naming and types

James Lyons-Weiler weiler at ERS.UNR.EDU
Sat Nov 9 08:28:31 CST 1996


On Sat, 9 Nov 1996, JOSEPH E. LAFERRIERE wrote:

>
>
> Dear Dr. Weiler,
>    Thank you for your thoughtful comments. I find, however, that
> some of your arguments miss my point while others actually
> support my position.
>    First, I never said non-European societies do not possess
> folk taxonomy. With my experience in ethnobiology, I am very
> well aware of that. The fact that every known language on Earth
> has a folk taxonomy very similar to the European model bolsters
> my contention that the categorization of nature is rooted in
> human sensory input patterns.

That's one interpretation.  The other interpretation is that the
patterns are real; i.e., that a species level of of diversity tends to
exist independent of humans, and that cross-cultural similarites is
classification modes reflect that reality.  I'm not for one or the other
position here.

 >    Second, I am baffled by your comment that I am
"killing > myself," or that it means taxonomists should be separated into
> dusty old museum separate from other biologists. The truth is
> that yes, there is a difference between nomenclature and other
> aspects of biology. Nomenclature is not a science at all, and
> never has been. It is far older than most other aspects of
> science, having a very different historical origin. We have
> very wisely decided to make nomenclature reflect the course of
> evolution, thus making an interplay between nomenclature and
> science inevitable.

There is an astounding correspondence between what we call things and they
way we think about them and react to them.  I could phrase the problem in
as an exercise in hypothesis testing:

The null hyothesis is that there is no significant difference between
populations A and B with respect to (morphology? no) _biology_.

What I take your position to be is an argument that restricts the type of
information that can be used to reject this null.  It's quite analogous in
fact to experimentalists' convention of requiring p-values of less than
0.05 before a null can be rejected; you're arguing for a convention that
diagnostic macrocharacters need to exist before society can find any
utility to distinguishing by name species that indeed are on separate
trajectories.  I would argue that at the very least a contextualist
rationalist approach would be appropriate; but in addition, a falsehood is
conveyed when separate species are labelled as identical.  The negative
consequences of failing to convey
accurate information may also depend on circumstance, but sometimes a Type
II error is more costly than a type I error.  (Type I error = failing to
reject a null when it is true (saying they are different species when in
fact they are not; Type II error = rejecting a true null = saying the
species are the same when in fact they are different).  Type IIs result in
ignorance; Type Is result in oversplitting a taxonomy.  Which is worse?  I
would say it depends (contextualist rationality); others would say the
onus is on the taxonomist to make sure that the negative consequences of
each are considered, and both should be communicated.  The argument hinges
on how one defines a "significant difference".  You might (?) say that a
significant difference is one we can diagnose.  I would disagree; a
significant difference is biology may happen to produce diagnostic
characters, or it may not.  Just because verification would be made
difficult should not cause one to prefer ignorance over identification
error.

 >    Third, the first four of your five points could be handled
> without assigning Latin names to the "cryptospecies" involved.
> As for your fifth point, i.e. the idea that molecular studies
> may help morphologically oriented systematists to pay closer
> attention to minute morphological differences between taxa,
> this supports my premise precisely. If you can find consistent
> morphological differences, go ahead and write it up. Fine.
>    Fourth, I am gravely disappointed that you do not want
> any pins, as I could use the money.
>
BTW, I'm not a Dr., but I'm working on that.

James Lyons-Weiler




More information about the Taxacom mailing list