The Westermarck effect: why we rarely partner with the people we grew up with

The Finnish anthropologist Edvard Westermarck proposed in 1891 a hypothesis that ran counter to Freudian dogma of the era. Where Freud argued that incest taboos existed because humans wanted to mate with kin (and culture had to forbid it), Westermarck proposed the opposite: humans who grow up in close proximity from early childhood develop automatic sexual aversion to each other, regardless of biological relatedness.

The hypothesis was largely dismissed for most of the 20th century. The empirical data that has accumulated since 1970 has revived it. The findings are striking.

1. The Israeli kibbutz evidence

Joseph Shepher's 1971 study of marriages among adults raised in Israeli kibbutzim provided the first systematic test. Children in kibbutzim were raised collectively in age-graded peer groups, spending most of their childhood with the same dozen or so children of the same age, regardless of biological relationship.

Shepher analyzed marriage patterns among 2,769 adults raised in this system. The result: across the entire sample, not one marriage had occurred between two adults who had spent their first six years together in the same kibbutz peer group. The expected rate, given random mating, would have been several dozen (Shepher, 1971).

This was the first strong empirical evidence for Westermarck's hypothesis. Co-residence in early childhood produced sexual aversion in adulthood, even between unrelated individuals.

2. The Taiwanese minor-marriage evidence

Arthur Wolf's 1995 analysis of Taiwanese shim-pua marriages — arranged marriages where the bride was raised in the groom's household from early childhood — found similar patterns. These marriages, where partners had effectively been raised as siblings, showed:

• Lower fertility (controlling for everything)
• Higher divorce rates
• Higher rates of extramarital affairs
• Lower marital satisfaction

The longer the partners had co-resided in childhood, the worse the outcomes. Marriages where partners met as adolescents or adults didn't show these patterns (Wolf, 1995).

3. The mechanism

The proposed mechanism: an evolved psychological adaptation that detects co-residence in early childhood (roughly birth to age 6) and uses it as a proxy for genetic relatedness. The proxy was reliable in ancestral environments where children mostly grew up with kin. The adaptation produces sexual aversion to the detected "kin."

The mechanism works correctly when applied to biological siblings (preventing incest). It works incorrectly when applied to unrelated co-residents — but the false positive is rare in evolutionary terms, and the cost of incest avoidance is presumably worth the occasional non-matches.

4. The genetic data

Modern genetic studies have tested the related question: do humans avoid mating with kin without conscious knowledge of relatedness? The evidence suggests yes, partially:

• Anonymous donor-conception adults show some discomfort when finding out a romantic partner is a genetic half-sibling, even when they had no childhood contact
• This suggests additional mechanisms beyond Westermarck operate (possibly olfactory MHC-based kin recognition)
• But Westermarck's co-residence mechanism appears to be the strongest single signal in humans

5. The cultural implication

The Westermarck effect explains a striking cross-cultural pattern: incest taboos are universal across human societies, but they typically focus on close kin and don't extend to distant cousins. The taboo distribution aligns with kin categories the Westermarck mechanism would normally cover.

This is an unusual case of a 19th-century evolutionary hypothesis being repeatedly confirmed by 20th and 21st century data. Most pre-Darwin and early-Darwin psychological speculation has aged badly. Westermarck's specific claim — that early co-residence produces adult sexual aversion — has aged well.

For modern individuals: the effect operates without conscious awareness. Adults rarely think about why they don't find childhood-companion adults attractive. The brain has decided the question without consultation, using a heuristic developed millions of years ago.