[Fis]: Re: CONSILIENCE: When separate inductions jump together

To: "fis-listas.unizar.es" <fis@listas.unizar.es>
Subj: [Fis]: Re: CONSILIENCE: When separate inductions jump together

Dear Malcolm, dear FISers,

You ask:
> Another question that has been raised is whether the attempt to apply the
> same concept, such as 'entropy' across various disciplines counts as a
> consilience; and if not, why not. Obviously, there is a danger in being too
> liberal about what counts as a consilience....

When Boltzmann proposed a model to compute the Clausius entropy,
he created a bridge between two worlds: the "micro" and the "macro"
Undoubtly this was a major victory of science. It linked also
two areas of science, relatively to the approximate classification
of disciplines in vigor at the end of the 19th century. Now the
intersection of the two disciplines is recognized as a full
subdiscipline: statistical mechanics (it's a fun that the name
evokes statistics and mechanics but not chemistry). The main
point here about entropy is that we refer to only one concept:
a measurable parameter, and its calculation from a physical model.
The situation is different after 1948, when Shannon picked
the entropy word and used it in communication sciences.
Despite the formal similarity in the equations, entropy
exists now as a mathematical concept outside any reference to
thermodynamics (except historically). E.g. you can compute
the entropy of some distribution you like, even if it makes
sense in a context far from thermodynamics or statistical
mechanics. Going back to the consilience viewed by Whewell,
or by Wilson, I would agree with you, Malcolm, that there is
a danger to count entropy as an example of consilience:
owing from what I have read about the term (but I am not an
expert), it seems abusive to exhibit an example of consilience
when a (meaningless) formal analogy is found somewhere.
Ihe historical motivation of Shannon to choose the term entropy
is neglected here: it adds confusion to the debate.

Let me add the following about the situation in France, and
please tell me if it is different in other countries.
The need to classify things is commonly encountered in sciences.
and also outside sciences. On the other hand, many scientists
claim that the vertical division between scientific areas is
harmful. Now, I observe that it is difficult to access to
the A positions such as full Professor or Director of Research
(this is the French equivalent of full Professor for researchers)
when your work is pluridisciplinary, and reaching the highest
grades available to the "A" scientist needs first to be clearly
classified in a well known and easily recognized area: this is
called the "easy identifiability criterion".
In other words, you should both tell that you are pluridisciplinary
and absolutely avoid to do a pluridisciplinary work.
A well known example of that (still in France) is people working
on chemical data banks (involves graph theory, algorithmic,...)
The computer sciences and communication sciences are well
recognized, and most areas of chemistry too, but chemists working
on data banks fall most time in organic chemistry sections. It means
that some works involving a high part of informatics are evaluated
by people working on synthesis of some natural product in
20 or 30 steps. Of course organicians like data banks, but
when a decision is needed, they support primarily organic
chemistry projects, because they act in an organic chemistry section.
In such situations, how could we expect progress in
interdisciplinary fields: funds are never attributed to
to scientists considered as being "marginal".
The situation is going worse and worse: the trend is to support
only some large research groups working on a limited number of
topics of industrial importance, eliminating progressively
the smallest research groups. This phenomenon seems to extend soon
at the European level. Most of the french scientific community
do not approve this evolution. But the decision power is rather
elsewhere.

People like what looks familiar. Progress processes by neighbouring,
not by jumps. I heard that De Brooglie presented his work first to
chemists, and was not welcome...

Michel Petitjean Email: petitjean@itodys.jussieu.fr
Editor-in-Chief of Entropy entropy@mdpi.org
ITODYS (CNRS, UMR 7086) ptitjean@ccr.jussieu.fr
1 rue Guy de la Brosse Phone: +33 (0)1 44 27 48 57
75005 Paris, France. FAX : +33 (0)1 44 27 68 14
http://www.mdpi.net/entropy http://www.mdpi.org/entropy
http://petitjeanmichel.free.fr/itoweb.petitjean.html
_______________________________________________
fis mailing list
fis@listas.unizar.es
http://webmail.unizar.es/mailman/listinfo/fis
Received on Fri Sep 24 13:09:22 2004