Re: (OPE-L) Re: Hume and constraint based theories

From: Ian Wright (ian_paul_wright@HOTMAIL.COM)
Date: Thu Nov 20 2003 - 13:30:46 EST

Hello Ajit,

Thanks for your response. I'm enjoying the exchange because our
ideas are so different.

I wanted to completely seperate the relevance of my model as a
model of economic reality from its status as an object that is part of
reality. The point I was trying to make was that any computational
system, for example all the computer programs out there, from
payroll software, email delivery agents, web crawlers, programs
that manage online discussion lists, such as this one, controllers for
autonomous robots, and so forth, are all dynamical systems that
operate in real, historical time. (Let's not confuse a computer program
running in historical time with a computer program that is designed to
model, correctly or incorrectly, the time evolution of another system).

I do not think it possible to refute Humean scepticism over a couple
of emails, nor do I think I necessarily can, but I'd like to point out
that Humean scepticism regarding the causality of computer programs
is unjustified because the causality of computer programs is founded
on logical, not natural, necessity. Computer programs do what they
do in virtue of their logical structure. It is logically impossible for them
to act otherwise, and hence an observer trying to induce a theory over
the event regularities they generate is justified in doing so,
whether they know it or not. An observer that attains a correct theory,
in this case the discovery of the logical structure of the program
implemented in the material system, is justified in inducing
that the mechanism identified will continue to act in this way, for
it is logically impossible for the mechanism identified to act otherwise.

This is not to deny that computational systems may act unexpectedly
due to the vagaries of their implementation in circuits or other
structures of matter. But due to the robustness of the implementation
when something does go wrong then software engineers typically do
not look for errors in the circuitry, or wonder whether a stray
photon upset the CPU at the moment of calculation, or check
whether a strong mangetic field is affecting the machine. They
re-examine the logic of their computer programs because they (tacitly)
know that, excluding some essential breakdown of the system, the events
that the program generates occur with logical necessity, a kind of
necessity that cannot seriously be questioned without questioning
the very basis of rationality, which is a condition of possibility
of Humean scepticism. The demands of logic dictate that the
causality of a computer program be considered objective and real,
and in practice this is what happens everyday in businesses worldwide.

If this argument is accepted, then at the very least a subset of
the natural world is immune from Humean scepticism. A longer
argument would claim that it is justified to hypothesise
that this particular case is in fact universal, but I don't really
want to get into this. But I hope the foregoing provides reasons
why I think it very strained to maintain that the casuality of
an instantiated, running computer program is a figment of
human imagination, irrespective of whether the system was created
by humans (or not).

I'll return to your particular criticisms of my model of a simple
commodity economy in a later post.

All the best,


MSN 8 with e-mail virus protection service: 2 months FREE*

This archive was generated by hypermail 2.1.5 : Fri Nov 21 2003 - 00:00:00 EST