Defending Alva Noë against Jesse Prinz

If you haven’t noticed, Alva Noë’s enactive theory of perception has been taking a beating lately. Many philosophers have pushed back against Noë’s supposed strong claims about the constitution of visual perception. For example, take the following claim of Noë:

The central claim of what I call the enactive approach is that our ability to perceive not only depends on, but is constituted by, our possession of this sort of sensorimotor knowledge.

Philosophers like Jesse Prinz read this in the strongest possible sense. Responding to Noë, Prinz says:

In vision, for example, we deploy skills of pattern recognition; we know how to discriminate colors without knowing that they are constituted by thus and such physical magnitudes; and we explore the scenes by focusing, scanning, and surveying with our senses. Moreover, no one would deny causal interactions between action and perception: shifting our eyes can change what we see, and what we see can incite us to prepare different motor responses. The thesis becomes controversial when Noë says, “to perceive you must be in possession of sensorimotor bodily skill” (p. 11). Two things stand out about such claims. First, the term “sensorimotor” makes explicit reference to output capacities, and, second, the term “must” implies that the perceiving necessarily involves such capacities. I interpret Noë to mean the following: perception cannot take place (in us? in any naturally evolved creature?) without engaging the mental processes that underwrite our capacity to move. One way to make sense of this necessity thesis is to argue that there is a constitutive relation between perception and action. On this interpretation, the mental processes that enable us to act are literally components of the mental processes that allow us to perceive. I am fairly confident that this is the view that Noë wishes to defend. He often says that, in perceiving, we must use knowledge of sensorimotor contingencies, which implies, that perceptual mechanisms work by registering how stimulation will change when motor effectors are engaged.

In order to defend Noë, I will argue that Prinz misunderstands the scope of Noë’s claim about perception “depending” on certain sensorimotor skills. Prinz takes this  claim to mean that there is no visual experience without activating the mental processes underwriting locomotion itself i.e. broad limb control. Prinz takes “sensorimotor skill” to then mean moving the body around through efferent nerves. But this obviously leads to problems when you consider how people can have visual experience without those motor programs being operative. How then to rescue the strong constitutive claims of enactive theory?

First of all, as far as I can tell, when Noë is careful, he only claims that our ability to perceive depends on certain sensorimotor skills. And as far I see it, this thesis is almost certainly true in terms of visual perception provided “sensorimotor skill” is understood within its proper scope. For example, it is almost certainly true that normal visual perception depends on saccadic motion moving the eyes 3-4 times a second. Saccadic motion depends on tacit sensorimotor skills. Therefore, vision depends on sensorimotor skills.

I don’t think Noë’s argument about dependence needs to be read any stronger than this. And if I am right, then Prinz’s entire argument against Noë is misplaced since his reading of enactive perception is too strong. This also applies to Andy Clark’s recent criticisms of Noë.




Filed under Phenomenology, Psychology

13 responses to “Defending Alva Noë against Jesse Prinz

  1. I’ve noticed Noe’s stuff being criticized a good bit lately, too. Thanks for sharing your thoughts on it. You wouldn’t by chance know if Noe himself has responded anywhere?

    I have his 2009 book but haven’t really gone through it seriously yet, and it looks like it’s written in a more informal style, likely not responding to criticisms.

  2. M.H.

    I think Noe is committed to a more robust notion of action than saccadic eye movements or other tacit sensorimotor skills. For one thing, saccadic eye movements are not attributable to the whole organism; while interesting, such eye movements are “subpersonal”. Strictly speaking, animals don’t *do* saccadic eye movements. I think it doubtful that they can be considered ‘actions’ in any relevant sense.

    I would defend Noe according to a different line. I would criticize Prinz’s interpretation of Noe’s invocation of ‘constituent conditions.’ For it may not be that “the mental processes that enable us to act are literally components of the mental processes that allow us to perceive.” It may rather be that the perceptual environment (or the Umvelt, or whatever pet-tradition you have in mind) is configured according to the demands impressed upon an organism that has to act in meaningful ways.

    I admit this is cryptic and probably ripe for criticism. I’ve only just begun to read into the literature on philosophy of perception. I appreciate your blog, however, and look forward to starting my own graduate studies this Fall–in fact, I’m pursuing some of your same interests.

  3. Gary Williams


    On the contrary, I think Noe is far more committed to an account of “tacit” skills than he is to “personal” skills. In fact, he often talks about unconscious processing being the dominant mode of experience. Indeed, he says “Insofar as we are skillful and expert, we are not deliberate in what we do. Our skill enables us to respond appropriately to the world and in an automatic way”. This is what he calls our “wide mind” or the cognitive unconscious. Noe, like William James, emphasizes that we are primordially creatures of automatic habit. Accordingly, this is why Noe is committed to saying that perceptual abilities depend on tacit sensorimotor skills which cannot be ignored in computational accounts.

    As for your suggestion, I think you may be on the right track but I am not quite sure of what you are claiming. If you read the Gibsonian literature on ecological optics, there is a clear emphasis on how the pregiven structure of the environment makes it unnecessary to “compute” certain information that is already contained within the ambient array of light in the Umvelt. And since we have been evolving in coordination with these invariant Umvelt structures for millions of years, it makes sense that we see such structures as behaviorally relevant and meaningful.

    I appreciate your blog, however, and look forward to starting my own graduate studies this Fall–in fact, I’m pursuing some of your same interests.

    Thanks for reading and commenting! Where will you be attending graduate school? And this might just be my bias speaking, but I think philosophy of perception is a hot ticket right now, particularly if you are coming out of the new enactivist school. Good luck with your academic career, philosophy is very competitive!

  4. M.H.

    But we can be skillful and expert in actions that are not, for all that, deliberate–but which are, for all that, attributable to the whole organism. I don’t think saccadic eye movements are a secure bastion against the kind of criticisms that Noe has been garnering.

    I have a long reading list ahead of me that includes recent work in perceptual psychology. Gibson is certainly on there too.

    As for the competition in philosophy, tell me about it: I applied to 11 PhD programs and didn’t get into a single one. I’m starting my MA this fall at San Francisco State, with the hopes of jumping into a doctoral program just after. Besides CUNY and Toronto, what schools do you find amenable to the enactivist approach to perception?

    I look forward to reading/commenting on more of your posts. Someone posted it to Reddit, if you were wondering where I found it.

  5. Gary Williams

    Unfortunately, there are only a handful of enactivists around the world (although growing rapidly I might add). Here are some of the more empirical leaning PhD programs I was planning on applying to. They might not necessarily be enactivist programs, but I think I could pursue the research I am interested in.

    -Indiana University bloomington
    -Washington U at st louis Psychology Neuroscience Philosophy program
    -U. of Maryland
    -Oregon (Mark Johnson)

    CUNY is perhaps my top pick, as is Indiana.

  6. Ken Aizawa

    Sorry not to have happened onto this post sooner, but here are some quotes from Noë that suggest that it’s more than just abilities to perceive that depend on bodily movement. It is the very act of perception that depends on moving.

    The main idea of this book is that perceiving is a way of acting. Perception is not something that happens to us, or in us. It is something we do. Think of a blind person tap-tapping his or her way around a cluttered space, perceiving that space by touch, not all at once, but through time, by skillful probing and movement. This is, or ought to be, our paradigm of what perceiving is. The world makes itself available to the perceiver through physical movement and interaction.(Noë, 2004, p. 1)

    What perception is, however, is not a process in the brain, but a kind of skillful activity on the part of the animal as a whole. (Noë, 2004, p. 2)

    For mere sensory stimulation to constitute perceptual experienceBthat is, for it to have genuine world-presenting contentBthe perceiver must possess and make use of sensorimotor knowledge. (Noë, 2004, p. 10)

    It is misguided to search for neural correlates of consciousness—at least if these are understood, as they sometimes are, to be neural structures or processes that are alone sufficient for consciousness. There are no such neural structures. How could there be? (Noë, 2009, p. 185)

    Consciousness is not something the brain achieves on its own. Consciousness requires the joint operation of brain, body, and world. Indeed, consciousness is an achievement of the whole animal in its environmental context. I deny, in short, that you are your brain. (Noë, 2009, p. 10)

    Seeing is an activity of exploring the world, one that depends on the world and on the full character of our embodiment. (Noë, 2009, p. 146)

    Regarding the claim that “For example, it is almost certainly true that normal visual perception depends on saccadic motion moving the eyes 3-4 times a second.” This is a common mistake that philosophers make when reading the scientific literature. Scientists often write things that give this misleading impression, so you have to move beyond simple quote mining to get this. Normal visual perception depends on changing retinal stimulation; saccades are one mechanism for providing this. You can get a sense of this by considering the discussion of retinally stabilized images in Palmer, S. (1999). Vision science: Photons to phenomenology. Cambridge, MA: MIT Press.

    • Gary Williams

      Ken, thanks for the comment; those quotes are a very nice summary of Noë’s account of perception. But I still don’t think any of those quotes commit Noë to the strong claim which Prinz attributed to him, namely, that “perception cannot take place (in us? in any naturally evolved creature?) without engaging the mental processes that underwrite our capacity to move.” Prinz thinks Noë is committed to a view of perception that is grounded in “output”. But clearly, we can still perceive when our output capacities don’t work properly. So either Noë is wrong or Prinz misinterpreted the central claim of the enactive approach. I am inclined to say that Noë has simply been misunderstood.

      Indeed, Noë says, “According to actionism, perception requires sensorimotor knowledge, that is, knowledge of the sensory effects of movement. But neither the possession nor the exercise of such knowledge requires movement.” And in the next line he mentions that whereas the possession or exercise of tacit sensorimotor knowledge doesn’t require movement, it is certainly plausible that the development of such systems into behavioral maturity requires movement.

      So when Noë says that “The main idea of this book is that perceiving is a way of acting. Perception is not something that happens to us, or in us. It is something we do”, it is important to realize that not doing anything is itself “something we do”. As long as we are alive, we are always “doing”. Doing is our mode of being. Since conception, our metabolisms are dynamically active and our nervous system is made of bundles of recurrent circuitry which has a life of its own independent of task-driven stimuli. If I recall correctly, only 20% of the input into the LGN is from the optic nerves; the rest is from recurrent, top-down control with its own internal dynamics and attractor spaces.

      To me, this is a very Heideggerian notion. Accordingly, it is also very externalist and pragmatic, because for Noë, the object of perception is not re-presentations within an inner theater, but rather, the real objects of the environment laid out in space which are significant to us. This is because Noë takes a “differentiation” approach to perception instead of an “enrichment” approach. Accordingly, Noë follows Gibson is assuming that the perceptual stimulus of the Umwelt is usually not insufficient for the control of behavior. Moreover, Noë follows Gibson in claiming that sensation is not sufficient for perception e.g. the Ganzfeld effect. So Noë is committed to the idea that brute sensation is not enough; there has to be some kind of representational activity in the brain that codes for the sensorimotor knowledge that turns brute sensation into meaningful perception. However, as the Ganzfeld effect demonstrates, perception cannot be achieved without a differentiated environment that contains sufficient stimulus information. This is why “Consciousness is not something the brain achieves on its own. Consciousness requires the joint operation of brain, body, and world.” Luckily, humans live in a visual environment which affords rich behavioral opportunities and endless detail upon which to search for significant information.As Noë says, “Seeing is an activity of exploring the world, one that depends on the world and on the full character of our embodiment.”

  7. Ken Aizawa

    Gary, my view is that Noe equivocates on whether humans (or all perceptive beings) have to physically move their bodies in order to perceive. There are clearly points where he denies this. Nevertheless, there are points where it seems to me strained at best to suppose that movement is not required. So, to my mind, the challenge is not to find passages where Noe denies that movement is necessary. It is to make sense of certain other passages that seem to claim that movement is indeed necessary. So take, “The world makes itself available to the perceiver through physical movement and interaction.”. How can you read this as not committing Noe to the need for physical movement in order to perceive? This links back to Prinz’s claim “perception cannot take place (in us? in any naturally evolved creature?) without engaging the mental processes that underwrite our capacity to move.” Noe does not seem to me to be sensitive to the distinction between action and mere bodily movement, but I think he typically wants to claim that action is necessary for perception, and indeed constitutive of it. But, if action is necessary for perception, then one needs the brain apparatus for action, right? So, I think that Prinz is still on target.

    Or, take “Consciousness requires the joint operation of brain, body, and world.” People are conscious under complete neuromuscular blockade. How can one say that while being completely immobilized by neuromuscular blockade there is nonetheless join operation of the brain and body? There’s probably a way around this, just because you can almost never pin down a philosopher, but that looks like it will take a lot of fancy philosophical footwork.

    But, let’s do this. We can distinguish strong and weak enactivism, where the strong requires the performance of physical actions with bodily movements of the head and eyes, where the weak only requires some entirely brain-bound knowledge. So, it sounds as though you are willing to reject strong enactivism, but want to hold out for some form of weak enactivism. If everyone agrees to reject strong enactivism, then that’s progress in my book. Then we can take up weak enactivism.

  8. Ken Aizawa

    Incidentally, I have a forthcoming paper that explores at least one version of something like weak enactivism. I got some great feedback from Anthony Morse, (among others), who is very well versed in enactivism and perception, but I appreciate feedback from anyone sympathetic to Noe or ecological psychology. I try to read through these papers, but there seem to me to be so many incredible claims I typically think “They can’t really mean what this seems to mean.”

    The paper is “Consciousness: Don’t Give up on the Brain” It’s available here to anyone:

    • Gary Williams

      Ken, I think your distinction between strong and weak enactivism is a good one. It seems like clinical counter-examples compel us to accept a weak version. In regards to Noe then, the question is whether his oeuvre is most compatible with weak or strong enactivism. I would be inclined to say he endorses a weak version, but this might be because I want to make Noe “one of the good guys” and accordingly, I read him very sympathetically (although I too sometimes think “he can’t really mean what this seems to mean”).

      You mention several quotes which seemingly go against a weak enactivism such as “The world makes itself available to the perceiver through physical movement and interaction.” I think this quotes is ambiguous in respect to strong and weak enactivism. The strong reading is about the constitutive nature of perceptual experience requiring movement. The weak reading would paraphrase the statement as saying something like “Opportunities for behavior are made salient to the perceiver through movement and interaction”. On this reading, it wouldn’t be so much that the chair can only be perceived when we are moving, but rather, the chair would only be significant to us if we had some kind of history of bodily interaction with or exploration of the chair.

      Movement brings forth a world because “world” is simply defined as a world of significance, but not a visual world. “Availability” then must be read in terms of behavioral opportunities, not the generation of perceptual content. The world of my living room is significant (available) to me because I am capable of moving through it and interacting with it. Noe seems to be trying to articulate the Heideggerian notion of availability or “readiness-to-hand”. Accordingly, action is only constitutive for perception insofar as (1) the world is usually taken as significant through our active bodily interaction with it and (2) the normal development and maturation of perceptual systems depends upon some modicum of bodily movement, even if that is just the eyes or head moving.

      I’m reading your paper at the moment. Hopefully I will be able to send you some comments when I am through.

  9. Ken Aizawa

    Hi, Gary,

    In your second paragraph above, you suggest a kind of “developmental” version of enactivism. That perception requires a history of action. Ok. I think that that is very close to being right. There are literally hundreds, if not thousands, of articles on the effects of, for example, lid suture on the development of vision in cats, ferrets, etc. So, if that is what Noe is after (and I’m sure that when he talks about the Hein and Held cat experiments that is what he is after), then he’s not offering something that is going to overturn the orthodox applecart (to use Block’s metaphor).

    Moreover, I think it is quite a challenge to read the whole of paragraph one of Action in Perception as merely a developmental claim. “Perceiving is a way of acting …” That doesn’t sound like a clipped version of the developmental hypothesis. “we enact our perceptual experience; we act it out”. That too does not sound like a developmental hypothesis. Or how about this

    “To perceive like us, it follows, you must have a body like ours. In general it is a mistake to think that we can sharply distinguish visual processing at the highly abstract algorithmic level, on the one hand, from processing at the concrete implementational level, on the other. The point is not that algorithms are constrained by their implementation, although that is true. The point, rather, is that the algorithms are actually, at least in part, formulated in terms of items at the implementational level. You might actually need to mention hands and eyes in the algorithms!

    This sounds like a synchronic claim.

    There is also a paper by Nivedita Gangopadhyay coming out here which also defends Noe is something like the ways to which you are sympathetic:

    • Gary Williams


      I don’t think Noe’s enactive claim is merely a developmental claim or a weak constitutive claim, but rather, both. Noe can then be read as saying (1) normal development of perception requires action and (2) normal perception is usually something we do, rather than something that happens to us. I think when we evaluate the strength of Noe’s account of perception, we shouldn’t rush to find clinical counterexamples to falsify his claims. Instead, we should ask ourselves whether the claim that perception is a way of action captures the “everyday” perception of adults e.g. the kind of perception we use to shop at the grocery store, or ride a bicycle.

      In this way, the fact that sometimes perception can be passive doesn’t immediately falsify the notion that perception is usually active and exploratory in the way Noe describes. If we attend to phenomenology, it is apparent that enactivism is more or less true for the average person under average circumstances. In our everyday lives, important behaviors such as driving to work or getting a beer out of the fridge are precisely “touch-like” and exploratory in the way Noe describes. Accordingly, when Noe claims to be developing a general theory of perception, I simply assume that he is talking about these everyday kinds of situations, not clinical case-examples. So if enactivism fails to capture the mechanics of a clinical case-example, we shouldn’t see this as a falsification of enactivism, but rather, a problem to be met by future research (or current research integrated into Noe’s account). I think we should apply this same criterion to orthodox theories of perception. Just because classic representations don’t seem to play a major role in everyday action-perception cycles doesn’t mean that they aren’t important in other situations. In this respect, I think Andy Clark is right to seek an ecumenical account of cognition that differentiates “representation hungry” problems from those which can be tackled by enactivist theory.

      • Ken Aizawa

        Hi, Gary.

        So, we agree that Noe aims for claims about both the normal development of perceptual capacities and normal perceptual acts.

        The difficulty for Noe’s view, I think, is the way in which he explicates the idea that action and exploration are involved in everyday perceptions. “Involved” is a pretty loose term. So, I would say that action and exploration are involved in everyday perceptions in roughly the sense that actions and perceptions occur in close spatiotemporal, causal relations. Noe, on the other hand, I think, holds that action and exploration are involved in everyday perceptions in the sense that they are constitutive of everyday perceptions. To speak a bit more technically, he seems to think that actions and explorations are part of the minimal supervenience base for perceptual experiences. This is at least part of what Noe means when he says that conscious perception is not something that happens in the brain. It is not a minor part of his view. This idea is what exercise Block and Prinz and me. Such a view seems to predict that if you fall below this minimal supervience base, as by losing the capacity to perform actions and explore, then you will lose your perceptual capacities. One way to try to falsify this is complete neuromuscular blockade. But, you don’t like clinical cases. So, what about hearing an alarm clock that wakes you up in the morning. Is the sound of the clock something you perceive by performing actions and explorations? This looks to me to be an everyday, non-clinical case. Or, what about the perception of a camera flash? The flash takes place in less time that it takes to perform an action or to explore. That also looks to be an everyday, non-clinical case.

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s