humans who looked exactly like us behaved, as far as can be told, 
pretty much like Neanderthals, for upward of 50 kyr. These humans 
had brains that were externally like our own, but that evidently did 
not function in the way that the Cro-Magnons’ did in later times. 
What happened? Did the earliest anatomically modern and the ear- 
liest behaviorally modern humans represent separate but skeletally 
identical species, the latter eventually replacing the former? This 
scenario seems inherently improbable, since any such dramatic Old 
World-wide replacement would have had to have taken place in an 
implausibly short window of time; and there is, in any event, no 
direct evidence for it. The only evident (and as we’ve seen, in terms 
of evolutionary mechanisms far from unusual) alternative is that the 
potential for the unique human capacity was born with our species 
Homo sapiens as a byproduct of some other change, and that it lay 
fallow, as it were, until unleashed by a cultural (rather than biolog- 
ical) stimulus. This capacity, once declenched, would then have 
spread quite easily by cultural contact among populations that al- 
ready possessed the latent ability to acquire it. 
What might that releasing stimulus have been? Like many others, 
I am almost sure that it was the invention of language. We must 
bear in mind here that by the time Homo sapiens came on the scene 
the peripheral equipment that allows articulate speech had been 
around for several hundred thousand years, having emerged initially 
for other purposes entirely. The archaeological record is but a dim 
record of the full panoply of behaviors of any early hominid, but if 
it shows us anything at all it is the starkness of the contrast between 
the torrential outpouring of symbolic behaviors by the Cro-Magnons 
and the essentially symbol-free behaviors of their predecessors. The 
fundamental innovation that we see with the Cro-Magnons, under- 
writing all their varied achievements, is that of symbolic thought, 
with which language is virtually synonymous. Like thought, lan- 
guage involves forming and manipulating symbols in the mind, and 
our capacity for symbolic reasoning is virtually inconceivable in its 
absence. Imagination and creativity are part of the same process, for 
only when we create mental symbols can we combine them in new 
ways and ask ‘‘What if?’ Nonverbal intuitive reasoning can, of 
course, take one a long way (recall the toolmaking experiment); and 
24 
