The Problem of Complexity: Definition and Knowledge

I will write more about narrative theory, but I won't stop reading about complexity. I'm reading a dissertation by Rika Preiser entitled The Problem of Complexity: Re-Thinking the Role of Critique (Dec. 2012, Stellenbosch University). I find it most engaging, and I want to write about it before I forget what she says. I came across Preiser's work through her association with Paul Cilliers, who was her dissertation director until his untimely death 2011 July 31. I have read much of Cilliers, and quickly realized that he was helped greatly by two of his students Rika Preiser and Minka Woermann, both of whom I started reading. Their own work has helped me understand Cilliers. I suspect Preiser's dissertation will do the same.

In her dissertation, Preiser frames the problem of complexity in two ways:

1. the problems with the definition of complexity, and
2. the problems with observing, knowing, and describing complexity.

In other words, she poses an ontological question: what is complexity? and an epistemological question: how can we understand complexity?

The Idea(l)s of Complexity

Preiser insists that there is no unifying Theory of Complexity (24). At best, we can recognise a certain "economy of concepts" that arranges itself around the characteristics of complex systems to form a "commonplace structure of intelligibility" (38) that Edgar Morin calls a paradigm of complexity and Paul Cilliers calls an attitude of complexity. Preiser lists 10 common denominators that inform the various theories of complexity:

1. The history and origins of theories of complexity are directly linked to General Systems Theory, cybernetics and artificial intelligence. 
2. Theories of complexity follow two distinct tracks: 
1. a track that claims complexity has the duty to measure and formalise complex systems by means of mathematical computation, called restricted complexity by Morin and Cilliers, and 
2. a paradigm that argues that complex systems ultimately cannot be measured and calculated but remain in principle too complex to model in theoretical equations. Called general complexity by Morin and critical complexity by Cilliers. 
3. Theories of complexity are all concerned with the study of complex phenomena in states of non-equilibrium that display characteristics of non-linearity, self organisation, and emergence and behave in a manner in which time and energy expenditure is irreversible. 
4. Theories of complexity use technical and metaphorical vocabulary to describe complex phenomena and provide scientists with a language for dealing with complex phenomena. 
5. Theories of complexity shift from a paradigm of classical Newtonian/Cartesian science to a non-reductionist paradigm, in direct opposition to linear, atomist, determinist and reductionist explanations of the world. 
6. Complexity studies prefer organisation over static structures, ‘relationships over entities’, stochastic above determinist mechanism, and phenomenon in its context over isolated objects. 
7. Complexity theories express the limits of human understanding in relation to complex natural and social phenomena and problematizes instruments and strategies used to model the relation between natural and formal systems. 
8. Theories of complexity devise few problem-solving tools and clear-cut solution kits, but rather expose, challenge, and problematise the assumptions of conventional theories and practices. 
9. Theories of complexity influence the way in which we do science and how we practically implement scientific findings and demand methods of inquiry and knowledge generating practices that draw from a plurality of epistemologies or positions. 
10. Complexity discourses affect and cross-pollinate a variety of fields of study.

Describing Complexity

The general, interdisciplinary approach to complexity of Morin and Cilliers leads to Preiser's second problem: knowing and describing complexity. Study of complex systems requires a complex, interdisciplinary, integrative theoretical approach that remains critical of each approach used, challenging its own knowledge generating practices. Complex study cannot ignore its own complicity in and influences on the complex systems it studies. Once inside a system to observe it, the observer must account for being there. This is always a problem inside the problem one is trying to observe, understand, and describe. Complexity, in other words.

Preiser lists five descriptions of complex empirical phenomena that seem not to fit the traditional "Cartesian/Newtonian prescriptions of analysis" (74):

1. Openness - Complex systems are open to their environments -- exchanging energy, matter, information, and organization -- so that according to Cilliers clearly defining the boundary of the system is problematic and is often "a function of the activity of the system itself, and a product of the strategy of description involved".
2. Relationality, non-linearity and non-equilibrium - Complex systems are constituted relationally both inside and out, and the relations between internal components and the environment are dynamic, manifold, and nonlinear, which means that output is not directly proportional to input. The behavior of interactions is to some degree unpredictable and uncertain and functions in a state of asymmetrical non-equilibrium. The survival of complex systems depends on this nonlinear relationality.
   
3. Non-homogeneity - Complex systems are comprised of a number of heterogeneous components with multiple, dynamic pathways among them that create rich and diverse interactions which become too complex to calculate. The elements and interrelationships change over time and scale.
   
4. Emergence & complex causality - Because of the dynamic nature of internal and external interrelationships, complex systems manifest emergent properties that can be understood only in terms of the organizational structure of the system and not in the properties of the components. Emergent phenomena depend on and yet are independent of constituent parts and display certain properties:
   
1. radical novelty: emergent phenomena are neither predictable nor deducible from micro level components, which are necessary but insufficient for understanding emergent phenomena.
2. coherence: emergent phenomena are integrated wholes likely to maintain some identity over time.
3. macro level: emergent phenomena occur at a macro level compared to their micro level components.
4. dynamical: emergent phenomena are not a priori wholes but gradually appear as a complex system dynamically develops over time.
5. ostensive: emergent phenomena show themselves and are ostensively recognized in terms of their purpose and meaningful behaviour.
Complex systems operate through both upward and downward causation, such that emergent properties are the result of the organization and interactions of constituent parts at the micro-level but also in turn cause changes in the constituent parts.
5. Self-organisation - Complex systems are able to evolve within themselves their internal structures in order to cope with their environments. They are able to learn and to adjust to ensure their survival.

Complexity, then, is first a problem of observing and studying complex phenomena that themselves have incalculable interrelationships and interactions and unpredictable properties and then second being able to observe only from the inside as part of the system. The observer has no objective, outside point of view, but only a subjective, inside point of view that affects -- often non-trivially -- the complex system under observation and study. The presence of a thermometer changes -- however slightly -- the temperature of the body being measured.

Following Morin's lead, Preiser  concludes by positing complexity not as a theory but as a pilot notion "that allows for an integrative theoretical approach that remains critical of the scientific assumptions that emerge from studying complex phenomena ... exposes the limits of each discipline and ... [problematizes] the status of knowledge and knowledge generating practices (75).

In my next post, I will use my own house and body to explore these five characteristics of complexity.