selective eYcacy of environmental pressures. So one might wonder
whether empiricists could not avail themselves of a similar response and
urge that, since childrendolearn, the stimuli to which they are exposed
must be considerably richer than nativist theorists imagine.
However, there is a signiWcant disanalogy between ontogeny and
phylogeny which needs to be taken into account in these two areas of
theoretical debate. The diVerence is that individual development (on-
togeny) conforms to a cognitive pattern for the species as a whole, whereas
there is no comparable phylogenetic pattern which constrains speciation.
A well-known, yet only partly acceptable, dictum of evolutionary specula-
tion has it that ‘Ontogeny recapitulates phylogeny’. As far as the funda-
mentals of psychological development are concerned, it is a lot safer to say
that ‘Ontogeny recapitulates ontogeny’ – that is, that the development of
individuals follows a similar course to that of other individuals of the
species.
Suppose, by way of thought-experiment, that we intervened in the
evolutionary process to produce geographical isolation, taking similar
stocks of a species and depositing them in quite diVerent environments –
sending one batch to Australia, another to a tropical rain-forest, a third to
temperate grasslands, and so on. Assuming the shock of relocation does
not lead to complete extinction, and allowing a short interval for speci-
ation (say a million years or so), we return to survey the results. Supposing
that the environmental conditions encountered by the several branches of
the ancestral stock had continued to diVer, would we expect toWnd parallel
development and closely similar species in all those diVerent environ-
ments? No, surely not. Geographical isolation leads to divergent speci-
ation – as indicated by Darwin’s study of the GalapagosWnches, and by
what has happened on Madagascar since it was separated from the African
mainland.
If we were toWnd that while the species had evolved it had evolved in
much the same way in those diVerent locations, in apparent deWance of
environmental variation, then we really might start to think there must be
something in the idea of a pre-determined path for phyletic evolutionary
development, somehow already foreshadowed in the experiment’s initial
gene-pool. Arthur Koestler believed that something like this is true, and
that evolutionary development follows certainchreods(Greek for ‘pre-
determined paths’; see Koestler and Smythies, 1969). There are indeed a
few cases of remarkable similarities between species which are quite distant
from each other in terms of descent – such as the Siberian wolf and the
Tasmanian wolf. But only a few. So there is no good case for chreods in
phylogenetic development. The strength of the nativist case, by contrast,
derives from the fact that there really do seem to bechreodsfor human
The case for nativism 53