Re: Croc classification (was Re: Sarcosuchus and Dumbing things down)

["Oliver Rauhut" <orauhut@mef.org.ar>]

Hi Steve, nice to hear from you again!

Well, to the point. 
Concerning the question of phylogeny and taxonomy, I think it is important
to make a distinction between species-level taxonomy and higher-level
taxonomy, at least in paleontology. At the species level, taxonomy will
always be independent from phylogeny to a certain degree. If you have the
last branching point between two specimens in a cladogram, it is pretty much
up to you whether you think that these two specimens represent the same
species or not. Likewise, the creation of new species also just depends
indirectly on the phylogeny: If you can narrow own the phylogenetic position
of the creature, at least to some group like "Dinosauria", you don't need to
run a cladistic analysis to show that it is a new species, if it clearly
differs from all other members of that clade in some way, be it in the
possession of discrete autapomorphic characters, or in an apomorphic
combination of homoplastic characters.
Another problem that we as paleontologists have to deal with in species
taxonomy (and, to a certain degree that also has an impact on phylogenetic
reconstructions), is that, according to our recent views of evolution, each
species has their origin as a population of a, (like it or not) by cladistic
definition, paraphyletic mother-species (which, at some earlier point in
time, represented a monophyletic unit, of course. Thus, I am always mildly
irritated if you see in some cladistic textbooks the equation of
paraphyletic = unnatural grouping. Without this "historical paraphyly",
there would be no descend by common ancestry (since every ancestor, be it a
species, or a population has to be paraphyletic to give rise to another
species) and thus no evolution!). Thus, our problem is not only to find out
what distinguishes a species from another species, but to also find out at
what point in time it does so. Admittedly, this is a rather philosophical
question, since we will (at least in vertebrates) never have a fossil record
that is complete enough to get this detail. However, this shows that, in the
context of evolutionary theory, there cannot be a simple solution to the
question of species-level taxonomy, at least not for paleontologists.

But I'm diverging from the topic. To a certain degree, I agree with Steve
that higher-level taxonomies depend on the framework that you are working
in. We have widely come to accept that taxonomies reflect phylogenetic
relationships, but that does not necessarily need to be so. If you are
interested in, say, the biomechanic principles of bipedal walking, a
phylogenetically completely polypheletic group "Bipedalia", including birds
and humans and so on, has the same merits for your and your colleagues' work
as do the phylogenetic groupings for evolutionary studies: It denotes a
sharply defined group (everything that walks on two legs) and thus helps you
to communicate with other workers working in the same framework. Of course,
if you are interested in the evolution of a group, or, to stay with the
example, in the evolutionary origin of bipedality, a taxonomy based on
phylogentic relationships provides the right framework for this study.

One thing we should never forget is that the names given by us to any group
of organisms represents a human concept of this group, and does not reflect
some underlying "fundamental truth". In this context, I have to admit that I
am also opposed to the rigid use of phylogenetic taxonomy, since it only
creates a stability of a name rather than reflecting a real concept of
phylogenetic relationships (as pointed out, among others by Dominguez, E. &
Wheeler, Q. D. (1997): Taxonomic stability is ignorance. - Cladistics, 13 :
367-372). Although a group defined by only two of its members (be it stem-
or node-based) will always remain monophelitc, no matter under which
cladistic hypothesis, it is simply not practical to retain a name if the
content, stratigraphic range and characters defining this group change
completely from one hypothesis to the next. I am not against using
phylogenetic definitions for taxa, but think that they should always be seen
in the context of the cladogram they are created for, and, if a new
cladogram comes to different results, it might be better to abandon a name,
or change its definition slightly, than to just keep it, because there is a
clade it can be applied to. Remember: the name is a means for us to
communicate with each other, not some purpose in itself!

Phuu, that was rather a lot...

Cheers,

Oliver


___________________________
Dr. Oliver Rauhut
Museo Paleontólogico Egidio Feruglio
Fontana y Lewis Jones
9100 Trelew
Chubut, Argentina
Tel: 0054 / 2965 / 420012
Fax: 0054 / 2965 / 432100
email: owmrauhut@hotmail.com
       orauhut@mef.org.ar

----------
>Von: "Steve Salisbury" <steve_salisbury@bigpond.com>
>An: Dinolist <dinosaur@usc.edu>
>Betreff: Re: Croc classification (was Re: Sarcosuchus and Dumbing things down)
>Datum: Die, 30. Okt 2001 6:23 Uhr
>

> Some parts of the previous chain have been deleted.
>
> ----------
>>From: chris brochu <cbrochu@blue.weeg.uiowa.edu>
>>To: dinosaur@usc.edu
>>Subject: Re: Croc classification (was Re: Sarcosuchus and Dumbing things down)
>>Date: Tue, 30 Oct 2001 2:48 PM
>
>> Crown groups have a number of benefits, one of which is stability -
>> and here is where I really think Steve Salisbury is missing the point
>> of stability.  We can never achieve stability of content, which is (I
>> think) what he meant by "Crocodilia in the old sense had stability" -
>> everyone agreed on the content.  I disagree with that (witness his
>> agreement with me that the lower bound of "Crocodilia" was poorly
>> defined), but more importantly disagree on whether everyone agreed on
>> the underlying  *meaning* of Crocodilia.
>
>> The truly important point is that phylogenetically-defined names have
>> stability in meaning; stability in content, in diagnosis, and in time
>> of origin can never be stable (under ANY taxonomic system)
>
> Eh?  How can a cladistically defined name have meaning in a
> non-cladistically based taxonomic system?  The truly important point is that
> there is more than one set of rules.  There's also more than one game, and
> no one should be telling us which one we have to play.
>
> At the end of the day (and hopefully this chain!) it all depends on whether
> you believe taxonomies should reflect phylogenies.  I don't think taxonomic
> categorisations have anything to do with phylogeny or, for that matter,
> evolutionary transformations.  To my mind, taxonomic definitions that rely
> on assumptions of relationship are therefore meaningless (sorry Chris).
> Taxonomies are much easier to defend when they don't involve phylogenetic
> presuppositions.
>
> Steve
>
> ---------------------------
> Steve Salisbury
>
> Palaeontology and Geology, Queensland Museum
> PO Box 3300, South Brisbane, Q 4101, Australia
>
> www.Qmuseum.qld.gov.au
>