Re: Judith's challenge #1

[Howard Pattee <***>]

Judith and Tim,

I see from your responses, that I have not been clear about what I see as an unnecessary 
problem. I see it as a misunderstanding on Bob's part the von Neumann's thinking. It was 
certainly not intentional, but nevertheless Bob's repeated charge that von Neumann 
"confounded computation with construction" has had a very adverse effect over the years 
on Bob's reputation. This is because it is such an absurd charge. Perhaps it is my fault 
by assuming too much history of the problem, so let me restate it.

The first point I was trying to emphasize is that, according to all he evidence I have, 
written and oral, Rosen and von Neumann agree on the fundamentals of modeling. Let?s 
begin where Bob does, with the Hertz modeling relation that Bob expressed as a graph. 
There are two sides of the graph, the left side represents the ?natural world? that runs 
on energy/matter and laws. The right side represents our image of nature that runs on 
symbols and rules we call syntax. The two sides are separate causal categories but they 
can be related by measurement (observation, detection, meters, encoding, etc.). Von 
Neumann has discussed this categorical separation in depth because it is crucial in 
physical theories. This categorical separation is often called the epistemic cut. The 
necessary but not sufficient condition for a good model is that the two sides commute or 
run in parallel. Some philosophers have called this the psycho-physical parallelism. As 
Hertz says it, the syntactic (symbolic, inferential) consequent of the image of nature 
must be the same as the image of the consequent of laws of nature.

If any list members feel this much is not absolutely fundamental for Bob?s subsequent 
arguments I can only suggest reading Anticipatory Systems, pp. 73-75, or Life Itself, pp. 
57-61. You will find that Bob?s wording is different than Hertz?s or von Neumann?s. Hertz 
uses ?consequent of laws? and ?consequent of image? instead of ?cause? and ?inferential.? 
( I think Bob uses cause because of Aristotle?s causes.) I see no substantive difference 
in von Neumann and Rosen as indicated by the following quotations:

Here is one of Von Neumann?s statements of the Hertz condition:
Indeed, subjective perception leads us to the intellectual inner life of the individual, 
which is extra-observational by its very nature (since it must be taken for granted by 
any conceivable observation or measurement). Nevertheless, it is a fundamental 
requirement of the scientific viewpoint -- the so-called psycho-physical parallelism 
--that it must be possible so to describe the extra-physical process of the subjective 
perception as if it were in reality in the physical world -- i.e., to assign to its parts 
equivalent physical processes in the objective environment, in ordinary space.

Here is Bob?s statement of the Hertz condition:
The ?whole purpose of modeling is to bring causal entailment, in the external world of 
natural systems, into congruence with the inferential entailment in a formal system or 
model.? (LI, p. 228)

Von Neumann showed by a clever argument that any semantic relations between these two 
categories, like the measurement process or perception, is not derivable from or 
reducible to the laws of nature:
?First, it is inherently entirely correct that the measurement or the related process of 
the subjective perception is a new entity relative to the physical environment and is not 
reducible to the latter.? (e.g., Mathematical Foundations of QM, pp. 351-354)

Bob also recognized this. In his wording:
What is the status of the encoding (measurement in AS) and decoding arrows in that 
diagram [3H.2, LI, p. 60)? We already saw, in the exactly similar diagram of figure 3F.2, 
that the encoding and decoding arrows were themselves unentailed.? (LI, p. 61) That 
means, as von Neumann shows, that measurement is not reducible to either physical laws, 
or inferential syntax.

So, both von Neumann and Rosen in their own wordings are actually recognizing  the same 
three disjoint categories, (1) the laws of nature themselves, (2) our models of the laws, 
and (3) the processes of mapping the one into the other. They also agree that in order to 
make any sense of these causal categories they must not be confounded. Bob says, 
?mischief arises by confusing them.? (LI, p. 234)

I think I will stop here to make sure I have not pushed anybody?s buttons the wrong way 
so far. If there is any reasonably close consensus on the above, I will next outline von 
Neumann?s adaptation of the formal Universal Turing Machine to real computation. Finally, 
I will repeat his words on how he used the logical requirements of a UTM as an informal 
analogy to describe a universal description-construction device. In the interest of 
getting through this discussion in a reasonable time, please limit your responses to just 
what I have written above. Don?t project, or we?ll never get to the next stage.

Howard