[Taxacom] Symposium: New Applications of Bayesian Inference in Phylogenetic Biology

[J. Kirk Fitzhugh]

Peter,

Bayesianism addresses changes in one's belief in a hypothesis based on the 
introduction of test evidence.  There are three roughly distinct activities 
that can be identified in the sciences: (1) conceiving of a hypothesis, (2) 
developing potential tests of a hypothesis, and (3) the actual testing of a 
hypothesis.  The inferential structures for each of these activities are 
abductive, deductive, and inductive, respectively.  It is in the act of 
testing a hypothesis that one determines whether or not predicted test 
evidence, inferred via deduction, is the case or not.  The inference drawn 
from such an action is inductive because it is on the basis of that 
observed test evidence that we make a conclusion regarding our belief in or 
acceptance of that hypothesis.  As no act of testing can guarantee the 
truth of a confirmed/corroborated hypothesis, the acceptance of that 
hypothesis is strictly inductive.  How we act in the future is contingent 
on that acceptance.  Similarly, disconfirmation/falsification leaves us 
with the conclusion that any number of other hypotheses must be 
entertained, which is itself an inductive conclusion.

So, you are correct that Bayesian reasoning formally spells out how (test) 
evidence (which could never be shared characters) is positively or 
negatively relevant to our belief in a hypothesis.  Changes in belief are 
matters of induction, which is why Bayes theorem cannot be used to 
initially infer the hypothesis for which the theorem is used to measure our 
change in belief in that hypothesis.

Best,
Kirk

At 09:35 AM 10/6/2006 +0200, you wrote:
>Kirk,
>
>Can you explain how Bayesian reasoning can be considered inductive?
>As far as my understanding goes, it formalizes (or attempts to) the way
>we informally assess the relevance of a particular piece of evidence to
>a hypothesis. But the hypothesis is not inferred by the theorem, in no
>...ductive way.
>
>Thanks,
>
>Peter Hovenkamp
>Nationaal Herbarium Nederland, Leiden branch
>www.nationaalherbarium.nl
> > -----Original Message-----
> > From: taxacom-bounces at mailman.nhm.ku.edu [mailto:taxacom-
> > bounces at mailman.nhm.ku.edu] On Behalf Of J. Kirk Fitzhugh
> > Sent: Wednesday, October 04, 2006 11:06 PM
> > To: taxacom at mailman.nhm.ku.edu
> > Subject: Re: [Taxacom] Symposium: New Applications of Bayesian
>Inference
> > in Phylogenetic Biology
> >
> > Nico,
> >
> > Could abstracts of the talks from the symposium be made available via
> > taxacom? It would be fascinating to see how Bayes theorem, which is a
> > strictly inductive procedure, will be justified as a means for
>inferring
> > explanatory (i.e. phylogenetic) hypotheses, which is strictly
>abductive.
> > The theorem is only capable of addressing changes in hypothesis belief
> > contingent on test evidence (which cannot be shared similarities), not
>the
> > inference of the very hypothesis to which the theorem is to be
>applied.
> >
> > Thanks,
> > Kirk
> >
> > At 04:39 PM 10/4/2006 -0400, you wrote:
> > >Announcing a Special Symposium! New Applications of Bayesian
> > >Inference in Phylogenetic Biology
> >
> > -----------------------------------------------------
> > J. Kirk Fitzhugh, Ph.D.
> > Curator of Polychaetes
> > Invertebrate Zoology Section
> > Research & Collections Branch
> > Los Angeles County Museum of Natural History
> > 900 Exposition Blvd
> > Los Angeles CA 90007
> >
> > Phone:   213-763-3233
> > FAX:       213-746-2999
> > e-mail:   kfitzhug at nhm.org
> > http://www.nhm.org/research/annelida/staff.html
> > http://www.nhm.org/research/annelida/index.html
> > ----------------------------------------------------
> > _______________________________________________
> > Taxacom mailing list
> > Taxacom at mailman.nhm.ku.edu
> > http://mailman.nhm.ku.edu/mailman/listinfo/taxacom

-----------------------------------------------------
J. Kirk Fitzhugh, Ph.D.
Curator of Polychaetes
Invertebrate Zoology Section
Research & Collections Branch
Los Angeles County Museum of Natural History
900 Exposition Blvd
Los Angeles CA 90007

Phone:   213-763-3233
FAX:       213-746-2999
e-mail:   kfitzhug at nhm.org
http://www.nhm.org/research/annelida/staff.html
http://www.nhm.org/research/annelida/index.html
----------------------------------------------------