- Supersizing the Mind: Embodiment, Action, and Cognitive Extension
Andy Clark’s works set the standard for theoretical cognitive science corroborated by wide-ranging surveys of empirical results. His philosophical claims are always simultaneously provocative and productive: he not only criticizes but always offers new directions for research. Supersizing the Mind is Clark’s most recent book of this kind.
Supersizing the Mind supersizes the arguments made by Clark and David Chalmers in their seminal 1998 Analysis paper, “The Extended Mind.” That paper (handily included as an appendix to this book) argued for the then radical, but also immensely compelling, claim that cognition can extend not only outside the skull, but outside the body. The arguments in that paper made use of a simple parity principle: “If, as we confront some task, a part of the world functions as a process which, were it to go on in the head, we would have no hesitation in accepting as part of the cognitive process, then that part of the world is (for that time) part of the cognitive process” (77). Thus, they argued, a man Otto who reliably used a notebook to guide his actions can be said to have beliefs that are recorded in the notebook—at least when using the notebook in certain kinds of ways. The parity principle provides a powerful argument for extending cognition; those who oppose the extended mind hypothesis must offer some non-arbitrary reason to rule out these particular instances of potential extension. Many have tried, of course, and in this book, Clark sets out to explain, emend, and defend the hypothesis.
The book has three distinct parts. The first provides the context for the extended mind. Clark reviews research and provides a theoretical perspective, that reveals (mostly) human minds as economically exploiting their environment in ways that minimize the demands for internal representation when the environment itself can do the required work. Thus, “the canny cognizer tends to recruit, on the spot, whatever mix of problem-solving resources will yield an acceptable result with a minimum of effort” (13; Clark’s emphasis). This can include not bothering to remember elements of the environment if one can merely look at them to check on their status. Clark also reviews the kinds and complexities of mind-and-world interactions and entrainings. This guiding principle of cognitive economy also elegantly relates Clark’s work on embodiment to his work on the extended mind; the main lesson of [End Page 415] embodiment is that the body and its environment can solve problems in place of complex representations or algorithms.
The second part of the book defends the extended mind hypothesis from a range of critical responses to the original paper. Most of these criticisms are, as Clark shows, off the mark. For example, they include the claim that the notebook referred to in the original paper is, on the extended mind hypothesis, itself conscious or cognitive. Clark patiently explains that it is the agent and the notebook that have cognition, never the notebook alone. He unequivocally clarifies that the human body is necessary for human cognition, the body is cognition’s source and font, and that there is no inconsistency between his work on embodiment and the extended mind hypothesis. Here, Clark’s ecumenicalism—his willingness, shared by most other cognitive scientists, to approach the study of mind from a very diverse range of perspectives, bringing to bear many different kinds of models—comes to the fore. Extending the mind is an important strategy but only one that thinking agents use.
The final part of the book is a different kind of clarification: an exploration of the limits of embodiment and the extended mind. This begins with Clark’s criticism of proposals by Noë and others that perception is always actively constructive and dependent upon motor engagement with the environment. Clark rejects this by—unsurprisingly—proposing a mixed model of perception in which the role of motor engagement in perception requires that there be some dedicated and stable representations of which perception can make use. These explain...