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

>Here are three questions that might be discussed:
>
> (1) Is the consilience of inductions a clear notion? What
> does it mean for facts to be of a different kind?
> Why is this especially significant?

I wonder what makes consilience different from prediction of untried
instances. There is a question of the level of abstraction that we are
operating at. In "prediction of untried instances" we are examining the
ability of a particular model to be applicable to situations that were not
used to build the model. In "consilience of inductions" we apply the model
built with observations of apples and stones to observations of moons and
planets.

A somewhat different view is that consilience is about the creation of a
variable and/or a pattern that is defined for a large number of diverse
situations. Consilience = generality. Such an interpretation would make "4)
Convergence of the theory" redundant, as increasing the generality of a tool
is how simplicity is acheved through a reduction in the number of cognitive
tools.

> (2) Do examples of consilience occur in sciences outside of
> physics?

How about mathematics? Ultimately, all branches of science make good use of
notions of summation, multiplication, integration, of histograms and
probability distributions. Any kind of an abstract concept that can be
applied to various variables of a particular type are subject to
consilience. Entropy and energy are particular concepts that, at an abstract
level, can be applied to many areas. If you allow me to oversimplify: in
physics, energy is, well, energy; in economics, energy is money; in biology,
energy is food; in chemistry, energy is heat; in psychology, energy is
motivation; etc.

In these examples, I am examining the consilience not of hypotheses, but of
scientific "tools" (calculus: applicable whenever there are variables and
infinitesimal quantities) and "patterns" (second law of thermodynamics:
applicable whenever there is a notion of energy and a notion of space).

Best regards,

                Aleks

--
mag. Aleks Jakulin
http://www.ailab.si/aleks/
Artificial Intelligence Laboratory, 
Faculty of Computer and Information Science, University of Ljubljana.
====
For reference:
> Whewell distinguishes four tests of scientific hypotheses:
>
> (1)   The Prediction of Tried Instances (used in the 
> construction of the hypothesis).
> 
> (2)   The Prediction of Untried Instances;
> 
> (3)   The Consilience of Inductions; and
> 
> (4)   The Convergence of a Theory towards Simplicity and Unity.
_______________________________________________
fis mailing list
fis@listas.unizar.es
http://webmail.unizar.es/mailman/listinfo/fis