[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Archaeopteryx

--- Colin McHenry <cmchenry@westserv.net.au> wrote:
> >"Oh, I feel like this should
> > be in _Aves_. No, this should not be in _Aves_." That's completely
> > subjective.
> It's completely subjective to say that a reasonable definition of Aves
> should not include crocodiles?  Snakes? Turtles?  Mike, where would you draw
> the line?

At the last common ancestor of _Archaeopteryx_ and _Neornithes_, of course.
It's traditional taxa that don't draw a clear line. Let's say you "define"
Classis Aves by listing all known species. Then we find something new,
something slightly more basal than the most basal species you've listed. How do
we decide where to put it?

Phylogenetic taxonomy provides a clear line. Yes, we may not always be entirely
certain which side a given organism falls into, if it is not a specifier, but
that's the nature of the beast. To mandate total membership is to decide ahead
of time how the relationships should be -- that's not science.

> The point is that, potentially, no matter how well you think you understand
> a taxon and its place in the tree of life, it is theoretically possible for
> even a well understoof organism to be shifted around the tree quite
> radically in the light of new fossils and/or analyses.  It's happened
> before, and it'll happen again.  If you think that, in 2003, we've finally
> got the right topology of the tree, I think you're in for a rude surprise.

But our taxonomy should update itself to fit the tree.

The real problem is a conceptual one, and would probably be mitigated if PT
used entirely new names, insteading of co-opting traditional ones. We have a
tradition of taxa being typological, with borders defined by authority and
traditional usage. It's hard for some to get used to the idea of a taxon as a
specific ancestor plus all descendants, even if that is much more in accord
with evolutionary history.

> Actually, one of the ideas behind taxonomy is to provide a stable
> classification that can be used to enhance communication between natural
> scientists of different disciplines, and between science and the general
> public.

Linnaean taxonomy is not stable, because it does not define anything rigorously
(possibly excluding species, possibly not). As an example, Familia Hominidae is
simply "the family that includes _Homo_". With no clear indication of exactly
what a "family" is, workers have been free to include or exclude great apes as
they feel fit.

The membership of phylogenetic taxonomy may not be stable, but the definitions
are. (Well, they will be once workers agree to a code -- but we're getting
there.) _Hominidae_ will be Clade(_Homo_ <- _Pongo_) -- that's it.

And I would argue that a classification should not be entirely stable. To
create a rigid classification is to assume that you have everything figured
out. But I think that where the classification changes, it should change
according to new data (as in phylogenetic taxonomy), not according to personal
opinion (as in Linnaean taxonomy).

> As far as I can see, phylogenetic taxonomy is a quasi-religious
> tool used to beat about the head those who are not true-believing members of
> the high priesthood.  I think it's missing the point by such a large margin
> that it no longer has anything to do with traditional taxonomy.  Whoever
> first said that taxonomy had to equal phylogeny was a fool who didn't
> understand (a) taxonomy, (b) how the human brain perceives and recognises
> patterns, and (c) phylogenetics.

Well what should it equal, then? Phylogeny is a truth which we are uncovering
(or attempting to). I don't see any other suitable basis. Size? Color?

Traditional taxonomy, at least since around the time of Darwin, is mostly
phylogeny-based, anyway. Would you advocate polyphyletic groups? I think you'd
have to at some point, if you wanted your taxonomy to be completely stable.
> > Personally, I really don't think we have to worry about ceratosaurs being
> > avians. But if dromaeosaurids, troodontids, oviraptorosaurs,
> > etc. fall in Clade(_Archaeopteryx_ + _Neornithes_), then, yes, they should
> > considered avians.
> Excuse me, but this is bullshit.  If you found that any one of the taxa
> currently included in Aves (Archaeopteryx + Neornithines) actually lay
> outside of, say, the Tetanurae, you'd shift it without missing a beat and
> keep Aves (in terms of included taxa) almost identical. And so you should -
> you're only moving one taxon out of maybe 10,000 in this group. Why should
> it be any different if the one taxon you are forced to shift (as a result of
> your new improved phylogeny) happens to be the one that (for reasons to do
> with accidents of history and convenience) you had previously used to
> bracket a higher taxon?  You are then completely changing the content and
> practical definition of a major group for no better reason than it is
> consistent with your ideology.  To me, this is throwing the baby out with
> the bathwater.

I will agree that there should be some connection with traditional usage of
taxa, but I don't think we should be completely ruled by this. Formal taxa
should not be subject to aesthetic ideas, but to rigorous definitions.
Otherwise, it is nothing more than folk taxonomy in Latin.

> I doubt that Nature has any preference for following formulae that we might
> possibly be able to understand, at least in the context of this discussion.
> As for 'opinion' and 'authority'?  Well, as I've argued before on this list,
> I think they count.  I think that humans are a much more sophisticated
> pattern recognition device than any algorithm that will be devised in our
> life times.  I think that a human that knows its subject inside-out and
> back-to-front is worth listening to - as do you, I'm sure, if you stop to
> think about it.

But this is not science. Each human is going to have a different outlook --
science is supposed to be about approaching the objective truth, not about
creating a lexicon of ill-defined insider lingo based on cultural traditions.

> Just because people such as Simpson did not explicitly frame ways of testing
> their theories of phylogeny, doesn't mean the whole notion of 'prefered
> phylogenetic hypothesis as influenced by expert opinion' is inherently
> wrong.  You need to listen to everyone with a critical ear.  That includes
> modern phylogeneticists as much as it does the old time 'experts'.  In fact,
> your modern phylogenetic theories are still framed by the work of the
> previous generation, even if they did not use 'explicit formulae'.  And the
> opinion of the current generation of experts does counts for something in
> today's world of 'rigorous' and 'explicit' methodologies.   And part of that
> opinion will be formed by cognitive processes that lie outside of the strict
> bounds of modern phylogenetic taxonomy methodologies.  What are you going to
> do about that?

This paragraph seems to be more about cladistic analyses than phylogenetic
> Taxonomy and phylogenetics will never be free from preconceptions.  I think
> you're better off developing a theoretical framework that can accomodate
> this, rather than dogmatically pursuing an unattainable ideology.

The truth may not be attainable, but it is approachable.
[more on cladistic analyses, rather than taxonomy]

> A dogmatic obsession with phylogenetic taxonomy assumes that the One True
> Tree Of Life is discoverable.  It is not.  It can only be approximated, and
> probably not all that well.  The level of accuracy with which we can
> reconstruct the One True Tree Of Life is logically and philosophically
> constrained by a large number of factors, some of which include the quality
> of the fossil record and the vagarities of the methodologies used to
> construct phylogenetic theories.

I just can't identify with that attitude. It really starts to sound like PT
advocates are saying, "Let's clean this up," while traditionalists are saying,
"It'll just get dirty again, and we like the mess anyway."

I think we can more and more successfully approximate the truth through more
data and refined analsyses. Furthermore, you aren't offering any alternative
criteria other than expert opinion, which is based on phylogeny, anyway. (And
what happens when the experts are mostly PT practitioners?)

> Additionally, one of the most important requirements of any taxonomic system
> is that it should be stable enough to promote communication, not hinder.
> The results of phylogenetic theories change with new fossils, new
> algorithms, and new analyses.  I personally do not think that the taxonomy
> that I use in communicating my work to other scientists, and the public,
> should have to be revised every time some kid with a copy of PAUP updates
> their character matrix.   Taxonomies should work for us, not the other way
> around.

I agree that, in general, only the most well-supported nodes should be named.
But, as I stated before, if your taxonomy is rigid, it is going to fall out of
alignment with the truth. Otherwise we'd still have Pachydermata, and
caenagnathids in Arctometatarsalia.
> Besides which, my personal observation is that an desire to keep taxonomies
> in line with the latest phylogenies can very quickly develop into an
> obsession.

Guilty as charged.

> You will soon find that you have little time for considering
> anything but phylogenies.  To suggest that any group, least of all one as
> magnificent as the dinosaurs, is worthy only of phylogenetic analysis is, in
> my opinion, a great shame.

Who suggested that?
> An example of this?  Even on this list, most posts (especially in recent
> times) are concerned with details of phylogenetic taxonomy.  I personally do
> not think that this is a healthy situation.  I notice, with the posts I make
> to the list, that most go largely ignored, but the few times I make a
> comment (usually critical of the status quo) on cladistics or taxonomy there
> is always a flurry of hurried, indignant responses.  I am not trying to
> bellyache about this - it doesn't really bother me if my other stuff gets
> little response, because my motive for writing it is usually to try and
> clarify my own thoughts on a subject through the act of writing them down -
> but I wonder if, for the large number of non-professional dinosaurologists
> involved with this list, the emphasis upon phylogenetic systematics reflects
> the focus of their interest in dinosaurs?

It does for me. So what?
My interest in dinosaurs has always been in their diversity and relationships
to each other. Studying phylogeny and taxonomy is one of my ways of
appreciating that. I like to have a framework to see how these creatures
evolved. And I think having a more science-based taxonomic system can only help
to discuss other subjects (behavior, biogeography, etc.)

So I like discussing taxonomy more than most other topics. That's me. Why would
you want to change that, or think that you could?

> Anyway, before I get flamed again, I just want to make clear that I don't
> mean anything personal against Mike (on the contrary, I have a great deal of
> respect for his knowledge on dinosaurs), and I certainly don't regard
> phylogenetics as a waste of time.  But I strongly disagree with the thrust
> of his argument on this subject.

I, too, hope I haven't made my arguments too personal.

=====> T. Michael Keesey <keesey@bigfoot.com>
=====> The Dinosauricon <http://dinosauricon.com>
=====> BloodySteak <http://bloodysteak.com>
=====> Instant Messenger <Ric Blayze>

Do you Yahoo!?
SBC Yahoo! DSL - Now only $29.95 per month!