Comparing Rosennean Complexity

[Judith Rosen <***>]

>Howard Pattee wrote:
Von Neumann's fundamental question is: What is the threshold of complexity
of a system that can evolve increasing complexity rather than degrade?
Increasing complexity can be interpreted broadly as open-ended emergence,
endless variety, or evolutionary novelty.<

I did some reading over the weekend and I've got a handle on what most
bothered my father about von Neumann's theoretical stuff on complexity. The
set of ideas above is already in contradiction with what my father's
theories posited, I'm afraid. His contention was that all  levels of
complexity are innately stable (i.e. the atom) because complex organization
is-- by virtue of its organization-- a naturally balanced kind of system. He
felt that only simple systems tend towards destabilization over time
(entropy). The threshold that von Neumann was talking about is similar to
what, in Rosennean Complexity would be the threshold (except that it isn't a
threshold- see below)  between simple systems and complex systems.

As such, evolutionary processes, as used/defined in biology, are not
happening at this level of complexity. I agree that the word "evolution"
suits a number of different realms, but the kind of thing that happens in
living systems is different in a fundamental way from the evolution of
elements, say, or the evolution of airplane flight, etc. Atoms are complex
systems that possess the innate stability of their complex organization, but
they don't reproduce and don't exhibit anticipatory behavior. Similarly,
ecosystems don't reproduce and don't exhibit anticipatory behavior, except
via the collective behavior of the life forms that make an ecosystem. But an
single celled algae does reproduce, does exhibit anticipatory behavior. The
threshold that my father was most interested in was the one between life and
non-life because this is where his own answers were to be found.

He didn't like the term "threshold" though, which strikes me as being kind
of funny. He could use a term like "internal predictive model" but eschew
the term "threshold"-- why? Because "threshold" suggested that you could get
there from simplicity, just by adding more complicatedness. But what else
would you call a boundary that, once crossed over, has a whole new world on
the other side? If threshold, as a term, connotes a reductionistic or
mechanistic idea, how does "internal predictive model" not do the same?
(That's the kind of thing I would ask him. Anybody game to fill in?) Anyway,
I have found the word threshold to be necessary when discussing my father's
ideas in plain English.

The reason I wouldn't say there is a threshold between simple systems and
complex systems is that, in my father's theoretical framework, there is no
way to cross over from simplicity. In order to generate a complex system one
would have to start with complex organization. No amount of complicated
simplicity, added together, is going to yield a complex system. From his
vantage point, he said von Neumann was saying the opposite of this.

Judith