Monday, April 02, 2007

Against "Appearances"

Chisholm (1957) and Jackson (1977) distinguish an epistemic from a phenomenal sense of the word "appears".

The epistemic sense: If I say "It appears the Democrats are headed for defeat", normally I'm expressing a kind of hedged judgment. I'm expressing the view that the Democrats are headed for defeat, qualified with a recognition that my judgment may be wrong. I'm not saying anything in particular about my "phenomenology" or stream of experience. I'm not claiming for example, to be entertaining a mental image of defeated Democrats or to hear the word "defeat" ringing in my mind in inner speech.

The phenomenal sense: If I say, looking at a well-known visual illusion, "The top line appears longer", I'm not expressing a judgment about the line. I know the lines are the same length. Instead, I'm making a claim about my phenomenology, my visual experience.

These senses sometimes come together in perception: If I say, looking at two peaks in the distance, "The one on the left appears higher", I might be saying something about the peaks, or I might be saying something about my experience, or (more interestingly) I might be saying something about the peaks by way of saying something about my experience. I might be saying: "My visual experience is a left-looking-higher kind of visual experience; based on that I tentatively conclude that the left peak is higher."

Now how often do we actually do that last type of thing? If not in conversation with others, in our own cognition? How often, that is, do we reach judgments about outward objects on the basis of our knowledge of our own experience?

Traditional "veil of perception" views of perception (e.g., Descartes and Locke, on standard interpretations) suggest that that is what we do all the time in perception. We know outward things only by knowing our own minds first (and better), in particular by knowing our inner experiences of those things. What we know directly is our own sensory experience; our knowledge of the outside world is derivative of that. (Thus, it is as though our sensory experience stands like a veil between us and the world, preventing direct contact with things as they are in themselves.)

I worry that the word "appearance", as philosophers of perception typically use it, invites something like this view, by blurring together the phenomenal and the epistemic senses of "appears". I worry that it invites the view that our judgments about the things we see -- the real, physical objects around us -- are grounded in facts about how those objects are experienced phenomenally. I worry it invites the view that when I say "It (visually) appears that there is a coffee cup on the table" I mean both that I (visually) am inclined to judge that there is a coffee mug on the table and that I am having a visual experience of a certain sort -- a coffee-muggish experience; and that these two events are integrated in a certain way, as different aspects of the "appearance" perhaps. It's because I have the visual experience that I reach the visual judgment.

But here's the question: Do we reach judgments about the properties of objects based on the sensory experiences they produce in us? Or is the visual experience we have of an object the product of, or something created in parallel with, our judgments about the object? If I'm right that talk of "visual appearances" tends to invite the former view, that's reason to be wary of it, if the latter view has merit.

14 comments:

michael metzler said...

This is a great post that I think is directly salient to something I’ve been thinking of for a little while now. The hypothesis I’m working on is that our judgments about the object are based on something created ‘parallel’ (to be less than precise) with our visual experience and also on that visual experience itself; we would therefore always have phenomenal and non-phenomenal access to information about that object. If what the object ‘is like’ (Pitt 2004) is a simplified product of our more sophisticated unconscious intentionality directed at this object, I’m not sure what else could be the case. A big ‘if’, but one I think many now accept.

In the case of a juiced up blindsight scenario (see Rust’s recent post and comments), our judgments would be strictly based on non-phenomenal access. But in the case of more poetic judgments, we might be inclined to think our access is only phenomenal (e.g. literary metaphor). But in the typical scenario, it seems we should assume that our abilities at forming judgments about a particular object are based on the way our unconscious mind has already discriminated the object from the environment and also on the resulting ‘what it is like’ to see the object.

That’s the rough idea anyway. Is there something wrong with the above conditional or the claimed truth of the antecedent? Thanks!

michael metzler said...

. . . and perhaps another way of answering the question, although less completely, is to say that our visual experiences are the products of our unconscious mind’s more sophisticated ‘judgments’.

Tanasije Gjorgoski said...

Hi Eric,

I'm reminded of the post at Brain Hammer: Bursting on The Seems., where Pete also argued against this distinction.
My main concern, as I said there, is that I take it that phenomenal appearance can be used as base for concepts. For example a specific color of some object (even before we have concept green) appears somehow to us, and it is that appearance on base of which we "create" concept - green.
Epistemic appearance, on other side, seems to require conceptual judgments.

On other side, I'm suspicious too of this distinction between epistemic and phenomenal seemings. But instead of reducing them to one or the other, I think there is some common (more abstract) meaning in both those senses.

Tanasije Gjorgoski said...

Oh, BTW, I tend to think that in the "The top line appears longer" example, "appears" is used as a kind of hedged judgment (maybe you remember, the issue came up previously when discussing if the reports of how things seem to us be false).

Matt Brown said...

This reminds me of John Dewey, who was clearer than most in separating experiential and epistemic concepts. He thought (and I think you might agree) that much of the confusion in philosophy came from conflating the two.

Pete Mandik said...

Hi Eric,

I'm dubious that there are such things as phenomenal appearances (irreducible to epistemic appearances) and the contrary assumption goes along with assuming that there's such a thing as phenomenologiy, such things as qualia, etc.

In addition to the post of mine that Tanasije pointed out, there's another one with arguments I like better: Bursting at the Seems 2: Electric Boogaloo

Eric Schwitzgebel said...

Thanks to all of you for the kind and generous comments on a post that I'm now thinking was somewhat unclear about the key point: why, exactly, the word "appearance" invites the view I suggest it does.

Michael: I think I am attracted to something like the view you suggest, though I'm not feeling entirely clear and decided about it. The Carnapian in me wants to think about what the empirical differences might be between the two views -- and until I understand that, I'm not sure I understand exactly what the issues are. But I do feel clear that hanging any view about such matters on assumptions implicit in the use of the word "appears" is not a good way to do philosophy!

Tanasije: I wish I knew what it was to have a concept! I guess that's another word I'm a little suspicious of -- or at least, I'm suspicious of philosophical arguments that turn essentially on a specific understanding of the "concept" concept! Those qualms aside, though, I don't see why we couldn't sometimes have concepts that were somehow built out of prior knowledge about our experiences. I don't think my comments in this post depend on denything that. But I do think that it's a mistake to suppose that normally our judgments are grounded in a prior apprehension of our experiences.

Thanks for the tip on Dewey, Matt. I'll have to go look that up! Do you have a reference?

Thanks for the pointers to Pete's interesting posts, Tanasije and Pete. I do think we're up to slightly different things, though. I don't like the word "appearances" for how it seems to conflate the phenomenal and epistemic and invite the idea that the epistemic depends on the phenomenal. You, Pete, as I understand it, deny that there is a kind of thing, a "phenomenal appearance" that is distinct from epistemic appearances. Yes?

Gabriel said...

Nice post! I find appearance really interesting, but difficult to make sense of.... a couple of points:

I take it that when talking about visual appearance, we talk about looks. But looks locutions differ from appearance ones - you can say that it appears that p, but not that it looks that p. The looks locution that takes a sentential complement clause is 'it looks as if p'; it's not so obvious that that should be read as expressing a propositional attitude. For this reason, one might not understand epistemic looks as expressing a judgment that has a content expressible by the sentential complement clause that completes the locution. It also seems fine for me, when looking at my neighbour's uncollected post, to say that it looks as if they are away, even though I know they are just lazy.

Second - you suggest that conflating epistemic and phenomenal looks invites the view that judgments about objects are based on sensory experience. I can see that conflating the two might invite the view that judgments about objects are based on how they phenomenally look. That may be a mistake, but it's not trivially equivalent to what you say unless one thought that phenomenal looks were sensory experiences, and there's no need to think that. In your well known visual illusion, one could say that the judgment of unequal length is made on the basis of the distinctive look of the lines, without taking that to be something about one's sensory experience. It could just be some other property of the lines that you are aware of.

Eric Schwitzgebel said...

Thanks for the thoughtful comment, Gabriel! You make some interesting points about potential differences between how something looks and its "visual appearance". I like the term "looks" better than "appearance" because "looks" does not as easily invite the idea that there is a thing or event -- an "appearance" -- that stands between the object and our judgment about it, as it were. At least to my ear, it doesn't invite that as readily.

That said, though, I'm not sure I'm getting your second point. I'd have assumed that "phenomenal looks" are just visual sensory experiences. Maybe you could expand a bit on the distinction you have in mind here?

gabriel said...

Austin has a discussion of different looks/seems/appear locutions somewhere in Sense and Sensibilia - I think he makes the point there that there isn't a looks-that locution.

I was thinking that, e.g., in the Muller-Lyer case, you could say that the lines have a particular look - that's just something the lines have, that they get to have by having the distinctive wedges drawn on them and so on. People go wrong when they take that look to support the claim that the lines are of different length. I think Travis says this in 'The Silence of the Senses'.

That still leaves open what to say about what it is to have a particular look. It does seem that there is a sense in which a penny viewed slightly from the side looks elliptical. Some people have said that this is because it, e.g., subtends an elliptical region of the visual field, a property it would have in common with an elliptical penny placed at right angles to the line of sight of the subject. Inspired by this, one could say that having that (elliptical) look is just a matter of having this property. Perhaps something similar is going on in the Muller-Lyer case...? They have some spatial property in common with, or similar to, that which lines of different length do; that's what is is for them to look as they do, and this leads to the false judgment of difference in length.

Justin Kalef said...

Hi Eric,

Justin here -- I'm the Canadian hayseed in the straw hat who keeps harassing your experiment table out of sheer delight that someone's actually doing the stuff you and your team are doing. So cool!

Anyway, I thought I'd mention that C. W. K. Mundle's superb but almost entirely unknown book _Perception: Facts and Theories_ proposes a notation for using words like 'seems', 'perceives', etc. that disambiguates these terms in what turns out to be a very important way (along the lines you suggest).

Also, Mundle presents some important -- and surprising -- experimental findings regarding the 'which coin is larger' issue and the issue of disagreement in colour perception. I notice that in the bibliography of your essay on whether coins 'appear' flat, you don't mention Mundle. But I think you'd find it very interesting, thought-provoking and pertinent reading. over 20 years ago, he was throwing these experimental findings in the faces of those philosophers who were assured a priori of their folk psychological assessments, and getting systematically ignored (well, more or less) for his efforts.

Keep it up, man, this stuff is great! So glad I met you and learned about this blog. I'm going to pass the word along.

-Justin

Eric Schwitzgebel said...

Thanks for the interesting posts and references, Gabriel and Justin. Sorry it has taken a while for me to get back to them!

Gabriel, there's something that seems wrong to me in the idea that things present "looks" on the basis on which we reach judgments about how things really are. Partly, I'm nervous because I doubt our capacity to know our phenomenology or experience very well. I think we know the outside world better, and first. In fact, visual illusions and the experience of ellipticality when looking at a coin are exactly some of things I think we can go pretty wrong about in our experience!

On the other hand, I also suspect there's something right in the view. We have, I'm inclined to say, some sort of "attunement" to our experience that plays a certain kind of role in reaching our judgments. I hope to work out some of the details of this view in future blog posts.

Justin: Thanks for the Mundle reference. Someone else, I think, pointed that out to me a while back, but I haven't looked at it yet. It sounds like I should!

Thanks for the kind words. I enjoyed meeting you at the APA.

Pete Mandik said...

Hi Eric,

Yikes! Sorry for such a long delay. The answer, however, is "yes".

max said...

Justin, I am so glad to see Clem Mundle quoted - he was my uncle and also wrote 'A critique of linguistic philosophy' which also flew in the face of much contemporary opinion.