[Taxacom] Rummaging in the basement

Bob Mesibov mesibov at southcom.com.au
Tue Oct 9 00:07:03 CDT 2007

The barcoding discussion last week, and especially Rich Pyle's excellent
contributions, seemed to me to 'loosen up' thinking on the purpose of
taxonomy - thinking that's been rigid for a very long time.

IMO the rigidity set in with Darwin. There had been a number of workable
classificatory systems before his time, but Darwin said that a system
needed to be genealogical. A natural classification should be structured
like the true evolutionary history of Life.

We are much better today than we were in Darwin's time at coming up with
defensible evolutionary hypotheses. We can generate hypothetical
phylogenies quickly and easily from all sorts of character data. In
fact, it is much easier to produce a hypothetical phylogeny than it is
to produce a classification, because not everything in the phylogeny
(even for PhyloCoders) needs to be named, and there is no Phylogenetic
Code. Even poorly supported trees get published.

To go one step further, *nothing* in a phylogeny needs to be named in
the taxonomic sense. A phylogeny consists of a tree diagram with
terminals representing individual specimens, or small groups of
specimens. To call these specimens 'exemplars' of 'species' or other
taxa is to generalise unnecessarily. The process of generating a
hypothetical phylogeny does not require 'species', 'genera' or any other
intellectual constructs. You get the same tree diagram using simple
code-names (museum registration numbers will do) for the individual
specimens you analysed.

We can all think of many uses for taxonomy that have nothing whatsoever
to do with evolutionary hypotheses. I assume these are the
'communication' uses that Rich was referring to. While it would be nice
and consistent for the taxonomic system and the phylogenetic system to
be congruent, precisely what scientific purposes are served by insisting
on congruence?

To answer my own question, one is to help taxonomists assign new
material (after doing a character analysis) to an existing taxon, or to
justify establishing a new one. The closer the congruence between the
taxonomic hierarchy and the tree hierarchy, the easier this job becomes.

Anything else? How about the people on this list who believe that the
overarching mission of biology is to reconstruct the one true Tree of
Life? How are you aided (if you are) by the notion that taxonomy should
be congruent with phylogeny?

And how about any barcoders out there? If we can (in theory) lay out
unique identifying nucleotide sequences for every individual or every
group of genetically identical individuals, to what extent does that
make taxonomy redundant? How many of the 'communicating' uses of
taxonomy could be taken over by UIDs, instead of a hierarchy of names?

Disclaimer: I am not 'against' taxonomy, or constructing evolutionary
hypotheses, or trying (Darwin-fashion) to aim for congruence, or
barcoding. I'm also aware that a character analysis can be thought of as
producing a robust, defensible classification *only*, and that *some*
people think that if you call the tree a history, you're an idiot.
Dr Robert Mesibov
Honorary Research Associate, Queen Victoria Museum and Art Gallery
and School of Zoology, University of Tasmania
Contact: PO Box 101, Penguin, Tasmania, Australia 7316
(03) 64371195; 61 3 64371195

Australian millipedes checklist
Tasmanian multipedes
Spatial data basics for Tasmania
Biodiversity salvage blog

More information about the Taxacom mailing list