Re: [OPE] epistemological and scientific questions

From: Dave Zachariah <>
Date: Mon Dec 15 2008 - 15:13:24 EST

on 2008-12-14 12:34 wrote:
> Scietific theories have to do with what is true or not. You seem to
> have however a relativist theory of knowledge. If fact you accept in a
> limited sense when you say some models may be ranked "better than
> others". For to rank some as better than others is a statement on what
> is true.
> Which developments in science suggest that [I] have to give up mirror
> theory?

No, concern of scientific theories is not what is "true or not" but the
underlying mechanisms and laws that generate empirical observations. The
underlying mechanisms can by definition not be observed, hence the
notion of a 'true' or 'false' theory is a category error. To rank
something better than others certainly does not state what is 'true'.
Consider ranking airplane models according to some criteria. To say that
one airplane model is 'true' is a category error.

This sounds awfully abstract and strange until one looks at how science
has evolved historically. Compare Newtonian physics, special relativity
and general relativity. Is one 'true'? Certainly they cannot all be true
because they are not equivalent. Are all 'false'? None of the above,
this is a category error. Similarly, this is why the mirror analogy is
very misleading. Mirrors are one-to-one mappings between the surface of
the objects they represent. Which theory is a mirror that perfectly
'reflects reality'? Certainly no one would say that about either of the
theories above.

So, neither truth or falsehood of theories nor the mirror analogy are
helpful to understand the evolution of science. Rather the above
theories are successively better models of reality. And they are ranked
as 'better' based on the criteria of accuracy, generality and consistency.

> I am still not sure what you mean by this. Dou you mean by "predict
> observations" that scientific theories have merely to do with what
> things moy become in future? If this is true, what abou the past and
> present: what was, what is and what might be?

This may be lost in translation. By 'prediction' I did not mean
'forecast', i.e. how the future will become. Prediction here means that
given a certain set of input data a theory allows one to deduce a
certain set of output data. The expected outcome from the theory is the
'prediction'. Whether the outcome occurs tomorrow, over the next 20
years or occurred some 2 million years ago makes no difference in

E.g. Mendelian genetics predicts certain distribution of biological
traits of mice. This is not a forecast, it is a prediction of a process
that was ongoing in the past, although the 'observation' from the
standpoint of the theorist may occur in the future, after the theory is
formulated. The accuracy of the prediction to the actual observation is
one of the criteria used to rank theories.

I urge you to read the brief article by Jennings, who is a physicist,
that I attached before. He makes it quite clear.

//Dave Z

ope mailing list
Received on Mon Dec 15 15:18:40 2008

This archive was generated by hypermail 2.1.8 : Wed Dec 31 2008 - 00:00:05 EST