In "Memento's Revenge", Andy Clark responds to some of the criticisms of the Extended Mind thesis (that particular kinds of organism/environment coupling count as cognitive processes; cf Clark and Chalmers 1998), and further clarifies this claim. While sympathetic to the general thrust of the thesis of Active Externalism, I nonetheless argue the following:
1) The concessive remarks that Clark makes concerning the essential role that development may play in cognition suggest that we may need to add a further requirement, that of "proper developmental history" to the conditions (reliably available, typically invoked, automatically endorsed, and easily accessible) that must be met for an external process to qualify for equal treatment with respect to cognition.
2) A simple thought experiment, and reflection on the use of the concept of "organism" in stating the thesis, reveal that Active Externalism is not so much about location, but provenance. The thesis then amounts to claiming that items that were not selected to perform a cognitive function can nevertheless do so. Thus, there is a conflict between the Extended Mind view, in which ahistorical functionality is taken to trump all, and teleosemantic-cum-evolutionary theories of mind, in which history of selection is of ultimate importance. Arguments that defend the latter's historicism (against, e.g., Swamp Man objections) can be used to cast doubt on the support offered for the Extended Mind view. This is a special case of a more general conflict between the Extended Mind view (and related positions, such as virtual embodiment; cf “The Twisted Matrix: Dream, Simulation or Hybrid?”, Clark forthcoming) on the one hand, and theories that place emphasis on actual embodiment, on the other.
3) Central to the Extended Mind view is The Parity Principle: "If... 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." As Menary (forthcoming) has argued, this principle stands in need of clarification. I show that when considering the counterfactual in the Principle, one gets different results depending on what one does and does not keep fixed in the hypothetical move into the head. Without a prior motivated basis for selecting what aspects of the relevant part of the world will and will not remain constant, the Principle is of little assistance in supporting the Extended Mind thesis.
4) A principled distinction can be made between two kinds of components of cognitive processes: indirect components, that are part of the cognitive process by virtue of having been available as the object of an agent's awareness; and direct components, in which there is no such dependence on availability in awareness. It seems most compatible with our intuitions that only the direct components of cognitive processes can constitute states of the subject. A thought experiment is offered to support this intuition. It follows that Otto's beliefs cannot be located in his notebook, since the notebook's role in Otto's cognitions is indirect. I consider the Clark-esque response that Inga's ability to recall an address, like Otto's, is (or at least may be) by virtue of something being the object of her awareness (e.g. visual images of MOMA or the address in writing; aural images of someone speaking the address, etc.). Therefore, parity is preserved. I maintain that this reply trades on a disanalogy. The relevant processes in Inga are not mental images (even if present), but the neural processes that enable (or even constitute) her ability to recall the address. Inga is not aware of these processes; and even if she were (by self-examination in an FMRI scanner, say), it would not be by virtue of such awareness that she has the ability to recall the address. Thus, the intuitive view that Inga's memories are located in her head, while Otto's are not located in his notebook, is given a principled defense.