Cladistic analysis, Gregory Paul...
kinman at HOTMAIL.COM
Mon Sep 16 18:25:21 CDT 2002
First let me clarify that the "silly notion" Gregory Paul was
referring to is that held by Feduccia (and a few other workers) that birds
did not evolve from dinosaurs. I have never taken that "notion" very
seriously, and I continue to show Class Aves evolving from saurischian
dinosaurs (Order Saurischiformes, Class Reptilia). Whether we should say
"birds ARE dinosaurs" instead of "birds are dinosaur descendants" is largely
a matter of semantics and classificatory convention that has been debated
here (and elsewhere) ad nauseum for years (and I'm not really in the mood to
open that "can of worms" yet again).
Since I do not think "Cryptovolans" is a dromaeosaur, I disagree with
Gregory Paul's statement that the idea of "dromaeosaurs as protofliers" is
dead, dead, dead. Of course, dromaeosaurs may very well turn out to be
secondarily flightless, but I just don't think Cryptovolans is evidence of
this. Thus I think he is "burning bridges" prematurely.
I try to take a lot of different "complementary approaches" to such
problems of phylogeny, but in the face of the enormous amounts of homoplasy
that occurred during the transition from theropods to early birds, I'm not
sure just how objective one can truly be at this point in time. Cladistic
analyses have generated taxa like Arctometatarsalia and Bullatosauria, and
further cladistic analyses several years later showed they were based on
homoplasies. This helps to demonstrate the long-term value of cladistic
analysis, but also the growing problems associated with purely cladistic
As for using "fossil series", this should cause no big problems at
species or generic levels (especially for groups that fossilize easily).
Otherwise you start running into "ghost lineage" problems, something
Feduccia has long ignored when arguing that the fossil record supposedly
shows birds couldn't have evolved from dinosaurs (newer fossils are showing
that such an argument lacks any merit). Besides the problem of lack of
fossils (such as ghost lineages), there is also the problem of
misinterpretation of fossils---and I think that is one area in which
malacology has gone badly astray over the years. But I've just begun to
explore that whole mess.
Cladistic analysis is just a tool, with great potential once we learn
how to use it better. But like any tool, it can also do a lot of damage
when not used properly. If you prematurely freeze cladistic findings into
formal taxa (formal nomenclature, you are just asking for trouble.
Obviously, I find Lophotrochozoa and the Three Domains to be notorious
examples of misinterpretation, misclassification, and continuing
misinformation. But perhaps they will be of value to future generations of
biologists about the pitfalls of circular reasoning on a large scale.
Pierre Deleporte wrote:
>More seriously (than my previous post), it would be interesting to check
>this "Aves" example a little bit more. The following comments (in the
>dinoarch list) raise two points : the cladistic method would not be
>involved per se in the "mistake", and a plea for integrating different
>As far as I understand the problem, there is no strong cladistic resolution
>due to a lot of homoplasy, and fossil "transition series" may be
>also, some homologies may have been overlooked ("hallux" problem), but this
>is not a "cladistic" recommendation at all, just the possible weight of
>received wisdom, as possible in any approach.
>These are different questions, and to take just one : I know of "cladists"
>(and just here in Rennes University paleontology dept) who are ready to
>take fossil series ("as read in the rocks") into account when they are
>deemed fairly "complete". This the more when cladistic resolution is
>doubtful or lacking.
>Isn't the problem one of integrating complementary approaches, and which to
>be more relied on in cases of contradictions, depending on the relative
>"strength" of historical signal issuing from these complementary approaches
>? And how to assess objectively this relative reliability...
>Also : clearly distinguish "cladistics" as phylogeny reconstruction
>methods) from cladistic classification (strict of relaxed...). Seems that
>G. Paul, in this paper, aims at the former, not the latter.
>A 19:54 13/09/2002 +0000, Ken Kinman wrote :
> (...) Here's a link to Gregory's post for those who wish
>to read it:
> --- Cheers, Ken Kinman
Join the worlds largest e-mail service with MSN Hotmail.
More information about the Taxacom