Cladistics and "Eclecticism"

[Pierre Deleporte]

At 10:44 05/02/2002 +0100, Peter Hovenkamp wrote:

>1. If a group of people (say, a mixed group of 100 men, women and children)
>embarked on a spaceship and were to settle on Mars - would that effectively
>split Homo sapiens into two species (assuming contact would be scarce and
>limited to raw materials, not frozen sperm etc.) the moment the spaceship
>took off? (...)

Hi Peter and taxacomers,

Maybe we could stick to the way systematics are performed, effectively ?
We don't deal with direct observation of populations history, but with 
specimens carrying characters.
The population split evoked by Peter does not provide different characters, 
thus there is no ground for classifying the two sets of individuals as 
different terminal taxa, from a systematists point of view (different from 
the ecological point of view, as reapetedly underlined by others in this 
list: conservation ecology may be concerned by the two populations).
No difference in characters, no concern for classification.

A problem seems to arise if only one population evolves, and the other 
remains exactly the same in all its characters: should we apply the 
cladistic convention of giving them both terminal taxon status?
But in the real life, we did not observe the evolution of the populations. 
Then we are facing two different taxa, because they carry different 
character sets. One of them could be the strict copy of the "ancestor" of 
the other one. Anyway it is not properly the ancestor, the two populations 
being contemporaneous.
Is there any problem in naming them as sister terminal taxa, that is 
"closest contemporaneous relatives", identified as such by their 
synapomorphies, and differing in a series of characters?
In the present case, some symplesiomorphies of the two taxa could be 
mistakenly taken as autapomorphies of the "unchanged" terminal taxon. 
Correct, but is there a remedy to such a mistake, given than we have no 
time machine to go back in time and observe the evolution of the ancestral 
population? There is no label "false autapomorphy" written on the character 
of the specimen at stake. Thus, is this an issue at all for systematics? 
Misleading data may mislead us, and what can we do to that?
I think that the key point is that two sister terminal taxa simply may, or 
may not, be two "equally diverging" taxa; one of them may, or may not, be 
the more or less strict copy of their exclusive common ancestor, and maybe 
we have no way to know that. Where is the problem ? Just agreeing on the 
full acception of the naming convention: a sister group, i.e. a 
contemporaneous closest relative, may effectively be unchanged since a long 
time in some characters. Coelacanths do exist, as a sister group nearly 
unchanged in some morphological traits.
For a phylogenetic classification, the key feature for the two sister taxa 
is their separation with no back crossing, not their differential 
evolutionary rate.

Identifying reticulate evolution is a different problem, as well as 
conventions for naming taxa in case of reticulate evolution. We could have 
a discussion focusing on this point specifically.

Identifying an ancestor taxon in fossils is still another problem, well 
debated in the cladistic literature: there logically cannot be positive 
evidence of being an ancestor in the characters of fossil specimens (only 
lack of synapomorphies in the known characters), thus some specimens may be 
placed either as terminal taxa, or pulled down at the internal node of the 
cladogram as possible ancestors. If placed at the top of a branch, this 
branch will carry no synapomorphies, but this graphic representation could 
perhaps be worth translated in the naming, with a marker indicating this 
ambiguous status "sister or ancestor".

But can't we separate the questions, and deal specifically with problems like:

- possible difficulties to infer the phylogeny (including possible 
reticulation): not directly a problem of naming conventions;

- rules for naming taxa in case of doubtful phylogenetic resolution: should 
we better urge the change in naming or name content, or wait and possibly 
use a marker indicating doubtful phylogenetic status of a taxon?

- problems with possibly adjusting classic taxon names to new discoveries, 
e.g.: where should "birds", or "Aves", begin now? And why not keep their 
ancient acception ? Unless we would consider that an ideal pre-defined 
"bird" taxon should be decifered in nature, changing the content of "Aves" 
when new close relatives are discovered is perhaps not a necessity (think 
of stability, Ken!  Maybe it is not forbidden either, but then let's stop 
arguing of stability).

- conventions for naming taxa in case of a well documented reticulate 
phylogeny: botanists could provide nice real examples here. How should we 
name such things? With markers?

- conventions for naming taxa in case of a well resolved, strictly 
dichotomous phylogeny. Do eclecticists have a general criterion for 
defining a relevant (or useful) paraphyletic group? (better answered by 
eclecticists). If yes, does this criterion applies to non-bird dinosaurs 
for example?

- is eclecticism (implementing simultaneously two or more incompatible 
criteria) a problem in itself for a classification? For what practical 
reasons? What do we gain, what do we loose, compared with a nested 
hierarchy of taxa based on a unique criterion?

-  if eclecticism is a problem, and if some paraphyletic groups are worth 
naming anyway, why not call such groups "non-something" taxa (like 
"non-bird dinosaurs" or "non-human primates"), and everybody is happy? 
(exactly a complement to the Kinman system: names for holophyletic groups, 
with "non-something markers" for paraphyly, while the Kinman system uses 
names for some paraphyletic groups, with markers for providing the 
corresponding phylogenetic information). Exactly the same information 
content may be conveyed with and without eclectic definiton of names, hence 
my insistence in discussing eclecticism in itself.

- the "too many names" issue: apparently nobody suggests to name all 
clades. Not all possible paraphyletic groups either. Choosing which groups 
to name seems a practical question, not "natural" or "scientific". But is 
there any limit, besides imagining the names? Is this an issue at all?

- still another question is the subjective feelings facing some conventions 
for some groups. Just a test:
   * Humans are primates, and of course they are not non-human primates.
   * Birds are dinosaurs, and of course they are not non-bird dinosaurs.
These formulations convey information about both monophyletic and 
paraphyletic groups, and they do not hurt any of my non-scientists friends, 
nor even most of my scientist colleagues. Given that the information 
content is not ambiguous, why should they hurt some people ? (better to be 
answered by people feeling hurt or uneasy with such conventions).

This series of questions is not limitative, I simply suggest that shifting 
from one topic to the other is a very efficient way (intentional or not) to 
empede a thorough debate on any given point.

Not to mention questions of cloud castles, pedantic head banging from top 
of ivory towers, giving somebody a break, good ol' fashioned criteria, 
rafts and rocks, and apocalyptic prophetia...
(sorry, I fear I'm getting somewhat trolled   ;-)

Cheers,
Pierre



Pierre Deleporte
CNRS UMR 6552 - Station Biologique de Paimpont
F-35380 Paimpont   FRANCE
Téléphone : 02 99 61 81 66
Télécopie : 02 99 61 81 88