The lack within AL of an agreed-upon notion of life and of a set of criteria for identifying life is considered. I propound a reflection upon the codified nature of the organization of living beings. The necessity of a guiding notion based on the coding is defended. After sketching some properties of the genetic code I proceed to consider the issue of functionalism as strategy for AL. Several distinctions ranging from plain multiple realizability to total implementation independence are made, arguing that the different claims should not be confused. The consideration of the semantic and intrinsically meaningful nature of the code leads to discuss the “symbol grounding” in AL. I suggest the principle of Semantic Closure as a candidate for confronting both problems inasmuch as it can be considered an accurate guiding notion to semantically ground Artificial Life.

This paper describes a method for a computer exploration of formation of structures based on the network of autonomous units. This method has a biological correspondence with morphogenetic processes. The interactions in the network of autonomous units are modeled by two kinds of forces: repulsive and attractive forces. When and what kind of forces are active at each unit is based on genetic information and environmental factors. Genetic information enables the use of evolutionary algorithms to evolve the interactions and thus to create new structures. Environmental factors provide the needed restrictions for the space of possible structures. Depending on what meaning is given to the units, the system is capable of simulating various kinds of emergent phenomena. For example, in the case where units are interpreted as cells, where the repulsive and attractive forces represent collision and adhesion forces, a formation of multicellular organism can be achieved.

Abstract: We argue that the significance of the spatial boundary in autopoiesis has been overstated. It has the important task of distinguishing a living system as a unity in space but should not be seen as playing the additional role of delimiting the processes that make up the autopoietic system. We demonstrate the relevance of this to a current debate about the compatibility of the extended mind hypothesis with the enactive approach and show that a radically extended interpretation of autopoiesis was intended in one of the original works on the subject. Additionally we argue that the definitions of basic terms in the autopoietic literature can and should be made more precise, and we make some progress towards such a goal.

Key words: Operational limit, operational closure, physical boundary, relational domain, spatial boundary

In this paper, I propose a neo-Heideggerian framework for A-Life. Following an explanation of some key Heideggerian ideas, I endorse the view that persistent problems in orthodox cognitive science result from a commitment to a Cartesian subject-object divide. Heidegger rejects the primacy of the subject-object dichotomy; and I set about the task of showing how, by adopting a Heideggerian view, A-Life can avoid the problems that have plagued cognitive science. This requires that we extend the standard Heideggerian frame-work by introducing the notion of a biological background, a set of evolutionarily determined practices which structure the norms of animal worlds. I argue that optimality/ESS models in behavioural ecology provide a set of tools for identifying these norms, and, to secure this idea, I defend a form of adaptationism against enactivist worries. Finally, I show how A-Life can assist in the process of mapping out biological backgrounds, and how recent dynamical systems approaches in A-Life fit in with the neo-Heideggerian conceptual framework.

Key words: adaptationism, Cartesian, enactivism, Heidegger, significance, subject-object dichotomy.

We present two original computational models – globular universe and autopoietic automata – capturing the basic aspects of an evolution: a construction of self–reproducing automata by self–assembly and a transfer of algorithmically modified genetic information over generations. Within this framework we show implementation of autopoietic automata in a globular universe. Further, we characterize the computational power of lineages of autopoietic automata via interactive Turing machines and show an unbounded complexity growth of a computational power of automata during the evolution. Finally, we define the problem of sustainable evolution and show its undecidability.