Punctuated Equilibrium and the Validation of Macroevolutionary Theory 915
and exemplify the fractal reach of this speciational reformulation into the smallest
nooks and crannies of conventional wisdom.
Throughout my professional life, I have read, over and over again, the almost
catechistic claim that the increase in cranial capacity from Homo erectus to Homo
sapiens (variously specified as a 50 percent growth from about 1000 cc to 1500 cc
as a lower estimate, to a doubling from 750 to 1500 as an upper bound) represents
a stunning example of evolution at maximal rate, something so unusual and
unprecedented that we must seek a cause in the particular adaptive value, and
potential for feedback, of human consciousness. (Again, I suspect that our mental
predisposition to commit such errors arises from an overwhelming desire to find
something unique not only in the result—a defensible proposition—but also in the
biological mechanism of human consciousness.) This claim for a maximal rate
presumes anagenetic transformation, with the high value then calculated by
spreading the total increment over the short amount of available time (from
100,000 years as a lower bound to a million years as an upper limit).
Two major errors, one obvious and almost ludicrous, the other more subtle
and speciational, promote this "urban legend" that would disappear immediately
from the professional literature if people only stopped to think before they copied
canonical lines into their textbook manuscripts. First, the claim isn't even true
within its own assumption of anagenetic transformation. Such a rate should not be
designated as rapid at all when we recognize the proper scaling between our
estimates of selection's strength in ecological time and its effect in geological time.
Williams (1992, p. 132), for example, cites a standard claim and then presents
some calculations:
Even some widely recognized examples of rapid evolution are really
extremely slow. Data on Pleistocene human evolution are interpretable in
various ways, but it is possible that the cerebrum doubled in size in as little
as 100,000 years, or perhaps 3000 generations (Rightmire, 1985). This,
according to Whiten and Byrne (1988) is "a unique and staggering
acceleration in brain size." How rapid a change was it really? Even with
conservative assumptions on coefficient of variation (e.g. 10%) and
heritability (30%) in this character, it would take only rather weak selection
(s = 0.03) to give a 1% change in a generation. This would permit a
doubling in 70 generations. An early hominid brain could have increased to
the modern size, and back again, about 21 times while the actual evolution
took place. Indeed, it is plausible that a random walk of 1 % increases and
decreases could double a quantitative character in less than 3000
generations.If this first error of scaling has been identified before, the second and deeper
fallacy of false assumptions about mode has generally escaped the notice of critics.
The anagenetic assumption that trends represent the flux of a central tendency in a
species's global transition to a better form must be replaced, in most cases, with a
speciational account of trends as differential success of certain species within a
clade. When we add the additional observation that