Rosen and the Santa Fe Institute

Jeff Prideaux (JPRIDEAUX@GEMS.VCU.EDU)
Thu, 23 Feb 1995 17:14:40 -0400


A PCP member (offline) asked for the Rosen reference and what was
the difference between Rosen's theories and the work being done at
the Santa Fe institute..

Robert Rosen, Life Itself: A Comprehensive inquiry into the nature,
origin, and fabrication of life, New York: Columbia University
Press, 1991

Also, although I haven't read it yet, here is another one Don
Mikulecky recommends:

John L. Casti, Alternate realities: mathematical models of nature
and man, New York: Wiley, 1989

The big difference between Rosen and the Santa Fe group is over the
concept of simulation. The Santa Fe group, to the best of my
knowledge, are simulating life-like processes on computers. In a
simulation, the phenomenology of one process is acted out in the
hardware of another device (independent of the process being
simulated). In short, some information about the causality is
lost. For example in a computer simulation, the functional aspects
become the program, and the measurable entities become the data.
But in a computer, there is no fundamental difference between
program and data. It is all software. The hardware is uncoupled
from the problem being run. (This is the power of the universal
computer...the same computer can be used for many different
things...but it sets up certain limitations). In short, the
distinction between material and efficient cause gets blurred in a
simulation. This distinction is what Rosen used to stop the
infinite regress referred to in earlier messages. Rosen claims
that a simulation, no matter how elaborate, cannot approach the
true complexity of life. Rosen even defines complexity as
something that is not simulatable. In short, if you are doing it
on a computer, it isn't life.

It appears, if one wants to be true to Rosen's definition of
complexity, that one must dabble in actual hardware. The material
stuff of which one works must be the actual the phenomena of study.

Think of a realized complex model as the "inclined plane" of
biology. Remember that historically it was the inclined plane
which enabled people to understand gravity.

-----------------------------------------------------------------

For another circle of ideas, refer to the excellent book Godel,
Escher, Bach by D.Hofstadter. Although he arrives at a different
conclusion than Rosen, he uses the term "Strange loop" which has
some bearing. For Hofstadter, a strange loop is one in which there
seems to be some logical impossibility. For example, the authors
triangle: A invents the character B. B invents the character C.
And C invents the character A. Another example would be Escher's
drawing depicting one hand drawing another hand which is drawing
the first hand. In both of these cases there is a hidden layer
that is responsible for the apparent strange loop. In the case of
the drawing, it is Escher himself that drew the entire drawing.
The authors triangle only exists in someone's imagination.
Hofstadter explains away these "strange loops" by always having a
hidden uncoupled level separated from the strange loop. He goes on
to say that our consciousness is a form of a strange loop that is
possible because of the hidden level of our neurons that are not
directly perceptible to us.

I believe that Rosen, though, is arguing for an actual strange
loop. Not one made possible by another hidden layer, but one
manifested in the material of which it is made alone. I believe
this distinction may be at the core of the concept of simulation
itself. I'm still struggling with these concepts though... If
anybody has any comments, I'm all eyes.

Jeff Prideaux