Paraphyly=mistakes? (There's the rub)

[Richard Jensen]

I suggest that Herb's objection to "best" is similar to my own.  By what
criterion, that most systematists will accept, is any depiction of the
phylogenetic history of a group designated the "best."  It might be the most
parsimonious, or that supported by the greatest number of cliques, or that
representing a maximum likelihood solution.  But, all of these (and any others)
are contingent on so many assumptions about the data, evolutionary processes,
etc., that the best one can hope for is a solution that is consistent with the
data and these assumptions.  I would avoid using "best" and, instead, describe
the depiction as the most consistent with these data and assumptions, as
currently understood.  After all, we all know, that for virtually any large data
set, there may be hundreds, if not thousands, of equally parsimonious solutions.
We can choose a consensus of these (strict, majority rule, etc.) to "best
represent the overall results of the analysis," but that is the only sense in
which it might be the best.  Is that what you had in mind?

Thomas DiBenedetto wrote:

> -----Original Message-----
> From: HJJACOBSON at aol.com
> What cladist ever said that cladistic analysis produced the “best inference
> as to the true historical branching pattern.” Perhaps they said it was the
> most logically consistently inference, but never the most consistently
> logical inference of the “true” historical branching pattern.
>
> Herb,
>  Although I am sympathetic to the deep epistomological arguments about
> "truth" and our ability to ever confidently assert that we have detected it,
> I nonetheless have no problem proceeding on the basis of an assumption that
> there really really is a true singular history to taxic divergence. I
> further take as a basic assumption that the field of systematics is, or
> should be, dedicated to discovering. to the best of our ability, what this
> true history was, and to then represent it, both graphically and textually,
> to the rest of the scientific community and to the public at large. As a
> well developed rational enterprise, systematics has its own set of rules and
> practices that have developed over the years (as in all other sciences) to
> assure that we make consistent progress toward that goal. Thus we develop
> methods by which various hypotheses can be tested, and various inferences
> can be evaluated. I dont understand your objection to the use of the word
> "best" - clearly every science has methods for evaluating inferences, and
> consensus forms around those inferences deemed the "best". The fact that at
> any given point of time there might be disputes as to which inference is the
> best on any particular issue is hardly a legitimate objection - such is the
> necessary nature of scientific inquiry and argumentiation in all fields.
> So I repeat - I want the best inference (should I prefer the worst???) as to
> what really really happened in history (da truth!). I guess I really dont
> understand the nature of your objection.
>
> Tom diBenedetto

--
Richard J. Jensen              TEL: 219-284-4674
Department of Biology      FAX: 219-284-4716
Saint Mary's College         E-mail: rjensen at saintmarys.edu
Notre Dame, IN  46556     http://www.saintmarys.edu/~rjensen