Real Species

[pierre deleporte]

At 12:35 20/04/2004 -1000, Richard Pyle wrote:
>Pierre Deleporte wrote: > Individuals are rather well circumscribed 
>biological systems in
> > themselves.
>
>R.P. Yes -- certainly more so than populations or taxonomic clusterings.  But,
>even individuals are dynamic entities (cycling from a pair of gametes,
>through a succession of material turn-over, through entropy and
>dissolution).  Subtly so in mega-fauna; less subtly so in colonial and other
>more "ambiguous" individual organisms.

Agreed. But I suggest to distinguish:
- systems: individuals, colonial organisms (= cohesive biological systems), 
populations and social groups (= a number of really interacting 
individuals, these interactions constituting the population or group in a 
"real system" out there),
- from taxa: whose only "cohesion" is that all the individuals fit our 
definition for being member of the same taxon. Taxa don't exist outside our 
classification, while interacting groups or populations do.

Thinking of "reality" in terms of concrete systems is of some help.
Fuzziness is one thing, being a real system or a human construct is another.

>R.P. species (or any other taxonomic or other aggregate unit -- including
>subspecies, populations, etc.) are "defined", not "discovered".

Again, a population of interacting individuals may be said to be discovered 
(given a clear definition of "interactions"...! see R.P. below). Because if 
you don't study the population, the individuals go on interacting. Thus a 
population should not be confused with a "subspecies", which includes 
non-interacting individuals of the same "subspecies".
Some "aggregate units" you list above are classes, while a cohesive 
population is a system which can be "discovered".

>But even
>here we have a semantic thorn, because given a definition of "species", one
>could argue that boundaries between those defined units are "discovered", by
>way of acquiring information relevant to the definition.

This is a crucial point. There is possibly a semantic thorn, but not an 
ontologic one.
There is nothing wrong in "discovering" members of a class according to 
one's criteria (and being conscious of applying these criteria). We cannot 
class another way. But why not call it classifying (with more or less 
problems of fuzziness) rather than "discovering", for sake of clarity? If 
"I discover classes" means exactly "I class", there is redundant, and 
possibly misleading, vocabulary. I may "discover" that my classification 
conventions work well or badly for some specimens, but this is obviously 
linked to my criteria.

On the other hand, what I call positivist illusions in classification is to 
consider that there exist "self-evident classes" in the world out there 
independently of our classification criteria. Biological systems are more 
or less self-evident (i.e. materially well-delimited): individual, coherent 
social group, population... but classes are not systems, they are human 
constructs by definition.

I suggest accordingly that we try to clearly distinguish "population" from 
"subspecies" in the discussion. Discovering a cohesive population is not 
the same thing as classifying specimens into a subspecies with or without 
fuzziness. Not at all. If it were the same, then let's call it a 
population, not a "subspecies", and find another name for specimens grouped 
together in terminal taxa while they don't belong to the same "real" 
populational system of contemporaneous, living, effectively interacting 
individuals.

>To be fair, I believe the implied meaning of "real" in this context is that
>they (taxonomic entities like "species") exist and are identifiable outside
>the scope of "artificial" (=created by humans) definitions of taxonomic units.
>I.e., they are "discovered" in nature; rather than defined.

See positivist classificatory illusions above. I admit that there is little 
"scientific discovery" without scientific theory (again contra naive 
positivism), while scientific realism means that we believe that there are 
real, material systems out there.

But species are not systems, they are classes, they don't "exist" outside 
the classificator's brain processes. This doesn't mean that they lack 
scientific pertinence. Concepts matter in science.

> > they are "useful or not useful" in a specified context,
>This is a critical point that usually gets forgotten in the various debates
>(e.g., cladistic vs. eclectic applications of Linnaean nomenclature).

Effectively... and the search for the holy grail of a unique-and-optimal 
classification may tend to mask the logical point that different purposes 
may require different optimal classificatory conventions. Accepting 
different classifications for different purposes (including different 
species concepts) complicates the picture, but reality is not that simple 
and science (and communication) must follow.

Best,
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