Chaos and complexity: philosophical implications

This is a first attempt to systematically evaluate the philosophical implications of the phenomena of chaos and complexity. Basing on the results of nonlinear science and fractal geometry, I show that: a) explanation is really different from (and not reducible to) prediction; b) different levels of knowledge and reality actually exist; c) the aristotelian concept of "form" in many senses (although not in every sense) is very similar to the concept of "strange attractor"; d) in nonlinear dynamics and fractal geometry there is a kind of similarity, mathematically definible, which is not reducible to identity (as in classic geometry) and therefore is a good ground for the realistic interpretation of the universal terms; e) authomatical pattern recognition systems based on nonlinear dynamics and their limits demonstrate that human intelligence must have a non-logic component, i.e. intentionality; f) in a sense (although not in every sense) also mathematics can be considered an experimental science.