Jeff:
> I am personally very interesting in exploring this change in context.
Me too.
> This seems to be the aspect that is not amenable to the Cartesian/dual
> logic world view.
That depends on how narrowly you define that view (not mine!). If
you mean the position that (at least the important) truths are
stateable that are independent of context, then you are probably
correct. On the other hand, hard AI-ers like John McCarthy have
developed formal extensions of classic first-order logic which encode
at lease some contextual effects. It is unclear, as yet, as to the exact
relationship between these systems and views.
> How does a system undergo a change in context?
The first question is where this change takes place: is it in the
observer's model or the system itself? Maybe the key change (in this
case) is that living systems model themselves to some extent - for
them it is not a change in context but the _creation_ of context.
For us, this can result in a _change_ of context, as we seek to
better model the situation (of course we can choose to keep
considering the system from the old context, but it probably won't be
very successful).
> What extra first principle do we need to include (or remove) from our
> world view?
"Need" is very subjective. Do you mean "need" as in 'need to produce
more accurate/predictive models', or "need" as in 'need to produce
simpler/more comprehensible models", a combination of these or what?
What are you needing to understand or predict the live state or maybe
the transition?
Assuming that the "need" is for understanding (i.e. simpler/more
comprehensible models) the living state, then we need to select a
context which takes into account whether the system is modelling
itself (has a self context), to understand such contexts, to find out
about constructing relevant and efficient contexts to examine
different aspects of the system.
> Could multiple contexts always be implicit in all of material
> nature, but they only become manifest in certain types of organization?
I think context is inseperably bound up with meaning and modelling -
therefore context is not implicit in material nature (this is, of
course, conjecture). I think that multiple contexts are implicit in
US, that we can *always* find inequivalent contexts/langauges/models
for anything real (?definition of reality: that which can be made
arbitrarily complex by changing your view?). This may be a
consequence of the self-modelling relation, thus it would be
implicit in many systems.
> Is there a consensus on the second-order cybernetic ideas?
No idea, Cliff?
----------------------------------------------------------
Bruce Edmonds
Centre for Policy Modelling,
Manchester Metropolitan University, Aytoun Building,
Aytoun Street, Manchester, M1 3GH. UK.
Tel: +44 161 247 6479 Fax: +44 161 247 6802
http://bruce.edmonds.name/bme_home.html