Re: Philidor: No Class

["David Marjanovic" <david.marjanovic@gmx.at>]

----- Original Message -----
From: <philidor11@snet.net>
Sent: Thursday, August 29, 2002 6:50 AM

> And, of course, his effort would claim greater attention from
> others in the field than my own.

That's probable. But if you publish in the right journal, and if your data
matrix is bigger, then you have good chances of getting more attention than
anyone else in the field (before the next analysis is published).

> By the way, cladistic analysis has used alternatives to parsimony
> without being less 'scientific'.

Hm. Molecular folks like to use maximum likelihood instead of max.
parsimony. Fine. Just inapplicable to fossils, because how much more
probable is it, _in percent_, that the 3 central metatarsals fuse than that
the quadrate gets a second head? Such numbers can be found out for
molecules, e. g., how much more likely it is that an A in the third position
of a codon will be substituted with a G than with a T.
        Many other "alternatives" have been used -- but they, e. g.
neighbor-joining or DNA-DNA hybridization, are methods that show distance,
in how many features taxa differ, but don't tell apomorphies and
plesiomorphies apart. They are methods of phenetics, not of cladistics. (And
they have only been used for molecules AFAIK.)

> Any logical principle applied
> to data is potentially acceptable,

Any _scientific_ principle that makes testable predictions on phylogeny is
potentially acceptable when we want to find out phylogeny, which is what we
do. Scientific means that it makes testable predictions.

> Past evolution can be neither observed nor replicated,
> so the ultimate arbiter is unavailable.

So we should stop _all science_?

> On the other hand, the Linnaean system is so stable that your
> theorizing would be at the fringes.

:-D :-D :-D
Have a look at this classifications of Tetrapoda except Aves and Mammalia.
www.cmnh.org/dinoarch/2002Jan/msg01091.html It's just 70 years old. You can
see the changes. More has changed than has stayed the same! Or have a look
at your own classification. How big is Hominidae? There is no way to decide.
Apart from "I say it because I am an expert", or just "I say it because I
feel like it", and then I'll find heaps of other experts who'll disagree.
Because names are not defined in the Linnaean system, anyone can do as they
like. If I say "no, crocodiles are too much like birds, they're completely
unlike lizards, I say crocodiles are not reptiles, I'll give them their own
class", who on that planet can stop me? Huh? I am not exaggerating. Things
like that have been done. Often. And no ICZN or suchlike has ever changed
that.
        If, on the other hand, I say that crocodiles share a more recent
ancestor with lizards than with birds, and if I publish my evidence, then
this is a testable (and already falsified) hypothesis. And everyone who can
falsify the hypothesis can stop me. (If I don't simply switch to being
stubborn. But then I'll be called unscientific, and rightly so.)
        To return to the example above: How big (inclusive) Hominidae is is
a matter of _definition_, and only in phylogenetic nomenclature are names
defined. The shape of the tree, part of which is Hominidae, can be
_discovered_. So let's talk about that.

> If you have difficulty with placing
> a long extinct animal which might be a bird or dino,
> there will be many erudite people to be persuaded.

Let's make it simple. Let's all agree it is a dino. Then we can limit the
discussion to whether it is a bird. And as soon as Aves is defined, this
discussion becomes _identical_ to the discussion about its place in the tree
of life. And this is the realm of science. We can make testable hypotheses
on that. These are now suddenly hypotheses on classification. Look what
we've done: We've based classification on science, rather than mood swings.
Great job!
        A consequence of this is that ranks are abandoned. Who needs them?
Do they correspond to anything in reality?

Have you read the first few pages of the PhyloCode? I really think you
should. Might clear up some misunderstandings.

>  Those uninterested need not be in suspense about whether their
> children have to learn new current majority opinion.

Which never happens. At least over here. I had old biology teachers,
therefore I learnt what was the majority opinion when _they_ had gone to
school: _Two_ kingdoms, animals and plants.

> By the way, you can tell your brother that the new kingdom matches
> Leichtenstein or Monaco in size of population and influence.

I want to see your face if suddenly all Archaea die. Liechtenstein, BTW.

>  Shame there are no Duchies in the plan of biology.

There have once been legions, cohorts, divisions, series, all with super-
and sub- ... few people still use them. Did you write the Linnaean system
were stable?

> And you
> can add that at least he doesn't have to learn cladistics, where
> the text would hourly be replaced with new interpretations recorded
> on self-stick paper.

Well, firstly, most of the hourly changes are far too affect a schoolbook.
Secondly, if classifications in biology textbooks would follow the newest
opinions, then the text would have had to be replaced every hour for the
last 100 years, and every day for the century before.