What does it mean for something to be alive? Why are some things alive, and some not? Is there only one way of being alive, or many different and distinct ways?
These are the most fundamental questions which should define biology--and thus, in turn, Artificial Life--as a distinct field of enquiry. Yet the answers are not straightforward, and certainly not the subject of clear consensus. With the demise of vitalism, it was recognised that living systems are made of the same kinds of things as the non-living. It follows that the distinctiveness of living systems must arise from the particular manner or modes of organisation which they embody. But it has proved surprisingly difficult to articulate what this mode (or these modes) of living organisation consist in.
The project of autopoiesis was motivated inter alia by the attempt to answer just such questions. It proposes that, at some level, and in some way, all living things share a common organisation; and that this is characterised by two intimately entwined closure properties:
The central claim of autopoietic theory is that autopoietic organisation is necessary (though hardly sufficient) for the emergence of living phenomenology. If that claim is taken seriously, it naturally defines a particular programme within Artificial Life: namely the attempt to realise autopoietic organisation in artificial, especially computational, media. Indeed, the attempt to do just this has been an integral part of work on autopoiesis from its very inception. Accordingly, we will turn now to the historical setting and context for the formulation of autopoietic theory, and its seminal casting in a specifically computational context.
Copyright © 2004 All Rights Reserved.