Re: [OPE-L] Hume

From: Paul Cockshott (wpc@DCS.GLA.AC.UK)
Date: Fri Jan 21 2005 - 06:20:24 EST

Replying to an old posting here by Ajit.

The issue raised here is what the nature of 
causation in science and economics should be.

I would argue that what science does is not so
much uncover causes but uncover symmetries and

To take a paradigmatic instance: the Newtonian
laws of motion do not deal with causes, they
uncover constraints on the feasible configurations
of physical systems in time and space. They 
are constraints on 4 dimensional space.

The structure of something like Sraffa's price
system is similar. It establishes a set of
constraints on what is feasible, given certain
assumptions. In this sense it is not causal
> Andy

Hi, Andy!
Because economic theory does not have to be an
empiricist philosophy of knowledge. Hume himself did
not follow his empiricist philosophy in his other
works because his philosophy ultimately leads to
nihilism. But that does not mean that the
philosophical problem he raised for empiricist
knowledge, particularly for the implied relation of
cause and effect, is all bunk. It shows us the
limitation of what we claim to know. Now, we all know
that all sciences are predictive, i.e. built on the
relation of cause and effect. But science is not in a
business of proving anything--it is neither philosophy
nor mathematics. In some Kantian sense science simply
takes the relation of cause and effect as a priori or
its fundamental belief or axiom. On this basis it only
tentatively suggests certain causal explanations for
various phenomena. But these theories must always
remain tentative and can never prove its correctness
beyond doubt. The main role of science is to act as a
medicine that sooths our mind by giving some sort of
order to desperate phenomena--it keeps us from going
crazy! That's an admirable job and economics can be
part of it. But it is also good to know the
limitations of what we claim to know. Cheers, ajit sinha

Do you Yahoo!?
Protect your identity with Yahoo! Mail AddressGuard

This archive was generated by hypermail 2.1.5 : Tue Jan 25 2005 - 00:00:01 EST