generic oversplitting (heart of the problem)

Thomas DiBenedetto tdibenedetto at OCEANCONSERVANCY.ORG
Thu Jul 19 11:27:08 CDT 2001

I agree with the ideas that have been expressed here regarding the
importance of stability and usefulness in classifications. The advent of
molecular systematics has led to the publication of a great many
phylogenetic analyses that are restricted to single genes or parts of genes,
and often the authors of these works have advocated revisions to
classification based on their results. This has been rather unfortunate.
Perhaps there has been an irrational exuberance about the reliability of
genetic evidence, or a lack of firm foundation in the principles and
standards of systematics. I am of the opinion that changes in classification
should only be proposed by those who present the results of major
revisionary works - works that include all species in the particular group,
and that incorporate all the character evidence that is available in the
literature for that group. I think this standard would go a long way to
dampening nomenclatural instability.

I also feel that the cladistic approach is likely to be the most stable
system in the long run, even if there may be a period of instability in the
short run. The cladistic approach establishes as an ideal standard something
which is real and unchangeable - the actual real-world history of lineages.
As with all other scientific enterprises, we can expect that with rigorous
application of our methods, there will be a continual process of convergence
on an accurate representation of that reality. This process cannot be
expected from eclectic or hybrid classification schemes since by their very
nature there is a core of subjectivity to their arrangements. If all
scientists in the world could agree on a single authority (you interested
Ken?) then perhaps this problem could be avoided, but lacking that
agreement, there is nothing to prevent anyone else from coming up with a
competing system - and thus increasing confusion and instability for all. I
think that it is only by establishing the real lineage history as the
standard, or more precisely, the most convincing set of arguments based on
evidence for a particular history, that we can arrive at a reasonably stable
and maximally accurate classification.

Tom DiBenedetto

More information about the Taxacom mailing list