Simon opens a central chapter of Sciences of the Artificial with the tale of an ant picking its way across a windblown beach toward its nest, encountering many small obstacles along the way. To an onlooker, the ant’s path might seem to reflect a complex internal process. According to Simon, though, the path’s “complexity is really a complexity in the surface of the beach, not a complexity in the ant” (1996, 51). He extends the moral of the story to human cognition and behavior: “Human beings, viewed as behaving systems, are quite simple. The apparent complexity of our behavior over time is largely a reflection of the complexity of the environment in which we find ourselves” (1996, 53).
This is a striking claim, not entailed by adaptivity; for, various highly organized systems might all nevertheless be organized so as to behave adaptively, and one might even think it necessary that a system have a complex structure in order to respond effectively in a wide range of task environments. I emphasize the claim of inner simplicity at this point in the discussion because the claim seems to stand so deeply at odds with the embodied program, which takes human cognition, in all of its nuance and complexity, to be largely a function of fine-grained facts about the distinctive structural and causal organization of the inner workings of the human system.
Throughout II.B, I have portrayed Simon as a representative of a brand of computational functionalism often derided by embodied theorists for its utter disregard for the material basis of cognition and for its tendency instead to fetishize such projects as a formal analysis of task domains and the exploration of algorithms of search through them, algorithms that have no connection to distinctively human, body-based strategies.
II.C Embodied functionalism and contingency-making
The preceding portrayal significantly misrepresents Simon’s approach to the mind, and here’s a way to begin to see why. One of the most important aspects of the embodied view is a commitment to what might be reasonably called ‘contingency making’. On the embodied view, contingent facts about our bodily existence color and shape human cognition (and presumably, something similar holds for other creatures as well). Humans have a particular body shape and orientation and particular networks of muscles that interact in a finely-tuned orchestra of counterbalancing forces to keep the organism alive and functioning in the face of ongoing perturbation by environmental forces. Moreover, patterns of neural firings involved in such orchestration provide resources to be co-opted for other cognitive purposes, from the encoding of items in working memory (Wilson 2001) to the metaphorical understanding of abstract concepts (Lakoff and Johnson 1999). On this picture, the resources used in real time by human organisms – in all of their ad hoc, makeshift, exapted, kludgey, but stunningly effective, glory – bear the marks of this distinctive body-based orchestra. Thus, according to embodiment theorists, there is simply no way to understand what’s going on with human cognition absent the thorough investigation of these contingency-making, bodily contributions.
Simon’s picture, too, is rife with contingency making, arguably of the bodily sort (although he does not often use such language), which places him – or at least the second cluster of his views – squarely in a camp with the embodiment theorists. In fact, one might reasonably contend that Simon’s uncompromising interest in the contingencies of the human case helped to set the stage for the embodied movement.
It is the burden of Section III to make a detailed case for this claim. But, first, a door must be pried open. Don’t embodiment theorists routinely criticize, reject, and even demonize computationalism, functionalism, and multiple realizability? How could I possibly, in all seriousness, present Simon as a proto-embodiment theorist, given his clear commitments to computational functionalism and related views, as documented above? The answer requires that we revisit these much-derided orthodox positions and clarify their relation to embodied cognitive science. In doing so, we clear the way for a proper understanding of Simon’s view and of embodied functionalism, generally speaking.