Re: Nomina Dubia Part II: Rapator

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

Tim Williams wrote:

================================================

>  When we split or lump genera, nothing is being tested.  Nevertheless,
>  we are still doing science.

No, nomenclature.

>  For example, if somebody finds that _Hesperosaurus_ and _Stegosaurus_
>  are sister taxa, and consequently decides to lump them in the same
>  genus, they must first build a case using the available evidence for
>  regarding _Hesperosaurus_ as a species of _Stegosaurus_.

Not under the rank-based codes.

Remember, "genus" is not defined. There isn't even a way to build a 
case. The ICZN gives you the right to _simply claim_ they're the same 
genus; as long as you're following the ICZN in all other respects (such 
as providing a diagnosis for the new & improved *Stegosaurus* -- such a 
diagnosis does not need to be anywhere near correct, BTW), this is a 
valid nomenclatural act.

And then everyone (yourself included) has the right to call you full of 
shit and separate the two genera again, for any or no reason.

And both opinions are equally valid forever as far as the ICZN is concerned.

If you believe that genera should be synonymised if they taxa they used 
to describe are sister-groups, and if (...whatever...) the morphological 
diversity in the whole clade is below whatever limit you wish to define 
or not define _or_ keeping them separate would make the classification 
unmanageable _or_ whatever, and if you _then_ make a case that *H.* and 
*S.* fulfill _those_ criteria, that's fine, but you're satisfying 
yourself here, not the ICZN. The ICZN couldn't care less.

Phylogenetics is science. Nomenclature is not, it's a set of arbitrary 
conventions.

>  So although it's not testable (because ultimately it is a subjective
>  decision), I would still regard it as 'doing science'.

It's not "ultimately", it's a subjective decision from start to finish, 
because "genus" isn't defined, and "*Hesperosaurus*" and "*Stegosaurus*" 
are incompletely defined ("whatever genus contains *H. mjosi*", 
"whatever genus contains *S. armatus*"). All, literally all, the ICZN 
does is tell you which names to sink if you declare them synonyms. It 
does _not_ say _anything_ about which reasons you might have for 
declaring anything synonymous. You're in all seriousness allowed to do 
as you please.

Scary, I know. But that's what rank-based nomenclature is like.

> > As soon as you can't answer the question "if I were wrong, how
> > would I know?" any longer, you're not doing science.
>
>  That's a little harsh.  Some scientific questions have no "right" or
>  "wrong" answer.  Is _Brontosaurus_ the same as _Apatosaurus_?  The
>  evidence suggests "yes";

The evidence can't suggest anything, because "genus" is not defined. OK, 
science can in principle tell us (even if only by parsimony arguments) 
if *B. excelsus* and *A. ajax* are, say, different growth stages or 
different sexes or individual variation of the same... the same... the 
same LITU, but that basically is it. If they're separate species by 
whatever species concept you might prefer*, there is no way evidence 
could suggest or not suggest they're the same genus.

Unless you define "genus" first. Again, the ICZN has never even tried to 
do that.

* It goes without saying that the ICZN doesn't say anything about 
species concepts and couldn't care less.

>  but we're never going to KNOW for certain.

If you want to go down that road... we can't know anything for certain. 
It's not even possible to disprove solipsism.

================================================

Mike Keesey wrote:

>  what is science, if not a body of knowledge gleaned from tests?

The methods of testing themselves... science is something you do, an 
activity, not a collection of facts.