Every people that sailed the Mediterranean stopped in Sicily, and most of them stayed. That is the island's reputation, and for once the genetics largely agrees — with the crucial caveat that the layers are thinner and better-ordered than the folklore suggests. Sicily also holds real ancient-DNA anchors for most of its famous chapters, which makes it one of the few places where "crossroads ancestry" can be read as a sequence instead of a shrug.
The short answer: Sicilian ancestry stands on a Neolithic farmer base, took steppe-related ancestry late (from ~2200 BCE), was substantially reshaped by Greek and eastern-Mediterranean settlement in antiquity, and carries a real North African layer deposited mostly in the Islamic centuries and diluted, not erased, afterwards. Modern Sicilians sit with southern mainland Italians and Greek islanders at the Mediterranean's genetic centre of gravity.
Foragers, farmers, and a late steppe arrival#
Sicily's Mesolithic foragers (Grotta d'Oriente) were western hunter-gatherers of the standard type. Neolithic farmers arriving by sea (~6000 BCE) largely replaced them, and for two millennia Sicily was a farmer island with Sardinian-like genomes — the profile Sardinia then kept while Sicily moved on.
Fernandes et al. 2020's island transect dates the change: steppe-related ancestry reaches Sicily around 2200 BCE — several centuries after the mainland — arriving already blended (Beaker-derived, via Iberia or the mainland rather than from the steppe directly) and modestly: Bronze Age Sicilian genomes take on a minority steppe share and the first R1b-M269 lineages, while remaining farmer-dominated. The same study finds Iranian-related ancestry appearing in Middle Bronze Age samples — the eastern Mediterranean already leaking westward before any Greek ever sailed.
The colonial millennium#
Iron Age Sicily held Sicani, Sicels and Elymians — genetically, local Bronze Age profiles — and then the ships arrived. Greek colonisation (from 734 BCE) was demographically real: sampled classical-era Sicilians from Greek cities carry strong Aegean-related ancestry, and the Himera soldiers study showed the colonial world's reach — a Greek army whose fallen included men of steppe, Baltic and Caucasus origins alongside locals. Phoenician Palermo and Motya added Levantine-related ancestry on the west coast — the Punic network's genetics show colonies that were themselves melting pots. The Roman centuries broadened the same eastern-Mediterranean drift visible across the imperial world. Net effect: by late antiquity, Sicilian genomes had moved decisively east-and-south of their Bronze Age position — toward where they still sit.
The Islamic and Norman centuries#
The Aghlabid and Fatimid centuries (827–1091) settled Muslim communities across the island; the Norman conquest and its Hohenstaufen aftermath re-Latinised it, ending with the deportation of Sicily's Muslims. The genetic ledger of those events, read from modern samples (Sarno et al. 2017 and kindred work) and the first medieval genomes: a real North African- related layer — typically estimated at a few percent, strongest in the island's west — plus continued Greek-and-Levantine affinity, over a base the conquests never displaced. Norman, Lombard and later mainland settlement left its own northern texture (the Gallo-Italic dialect pockets of the interior are its cultural fossil), modest at genome scale.
Modern Sicilian genomes accordingly cluster with Calabrian and southern Italian samples, close to Greek islanders — Mediterranean central, farmer-heavy, steppe-light by European standards, with small but real Levantine and North African edges. Y-DNA is a harbour registry: R1b and J2 leading, with E-V13, G2a, E-M81, J1 and T all present in colonial-era proportions.
Reading your own results#
- G25 distances: modern era led by Sicilian and southern Italian averages with Greek islanders close; classical eras full of Aegean- and eastern-Mediterranean company; the Bronze Age era noticeably more local. That era-by-era eastward drift is the colonial millennium, in one tool.
- Admixture fits: panels need an eastern-Mediterranean source and benefit from a North African one — a Sicilian coordinate fitted against a purely farmer-plus-steppe panel pushes its real signals into the wrong components. Expect farmer-related weight to dominate and steppe-related weight to run below the European average.
- qpAdm: the productive formal questions are the layered ones — does my genome require a North African-related source? an Aegean-related one beyond the Bronze Age base? — precisely the contrasts a tested model resolves and a calculator merely decorates.
Frequently asked questions#
How much Arab ancestry do Sicilians have?#
The North African-related layer averages a few percent, west-tilted — real, measurable, minor. Note the label: the medieval settlers were largely Berber and Arab-led North Africans, so "North African-related" is the precise genetic description.
Are Sicilians more Greek or more Italian?#
Falsely posed — southern mainland Italians and Sicilians both carry heavy Aegean-related ancestry from the same colonial-to-Byzantine continuum. Sicilian genomes are what "Greek and Italian" look like after two millennia of sharing one sea.
Is there Norman (Viking) ancestry in Sicily?#
Traceable at most as a small northern-European texture — the Norman state mattered politically far more than demographically. A large "Scandinavian" percentage from a hobby tool is a panel artefact.
Terms used here are defined in the glossary.
Sources and further reading#
- Fernandes, D. M. et al. (2020). The spread of steppe and Iranian-related ancestry in the islands of the western Mediterranean. Nature Ecology & Evolution, 4, 334–345.
- Reitsema, L. J. et al. (2022). The diverse genetic origins of a Classical period Greek army. PNAS, 119, e2205272119.
- Sarno, S. et al. (2017). Ancient and recent admixture layers in Sicily and Southern Italy trace multiple migration routes along the Mediterranean. Scientific Reports, 7, 1984.
- Antonio, M. L. et al. (2019). Ancient Rome: a genetic crossroads of Europe and the Mediterranean. Science, 366, 708–714.
