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

Re: Classification: A Definition

On 10/05/2007, at 11:49 AM, T. Michael Keesey wrote:

On 5/9/07, Colin McHenry <cmchenry@westserv.net.au> wrote:

Hear, hear. The question is, if you're not putting together the phylocode as an alternative classification system, then why are you doing it?

Well, it really boils down to your definition of "classification". Phylogenetic nomenclature might be said not to be a form of classification because it does not deal with classes (in the philosophical sense, not the Linnaean sense) but with entities. It provides methodologies for associating names with real entities (clades). Our understanding of that entity is determined by the scientific process (e.g., cladistic analysis). By contrast, classes are not real entities (although they may correspond to real entities), but groups whose composition is determined by the classifier, rather than by science.

(Maybe someone else with more experience in philosophy can explain it
better--my head's starting to hurt.)

I expect Allan Hazen to step in here and correct me, but here's my take.

A class can be a number of things in different logics. But it is often understood in the taxonomic context to mean a set that has a definition; which is to say, it is a general term. So on this, Aristotle's definition of "tree" is a class. Irrespective of its ancestry or ecology, the morphology is definition enough. Whether these sorts of classes are real or not is the central question of the middle ages, after Porphyry declined to answer that question in his introduction to Aristotle's logic (the Isagoge).

In contrast, a clade is not that sort of class. It lacks a definition of content, instead being anchored by particular taxa. The synapomorphies that are used too diagnose a clade are not necessarily what *makes* it that clade, because apomorphies can themselves become plesiomorphies without removing the derived taxon from the clade.

There are a number of taxonomically-driven conceptions of classification. The "evolutionary systematics" form of Mayr and Ashcroft were a hybrid form, involving both clades and classes (which are effectively grades of organisation). Clades simpliciter are notionally formed by descent, so they represent, in effect, the ancestral species of that clade. Since a species lacks an essential definition, so too do clades, although there can be diagnostic definienda of course.

Classification is basically, in my opinion, the saying of the most number of descriptive things in the shortest number of unambiguous words. Some of it is ordered and ranked, some not. Some of it is defined, some ostensively indicated ("this specimen and those most closely related to it"). I think that descriptive classes and ostensive classes are two distinct classificatory groups. In short, classification (in biology) is by both clade and grade. A clade gives you the historical relations, or some aspect of it, while grade gives you defining characters. They need not overlap substantially.

Insistance of the importance recognising major monophyletic groups?
Sure.  Committment to abandoning ranks?  No.

But strict monophyly cannot be used with absolute ranks, at least, not as they are currently used. For example, if we recognize Dinosauria as a class (Linnaean sense) as well as a clade, then the direct ancestor of dinosaurs cannot possibly belong to a monophyletic class. The only solutions would be 1) to allow taxa of the same rank to include each other (which obviates the whole idea of ranks, anyway), or 2) to allow organisms not to belong to any taxon of certain ranks. Neither is permissible under any taxonomic code.

Also, think of Bakker's reclassification of Dinosauria (=Ornithodira,
under his scheme), where Dinosauria is raised to a class and birds can
still only fit by becoming a family!

Of course, if Margulis is right about the eukaryote cytoskeleton being a
result of endosymbiosis with a spirochaete-type bacteria, then the
Eucarya are merely a symbiosis between one branch of the Bacteria and
one branch of the Archea. I'm still confused about how the Phylocode
will deal with these sorts of tree topologies; if the e.g. Archea
includes all of its descendants, including the Eukarya, and the Eukarya
include the descendants of one branch of the Spirochaetes, then surely
the clade that contains all spirochaetes contains the Eukarya (which are
a subclade of the Archaea) as well.

Clades can overlap. It's right there on the very first page of the rules: http://www.ohiou.edu/phylocode/art1-3.html

Maybe I should just read the
Phylocode - does it cover this issue?


I am constantly astonished by how many people critique the PhyloCode
without having actually read it, especially considering how concise it
is compared to other codes. And it is extremely rare that I see
someone bring up an issue that it does not cover. (There are a few, as
it's still in draft form.)
Mike Keesey

-- John S. Wilkins, Postdoctoral Research Fellow, Biohumanities Project University of Queensland - Blog: scienceblogs.com/evolvingthoughts "Darwin's theory has no more to do with philosophy than any other hypothesis in natural science." Tractatus 4.1122