Andy Clark’s central, and surprising thesis, is that mind is not what exists between the ears, but is rather the mesh of brain, body, and world itself. At the beginning of Supersizing the Mind, Clark relates an anecdote from Feynman to illustrate this idea. Charles Wiener had expressed delight in discovering a collection of Feynman’s notes and sketches, indicating how wonderful it was to have a record of Fyenman’s day-to-day work. As Clark puts it, Feynman reacted with unexpected sharpness:
“I actually did the work on the paper,” he said.
“Well,” Weiner said, “the work was done in your head, but the record of it is still here.”
“No, it’s not a record, not really. It’s working You have to work on paper and this is the paper. Okay?” (from Gleick 1993, 409)
The point of this anecdote, in Clark’s view, is that the pencil and paper are quite literally parts of the mind in the process of cognition. What is on the paper is not, for Clark, something that merely records a trace of cognition, but rather the brain, body, and these external artifacts are all the cognition. In a manner that immediately brings Morton’s Ecological Thought to mind, it is this mesh that is cognition. Here it’s important to note that there is no Hegelian style idealism here. Clark is not asserting “the identity of substance and subject”, such that substance is subject and subject is substance. Clark’s position is thoroughly materialist. The pencil and paper are material entities. His point is that they are not merely props or tools for cognition, but that entities such as this play a key role in cognition by affording and constraining possibilities of cognition through their use.
Throughout his work Clark’s emphasis is on real-world cognition in insects, animals, artificial lifeforms (like cockroach robots) and humans. Biological and technological lifeforms, argues Clark, perpetually offload problems of cognition on the external environment so as to maximize real-time responses to situations and to minimize “expensive” computation (representation). Why for example, have a complicated mental map of my living room, when I can use the living room itself (as perceived by entities such as humans) as its own best model? In other words, organisms perpetually rely on the scaffolding of the world in their cognition. This scaffolding consists of relatively stable regularities in the environment. In the case of humans, Clark argues, a large part of this scaffolding consists of culture in the form of institutions, technologies, and language. Thus, for example, my iPhone is literally, for Clark, a part of my memory. Rather that relying heavily on internal memory to recall everyone’s phone number and email address, rather than encoding all of the dining recommendations I’ve heard from friends, family, and the media, I can instead simply turn to my iPhone and pull these things up. The iPhone itself becomes a part of the cognitive process. However, Clark’s thesis is much stronger than this. Cultural institutions and technologies begin to think for us. In Being-There Clark gives the example of an office where there are all sorts of subroutines for particular actions (“place the pink form in the bin labeled x”). The institutional structure does not require any centralized planner nor agents that have an overall representation of how the office works, but rather all the subroutines, including their material elements, collaborate in a distributed fashion together to produce a set of regular results. The institution as a whole has cognition in and through its mesh. This mesh wouldn’t be able to function without brains, but those brains are only a component in these cognitive processes. This is what allows us to claim that cultures and societies think. A big part of this thesis, and I can’t develop it in detail here, is that there are a variety of ways in which natural and cultural environments channel and structure cognition.
My intuition is that the thesis of extended mind has tremendous social and political implications (which sadly Clark doesn’t explore in his work as far as I can tell, but which is a boon for all of us working in the vein of OOO). Here I will only bookmark some of these implications, opening a space to develop them in the future. First, if Clark’s thesis about the extended mind is true, we can’t speak univocally about the “human”. Foucault had already recognized this in the close of The Order of Things when he spoke of “the death of man”. In speaking of the death of man, I believe Foucault had dimly glimpsed the death of man thesis (in OT he showed how “man” was the product of a set of institutional and discursive constructions, i.e., what Clark calls “scaffoldings”). If the extended mind thesis is right, then there will be as many different minds as there are brain-body-world assemblages. Marx glimpsed this when he argued that the factory worker and the farmer were two entirely different species in the Manifesto. He develops this further in his chapter on the working day in volume 1 of Capital where he shows how the industrial factory fundamentally transforms the nature of homo sapian existence. Here we sorely need a well developed version (not just nods, but fine-grained analyses) of Deleuze and Guattari’s ethology as developed in A Thousand Plateaus, where the being of entities is understand in terms of what they can do, not by representational resemblances (recall the famous thesis that “the work horse is closer to the ox than the race horse”). Here sorting of entities isn’t based on embodied resemblances, but on capacities to do. When this is meshed with the extended mind thesis, we begin to sort cognizing beings based on extended assemblages involving brains but also world and technologies (this, incidentally, is what allows us to take into account arguments that discuss the role that privilege plays for particular groups and that the absence of privilege plays in other groups).
It will also be noted that part of the scaffolding of what we often call “humans” (our language isn’t fine grained enough to denote all the entities that are here referred to) will necessarily involve all sorts of animals. Part of extended cognition and minds will include the worms, bees, and other small bacteria that make soil possible for farming. It will involve livestock. It will involve ecosystems. However, these perspectives will be reducible. It will not only be the case that humans cognize in and through the agency of other animals (think of Haraway and her dog, together they form a mind), but it will also be that many organisms (grass, cows, dogs, cats, wheat, cane toads, etc) use humans as a part of their minds.
Aside: Some folks have objected that nonhuman objects have no moral status because they are incapable of reason and therefore have no agency. This is a perverse and bizarre justification for not considering nonhuman entities in our ethical deliberations. As Wolfe argues, the issue isn’t whether they are responsible but whether they can suffer. It doesn’t matter whether or not they are capable of deliberation in the same way humans are. The argument for this position– and I believe it’s deeply convincing –actually comes from the domain of the human. In those instances where humans are incapable of rational agency– severe mental illness, dementia, mental deficiency, catatonia, coma, alzheimer’s, amnesia, etc. –we don’t suddenly say that biological experimentation is okay with these folks, or that it’s okay to use them as food or throw them on a fire as fuel. This point categorically recognizes that the condition of whether, in Kant’s terms, to not treat something as a means is entirely independent of whether that other entity is itself a rational agent. Ergo, nonhumans.
The extended mind hypothesis also has a number of implications for feminist, racial, queer, and Marxist thought. If it is true that mind is extended, then a big part of understand race and gender will involve careful and nuanced investigation of the worldly scaffolding that comes to structure race and gender. This will also be the case where class relations are concerned (and, of course, all these things intersect such that we can’t treat any of these categories as overdetermining the others. We need to ask, for example, what the extended mind is that comes to structure sexual institutions. What are the scaffoldings, in other words, that ground heteronormativity? What are the scaffoldings that ground patriarchy and male privilege? These scaffoldings, additionally, should be seen as simultaneously channeling men and women, queer and straight, white and brown, etc. If we don’t engage in object-oriented archaeologies of these scaffoldings then we will be unable to develop universal-egalitarian political interventions that respond to them. There is so much more to say here, but I’m in a rush to make dinner so I’ll bookmark these things for later discussions.