Positivism in evolutionary science

[James Francis Lyons-Weiler]

On Wed, 3 Dec 1997, Doug Yanega wrote:

> My two cents again...
>
> James Francis Lyons-Weiler wrote:
>
> >        In light of all these possible sources of noise, the central tenet
> >        of phylogenetics should not be that evolution is, in general
> >        or in detail, a parsimonious process.
>
> Maybe I'm missing something, but I've never heard anyone state that they
> thought evolution was parsimonious, or that this was why parsimony was used
> in phylogenetic inference. The idea is that we want as objective and simple
> an explanation as possible; if you condemn that, you're condemning the
> essential underlying tenet of science itself. Parsimony methods minimize
> the number of rejected hypotheses of homology in an analysis, with NO
> reference to process assumptions whatsoever. Evolution does not need to be
> parsimonious in order for parsimony methods to recover the overall pattern.
> If you start adding process assumptions and incorporating *models* of
> evolution, you've no longer got a simple hypothesis, but nested hypotheses
> (one that your model is accurate, the other that the final result is the
> simplest *within* the context of that model - ALL methods ultimately use
> the *criterion* of the most parsimonious hypothesis).

        I rather think that many modes of scientific inference
        exist that are independent of parsimony.  For example,
        the more complex theories of physics that are lent
        credibility owing to the outcome of experiments (tests).
        Parsimony summarizes, but does not test.
>
> > It is, as I indicated in
> >        the second post in this particular thread, positivistic to presume
> >        that parsimony will result in an accurate accounting of which
> >        characters correctly indicate homology that indicates shared
> >        relationships, rather than the alternative, namely, that
> >        parsimony may be mislead by "something else".
>
> Presuming that a tree is an ACCURATE hypothesis is not the same as
> presuming it is the SIMPLEST hypothesis. Right?

        If the tree that is the set of simplest hypotheses isn't
        accurate, then what's the point?  I don't mind that the MP
        tree(s) may be the set of simplest hypotheses - I'm interested
        in a accurate image when I view through the windows to
        the past - simple or not.

James Lyons-Weiler