Cognition on the Ground by Douglas Maynard (2006)

(Editor's Note: Doug Maynard is a part of my sociological family, being trained at UCSB with many of the same faculty as me (Zimmerman, Wilson, Molotch, Cressey, Shibutani) but graduating with his Ph.D. several years before I arrived.  He has always been a role model of an outstanding scholar and researcher, and is a really nice guy!)

Cognition on the ground

Discourse Studies
Copyright © 2006 SAGE Publications. (London, Thousand Oaks, CA and New Delhi) Vol 8(1): 105–115. 10.1177/1461445606059560
A B S T R A C T Suggesting that much of social science is still wedded to the ‘dogma of the ghost in the machine,’ I discuss my ethnomethodological and conversation analytic approach to the assembly of cognitive objects. It is important to reverse the usual social psychological metalanguage of mind causing behavior, and see how practices in interaction operate to display cognitive states of participants. Two examples are given: one in regard to the assembly of gestalts, including social actions in talk, and the other concerning the production of responses as accountable phenomena in the survey interview. While advances are being made in brain imaging, the connections between the neurobiology of the mind and the practices embedded in human conduct are tenuous at best. For the social scientist, approaching cognition through practices puts ghosts and other ephemera in abeyance and allows for analysis of the detail exhibited in behavior when minds are purportedly at work.
K E Y W O R D S : cognition, conversation analysis, embodiment, ethnomethodology, gestalts
People, Gilbert Ryle (1949: 32) has said, ‘are wedded to the dogma of the ghost in the machine,’ the Cartesian idea and ‘philosopher’s myth’ that mind and body are separate entities such that when people act, there are two separate processes – a ‘theorizing’ or cognitive operation and a ‘doing’ or embodied performance. Whether in philosophy (Wittgenstein), political thought (Pitkin), sociology (Garfinkel), or elsewhere, so many critiques of the ‘ghost’ view are available that it is a wonder why it is so strong. But strong it is. For example, in one of the fields where I teach and do research – social psychology – practitioners (Cartwright, 1979), discursive psychologists (Edwards and Potter, 1992), and historians (Danziger, 1997) alike characterize it as highly cognitivist. By this, they mean 

106 Discourse Studies 8(1)
that investigators are concerned with the subjective and internal states of actors that have causal relationships to what people actually do.
The reasons for the cognitivist preoccupation are interesting though not relevant here except to say that there are good historical accounts for why social scientists, including social psychologists, maintain a cognition-based ‘meta- language’ that frames inquiry (Danziger, 1990, 1997). As I will note in the Conclusion, there is nothing wrong in a preoccupation with ghosts – the presently not seen but nevertheless measurable attitudes, values, norms, or other internal factors such as neurobiological ones that may be partially predictive of behavior in the world. It’s that partial predictability – what goes under the category of unexplained variance – that is the rub. Even with the best models, the most sophisticated technology, and systematic empirical investigations, immense amounts of behavior go unaccounted for. One can take the approach that we need better models, yet more advanced technology, improved methods of data collection, or more precise analytic procedures, and we’ll find and nail those ghosts to the ground. But if one is interested in the panorama of human behavior in all its detail, and wants to do a science that is committed to capturing the range of concerted behavior and the organization it so manifestly displays, then other possibilities present themselves. First, the investigator can do away with the dualisms of mind and body, outer and inner, public and private, and so on, using tools developed in philosophy, political theory, sociology, and so on. This does not necessarily entail critique of cognitivist accounts, but rather, as with phenomenology, it just means being agnostic by not taking a prior position about whether the mind exists and if so, what it consists of. Second, we can appreciate cognition as a manifest part of interactional discourse rather than its progenitor. Such a strategy enables initial appreciation of cognition on the ground rather than in the airy space of heads or brain cells or neural networks. To the degree that we want to know about spaces in heads and cells and networks, this means the possibility of mapping them from what we know about actual behavior rather than the reverse. Along these lines and from my own research, I will give two examples of studying cognition on the ground: how participants may assemble gestalt-like social actions in talk, and how survey researchers manage to measure cognitions.
Gestalts in talk
When teaching courses in ethnomethodology, social psychology, or language and social interaction, I discuss various gestalt figures – organized groupings with ‘intrinsic articulation and structures’ whereby the whole is larger than its constituent parts (Gurwitsch, 1964: 115) – with the purpose of examining what they are as social rather than psychological objects. That is, the literatures on gestalt formation are highly subjectivist, psychological, cognitivist, and sometimes neurobiological in their attempts to explain the phenomenon, and they mostly consist of experimental attempts at such explanation. This is not to

decry the considerable accomplishments of gestalt theorists and researchers, particularly their challenges to the elementarism of structuralist and behaviorist psychologies (Rock and Palmer, 1990). In fact, it is fun to teach about gestalt formation using well-known illustrations including the rabbit-duck, Rubin’s vase (which vacillates between being a vase and two faces looking at one another), and the Necker cube, named after the Swiss crystallographer who saw cubic shapes spontaneously reverse in perspective.
When I teach about gestalts, however, I do so in a particular way. In classes and seminars, I use an overhead screen and handouts, videotaping our sessions. The videotape shows that whatever goes on cognitively when we deal with gestalt objects, these objects only have a social reality in the actions and interactions by which participants assemble them in concert with one another. For example, when we deal with the Necker cube, the first thing I do is invite the seminar members to name the figure they see in the handout or on the screen (‘Ins’ = Instructor; ‘Stu1’ = student 1):
(1) Ins:
Stu1: A box.

(2) Stu1:
Ins: Stu1: Ins:
I don’t know how to explain it exactly, but you can basically see the two [gesture with both hands] panels, and see them either as being in the back of the cube like this [hands in concave shape]
Uh huh
or in the front of the cube. [turns hands to convex shape.] [nods head] Okay.

Whaddya see here. What’s the name of the object.
Ins: Okay. Stu1: A cube. Ins: Cube.
When students call the figure either a box or a cube or both, and the rest of us agree with one another on this name, there is a proposal and ratification structure whereby some three-dimensional perception of the object on the paper and on the screen preliminarily comes into being. Then I ask students to describe ‘how they see’ the box or cube. The transcript at (2) has what I term an embodied telling of a seeing: Student 1 (Stu1) both describes what she sees and instructs the others in how to see likewise.
As she talks, this student has two hands up with palms in front with concave shape; she angles her body, head, and arms, so that her hands, from her own as well as others’ view, can be seen in three-dimensional space with a depth that conveys first a concave shape and then a convex shape that are to be iconic analogues of what everyone can perceive on the two-dimensional paper or the screen in front of the class. In the end, the instructor and the other students, through their own looking, there own bodily positioning, and their verbal and nonverbal acceptance of Student 1’s proposals, align to these purported 

108 Discourse Studies 8(1)

viewings. In these ways, the proximal gestalt perception comes to have not just subjective status but intersubjective status.
Not all such perceptions gain this kind status. Just after Student 1’s telling, Student 2 offers this:
(3) Stu2: Uhm [pause] essentially I’m- sometimes I see it, when I look at it, it looks as though the, the- [gestures with left hand
vertically] it’s taller than it is [gestures accordion-style with both hands] wide, but other times it looks like it’s [draws two hands farther apart] wider than it is tall.
[SILENCE: others are looking at handout or up at screen]
The silence after Student 2’s proposal concerning height and width marks the absence of alignment, and Student 2 gradually withdraws:
(4) Stu2:
It is more clearly [pointing at handout] when I- when I look at
it on my- my sheet rather than that one up there.
Ins: Ah okay. [pause; looking at handout] And can you describe when

you see it as being um taller than it is wide. What- how does it
present that [pause] feature? Stu2: I’m- I’m not sure how it is. Ins: Uh huh.
Stu2: I’ve got a feeling you can’t. Many: [laughter]

At least for this moment, the proposal that the cube is wider than it is taller or vice versa according to its perceptual oscillations has no intersubjective status. There is no mutual agreement, and consequently no social reality to these possible gestalts. That there is no social reality also undermines psychological experience of the gestalt, for ‘there are no criteria for correct or incorrect description,’ as Hilbert (1984: 374) puts it.
But then slightly later, a third student presents an understanding of, and further instructions on, how to see what Student 2 had proposed. Student 3, who is seated close to the screen, gets up and goes to the screen:
(5) Stu3:
If you look at [pause] this [tracing two lines] as the front of
the box [turns to class] Ins: Uh huh
Stu3: then it [gestures with right extended index finger up and down] seems to be taller than it is [two hands doing accordion] wide.
Ins: Okay.
Stu3: [turns to screen and points to

vertical line, Figure 1] But if you look at this as the front of the box [turns to instructor]
Ins: Uh huh

A Link to the full article is provided below: