Re: Defending grades (Was: Re: Archaeopteryx (rant))

[David Marjanovic <david.marjanovic@gmx.at>]

> T. Michael Keesey wrote: 
> 
> >> That one group of animals can give rise to another group that should 
> >> not be included in the ancestral is a biological fact. 
 
"should" and "fact" is a contradiction. :-) 
 
> > No, it's a human construct, and, I would argue, not a very useful one. 
> 
> In my original email the statement above was followed by an example, 
> namely how a species becomes paraphyletic [...] 
 
Yes, a species. But why should we carry this on to other taxonomic 
categories? (Especially if we consider clades as being composed of 
_individuals_ and not of species?) We don't need to. So why should we 
burden ourselves with this? 
 
> If this didn't happen every organism on this planet would 
> still belong to the same species, so this phenomenon is rather 
> central to evolution. 
 
To the origin of species. Not to the rest of evolution. :-) 
 
> Regarding how we humans treat higher taxa there is of course a lot of 
> construction involved. 
 
Don't you think we should minimise it? 
 
> (Once upon a time I studied the various levels of reproductive isolation 
> in a widespread living superspecies complex, so I know a little about  
> what "speciation" means. 
 
Congratulations! But I don't think speciation is what we're talking about 
in this thread... 
 
> Unfortunately, it is difficult to do the same with 
> fossil taxa...) 
 
(So much so that I think most fossil species shouldn't be recognised as 
species at all.) 
 
> The reason we still use the Linnean system is that it is a very good 
> tool for information storage. Once you know the system the position of 
> one taxon gives you a lot of information about its affiliation, 
> anatomical and other characteristica and what it includes. 
 
For information about its affiliation phylogenetic nomenclature is much 
better. And anatomy etc. are only half as interesting without phylogeny. 
:-) Also don't forget that, _unlike_ Linnaean taxonomy, phylogenetic 
nomenclature allows you to precisely define (just not name) _any_ 
paraphyletic group that you want! (The most famous example are the 
nonavian dinosaurs.) 
 
> A cladogram can be based on any inheritable character. 
 
Very few -- and still too many -- cladograms are based on only one 
character. But this isn't part of the debate -- phylogenetic nomenclature 
presupposes that there _is_ a phylogeny, not that we've used cladistics or 
any method to find it. 
 
> [...] analyses are now often almost exlusively based on DNA 
 
Combined analyses are pretty common. 
 
> because it is more reliable regarding ancestry. 
 
It generally gives more characters. But at the same time it is more prone 
to long-branch attraction... it's complicated. :-) 
 
> Working with extant anatomy you would 
> e.g. never get the whales dumped in the middle of Artiodactyla. 
 
Working with fossil anatomy you get Artiodactyla and Cetacea as 
sistergroups. (IMHO using extant anatomy of vertebrates, where lots of 
fossil anatomy is known, is a crime.) 
 
> BTW, this is another example where the anatomy of a group of animals 
> have changed so much that it is defensible to classify it outside the  
> parent taxon instead of squeezing it inside, e.g. as a sistertaxon to 
> Hippopotamidae. 
 
And here we have it! I assume you'd like to keep Cetacea separate because 
it's so big (diverse)? Then how big is so big? Can you quantify that? 
Because if you can't, I can come and say "no way it's big enough, look how 
small it is" and squeeze them all into Suborder Whippomorpha, and _there 
is no way at all_ how we could resolve our dispute. :-o 
        If instead you'd like to keep it separate because it's 
morphologically so distinct, then the same applies -- in the absence of a 
genericometer. And in the presence of one, where are the enormous 
differences between *Ichthyolestes* and... well, some contemporary 
artiodactyl? 
        Of course _this_ is the state of the art of Linnaean taxonomy. :-( 
:-( :-( There are many books which have e. g. a classification of all 
fossil animals somewhere in the back, and _no two_ of those are the same. 
The reasons are 2 sources of instability: changes in phylogenetic 
hypotheses, and mood swings (above: "wait, you're right, it is big 
enough"). Phylogenetic nomenclature retains the former, but completely 
eliminates the latter. Therefore I think phylogenetic nomenclature 
_increases_ stability, contra Benton et omnes. (Who seem to think that the 
2 eliminate each other -- sometimes this does happen, but most of the time 
it does not.) 
 
> > Classification is intended to ease our understanding of a species' 
> > origins. 
 
And phylogenetic nomenclature accomplishes this best. 
 
> > One can note that aves evolved from reptiles, while still maintaining 
> > them as a separate class. 
 
I'm not arguing for the abolishment of the name Aves. I'm just arguing 
against taking them out, against chainsawing the tree of life apart and 
arranging its branches in mid-air in some chessboard pattern. 
 
However, for the record, I am arguing for the abolishment of the name 
Reptilia. The existing phylogenetic definition contradicts a PhyloCode 
recommendation against huge changes in content, and changing it from 
"horizontal" (all basal amniotes except 2 clades) to "vertical" (almost 
all of one of the 2 component clades of Amniota) is such an enormous 
change in content. Why would anyone want to lump crocodiles with turtles 
instead of with birds? Why would anyone want to lump snakes (which produce 
lots of uric acid) with turtles (which keep producing lots of urea) 
instead of with birds (which produce lots of uric acid)? Let's have 
Amniota, Theropsida and Sauropsida. Is way cooler. :-) More useful, I 
mean. Yes, I mean it. 

-- 
+++ GMX - Mail, Messaging & more  http://www.gmx.net +++
Bitte lächeln! Fotogalerie online mit GMX ohne eigene Homepage!