Embodying the mind by extending it

To subscribe to the embodied mind (or embodiment) framework is to reject the view that an individual’s mind is realized by her brain alone. As Clark (2008a) has argued, there are two ways to subscribe to embodiment: bodycentrism (BC) and the extended mind (EM) thesis. According to BC, an embodied mind is a two-place relation between an individual’s brain and her non-neural bodily anatomy. According to EM, an embodied mind is a threeplace relation between an individual’s brain, her non-neural body and her non-bodily environment. I argue that BC can be given a weak and a strong interpretation, according to whether it accepts a functionalist account of the contribution of the non-neural body to higher cognitive functions and a computational account of the contents of concepts and the nature of conceptual processing. Thus, weak BC amounts to an incomplete version of EM. To accept a weak BC approach to concepts is to accept concept-empiricism. I raise four challenges for concept-empiricism and argue that what is widely taken as evidence for concept-empiricism from recent cognitive neuroscience could only vindicate weak BC if it could be shown that the non-neural body, far from being a tool at the service of the mind/brain, could be constitutive of the mind. If correct, EM would seem able to vindicate the claim that both bodily and non-bodily tools are constitutive of an individual’s mind. I scrutinize the basic arguments for EM and argue that they fail. This failure backfires on weak BC. One option left for advocates of BC is to endorse a strong, more controversial, BC approach to concepts. Rev.Phil.Psych. (2012) 3:33–51 DOI 10.1007/s13164-012-0087-2 Earlier versions of this paper were presented as comments on Ian Hacking’s third Descartes Lecture, at the University of Tilburg, on October 8, 2010; at the Conference on Embodied Mind: Perspectives and Limitations at the Radboud University in Nijmegen, on October 28, 2010; at the Workshop on New Directions in Cognitive Science at the University of Bucharest on June 20, 2011. I am grateful to Stephan Hartmann, Harold Bekkering and Radu Bogdan for inviting me to deliver a paper at these meetings. I am also grateful to Ian Hacking, Martin Kusch, Guenther Knoblich, Nathalie Sebanz, Dan Dennett, Nicholas Humphrey, Dan Sperber, Frédérique de Vignemont for discussions on the topic of this paper. I am grateful to two anonymous referees for this Journal for their useful criticisms and especially to Adrian Alsmith for his acute criticisms and comments. This work was supported by a grant from the French ministry of research (ANR-BLAN SOCODEV). I dedicate this paper to the memory of my friend Marc Jeannerod who died on July 1, 2011. P. Jacob (*) Institut Jean Nicod, ENS/EHESS/CNRS, Paris, France e-mail: [email protected] Author's personal copy