[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: FW: Re: FW: Re: FW: Re: FW: Help with Cladistics

In a message dated 95-08-28 22:41:09 EDT, grenard@herpmed.com writes (quoting
Chris Brochu):

>1.  Such groups are inherently arbitrary and subjective.  Birds share a
>common ancestor with dinosaurs.  A natural group defined by their common
>ancestor thus exists.  The name we choose to fix to this group is
>arbitrary, but the group itself exists regardless of our ability to recover
>it.  The practice of separating groups on the basis of some "key character"
>is very subjective - who, after all, is to decide which characters are
>"key?"  Why are feathers so much more important than a wishbone or a hollow
>skeleton - characters that are arguably requisite for avian flight, but
>which are distributed more generally among theropod dinosaurs.
>2.  We will always be finding fossils that smear the transition between
>such groups.  When the hands and feet of Ichthyostega were found back in
>the '80's, they were seen to have 7 or 8 digits and to resemble fins more
>than legs.  Was it still so obviously a non-fish?  As our knowledge of the
>fossil record improves, our ability to resolve paraphyletic assemblages

It is impossible to remove subjectivity from taxonomy, even in cladistic
analysis. Cladistic analysis merely spreads the subjectivity around,
fostering the illusion that it has been entirely removed from the
classification process. But subjectivity exists when a decision is made about
whether a feature should be scored as a character; it exists when a decision
is made about how much weight to give to a character, or in deciding whether
all characters should be equally weighted; and it exists when a decision is
made about whether a character should be scored as primitive or derived. And
how, for example, should a feature such as "tail reduced" be scored? As one
character? Or as, say, 25 characters (one for each caudal vertebra lost)? Or
as some intermediate number of characters, say 5 or 10 or 13?

At its base, even cladistic methodology acknowledges the existence of
paraphyletic taxa: Any species that speciates is by definition paraphyletic.
Suppose a small subpopulation of a species becomes stranded on an island
somewhere and, through isolation from its parent species becomes a distinct
daughter species, whereas the main population of the parent species continues
unchanged. Do we consider the parent species as it exists after the
speciation event to be a different species? If not, why not? Its members
share a closer common ancestor with the isolate daughter species than with
any members of the parent species prior to the speciation event. But if so,
why? Why should the subsequent history of a small, isolated subpopulation of
a species have anything to do with how the main population is classified?

If cladistic analysis weren't subjective, we wouldn't have a dozen different
competing phylogenies from a dozen different competent workers on the same
group of organisms. Given a competent analysis of a group, we would find only
the One True Phylogeny.

Subjectivity and paraphyletic taxa can be kept under control by a consensus
of the scientists who are engaged in studying the organisms being classified.
We should acknowledge the arbitrary and subjective nature of any system of
classification that divides into discrete segments or categories the
continuous one-dimensional branching manifold that is Tree of Life. And then
we should make the leap: decide which will be the one key character that in
any instance defines the point where a genuinely new taxonomic level has been
attained, and work with it.