Kevin Alan Brook review: 8 shocking flaws in his treatment of Jewish ancestry
This Kevin Alan Brook review examines a fundamental problem in The Maternal Genetic Lineages of Ashkenazic Jews: detailed maternal and paternal genealogies repeatedly carry historical conclusions that need much stronger autosomal support. Jewish ancestry claims cannot be settled by treating surviving Y-DNA and mitochondrial lineages as a census of an entire population’s ancestors.
Autosomal DNA represents the overwhelming majority of the nuclear genome and draws ancestry from many family lines. Y-DNA follows one paternal line; mitochondrial DNA follows one maternal line. Both are valuable. Neither can independently establish the proportions, geographical origins or demographic history of the rest of the genome.
Two ancient W3b1 carriers make that distinction concrete. PCA0376, from medieval Poland, has a supplied autosomal comparison dominated by eastern European references. KFJ034, from Avar-period Hungary, has a markedly different profile whose closest supplied references are Turkmen populations. The maternal label is shared. The broader genetic backgrounds are substantially different.
1. Autosomal evidence receives too little analytical weight
Brook’s principal autosomal overview occupies a few paragraphs on pages 3–4. The maternal-lineage encyclopedia occupies pages 13–126. Autosomal observations also appear elsewhere, so saying he never discusses autosomal DNA would be inaccurate. The problem is the unequal analytical weight given to different kinds of evidence when making broad Jewish ancestry claims.
A maternal-lineage reference book can reasonably concentrate on mtDNA. But once it estimates population ancestry or reconstructs the scale of historical intermarriage, its conclusions extend beyond a maternal catalogue. Those arguments require explicit demographic models, representative reference populations and serious treatment of drift.
| Evidence | What it follows | What it cannot establish alone |
|---|---|---|
| Autosomal DNA | Ancestry across many inherited chromosome segments | A unique historical explanation without appropriate models and references |
| Y-DNA | One direct paternal lineage | The total contribution of all male ancestors |
| mtDNA | One direct maternal lineage | The ancestry proportions of an individual or population |
2. Genetic drift breaks the shortcut from present frequencies to past intermarriage
On page 8, Brook moves from a small present-day proportion of paternal lineages attributed to northern and eastern European sources to a restrictive conclusion about historical reproductive relationships. This is one of the most consequential weaknesses examined in this Kevin Alan Brook review.
A man can leave autosomal descendants through daughters while his Y-chromosome lineage disappears. A woman can leave autosomal descendants through sons while her mitochondrial lineage disappears. Over generations, bottlenecks, chance and unequal reproductive success can radically alter the frequencies of surviving uniparental branches.
No uninterrupted father-to-son or mother-to-daughter transmission survives.
Autosomal segments can continue through descendants of either sex.
Founder effects and drift reshape which uniparental branches remain common.
Brook cites bottleneck research and recognizes that some lineages may have disappeared. The criticism is that these implications are not adequately integrated into particular historical conclusions. Present-day Y-DNA frequencies cannot simply be read backward as the percentage of past non-Jewish male contributors. Jewish ancestry claims built on that shortcut require substantial revision.
3. A “Middle Eastern percentage” depends on the European comparison
A model does not discover an ancestry percentage independently of its reference populations. If the European proxy inadequately represents Mediterranean ancestry, another source can absorb the mismatch. Southern Italian, central Italian and northern Italian references are therefore consequential choices.
Table 7 of my preprint reports the following results when the Italian reference changes in models also using Lebanese and Russian sources. These are the preprint’s reported outputs, not newly rerun estimates.
The reference changes. The Lebanese coefficient changes.
Bar scale: 0–50% of the fitted model. Source: preprint, Table 7.
p = 0.1445; reported feasible
p = 0.0514; reported feasible
p = 0.0256; reported not feasible
The Lebanese coefficient shifts by 20.51 percentage points between the displayed southern and northern models. That is a substantive modelling issue. It also matters that modern Mediterranean references already contain older shared ancestry: a Lebanese coefficient is not automatically the total amount of deep Levantine-related ancestry.
This Kevin Alan Brook review therefore asks a specific question: where is the systematic demonstration that the book’s broad ancestry allocation remains stable across plausible European sources?
4. Brook’s own sources complicate a fixed ancestry split
Brook cites the Crete study by Drineas and colleagues in note 16. Mediterranean affinities are consequently present in his source base. They deserve sustained analysis alongside the book’s geographical assignments, especially when Southern Italian and Aegean comparisons can influence an ancestry model.
His note 17 cites Xue and colleagues, whose paper does not provide only one approximately half-European estimate. Its RFMix analysis estimates about 53% European ancestry. Its CHROMOPAINTER-derived ancestry profile gives 55% Southern European, 10% Eastern European, 5% Western European and 30% Levantine: approximately 70% European combined, or approximately 67% after simulation calibration.
These are different methodological estimates, not interchangeable measurements or proof that the largest number must be correct. Their disagreement is precisely the issue. Reducing this source to a settled 50/50 formula strips away information central to evaluating Jewish ancestry claims.
The criticism is not that Brook lacks citations. It is that the qualifications and competing results within those citations need to shape the historical argument.
5. Precise-looking percentages need reproducible foundations
Several specific ancestry figures in the book trace to online modelling discussions. Forum research can generate useful hypotheses, but numerical precision does not replace a documented reference panel, reproducible settings and comparisons against alternatives.
| Claim presented | Book reference | What requires examination |
|---|---|---|
| 42.4% Megiddo-like ancestry in a German Ashkenazi model | Note 15, Leo Cooper’s forum model | Alternative Mediterranean sources and model dependence |
| 10.8% northern Italian ancestry | Page 4 and note 19, Ariel Lomes’s modelling | How Italian regional references partition shared ancestry |
| 4.2%, 0.8% and 0% medieval Alan contributions in different Ashkenazi groups | Note 27, Leo Cooper’s modelling | Whether the source is uniquely identifiable against alternatives |
A close fit establishes that a combination approximates the target under specified conditions. It does not independently establish the identity of the historical contributors. That distinction should accompany the book’s Jewish ancestry claims wherever percentages become migration narratives.
6. A geographical match is not an ancient ethnic identification
On page 6, Brook discusses his probable E-Y87732 paternal branch and identifies it with Israelite descent, drawing on modern matches largely associated with the Middle East. The missing step is the demonstration connecting that branch specifically to the proposed ancient population rather than to a broader regional history.
Close Y-DNA subclades remain important evidence. Italian and Jewish branch relationships deserve full attention, including Brook’s own discussion of a probable Roman Italian connection for J2-Y33795. The solution is to integrate branch structure, estimated dates, ancient carriers and autosomal context consistently.
My earlier articles examine Southern Italian and Ashkenazi genetic affinities and the corresponding paternal-lineage connections. These connections belong inside the analysis, rather than being overshadowed by a simple Europe-versus-Levant division.
7. Ancient W3b1 exposes the limits of a modern-match origin story
Brook’s W3b1 discussion on page 123 identifies Ashkenazi maternal lines from Poland, Ukraine, Belarus and Lithuania, then uses a match associated with Graz, Austria, to propose a central European convert as the first Ashkenazi carrier. By page 138, that proposed origin is presented categorically.
A modern Austrian match does not locate the shared ancestor, date entry into a Jewish population or establish migration direction. The ancient profiles below widen the evidence substantially. Their studies appeared after Brook’s 2022 book, so they test his conclusion today rather than demonstrate that he ignored publications already available to him.
PCA0376: medieval Poland, W3b1 and eastern European autosomal affinity
PCA0376 was a male buried at Santok in Poland, dated contextually to 1000–1200 CE. The Stolarek study’s supplementary records identify W3b1 and the paternal branch CTS1211. The commercial page’s “Iron Age” heading and its narrative about 600–400 BCE conflict with the medieval sample dating and should not be repeated.
All ten closest modern references supplied for PCA0376 are Russian, Belarusian or Ukrainian. Russian Kursk ranks first at 1.6584, followed closely by Belarusian at 1.6630. His W3b1 lineage therefore occurs alongside a distinctly eastern European pattern in this reference comparison.
PCA0376: ten closest supplied modern references
DNAgenics displayed distances. Smaller is closer. Shared chart scale: 0–10.
The supplied commercial model assigns PCA0376 100% to its European categories: 46.4% Northwestern European, 31.7% Eastern European and 21.9% Southern European. Those fitted labels should be read alongside the direct population ranking, not as a literal list of his ancestors’ modern nationalities.
KFJ034: Avar-period Hungary, W3b1 and a contrasting autosomal profile
KFJ034 was a female from Kunszállás-Fülöpjakab in Hungary. The supplied profile dates her to 670–720 CE; the authors’ dataset places her contextually around 700 CE in the Middle Avar period and records W3b1. She belongs to the dataset associated with Gnecchi-Ruscone and colleagues’ 2024 Avar study.
Her supplied modern-population ranking begins with Turkmen, Turkmen Uzbekistan and Turkmen Iran. Crimean Steppe Tatar, Uzbek and Tajik references follow. This is a striking contrast with PCA0376’s eastern European ranking, despite the shared maternal haplogroup.
KFJ034: ten closest supplied modern references
DNAgenics displayed distances. Smaller is closer. Shared chart scale: 0–10.
KFJ034’s supplied commercial model allocates 83.5% to Asian categories, 15.7% to European categories and 0.8% to an American category. Its largest regional terms are Northern West Asian and Central Asian/Northern Indian/Pakistani. These are model labels; small Japanese or Native American assignments are not established genealogical events.
The supplied commercial models, side by side
Each bar represents 100% of that website’s fitted output.
PCA0376
Blue: Northwestern Europe 46.4% · Teal: Eastern Europe 31.7% · Gold: Southern Europe 21.9%
KFJ034
Purple: Asia 83.5% · Blue: Europe 15.7% · Orange: America 0.8%
Expand the complete supplied model breakdowns
| Sample and category | Components supplied |
|---|---|
| PCA0376: Northwestern European 46.4% | English 15.8%; Finnish 13.3%; Northwestern European 9.6%; Scandinavian 7.8% |
| PCA0376: Eastern European 31.7% | Eastern European 31.7% |
| PCA0376: Southern European 21.9% | Balkan 19.2%; Iberian 2.7% |
| KFJ034: Northern West Asian 34.5% | Caucasian 20.1%; Mesopotamian 6.6%; Anatolian 5.0%; Cypriot 2.8% |
| KFJ034: Central Asian, Northern Indian & Pakistani 21.8% | Central Asian 11.6%; Pakistan 7.3%; Indian 2.9% |
| KFJ034: Arab, Egyptian & Levantine 9.7% | Levantine 6.1%; Arabian 3.6% |
| KFJ034: Northern Asian 9.1% | Siberian 9.1% |
| KFJ034: Japanese & Korean 5.3% | Japanese 5.3% |
| KFJ034: Chinese & Southeast Asian 3.2% | Chinese 3.2% |
| KFJ034: Northwestern European 10.6% | Northwestern European 6.3%; English 4.3% |
| KFJ034: Southern European 5.1% | Italian 5.1% |
| KFJ034: America 0.8% | Native American 0.8% |
Values preserve the supplied output. Some subtotals differ by 0.1 percentage point because of rounding.
G25 coordinates: the shared maternal label does not erase autosomal separation
The following plot uses the first two coordinates from the supplied G25 vectors directly. It shows their different positions without inventing reference-population coordinates. Full 25-dimensional comparisons contain additional information; this two-axis view illustrates the contrast rather than assigning an ethnicity.
Two ancient W3b1 carriers in supplied G25 space
PC1 horizontally; PC2 vertically. Positions calculated directly from the supplied coordinates.
PC1
PC1: 0.12919662
PC2: 0.11632982
PC1: 0.07839730
PC2: −0.03519232
Show both complete G25 coordinate strings
PCA0376
PCA0376,0.12919662,0.11632982,0.07699966,0.0654419,0.03951832,0.026016,0.01137312,0.01468788,-9.948E-05,-0.02882956,-0.00094946,-0.00979548,0.01714488,0.02179304,-0.00614572,-0.00422932,-0.00459256,-3.466E-05,0.00319488,0.00101622,-0.00049568,-0.00222536,0.00364356,0.00038314,4.464E-05
KFJ034
KFJ034,0.0783973,-0.03519232,-0.0083328,-0.01546124,-0.0315155,-0.0094135,0.00548156,0.0030949,-0.000373,0.00462178,-0.00241568,0.00163724,-0.00073758,-0.00195734,-0.00049584,0.00020624,-0.00221882,0.0037229,0.00518174,0.0001227,0.0042918,-0.0031155,-0.00299126,-0.00159282,0.00188455
These samples establish ancient W3b1 in different archaeological settings and show sharply contrasting supplied autosomal profiles. G25 cannot establish either person’s religion, and the broad W3b1 label alone does not establish that either belonged to the precise maternal branch ancestral to modern Ashkenazi carriers.
The challenge to Brook remains substantial: W3b1 cannot be geographically resolved through one modern Austrian match. A persuasive reconstruction needs complete mitochondrial branch relationships, ancient context and the broader autosomal evidence shown here. For Jewish ancestry claims about this lineage, those are central requirements.
8. Confidence in the prose exceeds confidence in the reconstruction
The recurring problem is a shift from an observation to a specific historical story: a modern match becomes a founder’s homeland; surviving lineage frequencies constrain centuries of intermarriage; a fitted coefficient becomes an ancestral population.
These transitions need arguments of their own. Brook’s catalogue contains valuable leads, but its breadth does not automatically validate each demographic conclusion. As my earlier discussion of reference sample sizes explains, the composition of the comparison dataset matters as much as the apparent precision of its output.
Kevin Alan Brook review: the genome must remain central
The strongest criticism of this book is methodological. Jewish ancestry claims deserve more than a lineage catalogue followed by confident population narratives. Autosomal evidence, Mediterranean reference choice, genetic drift and ancient samples must actively determine the reconstruction.
PCA0376 and KFJ034 make the issue visible: one maternal haplogroup, two strikingly different autosomal profiles. Close paternal and maternal relationships remain meaningful, but the rest of the genome cannot be pushed into the background when the question is where an entire population came from.
Sources and data
- Kevin Alan Brook, The Maternal Genetic Lineages of Ashkenazic Jews (Academic Studies Press, 2022), especially pages 3–8, 123 and 138; notes 15–19, 27 and 686–687.
- Steven Parker, preprint, version 2, especially Table 7. The model chart reproduces author-reported results.
- Xue et al. (2017), “The time and place of European admixture in Ashkenazi Jewish history.”
- Drineas et al. (2019), “Genetic history of the population of Crete.”
- Stolarek et al. (2023), “Genetic history of East-Central Europe in the first millennium CE.” See supplementary tables for PCA0376.
- Gnecchi-Ruscone et al. (2024), “Network of large pedigrees reveals social practices of Avar communities.” See the associated authors’ sample metadata for KFJ034.
- DNAgenics sample profiles supplied for PCA0376 and KFJ034: source of the displayed modern-population rankings, commercial model outputs and G25 coordinate strings. These outputs are separate from the original studies’ analyses.