[Taxacom] Predatory Open Access Publishers

Richard Zander Richard.Zander at mobot.org
Sat Sep 15 12:06:16 CDT 2012

Oooo...steamrollered. Okay, Kirk, Hempel rules in one sense, yet I think
I am also right.

What seems like equivocation is because we are talking past each other.
In the sense of a universal logic, well sure anything at all relates in
some way to everything else. Some few degrees of association with
Francis Bacon, say. 

But modern science uses a probabilistic basis, and does not see a
problem as a sorites that includes a huge number of variables that can
be worked out by a Mensa type. Although Francis Bacon may have something
to do with some systematics problem, we can discount the relevance until
we see some indication that we need to expand the set of evidence.

We agree, however, I think, in that the set of molecular evidence is way
too small to help solve a taxonomic problem, and I suggest that
morphological data helps do this. In Bayesian terms, total evidence is
reflected in the fact that if morphology says ((AB)C) at 99% confidence
and molecular evidence says ((AC)B) at 99% confidence, then the Bayes
formula says no hypothesis has better than 50% confidence. This needs
explanation, not rejection of the problem. Such as, one can redefine the
problem as evolution is molecular trait changes, period. This is as bad
as defining evolution as morphological trait changes, period. 


Richard H. Zander
Missouri Botanical Garden, PO Box 299, St. Louis, MO 63166-0299 USA Web
sites: http://www.mobot.org/plantscience/resbot/ and
Modern Evolutionary Systematics Web site:
UPS and FedExpr -  MBG, 4344 Shaw Blvd, St. Louis 63110 USA

-----Original Message-----
From: taxacom-bounces at mailman.nhm.ku.edu
[mailto:taxacom-bounces at mailman.nhm.ku.edu] On Behalf Of Kirk Fitzhugh
Sent: Friday, September 14, 2012 7:26 PM
To: taxacom at mailman.nhm.ku.edu
Subject: Re: [Taxacom] Predatory Open Access Publishers

Only because I prefer following principles that are developed well
beyond the limits of systematics thinking, such as evidential relevance
and the total evidence requirement:

"It might seem that... the observation that any particular scientific
investigation is aimed at solving a specified problem, and that the
initial selection of data should therefore be limited to facts that are
relevant to the problem. But this will not do, for the statement of a
problem does not generally determine what kinds of data are relevant to
its solution.... The notion of 'relevant' facts acquires a clear meaning
only when some specific answer to the problem has been suggested,
however tentatively, in the form of a hypothesis: an observed fact will
then be favorably or unfavorably relevant to the hypothesis according as
its occurrence is by implication affirmed or denied by the
hypothesis.... Generally, then, those data are relevant and need to be
gathered which can support or disconfirm the contemplated hypothesis and
which thus provide a basis for testing it." C.G. Hempel, 1966, 'Recent
problems of induction.'

In the scope of systematics inference, which is abductive, evidential
relevance is determined by whether or not effects in need of explanation
need to be accounted for by the same theory or set of theories. If the
same theory(ies) apply, then explaining each of the effects is relevant
to each other, for otherwise the opportunity for rationally determining
(abductive) support would be compromised. For instance, we don't take
each individual shared character and explain its occurrence to the
exclusion of all other shared characters. We know that at least some of
those other characters need to be explained in the same causal context
(e.g. 'speciation' or population splitting events), such that inferring
an explanation for one character will be relevant to the others. And
since shared characters form part of the premises of the inference, they
are part of the evidential support for the concluded hypothesis. Hence
the reason the requirement of total evidence matters, and why
partitioned analyses are irrational, and why cladograms derived from
partitioned analyses can't be compared.


More information about the Taxacom mailing list