[Re:] reweighting characters, few and many

Thu Oct 28 14:05:49 CDT 1999

It is no wonder and no coincidence that Karl Popper had an evolutionist,
darwinist world view. Charles Darwin was one of the most influential and
important intelectual authorities for Popper.

Popper got to the point of applying Darwin´s theory of selection to the working
of his own mind.
That is Popper was astonished how often he had to write, delete and re-write a
certain line of argument, before it got ready.
This selection process of intelectual things is called learning by "trial and
error".
One of Popper´s colleagues, Bertrand Russel, however, always wrote his
arguments down in a nearly perfect manner, without much deleting and re-writing.
Popper stated this in one of his essays.

          Thomas

On (         Thu, 28 Oct 1999 16:51:02 +0200
),         Pierre Deleporte <Pierre.Deleporte at UNIV-RENNES1.FR> wrote:

>Dear all,
>
>
>first, a comment on Thomas Schlemmermeyer's message (maybe it was largely
>a joke, but I just play it serious a little while)
>
>
>asking for suggestions, Thomas wrote :
>
>
>>In PAUP 4.0, I ran a matrix
>
>
><underline>suggestion</underline>:  expose your evolutionary assumptions
>justifying the use this parsimony program for phylogeny reconstruction
>
>
>>which gave 423 most parsimonious trees. As the
>
>>consensus tree was highly unresolved,
>
>
><underline>s</underline>:  ... you could very well have stopped here, and
>went and look after more characters and more taxa, or try to improve
>primary homology assessments...
>
>but I still follow you anyway
>
>
>>I ran the command "reweight"
>
>
><underline>s</underline>:  justify reweighting
>
>
>>which gave,
>
>>after a new search, 21 parsimonious trees, and a nice consensus tree,
>but one
>
>>that I personally dislike. So,
>
>
><underline>s</underline>:  there is no shame in disliking (but have you
>any biological justification for rejecting?)
>
>
>>in the next step I widened up slightly, but only slightly, the upper
>bound
>
>>of the trees to be kept and searched the unweighted characters again.
>
>>Excitingly, this enlargened the tree file from 423 trees to
>approximately >25000 trees. Now, I reweighted those 25000 trees, and run
>parsimony search >again.
>
>>I got only three trees, and Imagine!!!, the consensus tree corresponds
>exactly
>
>>to the tree, which is my person favourite!!!
>
>>
>
>>  Any suggestions?   Thomas
>
>
>
>Well, you may convince some colleagues to accept the two consensus trees
>as acceptable alternatives (if the results are very "close to one
>another"), but surely not to prefer the second one... unless you provide
>convincing external justification, but presumably your arguments are
>already "coded in the data matrix" under the form of (possibly weighted)
>characters and polarisation rules.
>
>Perhaps you could look after some possible external arguable reasons
>supporting your "preference", but, in itself, this personal preference is
>not an  argument, and not even a problem for biology.
>
>
>
>Further in this thread, Richard Zander wrote:
>
>
>>What would Karl Popper say relevant to this exercise? Probably that
>
>>phylogenetic estimation is just historicism in reverse.
>
>>(...)
>
>>I guess he hadn't heard of cladistics yet.
>
>>Check out: Popper, K. R. 1957. The Poverty of Historicism.
>
>>Harper & Row, N.Y.
>
>
>
>It happens that I am presently reading this book (French translation,
>sorry for possible errors), and, to my disappointment, as far as I
>understand it it does not bear directly on the subject.
>
>
>Popper condemned the possibility for <underline>social
>sciences</underline> to scientifically make <underline>long-term
>predictions</underline> about the future evolution of human societies
>(and for physics to predict the long-term future of real, complex
>situations as well). But he said nothing in this book about
>retrodictions.
>
>His notion of "historicism in social sciences" should not be confused
>with other "historical approaches" (reconstructions of the past), and
>surely not with phylogeny reconstruction. The notion of "historicism in
>reverse" is apparently Richard Zanders creation and should be defined.
>
>
>More than this, Popper explicitly <underline>accepted</underline> the
>notion of general tendencies in the evolution of organisms, namely the
>tendency toward an increasing number and an increasing variety of
>biological forms. He called them: <underline>explained</underline>
>tendencies, which may be the basis for some general predictions given
>that the conditions ("initial conditions") for the expression of this
>tendency do not vary too much.
>
>Well, with the notions of multiplication and diversification, we are not
>too far from "descent with modification", isn't it? And if this explained
>tendency  is good for prediction, I think it may be useful for
>retrodiction too...?
>
>
>Of course we must assume some "general law of evolution" (however
>careful, e.g. descent with modification with no imposed tempo) to
>reconstruct the phylogeny, but... the whole story already happenned, was
>unique, and left some traces behind it, thus we are NOT facing the
>problem of guessing which of multiple ways evolution will follow in the
>future, we are instead facing the more accessible problem of "reading"
>the traces of a unique past, given the knowledge of explained
>tendencies.
>
>
>In conclusion, please let's not feign confusing phylogenetics with
>Popper's "historicism", and Sherlock Holmes with Nostradamus.
>
>
>
>sincerely,
>
>Pierre
>
>
>
>
>
>Pierre Deleporte
>
>CNRS UMR 6552 - Station Biologique de Paimpont
>
>F-35380 Paimpont   FRANCE
>
>Téléphone : 02 99 61 81 66
>
>Télécopie : 02 99 61 81 88
>

------------------------------------------------------------------------------
Thomas Schlemmermeyer
Museu de Zoologia, Universidade de São Paulo
Caixa Postal 42694
CEP 04299-970
São Paulo, SP, Brasil

Residência:
Thomas Schlemmermeyer
Caixa Postal 00276
CEP 14001-970
Ribeirão Preto, SP, Brasil

Fone, Fax: 016 6371999
------------------------------------------------------------------------------