[Taxacom] burn out (was: classification of Class Rosopsida)
Richard.Zander at mobot.org
Thu Apr 23 11:22:16 CDT 2009
There continue to be good responses to my inveighing against holophyly
as a method of classification, so the flame seems eternal.
ASL: It is very cunning to make the cut where you made it
Reply: Welcome to the Marketplace of Ideas.
ASL: What I am saying is:
the ancestors did not survive as a paraphyletic assemblage of lineages
but instead they survived as both this paraphyletic "taxon" and the one
that would have to be included to make it monophyletic.
Reply: Ah, I see. In my opinion the extant taxa (by exemplars) did not
arise out of a confusion of ancestral taxa, but, because a section of
the phylogenetic lineage consists in part (basally) of a series of lines
coming off and all of the same taxon, then the direct ancestor of THOSE
lines was that taxon. AND the autophyletic taxon descended from THAT
ASL: I still have to be shown how the history of evolution can be
anything else but basically, largely, a tree.
Reply: I'm not trying to do so. Darwin's Tree of Life was an
ancestor-descendant tree, one taxon derived from another. Modern trees
are usually sister-group trees, with taxa derived from nodes. They are
both trees and address evolution. I think the Darwinian tree represents
evolution more directly than the Hennigian tree.
ASL: Mentha cannot be placed in a bifurcating tree because it has two
extant ancestral species does not keep us from circumscribing the genus
Mentha as monophyletic. Nobody says that bifurcation is the only thing
that happens, but for higher taxa it is typically the only thing that we
can possibly infer. We use morphological, chemical and molecular markers
for or inferences, and they all undergo lineage sorting over time.
Reply: You must make a distinction in whether you are referring to
phylogenetic monophyly (=holophyly) or evolutionary monophyly.
Phylogenetic monophyly is an artificial classification, and you can
classify anything artificially. Evolutionary monophyly simply refers to
the dictum that every taxon is ultimately derived from another more
ancestral taxon whether sequentially or bifurcating tree or bush.
ASL: How the long-dead ancestral populations looked like can
approximately be inferred by subtracting all apomorphies back to the
node of interest. And what would you put in the place of this approach?
Reply: You refer here to character mapping on a molecular tree, I think.
Okay so far as it goes, but it assumes gradualist (not punctuational)
evolution and the biological species concept (every split becomes a new
taxon). In place of this I put taxon mapping. We can fine ancestral TAXA
by recognizing paraphyly for what it is, not just ancestral TRAITS,
though the taxa are fully diagnosable by extant traits as a kind of
ASL: As far as I understand, this: simply defining the ancestors as
identical with only one fraction of their descendants, no matter how
much they have changed in the intervening millions of years (just as
long as they have not changed with regard to the apomorphies that the
rest of the descendants have evolved).
Reply: Yes yes! Paraphyly distinguishes the fraction of the descendants.
Oh, and using TAXA instead of traits makes use of the single most
important evolutionary feature of taxonomy, namely that species (and
doubtless higher taxa) are morphologically static for thousands and
millions of years, which is why we can do taxonomy at all.
ASL: Of course genus is an arbitrary rank, as are all ranks except
perhaps, in some ideal cases, that of species. The most important reason
behind the decision where to apply that rank appears to be tradition.
Reply: No it isn't. Genera are real (as distinguishable, describable
clusters). Prove they are not. "Tradition" refers, I expect, to 250
years of taxonomic endeavor, sloughed off in favor of an artificial
classification (holophyly). The only way I can think of to approach
problems in standard taxonomy is to subject it to Quine's tests of
ASL: Otherwise, how could we explain that mammal genera usually contain
1-2 species, while genera are much more broadly described in all other
taxonomic groups? That large genera such as Justicia s.l. (Acanthaceae)
or Satureja s.l.
(Lamiaceae) have been circumscribed in a very wide manner by some
botanists but split into myriad segregate genera by others?
Reply: This is called Science on the March. It involves dealing with
data using every method that seems relevant to constructing a
classification that reflects what we know about evolution in a group,
struggling with the results, forming hypotheses and theories that may
contradict each other, and so on. A quick fix, like automatic
classification by holophyly, is no solution.
By the way, I'd like to state here that the endless wrestling over
certain concepts to the point of burn-out or ennui is to a large extent
what Taxacom is for. We all have made mistakes, gaffes, lapses, and
foolishness, and have had moments of grace and discovery, but what I
expect comes across to most members of Taxacom is how much all
participants in these discussions love and respect science. I, for one,
find it a privilege to live in a time when such easy colloquy is
Richard H. Zander
Missouri Botanical Garden
PO Box 299
St. Louis, MO 63166-0299 USA
richard.zander at mobot.org
Web sites: http://www.mobot.org/plantscience/resbot/
Non-post deliveries to:
Missouri Botanical Garden, 4344 Shaw Blvd., St. Louis, MO 63110
From: Alexander.Schmidt-Lebuhn at biologie.uni-goettingen.de
[mailto:Alexander.Schmidt-Lebuhn at biologie.uni-goettingen.de]
Sent: Wednesday, April 22, 2009 3:40 PM
To: Richard Zander
Subject: Re: [Taxacom] burn out (was: classification of Class Rosopsida)
More information about the Taxacom