science = determinism?
Bill Todd
billtodd at foo.mv.com
Sun Nov 28 02:24:24 EST 1999
<amolitor-at at visi-dot-com.com> wrote in message
news:JLX%3.355$Sz5.21604 at ptah.visi.com...
> In article <81pden$h80$1 at pyrite.mv.net>, Bill Todd <billtodd at foo.mv.com>
wrote:
> >Mea culpa (see previous amending post) to the several people who
responded.
> >I'm still less than convinced that 'non-deterministic' and 'random' are
> >synonymous, since the word to me implies a degree of unpredictability
beyond
> >mere uncertainty - e.g., a truly 'random' event to me is one whose
> >probability cannot be evaluated.
>
> This doesn't even make sense except, perhaps, in the case of
> unique events. If you can imagine an event occuring repeatedly, then
> the probability of the various outcomes of the event is a meaningful
> quantity. You may not be able to evaluate it, since you can't carry
> out an infinite number of experiments, but it's meaningful.
Well, yuh - I think that's what I said, or at least what I was attempting to
convey. 'Random' to me carries the connotation that causality - and even
statistical predictability in the absence of understood causality - cannot
be perceived by the observer who so characterizes an event, or, in other
words, that the event appears completely haphazard.
I understand, for example, that something like a hypothetical ideal 'random
number generator' would operate in a statistically-predictable manner in the
aggregate, but the value of each individual number generated might still
satisfy my preferred definition of randomness (since its probability of
occurrence would not be evaluable without at least one piece of external
information - the fact that it was being generated in a 'random' manner -
and for that matter usually within some bounded range at some non-continuous
granularity). I'm not sure that a group of such numbers would, however, if
it was sufficiently large to have a statistically-significant 'random
pattern' to it, and I'd be even more leery of a group that matched something
like a bell curve (since there would obviously be some underlying principle
begging to be understood).
Are you really stating that, at least in mathematics and physics, 'random'
and 'not completely deterministic' are exact synonyms? If so, I don't
believe that this captures the entire sense of the English definition, and
my suspicion is that the English definition may have been what the original
post had in mind - which if read in that light might affect some of the
subsequent responses.
>
> I'd suggest some elementary text on probability and statistics,
> these concepts are quite accessible.
>
> No offense, but you're coming across like a slightly more
> coherent than usual Usenet Loon.
None taken: I don't worry excessively about what people may think of me,
and would certainly not give it higher priority than the opportunity to
improve my ability to communicate.
- bill
More information about the Neur-sci
mailing list