At 05:47 PM 09/14/2000, you wrote:
Marcio
>Hi, folks!
>
>I think there are two important points you guys are missing.
>
>First of all, I think that if we want to discuss something in scientific
>terms we must stick to the scientific jargon regardless of how people
>usually use a word. Otherwise we are comparing apples and oranges. Words
>such as random, chaos and homology may cause problems like that. In this
>sense, there is no association between the word "random" and the idea of
>purpose. Curiously, some people commonly use random variables to achieve
>the solution of optimization problems (e.g. Monte Carlo models).
>
>
>Also, the distinction between "deterministic" and a "random" is not
>necessarily true. Even though it is impossible to predict the outcome of a
>single coin toss, if you toss it a sufficiently large number of times you
>will get a frequency distribution (a probability distribution) that is
>highly deterministic.
Chris
This is an interesting point, but you are not distinguishing between
determinism as it applies to *an* event and determinism as it applies to a
probability distribution. Further, the *actual* distribution may not match
the probability distribution exactly. Also, as has been pointed out by
others as well as myself, some seemingly random sequences are definitely
deterministic if you know the mechanism behind them. A computer might be
printing out what looks like random digits but they may actually be digits
of the square root of a given number, being calculated one after the other
and printed out.
There is, this suggests, no magic way of distinguishing *true* randomness
(if it could occur) and something that may merely be the "chaotic" working
out of a deterministic process.]
But, Darwinian evolution does not need true randomness, anyway. For it to
be true, it is only required that the process not be manipulated and that
there be enough variation in the variations to provide sufficient "raw
material" to ensure that some variations in some cases will be beneficial
to the genotype in which they occur, and often enough to enable the
organism to keep up with changes in the environment, etc. A strictly
algorithmic process (such as calculating digits of pi (or even, possibly,
merely cycling through a large number of alternatives in the same order
every time)) would be sufficient for evolutionary processes to occur. I
emphasize randomness only to make it clear that the variations do not need
to be "designed" (and it is probably best that they aren't in the early going).
This archive was generated by hypermail 2b29 : Fri Sep 15 2000 - 00:43:58 EDT