🚀 go-pugleaf

W. H. Newton-Smith, THE RATIONALITY OF SCIENCE, Routledge, London,
1981, p. 199:
"By a theory I shall mean the deductive closure of a set of
theoretical postulates together with an appropriate set of auxiliary
hypotheses; that is, everything that CAN BE deduced from this set."

This definition is nice but "CAN BE", a typical postscientific
distortion, spoils everything. "CAN BE" presupposes an infinite number
of arguments and, accordingly, an infinite number of conclusions or
assertions in the theory. So "CAN BE" vindicates established
postscientific accounts of deductive science:

W. H. Newton-Smith, THE RATIONALITY OF SCIENCE, Routledge, London,
1981, p. 55:
"A theory is a set of assertions and if the number of assertions in a
theory were finite we might initially seek to explicate the notion of
relative verisimilitude in terms of the number of truths and the
number of falsehoods contained within the theories."

That is, since, according to Newton-Smith,  the number of assertions
in a theory is infinite, we cannot "explicate the notion of relative
verisimilitude in terms of the number of truths and the number of
falsehoods contained within the theories" and therefore should take
Popper's speculations very seriously. In the absence of postscientific
distortion Newton-Smith's definition would be slightly different:

"By a theory I shall mean the deductive closure of a set of
theoretical postulates together with an appropriate set of auxiliary
hypotheses; that is, everything that HAS BEEN deduced from this set."

Then the verification of any deductive theory would be reduced to a
small number of verifications of the truthfulness of the postulates
and the auxiliary hypotheses, on the one hand, and a finite number of
verifications of the validity of the arguments, on the other.

Pentcho Valev
pvalev@yahoo.com