3TA (correction)

[pierre deleporte]

At 15:28 16/04/2003 +1000, Don Colless wrote:

>This is a neat example of the "character = part" usage, which of course
>doesn't mesh at all with my usage. And of the curious metaphysical
>notion, that the fin is still present in the limb. I recall an argument
>with Gary Nelson, who insisted that a creature descended from elephants,
>but which lacked a trunk, REALLY still had a trunk. O.K., if it makes
>you comfortable to think such things, who am I to object - just as long
>as the taxonomy still gets done.

This remark of Don Colless remembers me that in a recent thread on this 
list about "natural classification", I happened to be wrong in arguing (in 
the beginning) that biological systematic groups (clades...) were 
"classes". They are not, according to the classic definition of classes, 
because of the point raised above by Don Colless: features inherited by 
members of a clade can further transform i the course of evolution and no 
more be similar, nor even present.

I later realized that Mahner and Bunge (1997), although generally arguing 
for "natural classification", nevertheless aknowledged the point that 
synapomorphies can transform, and thus proposed the term "biological kinds" 
(kinds are classes of objects sharing more than one common property) to 
account for this property of taxa (clades) due to biological processes, to 
the difference of possible "natural" kinds of non-biological objects.

In a biological classification, biological taxa may thus be defined as 
possessing a feature OR descending from an ancestor biologically inferred 
as having possessed the feature. This is different from arguing that a 
trunkless elephant "really possesses a trunk": in fact the poor animal 
merely descends from a trunked ancestor, which biological property can be 
taken into account for biological classification.

To make those elephants (trunked and trunkless) a "biological class" based 
on a single shared property (e.g. "trunk"), the notion of "possessing a 
trunk OR descending from a trunked ancestor" is necessary, and sufficient. 
Implementing this notion requires considering and phylogenetically 
analysing characters other than "presence of a trunk in the pheneotype" (I 
don't develop this point, which could involve many sources of biological 
assessment) in order to fit the trunkless species inside elephants in the 
first place.

I suggest that this makes biological sense viewed this way, that there is 
no metaphysical notion involved (e.g. that trunkless things "really" carry 
a trunk), and that, more generally, biological processes producing 
biological objects can legitimately suggest corresponding biological 
classificatory logics and conventions.

I see no reason that a unique, universal classificatory convention should 
fit all possible goals and needs. I would nevertheless demand that goals 
and needs (including background knowledge, theories and assumptions) are 
clearly stated, unless no appreciation is possible of the relevance of a 
classificatory logic... hence some lasting, pure dialogues of the deaf, 
where "why" questions are systematically given "why not" answers.

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