550 THE STRUCTURE OF EVOLUTIONARY THEORY
stocks with more successful systems: it must be by this process that group-
characters slowly evolve.The abstract logic of this argument cannot be faulted, but we must ask if the
required conditions are encountered frequently enough in nature. Do social groups
remain sufficiently exclusive; is group selection strong enough to overcome
Darwinian organismic selection; do social behaviors originate by organismal or
group selection? Nearly all evolutionists would now agree that groups rarely
maintain the required cohesion, and that group selection (in Wynne-Edwards's
mode) will usually be far too weak a force to prevail over the conventional
Darwinian mechanism of organismic selection.
George Williams's brilliant book, Adaptation and Natural Selection (1966),
provided the historical focus for general rejection of Wynne-Edwards's theory.
Williams wrote with other sources and targets in mind as well. (He told me, for
example, that he had originally been most strongly motivated by the false
arguments of Allee, Emerson and the Chicago school.) But Wynne-Edwards stood
out as the main group-selectionist game in town when Williams wrote his book.
Williams's book won deserved influence by its incisiveness of logic and
argument, and for its persuasive style of composition. (I know no better example of
a work that prevailed primarily by the entirely honorable sense of the unfairly
maligned word "rhetoric," properly defined as "the effective use of language.")
Williams begins by characterizing adaptation as an "onerous" concept that should
be invoked only when all simpler explanation fails. We should then become all the
more impressed when we find that we need to invoke adaptation so often! Having
established one "tough" criterion by permitting the invocation of adaptation only
when all else fails, Williams then proposes another—this time governing the
advocacy of levels higher than the conventional Darwinian focus upon organisms.
In short, Williams states, don't make a claim about higher levels unless both logic
and empirics permit no other alternative. Adaptation is onerous enough in any
case; if we must call upon such a mechanism, we should do so at the lowest
possible level of the genealogical hierarchy. This appeal to some form of
parsimony or reduction leads Williams to reject all claims for group selection, so
long as Darwinian organismic selection can render the same phenomenon in
principle. Williams presents his argument largely as a theoretical proposition, and
only rarely as an empirical claim. If the phenomenology of a situation can be
rendered by an organismic interpretation, he asserts, one should then advocate this
lower level of causality—even if a group selectionist scenario violates no tenet of
logic or plausibility.
In his introductory pages, Williams tells us that claims for group selection
have inspired his attempt at cleansing and simplification:
Even among those who have expressed the opinion that selection is the sole
creative force in evolution, there are some inconsistent uses of the concept.
With some minor qualifications to be discussed later, it can be said that
there is no escape from the conclusion that natural selection...