As "Against 'Distributed Cognition'," or, The meaning of pain

Jacoby: I’m gonna teach you the meaning of pain.
Elizabeth: You like pain? Try wearing a corset.
- Pirates of the Caribbean: The Curse of the Black Pearl

I've been thinking about using the discussion of navigation in Melanesia from Hutchins's Cognition in the Wild as part of a "retelling" of the story of phatic communication -- however I'm not entirely sure how that will work, and as I don't have the book in my hand at the moment, I decided to warm up with a critical look at Hutchins's theory.  The main source I'm looking at is:

• Button, Graham. "Against 'distributed cognition'." Theory, Culture & Society 25.2 (2008).

Button describes distributed cognition as an attempt to extend "cognitivist" ways of thinking out into the world.

This cognitivist way of thinking is derived from computational metaphors and terms like memory, plans, and intentions all figure in this way of thinking with technical meanings layered on top of their everyday meaning.  Button is skeptical of cognitivist thinking per se, and views an attempt to extend it out into the wider world as a compounding mistake.

One of the key themes in "cognition" is the idea that there is such a thing as an inner world that represents things in the outer world.  (The quote from Winnicott in a recent post adds a third neither-inner-nor-outer world.)  Button doubts the usefulness of a "representational" conception of thinking, and argues in favor of shared social meanings.

To form his argument, he makes use of Wittgenstein's analogy of the "beetle in a box":

Because I cannot see into your box or you into mine I just cannot know what is in your box or you what is in mine. They may or may not be the same thing but we have no way of telling. The only thing we could say of the beetle is that it is what we are calling ‘the thing in the box’, but whether or not that thing is the same we will never know, and even then, there may not be anything inside the box. As with the beetle in the box, my inner sensations of pain are just not available to you, so become irrelevant in our descriptions of pain. In this respect, we can know that someone is in pain without recourse to their sensations; we know that someone is in pain because of the circumstances in which they are in, or because they are acting as if they are in pain. - Graham Button, Against ‘Distributed Cognition’, p. 97

Whereas communication about any private world breaks down, communication in terms of shared patterns and norms works just fine.  This opens a view on communication that can only be contextual, and this seems like a way of looking at things that is "begging the question" of phatics.  But I think we need to be a bit careful.

Specifically, I don't think the idea of phatics is making the same assumptions as distributed cognition, but it is talking about a rather similar domain.  Whereas distributed cognition would refer to shared memory that in inherent in an artifact like a chart -- for example -- phatics would refer to the means by which communication (and/or other "relations whose obtaining entail the existence of their relata", to paraphrase Harold Langsam) become possible.  As I put it in my thesis, phatics and what we're now calling "phatic techinques" does not simply rely on the concept of context but looks at how "a context" is created.

If there are assumptions here, they are not the standard assumptions of computation -- nor those of linguistics, semiotics and dialectic -- precisely because phatic studies looks into the "genealogy" of those concepts and disciplines.  Accordingly, the assumptions we make are (more likely!) those of genealogy or Simondonian onto-genesis (ontogenèse).

One place where I think Button really dates his critique is with the remark: "We cannot observe brain processing and we cannot measure it." It seems to me that if that was true in 2008, it is increasingly not true now.  Not only is brain processing observed and measured, we are seeing the rise of "brain interfaces".  Button describes the "problems" that pertain to correlating neural activity with cognitive processing, but the existence of such problems gives a raison d'etre to a certain science.  (Whether cognitive science actually takes up the challenge is another question entirely.)

Button is (after all) aware of direct observation of brain activity, he just doubts if it is relevant or useful.  Note the classically "phatic" example embedded in this quote:

Correlating data about brain processing with data about the social world does not tell us anything about how that social world is made up and organized. It is merely to say that, when engaged in an activity such as greeting someone, there is a pattern of electrical activity going on in the brain. It does not tell us anything about the organization of greetings in our culture.  The problem with ‘distributed cognition’ is that, like many other explanatory frameworks in the human sciences, it looks outside of the setting in order to explain how the setting is being organized. It thus looks to the brain rather than to the setting itself. - p. 102 (emphasis added)

The anti-brain stance is a bit misplaced (the brain is, after all, a part of the situation), but that's a relatively small point.  The basic idea is summed up well in this way: "[Harold] Garfinkel’s point is that settings are self-organizing, that the social action is ongoingly ordered as it is conducted."


From the third note in this paper:  "The point I want to hold on to is that the inner sensation of pain is not at issue in the description of pain."

To put it otherwise,

Morty: Well, I got news for ya — he's saying it ironically.
Bird Person: No, Morty. Your grandfather is indeed in very deep pain. That is why he must numb himself. - Rick and Morty, S01E11