|
page numbers from Internet copy of article http://blogs.cim.warwick.ac.uk/complexity/wp-content/uploads/sites/11/2014/02/Cilliers-2001-Boundaries-Hierarchies-and-Networks.pdf
p.1 We want to predict the behaviour of complex systems, and for that we need good models.
p.2 Is it possible to have a general theory of complex systems? In this paper I suggest that although we can say a lot of important things about complexity in general, it is not possible to develop a general model for complex systems... In what follows I will look at the limitations of models of complex systems be examining the status of boundaries and hierarchies.
p.2 The notion of a model is central to scientific understanding. The notion will be used here in a wide sense (i.e. theories and systems of rules can also be seen as models). In the context of complexity, the role of models is described in the following way by Csanyi (in Khalil and Boulding 1996: 148):
Any kind of scientific statement, concept, law, and any description of a phenomenon is a model construction which tries to reflect phenomena of the external world. Reality is extremely complex; it consists of strongly or more weakly related events. Science makes an attempt to separate and isolate different effects and phenomena. It seeks the simplest relationships by which examined phenomena can at least be described or demonstrated. It creates simplified models which only partly reflect reality, but which allow contemplation, and what is most important, pragmatic, even if sometimes modest, predictions.
We cannot deal with reality in all its complexity. Our models have to reduce this complexity in order to generate some understanding.
p.3 we would hope that that which is left out is unimportant... Models have to reduce the complexity of the phenomena being described, they have to leave something out... We have no choice but to make models if we want to understand the world. However, we have no way of predicting the importance of that which is not considered. [JLJ - perhaps there is a way around this. If we are modeling the current position on the board of a complex game of strategy, we can construct tournaments of games and experiment with leaving certain details out of the model. One model, out of the several, dozens or even hundreds considered, will perform the best (of those considered) in these tournaments. We can use a model which works best overall, and just take in stride an occasional mis-step.]
p.3 what is it then that is described by our models? Are they merely constructions or instruments, or do they reflect reality in some way? ...Emmeche (1997: 46) argues that we can only deal with complexity if we adopt elements from both kinds of ethos.
p.4 it is important to realise that the notion of a constraint is not a negative one. It is not something which merely limits possibilities, constraints are also enabling. By eliminating certain possibilities, others are introduced. Constraints provide a framework that enables descriptions to be built up around it. When dealing with complexity, though, these frameworks cannot be fixed. They are constantly being transformed, and therefore our models will always be provisional.. I would argue that models attempt to grasp the structure of complex systems. Complex systems... have structure, embodied in the patterns of interactions between the components... These structures are also intertwined in a complex way.
p.5 complex systems are open systems where the relationships amongst the components of the system are usually more important than the components themselves. Since there are also relationships with the environment, specifying clearly where a boundary could be, is not obvious.
One way of dealing with the problem of boundaries is to introduce the notion of "operational closure". For a system to maintain its identity, it must reproduce itself (internally). These arguments often follow from the work by Maturana and Varela on autopoiesis. Zeleny (in Khalil and Boulding 1996: 123) defines an autopoietic system as
...a system that is generated through a closed organization of production processes such that the same organisation of processes is regenerated through the interaction of its own products (components), and a boundary emerges as a result of the same constitutive processes.
p.5 The boundary of a complex system is not clearly defined once it has "emerged". Boundaries are simultaneously a function of the activity of the system itself, and a product of the strategy of description involved. In other words, we frame the system by describing it in a certain way (for a certain reason), but we are constrained in where the frame can be drawn. The boundary of the system is therefore neither purely a function of our description, nor is it a purely natural thing. We can never be sure that we have "found" or "defined" it clearly
p.5 We often fall into the trap of thinking of a boundary as something that separates one thing from another. We should rather think of a boundary as something that constitutes that which is bounded. This shift will help us to see the boundary as something enabling, rather than as confining. To quote Zeleny (133) again:
All social systems, and thus all living systems, create, maintain, and degrade their own boundaries. These boundaries do not separate but intimately connect the system with its environment. They do not have to be just physical or topological, but are primarily functional, behavioral, and communicational. They are not "perimeters" but functional constitutive components of a given system.
p.6 in a critically organised system we are never far away from the boundary. If the components of the system are richly interconnected, there will always be a short route from any component to the "outside" of the system. There is thus no safe "inside" of the system, the boundary is folded in, or perhaps, the system consists of boundaries only. Everything is always interacting and interfacing with others and with the environment; the notions of “inside” and “outside” are never simple or uncontested.
|
