From: Jim Ledbury (jim.ledbury@dsl.pipex.com)
Date: Sun Oct 17 2004 - 00:58:10 BST
Hi Platt, Mark, Scott ...
the context of random in terms of evolution means that mutations are
supposed to occur without purpose. This is because they are imagined to
arise through copying errors in cellular replication or through
environmental factors (radiation, chemical mutagens...) hence they occur
without any regard to consequence. Whether there are other causes of
mutation a different matter, but given that such purposeless mutations
*do* occur it is currently considered as sufficient by biologists to
drive the variations required for evolution by natural selection: the
deciding factor is that they are survivable and are advantageous. As I
have said elsewhere, given the tangles that evolution does get into, I
have little reason to believe that there is anything other than an
essentially purposeless, haphazard variation involved. This fits the
dictionary definition of random perfectly.
Randomness as used as a modelling tool in statistics is a different
matter. It simply means that we are not interested per se in the causes
for variations in initial behaviour, we simply assume that there is a
certain distribution of tendencies and run with it. Whether or not we
can attribute an underlying cause for the distribution is generally
immaterial. The turn of a dice or the fall of the balls in the UK
lottery is assumed to involve sufficient mixing of initial conditions
that an even distribution of outcomes is an article of faith, although
it can be backed up (or refuted) by statistical analysis.
In physics, randomness is asserted at the quantum level. Some
interpretations of quantum mechanics state that the outcome of a given
experiment is actually globally determined. It's just we can never
access the global information necessary to determine the outcome of a
given experiment, so we are forced to assume a random albeit weighted
outcome. Other interpretations say the randomness is intrinsic, so we
end up doing the same calculations anyway. Here I would add that
'consciousness' obviously does have an influence on QM randomness as I
end up typing the keys I want to (usually).
'Random' number generators used by computers are actually deterministic
algorithms, frequently involving modulo arithmetic, and are correctly
termed pseudo-random. They just give a decent distribution for
modelling say dice throws or card orders in games but not for instance
in deriving the keys for encryption where decent analysis will uncover
the fact that the 'random number' generator is not actually that
random. There are web-sites devoted to providing a better set of random
numbers than are produced by the bog standard random number generators
available to most computers (http://www.random.org/). The best
description of a random distribution is rather subtle and requires that
there is no simpler description of the distribution than the set of
random numbers themselves (as opposed to a generating function). This
has no bearing on the assertion that evolution is random.
Randomness in chaos is down to the existence of 'strong mixing' in
strange attractors. This is similar to random number generators
discussed in the last paragraph but all it means here is that
arbitrarily small differences in initial conditions ultimately lead to
vastly different end points (the so-called butterfly effect).
Mathematically this is modelled by completely deterministic formulae,
but is generally taken to completely independent of quantum mechanical
randomness (although a different formulation of QM could I guess make
use of strong mixing) and in any case has absolutely no bearing on what
is considered to be the random nature of mutations as the source of
evolution. Even if the world is wholly deterministic, the cause of
genetic mutation is completely without regard to the consequence of the
mutations so can safely be presumed to be haphazard, purposeless and
hence 'random' - unless you want to get into universe spanning
manipulators of fate, which personally I don't. I don't believe in the
synchronicity espoused by Jung to explain the utility of the I Ching
either.
I accept the possibility of purposeful manipulation of the genetic code
on a cellular level as determined by the cells themselves
(neo-Lamarkism): I just think it would inevitably be incredibly
shortsighted therefore not much better than randomness if at all, but
given the actual existence of random, purposeless, haphazard mutations
as opposed to a hypothetical puroposeful adjustment, I am comfortable
with this as an explanation for evolution.
In short, I'll accept the assertion of "random mutation acted on by
natural selection" as a reasonable working hypothesis, and I don't think
I'm being dogmatic about it. :-)
Regards,
Jim.
MOQ.ORG - http://www.moq.org
Mail Archives:
Aug '98 - Oct '02 - http://alt.venus.co.uk/hypermail/moq_discuss/
Nov '02 Onward - http://www.venus.co.uk/hypermail/moq_discuss/summary.html
MD Queries - horse@darkstar.uk.net
To unsubscribe from moq_discuss follow the instructions at:
http://www.moq.org/md/subscribe.html
This archive was generated by hypermail 2.1.5 : Sun Oct 17 2004 - 01:12:44 BST