Positivism in evolutionary science

[Tom DiBenedetto]

James Francis Lyons-Weiler wrote:

>>, Tom DiBenedetto wrote:

>> Parsimony is a crucial factor in scientific tests. The corroboration
>> of a hypothesis represents the parsimonious explanation for the
>> concordance between a prediction and an experimental result. It is
>> also a ubiquitous standard to make recourse to as few ad hoc
>> supplementary hypotheses as necessary to explain discrepencies
>> between predictions and results.

>       You're making the same mistake others have made in
>       the relationship between parsimony and corroboration.
>       In practice, the degree of corroboration afforded
>       to a hypothesis h increases when it has survived
>       a critical test T as the hypothesis (1) becomes more outlandlish.
>       For such hypotheses, the same type of test T represents
>       greater risk, so, when h survives the test, we are
>       surprised. The degree of corroboration affored to h
>       is linked to the degree of surprise.

I dont understand what you think my mistake is. I agree with the
above.
In the first part of my statement I was not addressing the question
of the degree of corroboration; I was addressing the basic concept of
corroboration; and saying that the acceptance of the hypothesis as a
valid explanation of the result (no matter how "strongly" that
supports the hypothesis) represents a parsimonious conclusion. One
could always dream up all sorts of reasons why the concordance
exists, other than
that the hypothesis might actually be true. Perhaps this could be
seen as a trivial point, but I make it to emphasize how parsimony is
fundamental to our way of thinking. The second part of my point (the
reference to minimizing ad hocs) is directed to the explanation of
discrepancies. Popper goes on at length about the desirability to
depart from ad hocness; it is directly related to the
"outlandishness" of the hypothesis (actually it might be better to
simply refer to the information content of the hypothesis). To relate
this back to phylogenetics, the most parsimonious tree is the
hypothesis which has the greatest information content; it can account
for the data more efficiently than any other hypothesis. A hypothesis
of independant origin for every taxon (creationism) can account for
the same amount of data, but it requires a separate (ad hoc)
hypothesis for each novelty, and is least parsimonious.

>       What Popper saw as the only link between the degree of
>       simplicity of a hypothesis and corroboration is often
>       confused, too.  He did not say that simpler hypotheses were,
>       automatically, better corroborated.  He said that they
>       were, in general, easier to test (i.e., better corroborable),
>       and that therefore we should examine the simple hypotheses
>       first.

No, you are missing something here. Simpler hypotheses are better
corroborated because they inherintly say more; they have higher
information content; they are inherently more bold. A homology
hypothesis which postualtes a single transformation, and thus implies
a grouping (of say, 8 taxa), is saying more than a hypothesis which
postulates three transformations and thus implies several smaller
groups, or a hypothesis which postulates many transformations and
implies nothing about relationships.




Tom DiBenedetto                 http://www-personal.umich.edu/~tdib/
Fish Division                                   tdib at umich.edu
University of Michigan Museum of Zoology