[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

RE: ...and how ranks don't work was Re: Sauropodz r kewl



> > Rhynchocephalia and Squamata are sister-groups. To put
> *Sphenodon* (and 
> > the extinct rhynchocephalians) into Squamata would just
> mean we'd 
> > confuse everyone and need to come up with a new name
> for what we used to 
> > call Squamata. Note how I wrote this without mentioning
> a rank once.
> 
>  
> 
> Oh I understand how that's possible for a discussion between
> people who both know what the names imply.
> 
>  
> 
> It's like if we held conversations in German or in Damin -
> we would each know what the other is saying and
> meaning...but if we want to tell anyone else about what
> we're talking about and working on, we would have to use a
> language they know: ranks.

It may be significant to remember that even since before Darwin, the 
nomenclatorial problem (ranks vs no ranks) is secondary to the problem of 
determining the shape of the tree/(before Darwin)relationships between groups.

Consider de Blainville's "Secondates", "Tertiates" and "Quaternates", Vigors' 
"Grallatores", "Insessores", "Natatores" etc, or just *any* taxonomic scheme 
that was ultimately rejected.

The problem of taxonomic relationships/validity is neither independent from nor 
identical to the problem of nomenclatorial terminology. Given that the true 
tree can only ever be inferred, some problems associated with rank-based 
nomenclature (merely for reasons of tradition - there has only been rank-based 
nomenclature until fairly recently) won't go away by changing the 
nomenclatorial paradigm. Even with the best methodology, incompleteness of 
phylogenetic information will occasionally result in the erection of 
paraphyletic or polyphyletic taxa. No matter whether ranks are used or not.

Consider: ever since Darwin, monophyly has been (at least to most researchers) 
the prime criterion by which to judge a taxon's validity. If not explicity so, 
then implicitly so: even before Hennig, taxa were rejected on grounds of 
nonmonophyly. That the problems of rank-based nomenclature have been noted at 
all is a direct consequence of this. A non-rank-based nomenclature can fix the 
semantics, but that's all.

I prefer keeping the two problems as separate as possible. On one hand the 
problem of nomenclatorial precision (unranked wins) vs ability to communicate 
to nonspecialists (ranked presumably wins*). On the other hand, the 
methodological problems of taxonomic inference, which equally apply to rank- or 
nonrank-based nomenclature if we accept evolution as a fact (as we do).

In the end, the core problem with ranked clades exists for unranked clades all 
the same, it is only a bit less visible: A clade is quantitatively superior**  
to a ranked taxon or an Aristotelian ideos, but that only matters so much; it 
is qualitatively as much a fiction as are the others.
ALL TAXA ARE UNNATURAL CONSTRUCTS. You cannot draw dividing lines through a 
tree of *unbroken descent* and be factually correct. Nature knows only ancestry 
and progeny, mutation and selection. "Clades" don't exist in nature, just as 
"ranked taxa" don't exist. Unfortunately, evolution gave our species an 
inordinate fondness for thinking in neat rows and stacks of nice little square 
boxes.


Regards,

Eike

* Nonspecialists seem to find it easier if things are neatly - if wrongly - 
labeled; this would have to be tested though. Ask any specialist in any field, 
and they can rant for ages on how the public at large is incapable of 
understanding the finer points of the "taxonomy" with which the specialist in 
question works. The public at large has a hard time distinguishing things like 
iron and steel, property and possession, arthropods and insects, etc. I presume 
this cannot be fixed, because for everyday communication *in general* (i.e. 
among people that neither know each other nor share previous knowledge 
regarding the subject of communication), expediency is a more pressing 
criterion than precision.
A crude and technically wrong representation of evolutionary relationships may 
be preferrable to one that is more correct, but so unfathomable that for most 
people it would amount to "no evolutionary perspective at all". There's 
science, and there's the politics of science (and politicians who still 
misinterpret the term "evolutionary theory", among othe things), and it is the 
latter who decide on funding (or not funding) the former.

** Though not necessarily. Imagine a branch-delimited or node-delimited clade, 
for which we have a complete set of specimens from the node and immediately 
around it. They need to be stored somewhere, in a labeled box arranged with 
lots of other such boxes in a cabinet or similar. This practical problem cannot 
be solved in a workable way if taxa are strictly made to correspond to clades; 
a bunch of closely-related individuals must be sorted into 3 different taxa. 
Rather than clarifying the evolutionary relationship, it is obscured by 
artificial boundary lines. 
This is not entirely fictional; for some mammalian taxa (horses, rodents) the 
hypodigm (teeth, especially) may be comprehensive enough to make it a real 
issue. In fact, I'd wager that it *is* a real issue for some late Neogene/early 
Quaternary rodents. Even for theropods, we may have it in the case of _Corvus 
corax_ and _C. corone_ (secondary parapatry after separation/speciation in 
allopatric glacial refugia; abundant material of the presumed LCA exists from 
Mediterranean cave deposits), and perhaps also _Alectoris rufa/graeca_ and taxa 
with a similar spatiotemporal pattern.