Re: A contradiction? Not.

[Judith Rosen <***>]

Not to worry, John (K.). He wasn't contradicting your understanding of his
work at all. My father viewed the "internal predictive models" within
organisms as naturally occurring aspects of those systems, not as
"formalisms". So, models as formal systems, as he defined models in that
reference, means our ways of representing natural systems, be it in
language, in mathematical notation, in schematics, drawings, what-have-you.
So the operative word is "formal" in defining whether a given model is human
created or naturally occurring. Bear in mind that the kind of model that is
an integral aspect of the organization of living systems is not anything
like our human creations. He used the word "model" to refer to this aspect
of living systems, but it's misleading to some degree.

Judith
----- Original Message -----
From: "John Kineman" <***>
To: <***>
Sent: Friday, July 16, 2004 12:25 PM
Subject: [ROSEN] A contradiction?


> Hi list,
>
> I was reading in the limited edition "Rosennean Complexity" at the back
> where there is a section titled "Personal Notes." (P 332-333 in the
> book, p22-23 in Personal Notes). These notes seem to be recapping a lot
> of ideas in a more conversational tone, but there were some statements
> that, if taken literally, would contradict my entire understanding of
> Rosennean Complexity. In particular, he identifies three kinds of
> systems, Natural Systems, Formal Systems, and Volitional Systems, and
> writes the following about Formal Systems:
>
> "B. Formal systems, which could not exist without us. This is the
> province of natural language and mathematics. Clearly, formal systems of
> this type could not exist without human mind...."
>
> The problem occurs if this is taken generally to restrict his definition
> of formal system to human epistemology (a discussion we have had before
> on the list, noting similar confusion). In that case the following
> problem arises:
>
> Given the following well-established statements in Rosennean Complexity:
> 1. Organisms have internal predictive models
> 2. The development of predictive models ("modeling") involves acts of
> abstraction as represented in a modeling relation.
> 3. Acts of abstraction are represented as "encoding" and "decoding" maps
> to some type of formal system.
>
> Then it follows that:
> 1. Models implicate formal systems of some natural kind, at least in
> their construction (once constructed we may think of them as fully
> realized system acting in a formal way with another system).
> 2. The presumed existence of formal systems cannot then be restricted to
> only the human case.
>
> Intuitively we would have to question the restriction of formal systems
> to humans alone on empirical grounds as well, as we know the boundary
> between humans and other life forms is a fuzzy one and that many other
> animals use symbolic formal representation.
>
> One possible interpretation would be that the emphasis on "human" and
> "us" was meant metaphorically, but nevertheless meant to indicate
> applicability only to some higher form of life (humans being a type
> example).
>
> Another possible interpretation would be that these statements were
> meant only in the context of human experience and that formal systems of
> some other type are indeed implied in the rest of nature. This is
> supported by his use of the phrase "formal systems of this type." Hence
> the interpretation would be that he is saying we can be directly aware
> of formal systems only through our own human experience, which of course
> must be true since they are by definition unrealized in our external
world.
>
> This possible ambiguity suggests to me perhaps that the terms of
> reference (precise, general definitions) for this aspect of the ideas
> may not have been fully worked out at the time of his passing, and/or
> that there was some tendency to shift contexts (and thus definitions) in
> an attempt to be convincing to different audiences.
>
> We also have the problem in these "Personal notes" that they were not
> finalized by Rosen himself for publication. Quite often in my own
> writing I will state something in a certain limited way because of the
> context in which I developed the thought, and then I will  have to alter
> the statements for publication as I become aware of equivocation
> problems in contexts and general application  - i.e., to clean up the
> definitions of terms for publication to remove any apparent
> contradictions in a broader context.  However, it seems unlikely that he
> would have been thinking in a narrow context at the end of his career,
> when attempting to write a summary (The "Personal Notes" are not dated,
> but I assume they were written toward the end of his career, which makes
> it all the more intriguing and not dismissable).
>
> Certainly my inclination would be to give priority to the final
> published statements, which presumably have been scrutinized more; but I
> also am reluctant to dismiss these comments as irrelevant - surely he
> had some useful distinction in mind when defining these three kinds of
> systems.  Yet in my view the entire ediface crumbles if formal systems,
> and thus abstractions, and thus models, are allowed to exist only in
> higher intellectual forms, or the definitions are such that "models" in
> non-human organisms must be said to not involve formal systems, and thus
> not involve encoding and decoding, which are clearly processes of
> formalization.
>
> I thought Judith might have some further insight on this.
>
> All the best,
> John K.
>