Computational Autopoiesis: The Original Algorithm

Barry McMullin

97-01-001

This report presents a detailed review and re-presentation of the algorithm for (computational) realization of autopoiesis, originally presented by Varela, Maturana and Uribe (1974). The review is from the perspective of one seeking to reimplement this algorithm. It arises from an on-going project to develop such a reimplementation using the Swarm simulation system. The motivation for such reimplementation is firstly to critically reexamine the phenomenology generated by this model chemistry, and secondly to use it as a basis for internal consistency and clarity; it is then separately reviewed for consistency with the experimental results which originally accompanied it. A number of significant discrepancies are discussed. By kind permission of Francisco Varela, this report also includes, as an Appendix, some previously unpublished documentation and source code (in FORTRAN IV) associated with the original model. By critical consideration of the experimental results in conjunction with this code, an important---perhaps crucial---interaction, not included in any previous description of the model, has been rediscovered. This interaction is presented and discussed briefly.