ABSTRACT
Holland (1998) takes a similar tack, but with a rather different emphasis. He holds that emergent patterns are predictable and regular. He accepts the long-term unpredictability of complex systems but claims that this does not matter because the scientist should focus on levels of detail and time spans for which predictability is possible. He dismisses the importance of long-term unpredictability and holds that it is possible to get by through focusing on the short term. Holland holds that if the current state of a model is specified then that state, processed through the structure of the model, determines the next state, which in turn determines the state after that. He claims that the only uncertainty relates to the appropriateness of the level of detail in which the current state is specified, and to the faithfulness of the correspondence between the model and reality. In other words, uncertainty lies in the interpretation that the modeler makes of a deterministic reality. The future, for him, is fully and unambiguously determined. It is a hidden, pre-given reality. Presumably, unpredictability lies in the models of reality being used, not in the unpredictability of that reality itself. For him, complexity theory is not a fundamental challenge to the dominant scientific discourse but simply another model that is useful in the pursuit of “science as usual.” He retains the centrality of efficient causality but adds to it notions of formative cause. Holland’s interpretation of complexity theory is quite clearly made within the framework of Formative Teleology.
