rankless nomenclature
Richard Pyle
deepreef at BISHOPMUSEUM.ORG
Wed Oct 11 10:44:19 CDT 2000
I've been fascinated over the years by the awkward dance between the
disciplines of nomenclature and phylogeny; the aged former being
characterized by long-held tradition, and the youthful latter being
characterized by ever-dynamic methodologies and interpretations. Whenever
I've participated in discussions/seminars/arguments about this awkward
dance, I pose a simple question, which usually leaves the dialog
participants somewhat befuddled:
"Why do we assign names to taxon groups?"
The obvious answer, of course, is to facilitate communication. But more
specifically, the question relates to the issue of what the nomenclatorial
system should represent and reflect. The unspoken assumption among modern
researchers is that the naming system should reflect evolutionary lineages,
without regard to subjective human interpretations of morphological
similarity. This seems a safe assumption, because it more or less is the
way taxonomy has been practiced all along -- even since before the whole
concept of evolution was introduced. For quite some time now, bats and
whales have been lumped with mammals, rather than with (respectively) birds
and fishes. The reason for this is that taxonomists recognize that
"similarity" is revealed in the details, more so than in overall gross
design. This practice has continued to the present, with the "details"
becoming ever finer in scale (currently down to the fundamental scale of a
DNA sequence). At some point, our ability to resolve the "details" has
allowed us to recognize evolutionary patterns that defy even the most
careful external examination. So much so that we now are faced with
phylogenetic relationships that, when applying strict rules of monophyly,
leave us with a system of nomenclature that seems at times almost
nonsensical to a human observer.
There is a dark side to this progression towards finer detail and cladistic
methodology, which I believe is a large component underlying all the
controversy between traditional taxonomy and phylogenetic approaches to
biological classification. That dark side ultimately boils down to
overconfidence in our ability to use these new techniques to accurately
interpret and detect phylogenetic patterns. The "confidence" part comes
from the fact that the newer methodologies are fundamentally more objective
than the more subjective classical approach, and in science, objectivity has
consistently led us to more satisfactory and convincing conclusions. The
"over" part comes from the fact that cladistical application is imprecise,
and does not necessarily give us absolute answers. This is demonstrated by
the many, many examples of amply-qualified cladistical taxonomists arriving
at dissimilar phylogenetic interpretations of the same group of organisms,
based simply on examining different sets of "details".
So what's my point? Even though practitioners of Linnaean classification
have virtually always strived for names that reflect the details more than
gross external design, the nomenclature was intended to represent how humans
perceive organisms. This system of nomenclature works *almost* well enough
for the new phylogenetic way of thinking, that as phylogeny overtook human
perception as the underlying pattern that should be reflected in the names,
the nomenclatorial system as been contorted to represent a pattern of
evolutionary phylogeny (a recent Gould essay in Natural History dealt with
issues of the Linnaean classification in the context of patterns of
evolution). The "overconfidence" effect has led to an overall decrease in
stability of names in recent decades, much to the chagrin of taxonomic
researchers who follow a more classical approach.
The bottom-line point of contention comes down to two questions: 1) How
rigidly the nomenclature should reflect strict monophyly in cases where
doing so creates groupings that are nonsensical to most observers?; and 2)
Which is more important to the field of taxonomy: stability of names, or
ensuring that current nomenclature reflects the phylogeny du joir.
Recognizing that "almost" only counts for Horseshoes and Hand Grenades,
PhyloCode has been developed to address the fact that the traditional
classification system doesn't quite meet our needs in reflecting
phylogenies. This happens to also give us a wonderful opportunity to dispel
with all the aforementioned controversy. The PhyloCode system, as I
understand it, embraces the idea that phylogenetic interpretations change
as new evidence becomes available, and as the whole body of evidence is
re-interpreted by different systematists. It will provide a more effective
language of classification to the thousands of researchers interested in
sleuthing out phylogenetic relationships among life forms. It will also
allow the return of the Linnaean system to its original purpose as used by
millions of other humans: grouping organisms based on intuitive perceptions
of similarity. Linnaean "nomenclature" is intended to reflect human
subjective interpretation, whereas PhyloCode's "numenclature" is intended to
represent the objective phylogenetic relationships among organisms (to the
extent that we are able to accurately determine them from our ever-expanding
ability to scrutinize the details).
Now, one final point: Before I get attacked for my use of the phrase "human
subjective interpretation" as applied to the Linnaean classification system,
be aware that I understand that the Linnaean nomenclature should, in the
vast majority of cases, directly reflect PhyloCode numenclature; simply
because evolution *usually* produces phenotypic groupings (I'm not talking
about applying different names to males and females of sexually dimorphic
species, or anything like that). However, discrepancies will exist
primarily for two reasons: The first is that it just plain makes sense to
the vast majority of people in the world to draw lines at certain grades,
rather than at clades (e.g., birds/reptiles, whale sharks/nurse sharks,
etc.). In other words, Linnaean nomenclature should be allowed to
"recognize" extreme divergence of form, even when doing so violates strict
monophyly. The second reason is that the Linnaean system abhors
instability. For this reason, I envision that nomenclature will generally
lag behind numenclature, until the numenclature has achieved some sort of
stable ground. At that point, the nomenclature can then be changed to
reflect the numenclature (where appropriate) *once*, rather than flip-flop
multiple times while the phylogeny is still being worked out. There should
also be no difficulty in mapping the nomenclature to the current
numenclature. The reason is that every taxon name in the Linnaean system
(valid or synonymous) should ultimately tunnel down to a single type
specimen (i.e., species have type specimens, genera have a type species,
families have type genera, and so on). Each specimen should also fall
within a defined numenclatorial clade. In both systems (Linnaean
nomenclature and PhyloCode numenclature), the type specimen is ultimately
what anchors the name to the populations of critters and weeds. Those type
specimens allow cross-mapping. In the end, everybody is happy.
I'll shut up now.
Aloha,
Rich
Richard L. Pyle
Ichthyology, Bishop Museum
1525 Bernice St., Honolulu, HI 96817
Ph: (808)848-4115, Fax: (808)847-8252
email: deepreef at bishopmuseum.org
"The views expressed are the authors, and not necessarily those of Bishop
Museum."
More information about the Taxacom
mailing list