Re: On Robert Rosen's travels...

[Howard Pattee <***>]

Judith,

Following your suggestion, I will say that it is my opinion that Rosen and von Neumann 
reached similar, or at the least, consistent conclusions about the inadequacy of 
Newtonian dynamics for describing life. It is also my opinion that their apparent 
difference over the very narrow technical issue of self-replication was a 
misunderstanding.
I think it is more than my opinion that von Neumann's model is now accepted by many 
active researchers as a fundamental contribution to theoretical biology.

> Pattee wrote: "I also do not consider different definitions as a basic disagreement. 
> Definitions are assessed by their usefulness in a particular context often by their 
> communication value or in constructing specific models to answer specific questions. I 
> agree that Rosen's definition of complexity is different from von Neumann's as well as 
> Chaitin's and others, but I do not see them as inconsistent with each other. They are 
> all useful for models that answer different questions."

Judith: I find this statement extremely slippery. You "do not see them as inconsistent 
with each other"?
Please define "consistent/inconsistent".

HP: I use ?consistent? only in the formal sense. That is, I use it like a logician or 
mathematician. There are several meanings in logic, but for normal discussion I use 
two-valued logic. If we agree ?either A or not-A,? then ?A and not-A? is inconsistent.

Inconsistency can arise only in formal symbol systems. Nothing in nature can be 
inconsistent (What would that mean?). It is only in our formal models that inconsistency 
can occur.

Judith: In my mental dictionary... When scientific definitions, which are purportedly 
describing actual properties of reality, cannot co-exist IN reality, I define that as 
being "inconsistent" and in my opinion, it means that one or both of those scientific 
definitions is wrong.

HP: To change your mind I think would require more experience with physical models. 
Scientific definitions do not describe ?actual properties of reality.? They are not right 
or wrong in any scientifically verifiable sense. Only in the formal sense can they be 
inconsistent. Two well-defined valid models can describe properties of reality even 
though the models if combined formally would be inconsistent. For example, the 
microscopic laws of motion are reversible, that is, formally symmetric in time. A box of 
gas obeys these laws. The second law of thermodynamics is irreversible, and the same box 
of gas also obeys this law. You cannot formally combine these two models consistently, 
but to fully understand a box of gas you need both models.

Physics usually call these complementary models. Mathematicians usually call them 
inequivalent models. The essential property of such multiple models is that neither can 
be derived from, or reduced to, the other. As Rosen points out, this is an essential 
condition for modeling complex systems. To fully understand a complex system requires 
such multiple inequivalent models.

Judith: There is a danger in ignoring such inconsistency.

HP: Believe me, physicists do not ignore inconsistency and try to avoid it. But the 
ultimate test is experiment. There is a greater danger in believing that inconsistency is 
in reality and ignoring experiments that appear inconsistent only because of our models.

Judith: The point of the story is that there is always another side, another perspective, 
and another opinion. In Robert Rosen's case, he could hold them all simultaneously. His 
was a very complicated mind, where multiple points of view about the same thing could be 
generated and held, even when many of those points of view conflicted with each other.

HP: We all agree. That is exactly the type of thinking that complementary models engender 
and that physics has shown to be necessary.

Howard