I am similarly frustrated by this discourse. I thought that I had tried a
number of different conversational approaches, but I do admit that I tend
towards the analytical.
I THINK I understand your perspective, but honestly, it's too "squishy" for
me to be sure. That's why I try to pin you down, and make you more
specific. Are you sure that you've tried as hard as you can to see things
my way, as well?
My interest is not to convince you that everything is hierarchy, and that
recursion is invalid. I just want to make it more clear what we mean by
those terms, because, honestly, I've used them before in other contexts: to
me, the recursion you (Beer) describe is a FORM of hierarchy, and hierarchy
means much more than just a particular traditional form of top-down control
social organization.
But we don't need to go on if you don't want.
>A hierarchic system is one in which the relationship between the component
>parts is premised on a command/obey correlation -- hence "The Hierarchy" in
>the Roman Church.
TO YOU.
To ME: a hierarchy is any system which admits to a description in terms of
LEVELS. These levels can be partially ordered, not just linearly ordered,
and form the levels of the hierarchy.
THAT's why we need to stick little hierarchy_B and hierarchy_C labels on
things, so we (or at least I) don't get CONFUSED. The way you talk CONFUSES
me.
>Here are some other observations that, generally, apply to relationships
>between components in systems (at least, social systems) that are structured
>hierarchically: linear, centralized, closed, reliant, dependent, competitive.
>
>And here are some observations that generally apply to relationships in
>recursive structures: neural, decentralized, open, autonomous,
>interdependent, collaborative.
I'll be happy to put the former terms into the hierarchy_B basket and the
latter into the recursion_B basket. But I aspire to more: I aspire to the
development of a THEORY which would be able to describe and predict these
characterstics in terms of each other, and based on some underlying
structural and/or functional properties of the system in question. For
example, why do linear and centralized go together? What is something which
is non-linear and centralized, or linear and decentralized? Are these
common characteristics of something else? How do they jointly related to
the closed vs. open distinction? How do we verify that your distinctions
really hold up? Etc. etc. etc.
Now the point is that I've already done this for myself a fair amount
(although not in terms of these specifics). And for me, and I think also
for Francis and Val, the concept of a HIERARCHY in terms of a
partially-ordered structure is very central. These systems are both
somewhat linear and non-linear, centralized in some aspects, not in others,
etc.
Now if you want to have a conversation, let's make it two-way: I'll try to
understand what you mean by all those terms, and how they relate, and you
try to wrap your head around this more general concept of hierarchy.
>Please visualize the component parts in a recursive system as being embedded
>in each other like a number of Russian dolls. Then re-visualize a hierarchic
>system as that familiar, pyramid or tree-like structure.
Yeah, sure, no problem, did that long ago. Now, do you care that there are
isomorphisms between these two structures? That nested dolls can be
described in terms of trees, and vice versa? If not, then I guess we've got
nothing to talk about.
> --- Beer quote deleted ----
>
>Am I helping you to understand how hierarchic and recursive systems differ?
No: like Don with Rosen, mere appeal to authority is not sufficient. Of
course a rabbit in a wood or a star in a galaxy is not part of a
command-control hierarchy. So what? Did anyone suggest that they were? Is
the point ONLY that there are non-hierarchical_B ways to organize societies
or ecologies? Did anyone honestly suggest that you can model a fish pond
like a military?
>My thesis/proposition is that our world is more clearly understood in terms
>of models that are recursive rather than hierarchic; and that we, therefore,
>might more competently manage our complex human affairs if we used recursive
>systemic models and logic. To me, that is consistent with "tackl[ing]
>age-old philosophical questions with the help of the most recent cybernetic
>theories and technologies."
OK, if here you mean hierarchic_B and recursive_B, I'd go further: I'd say
the world is more clearly understood in terms of models which extend beyond
all these forms of organization.
So what's the real point here?
O---------------------------------------------------------------------------->
| Cliff Joslyn, NRC Research Associate, Cybernetician at Large
| Mail Code 522.3, NASA Goddard Space Flight Center, Greenbelt, MD 20771 USA
| joslyn@kong.gsfc.nasa.gov http://pespmc1.vub.ac.be/joslyn 301-286-5773
V All the world is biscuit-shaped. . .