On 2008-01-27 12:03, Richard Wordingham wrote:
> I think this particular example only applies to living organisms. If
> you put Archaeopteryx in Aves, you again open up a host of problems!
> Bakker possibly resolved this problem by putting Tyrannosaurus in Aves!
Perhaps I should have written Neornithes, the crown group of Aves. Its
definition is uncontroversial and the clade is robustly supported by all
analyses. If we wanted to define birds as e.g. all taxa more closely
related to the house sparrow than to crocodiles, we'd classify not only
_Tyrannosaurus_ but _all_ dinosaurs (and pterosaurs to boot, not to
mention some less known Mesozoic critters) as Aves, just renaming an
already existing taxon (Ornithodira, I think). So in order to make Aves
do some useful work a more restrictive definition has to be used.
Recently many palaeontologists have been arguing in favour of abandoning
the traditional view that Archie is the basalmost bird just to avoid the
instability of the clade Aves in the light of new discoveries. That
would leave quite a few small feathered dinos outside the pale, but then
questions like "was _Rahonavis_ a bird or a dinosaur?" are meaningless
anyway. Clades may be non-arbitrary, but the names we give them are. We
would have similar problems with IE if we could learn anything about the
closest cousins of PIE (unless it had been an isolate for ages).
Conveniently for us, they have vanished without trace (unless one counts
the Tocharian and Anatolian groups, but they've been incorporated).
Piotr