Expertise and the Conscious Mind

41xvC4cQ40L._SX323_BO1,204,203,200_Over my last few posts, I’ve been working my way through Hubert Dreyfus’s and John McDowell’s accounts of concepts in perception and action. I tend to follow McDowell in seeing the conceptual as co-arising with perception, as opposed to Dreyfus, who sees practical action as largely nonconceptual and automatic. However, in seeing the conceptual as basic to perception, one need not assume a definition of concepts that requires explicit propositions, declarative speech, or conscious rule following, as Dreyfus suggests.

Barbara Gail Montero is persuasive on this issue in her view that language can and does mediate sensory attention, but that we as actors are quite capable of attending to details without articulating in words what, exactly, we are attending to. “For example,” writes Montero, “you are not or at least need not be thinking this sunset exhibits the most spectacular array of colors I have ever seen. Nonetheless, you may be attending to the scene before you.”[1] Montero continues, “According to Dreyfus as well as a number of psychologists the expert performs best when she does not conceptualize her actions. I take conceptualization roughly to be a process of understanding one thing as falling in a certain category, a process which can be, though is not necessarily, verbalizable.”[2]

Conceptualization and verbalization are on this view discrete tasks. In other words, Montero is suggesting that meaningful conceptual attention, interleaved with empirical particularity, makes attending without articulating possible, if not a common mode of engagement with environments. McDowell agrees when he writes, “Our rational mindedness is in our sensory awareness and intentional bodily movements, whether or not they are explicitly self-conscious.”[3] The order of conceptual syntheses that make available the meaning and appearance of phenomena are in this way tied to and ingredient in the causal order of the larger ecological system, even at the level of motor intentionality.

For her part, Montero goes farther than McDowell in suggesting that, not only is the conceptual a key component of skillful action and absorbed coping, but argues that the zone of skillful action is itself punctuated by moments of deliberation and second-order perception. “Contrary to the idea that expert action proceeds automatically,” writes Montero, “I argue that experts in fact present a model of Socratic rationality, exemplifying both grounded knowledge of their actions and self-awareness.”[4] In other words, argues Montero, “effort, thought, bodily, awareness, and other such psychological factors are generally integral to the smooth, apparently effortless execution of expert-level skills.”[5]

In supporting the view that expert action often involves conscious monitoring, Montero paints quite a different picture from those of Dreyfus and McDowell. However, Montero’s point is not that unconscious execution of movement plays no role in expert action—it clearly does—her point is more that modes of self-reflective attending to action are just as trainable, and are just as advantageous to accomplishing a goal, as are the movements themselves. Moreover, Montero suggests that the real conflict may be between declarative and nondeclarative modes of attending and monitoring, rather than between thinking and absorbed action per se.[6] In other words, Montero argues for nondeclarative modes of sensory and cognitive attention, leaving “open the possibility that we may have concepts that are not verbalizable.”[7]

But in what ways can we justify the claim that skilled attending without discursive articulation is possible? One path forward is to suggest that, if language is a medium of thought, it is not the only medium of thought, and, closer to the argument of this paper, it is not the only medium through which the conceptual emerges. One can imagine, for example, a nonverbal thinking in the mode of geometric images or graphic movie-like scenes rather than words (see for example my earlier post on the artist Kim Jung Gi). There is no reason to assume that thinking cannot take place through a multi-modal display. More to the point, what is required here is not the ability to express things in words, but to make conceptual judgements through a variety of a practices and communicative processes. For such images to convey meaningful content, they would still need some conceptual comportment that could deliver an understanding of meaning, with or without exclusively verbal expression.

In other words, embodied action on this account still deals in a kind of judgment, though not necessarily linguistic, expressively conscious, or reflectively distanced kinds of it. In this context, Montero appeals to a class of nondeclarative concepts that give meaning to intuitive response and expert action. In other words, for Montero, the intuitive is the site of sedimented conceptual understanding, geared towards the solicitations of a specific environment.[8] (This is also what McDowell, appealing to Aristotle, calls second nature, or “cultivated rationality.”[9])

On this view, the embodied conceptual structure of intuition, as found in, for example, learned movement, may be separable from the abilities tied to language use. A judgment is more like a habit in this sense, an automatic and preconscious execution of context-bound semantic associations, acquired in experience. In fact, it appears from this view that the nonlinguistic capacity for conceptual discrimination is the very condition of possibility for language, rather than the other way around. The linguistic on this view is a translation of a prior conceptual mediation, rather than a prerequisite for conceptual competence, as is sometimes argued.

At this point, the reader can perhaps anticipate that Dreyfus, still following Heidegger and Merleau-Ponty, is skeptical of the idea that all intelligible perception and skillful coping is implicitly conceptual.[10] Again, though, Dreyfus is concerned with a specific kind of context-invariant, propositional, and linguistic concept—the kind of concept that could never adequately capture the particular singularity of actual contextual events—as seen, for instance, when he invokes Robert Brandom’s idea that to grasp a concept is to master the use of a word.[11] Separating expert action from word use is the right move, but when Dreyfus then appeals to a “space of motivations” in the place of a “space of reasons,” Dreyfus drops us back into a zone where we are left to explain how solicitations and motivations from the environment draw us to act in intuitively immediate ways. In other words, we are back to the question of intuition.

[1] Montero, Thought in Action, 44.

[2] Ibid., 45.

[3] McDowell, “The Myth of the Mind as Detached,” 56.

[4] Montero, Thought in Action, 31.

[5] Ibid.

[6] Ibid., 43.

[7] Ibid., 45.

[8] See Montero, Thought in Action, 233–236.

[9] McDowell in “The Myth of the Mind as Detached” writes, “The concept of second nature applies to any responsive propensities that are not inborn or provided for by ordinary biological maturation but acquired through, for example, training,” 51.

[10] Dreyfus, Skillful Coping, 110.

[11] Ibid., 115.


  1. why conflate all judgment with conceptualization? why not just talk in terms of varying kinds/blends of know-how/skills?


      • ha yeah, I guess some of us don’t see a lacuna there and some do, so it goes I suppose.


      • That little gap between environmental solicitation and intuitive response, you find that compelling?


      • no I’m not with Bert there (M-Ponty and all get too close to pantheism for my tastes, for me it’s more here affordances/ resistances via Gibson) so much and I agree with Gail that conscious reflection can, and often does play some vital role in our doings, but this idea of concepts that aren’t conscious (can’t be said) seems like an analytic hangover. I don’t think that for every action I take there needs to be some kind of belief at work and that the giving of reasons/beliefs (or categorizing things, etc) is just a kind of know-how not an underlying/overarching principle or the like. In some sense not so much a flat ontology in the OOO mode but a flat epistemology, no special realm (pace Brandom and all) for the social/conversational.


      • OK. But the direct perception / affordances line has the exact same gap. How do you get to perceiving affordances without conceptual inference?


      • I’m just about finished with this talk and now I’m even more confused.

        Clark himself says “percepts are always at least weakly conceptualized” and that perception, understanding, and imagination (construed as a kind of inferential/anticipatory model for predicting future events) all arrive at the scene together in our experience, implying (and he even says this) that action always involves belief (because it involves inference).

        There’s other stuff, too: For example he has a whole bit on top-down processing views of cognition that nix the empirical/sense datum/atomistic view of building up experience out of small bits at a time. All of this seems consistent with what I’m saying. It’s even in broad strokes consistent with a certain Kantian/Merleau-Pontian view of mind. Is there some contradiction here I’m not seeing?


      • I think it is consistent with what you’re saying, see also


      • Ah, OK. I thought Clark was being marshalled as a counterpoint. Do you disagree with Clark too, then? Trying to get my head around the differences.


      • if you get a chance check out the Ward, especially ( the rest is important as he talks thru McDowell) around 24 mins in where he contrasts M-Ponty vs Kant in terms of contingent forms of bodily activity vs transcendental ego as the bundlers of signals.


  2. so these are just some examples (I wish they wouldn’t talk of “storing” and the like but old habits are hard to break, we are actually just making circuits) and there is a lot of science yet to be done but I think something like this will add up to most of an account of how we make our ways/in of the world. Just as face perception is an assembly of neural networks (various neurons all tuned and firing in response to various features, some to shapes, some to angles, motion, etc) so are other skills, other manipulations.


    • You know I’m for an increasingly rich neuroscience. At the same time, I feel like we’re switching registers here. However detailed the third-person description gets, it’ll never show up for us that way on the side of our phenomenological experience, right? So if we want to talk about learning, skill building, etc. we need a language adapted to that level of activity, in other words, philosophical and phenomenological language.


      • I don’t see any in principle reason that any particular vocabulary can’t be put to any use, and even along the lines you suggest I don’t get what it accomplishes to (over) extend the meaning of concept this way anymore than I get extending mind (or communication) in many of the ways people do in relation to say plants or the like, which in addition to taking away from specificity seem to be making philosophical language serve scientific-type means. If we are talking making a more pragmatic move to talking heuristics in the interest of easy/ready adoption that’s fine by me, part of why earlier I asked about the tasks/skills that we might think of as philosophical as opposed to more everyday.


      • What a casually dismissive review. Here’s a better one for you:

        As for overextending the notion of concept, I reject the assertion, naturally. It doesn’t take away specificity or overreach into scientific territory, unless you’re already wanting to grind that particular axe before we even get off the ground.

        It’s one of the oldest philosophical notions, after all.


      • that longer review doesn’t address the central concern of the shorter one I would guess because Mendola is likely an analytic philosopher (they like to think that analytic philo is the defining mode of human-being and if we can’t find overt signs of that it must because it happens in some unconscious version of what they do explicitly) but leaving that aside, how is this “Montero argues for nondeclarative modes of sensory and cognitive attention” not a kind of 3rd person account, and don’t the folks above just flesh out nondeclarative modes of sensory and cognitive attention?


      • A nice list of assumptions. It’s her methods that are first personal.


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 )

Facebook photo

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

Connecting to %s