Hurrian Crete: The Short Version

Somewhere between 9 and 17 per cent of the ancestry of the Minoans — the people who built Europe's first palaces — came from the east. Nobody has been able to say who carried it. This is the short version of a longer paper that names them, and of the evidence that was already in the ground when the genetics arrived.

Two peoples, one question

The Minoans built the first palace civilisation of Europe on Crete: Knossos, Phaistos, Malia, Zakros, with frescoed halls, bureaucracies that kept records on clay, and a script called Linear A that has never been read. Arthur Evans began digging at Knossos in 1900, and the Minoans have been treated ever since as the beginning of something European.

The Hurrians are much less famous, though they were there first. They are attested from the late fourth millennium BC at Urkesh (Tell Mozan) in the Khabur headwaters of northern Syria, and by 2300 BC their kings were sealing documents in their own language. They spoke a language with no known relatives except Urartian, wrote in cuneiform borrowed from Akkadian, and their religion and religious vocabulary were taken up by the Hittite state. They are, in other words, exactly the kind of people who shape a region without leaving a famous monument behind.

The surprise in the bones

In 2017, Iosif Lazaridis and colleagues published genomes from Bronze Age Crete. Minoans carried a substantial eastern component — 9 to 17 per cent ancestry related to the Caucasus and Iran — and, remarkably, no measurable contribution from the Levant. The old picture of a self-made European civilisation collapsed. The old picture of "influence from the East" also turned out to be pointed the wrong way.

In 2023, Eirini Skourtanioti and colleagues localised the proximate source to Anatolia. That sharpened the question without answering it. Neither study named a population. A genetic signal was moving west through Anatolia towards the Aegean, and the people carrying it had no name attached.

That is the puzzle this paper addresses — not whether eastern ancestry reached Crete, which is now established, but who was carrying it.

A corridor with a name on it

Archaeologists have been excavating that route for a century, and the name is on it.

At Urkesh in the Khabur headwaters, the Buccellati excavations established a Hurrian foundation going back to at least 4000 BC. South-east along the Khabur and Trans-Tigris zone, the Tupkish sealing (c. 2300 BC) and the Tish-atal inscription (21st century BC) are the earliest direct evidence of a Hurrian-speaking state. North-west, Arslantepe near Malatya sits on the Anatolian side of the corridor. Further west, Hurrian religious texts and Hurrian prestige were taken up in Hittite central Anatolia. And south at Ugarit on the Syrian coast, excavations produced around fifty Hurrian tablets in four genres — including the hymn to Nikkal, tablet H6, the oldest known notated melody — and a room identified as the House of the Hurrian Priest.

Put those points on a map and they do not form a scatter. They form a line, running from the Hurrian heartland through northern Syria and eastern Anatolia towards the Aegean — the same line the eastern ancestry must have travelled. Hurrian is attested on that line, in that direction, for two and a half thousand years. If a population moved west along it, the chance that it spoke something other than Hurrian is small.

Things that travelled the wrong way

Contact between Crete and the east was not a rumour. The objects are in the museums.

In Crete: a Syrian silver cylinder seal at Mochlos (c. 2500 BC) and a second northern Syrian haematite seal from the same site; a hippopotamus tusk of about 45 cm in Early Minoan levels at Knossos (c. 2650–2450 BC); and metallurgical habits of North Syrian tradition in early Minoan weaponry. Going the other way: Minoan Kamares ware at Ugarit, Qatna, Byblos and Hazor; the Mari archives referring to Crete as Kaptara; and Minoan-style frescoes in the palaces of Alalakh, Qatna, Tel Kabri and Tell el-Dab'a — Cretan painters working in the Levant.

The dates matter as much as the objects. The earliest Syrian goods in Crete predate the earliest Minoan genomes that carry the eastern component. The corridor was already open and in use before the ancestry shows up. People were moving along it — and where people move, languages move with them.

It is worth being clear about what that movement looks like in practice. It is rarely a single fleet arriving with banners. It looks like craftsmen hired across a sea, marriages between the families of traders, specialists in metals and seals and frescoes moving to wherever the work is, and the slow accumulation of foreign habits in ordinary households. That is precisely the kind of movement that leaves no heroic narrative and no obvious layer in a trench — but does leave foreign objects in the ground, and sometimes a foreign signature in the bones.

The script problem

Linear A has never been deciphered, so no one can say what language it records. But two things about it can be said with confidence.

The first is structural: Linear A is open-syllabic — syllables of the form consonant-vowel, vowel-consonant-vowel, consonant-vowel-consonant-vowel, with few consonant clusters. That is a shape like Japanese, and unlike the heavy clusters that Indo-European languages permit. The second is morphological: the language behind the script builds words by adding affixes productively, a hallmark of what linguists call agglutination.

Very few languages of the ancient Near East combine those two features. Five candidates exhaust the field: Sumerian, Elamite, Hurrian, Urartian and Hattic. Geography then narrows it. Hattic is known almost entirely through Hittite transmission and has no attested presence in the eastern Mediterranean. Urartian appears only in the ninth century BC, is confined to eastern Anatolia, and is a close relative of Hurrian anyway. Elamite belongs to Iran. Sumerian has a deep scribal tradition but no sustained Mediterranean presence.

What remains is Hurrian — the only one of the five attested along the corridor through which the ancestry actually moved.

What Linear A seems to record is not literature. The surviving tablets are largely administrative: quantities, commodities, rations, offerings, lists. That limits what can ever be proved from them — but it also explains why the structural argument matters. Even a shopping list carries the fingerprint of the language that wrote it, in the way its syllables are built and its words are assembled.

And the fallback position? A completely unknown language isolate remains possible in principle. The surviving corpus is small, and a language attested nowhere in the record could satisfy the same structural constraints. But such a hypothesis would require an undocumented population, an undocumented archaeological record and an undocumented linguistic identity in order to explain exactly the evidence that Hurrian already explains with documented ones. It would be akin to a deus ex machina.

What this does not prove

Linear A is not deciphered, and this paper does not decipher it. The linguistic conclusion is an inference, not a demonstration: language shift is possible, and the western Anatolian linguistic record of the relevant period is essentially absent. The genetic side has its own limits — the published proxies are broad, and no ancient DNA yet exists from the specific populations that carried the ancestry through Anatolia.

None of that changes the shape of the result. Three independent records — the archaeology of the corridor, the genomes of 2017 and 2023, and the structure of the Aegean script — point to the same people. Each of them alone is suggestive. Together they converge.

Why it matters

If the Hurrians carried that ancestry into the Aegean, then a people who left no famous monument and appear in few school textbooks were present at the making of Europe's first palace culture. The century-long argument about "eastern influence" on Crete was not wrong about the East. It was looking at the wrong East — the Levant and the coast, rather than the highlands beyond them, and the corridor that ran through them for two and a half thousand years.

There is a wider point too. The story of the Bronze Age is usually told as a story of centres: Egypt, Mesopotamia, the Hittites, the Mycenaeans. The people who actually moved the genes, the metals and the words between those centres rarely get a chapter of their own. The Hurrians are one of them — and on this reading, they were not a footnote to the Aegean. They were among the people who made it.

This is the short version. The full paper, with the archaeological, genetic and linguistic detail and its sources, is here.