70 PRETEND GENESphenylalanine from the diet). Likewise, a heritability of 0 does not mean
that genes are not involved in the development of form/function and
phenotypic variation. It simply means there is little correlation between
them. Consequently, the heritability of a trait has no implications at all
for predicting individual potential— for example, a child’s IQ from that
of its parents—or the success of intervention.
In fact, genet ically minded psychologists will occasionally acknowl-
edge that demonstrable fact— but then rapidly forget it. For example, in
their book G Is for Genes, Kathryn Asbury and Robert Plomin warn
about the limitations of the concept of heritability but then subtitle their
book Th e Impact of Ge ne tics on Education— deduced entirely from heri-
tability estimates! Only in a world of pretend genes could we infer such
“impact” from heritability estimates. Again, we need to ask, what is really
going on in the pursuit of pointless heritability estimates?
We must also note the widespread practice of claiming, from heri-
tability estimates, that ge ne tic variation has caused this or that trait
variation. Terms like “ge ne tic infl uence,” “responsible for,” or “due to”
are routinely used to report what are only statistical associations. For ex-
ample, in the same book (p. 16), Asbury and Plomin use phrases like “we
use twins to estimate how much of the diff erences between people...
is due to ge ne tic infl uence.” I described in chapter 1 how this is quite mis-
leading wordplay.
Ideology infi ltrates science when its under lying concepts are hazy
and permeable. In this chapter I have shown how that is the case with the
genes said to determine IQ. Th e model of individual diff erences based
on such genes entails assumptions that are clearly false. Moreover, the
latitude built into the model seems to have invited a make-do empirical
culture in the methodology intended to test it. Apart from the lack of
test validity in IQ tests, generally, scores have oft en come from shortened
forms administered in far from ideal conditions. More seriously, the in-
trinsic fl aws in twin studies— especially the EEA— and of adoption
studies, make substantial (pseudo-) heritability inevitable in almost any
trait studied. Other statistical attempts to estimate heritability for IQ
have their own drawbacks. I have also discussed how the concept of
heritability tends to be misunderstood and is pointless in human
populations.
This content downloaded from 139.184.14.159 on Tue, 17 Oct 2017 13:51:26 UTChttp://www.ebook3000.com