PINNING DOWN POTENTIAL 130 to 1, or as a percentage). It is estimated from breeding experiments
under controlled environmental conditions.
In the hands of psychologists, however, heritability has— like the con-
cept of potential itself— been consciously or carelessly obfuscated. Instead
of a statistical ratio, it has come to be understood— and presented to the
public and media—as the deterministic concept of inheritability: that is,
telling us the degree to which a trait value, such as IQ, for any individual
is determined by heredity. Ideologically transmogrifi ed in that way, heri-
tability is read as the degree to which diff erences among individuals,
social classes, and ethnic groups are scientifi cally proven to be ge ne tic, or
immutable.
In humans, the heritability of “intelligence” has been estimated by
comparing resemblances in, say, IQ scores among pairs of twins. Identi-
cal twins share all the genes involved in variation, whereas nonidentical
twins share only half of those genes. If the identical twins are more alike,
then it is concluded that ge ne tic variation is impor tant. Formulas have
been devised for estimating the heritability from those resemblances, and
the estimates have usually been quite high.
Th is logic overlooks the simple (and, indeed, demonstrable) possibil-
ity that identical twins will also share more environment, in the sense of
being treated more alike by parents, teachers, and so on (an Achilles’ heel
of twin studies to be more fully discussed in chapter 2). In the heady
announcements of heritabilities from twin studies, such possibilities are
virtually swept aside. But there are many other fl aws and assumptions in
twin studies that I discuss at length in chapter 2.
Th e more impor tant truth, though, is that even a perfectly accurate
heritability estimate of human potential or intelligence— which we cer-
tainly do not have— tells us nothing about the ge ne tic makeup of individu-
als or subgroups reared in diff er ent environments. To suggest other wise is
an elementary falsehood (for reasons to be made clear in chapters 2 and 4).
Outside agricultural breeding programs, where environment is carefully
controlled, it is a completely pointless statistic— except, that is, for its
ideological thrust.
In spite of those prob lems, the past two or three de cades have seen an
intense new focus on twin studies and heritability estimates, backed by
streams of funding and turning out new waves of results enthusiastically
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:51:16 UTC