PKT message, Re. 'Debunking Economics'

Harry:

To my "brief answer":

> >Both "predictive models" are descriptive rather than explanatory with
> >respect to empirical phenomena in their domains, and owe their "predictive
> >success" to the stability of Nature's unknown physical attributes whereby
> >the phenomena are generated.,

you reply:

> This is *an* explanation of causality, so as such it does not negate
> the need to ascribe casual relationships between variables
> when making predictions. When we make predictions they rely
> on explanations of "cause". There are perhaps many explanations
> of "cause" and they need not be determinstic. There is no need
> to avoid ascribing causal relationships because we don't yet have
> an adequate explanation (or "mechanism"). We should not
> try to conceal our casual intuition behind a wall of mathematical
> formalisms. Newton was uneasy and even embarrassed
> by his "attractive" account of gravity because most phenomena
> in his day were being explained in terms of the interaction
> of material bodies.

The point at issue mirrors the differences between Milton Friedman and Paul A. Samuelson which Steve addresses in Ch. 10 - 'There is Madness in their Method'.

The epistemological considerations noted by Newton and Hume which I cited in an earlier message imply that, as Friedman put it somewhat inaccurately, "Truly important and significant hypotheses will be found to have "assumptions" that are wildly inaccurate descriptive representations of reality." (p. 101).

In the context, his "F-twist" rejoinder suggests that epistemology is not Samuelson's strong suit.

Gunnar

Re. 'Debunking Economics' - Ch. 12