Fwd: Re: More on the 'cladistics' of sequences

[pierre deleporte]

>A 08:56 15/06/2004 -0400, John Grehan wrote :
>>The definition of phenetics posted by Tindall (below) conforms to my
>>understanding and use of the term phenetics I posted on TAXACOM even
>>though such usage was problematic for other contributors.

If things are such as you state, you should thus have stopped talking of 
"phenetic characters" or "phenetic sequences", and talk instead of possible 
phenetic analysis of the data set (overall similarity criterion for 
comparing terminal taxa, not characters) versus possible cladistic analysis 
of the very same data set. This is why your usage is problematic.

Otherwise, making use of all available (and relevant) data is not 
particularly cladistic or phenetic, it is a logical requirement in 
scientific explanation (don't discard evidence).

A possible distinction in the data sets could be that some pheneticians 
don't care if some characters are not inheritable, while phylogeneticians 
(cladists or otherwise) should logically use inheritable characters only 
(plausible markers of common descent).
But you were always talking of polymorphism in character states in the 
outgroups (the discussion revealed that tis was the very meaning of 
"phenetic character" in your personal acception), inducing possible 
ambiguity in polarization and consecutive rooting, hence it doesn't seem 
that inheritability was your point when talking of "phenetic characters".

And it has been largely argued that rooting and optimizing a topology are 
independent operations. If you don't believe this, try it yourself. It's 
not a matter of "phenetic" or "cladistic" opinion, it's pure logics in 
basic graph theory.

Pierre


Pierre Deleporte
CNRS UMR 6552 - Station Biologique de Paimpont
F-35380 Paimpont   FRANCE
Téléphone : 02 99 61 81 66
Télécopie : 02 99 61 81 88