>My point was that intelligent input is required for the economy to
>"evolve". Whether this input comes from inside or outside the system,
>whether or not the intelligent agents are coginzant of their exact part in
>the whole -- these are irrelevant. Nonintelligent agency cannot generate a
>complex system such as an economy. Blind Watchmaker evolutionary theory
>argues for the evolution of complex systems with *no* intelligent input
>from any source; thus the economy analogy fails.
I think its your argument against Glenn's analogy that fails. The
intelligence or lack thereof of the individual agents in the economy is not
relevant to the analogy. For the most part the economy can be modeled as a
collection of various populations of automata whose actions follow a few
simple rules that can be reduced to algorithmic formulae. The most important
such rule is that the agents each attempt to act in a way that will maximize
their well being (i.e. profit/income/market share/ etc.) based on the limited
information available to them from their local environment (e.g. market
conditions, etc.) at the time. The effect of irrational actions and self-
denying altruistic behavior, and other acts that agents may exhibit by
actually being free (supposedly) intelligent agents can be modeled as a
background stochastic noise (mutations) as can the effect of the introduction
of various political, social and technological developments (such as wars
crop failures, and new inventions). As I recall (former physicist) Paul
Samuelson got a Nobel Prize in economics for applying mindless extremum
principles (analogous to the Hamilton's principle of least action) to the
field of analytical economics. Econometrics is not as precise as physics as
far as its models goes, but even physical models of complicated phenomena in
nature are not perfect (e.g. simulations of the global weather and climate,
etc.) either. Just because the models sometimes (ok, often) are incorrect in
their detailed predictions, that does not mean they do not capture the major
effects driving the phenomology, and the supposed intelligence of the actors
is not needed to understand how the processes work or how things unfold on the
large scale. Like the biological world there are fossil suboptimalities in
the marketplace that are the result of historical contingencies. (This is
certainly the case in the electronics/computer/software industries, e.g. the
longtime persistence of the 640k memory barrier and many other examples.)
Because of the (non-markovian) persistent effects of such random historical
contingencies it is doubtful that the economy would develop the same way in
detail a second time from the same inital state. However, to understand the
mechanisms of economic change one does not have to invoke intelligent design.
Come to think of it, I believe that recent history has shown that planned
economies tend to be at a competitive disadvantage.
Echoing Brian here, Good Analogy Glenn!
David Bowman
dbowman@gtc.georgetown.ky.us