If you go to http://pespmc1.vub.ac.be/cybswhat.html you will find a
mention of 2nd order cybernetics, though I would prefer to see a more
detailed treatment. But the main philosophical idea underlying 2nd order
cybernetics, constructivism, is discussed in detail in
http://pespmc1.vub.ac.be/construct.html. The epistemology of PCP is
constructivist, but without falling into the extremes held by certain 2nd
order cyberneticists, which tend to see all models accepted by a community
of observers as similarly "real".
> or is it the current position of PCP not to
>include such (speculative) ideas on the official home page? Or does the PCP
>home page not make a distinction between regular cybernetics and second-order
>cybernetics?
The position of the PCP editors is that the difference between "first" and
"second" cybernetics tends to be overemphasised by those of the second
order school. If you read the writings of the founding fathers of
cybernetics, such as Ashby, you will notice very little disagreement with
the writings of second order cyberneticians, like von Foerster or Maturana.
In practice, it is more a difference in emphasis than in fundamental
assumptions. Umpleby has also suggested to encompass 1st and 2nd
cybernetics in an integrated model characterized by the additional
dimension of "interaction with the observer". When observer influence is
small, you get 1st, if it is large you get 2nd cybernetics.
>(where it is all part of one big package called cybernetics)???
In practice, "cybernetics" is what is done by people who call themselves
"cyberneticians". If you look at the traditional associations, journals and
books on cybernetics, you encounter always the same people, without clear
separation between people doing 1st or 2nd cybernetics, or even between
cybernetics and systems theory. It is much more practical to just call the
whole bunch "cybernetics and systems science".
>Would it be fair to say that regular cybernetics is in line with computable
>engineering control theory (for example feed-back systems)...and SOC would
>be representative of a non-computable paradigm shifting idea?
That seems like an oversimplification. First, classical cybernetics
(feedback etc.) has never paid much attention to computability since its
models tend to be rather continuous or analog, in contrast to the discrete,
digital models of computer science and AI. Second, many people in AI and
Alife using computational models have been influenced by 2nd order
cybernetics, though there is still an ideological battle going on around
that.
On the other hand, it is obvious that if you take into account the
observer, all models become more open, complex, fuzzy and variable, and
because of that more difficult (if not plainly impossible) to compute.
Felix Geyer, in his excellent review of 1st and 2nd order cybernetics and
their applications to social systems
(http://pespmc1.vub.ac.be/Einmag_Abstr/FGeyer.html), which he presented at
our "Evolution of Complexity" Symposium last May in Brussels, has suggested
that because of that 2nd order cybernetics may have been "a bridge too
far": an important conceptual insight, which unfortunately makes all models
so complex that they become practically useless.
The search is on for models that take into account the observer but that
still allow some form of concrete predictions. The only real success in
that field until now, quantum mechanics, owes nothing to (2nd order)
cybernetics.
________________________________________________________________________
Dr. Francis Heylighen, Systems Researcher fheyligh@vnet3.vub.ac.be
PESP, Free University of Brussels, Pleinlaan 2, B-1050 Brussels, Belgium
Tel +32-2-6292525; Fax +32-2-6292489; http://pespmc1.vub.ac.be/HEYL.html