Nature versus Nurture

In brief

The old question was framed as a contest: are we the product of our biology or our upbringing? Behavioral genetics answered it by refusing the "or." Its tool is not the individual but the population, and its unit is not a trait but the variation in a trait across people. Twin and adoption studies partition that variation into a portion that tracks with shared genes and a portion that tracks with shared environment. Across thousands of studies and nearly every complex human trait, the verdict has been consistent and, to some, unsettling: genetic differences account for a substantial share of why people differ, usually somewhere between a third and two thirds. Yet the same field that produced those numbers now regards the nature-versus-nurture dichotomy as a badly posed question. Genes and environments do not add together like weights on a scale; they correlate, they interact, and they switch each other on and off.

The full treatment

The problem it answers

Francis Galton (1822 to 1911), Charles Darwin's cousin, gave the debate its modern shape in Hereditary Genius (1869) and popularized the "nature and nurture" phrase soon after. He noticed that eminence ran in families and concluded that ability was inherited, a conclusion entangled from the start with the eugenics he founded. The philosophical version of the question is far older and is treated elsewhere in this library: the empiricist claim, associated with John Locke, that the mind begins as a blank sheet and is written on by experience, against the rationalist and nativist claim that some structure is present at birth. This entry stays with the empirical descendant of that quarrel. Once you accept that both genes and experience matter, a measurable question remains: for a given trait in a given population, how much of the observed variation is attributable to each? That is the question behavioral genetics was built to answer.

How the method works

The logic exploits a natural experiment. Identical (monozygotic) twins share essentially all of their DNA; fraternal (dizygotic) twins share about half, like ordinary siblings. If a trait is more similar in identical than in fraternal twins, the extra similarity points to genetic influence, provided the two kinds of twins share their environments to a comparable degree. From the difference in correlations, researchers estimate heritability, conventionally written h squared: the proportion of variance in a trait, within a specific population and environment, attributable to genetic variance. Adoption studies supply the complementary test, comparing adopted children to their biological and adoptive parents, so that shared genes and shared homes can be pulled apart. The strongest design combines both: twins separated in infancy and raised in different families, who share genes but not upbringing.

The key study

The Minnesota Study of Twins Reared Apart, led by Thomas Bouchard from 1979, tracked down more than one hundred pairs of identical and fraternal twins separated early in life and reunited, in some cases, only for the study. Reared in different homes, often unaware of each other, they were flown to Minneapolis for a week of intensive testing. The 1990 report in Science found that on measured intelligence the reared-apart identical twins correlated about 0.70, a figure implying that roughly seventy percent of the variance in adult IQ in that sample was genetic. Personality traits showed heritabilities near fifty percent. The finding that drew the most comment was the near-irrelevance of the shared home: growing up in the same household seemed to make siblings barely more alike as adults than growing up apart. Eric Turkheimer later distilled the accumulated evidence into "three laws" (2000): all human behavioral traits are heritable; the effect of being raised in the same family is smaller than the effect of genes; and a substantial portion of the variation is explained by neither, leaving room for chance, unique experience, and error.

The distinctions that matter

Heritability is the most misunderstood number in the field. It is a property of a population, not a person: it says nothing about how much of any individual's intelligence or height is "due to genes." It is also not fixed. A trait can be highly heritable and still be highly malleable; height is strongly heritable, yet average height has risen with nutrition. And heritability rises as environments become more uniform, because when everyone gets similar opportunities, the differences that remain are more genetic by default. Two further ideas dissolve the dichotomy directly. Gene-environment correlation means genes and environments are not independent: parents who pass on genes for verbal ability also tend to fill the house with books, and children with a given temperament seek out matching surroundings, a typology Sandra Scarr and Kathleen McCartney sorted in 1983 into passive, evocative, and active forms. Gene-environment interaction means the effect of a gene depends on the environment, and vice versa.

Lineage

The empirical program grew from Galton's hereditarianism, was formalized by the biometricians and by R. A. Fisher's 1918 reconciliation of Mendelian genetics with continuous traits, and matured with the twin and adoption designs of the mid-twentieth century. It sits downstream of natural selection, which gives heritable variation its point, and alongside the statistical machinery of variance and correlation that makes partitioning possible. Robert Plomin and Turkheimer are its leading contemporary voices, though they read the same data toward different emphases. The philosophical parentage, the empiricist and nativist theories of mind, runs back through Locke, Leibniz, and Kant, and belongs to a separate lineage this entry deliberately does not retrace.

The strongest case for it

The evidence is broad, replicated, and convergent across methods that have different weaknesses. Twin studies, adoption studies, and reared-apart designs, each vulnerable to a different confound, point the same way, which is hard to explain if the signal were an artifact of any one design. The finding that shared family environment contributes little to adult personality and cognition was genuinely counterintuitive and forced developmental psychology to take genetic influence seriously rather than assuming parenting explained the outcomes. More recently, molecular methods have corroborated the twin-study numbers from an entirely independent direction: genome-wide association studies (GWAS) and the polygenic scores built from them detect heritable signal directly in DNA, without any twin comparison, and recover the same conclusion that many traits are influenced by thousands of variants of tiny individual effect. When two methods with nothing in common agree, the underlying claim earns confidence.

The strongest case against it

The critiques are serious and come from within the field as often as from outside it. The oldest targets the equal environments assumption: identical twins may be treated more alike than fraternal twins, dressed the same, mistaken for each other, expected to share interests, in which case the extra similarity attributed to genes is partly environmental. Leon Kamin and, more recently, the psychologist Jay Joseph have pressed this point hard, arguing the assumption is not merely imperfect but false in ways that inflate heritability estimates. A second problem is empirical and came from molecular biology itself: the "missing heritability" gap. Twin studies said intelligence was seventy percent heritable, yet for years the specific genes identified by GWAS accounted for only a few percent of the variance, a discrepancy that remains only partly closed and that some read as evidence the twin-study figures were too high. A third critique dissolves the number from a different angle. Turkheimer, Haley, and colleagues found in 2003 that among children in poverty, the heritability of IQ was near zero, while among the affluent it approached the familiar high figures. If heritability itself depends on the environment, then reporting a single number for "the" heritability of intelligence is meaningless, and the whole enterprise of ranking nature against nurture collapses. Caspi and Moffitt's celebrated gene-by-environment findings, that a variant of the MAOA gene raised the risk of antisocial behavior only in the maltreated (2002), and that a serotonin-transporter variant predicted depression only under stress (2003), initially seemed to vindicate interaction, but several failed to replicate at scale, a cautionary tale about candidate-gene studies that the field has largely absorbed. Finally, critics from the social sciences warn that heritability estimates are routinely misread as licensing fatalism about education or as underwriting claims about group differences, uses the mathematics does not support.

Where it stands now

Among behavioral geneticists the dichotomy is treated as resolved, though not in the way either historical camp wanted. Almost every trait is partly heritable; almost every trait is also shaped by environment; and the interesting action is in how the two are entangled rather than in their relative sizes. Epigenetics has sharpened the point: environmental signals, stress, diet, early care, can alter which genes are expressed without changing the DNA sequence, giving a molecular mechanism to the old intuition that experience gets "under the skin." The frontier has shifted from twins to genomes. Polygenic scores can now predict a modest fraction of variation in traits like educational attainment directly from DNA, which excites researchers and alarms ethicists in equal measure, since the same scores raise the specter of embryo selection and revive the group-difference disputes that have shadowed the field since Galton. What is settled is that "nature or nurture?" was the wrong question. What is not settled is how to talk responsibly about the answer.

Test yourself

Think of a trait you are sure you inherited, a temper, a knack, a vulnerability, and one you are sure you learned. Now ask which environments made the inherited one show itself, and which inheritance made you seek out the experience that taught you the learned one. If you cannot cleanly assign either to nature or to nurture, you have understood the finding: the two categories were never separable to begin with.