Colleagues Thanks to Alan Dean for that succinct account of Durkheim! However, I still have a problem with the idea that you only need supra-individual concepts in order to explain the 'social' and not the 'psychological'. I do agree with a lot of what is being said about having to make assumptions in order to do simulations. However, even the notion of a self-conscious individual 'self' is a product of thousands of years of social history. In the Egypt of the Pharaohs for example, only the Pharaohs were considered to be individuals, a status of quasi-Godhead (to use a desperately over simplified account for the sake of illustration). I am sure people do realise this in the back of their minds. It might just be that making it more explicit might help in certain cases where problems or paradoxes arise for models. I am also not sure about the idea that an emergent effect is top-down. As I understand it, an emergent effect is a bit like an unintended consequence (previously ages ago I used the example of increasting inequality in the educational experience and outcomes of comprehensive schools being produced by many individual parents' decisions). In this light they arise from many individual decisions but are not what was intended by any particular individual. In fact this is just why I am so interested in simulation! To end with a question: is anyone interested in the relationship of simulation results to Amartya Sen's work on 'preferences' Mel %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%