Why birds aren't bird-hipped

I was teaching a palaeontology class a couple of weeks ago, and showed a phylogenetic tree of the Dinosauria, with birds as the extant clade.  The discovery of Archaeopteryx in the 19th century first hinted that our avian friends might actually be 'living dinosaurs', but it's only in the last few years that the Chinese dino-birds have shed a whole new light on the issue.  No palaeontologist or evolutionary biologist would now dispute that birds really did evolve from dinosaurs.

An artist's reconstruction of Anchiornis, a Jurassic 'dino-bird' from China

Interestingly, though, one of the students in my class noted something I hadn't.  Although dinosaurs are divided into the bird-hipped (Ornithischia) and lizard-hipped (Saurischia), birds are placed in the latter, not the former.  How could birds not be bird-hipped, he asked?

"It's probably convergence," I speculated, but I didn't know the answer, and had to go away and look it up.  I was relieved to know that my speculation was correct, but it's still an interesting point, and one that I shall refer to in future.  And when doing so, I might just use this little passage of text from the University of California Museum of Paleontology website:

"The etymology behind the two names ("bird-hipped" vs. "lizard-hipped") is rather confusing, since some saurischians had bird-like hips, and ornithischians' hips were somewhat birdlike due to convergent evolution, not due to shared ancestry. Birds are apparently descended from saurischian dinosaurs, but have a reversed pubis like ornithischians do. Some close relatives of birds within saurischians have this same feature, too, so the ornithischian-saurischian dichotomy is not so simple. The names "Ornithischia" and "Saurischia" are used to refer to the common ancestry of their respective members. The names don't necessarily have to mean anything. They are just names, that's all."

"Hello Anchiornis huxleyi, how are you today?"  "Very tiny indeed, thanks. And truly ornithischian!"