Jewish DNA & the Levant

⬅️ Key Arguments (MOC) · Related: Indigenous · Jews Connection to Israel · Colonizers · Zionism · 3000BCE-1800 · History of the Jewish People · Jewish Conspiracy Theories

Earlier summary note: Jewish Genetic and Historical Connections to The Levant — this note supersedes and expands it.

One-line version

Four independent lines of evidence — genome-wide autosomal DNA, Y-chromosome, mitochondrial DNA, and ancient DNA — converge on the same conclusion: the world’s dispersed Jewish communities descend from a shared founder population in the ancient Near East, and their closest living relatives outside Jewry are other Levantines — Palestinians, Druze, Bedouin, Lebanese, Syrians. The Khazar hypothesis has no genetic support. But the same data show substantial European/local admixture in most diaspora groups, and they show that Palestinians are also descended from the ancient population of the land. Genetics settles a descent question. It does not, by itself, settle a rights question.

Read this before you use any number in this note

Admixture percentages are model outputs, not measurements. Every “X% Levantine / Y% European” figure depends on which reference populations the authors chose as proxies, which statistical method (ADMIXTURE, qpAdm, GLOBETROTTER, f4-ratio) they ran, and which ancient genomes existed when they ran it. The same Erfurt dataset yields Middle Eastern ancestry anywhere from 19% to 43% depending on whether you model the European source as South Italian, North Italian, or Greek — and the authors say so explicitly. Quote ranges, quote the method, and never treat a single percentage as a fact.

---

0. What the genetics can and cannot establish

Setting this out first, because most bad arguments in this space are category errors in both directions.

What the evidence supports well:

1. Most Jewish diaspora communities share a common ancestral core with each other that is not explained by their host populations.
2. That shared core is Near Eastern / Levantine in origin.
3. There is no detectable Khazar or Turkic signal at a level that could account for Ashkenazi origins.
4. Major diaspora communities passed through severe founder/bottleneck events, datable and independently corroborated by disease genetics.

What the evidence does not establish:

1. That any given Jew descends from the Israelites specifically — Israelites were themselves one branch of a broadly Canaanite-descended population. Genetics resolves to “ancient southern Levantine,” not to a tribe.
2. That Jews are more Levantine than Palestinians. On most measures they are less so (see §8).
3. Anything about who is entitled to sovereignty. Descent is not title. → §10.

The honest frame: genetics is a corroborating line of evidence for a claim that already rests on text, liturgy, language, law, archaeology, and continuous physical presence (→ Jews Connection to Israel, Indigenous). It is the weakest of those pillars to lead with and the easiest to have turned against you. Lead with continuity; use DNA as backup.

---

1. The ancient baseline — what the Levantine gene pool actually is

You cannot say Jews “come from the Levant” without knowing what the Levantine population was.

Canaanites = local Neolithic + eastern (Zagros/Caucasus) migrants

Agranat-Tamir et al. 2020, Cell — genome-wide data from 73 newly sequenced individuals (93 total with previously published genomes) from five sites across the Bronze and Iron Age southern Levant, c. 2400–900 BCE.

• Individuals sharing “Canaanite” material culture model as descending from two sources: earlier local Neolithic populations, plus populations related to the Chalcolithic Zagros / Bronze Age Caucasus.
• The non-local contribution increased over time — three outliers model as descendants of recent migrants.
• Modern Levant-related populations carry substantial ancestry from both of these sources, plus ancestry that cannot be fully modelled with available data — which the authors flag as evidence of “the critical role of post-Bronze-Age migrations into the region over the past 3,000 years.”
• Press-reported headline finding: modern Jewish and Arabic-speaking groups typically derive at least 50% of their genomes from this Bronze Age Canaanite–Zagros blend.

The load-bearing implication

“Canaanite” and “Israelite” are not genetically separable with current data. The Israelites emerge from within the Canaanite-descended population of the southern Levant — which is exactly what the archaeology (Finkelstein, Dever) already indicated. So “Jews descend from Canaanites” is not a gotcha against Jewish indigeneity; it is the same claim stated one layer down. → 3000BCE-1800

Continuity across five millennia

Haber et al. 2017, AJHG — five whole genomes from Sidon, a Canaanite city-state, c. 3,700 BP, plus 99 present-day Lebanese.

• Present-day Lebanese model as 93% ± 1.6% Sidon Bronze Age + 7% ± 1.6% Steppe Bronze Age.
• Canaanite-related ancestry was regionally widespread — coastal (Sidon) and inland (Jordan) alike — and itself derived from local Neolithic populations mixing with eastern migrants related to Chalcolithic Iranians.
• The ~7–10% non-Canaanite component entered roughly 3,000–2,000 years ago.

Conclusion: the Levantine gene pool has been remarkably stable since the Bronze Age. Conquests changed rulers, languages and religions far more than they changed genomes.

Ancient DNA from Israelites themselves — the real gap

This is the weakest link, and you should concede it rather than be caught on it. The southern Levant’s climate is hostile to DNA preservation; Iron Age Judahite/Israelite genomes are almost nonexistent.

• 2023–24: first recovery of ancient DNA attributed to First Temple period Israelites, from a rare Iron Age family tomb near Jerusalem (associated with Kiryat Yearim; research led with Israel Finkelstein), c. 750–650 BCE. Only partial data from two individuals. Treat as preliminary and not yet fully published — do not build an argument on it.
• Everything else about Iron Age Israel is inferred from Bronze Age Canaanite genomes plus modern populations, not from Israelite remains directly.

---

2. Genome-wide (autosomal) evidence — the strongest pillar

Y-DNA and mtDNA each track a single lineage. Autosomal data samples the whole genome and is what you should cite first.

Behar et al. 2010, Nature — “The genome-wide structure of the Jewish people”

~600,000 markers; 14 Jewish diaspora communities (114 individuals) vs 69 Old World non-Jewish populations (1,161 individuals).

“Most Jewish samples form a remarkably tight subcluster that overlies Druze and Cypriot samples but not samples from other Levantine populations or paired Diaspora host populations… These results cast light on the variegated genetic architecture of the Middle East, and trace the origins of most Jewish Diaspora communities to the Levant.”

• Principal component and structure analyses revealed previously unrecognised substructure within the Middle East.
• Exceptions: Ethiopian Jews (Beta Israel) and Indian Jews (Bene Israel, Cochini) cluster with their local autochthonous populations — despite a clear paternal Levantine link in the Bene Israel. Judaism spread by ideas as well as by descent, and the data show it honestly.
• Frequently quoted press framing of this result: in communities representing ~90% of world Jewry, Jews were more genetically similar to non-Jewish Levantines than their non-Jewish host populations were. (This is a press restatement, not a sentence from the paper — attribute it as such.)

Atzmon et al. 2010, AJHG — “Abraham’s Children in the Genome Era”

Seven Jewish groups (Iranian, Iraqi, Syrian, Italian, Turkish, Greek, Ashkenazi). Each forms its own distinct cluster, but all share a common Middle Eastern ancestry component. Two major branches: a Middle Eastern/Mizrahi branch and a European/Syrian branch, with an inferred split roughly 2,500 years ago.

Ostrer & Skorecki 2013, Human Genetics — the synthesis review

📄 In your library: 60.1.7 Books & Documents/Karl - The population genetics of the Jewish people (2012).pdf

Key findings worth quoting directly:

• Jewish populations of Europe, North Africa and the Middle East form a tight central cluster distinguished from their non-Jewish neighbours, each with its own subcluster — supported by Fst and allelic-sharing distance metrics.
• Identity-by-descent (IBD) sharing within Jewish populations is comparable to fourth-to-fifth cousins. Sharing between Jewish populations exceeds sharing between Jewish and geographically adjacent non-Jewish populations.
• European admixture of 30–60% among Ashkenazi, Sephardic, Italian and Syrian Jews.
• Money quote on neighbours: “The closest genetic neighbors to most Jewish groups were the Palestinians, Israeli Bedouins, and Druze in addition to the Southern Europeans, including Cypriots.” Their proximity “suggested a shared genetic history of related Semitic and non-Semitic Mediterranean ancestors who followed different religious and tribal affiliations.”
• Beautiful summary line: “Jews have a tapestry of shared DNA threads with other Jews and that no one thread is sufficient to define Jewish ancestry.”

Xue et al. 2017, PLoS Genetics — dating the European admixture

• Ashkenazi ancestry ≈ 47% Middle Eastern / 53% European.
• The European component is predominantly Southern European (~60–80% of it), remainder likely Eastern European.
• Inferred admixture ~35 generations ago, but this is an average over at least two events: one pre-dating the late-medieval founder event (Southern European source, bounded 25–55 generations ago), one post-dating it (Eastern Europe).

Carmi et al. 2014, Nature Communications

Ashkenazi Jews descend from roughly 350 founders, ~600–800 years ago, with approximately equal Middle Eastern and European components.

---

3. Paternal lineages (Y-DNA) — the strongest Levantine signal

The Cohen Modal Haplotype (CMH)

The most rhetorically famous result and the most commonly overstated.

• Skorecki et al. 1997, Nature / Thomas et al. 1998: self-identified Cohanim across Ashkenazi and Sephardi communities share a characteristic Y haplotype at far higher frequency than other Jews — consistent with patrilineal priestly descent from a common ancestor.
• Hammer et al. 2009, Human Genetics — the necessary correction. With higher marker resolution, the Cohanim resolve into at least two distinct founding lineages, not one:
  • J-P58 (a subset of J1) — coalescing ~3,200 years ago
  • J2a-M410* — coalescing ~4,200 years ago
  • The extended CMH (eCMH) accounts for ~30% of Cohanim Y chromosomes in both Ashkenazi and non-Ashkenazi communities, is virtually absent in non-Jews, and traces to a common male ancestor ~3,200 years ago in the Near East.
• ⚠️ Do not say “the Cohen gene proves descent from Aaron.” It doesn’t and can’t. What it shows: a Near Eastern paternal lineage of Bronze/Iron Age depth, shared across communities separated for ~2,000 years, tracking a religious status category. That is a genuinely striking result on its own terms.

Ashkenazi Levites — the case that reversed

A textbook example of the field self-correcting, and a useful one for arguing with people who claim the science is ideologically fixed.

• The Levite R1a cluster was long read as European/Khazar in origin — Behar et al. 2003 even floated the Khazar Kingdom as “one attractive source.” That reading appears in Sand and in Falk’s book.
• Rootsi, Behar et al. 2013, Nature Communications overturned it with whole-Y sequencing. The marker R1a-M582:
  • was absent in 922 Eastern Europeans tested;
  • was present in all sampled R1a Ashkenazi Levites;
  • present in 33.8% of other R1a Ashkenazi Jewish males;
  • present in 5.9% of 303 R1a Near Eastern males — where it shows considerably higher diversity (i.e. greater time depth → the source).
  • Also occurs at low frequency in non-Ashkenazi Jewish populations.
• Conclusion: a Near Eastern geographic source for the Levite founder lineage, likely present among pre-Diaspora Hebrews. Further refined in Behar et al. 2017, Scientific Reports (R1a-Y2619).

---

4. Maternal lineages (mtDNA) — the genuinely contested front

Be candid here. This is where the evidence is weakest and the debate is live. If you claim mtDNA is settled in favour of Levantine origin, you will be corrected — accurately.

The pattern everyone agrees on: Jewish diaspora communities show strong female founder effects — small numbers of founding mothers. Behar et al. 2006/2008: four mtDNA haplogroups account for >40% of Ashkenazi mtDNA, forming “star phylogenies” around a coalescent ancestral haplotype. In some communities (Azerbaijan, Georgia, Libya, Mumbai, Belmonte) a single founding mother accounts for ≥40% of modern lineages.

Where they disagree — the origin of those founders:

Ancient DNA arbitrates partially: at 14th-century Erfurt, 35% of individuals carried K1a1b1a — a higher frequency than the ~20% in modern Ashkenazim, and concentrated in the more Middle Eastern-ancestry subgroup (54% vs 20%). That establishes the lineage’s Ashkenazi centrality by the 1300s but does not by itself resolve where it originated.

How to state this honestly

“Paternal lineages are strongly and uncontroversially Near Eastern. Maternal lineages show a heavy European contribution in Ashkenazim, and the origin of the major founder mothers is actively disputed in the literature right now. The overall genome-wide picture — which weights both — still comes out with a substantial Levantine core.” That sentence is defensible against anyone.

Why the asymmetry, if real, is unsurprising: small numbers of Jewish men moving along Roman-era and medieval trade routes, taking local wives, whose descendants were raised Jewish. This is the ordinary demography of a merchant diaspora, and it is exactly what a sex-biased admixture signal looks like. It is not evidence of fabricated peoplehood.

---

5. Ancient DNA from the diaspora itself

Erfurt, Germany — Waldman et al. 2022, Cell

The single most important recent paper. Genome-wide data from 33 Ashkenazi Jews of the 14th century, from a salvage excavation of the medieval Jewish cemetery.

• Erfurt individuals are genetically similar to modern Ashkenazim — direct demonstration of continuity across ~700 years.
• Admixture ranges across plausible models — note the spread:
  • 65% South Italy / 19% Middle East / 16% East European
  • 37% North Italy / 43% Middle East / 20% East European
  • 51% Greek / 32% Middle East / 17% East European
  • Authors’ own caveat: “the specific identity of the source populations… should not be considered precise.” Mediterranean European 37–65%, Middle Eastern 19–43%.
• The founder event pre-dates the 14th century (this is the paper’s title claim).
• Demographic modelling from modern IBD data: bottleneck onset ~41 generations ago (95% CI 39–43) — about 1,000 years ago — with effective population size ~1,563 (95% CI 1,364–1,751), lasting 20 generations. The Erfurt-specific subgroup bottlenecked harder (Ne ≈ 627).
• Medieval Ashkenazim were more genetically diverse than modern ones, splitting into two subgroups: Erfurt-ME (closer to modern Western-European-origin Ashkenazim and modern Sephardim) and Erfurt-EU (extra Eastern European component). Stable isotope δ¹⁸O differences (p=0.0005) suggest different childhood water sources → different geographic origins. Modern Ashkenazim are the homogenised descendants of these once-distinct groups.
• 11 high-confidence pathogenic variants across 8 individuals — the Ashkenazi disease alleles were already circulating.

Norwich, England — Olalde et al. 2022, Current Biology

Six individuals from the base of a medieval well, radiocarbon-consistent with the documented antisemitic massacre of 6 February 1190 CE.

• Four of the six were closely related; all six show strong genetic affinity with modern Ashkenazi Jews.
• Four Ashkenazi-associated disease alleles identified — already at appreciable frequencies.
• Conclusion: the Ashkenazi bottleneck is centuries older than had been assumed. Together with Erfurt, this pushes the founder event back well before the 14th century.

---

6. Disease genetics as an independent clock

An underused argument. Founder-population disease alleles are a second, methodologically independent line of evidence for the same demographic history.

• Ashkenazi founder mutations cluster into categories: lysosomal storage diseases (Tay-Sachs, Niemann-Pick, Gaucher, mucolipidosis IV), glycogen storage diseases, clotting factor deficiencies (factor XI, factor VII, combined V & VIII), steroid biosynthetic defects, and DNA-repair/cancer-susceptibility defects (BRCA1/2, Bloom syndrome, Fanconi anemia).
• Coalescence analysis dates these mutations across Jewish history — some to the pre-Jewish era — and disease mutations in a given diaspora group typically coalesce to the founding of that group.
• Drift, not selection, is favoured as the mechanism (Risch et al. 2003; Slatkin 2004). The one proven selection case is G6PD-Med (malaria resistance).
• Erfurt and Norwich confirm these alleles were present by the 12th–14th centuries — matching what modern-DNA coalescence predicted. Two independent methods, same answer. That is what corroboration looks like.
• Practical footnote: Tay-Sachs screening since the 1970s reduced births from ~40–50/year to ~4–5/year.

---

7. The Khazar hypothesis — origin, collapse, and weaponisation

→ Fuller treatment in History of the Jewish People and Jewish Conspiracy Theories; see also 3000BCE-1800 > Khazar theory debunked.

The claim: Ashkenazi Jews descend not from Israelites but from the Khazars, a Turkic people whose elite converted to Judaism c. 8th–9th century, and who allegedly migrated west en masse after the khaganate’s fall.

Provenance: Isaac Baer Levinsohn (1828) → Poliak (1951) → Dunlop (1954) → Arthur Koestler, The Thirteenth Tribe (1976) → Shlomo Sand, The Invention of the Jewish People (2008). Koestler’s aim was explicitly anti-racist — to defuse racial antisemitism by denying a Semitic Jewish race. It was subsequently inverted into precisely the opposite use.

The genetic case for it:

• Elhaik 2013 — reported Caucasus links for Eastern European Jews.
• Das et al. 2016 — localised Ashkenazi origins to villages in northeastern Turkey along Silk Road trade routes, using the GPS (Geographic Population Structure) method.

Why it failed:

Fairness note: Elhaik has continued to defend his methods and disputes that Behar and Flegontov refute him. Raphael Falk (Zionism and the Biology of Jews, 2017 — 📄 in your library) makes the sharper meta-point worth engaging with honestly: “rather than the historians and anthropologists probing genetic-molecular data to provide clinching evidence, it seems that the geneticists seek historical and anthropological evidence to support their molecular data.” That is a real risk in this field. It is not, however, an argument that the Khazar hypothesis is true.

Why it matters politically: the theory’s entire modern function is to argue that Ashkenazim are “fake Jews” with no claim to the land — and from there into “Khazarian Mafia” material that is straight Protocols-descended antisemitism. Note the self-refuting structure: the argument concedes that real descent from ancient Israelites would ground a claim, then disputes only the descent. Anyone making it has accepted the indigeneity premise. → Jewish Conspiracy Theories · Antisemitism

---

8. The symmetry nobody on either side likes

This is the most intellectually important section in the note, and the one most likely to be weaponised against you if you haven’t already internalised it.

The same body of evidence establishes both of the following:

1. Jews descend substantially from the ancient population of the southern Levant.
2. Palestinians and other Levantine Arabs descend from that same ancient population — and by most measures, with higher continuity.

Palestinians are consistently identified as among the closest genetic neighbours of most Jewish groups (Ostrer & Skorecki 2013; Rosenberg et al. 2001; Kopelman et al. 2009). Haber 2017 puts modern Lebanese at ~93% Sidon Bronze Age ancestry. Agranat-Tamir 2020 has modern Levantine Arabic-speaking groups deriving well over half their genomes from the Canaanite-related population — with less subsequent dilution than diaspora Jewish groups experienced.

What follows:

• The Arab conquest of the 7th century was largely an elite and cultural replacement — language, religion, identity — not a demographic one. Most Palestinians are, in ancestry terms, Arabised Levantines, not Arabian settlers. (Nebel 2001’s high-frequency Eu 10 haplotypes among Palestinian Arabs and Bedouin do indicate some Arabian Peninsula input in the last two millennia — a real component, not the whole story.)
• Therefore: “Palestinians are recent arrivals with no roots” is not supported by the genetics. Don’t make that argument. It is false, it’s easily refuted, and it destroys your credibility on everything else you’ve said.
• Equally: “Jews are European colonisers with no Levantine ancestry” is not supported either. Both mirror-image claims lose to the same data.

The real conclusion

The genetics describes two peoples with overlapping ancestry in the same land, who diverged religiously and culturally rather than biologically. That’s the finding. It is a poor weapon for either side and a decent foundation for anyone arguing that neither population is going anywhere. → Solutions (Two-State, One-State & Beyond)

---

9. Group-by-group summary

---

10. Limits, critiques, and how to hold them

Methodological limits — concede these before they’re used on you:

1. Model-dependence of percentages. See the warning box at the top. Erfurt: 19–43% ME from one dataset.
2. Proxy population choice determines the answer. The Elhaik–Behar dispute is fundamentally a fight about proxies, not about data.
3. Uniparental markers are single loci. One Y lineage and one mtDNA lineage per person. They can tell radically different stories from the autosomes and from each other. They are not “your ancestry.”
4. Reference-panel dependence. Populations are proxied by modern groups that have themselves been changing for 3,000 years.
5. Ancient DNA scarcity in the Levant. The single biggest evidentiary gap. Iron Age Israelite genomes are effectively absent.
6. Category mismatch. Genetics resolves to “ancient southern Levantine.” It does not resolve to “Israelite,” “Judahite,” or “descendant of Abraham.”

Substantive critiques worth reading rather than dismissing:

• Nadia Abu El-Haj, The Genealogical Science (2012) — argues Israeli population genetics functioned as a “biopolitical project” tied to nation-building; that “DNA evidence does not generate historical stories”; that phylogenetic trees flatten “the messiness and density of the social world.” Her strongest point: genetic facts do not come with their historical narrative attached — the narrative is supplied by the reader. Her weakest: motive-questioning does not falsify a result, and the findings have been independently replicated by teams with no Zionist stake (Estonian, British, German, American).
• Raphael Falk, Zionism and the Biology of Jews (2017) — 📄 60.1.7 Books & Documents/Falk - 2017 - Zionism and the Biology of Jews.pdf. Traces how Zionist thinkers (Ruppin, Salaman) engaged early-20th-century race science and eugenics, including Ruppin’s adoption of Sombart’s claims about a Jewish “commercial instinct.” Uncomfortable and worth knowing: the history of Jewish biological self-definition is genuinely entangled with race science. That is a fact about the field’s past, not about the validity of a 2022 qpAdm model.
• Shlomo Sand, The Invention of the Jewish People (2008) — the maximal version; the genetic core of his argument has not survived (see §7), but his historiographical points about the modern construction of national narratives are made by mainstream historians too.

The philosophical limit — the one that actually matters:

Descent is not title

No amount of DNA evidence establishes a right to sovereignty. If it did, it would establish competing rights for Palestinians on the same data (§8), and would imply that peoples without ancient continuity have weaker claims to their homes — a principle no one making the argument would accept generally. Rights arguments rest on law (San Remo, the Mandate, UN 181 → Mandate & Legal Basis (Balfour, San Remo, League of Nations)), self-determination, and the facts of who lives there now.

The genetics does real but limited work: it rebuts a specific accusation — that Jews are foreign European colonists with a fabricated connection to the land. That accusation is empirically false. Refuting it is the whole job. Don’t ask the DNA to do more.

---

11. Using this in argument

Claims this evidence rebuts:

Claims this evidence does not support — do not make them:

• ❌ “Palestinians are foreign Arabs with no roots in the land.” False. → §8
• ❌ “The Cohen gene proves descent from Aaron.” Overclaim. → §3
• ❌ “Jews are X% Levantine, therefore they own the land.” Category error. → §10
• ❌ “The science is settled on maternal lineages.” It isn’t. → §4

If someone leads with genetics against you, the strongest reply is structural: “You’ve just conceded that ancient descent grounds a claim to the land. Fine — let’s test it. The genetics says both populations descend from the ancient inhabitants. So we’re arguing about a shared inheritance, not a foreign invasion.”

---

📚 Key sources

In your library (📄 60.1.7 Books & Documents/)

• Karl - The population genetics of the Jewish people (2012).pdf — Ostrer & Skorecki 2013, Human Genetics 132:119–127. The best single-document overview. Start here.
• The Y Chromosome Pool of Jews.pdf — Nebel et al. 2001, AJHG 69:1095–1112.
• Falk - 2017 - Zionism and the Biology of Jews.pdf — the critical/historiographical counterweight.

Primary literature
| Paper | Journal / Year | What it established |
|---|---|---|
| Behar et al., “The genome-wide structure of the Jewish people” | Nature 466:238–242, 2010 | Jewish communities form a tight cluster overlying Druze/Cypriots; origins traced to the Levant |
| Atzmon et al., “Abraham’s Children in the Genome Era” | AJHG 86:850–859, 2010 | Distinct clusters, shared Middle Eastern ancestry, ~2,500-yr split |
| Ostrer & Skorecki, “The population genetics of the Jewish people” | Hum Genet 132:119–127, 2013 | Synthesis review; IBD sharing; closest neighbours = Palestinians, Bedouin, Druze |
| Behar et al., “No Evidence… of a Khazar Origin” | Human Biology 85(6), 2013 | Khazar hypothesis fails on genome-wide data |
| Rootsi, Behar et al., ”…Near Eastern origin of Ashkenazi Levites” | Nat Commun 4:2928, 2013 | R1a-M582 is Near Eastern, not European/Khazar |
| Costa et al., “A substantial prehistoric European ancestry amongst Ashkenazi maternal lineages” | Nat Commun 4:2543, 2013 | The European-mtDNA challenge |
| Carmi et al. | Nat Commun, 2014 | ~350 Ashkenazi founders, 600–800 yrs ago |
| Flegontov et al., “Pitfalls of the GPS Approach…” | Genome Biol Evol 8:2259, 2016 | Methodological demolition of Das/Elhaik |
| Haber et al., “Continuity and Admixture in the Last Five Millennia of Levantine History” | AJHG, 2017 | Lebanese ≈ 93% Canaanite (Sidon BA) |
| Xue et al., “The time and place of European admixture in Ashkenazi Jewish history” | PLoS Genetics 13:e1006644, 2017 | ~47% ME / 53% European; ≥2 admixture events |
| Agranat-Tamir et al., “The Genomic History of the Bronze Age Southern Levant” | Cell 181:1146–1157, 2020 | Canaanite ancestry = local Neolithic + Zagros/Caucasus; ≥50% in modern Jews and Arabs |
| Olalde et al., “Genomes from a medieval mass burial…” | Current Biology, 2022 | Norwich 1190; Ashkenazi disease alleles pre-date the 12th century |
| Waldman et al., “Genome-wide data from medieval German Jews…” | Cell 185:4703–4716, 2022 | Erfurt; founder event pre-dates the 14th century; medieval Ashkenazim more diverse than modern |
| Livni & Skorecki, “Distinguishing between founder and host population mtDNA lineages…” | 2025 (+ commentary & response) | Reopens the maternal-origin question in favour of Levantine founders |

Critical / contrarian

• Nadia Abu El-Haj, The Genealogical Science: The Search for Jewish Origins and the Politics of Epistemology (Chicago, 2012)
• Raphael Falk, Zionism and the Biology of Jews (Springer, 2017)
• Shlomo Sand, The Invention of the Jewish People (Verso, 2008)
• Eran Elhaik (2013) and Das et al. (2016) — read alongside the Behar 2013 and Flegontov 2016 rebuttals, not alone

Circulating but unverified

A paper titled “Genetic Proximity of Modern Palestinians and Ashkenazi Jews to Iron Age Levantines: A Quantitative Paleogenomic Analysis” (2024–25) circulates online with headline figures of 81–87% Bronze Age Levantine ancestry for Palestinians and 37–65% European introgression for Ashkenazim. I could not verify peer-reviewed journal publication — it appears to be self-posted on ResearchGate. Its numbers are broadly consistent with the published literature, but do not cite it as peer-reviewed. Use Agranat-Tamir 2020, Haber 2017 and Waldman 2022 instead; they say substantially the same thing and are refereed.

---

🔍 Open questions / further research

→ Further Research

1. Iron Age Israelite aDNA. The Kiryat Yearim tomb material (§1) — has it been formally published? This is the highest-value gap in the whole field.
2. The mtDNA dispute. Track the Livni & Skorecki (2025) exchange to resolution. Currently the softest spot in the case.
3. Mizrahi genetics. Underrepresented in both the literature and in this note. Iraqi, Iranian, Yemenite and Kurdish Jewish communities retain the highest Middle Eastern component and are the strongest cases — yet almost all popular argument is about Ashkenazim. Worth building out, especially given that Mizrahim + Sephardim are a majority of Israeli Jews. → Israel
4. Ancient DNA from Roman-era Judea and from the early diaspora (Alexandria, Antioch, Dura-Europos) — would directly test the Roman-period migration model.
5. Palestinian population genetics specifically — mostly discussed here as a comparison group. Deserves its own treatment. → Palestine