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

Re: Some thoughts on cladistics

On Mon, Nov 05, 2001 at 11:22:12PM -0500, philidor11 scripsit:
> So, let's propose a rule:  the further a hypothesis gets from direct
> observation, and the fewer direct observations it's based on, the more
> doubtful it should be considered.
> The idea that a classification system should start with easily observed
> modern groups seems comfortable to me in this context.
> Simple, isn't it.  ;-)

Simple, obvious, and wrong. :]

Leaving aside that the modern groups aren't all that easily observed,
and that there isn't in principle any difference between an observation
of what kind of animals are alive today and what kind of animals are
fosilized in rocks, the essential thing about life is that it's the
product of natural selection.

Which is contingent, historical, accidental, and ongoing.

What we have _now_ isn't, on a scale of systematic classification of
life -- rather than *present* life -- a good representation of what
there has been, or might be; a sufficently general to be useful
classification system has to handle the known cases, or it's useless
when dealing with questions of modified descent, and therefor darn near
useless for any systematizing or generalizing study of life.

Once one deals with descent, one must deal with the continuous nature of
life in depth of time; at that point, it's a choice between arbitrary
division by the absence of known descent, or the formal recognition that
life is continuous.

'Continuous' has the virtue of being the nearest thing to an utter,
multiply observed, theoretically supported, rock solid certainty going
in biology, right up there with 'live things die when you boil them'.

               To maintain the end is to uphold the means.