tdib at OCEANCONSERVANCY.ORG
Fri Apr 12 13:06:12 CDT 2002
I've noticed confused over these terms.
Synapomorphies = Derived characters that unite two groups into a
monophyletic group after a phylogenic analysis.
Is there another term for derived characters that are shared between two
groups, before a phylogenic analysis?
Like using the term homology?
There is an interesting and confusing history to the semantics here. Your
proposal to use homology to refer to shared derived characters before
phylogenetic analysis would make things even more confused. You are correct
to point out that homology, from its original meaning, refers to _the same_
character in two groups. In cladistics, this has almost always been
interpreted to mean that homolgy must refer to characters that really really
are the same, historically - it is an ontological assertion that can only be
used when one is as certain as it is possible to be, that the character
really is the same. Before phylogenetic analysis therefore, one only has
proposed, hypothesized or putative homologies.
As coined and discussed by Hennig, synapomorphy clearly refers to characters
that are deemed to be homologies as the result of phylogenetic analysis.
Thus most cladists accept that homology and synapomorphy are effectivly
synonymous - the only difference being that synapomorphy refers to the
results of a defined operation, and homology is a term derived from basic
assuptions or theory. Thus I would slightly revise your comment:
Homology (truth) = Synapomorphy (hypothesis)
Homology (truth) = Synapomorphy (corroborated hypothesis)
The real confusion arose when some folks propsed using the term synapomorphy
to refer to character matches before phylogentic analysis. This proposal can
be found in Sober's book "Reconstructing the Past" and has been used by a
few cladists since then. Sober explicitly proposed to change the meaning of
synapomorphy so that we would have both an ontological statement of sameness
(homology) and an operational, or epistomologcal term - synapomorphy, such
that one could discuss phylogenetic analysis as a process by which
synapomorphies are tested, and if corroborated, be accepted as homologies. I
dont think this proposal has found much favor - most cladists simply would
say that pre-analysis character matches are putative synapomorphies or
putative homologies, and if they survive analysis, then they are
corroborated synapomorphies, or homologies.
Lets not even start on the absurdities that molecular systematists have
introduced with their use of the term "homology" as similarity. Grrrr
More information about the Taxacom