Re: Greg Paul is right (again); or "Archie's not a birdy"

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

>  Given that Ornithurae rests on a highly autapomorphic stem clade
>  (Hesperornithes),

This is one of about five phylogenetic definitions of Ornithurae.

>  the plesiomorphic part of which remains entirely unknown,

We do have *Pasquiaornis* and *Enaliornis*, fragmentary though they are, 
and *Gansus* is somewhere close.

>  and given the general lack of known basal diversity in
>  Ornithuromorpha,

*Archaeorhynchus* and *Yanornis* are not exactly the same thing.

(That's assuming they actually belong to Ornithuromorpha, which depends 
on the positions of *Patagopteryx* and -- under one of the two 
definitions -- *Vorona*. But in any case, they belong to Euornithes.)

>  In a similar vein, the supposed apomorphies of Pygostylia and
>  Ornithothoraces (whose non-apomorphy-based definitions implicitly
>  rest on the assumption that the supposed apomorphies *are*
>  apomorphies)

No, that's not how it works. The PhyloCode, at least, is not concerned 
with such etymological implications.

>  seem to be homoplasies at least for the largest part, or even - in
>  case of the pygostyle - hint at nonmonophyly.

The birds with a pygostyle -- as opposed to, presumably, the 
oviraptorosaurs and segnosaurs that may have one -- clearly do form a 
clade, even though *Sapeornis* is outside Pygostylia.

>  Neornithes is good as it is; contesting its validity is unlikely to
>  gain acceptance among neontologists. Hence, if we'd need a
>  phylogenetic definition of Aves RIGHT NOW, the only one that pertains
>  to a clade that is proven beyond all reasonable doubt, and to a clade
>  interesting enough to warrant such a familiar name, and that is not
>  wrought with a whole damn lot of controversy, would be: "Anything
>  closer to _Struthio camelus_ + _Vultur gryphus/Passer
>  domesticus/Gallus gallus_ than to _Enantiornis leali_ + _Sapeornis
>  chaoyangensis_ + _Confuciusornis sanctus_" or equivalent.

But this clade already has a name, Euornithes. Why give it a new one?

And why not include Enantiornithes in Aves? Remember that there are 
several pretty obvious ornithothoraceans of which it is not clear 
whether they are enanti- or euornitheans: *Eoalulavis*, *Eoenantiornis*, 
*Liaoningornis*, *Aberratiodontus*...

>  That clade would include only stem branches for which *non*survival
>  at the K-Pg event would be quite parsimonious due to their being
>  restricted taxonomically and/or biogeographically (it was a mass
>  extinction after all) but it would NOT contain any stem group that
>  was so successful that there is a real need to explain why they (i.e.
>  Enantiornithes) are *not* around anymore today. This makes it
>  attractive from an evolutionary standpoint too.

Hence its name Euornithes, "true birds". :-)

>  This solution might also be considered "not more bloody than
>  necessary", because it takes a widely-known name and restricts its
>  use to the one and only robust and major and significant clade within
>  Avialae and including Neornithes according to present data, which is
>  not very likely to be overturned by future data.

As now explained, I disagree.

>  In a nutshell: 1. It is probably not a good idea to use Archie to
>  define Aves

Agreed (unsurprisingly).

>  2. It is probably a good idea to retain Avialae

Yes -- there are so many clades that should be named that retaining this 
name for something (actually, why not the apomorphy-based definition by 
Gauthier & de Queiroz 2001) can hardly do harm.

>  3. It is probably a good idea to retain Neornithes

Yes.

>  4. It is probably not a good idea to use Aves for an insufficiently
>  robust "clade" or insufficiently large/significant clade

I agree partially, and disagree with your opinion on which clades are 
currently robust.

>  (It would also stand an actual chance of settling the question "what
>  is a bird?" in general knowledge, because it would at the present
>  state of knowledge very nicely fit with "Paraves" and "Avialae" and
>  "Neornithes" - we get "birdish dinosaurs" of which one part are
>  "bird-winged dinosaurs" of which one part are "birds" of which the
>  part that still exists today are "modern birds". As opposed to the
>  present mess, you could teach this even to Junior High kids and they
>  would understand it and learn nothing that is wrong now or likely to
>  be wrong in the future.)

Agreed.