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

Re: Definition of 'fish', & GSP is displeased ;-)



>> This may have been said many times, but to me, phylogenetics should
>> avoid names such as "Pisces" and the informal term "fish". Not only
>> because fish is used for a lot of animals which are not even
>> chordates, as David states. The same for reptiles, amphibians,
>> invertebrates, and all those non-monophyletic groups. Re-defining
>> leads to confusion, especially to new students and to professionals
>> dealing with the main public, which manage dictionary definitions.
>> Fortunately, this happened with some non-monophyletic clades such as
>> Pachydermata.

Just to add context to my previous ramble, I think re-definition is
actually a good thing for the public. Why? Two reasons for which I
think I can make a case. One, it preserves the familar name, which
helps avoid turning people off through an excess of terminology to be
remembered. Two, it actually underscores the importance of
monophyletic groups: Dinosauria had to be re-defined to include Aves,
because Aves was a component of a monophyletic Dinosauria.
Osteichthyes had to be re-defined to include Tetrapoda, because
Tetrapoda was nested within a monophyletic Osteichthyes. Elephants and
manatees had to be trimmed out of a thusly re-defined Ungulata because
they were not actually a part of the branch of interest (yes, I know
about the Pegasoferae hypothesis, just let me use my example).
Crocodiles had to be cut out of the really old-school but still
similarly re-defined Lacertilia because they turned out to be more
closely related to birds.

I think it's intuitive to the public that you have groups within which
there can be "correct" and "incorrect" members, where the incorrect
members need to be weeded out. If you have a bunch of apples and
oranges, but just want to have the apples, the solution is to remove
the oranges from the apple pile and be done with it, not to give all
the apples a new name because their previous association with oranges
no longer holds true. That's completely alien to common human
psychology. So that, I would argue, is actually a *more* confusing
practice for the layperson, because the layperson thinks in
cut-and-dried terms of "right/wrong" and "belong/doesn't belong," not
terms of "abandon familiar words when historical association could
prove confusing if any former member turns out to be misplaced."

And where to draw the line? Let's say we look at the Carnivora, and we
find some little herpestish thing that actually turns out to be the
sister group to Pholidota, not a member of Carnivora. Oops, I guess
that means the Carnivora doesn't exist, because the hypothesis that it
consisted of all other members plus that one critter has now been
proven false! Oh well, we'll just have to totally abandon a
centuries-old widely-known name, because the phylogeny we presumed to
be true was actually not. Them's the rules, and it'll be less
confusing for people looking at the literature later on, since they
won't associate the name of an incorrect hypothesis with a correct
phylogeny. Right? I think not. Any halfway sensible citizen would tell
you that you should just remove the offending member and move on,
because re-naming everything else would be silly. The rest of the
branch is what it is, the members are what they are, and the reality
of such observations should be reflected by a stable name that stays
with it intuitively and permanently - just like a car is always a car,
a pizza is always a pizza, a horse is always a horse, and a person is
always a person. Mundane life experience continually demonstrates that
there is no need to drop such terms just because they're sometimes
blurry around the edges, and John Q. Public benefits when we can meet
him halfway about the scientific jargon.

If it makes sense in a rhetorical example like this, then it should
make sense elsewhere. We have to be consistent one way or another, and
the form of consistency that doesn't drown onlookers in vexation is
preferable. People presume the existence of soild, "real" groups, not
groups so flimsy that they can't even retain their name if someone
messes up and throws in something different by mistake. The classes of
objects and ideas that the average person encounters day in and day
out don't work that way; they are what they are, errors can be
identified without changing the identity of the class, and everything
thus falls into a predictable order. In principle, so does phylogeny,
and we shouldn't muddle that fact when we could be using lay
psychology to our advantage.