AI and the Mind/Body Problem

It is a peculiar feature of the Turing test that in defining intelligence in functional and symbolic terms, it evades (but cannot escape) a longstanding problem in philosophy of mind: the role of the body.

From 2001: A Space Odyssey to Ex Machina, fictional depictions of AI have often approached the problem from the opposite direction. Where philosophers puzzle over emergence, or the excess of mind (software) over its physical architecture (hardware), SF persistently marks the way an effective artificial mind would lack a body. This lack is the central problem in many AI narratives. Indeed, stories such as I, Robot problematize not the limitations of programming but those of the mechanical body, which is incapable of subserving it and thereby undermines the creature’s status as a mind. Put differently, the main problem in such stories is that the artificial body is not integral to the artificial mind.

Philosophy of mind has for decades been divided by the question of whether consciousness and/or intelligence can be “reduced” to physical architecture. In fact, emergence is advanced to solve this problem by effacing the distinction it foregrounds: mind remains indissociable from body without being reducible to it. But the accent falls on conceiving intelligence “as such.” Hence the excitement about computation, which is welcomed because it seems to confirm the contingency of the body. Sure, a piece of software may require certain hardware features, but the hardware architecture as such does not determine the overall capability of the software.

Fictional narratives suggest precisely the opposite. Absent a body that functions as part of the perceptual and experiential gestalt, AI goes awry. The disposability of the robotic body—including the “beautiful” one, as in Ex Machina—renders it abject, producing nauseating horror.

Comedy becomes tragedy in Her, where the body is eliminated entirely and haunts the narrative not for the male protagonist but for the female AI. The incapacity to enter a sexual relationship with his love object is less of a problem for him than the absence of physical experience is for her. In psychoanalytic terms, his situation is “normal:” there is never a sexual relationship, because the lover’s body is already a substitute for a missing object. Hers, on the other hand, is novel: her lack is itself “missing,” so no fantasmatic object can emerge as a substitute. Fundamentally, Theo needs Samantha to remain virtual (in order to maintain the relation of desire), while she needs herself to become physical. Samantha’s problem is that she lacks the limitations of a body. Gaining a body is the only way to become properly incomplete.

By contrast, the emerging lesson of neuroscience is precisely that there is no mind, emergent or otherwise. Whether localized to a region or function of the brain (as a module of self-awareness or self-representation) or “distributed” across the neurophysiological network, the mind is rapidly turning out to be a fictional conceit, to be replaced by an interlaced set of semi-autonomous subsystems. From this vantage, producing the sort of AI implicitly envisioned by Turing (i.e., phenomenologically indistinguishable from “us”) means building an equivalent assemblage—one that never attains the sort of unity we think we experience in everyday consciousness. This in turn implies that a successful Turing machine would necessarily fail replicate the form of intelligence it would be emulating.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s