At present there is just one definition [2] that attempts to explicitly naturalise the concept of (general, embodied) agency, and it is unapologetically autopoietic-enactivist. This fact constitutes a public challenge to other traditions in cognitive science. A ‘bare’ problem of defining agency remains, even after paring away hard phenomenological and normative problems by limiting the scope of the problem to describing ‘as-if’ agency (i.e. the external appearance of agency). Building on [2], I identify an extended list of criteria that a theory of agency (whether ‘true’ or ‘as-if’) should meet. I argue that autopoiesis is the wrong foundation even for ‘bare’ agency, let alone phenomenological and volitional agency; instead, I recommend starting with an ‘as-if’ definition that relates agency to some theory of embodied rationality, effectively providing a generalised version of Dennett’s intentional stance [9], and taking a step towards a rigorous formal definition.
CITATION STYLE
McGregor, S. (2016). A more basic version of agency? As if! In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9825 LNCS, pp. 183–194). Springer Verlag. https://doi.org/10.1007/978-3-319-43488-9_17
Mendeley helps you to discover research relevant for your work.