Inadequacy of cladistics
A.P. Jason de Koning
apjdk at ALBANY.EDU
Fri Jan 23 15:24:58 CST 2004
I think the discrepancies in what everyone actually considers to be "the" cladistic methodology is interesting. For my part as a statistical molecular evolutionist / phylogeneticist, I am accustomed to referring to all methods of phylogenetic inference which involve functions of inferred character state changes as "cladistic" (i.e., maximum likelihood, bayesian likelihood, and parsimony). Whereas maximum parsimony is a logical generalization of the classical cladistic approach, likelihood modelling with Markov models is a logical generalization of parsimony. Indeed, maximum parsimony can be formally shown to be a special case of a full likelihood treatment using equal rates, equal equilibrium frequencies of states, and no branch lengths...
So with respect to your comments, Richard: 1) I disagree that 'cladistics' is a clustering approach (since methodologically, "clustering" usually implies a phenetic approach); and 2) I argue that the strict cladistic approach does indeed seek to reconstruct a process--it just does so more badly than other approaches which have generalized and improved upon it..
A.P. Jason de Koning, Doctoral Candidate Email: apjdk at albany.edu
Department of Biological Sciences Lab: (518) 442-4347
University at Albany, SUNY FAX: (518) 442-4767
1400 Washington Ave., Albany NY 12222, USA Mobil: (518) 210-4504
"Not ignorance, but ignorance of ignorance is the death of knowledge." - Whitehead
----- Original Message -----
From: <Richard.Zander at MOBOT.ORG>
To: <TAXACOM at LISTSERV.NHM.KU.EDU>
Sent: Friday, January 23, 2004 1:37 PM
Subject: Re: Inadequacy of cladistics
> I think I agree. Cladistics is essentially a cluster analysis that groups
> taxa. The theory is simplistic, and details of process are ignored. Of
> course, we don't know details of process for most taxa, so as long as
> cladists state we are using a simple model to cluster OTUs, and are not
> reconstructing a process, that's okay. The problem, is, as you say, that
> there is a confusion between such.
> In molecular terms, finding similarities between introns shows
> relationships, finding dissimilarities between exons shows process. I think.
> Richard H. Zander
> Bryology Group
> Missouri Botanical Garden
> PO Box 299
> St. Louis, MO 63166-0299
> richard.zander at mobot.org <mailto:richard.zander at mobot.org>
> Voice: 314-577-5180
> Fax: 314-577-9595
> Bryophyte Volumes of Flora of North America:
> Res Botanica:
> -----Original Message-----
> From: Thomas G. Lammers [mailto:lammers at vaxa.cis.uwosh.edu]
> Sent: Friday, January 23, 2004 12:07 PM
> To: Richard.Zander at MOBOT.ORG; TAXACOM at LISTSERV.NHM.KU.EDU
> Subject: Re: Inadequacy of cladistics
> At 11:55 AM 1/23/04, Richard.Zander at MOBOT.ORG wrote:
> >Excellent thought. What is this fundamental inadequacy?
> Your points are valid, but my criticism has always been the poor
> correlation between cladistic methodology and biological reality.
> Evolution is NOT perpetually dichotomous. In plants at least, there is a
> great deal of reticulation. Paraphyly is real; a parental population does
> not speciate when it gives rise to a daughter species. Etc.
> Since the early years of the 20th century, we have learned a great deal
> about the complex array of diverse biological phenomena by which species
> come into existence. With cladistics, we have chucked all that out the
> window and chosen to represent it by a gross oversimplification of the
> process. Those pretty stick figures are utterly detached from the complex
> reality they purport to represent.
> Cladograms may be objective, testable, and repeatable, but if the procedure
> by which they are created IGNORES underlying biological reality, they are
> (pardon the horribly mixed metaphor) a house of cards with feet of clay,
> built on sand.
More information about the Taxacom