Female Farmers, Jōmon Men, and the Hidden Polygyny Assumption
A new genetic study argues that continental women, not unusually successful Jōmon men, created Japan’s sex-biased ancestry. But that distinction only works if some men reproduced with multiple women.
A new preprint on the formation of the modern Japanese population makes a striking claim. According to it, the genetic asymmetry found in modern Japanese was produced primarily by a female-biased contribution from continental farming migrants, rather than by the male-biased Jōmon contribution traditionally invoked to explain the pattern. (bioRxiv)
At first sight, these sound like opposite explanations.
They are not, at least not under monogamy.
In a population formed from only two ancestral groups, more ancestry from continental women automatically means relatively more ancestry from Jōmon men. The two statements describe the same contrast from opposite directions.
The paper can distinguish them only because it estimates something more specific: the effective numbers of male and female founders within each ancestral population. Its results suggest that continental women greatly outnumbered continental men, while Jōmon men and women contributed in approximately equal numbers.
Taken literally, however, that demographic model implies polygyny or an equivalent form of male reproductive skew.
The Japanese “admixture paradox”
Modern mainland Japanese descend largely from continental East Asian migrants, together with a smaller contribution from the indigenous Jōmon population.
Earlier genetic estimates produced an apparent contradiction. Jōmon ancestry was estimated at around 10% on the autosomes, but a Jōmon-associated Y-chromosome lineage occurred in approximately 30% of Japanese men. The estimated Jōmon contribution to mitochondrial DNA was intermediate, at around 15%. This mismatch became known as the admixture paradox.
Because the Y chromosome passes from father to son, while mitochondrial DNA passes from mothers to all their children, the obvious interpretation was that Jōmon men contributed more to the emerging Japanese population than Jōmon women did.
The new paper revisits this problem using the X chromosome.
Autosomes receive exactly half their ancestry from mothers and half from fathers. The X chromosome is different: across generations, approximately two-thirds of X chromosomes come through women and one-third through men. Therefore, if continental ancestry is more common on the X chromosome than on the autosomes, it indicates that continental ancestry entered disproportionately through women. The authors find this directional pattern. (bioRxiv)
But this result alone does not distinguish female-biased continental ancestry from male-biased Jōmon ancestry. With only two sources, they are mathematical complements.
The paper’s more interesting result comes from its ancestry-specific drift analysis.
What the paper actually found
The authors calculate a statistic called (Q), which compares genetic drift on the X chromosome with drift on the autosomes. Under their model, the expected value with balanced male and female effective contributions is 0.75.
The results were:
Jōmon-related ancestry: (Q=0.78), not significantly different from 0.75.
Continental farmer ancestry: (Q=0.88), significantly above 0.75.
The difference between the two ancestry components was also statistically significant. (bioRxiv)
Under a simplified single-pulse founder model, the authors convert these estimates into approximately: 1.2 Jōmon women per Jōmon man and 3.1 continental women per continental man.
The first ratio is statistically compatible with an equal number of Jōmon men and women. The second indicates a strongly female-biased continental contribution. (bioRxiv).
This is the basis for the abstract’s claim.
The old model supposedly required an unusually male-biased Jōmon founder population. The new model instead has approximately equal numbers of Jōmon men and women, but about three continental women for every continental man.
The missing piece is how all these women reproduced.
What strict monogamy predicts
Let:
(F_f) = continental female contributors
(F_m) = continental male contributors
(J_f) = Jōmon female contributors
(J_m) = Jōmon male contributors
Under strict lifetime monogamy, every reproducing woman has one reproducing male partner and vice versa. Consequently, the total number of male and female contributors must be equal:
F_f+J_f=F_m+J_m.
Rearranging gives:
F_f-F_m=J_m-J_f.
This simple equation is the heart of the problem.
It says that, under monogamy, every excess continental woman must be matched by an excess Jōmon man. Therefore:
A female-biased continental contribution necessarily produces a male-biased Jōmon contribution of exactly the same absolute size.
Consider a population containing:
30 continental women
10 continental men
100 Jōmon women
100 Jōmon men
Suppose everyone who reproduces is strictly monogamous.
The 30 continental women can marry 30 different Jōmon men. The 10 continental men can marry 10 Jōmon women. That leaves:
70 unpaired Jōmon men
90 unpaired Jōmon women
Seventy further Jōmon couples can form. Twenty Jōmon women are left without reproductive partners.
The effective contributors are consequently:
The continental contribution is female-biased by 3:1. But the Jōmon contribution is simultaneously male-biased: 100 Jōmon men contributed, compared with only 80 Jōmon women.
Nothing about repeated migration changes this logic. Three continental women arriving every generation can each marry a different local man, so the marriages remain monogamous. But every generation, their local husbands are removed from the pool available to local women. Over time, the cumulative continental contribution becomes female-biased and the cumulative Jōmon contribution becomes male-biased.
Under monogamy, the two patterns are inseparable.
How polygyny changes the prediction
Now consider a different founder population:
30 continental women
10 continental men
20 Jōmon women
20 Jōmon men
The continental ratio is again 3:1, but the Jōmon sex ratio is perfectly balanced.
There are 50 women and only 30 men. All 50 women can reproduce only if some men reproduce with more than one woman.
For example:
Each of the 20 Jōmon men has children with one Jōmon woman and one continental woman.
Each of the 10 continental men has children with one continental woman.
The resulting contributors are:
No Jōmon women are excluded. The Jōmon founder population remains sex-balanced, while the continental contribution is strongly female-biased.
But this outcome requires the Jōmon men to reproduce with two women each.
That is polygyny.
The two mating systems therefore predict different ancestry-specific results:
The paper observes the second pattern: strong female bias in the continental component, without statistically significant male bias in the Jōmon component.
In other words, the result does not merely indicate female-biased migration. Under the literal founder interpretation, it indicates that the migrant women did not simply replace Jōmon women in monogamous unions. Instead, additional women entered the reproductive population without a corresponding exclusion of native women.
That requires some men to father children with multiple women.
Does this prove polygyny?
If the inferred founder ratios are interpreted literally, then yes: they imply polygyny in the biological sense, including serial polygyny.
The arithmetic is unavoidable. If more women than men reproduced, then at least some men must have fathered children with more than one woman:
Number of reproducing women> Number of reproducing men
This could have taken several forms:
one man having children with several wives or partners simultaneously;
a man having children with successive wives after death, separation or remarriage;
informal or extramarital reproduction;
the same men reproducing with women arriving in different migration waves.
All of these are forms of polygyny from a demographic and genetic perspective, because one male reproduces with multiple females.
Warfare, elevated male mortality or differences in social status could explain why fewer men reproduced. But they do not remove the implication of polygyny. If many men died or were excluded while most women still had children, the surviving or socially successful men must have reproduced with multiple women.
The genetic evidence cannot tell us which social arrangement produced this pattern. It cannot distinguish a chief with several simultaneous wives from a widower who had children with two successive wives, or from less formal multiple mating. For that reason, it would be too strong to claim that the study proves an institution of formal polygynous marriage.
But “male reproductive skew” alone is also too vague if the founder ratios are taken as literal numbers of reproductive participants. In that case, the skew necessarily involved polygyny or serial polygyny.
There remains an important statistical qualification. The paper’s estimates are effective genetic ratios, inferred from X-chromosomal and autosomal drift under a simplified demographic model. They are not direct counts of prehistoric husbands and wives. Complex migration histories, bottlenecks or differences in the long-term survival of lineages could cause the estimated ratios to differ from literal census numbers.
The appropriate conclusion is therefore:
If the estimated female-to-male founder ratios correspond approximately to the actual numbers of people who reproduced, the population must have practised simultaneous or serial polygyny. The genetics cannot determine the precise social form that this polygyny took. Female migrants and successful Jōmon men are not mutually exclusive.
Female-biased migration does not eliminate the need for polygyny
The abstract contrasts two explanations:
Jōmon ancestry was disproportionately transmitted by men.
Continental ancestry was disproportionately introduced by women.
These are not simply opposite descriptions once the paper estimates sex ratios separately within each ancestral component. The authors report a strongly female-biased effective contribution among continental migrants, while the Jōmon-related contribution is statistically compatible with equal numbers of male and female founders.
However, if these founder ratios are interpreted literally, the preferred model still requires some men to have children with multiple women.
Suppose continental women outnumbered continental men by approximately three to one, while similar numbers of Jōmon men and women reproduced. There would then be more reproducing women than reproducing men in the population as a whole. Strict lifetime monogamy could not produce this result. Some men must have reproduced with multiple women, either simultaneously or successively.
The genetic results do not show which men did so. They could have been Jōmon men, continental men, or men of both ancestries. A plausible scenario is that some Jōmon men had children with both Jōmon women and migrant continental women, but this is an additional historical interpretation rather than a direct result of the analysis.
The distinction from the conventional model is therefore narrower than the abstract suggests:
The conventional interpretation attributes the ancestry asymmetry to a disproportionately male Jōmon contribution.
The new interpretation attributes the clearest sex imbalance to the continental component, which contained many more effective female than male contributors.
The new result does not remove male multiple mating from the demographic picture. On a literal reading of the founder ratios, it requires polygyny or serial polygyny somewhere in the admixed population. What it does not establish is that the polygynous men were specifically Jōmon.
A more accurate summary would be:
Continental ancestry shows evidence of a strongly female-biased effective contribution, whereas the Jōmon component shows no detectable imbalance between male and female founder numbers. If these estimates represent literal reproductive participants, their combination requires simultaneous or serial polygyny, although the genetic data do not identify the ancestry of the men who reproduced with multiple women.
Thus, female-biased farmer migration and male multiple mating are not competing explanations. Female-biased migration describes the ancestry and sex of the incoming population; polygyny describes the mating system required to accommodate the excess female contributors without excluding a corresponding number of Jōmon women.