This report documents the results of the research on Artificial Life. The project was aimed at using simulations of evolving systems to help understand evolutionary creativity. There were two main products that serve these aims:
Both of these products were extensions of earlier research.
These products are important for two reasons. First, they help answer one of the deepest unanswered questions about evolving systems: the source of their creativity. The proof that we do not yet know the answer to this question is that nobody knows how to make a model of an evolutionary system that exhibits anything like the creative power observed in the actual history of life. The products developed with Murdock support are an important tool for understanding and assessing an evolving systems creative activity. Second, and just as important, the simulation and statistics provide a concrete illustration how new quantitative methods play a constructive and creative role in scientific and philosophical interpretations of evolution. Philosophers and scientists trying to understand the deep questions about evolution are increasingly forced to use new computer-based methods — what could be thought of as "computational thought experiments." There are no textbooks written about these methods; they are still under active development in the research community. Thus, instructors wanting to introduce these new methods to their students face a serious quandary: What is there to bring into the classroom? The products developed with this grant help solve this quandary.
Each term since the summer 2002 Murdock-funded project has seen concrete indications of the pedagogical usefulness of this work. The simulation of evolution and the methods for visualizing and quantifying evolutionary creativity played a central role in my advanced seminar on emergence last fall; they played a crucial role in two segments of my current philosophy of biology class at Reed; and they will play a central role in the advanced seminar on the metaphysics of life and mind that I will teach this coming fall.
The concrete accomplishments of Murdock support during 2002 are these:
M. A. Bedau and M. J. Raven. 2002. Visualizing adaptive evolutionary activity of allele types and tokens. In E. Bilotta et al., eds., Workshops Proceedings of the Eighth Artificial Life Conference, pp. 119-130. University of New South Wales, Sydney.
M. J. Raven and M. A. Bedau. 2002. A general framework for evolutionary activity statistics. Submitted to the European Conference on Artificial Life, ECAL'03.