has a genetic profile close to ancient individuals from the Balkans and Greece.
An ancient individual with the genetic profile of ancient individuals from Scandinavia is also found at this time in a rock-cut tomb at the Vounous Bellapais site dated between 2000 and 1800 BC. Interestingly, this man is also of the Y chromosome haplogroup: I1 typical of Scandinavia. His strontium isotope analysis confirms his foreign origin consistent with Scandinavia. This result suggests very long-distance interactions on the island of Cyprus during the Bronze Age.
In the following period, during the Iron Age, the population of Cyprus has a more homogeneous genetic profile containing a small proportion of Yamnaya ancestry. . . .
In conclusion, this study shows that the results of paleogenetics support the Italo-Celtic and Greco-Armenian linguistic models and disqualify the Indo-Greek and Italo-Germanic models.
Thus, the oldest occurrence of steppe ancestry in Italy comes from two individuals from Latium dated 2100 BC, linked to the Bell Beaker group and not to the Yamnaya group, similarly to the Celtic populations of western Europe.
In Greece, the oldest occurrence of steppe ancestry is dated 2200 BC in individuals directly linked to the Yamnaya group. They do not have a component from the globular amphora culture. This arrival precedes the emergence of the Greek language in the form of Linear B writing.
Greek is thus, more basal, within the Indo-European linguistic family, deriving directly from the Yamnaya people, rather than being mediated through intermediate Corded Ware and Bell Beaker people.
The Armenian language has been attested for about 1550 years in the southern Caucasus and eastern Anatolia. At the end of the Iron Age this region was under the control of the Urartian kingdom . This state was culturally diverse and contained Armenian linguistic elements suggested by borrowings between the two languages Urartian and Armenian. Steppe ancestry has been detected in several individuals from this region in the Bronze Age at the end of the third millennium BC coinciding with the fall of the Kuro-Araxes culture. These individuals with steppe ancestry are related to the Yamnaya group, like the ancient Greeks.
Armenian is probably hard to classify because it is a boundary of deeply linguistically divided branches of the Indo-European language family as well as non-Indo-European substrate languages. But the genetics of ancient Armenians suggest an ancestral Greek origin as a start point for its ultimate linguistic mix.
Huh, I read the paper (if not all the supplements) and I recall that they hypothesized that Greek and Armenian had the same Yamnaya root, not that Greek was upstream of Armenian.
ReplyDeleteYou are correct that they didn't correct the final dot. But given the geography, that is pretty much the only route that makes sense.
ReplyDeleteI don't see how an increase in Balkan-derived steppe ancestry in Iron Age Anatolia has anything to do with the origins of Anatolian speakers? Phyrigan is not part of the Indo-Anatolian language family to my knowledge, and that is awfully late for the arrival of Hittites and other Indo-Anatolians.
ReplyDelete@Ryan Your right (at least mostly). Personally, I think that the Indo-Anatolians probably arrived (with steppe ancestry) at a time roughly contemporaneous with the proto-Greek Indo-Europeans (ca. 2000-2200 BCE), not in the Iron Age (ca. 900 BCE) as steppe ancestry would suggest, or much earlier (the Middle to Late Neolithic) as urged by linguists but with almost no archaeological or genetic or historical support. And, the ancient DNA data from Late Neolithic and early Bronze Age Anatolia is still pretty thin. On the other hand, there is a decent case to be made that the Indo-European Anatolian languages were more of a product of elite dominance (a bit like the Magyars in Hungary), than one of demic replacement/significant displacement as seen in most other areas transitioning to Indo-European languages. On the other hand, it could be that Indo-Europeans somehow had a demic advantage whose seeds were sown in the Bronze Age when they practiced elite dominance, but which didn't real reap results until the Iron Age (e.g. maybe non-steppe ancestry ruled people took a harder hit during Bronze Age collapse than Hittite rulers with steppe ancestry). The latest paper doesn't address the Anatolian language origins question.
ReplyDeleteEven 2,200 BC may be late given how divergent Anatolian langues are from the rest of IE. If they weren't already in Anatolia by 3,000 BC then they were separated from the rest of IE some other way.
DeleteBut nobody is suggesting the speakers of the Anatolian (Hittite etc) branch showed up in the Iron Age, as that's chronologically impossible.
ReplyDeleteThe new paper shows steppe ancestry in Early Bronze Age Anatolia, with 2 samples from this period having the I2a-L699 patriline, which is described in the Genetic Origins paper from earlier this year as "an important lineage in the Dnipro area since the Neolithic hunter-gatherer period, continued to be prevalent among the Serdenii Stih, and in the Don Yamnaya was dominant (17/20 instances). This pre-Yamnaya steppe ancestry (modeled fairly successfully by Eneolithic Piedmont Steppe per the supplement ) matches the early divergence of the Anatolian languages from the rest of Indo-European languages
The Iron Age pulse of steppe ancestry into Anatolia would be the arrival of Phrygian speakers, whose variant of late Yamnaya ancestry resembling that of the Moldova is shared with the Greeks, which is in keeping with the increased linguistic support for the Greco-Phrygian node in the family tree. Armenians likewise share this variant of steppe ancestry, but this Late Yamnaya/Catacomb ancestry crosses the Caucasus, not Anatolia, as was shown as far back as the Southern Arc and Genetic probe papers in 2022. This fits with the diminished support for Armenian and Phrygian sharing a closer relationship than Greek and Phrygian (as per Obrador-Cursach) and fits the layers of loanwords in Armenian discussed by Rasmus Thorsø Nielsen in his recent dissertation.