Re: Howard's challenge

[John Kineman <***>]

I responded to that question, I believe, but I'll try to summarize:

Within a world view there can be multiple complementary theories.
There can also be multiple world views, which involve assumptions about
reality on which theory is constructed. These are often not commensurate
with each other, quite simply because they begin with different
assumptions about reality and the founding assumptions of a world view
are not themselves directly testible. Their test is through the
application and use of the view in formulating theories and the
subsequent value of those theories as a set. In other words, how much
new territory does the view allow us to explore, if any?

So, the point is that Rosennean assumptions, I believe, constitute a new
world view. In that sense it is not correct to say that it should
"complement" other views directly. In other words, there may not be a
1:1 translation to other views. This was the case with the quantum
theory and classical mechanics. Niels Bohr attempted to develop a
"correspondence" but failed. A new view can thus contain something new
that the other view cannot represent.

I also commented that it was important for us to develop the Rosennean
view, in relative safety, in order to understand its implications for
theory. It would not have been possible to develop quantum theory if
there were a very strong classical theory mafia breathing down their
necks every moment. The new assumptions need to be understood and taken
serioiusly by a group interested in exploring them. It does not mean
that everyone needs to be interested, but those who are need to have the
freedom, respect, and confidence to proceed with their work. As a
secondary matter - a very important one but nevertheless secondary - it
is important to evaluate how any new view's theories relate to other
view's theories - to check for consistency within the whole scientific
enterprise. If the new view predicts something different to occur in the
other views, that difference should be verifiable in both. Again, in
quantum theory the explicity representation of uncertainty is indeed a
phenomenon that can be discovered in the classical view. The difference
is that it can only be discovered in the classical view, it can't be
explained there. So there is no contradiction between these two views,
one is simply the larger set of phenomenon. This is precisely the
situation we suspect to be true of Rosennean theory and mechanistic
world views.

Respectfully,
JJK

Howard Pattee wrote:

> Judith, Tim, Jon, et al.,
>
> In my last post I irritated Judith and Tim by my wording, for which I apologize (it was not “willfull”). As a consequence no one responded to my central question that was: What happened to the need for multiple complementary models?
>
> Bob and I were in full agreement that one model doesn’t fit all. Bob proposed as a measure of complexity the number of inequivalent models necessary for adequate understanding of systems. The equivalent concept I used was generalized complementarity. Here is the basic idea: We make models to answer specific questions. One model can’t answer all questions. The more complex a system, the more questions are needed. That’s why we like Aristotle. His causes are simply answers to four different classes of questions: What is the house made of? Who built the house? Who designed the house? Why do you want a house? One could add more: What is a house? How do houses evolve? Why do houses decay?
>
> The point is that the answers are independent of each other. One cannot derive one model or one answer from another, nor are the languages in which they are expressed necessarily mutually consistent. “The whole thrust of the old Aristotelian analysis of causation is to make manifest that no one mode of causal entailment suffices to understand anything. At root, this is because the causal categories do not entail each other.” (Rosen, LI, p. 132)
>
> Now I don’t believe it is entirely a mischaracterization to say that Rosen and his followers as represented on this list are following this principle. I hear a lot of objection to, or dismissal of, other models answering other questions about life, reductionist questions in particular. There is some grudging acceptance that, well, reductionism is not so bad but only if based on or accompanied with Rosenean models. I never hear that Rosen’s views are only valuable if accompanied by reductionist models.
>
> I’m trying to get across the idea that modeling complex systems should not be a competitive sport. I know it is an easy trap to fall into, egos being what they are, but if you try to simply compete with reductionist models or the common sense of the likes of von Neumann you have already lost.
>
> I think we need to acknowledge the fact that other people’s models may indeed answer their questions. If you want to ask different questions and their models don’t answer your questions, you have no reason to criticize their models. The same goes for different definitions of the symbols and words, like constraint, complexity, causality, etc., used in different models.
>
> Am I getting through to anyone?
>
> Howard
>
>
>
> http://www.ws.binghamton.edu/pattee/
> http://www.c3.lanl.gov/~rocha/pattee/

--
© 2004 John J. Kineman
all rights reserved