Rapator in Agnolin et al. 2010 and the concept of nomen dubium

I recently engaged in a discussion about nomina dubia on the DML with Tim Williams and Jaime Headden, specifically regarding Ceratops.  Agnolin et al.'s (2010) extensive new review of Australian and New Zealand dinosaurs provides an excellent example of someone labeling a taxon a nomen dubium incorrectly.  Remember, a nomen dubium is an undiagnostic specimen.  It does not preserve any apomorphies, so cannot be distinguished from other taxa.  Here are the facts of the matter in regard to Rapator according to Agnolin et al.

- Rapator is easily distinguishable from most theropods- ceratosaurs, Tugulusaurus, Szechuanosaurus? zigongensis, Torvosaurus, Allosaurus, Acrocanthosaurus, and coelurosaurs including Ornitholestes and Nqwebasaurus.
- Rapator is most similar to Megaraptor and Australovenator, so is a megaraptoran.
- "Rapator and Australovenator differ from Megaraptor in the presence of a more dorsoventrally developed mediodistal condyle and a lateral facet for articulation with the metacarpal II lying in almost the same plane as the lateral margin of the shaft. Thus, albeit being very similar in morphology, Rapator and Australovenator are clearly distinct from Megaraptor."
- "Hocknull et al. (2009) recognized subtle differences between Rapator and Australovenator (e.g. more subequal distal condyles, flat proximal articular surface, straight lateral distal condyle) and we add here the presence of a distal medial condyle ventrally extended in Rapator."

I have no actual opinion on the validity of these statements, as I haven't studied Rapator yet myself.  Let's assume they're all correct.  What conclusion would we reach about the status of Rapator?  There are only two possibilities- it's diagnostic compared to Australovenator or it isn't.  If the former is true, it's a valid taxon.  If the latter is true, Australovenator is a junior synonym.  Yet Agnolin et al. confusingly state "However, due to the fragmentary condition of Rapator and the absence of autapomorphies and clear differences with Australovenator, we consider the taxon to be a nomen dubium."

Put bluntly, a taxon can't be a nomen dubium if it's only undiagnostic compared to one other taxon!  In that case, it's a synonym.  It has to be undiagnostic compared to TWO other taxa in order to be undiagnostic, since then we couldn't tell which taxon it came from.  If Agnolin et al. didn't want to sink the more complete Australovenator into Rapator, they could have relied on the differences they noted.  But if they don't think those differences warrant such a separation, then to be honest they'd have to synonymize the taxa.  But you can't have it both ways.

Another less explicit example is Timimus.  Agnolin et al. consider it Paraves indet., though they distinguish it from Troodontidae and eudromaeosaurs.  They end up viewing it as most similar to unenlagiines, but don't bother trying to distinguish it from any.  Yet Rahonavis is obviously different in having a trochanteric crest, while both Unenlagia and Buitreraptor have downturned femoral heads.  If Timimus is an unenlagiine, it's a valid taxon.

In general, the paper seems confused in regard to "indet." and "nomen dubium".  In their Table 1, the terms are listed separately, with some taxa they consider determinable being labeled as indeterminate (e.g. Kakuru, Muttaburrasaurus).

These are all great examples of why I don't trust pronouncements of nomen dubium status by authors without them going through a rigorous anaysis.  Agnolin et al. are actually much more detailed in their analysis than most authors when it comes to this, but they still end up declaring fragmentary taxa to be indeterminate even when their own written conclusions would indicate it cannot be so.

Reference- Agnolin, Ezcurra, Pais and Salisbury, 2010. A reappraisal of the Cretaceous non-avian dinosaur faunas from Australia and New Zealand: Evidence for their Gondwanan affinities. Journal of Systematic Palaeontology. 8(2), 257-300.