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

Re: Cladistics (was Sci. Am. - present)



Sorry for printing the entire original message, but it gets to the nub of
the whole business.

From: chris brochu   Date: 21 February 1998 08:03
Subject: Re: Cladistics (was Sci. Am. - present)

>> (JJ wrote:)
>>It is a pity we cannot have DNA details of dinosaurs since I do not
>>believe there was such a sudden burst of DNA changes on loss of flight,
>>as there was of morphological changes. I would place much greater faith
>>in a DNA cladogram for birds and related dinos.
>
>I'm not sure why this is. In the vast majority of cases, DNA sequence data
>and morphology tell you more or less the same thing. Molecular data are
>no less subject to selective pressures, and levels of homoplasy given data
>sets of the same size are no different.
>
>10 years ago, the question was "should it be molecules or morphology?"
>Today, the answer is "yes!"
>
>
>
>> Incidentally there is absolutely no reason why we cannot use a bit of
>>>brainpower to improve the performance of cladistics. When I said
>>"blind" cladistics, I did NOT mean all cladistics is blind.
>
>
>
>How would you "improve" a parsimony analysis? I'm not trying to be
>patronizing - this is a serious question. Given the principle of total
>evidence, the hypothesis that best matches all the available data is the
>preferred one - are you advocating some sort of significance test for
>character data? If so, on what objective basis would this be based? And
>bear in mind, switching to molecular data won't necessarily "improve" the
>analysis, except by adding additional relevant information.


You?re asking me to pose and justify an improvement to cladistic theory.  In
response, I suggest that I don?t need to, since the principle of
_imperfection in cladistics_ has already been accepted, and that is enough.
None of us can easily quantify the imperfection, or how it behaves under
varying conditions, and few of us would criticise it for this imperfection.
No tool is perfect.  However, the fact remains that we _are_ allowed to
question details of the output of any cladistic process.  ( I?ll briefly
mention ?loss of flight? as a special case, a ?Pooh-trap? for morphological
cladogenicists.)

You?re going to say ?Come on, unless you mention objectivity, parsimony,
etc, you?re just arm-waving!?  This reminds me of one of my professors who
was always saying ?Can you prove that??  Funny thing is, he could never
prove the necessity for proof himself since it was itself just another
heuristic.

We are looking into the dregs that time has left us, to make as accurate a
tree as we can.  This is really a task of perception.  Humans make machines
for visual perception etc, and so does nature.  In neither case is a
particular principle relied upon exclusively - whatever processes that can
be made to work well together are used.

Cladistics is just a tool.  So are objectivity, parsimony etc.  The only
indispensible tool is common sense.  (Warning!  Do not attempt this argument
in your grant application, where only a sub-set of reality will be in
operation!)

If you?re lucky enough to be able to earn a living out of your hobby, stick
to your guns and say ?the runes say _Velociraptor_ begat _Archeopteryx_.?
Me?  I just want to know the right answer!

John V Jackson    jjackson@interalpha.co.uk

"You're walking through the desert and you see a tortoise lying on its back.
Do you turn it over?"

"What's a tortoise?"