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

Re: Adios, "Brachiosaurus" brancai

> From: John Wilkins <john.s.wilkins@gmail.com>
> Date: 2009/9/11
> Subject: Re: Adios, "Brachiosaurus" brancai
> To: mike@indexdata.com
> That would be the proposal of the PhyloCode, along with those who,
> like Nelson*, Pleijel and others, think there is nothing particularly
> interesting about species *or* genera; they are just taxa.

This is a reference to: Pleijel, Fredrik. & Rouse, Greg W. 2000.
Least-inclusive taxonomic unit: a new taxonomic concept for biology.
Proceedings of the Royal Society of London B 267: 627-630.

I am not convinced by this proposal (and I guess its uptake shows that
no-one else is, either).  From the perspective of a vertebrate
palaeontologist, the important thing about a species (or genus) is
that it is defined using ONE specifier -- the type specimen -- and the
scope is "... and everything sufficiently close".  Uber-cladists will
cavil at its lack of rigour, but when naming organisms based on often
enigmatic type material, it's exactly the concept that's needed --
using a branch-based clade definition, as advocated by Pleijel, would
lead to extreme changes in content under different phylogenetic
hypotheses.  Don't get me wrong, clades are great too (hence the PN
section in the Brachiosaurus paper), but they don't obviate the need
for alpha taxonomy.

> I think that so long as genera are understood to be arbitrary and
> conventional, they are relatively harmless, and should remain pretty
> much as the first person to name them assembled them, unless they are
> para- or polyphyletic. Arguing over genera is like arguing over how
> best to put books on a shelf.

I hear this a lot; and the obvious conclusion we're invited to draw is
that it's a waste of time.  But that is predicated on the assumption
that how to put books on shelves is unimportant.  As it happens, my
day-job is as a computer programmer in the library realm, and I can
tell you that how books are put on shelves is VERY important if you
want to be able to find them again.  Same with biological taxonomy:
these things are finding aids, and we trivialise them at our cost.

> *Nelson, of course, is a known opponent of the Phylocode, but for
> other reasons, than objecting to rankless taxonomy.

... not that I want to get into defending the PhyloCode at this point,
but I will just say that I can't understand why ANYONE would be an
opponent of the phylocode, however dedicated a Linnaeist they may be.
To quote myself from this interview:

--- quote begins ---
The Phylocode is (or will be - read on) a formal code governing the
definition of clade names using phylogenetic nomenclature, just as the
ICZN (International Code of Zoological Nomenclature) governs the
definition of rank-based names for animals: species, genera, families
and all the variations thereof (subspecies, superfamily, and so on).
The ICZN has been around for quite a while now, and has proven its
worth in enforcing a much higher quality in species- and genus-level
taxonomy. In order to establish a new taxon to the satisfaction of the
ICZN, you have to do a specific set of things: nominate a holotype
specimen, state its specimen number, give locality and horizon
information, figure the specimen, give a diagnosis that can be used to
recognise other specimens, etc. Older acts of taxonomy often didn't do
any or all of these things - for example, the Wealden sauropod genus
Eucamerotus was raised and dismissed by Hulke (1872) in a single
paragraph: he named it and immediately synonymised it with
Streptospondylus, Ornithopsis and Cetiosaurus without even hinting at
a diagnosis along the way. Thanks to the ICZN, that sort of cavalier
behaviour is no longer tolerated, and now even the very perfunctory
new-taxon descriptions that appear in extended-abstract publications
like Science and Nature have at least some scientific value.

So, the ICZN governs species and genus-level taxonomy (and families,
if you care, although for some reason not higher-level ranks, such as
order and class). But at the moment, there is no equivalent code for
defining clades using phylogenetic nomenclature (PN). [snip]  Now,
here's the problem: since there's no code governing how to do PN,
we're still stuck at the stage that Hulke and his buddies were with
their species-and-genus taxonomy. Published papers can and do contain
some pretty vague and otherwise poor definitions, and it's not even
clear what is and isn't a definition: for example, Thulborne
(1984:124), in a paper on birds, mentioned "the crown-group
(Neornithes)" in passing. Is that a PN definition of Neornithes as the
most recent common ancestor of all extant birds together with all its
ancestors? Right now, it's up to individual authors to judge the
validity of such names. (My 2007 paper tried to establish some
guidelines for interpreting this kind of thing, but it carries no
authority - it's just suggestions.)

The Phylocode is the solution to this problem. It's a relatively short
and comprehensible document - at 97 pages including a long, discursive
preface and an index, it's a fraction the size of the ICZN - which
says what you have to do to define a clade name. For example, you have
to specify the anchor taxa either as species or as specimens which are
the types of their species. It seems clear that this is a great leap
forward: even die-hard Linnaeists like Mike Benton are defining clades
in their work these days, and it seems obvious to me that it's better
to do this under the governance of a code than not. And so far as
codes governing PN go, the Phylocode is the only game in town. So as
far as I am concerned, every biologist should be firmly behind it.

Unfortunately, quite a lot of traditionalists, notably Benton himself,
are dead set against the Phylocode. So far as I can tell, this is
mostly because they perceive it as an alternative, and therefore a
threat, to the traditional rank-based codes (ICZN for animals, ICBN
for plants, etc.) It ain't so: although earlier drafts of the
Phylocode did give the impression of intending to replace the old
codes, that language is, quite properly, absent from the current
draft, and the Phylocode is now best seen as a complement to the
rank-based codes. In fact, I would argue that the Phylocode actually
needs the rank-based codes, so they can define the species that it
needs as clade specifiers: when Apatosaurinae is defined as everything
closer to Apatosaurus ajax than to Diplodocus longus, what exactly do
we mean by Diplodocus longus? For an answer to that, we need to go to
the rank-based code. Bottom line, clades and rank-based named are
completely different kinds of thing, and neither code is a threat to
the other.

(A few years ago, I wrote a manuscript called Why Can't We All Just
Get Along? which argued for this separation-of-church-and-state
between the codes. The first and second versions of it that I
submitted were quite rightly rejected - it needed work - but while I
was making the changes suggested by the second set of reviewers, the
new Phylocode draft came out and disclaimed all intention of taking
over species nomenclature, which rendered my poor manuscript almost
completely redundant. Darn.)

So to come back to the actual question you asked, how will the
Phylocode change dinosaur phylogeny and systematics? It won't change
anything that's happening now in regard to species and genus names;
and it will give a sound framework to the definition of clades that
have so far been defined in an ad-hoc way. Everyone wins.
--- quote ends ---

Well, that ended up longer than I intended.