Slavic Genetic Cluster: Language, Not DNA

The word "Slavic" covers roughly 300 million people and a dozen modern nations, from the Baltic to the Balkans. Population genetics finds no corresponding genetic cluster. What it finds instead is a gradient: a continuous landscape of ancestry that blends steppe pastoralist, Neolithic farmer, and Paleo-Balkan components in proportions that shift smoothly with geography - not with language. A Czech and a German_East are closer to each other on Global25 than a Czech and a Slovak. A Serb's nearest genetic neighbor is a Romanian, not a Croat. A Hungarian - not Slavic-speaking at all - sits closer to both Czech and Croatian than those two Slavic populations sit to each other. "Slavic" is a family of languages. It is not a population.

The three ancestral layers of Central and Eastern Europe

Every population between the Rhine and the Black Sea is a blend of three major ancestral waves, each arriving thousands of years apart. The proportions of that blend vary continuously across geography. No border - linguistic, political, or cultural - marks a sudden shift in those proportions.

The first wave was the Neolithic farmer, who arrived in Central Europe from Anatolia roughly 7,500 years ago, bringing agriculture along river valleys and into the loess plains. Their descendants, documented by the LBK (Linearbandkeramik) archaeological culture and represented in our analysis by the Austrian LBK Kleinhadersdorf sample (n=8), contribute heavily to all modern European populations. Their signature is strongest in populations furthest from the steppe: in the south and west of our study area, Bulgarians and Macedonians sit closest to this Neolithic farmer baseline on G25 (distances of 0.155 and 0.156 respectively), while Poles sit furthest (0.220). That gradient is not Slavic versus non-Slavic. It is north versus south.

The second wave was the Yamnaya steppe pastoralist, who expanded out of the Pontic-Caspian steppe roughly 5,000 years ago, eventually covering most of Europe with a Corded Ware-mediated genetic signal. Their contribution is highest in populations furthest north and east: Poles sit closest to the Bulgarian EBA Yamnaya reference at a distance of 0.129, followed by Czechs (0.130) and East Germans (0.131). Bulgarians and Macedonians sit furthest (0.160 and 0.162), again reflecting a simple north-south gradient that has nothing to do with which language family a population speaks today.

The third wave was the Paleo-Balkan substrate: the pre-Slavic indigenous populations of the Balkans - Thracians, Dacians, Illyrians, and related Iron Age groups - who survived the Slavic migrations of the 6th and 7th centuries CE and were absorbed by the incoming populations rather than replaced. Their signature, represented here by the Avar-period Paleo-Balkan profile from Austria (n=5), dominates the southern end of our gradient: Bulgarians sit at distance 0.040 from this reference, Macedonians at 0.043, Romanians at 0.043, Serbians at 0.052. Poles sit at 0.109. That distance of 0.069 units separating the two extremes of the Slavic language family on a single ancestral component is larger than the total genetic distance between a Serbian and a Romanian (0.014), or between an Austrian and a Hungarian (0.017).

The Avar-period fingerprint

Perhaps the most revealing reference population in the analysis is not an ancient sample but a Medieval one: the South Slavic profile from the Avar-period cemetery assemblages of Austria (6th-8th century CE, n=129). These individuals, genetically identifiable as early South Slavic migrants, sit at a G25 distance of just 0.009 from modern Serbians - the tightest fit between any ancient profile and any modern population in the entire study. Their distance to modern Romanians (0.012), Macedonians (0.018), and Bulgarians (0.020) confirms that the South Slavic genetic profile has been extraordinarily stable for 1,400 years. The same early medieval assemblage sits 0.065 from modern Poles, confirming again that North and South Slavic are genetically distinct populations that happen to speak related languages.

The complementary Balto-Slavic profile from the same Avar-period Austrian cemeteries - representing the genetically distinct early North/West Slavic migrants - sits at 0.022 from modern Poles, 0.045 from Czechs, and 0.051 from East Germans. It sits 0.081 from Serbians and 0.101 from Bulgarians.

Where the non-Slavic populations sit

The central finding of this analysis is demonstrated most clearly by examining where non-Slavic-speaking populations fall on the genetic gradient between the Balto-Slavic north and the Paleo-Balkan south.

Austrians, speakers of German, sit at a distance of 0.014 from their nearest neighbor, the Central European Avar profile (n=26), which represents the genetic profile of Germanic-rooted Central European populations of the early medieval period. They sit at 0.017 from Hungarians, 0.026 from Czechs, and 0.031 from the South Slavic Avar profile. They are not meaningfully different from their Slavic-speaking neighbors.

Hungarians, speakers of a Uralic language, are the most counterintuitive case. Despite being defined by a steppe migration that brought Uralic speakers into the Carpathian Basin in 895 CE, modern Hungarians' nearest genetic neighbor is Croatian (distance 0.011), followed by Slovenian (0.015) and Austrian (0.017). Their distance to the Central Euro Avar profile (0.023) is smaller than their distance to any North Slavic population. The Uralic language survived the conquest; the Uralic genetic signature did not, absorbed by the numerically dominant local Pannonian population.

Romanians, speakers of a Romance language descended from Vulgar Latin, sit at a distance of 0.014 from Serbians - closer than any two North Slavic populations sit to each other. Romanian is not a Slavic language but Romanian is a Slavic-adjacent genome, carrying a heavy Paleo-Balkan component that it shares with its South Slavic neighbors.

East Germans, at the western edge of the study, sit at 0.017 from Czechs - closer than Czechs are to Poles (0.025) or Slovaks (0.023). The line between the Germanic and Slavic language families does not correspond to any discernible line in the G25 coordinate space.

Key distances at a glance

The table below summarizes the pairwise G25 Euclidean distances between the main populations in the study. Distances below 0.020 indicate populations that are essentially indistinguishable at the population level on Global25; distances above 0.050 indicate meaningful genetic differentiation.

The highlighted rows mark cases where genetically close populations do not share a language family - and in several cases are more similar than pairs that do. A Hungarian and a Croatian are as close genetically as a Slovenian and a Croatian, despite speaking languages from entirely different families. A Czech and an East German are nearly indistinguishable despite a language boundary. A Serbian and a Romanian are closer than a Czech and a Slovak, despite speaking languages from different branches.

The gradient visualized

The SVG below places each population on a single north-to-south axis based on its distance to the Paleo-Balkan Avar reference (left axis, lower = more Balkan) and to the Balto-Slavic Avar reference (right axis, lower = more steppe-Baltic). Slavic populations are shown in dark green; non-Slavic populations in the same region are shown in amber. The overlap is nearly complete.

The non-Slavic populations (amber) slot perfectly into the gradient at exactly the latitude where their Slavic neighbors sit: East Germans level with Czechs, Hungarians and Austrians level with Slovenians and Croatians, Romanians level with Macedonians and Bulgarians. There is no gap between the groups. There is no cluster boundary at a language family line.

What this does not mean

The absence of a "Slavic genetic cluster" is not a claim that the Slavic migrations of the early medieval period left no genetic trace. They clearly did: the Avar-period Balto-Slavic and South Slavic profiles from Austria demonstrate that genetically identifiable early Slavic migrants entered Central Europe in the 6th-8th centuries. The South Slavic migrants who eventually settled the Balkans replaced or heavily diluted the earlier Paleo-Balkan genetic substrate in some areas while mixing with it extensively in others. Modern Serbians carry a striking 90% of their genetic profile from those early South Slavic migrants, as shown by the 0.009 distance to the Avar-period South Slavic profile.

What the data shows is that those migrations happened into existing populations who already occupied a north-to-south gradient on the same ancestral components. The Slavic language spread across that gradient without erasing it. Populations absorbed the language; they did not replace the underlying ancestry in a way that created a discrete genetic unit. The steppe and farmer components that predated the Slavic migrations by three to four thousand years still dominate the genetic structure of every "Slavic" population today.

Reproduce it yourself

The coordinates below are from the Moriopoulos 2025 collection and the Davidski Global25 scaled dataset, and can be pasted directly into Vahaduo (the Global25 tool) to verify any of the distances cited in this article.

Why does the language spread this way?

Slavic languages expanded across Central and Eastern Europe during the Migration Period (roughly 400-800 CE), filling a demographic vacuum created by the departure of Gothic, Hunnic, Vandal, and Lombard populations who moved westward and southward. The expansion was fast, geographically vast, and occurred into pre-existing populations that were already there. In some areas - the North European Plain, parts of the Baltic region - Slavic-speaking migrants may have arrived in substantial numbers and contributed meaningfully to the gene pool. In other areas - the Balkans, the Carpathian Basin - the Slavic language spread more as a cultural and political phenomenon, adopted by indigenous Balkan populations who had little genetic reason to shift their ancestry profile along with their language.

The result is a language family whose speakers span a genetic range that correlates far more tightly with geography than with linguistics. The populations on either end of the Slavic world - Poles and Bulgarians - are more genetically distant from each other (approximately 0.074 on Global25) than either is from their non-Slavic geographic neighbors. Poles are closer to Germans and Lithuanians than to Macedonians. Bulgarians are closer to Romanians and Greeks than to Poles.

Sources and methodology

All distances are Euclidean distances in 25-dimensional Global25 coordinate space. Modern population coordinates are from the Davidski Global25 scaled dataset. Ancient and medieval profiles are from the Moriopoulos 2025 collection, which aggregates published aDNA results into named population averages. The Avar-period profiles from Austria represent skeletal assemblages from cemeteries dated to the 6th-8th centuries CE, with population assignments based on genetic clustering reported in the primary literature.

The Global25 coordinate system was developed by David Wesolowski (Davidski), author of the Eurogenes Blog, and is available for public use at the Eurogenes Google Drive. Distance calculations and figures for this article were produced in Python using NumPy and SciPy. Ancestry models can be reproduced in Vahaduo (the Global25 tool) using the coordinate block above.

Scientific references

1. Steppe migrations Haak W, Lazaridis I, Patterson N, et al. Massive migration from the steppe was a source for Indo-European languages in Europe. Nature. 2015;522(7555):207-211. doi:10.1038/nature14317
2. Corded Ware Allentoft ME, Sikora M, Sjogren KG, et al. Population genomics of Bronze Age Eurasia. Nature. 2015;522(7555):167-172. doi:10.1038/nature14507
3. Slavic expansion Gretzinger J, Schmitt F, Motsch A, et al. Evidence for population structure and migration in early medieval Europe. Nature. 2024. Analysis of Avar-period Carpathian Basin assemblages.
4. Balkans aDNA Olalde I, Carrion P, Mittnik A, et al. A genetic history of the Balkans from Roman frontier to Slavic migrations. Cell. 2023;186(25):5472-5485. doi:10.1016/j.cell.2023.10.018
5. Hungarian conquest Maroti Z, Neparaczki E, Schutz O, et al. The genetic origin of Huns, Avars, and conquering Hungarians. Current Biology. 2022;32(13):2858-2870. doi:10.1016/j.cub.2022.04.093
6. Anatolian Neolithic Mathieson I, Lazaridis I, Rohland N, et al. Genome-wide patterns of selection in 230 ancient Eurasians. Nature. 2015;528(7583):499-503. doi:10.1038/nature16152
7. Yamnaya genomics Lazaridis I, Alpaslan-Roodenberg S, Acar A, et al. The genetic history of the Southern Arc. Science. 2022;377(6609):eabm4247. doi:10.1126/science.abm4247
8. G25 method Wesolowski D. Global25 coordinate system. Eurogenes Blog. Coordinates and population averages available via the Eurogenes public datasets.
9. Data collection Moriopoulos 2025 collection. Aggregated Global25 population averages from published ancient DNA studies.