Skip to content
inkglot

Indo-European homeland

Debate

Where and when was Proto-Indo-European spoken?

Most specialists place the core of the family, everything except Anatolian, with Yamnaya-related herders on the Pontic–Caspian steppe about 6,500–5,500 years ago. Where the Anatolian branch joins that story, and how far back the root goes, are open: recent ancient-DNA work and Bayesian phylogenies point in different directions.

  • Proto-Indo-European was spoken by Yamnaya-related pastoralists on the Pontic–Caspian steppe around 4500–3500 BCE, and its speakers spread west and east from there.

    Evidence for

    • LinguisticShared inherited words for wheel, axle, wagon, wool and horse date the parent language to after these were in use.3
    • GeneticAncestry from the steppe entered central Europe around 3000 BCE with the Corded Ware culture.4
    • PhylogeneticTreating ancient languages as ancestors moves the Bayesian root to about 6,500–5,500 years ago.5

    Evidence against, critiques

    • PhylogeneticOther Bayesian analyses date the first splits earlier, about 8,000 years ago or more.67
    • GeneticHittites and other Bronze Age Anatolians lack the steppe hunter-gatherer ancestry typical of Yamnaya.8
    Held by
    Held by Anthony & Ringe 2015; Haak et al. 2015
    Sources
    Anthony & Ringe 2015; Haak et al. 2015
  • Proto-Indo-Anatolian was spoken by Caucasus–Lower Volga people around 4400–4000 BCE; Anatolian split off southward and the rest of the family came from Yamnaya.

    Evidence for

    • GeneticAncient DNA from about 430 individuals shows a Caucasus–Lower Volga cline that contributed about 80% of Yamnaya ancestry and about 10% of Bronze Age Anatolian ancestry.8
    • GeneticEarlier work on the “Southern Arc” traced gene flow between West Asia and south-eastern Europe.9

    Evidence against, critiques

    • PhylogeneticBayesian analyses of the languages date the first splits earlier, about 8,000 years ago or more.67
    • LinguisticHeggarty, a linguist, rejected the hypothesis that the first Indo-European speakers came from this region as “smoke and mirrors”.10
    Held by
    Held by Lazaridis et al. 2025
    Sources
    Lazaridis et al. 2025; Lazaridis et al. 2022
  • The family began splitting about 8,100 years ago south of the Caucasus; the steppe was a secondary homeland for several branches from about 7,000 years ago.

    Evidence for

    • PhylogeneticA new dataset of 161 languages, analysed with ancient languages as sampled ancestors, gives a root of about 8,100 years.7
    • GeneticAncient DNA shows gene flow between West Asia and south-eastern Europe along a “Southern Arc”, a possible route for early branches.9

    Evidence against, critiques

    • PhylogeneticGarrett called the new analysis “unconvincing”, while valuing its database.11
    • LinguisticThomason: the proposed 7,000 years of divergence “is still in conflict with the evidence from linguistic archaeology”.11
    • PhylogeneticConstraining ancient languages as ancestors yields a root of about 6,500–5,500 years.5
    Held by
    Held by Heggarty et al. 2023
    Sources
    Heggarty et al. 2023
  • Indo-European spread out of Anatolia with Neolithic farming; Bayesian phylogeography puts the root at 7,116–10,410 years ago (95% interval).

    Evidence for

    • PhylogeneticBayesian phylogeography of 103 languages places the root in Anatolia 8,000–9,500 years ago.6
    • ArchaeologicalFarming spread from Anatolia into Europe in the Neolithic, a known route for a population movement.12

    Evidence against, critiques

    • GeneticAncient DNA shows a large steppe-derived population movement into Europe around 3000 BCE.4
    • PhylogeneticConstraining ancient languages as ancestors yields a younger root.5
    Held by
    Held by Renfrew; Bouckaert et al. 2012
    Sources
    Bouckaert et al. 2012; Wikipedia: Anatolian hypothesis

What would settle it

  • Ancient DNA from Anatolia and the Caucasus before 4000 BCE that links a population to the earliest Anatolian speakers.
  • Agreement on how Bayesian models should treat ancient languages and borrowing.
History of ideas: proposals specialists reject

Out of IndiaFRINGE— Rejected by most specialists.

Proto-Indo-European originated in South Asia and spread outward.

Why it is rejected: The linguistic evidence points elsewhere: inherited vocabulary for wheeled vehicles and early contacts with Uralic place Proto-Indo-European north of the Black and Caspian seas (Anthony & Ringe 2015). Ancient DNA is consistent with this: steppe-related ancestry reaches South Asia around 2000–1500 BCE and is absent from the Rakhigarhi Harappan genome (Narasimhan et al. 2019; Shinde et al. 2019). Specialists do not accept a South Asian homeland. 31213

Genetic evidence shows where people and their ancestry moved. It does not show which language they spoke: linking ancestry to a language family is always an inference from several kinds of evidence together. As the geneticist Mait Metspalu put it: “Genes don't tell us anything about language, period.”10

Where each position puts the homeland

Loading map…

Schematic regions drawn for Inkglot from the cited descriptions; not boundaries. Colour and glyph follow each position's label. Select a region to find its card.Natural Earth (public domain)

When the family split, by published trees

Ages of selected splits, read from each study's own dated tree with its 95% interval. Studies with different methods give different dates; compare them on one axis.

Model estimates
  1. Indo-European (root)Hittite ↔ English
  2. Non-Anatolian Indo-EuropeanTocharian B ↔ English
  3. AnatolianHittite ↔ Lycian
  4. Indo-IranianVedic Sanskrit ↔ Avestan
  5. Balto-SlavicLithuanian ↔ Russian
  6. CelticOld Irish ↔ Welsh
  7. GermanicGothic ↔ English
  8. RomanceRomanian ↔ French

Years before present

Select a bar for its median, 95% interval and source.

Tree: Bouckaert et al. 2012, via Phlorest (CC BY 4.0). Bayesian phylogeography (Phlorest summary tree), 103 languages in the tree.Bayesian phylogenies estimate dates with uncertainty; they are one line of evidence. Some nodes in these studies are calibrated with historical dates, so ranges near them are narrow by design. “BP” counts back from the study's present.
View as table
SplitStudyMedian95% interval
Indo-European (root) (Hittite ↔ English)Bouckaert et al. 20128,470 BP10,340 BP – 7,070 BP
Non-Anatolian Indo-European (Tocharian B ↔ English)Bouckaert et al. 20126,830 BP8,100 BP – 5,670 BP
Anatolian (Hittite ↔ Lycian)Bouckaert et al. 20125,300 BP6,440 BP – 4,400 BP
Indo-Iranian (Vedic Sanskrit ↔ Avestan)Bouckaert et al. 20124,880 BP5,730 BP – 4,260 BP
Celtic (Old Irish ↔ Welsh)Bouckaert et al. 20122,870 BP3,540 BP – 2,240 BP
Balto-Slavic (Lithuanian ↔ Russian)Bouckaert et al. 20123,120 BP3,400 BP – 2,670 BP
Germanic (Gothic ↔ English)Bouckaert et al. 20122,440 BP2,890 BP – 2,080 BP
Romance (Romanian ↔ French)Bouckaert et al. 20121,990 BP2,230 BP – 1,730 BP

Proposed deeper relationships

Indo-European on the family tree · All homeland debates · Macro-family proposals

Sources

  1. 1.Narasimhan, Vagheesh M., Patterson, Nick, Moorjani, Priya, et al. (2019). The formation of human populations in South and Central Asia. Science 365(6457): eaat7487. doi:10.1126/science.aat7487
  2. 2.Shinde, Vasant, Narasimhan, Vagheesh M., Rohland, Nadin, et al. (2019). An Ancient Harappan Genome Lacks Ancestry from Steppe Pastoralists or Iranian Farmers. Cell 179(3): 729–735. doi:10.1016/j.cell.2019.08.048
  3. 3.Anthony, David W., Ringe, Don (2015). The Indo-European Homeland from Linguistic and Archaeological Perspectives. Annual Review of Linguistics 1: 199–219. doi:10.1146/annurev-linguist-030514-124812
  4. 4.Haak, Wolfgang, Lazaridis, Iosif, Patterson, Nick, et al. (2015). Massive migration from the steppe was a source for Indo-European languages in Europe. Nature 522: 207–211. doi:10.1038/nature14317
  5. 5.Chang, Will, Cathcart, Chundra, Hall, David, Garrett, Andrew (2015). Ancestry-constrained phylogenetic analysis supports the Indo-European steppe hypothesis. Language 91(1): 194–244. doi:10.1353/lan.2015.0005
  6. 6.Bouckaert, Remco, Lemey, Philippe, Dunn, Michael, et al. (2012). Mapping the Origins and Expansion of the Indo-European Language Family. Science 337(6097): 957–960. doi:10.1126/science.1219669
  7. 7.Heggarty, Paul, Anderson, Cormac, Scarborough, Matthew, et al. (2023). Language trees with sampled ancestors support a hybrid model for the origin of Indo-European languages. Science 381(6656): eabg0818. doi:10.1126/science.abg0818
  8. 8.Lazaridis, Iosif, Patterson, Nick, Anthony, David, et al. (2025). The genetic origin of the Indo-Europeans. Nature 639: 132–142. doi:10.1038/s41586-024-08531-5
  9. 9.Lazaridis, Iosif, Alpaslan-Roodenberg, Songül, Acar, Ayşe, et al. (2022). The genetic history of the Southern Arc: A bridge between West Asia and Europe. Science 377(6609): eabm4247. doi:10.1126/science.abm4247
  10. 10.Liberman, Mark (2025). New Indo-European genetic evidence. Language Log. Accessed 2026-11-15.
  11. 11.Mair, Victor (2023). Where did the PIEs come from; when was that?. Language Log. Accessed 2026-11-15.
  12. 12.Wikipedia contributors. “Anatolian hypothesis”. Wikipedia, The Free Encyclopedia. Accessed 2026-11-14. Orientation source only.
  13. 13.Wikipedia contributors. “Indigenous Aryanism”. Wikipedia, The Free Encyclopedia. Accessed 2026-11-15. Orientation source only.

Written for Inkglot from the sources above, with no generated text. Last reviewed 2026-11-15. Owner sign-off pending: labels follow the cited sources and await a final check against each source's own words. How we label knowledge