|
The trial-and-error method
different possible configurations are generated, after a test of their "fitness", the good ones are retained, and the bad ones or "errors" are eliminated
|
According to the neo-Darwinist view, evolution takes place through the
creation of random combinations of matter, with the subsequent struggle
for existence, as a result of which some combinations survive and proliferate,
while other perish. Popper describes this as the work of
the general method of trial and error-elimination. Campbell
uses the term blind variation and selective retention. Newell and Simon in their theory of problem solving called this mechanism "generate and test". Here we will speak simply of the trial and error method.
We do not need to use the term `blind', because in cultural evolution or in problem solving
we often have informed and guided choices. But even with regard to biological evolution we cannot be sure, much less prove, that the
variation is blind. It is true that we build our theory and check it
against facts in the assumption that variations are blind.
But we do not really use the fact that the variation is, indeed, blind or random,
i.e. all choices physically and logically possible are equiprobable.
The success of the theory proves that blindness, at the present
state of the theory, is sufficient, but does not prove it is necessary. The main requirement is that a large number of possible states or solutions is explored through a process of variation.
The principle is so powerful that any type of variation or trial, whether guided by foreknowledge or not, followed by the elimination of the "bad", or "unfit" trials, and the retention or propagation of the "fit" trials, will result in evolution. The fact that succesful steps are retained leads to an irreversible accumulation, a "ratchet effect" which only allows movement in a particular general direction, without going back (see the principle of asymmetric transitions).
Copyright© 1996 Principia Cybernetica -
Referencing this page
|
|
|
