Positivism in evolutionary science

[James Francis Lyons-Weiler]

On Wed, 3 Dec 1997, Tom DiBenedetto wrote:

> Who makes such claims? Advocates of parsimony merely claim that real,
> observed patterns do exist (I think you agree with that), and that
> descent is a suitable explanation for organism-wide consistent
> patterns in the distribution of heritable characters amongst taxa.
> Whether that pattern is formed by "most of the traits" aquired
> through descent does not seem to really be an issue, as far as I see.

        Really?  On what grounds?  Is there something intrinsic
        in the vast possibilities of past and present evolutionary
        processes that would guarantee that most of the evidence
        left by those processes is signal, and not noise (wrt
        shared geneaological relationship)?
>
> >       it's a matter of signal:noise, and to focus only on what we
> >       perceive as signal without (adequately) measuring the
> >       noise is positivistic.
>
> It seems to me that parsimony methods are quite efficient at
> distinguishing the hierarchical signal from the incongruent noise.

        I disagree.  I think they are efficient at portraying
        signal when signal is there and noise is rare and
        evenly distributed, and they are efficient at summarizing
        nonsense when the preponderance of "pattern" they tap into
        does not reflect evolutionary relationships.  Parsimony
        methods can do no more than provide an overall picture of
        the APPARENT signal in the distribution of states among
        organisms.  Occasionally, under a restricted set of
        conditions, MP will summarize true evolutionary
        signal - and sometimes they will not.
>
> >         Relying on (parsimony) trees
> >       to tell us about the reliability of the information
> >       in the distribution of character states among organisms
> >       is like trying to receive instructions on how to construct
> >       an antenna through the antenna you're trying to construct.
>
> The parsimony tree merely reveals the hierarchical order in the
> dataset. The question of the reliability of that data is another
> matter. If you claim that real patterns can arise coinidentally; i.e.
> that a real pattern might be in fact a result of more than one causal
> factor working to produce an artifical pattern, then I guess I would
> like to know what such factors are, and how that all works.

        When homoplasy is localized, the hierarchy in the matrix
        is distorted.  When homoplasy is not rare, the hierarchy in
        the matrix is displaced noise - which with varying states
        MUST assume (by definition) some type of "pattern".  To say
        that the pattern in the matrix is reflective of evolutionary
        relationships is a MAJOR assumption.  The reason why there
        will always be some finite degree of "pattern" in discrete
        matrices with varying states is that the matrix is finite;
        the state must be distributed this way or that.  I have
        posted in the past a number of sources of noise for
        phylogenetic inferences - I can post them again if you like.
        When the pattern or implied hierarchy in a matrix has not
        been caused by informative shared geneaological processes,
        parsimony will go right on and summarize the major emergent
        pattern.  I have focused on independent means to determine
        when and if the patterning in the distribution of states
        among organisms looks like it (1) indicates a significant
        degree of hierarchy , and if (2) the amount and type of
        implied hierarchy is the type that does not mislead
        algorithms that lead to trees.
>
> >       Some parts of the information may come through, but other
> >       parts are jumbled.  It's even worse than this analogy, because
> >       at least with radio or television signals we can immediately
> >       recognize the noise (scatter).  It would be as if the antenna
> >       would organise the noise into coherent sentences about the
> >       contruction of the antenna that were misleading - and we
> >       would have no independent means of checking the quality of
> >       the reception, because the antenna is making "sense" out
> >       of "nonsense".
>
> This seems to imply that distributions can be patterned,
> organism-wide, by a multiplicity of factors all working to produce a
> single coherent pattern. How can that be?

        There need not be a single, coherent pattern.  Parsimony
        will reflect the preponderant or major pattern - and
        it is hit or miss as to whether that pattern is, in fact,
        predominantely reflective of shared geneaological relationships,
        or something else.  That something else includes a mix of
        signal caused by shared genealogical, noise caused by common
        genealogy (organisms inherit homoplasy for deeper relationships
        from their progenitors), high rates of anagenetic evolution
        between speciation events, non-binary fission of species, non-
        independent anagenesis among characters in the same lineage,
        inheritance of ancestral polymorphism, concerted evolution (among
        lineages), high frequencies of convergence (adaptive or random)
        relative to the frequency of non-convergence, unequal distribution of
        homoplasy among lineages, and poor outgroup choice.

        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.  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".

        james lyons-weiler