Paraphyly is real (+ "strict cladist" defined)

Byron J. Adams bjadams at UFL.EDU
Tue Jan 15 10:15:43 CST 2002

on 1/14/02 5:22 PM, Ken Kinman at kinman at HOTMAIL.COM wrote:

> Tom,
> The above can perhaps be a useful Hennigian convention for cladistic
> analysis, but that is not what *really* happens.  There is one mother and
> only one daughter in a cladogenic event.  The daughter species has at least
> one synapomorphy that the mother species lacks (otherwise they would still
> be the same species).

What *really* happens?  No doubt this scenario occurs.  It's the logical
interpretation of this event that doesn't follow from the argument.  Shall
we recognize the mother and daughter lineages as two independent entities,
while at the same time acknowledge that the evidence for doing so is
privative?  (A privative group is one whose essence is the absence of
evidence).  That is, the evidence that allows us to recognize the daughter
lineage is an autapomorphy (I'm okay with that), yet the "mother" lineage
can be delimited only as "not the daughter".

The recognition of privative groups faces some serious problems of logic:

1.  The group is distinguished only by combinations of plesiomorphic
characters, which, for cladists anyway, are unable to inform phylogenetic

2.  The essence of the group is an absence of evidence in support of its

(Even Plato had this figured out when he asked his students to produce [in
effect] synapomorphies that would define and unite the "Barbarians" as a
group.  The best the students could come up with was that Barbarians
[basically, all the other people on the planet] could be defined only by
numerous plesiomorphic characters.  Since the Barbarians lacked
autapomorphic characters, they could be defined only as "people who are not
Greeks".)  Such groups cannot be subdivided, as it is absurd to think that a
group could be recognized based on the absence of nothing (see "Statesman"
by Plato, 320 BC).

3.  That a lineage can be at the same time a "mother" and a "daughter" is
absurd.  Sure, relationships _within_ lineages look like this, and that is
because they are tokogenetic, not phylogenetic.  If what we are looking for
is relationships among species and higher taxa, then we must look for
phylogeny.  Phylogeny produces lineages that cannot at the same time be an
ancestor and a descendent.  In the scenario you depict, the relationship
between the mother species and the daughter species cannot be distinguished
from the variation we would expect to find within or among outcrossing
populations.  If they truly are evolving independently, at some point we
should expect to see evidence of this from both lineages.

If this stems from cladists forcing logical discovery operations on nature
(what *really* happens), so be it.  It's the only way I know to delimit
these entities objectively.  The alternative is to go outside the realm of
science (hunches, crystal balls, psychic hotline, etc.,).

I'm not yet convinced that there are *real* instances where evolution
produced paraphyletic groups, or at least those that can be recovered
objectively.  And I hesitate to enthusiastically endorse those that aren't.


Byron J. Adams
University of Florida
Department of Entomology and Nematology
PO Box 110620
Gainesville, FL 32611-0620

Ph:    (352) 392-1901 ext. 132
Fax:    (352) 392-0190

More information about the Taxacom mailing list