Are current critiques of the theropod origin of birds science? Rebuttal to Feduccia (2002)

Auk, The, Apr 2003 by Prum, Richard O

Because Feduccia's scenario is untestably vague, he is reduced exclusively to criticizing the theropod hypothesis. Feduccia (2002) portrayed the theropod origin of birds as a vast cladisitic conspiracy to deceive the ornithological community. Like a conspiracy theorist, he unerringly interprets evidence-even the new, ultimate evidence of feathered theropods-as favoring his preconceived conclusions. Feduccia (2002) argues that hypotheses of character homology cannot be tested by congruence with other characters, advocating a character by character approach instead. This ad hoc method permits him to interpret each separate character in whatever manner he prefers to support his argument, and to avoid dealing with the overwhelming volume of character evidence from all parts of the body supporting the theropod origin. He frequently argues both sides of the same issue wherever it is convenient. For example, when the first nonavian theropod furcula was described for Velociraptor (Norell et al. 1998), Feduccia and Martin (1998) claimed that the presence of a furcula in the basal archosauromorph Longisquama demonstrates that those structures have evolved multiple times in archosaurs and are "weak evidence" of phylogenetic relationship. Now, applying his "character by character" approach in this commentary, Feduccia mentions that the furcula of Longisquama in an effort to support the hypothesis that Longisquama is closely related to birds, in direct contradiction to his earlier published conclusions. Simultaneously, he ignores the discoveries since 1998 of furculae or paired clavicles in six to nine major lineages of theropod dinosaurs, including dromaeosaurs, oviraptors, tyrannosaurs, allosaurs, and coelophysids (Tykoski et al. 2002). Of course, the only scientific way to resolve alternative hypotheses of character evolution and character conflicts are by applying a repeatable, explicit, analytical method to reconstruct evolutionary history of characters and organisms. But Feduccia (2002) openly rejects available systematics methods as forcing "into algorithmic form what is arguably the most subjective and qualitative field of biology."

Feduccia's (2002) dogmatic belief in rampant evolutionary convergence in morphology literally begs the question-that is, presumes exactly the facts that need to be demonstrated. The only way to demonstrate convergence is to show that the majority of character evidence supports a hypothesis of phylogeny in which the convergent characters are shown to have evolved independently (Patterson 1982, Pinna 1991). Yet, Feduccia rejects both repeatable systematic methods and explicit, testable alternative hypotheses, making it impossible for him to demonstrate the convergence in which he believes. Feduccia (1999a,b; 2002) repeatedly cited his favorite examples of failed phylogenetic hypotheses as a justification for rejecting repeatable scientific methods, but he failed to mention that it was subsequent phylogenetic analyses that established the preferred hypotheses. Actually, the repeatable scientific methods that Feduccia rejects are the solution, not the problem. Feduccia (2002) further states that the theropod hypothesis implies that dromaeosaurs are "avian ancestors," and that bird are "derived from dromaeosaurs," when it is well known that the hypothesis states that the birds and dromaeosaurs (probably including troodontids) are sister groups.