Paraphyly: A "Real" Saga (thought experiment)

Ken Kinman kinman at HOTMAIL.COM
Wed Jan 16 15:04:23 CST 2002

Dear All,
     I am so glad that we are getting this all out into the open and
discussing it thoroughly (especially the theoretical issues).  This is very
important, so I want to make my position as clear as I can with a sort of
thought experiment.
     Within the vertebrate branch (of the true tree of life), I think it is
widely accepted that stegocephaly (with tetrapody following soon after)
occurred once.  Much later one subbranch developed the amniotic egg
(allowing the amniotes to greatly diversify and "conquer" the land, just as
insects had done).
     By cutting the tree of life at these two important events, we create a
Class Amphibia which very clearly distinguishes it from both its ancestors
and its descendants.  While *any* cut on the tree of life is going to be an
arbitrary human decision, these two particular cuts appear to minimize the
degree of arbitrariness.
     Therefore, although even this particular delimitation of Amphibia
inevitably has some degree of arbitrariness, it is still a "REAL" section of
the tree life.  Whether you want to call Class Amphibia a paraphyletic
group, or by the rather pejorative term "grade", is up to you, but it is
still a "REAL" taxon.  You could argue against this philosophically, much as
you can argue that a solid object isn't really solid because it is mostly
empty space on the atomic level.  But that is just semantics.
      Likewise, a species is also real, even though its delimitation
(especially in the time dimension) is unavoidably arbitrary.  Unavoidable
because evolution is basically a continuous process, there's no getting
around that.  Ring species also demonstrate the arbitrariness of delimiting
species geographically, but this arbitrariness doesn't mean that they aren't
      Therefore, I have never understood the notion that species are real,
but higher taxa (bigger pieces of the tree of life) are not real (this makes
no sense to me at all).  What is true is that delimiting higher taxa is much
more difficult and arbitrary than delimiting at the species level.  But
cutting the true tree of life in two places will ALWAYS yield a "real"
paraphyletic group (like Class Amphibia).  With minimal arbitrariness, you
can justify some formal paraphyletic groups (like Amphibia), not only as
being useful, but real and natural.
     And if a population can cross a mountain (or any other type of
"barrier") and separate from its mother species paraphyletically, that
daughter species can occasionally diversify to create larger lineages which
we call higher taxa, and these taxa are real whether they are paraphyletic
or holophyletic.
     And FINALLY, and most importantly.  The tree of life has an origin, so
any cutting of the tree of life above the origin automatically creates a
real paraphyletic group.  Therefore, although it can certainly be
cladistically argued that we should *minimize* formal paraphyly (given the
incredible lack of data from extinction and scant fossil preservation), I
believe that it is logically impossible to justify the complete elimination
of all formal paraphyletic groups.  Cladistic conventions can help us deal
with lack of data, minimize arbitrariness, and maximize utility, balance,
and heuristic predictability----but take it to the extreme of no formal
paraphyly at all, those conventions will inevitably destroy these very
qualities.  Zero formal paraphyly is zero optimality (and thus a waste of
time in the long term).  We should instead be spending our time discussing
how much formal paraphyly is optimal.
            -----  Ken Kinman

Chat with friends online, try MSN Messenger:

More information about the Taxacom mailing list