revolution

James Lyons-Weiler weiler at ERS.UNR.EDU
Fri Oct 4 07:06:57 CDT 1996


On Fri, 4 Oct 1996, Alexey V. Kuprijanov wrote:

[snips]
>
> I agree. I attacked "cladists" for the following reason only: they dominate
> in the taxonomic community today (like communists in Russia yesterday).
> And the more recent careless papers are produced under the cladistic shell.
>

I agree about the more recent, careless papers, Alexey.  But part of the
carelessness comes from  overstatements by proponents of cladistic
parsimony.  Every method has limitations.  Why would anyone recommend a
method of inference while saying that it makes no assumptions, or has no
limitations?  Why should anyone else be convinced that this person fully
understands the method they recommend unless they can have considered its
weaknesses?  Such overstatements are rarer now, but they still pop up here
and there.  (As examples, I cite any and all authors who have written that
"parsimony will take care it".)

> I appreciate the role of cladists in the history of taxonomy. I really
> enjoyed Hennig's, Eldredge-&-Cracraft's, Nelson's, Wiley's books and numerous
> papers. They did very much to disturb the sleeping empire of the so-called
> evolutionary taxonomists. They did more than the numerical taxonomists had
> done before them. But the time of changes has gone. What is going on now, when
> they get power?
>
What's going on now is the development of statistical measures of
phylogenetic signal, for one.  In this probabilistic framework (without
assumptions of models of character evolution), everyone is equal; their
data may excellent or misleading, and they can find out.  Their data may
be improved upon by locating its weaknesses. Much current focus is on
phylogenetic accuracy, which requires the first step of asking when
methods (e.g., cladistic parsimony; maximum likelihood) might fail.
This is a step in the right direction towards stating their
limitations. Ergo, Huelsenbeck's and many other's work comparing different
methods with simulations.  The second step is to ask for the collection of
characters for YOUR group whether or not the data might trip up the
algorithms.

Of course, this approach has its limitations...


James Lyons-Weiler


_______________________________________________________________________________

  \  /   /    \  /           JAMES LYONS-WEILER           ______________
   \/   /      \/                                        |..............|
    \  /       /                                         |..............|
     \/       /              DOCTORAL PROGRAM IN         |..............|
      \      /               ECOLOGY, EVOLUTION, AND     |...***........|
       \    /                CONSERVATION BIOLOGY        |..*****.......|
        \  /                                             |.******.......|
         \/                  1000 VALLEY ROAD/186        |********......|
    ______________           THE UNIVERSITY OF            --------------
   | will perform |          NEVADA, RENO
   |  statistical |          RENO, NEVADA 89512-0013
   | phylogenetic |
   | analyses for |         "(Biology) is not religion; if it were, we'd
   |    food      |          have a much easier time raising money."
    --------------                       -Leon Lederman
_______________________________________________________________________________




More information about the Taxacom mailing list