"Fuzziness" (Continuity and classification)
kinman at HOTMAIL.COM
Tue Jul 24 15:24:58 CDT 2001
Actually utility was not originally the primary motivation for devising
the Kinman System. Heuristic considerations such as predictive value were
always given high priority, but I discovered that with a balanced approach
there is no good reason to sacrifice utility for predictive value (or vice
versa). In my introduction I state: "In order to retain both the greater
predictive ability of cladistic systems and the practicality, stability and
evolutionary-distance (anagenetic) information of eclectic systems of
classification, the use of 'semi-paraphyletic' groups has been adopted."
[If I had it to over again, I would substitute the term "semi-holophyletic",
even though it means the same thing].
And I agree with you that cladistic "techniques" are valuable in
biogeography and many other disciplines, but that is cladistic "analysis"
(which I use myself). I was talking about *strictly* cladistic
"classification" when I said the benefits are marginal at best. The Kinman
System uses coding that is primarily cladistic (main clades are numbered,
subclades are lettered), but to insist that all classification be strictly
cladistic is nonsensical, impractical, and besides it is fundamentally at
odds with the paraphyletic nature of speciation. American cladists swung
the pendulum too far the other way (i.e., overly reactionary) and have
become too dogmatic and rigid.
And as for a child being distinct from its parents, let's not take the
concept of "a species as an individual" too far. The edges (bounds) of
species are much more "fuzzy" or diffuse than the bounds of a human or other
organism. And the birth of a species is not the same thing as the birth or
an organism. We know that mother mammal and child will separate at the
umbilical cord, whereas speciation is much more unpredictable and tenuous a
process (in both space and time).
Cutting up a continuous series of fuzzy overlapping "species" will
always be arbitrary, and cladistic analysis only works as an analytical
methodology because extinction and a poor fossil record leave large gaps
(thus giving the illusion that there is no fuzziness).
With more information (morphological and genetic) and more fossil
intermediates, the fuzziness is slowly starting to show the fundamental
weakness of purely cladistic classifications, and even shows how difficult
it is to do cladistic "analysis" correctly (it's a powerful tool that can do
a lot of damage when used incorrectly). So I am still basically a
cladist----just not a strict cladist. And predictive value is important to
me, but I'm not going to let that one factor completely override the
importance of other factors necessary for what I consider "good"
>From: "John R. Grehan" <jrg13 at PSU.EDU>
>Reply-To: "John R. Grehan" <jrg13 at PSU.EDU>
>To: TAXACOM at USOBI.ORG
>Subject: Re: Continuity, reticulation, and classification
>Date: Tue, 24 Jul 2001 11:25:25 -0400
>At 05:07 PM 7/23/01 -0500, you wrote:
>> Before getting back to viruses and gene evolution, I'll quickly
>>the "human construct" issue. Any bioclassification is a construct and
>If it is 'arbitrary' does this mean that it has no predictive value about
>the nature of reality? If not then arguments over classification are
>>There is no real beginning or end to any species (especially when time is
>>considered as well as space). The tree of life is a continuity, and any
>>cuts to divide it into taxa are arbitrary.
>Depending on what one may mean by arbitrary. There is a continuity between
>myself and my parents, but a specific event may be identified to specify my
>individuality. Perhaps selection of that event is 'arbitrary' but the
>resulting designation is not (at lease I perceive a separate existence from
>my parents, as they do me).
>>Cladistic classification is just
>>as arbitrary in a different way, and many cladists believe that sister
>>groups actually exist (which I find very troubling).
>I'm not aware of 'troubling' being a criterion for the reality of
>>Its benefits are
>>marginal at best, and the drawbacks just keep getting more and more
>In biogeography at least, cladistic techniques have been more than marginal
>in helping to provide a more precise resolution of character geography.
>> Anyway, reticulation is just one more reason strictly cladistic
>>classification fails to reflect the reality of evolutionary history.
>But it can present potentially informative statements about the nature of
>This was demonstrated by Craw (1983) for biogeographic analysis.
> However, strictly cladistic
>>classification will most clearly betray its fundamental weaknesses (and
>>overly simplistic underpinnings)
>Perhaps all techniques have fundamental weaknesses?
> But a hybrid system will
>>allow some measure of stability over a wide range of detail---from the
>>evolution of organisms at kingdom and phylum levels down to minute details
>>of genetic variation among individuals.
>It seems to me that the overall theme for this system is utility, and it
>seems to me that utility is the hallmark of the arbitrary.
Get your FREE download of MSN Explorer at http://explorer.msn.com/intl.asp
More information about the Taxacom