http://groups.google.com/group/alt.philosophy/browse_frm/thread/3ad8e23dc2d07d34#
Is natural selection undeniable?"
The question isn't valid. It certainly isn't testable. Here's why.
We define only an existing life-form as being selected. We can't then
turn around and say that nature selected that form, for nature might
have had some other form lined up. We could, in fact, point to a
non-existing life-form and say that because it doesn't exist then
natural selection doesn't work.
What started out looking like a question that could be empirically
ascertained as being true or "undeniable", actually turns out to be a
species of tautology. In this case, synonyms (natural selection, and
life-forms that exist) are erroneously regarded as different events
in
relationship.
AN EXAMPLE to CLARIFY
For example. We could say that "it is undeniable that explosions
cause
things to fly apart". Here the tautology is to regard the synonyms
"explosion" and "fly apart" as two empirically real events in
relationship, when in fact there is only one event.
Similarly, "natural selection" and "existing life-form" aren't two
empirically real events in undeniable relationship. There is no
empirical relationship at all, as there is only one event.