[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Dinosaurimony (Re: Definitions...)



>In a message dated 98-08-10 16:58:55 EDT, znc14@TTACS.TTU.EDU writes:
>
><< "The simplest explanantion which accounts for the most data
> with the least number of unsubstantiated hypotheses and unsupportable
> assumptions is the most likely to be correct." >>
>
>Unfortunately, accounting for the most data, having the least number of
>unsubstantiated hypotheses and unsupportable assumptions, and being the
>simplest explanation are mutually independent, so a parsimonius hypothesis may
>be impossible to formulate.


"Parsimonious" is always used in a relative sense.  One rarely obtains a CI
of 1.0 with a comprehensive data set, but no one is saying "Hey - my
hypothesis is PARSIMONIOUS."  Rather, the point is that one hypothesis is
more parsimonious than another.




>
><<The use of parsimony in cladistics does not
>invoke a particular characteristic of the process of evolution (or of
>natural processes in general), it merely assumes that logical evaluation of
>evidence can lead you to the true answer.>>
>
>The use of parsimony in cladistics is simply one way to convert a character
>matrix into a cladogram. Nothing more or less. Here is a character matrix,
>here is the cladogram we get if we use parsimony analysis on the character
>matrix. If you want to interpret the cladogram as some kind of phylogeny, or
>want to use it to classify organisms, go right ahead--but beware.


Actually, this is only one of several approaches to parsimony in
phylogenetics.  The "pattern cladistic" school regards parsimony as a first
principle - the tree(s) one obtains is/are the best one(s) possible for the
data set, and one is free to interpret it/them as a phylogeny, but a priori
assumptions about evolutionary process are avoided.

The stripped-down version is what George stated - the cladist plugs and
runs, and that's that.  Of course, if it were that simple, it would have
taken far less than 5+ years to finish my dissertation.  As for "beware" -
THAT'S THE WHOLE POINT OF PUBLISHING THE MATRIX!  Of course cladistics is
not purely objective, and it never will be as long as bipedal primates are
in charge of it.  But it is explicit, at least when presented properly.  (I
know, I know - matrices are not always published.  Weekends aren't always
sunny, either - it isn't a perfect world, and I've never been an editor,
else any paper crossing my desk dealing with phylogenetics would either
have a matrix, refer to a matrix already published, or be swiftly
rejected.)  But as long as a data set is available, one can see *precisely*
where caution is needed.  I just finished writing up a discussion of one of
my phylogenetic analyses, and I could see *exactly* where the tree is
robust and where it's liable to fall apart under future study.  Without
such explicitness, one must beware of the entire tree, from start to
finish.

Another approach views phylogeny estimation as a statistical problem.  The
sharp increase in maximum likelihood methods over the past several years is
part of this trend.  In this school, parsimony is absolutely viewed as a
model of evolution - one of hundreds of possible models, in fact.  The
primary limitation for paleontologists is that very few models for
morphological evolution exist - there are several good molecular models
available, and with modern computing methods one can apply several.  But I
am unaware of a good morphological evolutionary model that can be applied a
priori to a search algorithm.  Several workers (Huelsenbeck, Wagner) are
working on stratigraphy-based models, and though I am EXTREMELY skeptical
of these, it will be interesting to see what comes out.


chris



-=--=--=--=--=--=--=--=--=--=
Christopher Brochu

Postdoctoral Research Scientist
Department of Geology
Field Museum of Natural History
Lake Shore Drive at Roosevelt Road
Chicago, IL  60605  USA

phone:  312-922-9410, ext. 469
fax:  312-922-9566

cbrochu@fmppr.fmnh.org