Presenter: Bence Nanay, University of Antwerp & University of Cambridge
- Podcasts: Audio & Video
- Get Bence’s paper in .pdf format
- Get the audio in .mp3
- Get the video in .wmv format or in Ogg format
Commentator 1: Kevin Connolly, University of Toronto
- Get the audio in .mp3 format
- get the video in .m4v format or in Ogg format or at Vimeo
Commentator 2: Farid Masrour, Harvard University
- Get Farid’s paper in .pdf format
Consciousness Online 
19 responses to “Perceptual Phenomenology”
Professor Nanay, I want to ask this question: Why should we think that action properties are phenomenological at all? Of course, we normally know what cups, hammers, and pencils are used for, but why think this knowledge has a phenomenology?
To explain the point of the question, I’ll suggest two accounts that do not seem to support a phenomenology for action properties. I’m not affirming the truth of either account, but I don’t see that they couldn’t be true, and my question is: What argument rules them out?
Scenario A. Hearing the action description causes patient MP to have kinesthetic imagery – e.g., imagery of a certain kind of finger grip when he hears “something to write with”, or imagery of a bent arm with hand near mouth when he hears “something to drink out of”. That imagery primes (by some as yet unknown, unconscious process) a binding of features that results in MP’s being able to pick out the corresponding object (e.g., pencil or cup) even when there are distractors present. (Background note: MP is capable of identifying objects when they are presented by themselves.)
In this scenario, there is relevant phenomenology, but what it consists in is kinesthetic imagery, not allegedly phenomenological action properties, nor a phenomenology of affordances, nor a phenomenology of knowing what common objects are used for.
Scenario B. Same as scenario A, except that there is no conscious phenomenology, i.e., no conscious kinesthetic imagery. Instead, hearing the description of the action activates neural patterns that would have led to the kinesthetic imagery of scenario A if they had lasted for 400 ms; but they are inhibited after, say, 200 ms, and the result is that they have some of the same priming effects as scenario A imagines, but do not provide any corresponding conscious phenomenology.
In this scenario, there is not even the phenomenology of kinesthetic imagery.
If either of these scenarios were to be assumed, we’d have no reason to think that action properties were phenomenological, and the question whether they were perceptually or nonperceptually phenomenological would not even arise. But again, I am not affirming that either of these scenarios does obtain. I’m just not clear what is supposed to rule them out. Can you clarify that?
First of all, I’d like to thank Richard Brown for putting together this great conference.
I’ll respond to the comments above in the order they are posted, so Kevin first:
Thanks, Kevin, I find it very helpful to bring in the concept of attention. And I completely agree that it would be great if we could make the contrast case method less reliant on intuitions and more empirically adequate. It is fun to think up contrast cases and that kind of fun is definitely missing from the methodology I am proposing.
But I am still very skeptical, both about the general idea of making the contrast case method more empirically adequate and about your more specific suggestion about how we could do so.
As I understand it, your suggestion involves two steps:
i. We can use the tracking of the micro-movements of the eye in order to assess what properties the subject is attending to.
ii. We can use differences in attention to establish whether two experiences differ in terms of their perceptual phenomenology.
About (i): I find the empirical literature on eye movements very important for a number of philosophical problems, but I’m not sure how it can help here, for the following two reasons:
a. As we know from the literature on ‘covert shifts of attention’, we can shift our visual attention without moving our eyes at all. If so, then the tracking of the eye movement will not track the change in attention.
b. Some philosophers (and some, much fewer, psychologists) make a distinction between attention to a region in space and attention to a property of an object. Even if we put the worries about covert shifts of attention aside, tracking eye movements will not track attention to properties (only, at best, attention to spatial regions).
About (ii): we can take difference in attention to be an evidence for difference in perceptual phenomenology – we don’t need to rely on our intuitions to tell whether two experiences differ with regards to their perceptual phenomenology. This strategy clearly presupposes that difference in attention implies difference in perceptual phenomenology. But some deny that difference in attention always implies difference in phenomenology at all and many philosophers deny that difference in attention always implies difference in perceptual phenomenology (they could say in the case of your goalkeeper example that the expert and non-expert goalkeeper’s perceptual phenomenology is the same, but their non-perceptual phenomenology differs.
But even if we accept that difference in attention always implies difference in perceptual phenomenology, I still have the following problem. The structure of the contrast case methodology is the following: if the only difference between E1 and E2 is that P is represented in E2, but not in E1 and if E1 and E2 differ in their perceptual phenomenology, then P is part of our perceptual phenomenology. But it is not clear that when the perceptual attention is different then the first premise of this contrast case argument is satisfied: it is not the only difference between E1 and E2 that P is represented in E2 but not in E1. Another, very important, difference between E1 and E2 is that our attention is exercised differently. As a result, if there is an attentional difference between E1 and E2, we can’t even use the classic contrast case method (which seems like an independent reason for doubting its feasibility).
Thanks, Farid, these are great comments and I’m especially happy that you went back to the Humphreys-Riddoch 2001 article and that you raised the issue of the standard interpretation of unilateral neglect in the literature (which, btw, I am in full agreement with, as you’ll see).
Four comments (or, rather, clarifications) in response:
Representation vs. phenomenology
It is true that I believe, and have argued elsewhere, that we perceptually represent action-properties. But this paper is not about representation, but about phenomenology (and I assume that these two come apart in the usual ways). This is important because it seems that you interpret some of my claims about unilateral neglect patients to be about their perceptual representations (or, as you say, their ’first-order representation’). I only talk about the phenomenology of these patients (not their representations). I do not talk about perceptual representation at all in this paper.
What’s missing in the case of unilateral neglect patients?
When I talk about unilateral neglect, I definitely do not say that what is missing in the perceptual processing of these patients is ‘the early perceptual representations’. I have been assuming throughout what I take to be the consensus in the unilateral neglect literature that you so nicely summarized (maybe I should have done so in the paper), namely, that unilateral neglect is a defect of some (relatively) higher order processing of the visual stimulus (which, presumably, includes attention). I was assuming that phenomenology would encompass this (relatively) higher order processing of the visual stimulus (including, presumably, attention). In short, I am endorsing a version of the claim you urge me on p. 3 to endorse, namely, that phenomenology encompass more than what you call ‘first-order representation’ (I do find this label misleading as it suggests that any such view would need to be a higher-order theory of consciousness, which I would want to deny). I may even be tempted to go along with a stronger claim that difference in attention brings about a difference in phenomenology (I do know that this is not uncontroversial). But the weaker claim, namely, that ‘first-order representation’ is not the full story about phenomenology, is, as you acknowledge on p. 3., enough to show that the case of the unilateral neglect patients does fit the general methodology I propose for telling perceptual and non-perceptual phenomenology apart.
Your alternative explanation
This assumption, you argue, opens up my account to the following objection: maybe what happens is that the ‘first-order representation’ of shape and color gives rise to the ‘first-order representation’ of action-properties, but while the former is part of the perceptual domain, the latter is not. And while, in the case of healthy humans, the ‘first-order representation’ of shape and color gives rise to the phenomenology of shape and color (which would be perceptual phenomenology), the ‘first-order representation’ of action-properties gives rise to the phenomenology of action-properties, which is non-perceptual. Importantly, it is not the phenomenology of shape and size that gives rise to the phenomenology of action-properties, but the ‘first-order representation’ of shape and color gives rise to the phenomenology of shape and color as well as the ‘first-order representation’ of action-properties, which, in turn, gives rise to the phenomenology of action-properties.
What happens with unilateral neglect patients? Their ‘first-order representation’ of shape and color is intact and gets processed further to give rise to the ‘first-order representation’ of action-properties (and then the phenomenology of action-properties. But because of the lesion, the ‘first-order representation’ of shape and color cannot give rise to the phenomenology of shape and color. I hope I did not misinterpret your alternative suggestion (let me know if I did).
If I understand you correctly, then this is exactly the picture I criticize on pp. 6-7 of my paper (the second horn of the dilemma in Section III.a and then the entire Section III.b. What I take to be the most important argument against this alternative explanation is that it makes MP’s perceptual phenomenology disappear altogether. MP has no perceptual phenomenology of action-properties (by supposition) and he has no perceptual phenomenology of shape and color either. So MP’s experience, which, presumably, has some kind of phenomenology, has no perceptual phenomenology of the object he is looking, performing visually guided actions with and talks about. This sounds to me a pretty wild consequence of the alternative explanation (there’s more on this in Section III.b).
Finally, action-properties vs. affordances:
This is a minor point, but I do want to talk about ‘action-properties’ and not about ‘affordances’, for a variety of reasons. Action-properties are represented by perceptual states, whereas affordances, according to Gibson, are supposed to be ‘picked up’ (whatever that means). According to Gibson, what we see are affordances; according to me, what we see are objects and we (sometimes) see them as having action-properties.
Thanks a lot for this very important question.
Would it be too much against the spirit of my paper to respond that I take it to be intuitively obvious that action-properties show up in our phenomenology?
There is something it is like to experience a ball as catchable. Suppose I am watching the Arsenal Barcelona football (soccer) game and see Robin Van Persie’s shot at the goal from that incredibly tight angle. Suppose also that I am an Arsenal supporter. Depending on whether I experience the ball as catchable for the Barcelona keeper, my overall phenomenology will be very different. It feels different. The big question (and the question I am engaging with) is whether this difference in phenomenology is a difference in perceptual phenomenology.
But, more seriously, given that my entire rhetoric in the paper is based on the claim that there is no point arguing about intuitions and, admittedly, I am using an intuition here, that you are free to question, you would be right to ask me to stop the intuition-mongering…
Luckily, the argument I present in Section III.b would work against your proposals (both your scenario A and your scenario B): if we accepted either of these scenarios, this would make MP’s perceptual phenomenology disappear altogether. As I said in response to one of Farid’s points, MP has no phenomenology (let alone perceptual phenomenology) of action-properties (by supposition) and he has no perceptual phenomenology of shape and color either. So MP’s experience, which, presumably, has some kind of phenomenology, has no perceptual phenomenology of the object he is looking, performing visually guided actions with and talks about. This sounds to me a pretty wild consequence of the alternative explanation (again, there’s more on this in Section III.b).
Oh, I forgot to specify that the previous post is the response to William S. Robinson’s comment above. Sorry (it’s late).
Thanks for the response, Bence. I have a follow-up question. I’m interested in giving a principle for the contrast case method like the one you give for your new method, so before I go on I want to be clear about what sort of principle you are offering. Here is your principle: “If we find patients who experience property P, but do not experience any lower level properties, such as shape, size or color, then we have good reason to conclude that property P is part of our perceptual phenomenology (p. 2).” That’s a sufficient condition, and not a necessary one. [One criticism: since the condition is not a necessary condition, a candidate property that fails to meet it might still be a part of the perceptual phenomenology. So, although you say on p. 2 that you intend your method to settle b at least for some kinds of properties, it looks like your method would settle only the first part of b, but not the second. You stated b as: “How can we tell which properties are part of our perceptual phenomenology and which ones are part of our non-perceptual phenomenology? (p. 1).” But, as far as I can tell, your principle fails to demonstrate which properties are not part of our perceptual phenomenology, let alone which ones of those are part of our non-perceptual phenomenology. And I don’t see anything else in your overall method that would demonstrate those things either.] That point aside, why do you include “good reason to conclude” in the consequent of your principle? Including that phrase makes me wonder whether you intend your principle as an only partially sufficient condition, and not as a fully sufficient condition. Might a candidate property satisfy your condition, but not be part of the perceptual phenomenology due to some overriding reasons? Or do you intend your principle to be a fully sufficient condition? These questions are important for me because I want to be clear about what sort of principle an empirical version of the contrast case method needs in order to match your own principle.
Thanks, Kevin, this is very important (and I should have been explicit about this in the paper). It’s intended to be sufficient only (not necessary), just as you say. So this methodology can tell us whether P is part of our perceptual phenomenology, but not whether P is not.
But as far as I can tell, this kind of sufficient condition is also what the contrast case methodology is shooting for: if you find the two contrasting experiences, such that etc. etc., then P is part of our perceptual phenoemenology. There’s no ‘only if’.
I’m curious to see what your principle for the contrast case method would be (and especially to see what role attention would play there).
Response to Bence Nanay’s post on 2/18 at 16:57
I agree that MP probably has a phenomenology, so if my scenarios implied absence of any such phenomenology, that would be a good reason to reject them.
But I doubt that my scenarios have this consequence.
I don’t find it comfortable to speculate on what the phenomenology of a unilateral neglect patient might be like. But that is the territory we are in, so I have no choice but to give it the following try.
I don’t see why you say MP has no color phenomenology and no shape phenomenology. The last sentence of p. 84 of the Humphreys and Riddoch says that “MP, MB and GK were each able to name all the objects and colors in single-item displays”. I’d suppose they have color and shape phenomenology when they do this. I don’t find it plausible that color and shape phenomenology simply disappears when there are several objects present at once. I’d imagine that they have plenty of color and shape phenomenology, and that the problem is that the various parts of the field don’t get bound into recognizable objects.
The table next to my desk has several stacks of several books each. When I look at the mess, I effortlessly see it as stacks of books. But I can easily imagine a condition in which the ends and the bits of spines and covers that are showing would fail to get organized into facets of books. The result would not be absence of color and shape phenomenology, but rather something like looking at a Mondrian – plenty of color and shape phenomenology, but no recognizable objects.
I can recall three or four occasions on which I awoke in the small hours and could not figure out what I was looking at on a chair a few feet away. Probably a piece of clothing, I thought, but for maybe ten seconds, I couldn’t get myself to see anything *as* this or that kind of object. But there was plenty of phenomenology – several shapes and several shades of grey.
It would be ridiculous to assert that either of these cases definitely is what it must be like for MP. But I do not see what is supposed to lead one to think that these models *could not be* pretty close to what it’s like for MP when there are several objects displayed. If this is how it were for MP in that condition, there would be plenty of phenomenology, but no picking out of the right tool.
Thanks, William (if I may) for the clarification. Let me also clarify:
What I take to be the relevant issue is these patients’ phenomenology of the objects they are looking for in the visual search task. I should have been more explicit about this in the paper. So it’s strictly speaking irrelevant what they experience ‘in single-item displays’. We have no reason to deny that these patients have perceptual phenomenology of shapes and colors in single-item displays. But my claim (it’s really Humphreys-Riddoch’s claim) is that they do not have perceptual phenomenology of shapes and colors when performing the visual search task.
I like your books/Mondrian example, but I think it in fact supports my point. You would be able to point to the red patch even without knowing which book’s spine it is – as you’re not an unilateral neglect patient. But MP cannot (or is slow to) do the same.
Did this clear up the disagreement? Thanks again for engaging with my argument.
Response to Bence Nanay’s post on 2/21 at 7:37
The relevance of my mention of single item displays was only to establish that MP has color phenomenology in those cases. Then my suggestion was that it would be extraordinary to think that introducing multiple items would remove color phenomenology. It’s much more plausible (to me, anyway) to think that there’s still color phenomenology and that what gets disturbed when there are several objects is the ability to bind parts of the field with each other in the normal way, i.e., into a set of coherent objects.
I do not think that Humphreys and Riddoch’s paper supports the view that MP does “not have perceptual phenomenology of shapes and colors when performing the visual search task” [i.e., I’m taking it, the search task in multiple item displays]. That’s because their fig. 2a, p. 86, leftmost panel, shows MP performing at 80% correct when asked to find objects by their colors. That seems compatible with MP’s having color phenomenology but, in 20% of cases, binding a color patch to the object next to it (instead of the correct one), and so pointing to one side or the other of the correct item.
(It would have been interesting to know whether MP’s errors in exp. 1 were usually pointings to an object adjacent to the correct one, or whether the errors were random, or distributed over a wider range than adjacent objects. Unfortunately, no such report is given in the Humphreys & Riddoch paper.)
Thanks again and sorry for being so slow to understand your proposal. So your suggestion is that the patient is aware of a lot of redness in that part of his visual field, but can’t bind the redness to the red objects. This is indeed a possibility, but unilateral neglect patients in general seem not to be aware of anything in the contra-lateral side of their lesion – if they are required to cross lines, what happens is not that they randomly make crosses in the contra-lateral side, but they just ignore the contra-lateral side altogether. So the conjecture is that they are not aware of patches of (unbound) colors in that side either, otherwise they would look again. Also, when they copy a scene, they tend to copy half of each object (with nothing in their contral-lateral side – not patches of color).
Maybe what is behind our disagreement is the difference between phenomenology and representation that I emphasized in response to Farid’s reply. I do think that we have no reason to deny that MP has (some kind of low-level) perceptual representations of the objects in the contra-lateral side of his lesion. But my claim is that he lacks perceptual phenomenology of them.
Does this address your worry?
Hi again, Bence. Regarding your 2/19 at 15:09 comment: Yes, I fully agree that the principle behind the contrast case method is a sufficient condition only (not a necessary one), and that in that respect your principle is no different.
I’m still wondering about one purported difference between the two principles, though. The consequent of the contrast case principle is: “then property P is part of our perceptual phenomenology” (p. 1). But the consequent of your principle is: “then we have good reason to conclude that property P is part of our perceptual phenomenology” (p. 2). Since you add “we have good reason to conclude that,” I wonder whether your principle is weaker than the contrast case principle. As it’s worded, your principle seems to leave open the possibility that overriding reasons could trump our good reason to conclude that property P is part of our perceptual phenomenology. This is not a possibility with the contrast case principle.
Also, I have a very minor point on the rat example on page two. Arguably, as the example is set up, E1 and E2 differ in two represented properties, not just one. E2 represents the property of being a rat, and E1 does not. Meanwhile, E1 represents the property of being a chicken, and E2 does not. If you change the set up slightly, it’s less controversial that the two experiences differ in only one represented property. Just remove any mention of chicken from the example. Have E1 and E2 both represent the property of being meat, while E2 represents the additional property of being a rat.
Thanks, Kevin, very good point about the rat, I’ll change this.
About whether it’s a straight sufficient condition: I was adding this ‘good reason to suppose’ lame-ass careful phrase because, strictly speaking, assuming the negation of my claim does not lead to a flat contradiction, only to a very very very implausible claim (about there being phenomenology of a consciously perceived object without there being perceptial phenomenology of it).
Response to Bence Nanay’s post on 2/22 at 11:36
This last reply of yours helps me see better how you are thinking about MP and why you say what you do. But I have to say that I am still very puzzled. I don’t have any confidence about my imaginings of what it might be like for MP, but I don’t see why there couldn’t be some color and shape phenomenology (not merely unconscious representation). For it seems *possible* that the difficulty could lie in inability to bind features and in inability to attend to left-side phenomenology.
It is very hard to repeat someone’s statement verbatim if it didn’t make any sense to you when you heard it. That *might* be a clue; maybe phenomenological material that is not properly conceptualized is especially hard to encode in memory anyway, and the lesion both prevents binding into coherent objects and increases the encoding difficulty (driving the resulting memory for the affected part of the field to virtually zero).
In drawing, one must shift attention from the scene to the drawing paper. If the left side is not coherently bound and not properly attended, it might not be encoded in short term memory. The subject draws something, and the result seems a good fit for all that’s in memory. It’s a representation of what seems right when he’s looking at the paper, and one consequence of the deficit is to prevent what seems right at that moment from faithfully corresponding to what was in the phenomenology a moment ago (as well as, of course, from corresponding to the objects causing that phenomenology).
Again, I am not affirming this speculation. But until we can rule it out, I remain skeptical of arguments based on a premise that there’s no phenomenology on the left side. I don’t see that what’s in Humphreys and Riddoch rules it out; but perhaps you can make something in their paper more explicit, or point me to something else that does rule it out?
One other question. *If* one becomes convinced that there is no color or shape phenomenology on the left side, why would it be implausible to think that there is no left side phenomenology whatsoever? (Or, am I wrong in thinking that you want to hold there is left side phenomenology of some kind in neglect patients?)
Thanks again. I completely agree that it is dangerous to speculate about what it is like to perceive for unilateral neglect patients – the whole point of this paper is to come up with a methodology for talking about perceptual phenomenology that does not rely on our intuitions and imaginings. In fact, what I take to be the most important big picture message of your comments is that although I am trying to get rid of intuitions and introspection, etc. in the discussion of perceptual phenomenology, the methodology I aim to replace it with does, in some important sense, need to rely on some kind of non-empirical assumptions. Namely, that MP has no phenomenology of colors in the contralateral side. This is a claim about MP’s phenomenology and it is not empirically based. A couple of comments on why I make this assumption nonetheless.
(1) I make this assumption because the empirical literature on unilateral neglect does so – unilateral neglect is even defined as the lack of awareness in the contralateral field. So I take myself to just repeat something that all unilateral neglect researchers agree on (it is a different question whether they are right to do so, see (3) below).
(2) This assumption is about phenomenology in general and not about perceptual phenomenology. So you could recapitulate the structure of my argument as taking a claim about phenomenology per se for granted and get a claim about perceptual phenomenology out of it. In other words, I do not use any non-empirical considerations when talking about perceptual phenomenology
(3) I think you are right that the general framework according to which unilateral neglect patients are not aware of the contralateral side of their visual field can be questioned, but it needs to be noted that this would be revisionary of the way these researchers think about unilateral neglect and, more importantly, it seems to work if we make some extra assumptions. More precisely, it seems to me that the disagreement between your picture and mine is that you take consciousness without attention to be possible (is this the case???). So, you can say that unilateral neglect patients fail to attend to the contralateral side, but this does not mean that they are not conscious of it (in the sense of having phenomenal character). I do deny, for independent reasons (see Jesse’s talk above, e.g.) that consciousness without attention is not possible. And so does the unilateral neglect literature in general – that is why they can explain the lack of awareness in terms of the lack of attention, which is what they take themselves to be doing (Kinsbourne is especially explicit about this).
About your last question: the argument is that although MP has no phenomenology of color, he does have phenomenology of action-properties (in the left hand side, while performing visual search).
Response to Bence Nanay’s post on 2/24 at 9:06
This’ll be brief. Yes, you’re right, I’m not collapsing consciousness and attention. But thank you for the pointer to Kinsbourne – I’ll go look at that.
On your last paragraph, I think we’ve about come full circle, to where I started in my first response. I can well see how someone might think neglect patients have no color phenomenology on the left side (even though my theme has been that I don’t see how we can rule out that they do). But I just don’t see the attraction at all of saying that they have a phenomenology of action properties on the left side (or, actually, of saying that there is a phenomenology of action properties anywhere). I know you’re assuming that, but I’m afraid I still can’t see why.
Hi Bence,
Cool paper. Sorry I’m late to the party. It’s reading week in these parts, which actually means that it’s skiing week. (Also, I apologize ahead of time if some of my comments/questions overlap with previous comments/questions. But, argh, if I’d taken it as a criterion to read through everything, it might’ve blocked me from ever getting to posting these…):
1- You’re right, of course, that if we’re going to debate whether all consciousness is perceptual, we need to be clear on what that debate is about. So we need to have some useful way of cutting the cake between the set of perceptual experiences and the set of non-perceptual experiences.
But it’s not like the literature is empty of options on this front – nor empty of options that don’t rely on contrast cases to spell out how to cut the cake. Two examples are Prinz 2007 and Carruthers and Veillet in Bayne and Montague in press. Do you have an argument that your proposal is preferable (or has some advantages) to the others out there, or are you mainly just trying to spell out another option in logical space?
2- I’m a bit lost in this talk of ‘experiences of what an object can be used for’. That concept is sufficiently unclear (to me at least) that I would’ve thought that these are odd experiences to appeal to as a means to illustrating your proposal. That is, it’s hard for me to see whether these are the best examples to appeal to if it’s broadly unclear what we’re talking about when we’re talking about them. (Though maybe this unclarity is local to me…)
Similarly, I find myself rather up in the air about whether to creed much of the data in the experimental evidence that you’re drawing on. Suppose I spend lots of time thinking about conscious experience but can’t put much traction on the concept of experiences of what an object can be used for. And suppose that I’m told (as you say in your paper) that the experimental subjects report that they have these experiences. I’m inclined to wonder: are the subjects all talking about the same phenomena? are they even terribly clear about what exactly they’re claiming to have, when they claim to have these experiences?
Thoughts like these at least caused me to spend my time thinking about your proposal independently of your illustrating example. Any chance you can provide some clarifications to quell my worries and general dubiousness?
3- You give us this conditional: “if we find patients who are capable of experiencing property P without being capable of experiencing the low level properties of shape, size and color, then we have good reasons to conclude that property P is part of our perceptual phenomenology”.
Thought experiment: Suppose we dope Jane up so that she’s blind and can’t feel anything she touches (or suppose she’s this way naturally). But suppose she has an otherwise ordinary life. It seems that she’s capable of experiencing all sorts of properties (she could still, presumably, lead a rich experiential life) but incapable of experiencing the low level properties of shape, size, and colour. It seems to follow (from your conditional) that any property P that she experiences is part of her perceptual phenomenology. (Right?)
But suppose that follows. Various question quickly follow in turn: Doesn’t this make your proposal look like it’s going to very quickly count basically any controversial case of a non-perceptual experience as being a perceptual experience (that is, just dope someone in the relevant ways, and all her experiences turn out to be perceptual)? Similarly, doesn’t that tell us that your proposal (even it it’s independently motivated) is locating the perceptual / non-perceptual divide in a very different place than (anywhere where) participants in the ‘Is all consciousness perceptual?’ debate have been putting it? And if so, doesn’t that temper the thought that your proposal is going to be of benefit to that debate?
Peace Jordan
Thanks, Jordan, great questions.
General remark: the ‘is all consciousness perceptual?’ debate that you allude to a number of times is a side-concern for my paper. I mention it in the conclusion as an example for a big debate that we can’t really address without answering my question about perceptual and non-perceptual phenomenology. But answering my question does not give us an automatic answer in the ‘is all consciousness perceptual?’ debate. My aim is to figure out which properties of the perceived object are part of the perceptual phenomenology. Even if P is, this does not necessarily imply that the thought that a is P cannot have what some refer to as ‘cognitive phenomenology’. So my question is really quite different from theirs.
Responses to your specific concerns:
1. I have independent worries about both the Prinz and the Carruthers ways of setting up the perceptual/non-perceptual distinction. And it needs to be noted that the distinction they make is supposed to serve a different purpose from mine. They are interested in when a STATE has perceptual phenomenology and I’m interested in when a PROPERTY is part of perceptual phenomenology. But you’re right, I should address these in the next version of my paper.
2. Action-properties: I take these to be very simple properties: edible, climbable. Arguably, even animals represent objects as having action-properties. MP told the experimenters that he can sometimes find objects in the left hand side of his visual field when he is looking for something that can be used for Q-ing. And this prompted the experimenters to do the study I’m using here. So MP did seem to know what he was talking about.
3. I talk about vision only through the paper and the ‘lower-order’ properties I talk about are also supposed to be understood as the ‘lower-order’ properties of vision. I probably should be more explicit about this. In the case of audition, they would, of course, be different.
Does this help? I do think that hard core cognitive phenomenology believers (like you) can go along with my methodology and still find a logical space for cognitive phenomenology…
Response to William S. Robinson’s last post: I hope it was not a full circle, in the sense that the correspondence failed to lead anywhere. I did learn a lot, especially about how I should have explicated the assumptions that I inhereted from the unilateral neglect literature (especially concerning phenomenology in the contralateral side).
I hope that the correspondence above was not completely pointless to you either. If you grant that MP has no phenomenology of shape and color in the contralateral side, then it follows from the argument in Section III. b. that we have very strong reason to believe that action-properties are part of his perceptual phenomenology. So I hope that the back and forth about shape and color did have some bearing on the issue about action-properties in the end…