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

Re: a rose by any other name(was fish & dogs)

Clarifying, please.  Even butter can be made clear.

<In the phylogenetic taxonomic system, the statement of ancestry is the
algorithm used.>

Definitional difference?  I'm using algorithm to mean a formalized process
of decision-making unrelated to any specific example, as in a computer
program.  Hence, the statement of ancestry is the result of the application
of the algorithm.
I'm thinking of the points you're discussing as concerning the algorithm and
also about the data going into the algorithm.  Both the algorithm and the
data entered are the result of selection among choices.  The process of
choosing (this IS the algorithm, these ARE the animals and characters and
codings to be used) does not lead to a single, inevitable result.
Therefore, some choices are relatively better than others.
How is it possible to determine relatively better; how can one set of
choices/decisions be authoritative, worth taking seriously, and another not?
I read with interest about your own recent presentation of your findings
(which concerned the data selection/entry, I think, and not the algorithm
selection part of the process of determining relationship).  You are
yourself very obviously an authority, someone able to make choices which are
taken seriously.  In this sense, a set of choices is authoritative because
you say so.
<Of course, someone first has to propose it, but then
it should (in principle) remain.  In contrast, an authority-based system
works on the principle of (pardon the temporary blasphemy...) "What Would
Simpson Do?" (or insert other authority in G.G.'s place).  Would Simpson
have included this new little fuzzball in Allotheria?  Would this new
ornithopod fit Lull & Wright's concept of Hadrosauridae, or Weishampel's, or
Horner's diphyletic hadrosaurid concept?>
Has no one, including you, ever tested to see what a new animal with a mouth
full of stabbers, choppers, and grinders would do to your work?
Also, do you expect that the algorithm you chose and the characters you
chose to code and the coding criteria decisions that you made and the
animals that you chose to include or exclude will have no influence later?
If someone chooses different alternatives, not on the basis of new
information, but in response to a different theory of relationship, wouldn't
you expect them to recognize your work and argue why their approach is more
productive?  Outside of your peers, wouldn't this impose some degree of
limitation on the discussion?
In short (too late), the reference to authority is pretty confusing.  I had
assumed that the greater the reliance on logic, the more authority-based a
system (whether that authority is based on an individual or on consensus in
a group, as in the 'battle fought and won').  Gives me hope for other
logic-derived systems of thought.

Have to think more about your other comments.  Ideas held because of
precision or utility...
Anyway, thanks.