Knowing Complex Sytems: the Limits of Understanding
Paul Cilliers
Summary

In different contexts there are different criteria for what qualifies as sound knowledge. To a large extent, science still makes use of a traditional understanding of knowledge which derives from Enlightenment thinking. This approach relies on objective experimentation, logical deduction and reductive thinking. Despite the usefulness of this approach, it cannot be used with success when dealing with complex systems. Complexity theory realises this, but much of the work done in complexity reverts to the traditional reductive criteria. It is argued here that an acknowledgement of the complexity of the issues we deal with implies an acknowledgement of the limits of the knowledge we can have about them. A complex system can be given several different descriptions which are not reducible to each other, but are not arbitrary either. Perspectives from the social sciences and humanities should not be made to fit traditional scientific criteria, they should be used to reflect on these criteria critically.

In this paper, following Morin, a distinction is made between “restricted complexity” and “general complexity”. It is shown that restricted complexity does not escape the old rationality. The introduction of a more radical understanding of complexity leads to a critical position which does not allow “scientific” knowledge to trump all other forms of knowledge. There is an inescapable normative dimension to all things complex. This implies that the claims we make about complex things are always provisional and limited. Such claims should thus be made with a certain humility.