Human and ape phylogeny

[P.Hovenkamp]

At 11:56 AM 4/5/03 +0200, René Zaragüeta Bagils wrote:
(...)
>(Characters are considered binary, to simplify the discussion).
(...)

>This is not the point. Cladistic characters are classificatory concepts that
>partition individuals into instances of the concept an non-instances of the
>concept. Absence cannot be an instance of a concept. So grouping by
>non-instances of concepts is logically inconsistent. "Reversals" can only be
>a posteriori interpretations of the presence of several non-instances of
>concepts in a clade. Reversals are not homoplasies (refuted conjectures of
>instances of classificatory concepts) but non-instances of concepts (or
>absences, or plesiomorphies). So it is logically inconsistent to optimize
>"reversals" because they cannot be "apomorphic", by definition (except in
>phenetics...)

This is indeed the point - and it goes back to Patterson, who (thought he) 
could distinguish "taxic" and "transformational" homology. In a 
non-evolutionary world, taxic homology (the presence of a homologous 
structure in a number of taxa) is indeed possible - in an evolutionary 
world, it is not. In an evolutionary world (and models do not enter, just 
evolution) homology describes a change from one state to another, and a 
full specification of a homology statement *needs* two states. There are no 
"concepts" that arise de novo - each character has had a state from which 
it developed.

>However, how can absences evolve (=transform, I guess you should say) in
>your model? What
>do you mean by the evolution of the "lack of a tail"? How can, for instance,
>natural selection work on the absence of something? How can you distinguish
>between the 'transformation' of the absence of a tail to the presence of a
>tail and the 'transformation' of the absence of a leg to the presence of a
>tail?
Exactly. That may be difficult (at least, in more realistic examples it may 
be). And it's called "character analysis". But just because it's difficult 
does not mean we have to give up.
Three taxon analysis gives up on character analysis even before it's 
becoming difficult. Discarding evolution by the way.

Peter Hovenkamp



>Pierre Deleporte wrote:
>"Phenetics did not care with evolution either, it was "overall similarity
>for overall similarity's sake", just so. Like 3TA is 3TA for 3TA's sake.
>Just so."
>Re:
>This is exactly the point concerning standard parsimony. If you make a
>conjecture of
>homology based on the presence of a feature (as everybody does), then why a
>computer program
>tells you that the absence characterizes a group if you didn't make the
>conjecture? Cladistics is supposed to test your conjectures, not computers'
>ones (unless PCs and Macs are delivered with evolutionnary models installed
>:-)).
>Optimizing instances and non-instances of characters at the same time is
>simply phenetic (just so). Organisms grouped together because they lack
>something live in a phenetic world (or a pre-evolutionary one...).
>
>A final question to the so-called "strict cladists" and to Pierre: Has
>anybody ever been criticised because his character definitions where not
>consistent with a particular theory of evolution or model? It has never
>happened to me, and I have never seen this kind of criticisms. People
>criticize my characters on topological grounds or on quality of
>descriptions, even on taxonomic sampling grounds, etc. I work on teleosts:
>can anybody tell me what cannot be
>coded because it doesn't evolve? Or everything, however described, evolves?
>As Jorge Wagensberg* has put it (modified): If evolution is always the
>answer,
>what is the question?
>
>* Wagensberg, J., 2002: Si la naturaleza es la respuesta, cual era la
>pregunta? Metatemas-Tusquets, Barcelona, Catalonia, 126pp.
>
>René
>
>René Zaragüeta-Bagils
>Département Histoire de la Terre UMR 8569
>Equipe SICC (Systématique, Informatique, Cladistique et Chronologie)
>Muséum national d'Histoire naturelle
>8 rue Buffon
>75005 Paris - France
>rzb at mnhn.fr