termites at USP.BR
Mon Jan 25 22:08:09 CST 1999
In response to the rather philosophical points made by Tom:
On ( Mon, 25 Jan 1999 14:35:45 -0500), Tom DiBenedetto <tdib at umich.edu> wrote:
>Yes, we have a historical science. But that does not mean that we do not
>develop testable hypotheses. We develop testable hypotheses of homology
>between character states, and test those hypotheses for congruence against
This may well be true. But, according to my knowledge of the philosophy
of science, the best sort of hypotheses are those that can be tested
That is still OK! If you have a number of hypotheses about homology, parsimony
may tell you which of them can be actually synapomorphies and which are better
seen as homoplasies. The quantitative criterium would be given, in this
procedure, by the number of steps needed for the tree.
But: A single, isolated hypothesis of homology, seen for itself, is not a
strictly quantitative hypothesis. It is a vague construct based on Remane's
criteria or whatever else.
But that is all OK.
>I think there is more to it than that. Simplicity is inherint in the scientific
>approach. If one does not attempt to formulate the simplest explanation, then
>one might as well rest content with a separate explanation for each
>phenomenon (the most complex explanation); i.e. no science.
I would not agree with that. Especially biologists should recognize that their
science is of quite idiosyncratic nature. Why were the dinosauros extinguished?
Why are there no leaf-cutter ants in Africa? and so on.
The diversity of life is so amazing and overwhelming that often simplicity
makes place for complexity.
In the beginning of a scientific paradigm there often is an elegant metaphor,
a simple phrase which enlightens and entrances its followers. Having past by
some decades, one discovers that maybe the anterior model or a third not even
thought out paradigm may find some use as well.
Another cause which may lead to some conflict between simpler and more complex
models is the level of organization one looks at.
Simple parsimony models were developped by systematists, the level of
organization would be, in this case, the whole organism. That all goes back to
Aristotle and his book about the parts of animals.
Evolution, ultimately, should be understood however, at the level of
populations and gene frequencies. This may be a field where one feels nearer,
in terms of thinking, to Felsenstein.
In the ideal case, in such conflicts between levels, the models of the
simpler level and the more complex level should not contradict with each other.
And I guess, that in the special case of morphological systematics, we
have to expect quite a lot from the Morphologists, as they work with large
Morphological structures are complex and diversified, so, in many cases,
their evolution may be easily tracked down when compared with poor DNA-sequences
where one position only allows evolution into 3 different states.
But one should not stir up unnecessarily conflicts between the geneticists and
the morphologists. I think unification is possible and will certainly reconfirm
our humble wit, that it is all hypothetical what we are working with, being
some things, however, more speculative and other things, on the other hand,
The art of being a biologist maybe is to find the adequate method for the
>of ever discovering historical relationships. We do assume that there is
>historical "signal" in character distributions.
I think the pattern cladists turned this somewhat around:
The fact that there is a signal in our character distributions, allows us to
look at our systematic knowledge within a historical framework.
>The homology we hypothesize is prior to evolutionary theory (as is
However, most people date beginning of evolutionary theory back to 1859
(Darwin's theory of natural selection).
If you set as starting point of evolutionary theory, however, Aristotle's
theory of common descent, however implicit this might be imbedded in his
work, starting point of evolutionary theory would be very much earlier.
But, one should mention as well Mendel's work about pea genetics. Only Mendels
work allowed the integration of classical naturalists' views like this of
Darwin and experimental genetical work. Seen that Mendels work has long
been neglected, starting point of evolutionary theory may thus be only about
1920, quite a long time after they started to homologize.
> Evolutionary theory must explain our
>discoveries of homology, they are not the source of it.
Yes and No. This is a very complicated problem I think.
Museu de Zoologia, Universidade de Sco Paulo
Caixa Postal 42694
Sco Paulo, SP, Brasil
Caixa Postal 00276
Ribeirco Preto, SP, Brasil
Fone, Fax: 016 6371999
More information about the Taxacom