[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]
Re: Philidor: No Class (long, too)
----- Original Message -----
Sent: Tuesday, August 27, 2002 3:34 AM
> <That what is important for one but can be ignored by another,
> and that none of us know the true picture, argues that none of
> us can see the truth, or know it exists, just argue for perfect
> agreement. This is parsimony, and its the #1 best tool
> in systematics.> [HP Jaime Headden]
> If I'm reading you correctly, I couldn't make a better argument
> against parsimony analysis.
> Myself, I'd say: Because evolution is known not to operate always
> and inevitably in the most parsimonious manner,
Is it, actually? There aren't many cases of evolution having been observed
in real time... not that we have here a truism in the making!
> any analysis using parsimony must remain guesswork.
But it is better guesswork than an analysis using nothing! :-) Parsimony is
certainly a good way to start, at least, because we know evolution is not
totally imparsimonious. (We know this because we don't see hopeful monsters
being born, not even a dolphin hatching from a shark egg.) What else would
you suggest to find out phylogeny?
> Whatever prior views were, the Linnaean system is not now hostile
> to science, obviously.
But it isn't scientific itself. It's impossible to test which group a living
being belongs to in the Linnaean system because the group names are not
defined in a way that would allow that.
> To my understanding, replacement is not
> a priority with many of the workers using the system.
Because those that see it as a priority already try to use the Linnaean
system as little as anyhow possible ( = required by journal editors).
> <One then must [...] accept a
> logical lineage of descent, which must be monophyletic.>
> The first comments would be: we can never determine for certain
> a single monophyletic lineage, so basing a classification system
> on current opinion raises flexibility to the level of the indefinite.
This is why phylogenetic _nomenclature_ has been invented. It is not based
on current opinion. As long as *Triceratops*, *Passer* and evolution exist,
Dinosauria cannot help but exist, regardless of whether jellyfish are
dinosaurs or not. A cladogram is not a classification system, it is a
hypothesis on phylogeny. A cladogram with names on it is a classification,
but it is not intended as a fixed classification system, instead it is the
inevitable consequence of applying phylogenetic nomenclature to a certain
> Remember that under this paradigm the ancestor is described
> but never named.
Yes, and what's the problem with that? How could we name the ancestor, for
example? We don't have it, so it can't be a type specimen. What the ancestor
looked like is the phylogenetic hypothesis, the cladogram itself.
> And, of course, unless you say that you frequently
> confuse present-day birds and reptiles, you can see a reason
> aside from the historical or the aesthetic for creating these
> essential groupings.
Yes, I can see it. But I can also see why this is misleading because
(traditional) Reptilia is paraphyletic with respect to birds, not its sister
> Strange that you'd consider a system based
> on undoubted observations less scientific than one based on malleable
That evolution occurs is not a malleable inference, it is even more than a
theory according to www.dinosauria.com/jdp/evol/evolfact.htm. That evolution
occurs within e. g. several species of Darwin finches, and many others, is a
fact according to the same webpage, an undoubted observation, and not any
inference at all. And evolution is the only thing phylogenetic nomenclature
is based on. :-|
On the other hand, the Linnaean system is based on things that are
_not even inferences_, for example ranks.
> More interesting, let's take the perfect case: you know the
> ancestry and descendants (if any) of every animal (and plant
> and whatever) that ever lived. You still don't have a classification
> system yet, just an incredibly complex squiggle of lines across
> a page.
> Classification comes when you decide to carve out a section of
> that page and give it a name.
Why do you have to carve it out? IMHO it's enough if you write a label and
tie it around a limb of the tree in a certain position, without cutting the
> For those who don't want to
> understand in detail, the large, simply defined groups. For
> those who do want the detail, you'd be meticulous and use less
> inclusive groups.
> You respond to your audience, but remain consistent.
I can remain even more consistent. Most definitions in phylogenetic
nomenclature will probably be equally simple. :-) No problem with using
Sauropoda for those not interested in the details and Sauropoda,
Euhelopodidae and Saltasaurini for those interested in them.
> And if the people that you're communicating with want concepts
> that center on what they can see, you go along with them.
Trust me: They'll see a fourth trochanter if I show it to them. :-)