Back    All discussions

2010-06-23
Dretske on seeing
Can somebody give me a quick reference?  Fred Dretske holds that to see something is to distinguish it from other things.  True/false?  Where?

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
Inclined to think that was in that SPAWN paper. zarright?
the view of course has no plausibility whatsoever! I see the gorilla but make no judgement about it, hence have no knowledge concerning its distinctness from anything else ...

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
I'm not sure if that's his exact view, but mustn't it be in:

Fred Dretske (1969). Seeing And Knowing. Chicago: University Of Chicago Press.

?

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
True, in Seeing&Knowing. More precisely: what is seen must be "differentiated" from "immediate surrounding" (not a matter of judgement for Dretske, obviously, but of how things look): pp. 20-28, with some discussion of  "limiting case". Also interesting discussion in Cassam's Possibility of Knowledge, ch.3.

2010-06-25
Dretske on seeing
Distinguish <> know to be distinct? Anti-Williamsonish, obviously, fwiw.
Inclined to think that 'how things look' means something along the lines of 'how X might (take themself to) know things are, going by looking', hence a matter of judgement after all -- though not obviously obviously so.

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
You might be interested in my discussion of Dretske--and others--on "non-epistemic" seeing in

Close, Daryl.  1976.  "What is Non-Epistemic Seeing?" Mind 85 (April): 161-170.

Close, Daryl.  1980.  "More On Non-Epistemic Seeing," Mind 89 (January):  99-105.



2010-06-25
Dretske on seeing
Reply to Benj Hellie
In one sense certainly (though in another "Sergent Garcia distinguished Zorro and Diego de la Vega" seems ok?). But for Dretske (as condition on seeing x--differentiation seems meant to be a different notion, no)? 
(+ what sort of "might"? I might take myself to know there's a gorilla if I notice/attend to it?)

2010-06-25
Dretske on seeing
Reply to Mohan Matthen

<?xml:namespace prefix = o ns = "urn:schemas-microsoft-com:office:office" />

Maybe this will help:

"Let us return, then, to the positive characterization of seeingn. If this state of affairs has no belief content, what content does it have ? [...] S seesn D = D is visually differentiated from its immediate environment by S.

The phrase "visually differentiated" is meant to suggest several things, the most important of which is that S's differentiation of D is constituted by D's looking some way to S, and moreover, looking different than its immediate environment."

 

F. Dretske (1969), Seeing and Knowing, Londres : Routledge &Kegan Paul. pp. 19-20


2010-06-25
Dretske on seeing
Thanks for the reference, Vivian, Philippe -- I'll look it up as soon as all these world leaders vacate Toronto and allow the University to re-open the Library.

I took it that for Dretske, seeing something meant being able to differentiate it from its background (whatever "being able to" amounts to).  You don't see a tiger in the jungle when its stripes blend into the background, even if you lay your eyes on it, and thus see it in Grice's sense.  As for the gorilla, I guess it depends on your theory of inattentional blindness.  Maybe you are not metacognitively aware of seeing it, though you see it in the sense of differentiating it from the background.  Or (on another view) maybe you don't see it at all, since you can't report it.

I am not up on my Zorro, but if Zorro = Diego, and if Garcia visually differentiated Z from his visual background, then he did the same for D, whether he knew it or not.  So if he saw Z in (what I take to be) Dretske's sense, he also saw D.  And that seems right, doesn't it?

2010-06-25
Dretske on seeing
Reply to Mohan Matthen

So blind people can see trees (by feeling them)? Dretske isn't stupid.

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
Suppose a uniformly colored expanse is before me; eyes open, lights on. I am preoccupied with thoughts of math, hence am not 'attending to' the expanse or any part of it. No differentiation in judgement of any part of it from any other part of it or of it from anything else. No part of the expanse looks different from any other part of it; it doesn't look different from anything else (because whether I see the expanse or not, I certainly see nothing else). So I don't 'visually differentiate' anything from anything. So I don't see anything. Evidently, I have gone blind. Scary!

2010-06-25
Dretske on seeing
Philippe: on the first point, the question is what this notion of differentiation without knowledge could be. People suggest it has something to o with looking a certain way; I maintain that the notion of looking has the notion of knowledge built in.
Mostly correct on the 'might'. The thought is that we have a hypothetical situation in mind in 'look' attributions. When we say 'O looks F to S', what we mean is this. Taking S's perceptual situation as built into the 'center' of the world: taking a certain (contextually specified) body of information for granted: going by looking: O is F. 

The notion of 'in a situation: going by looking: O is F' amounts to something like 'if we were in that situation, and we looked, we would judge that O is F'. But not exactly. First of all, the claim is not /about/ a judgement that O is F, it is rather a sort of relativized judgement that O is F. And second, the hypothesization of the situation does not work by building the situation into the antecedent of a conditional, but rather by hypothetically shifting in the indices against which various contextually specified parameters are evaluated to those in the situation.

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
See Chapter II of Dretske's Seeing and Knowing. He there makes the claim that someone sees something non-epistemically only if he differentiates the object from its immediate environment. He also elaborates on the idea of visual differentiation by suggesting that this has to do with a particular way in which the object must look to the perceiver in order for it to be true that it is seen non-epistemically.

2010-06-25
Dretske on seeing
Reply to Benj Hellie
A couple of things:
1) Mohan, for Dretske's view on inattentional blindness, his paper in Phil. Perspectives 2007 is useful. As for the Zorro case, I meant "distinguish" in a non-perceptual sense (perhaps wrongly assuming that Benj's notion wasn't particularly perceptual). For a similar perceptual case where, S visually differentiates x from y without knowing they're distinct, one might imagine that S has a hallucination of the pink background surrounding the kangaroo that's really in front of her (or alternatively, a giant bright yellow truck and a giant blue truck in the periphery of S's visual field, when S is fully focusing on and thinking about the kangaroo she is foveating).

2) Hugh, indeed. I thought differentiation was meant to be a necesary condition on seeing (though the formulation Dretske uses "=" is a bit unfortunate), and the first clarification he makes about the notion is that it'd be visual (check out p.20).

3) Benj. That's the limiting case I mentioned earlier. He discusses this sort of (no background) case on pp.24-6: upshot is, S sees(ne) x only if, unless everything in front of S is part of x&conditions are such that x has no environment, S differentiates x from its immediate surrounding. (Not saying this works, only that such cases are taken into account.)

4) Benj, on distinguishing and knowing distinctness. Why do you need the right-to-left entailment? Doesn't seem to play a role in your objection to the differentiation constraint + possible sense of distinguishing where: Matt knows the twins Aaron and Bryce are distinct but he simply cannot distinguish them when he sees them.

5) Benj, about looks: not sure there's any disagreement, but not sure I get all the bits in your proposal (relativized judgement? Is looking at o when o looks F sufficient for judging that o is F? Is looking at o when o looks F sufficient for selectively attending o being F and for noticing that o is F?).

I thought the  following was rather trivial. If C = whatever conditions are sufficient for S to be caused to believe that o is F (or looks F) on basis of S's experience of o when o looks F to S , then we can analyse looks in doxastic/gnosic terms: x looks F to S --> if C, then S believes o is (or looks) F. But, I assume, that's not what you take yourself to be doing, or is it?

+ on formal framework: I'm sure there's different ways to do this (long conditional, centered worlds, other contextualist approaches). One thing I don't get without hearing more is: is there any non-independent (e.g. absolute love of centered worlds in general) reason for preferring one way to do things over another? That is, any reason against conditional approach in this particular case?

Finally, not sure if this account of looks helps in the objection that the differentiation constraint is implausible (not sure whether it was meant to): seems compatible with S not actually believing anything about the gorilla and its background that S might.

2010-06-25
Dretske on seeing
Reply to Benj Hellie
Hi Benj: Based on what he argues in a public lecture, "What We See", I think Dretske would say that, with respect to your example, you do see each of the parts even though you are not attending to any of them.  His main argument is that, with respect to each part of the colored expanse, you received (by sight) from each part information about its color and its shape; and, he thinks, that as a consequence, you know of each part that it has such-and-such color and such-and-such shape.  You also know, says Dretske, of each part that it has the same color and shape as all the other parts - if any one part had a different color or shape from the rest, you would have noticed that it did.  According to Dretske, this is knowledge that you wouldn't have had if you hadn't seen each part, even though you weren't attending to any one part at the time.  Similarly, in the case of inattentional blindness, e.g., "the guy in the gorilla suit" experiment, Dretske holds that you do see the object(s) you are not attending to because you acquired, by sight, some de re knowledge of some of its sensible properties.  Dretske's lecture can be seen on YouTube.  Take care. --Ray

2010-06-25
Dretske on seeing
Reply to Mohan Matthen
Hi Mohan: I'm pretty sure Dretske would agree with you that if Z=D, then if you visually differentiated Z from his background then you will have thereby visually differentiated D from said background too (and conversely).  That would be an instance of his general view about object-perception: if Object X=Object Y, then you see X iff you see Y.  Take care.  --Ray

2010-06-25
Dretske on seeing
Reply to Benj Hellie
Benj: With regard to:

"Suppose a uniformly colored expanse is before me; eyes open, lights on. I am preoccupied with thoughts of math, hence am not 'attending to' the expanse or any part of it. No differentiation in judgement of any part of it from any other part of it or of it from anything ... (expand) else. No part of the expanse looks different from any other part of it; it doesn't look different from anything else (because whether I see the expanse or not, I certainly see nothing else). So I don't 'visually differentiate' anything from anything. So I don't see anything. Evidently, I have gone blind. Scary!"

I am not sure why you can't see without seeing anything.  That's what happens in a Ganzfeld, I think.   (Also, though I think you can easily reformulate the example:  I was not talking about attending or about judgement, but about seeing.)

2010-06-26
Dretske on seeing
Reply to Mohan Matthen
Yes, Dretske's does hold that seeing something requires one to "visually differentiate it from the immediate environment". Check out his Seeing and Knowing, chap. II, pp. 20ff.

2010-06-26
Dretske on seeing
Reply to Mohan Matthen
The Gestalt school of psychology claims that to see something we have to distinguish it from its background, i.e. other things.

2010-06-26
Dretske on seeing
Reply to Mohan Matthen
It's important to note that Dretske's definition of "seeing" discussed in this thread is a definition of "non-epistemic seeing," viz., seeing that is "devoid of positive belief content" (Seeing and Knowing, p. 6).  More specifically, Dretske defines non-epistemic seeing as "a way of seeing such that for any proposition, P, the statement 'S sees D' does not logically entail the statement 'S believes P' " (p. 6).

I analyze this definition in the 1976 paper cited earlier in this thread, arguing in part that Dretske does not present a clear case of someone seeing something non-epistemically on any interpretation of his definition.  In the 1980 paper, I argue that subsequent attempts to present cases of non-epistemic seeing fail.  I have yet to encounter a successful case, so if anyone can provide one here, that would be a significant advance in the philosophy of perception.

N. B.  Assume that "S non-epistemically sees D" does not merely mean something like "Light is reflected off of the surface of D into S's working eyes, with working optic nerve(s), working visual cortex, etc."  That's presumably why Dretske proposes "visual differentiation" as an alternative definition.

2010-06-26
Dretske on seeing
Reply to Mohan Matthen


Dretske reflects on 'Seeing and Knowing' in a later paper 'Simple Seeing' reprinted in his Cambridge 2000 collection. In that paper he talks about 'depriving' seeing of ' ... its logical or conceptual links with the observational data base' and about that deprivation as an epistemologically consequential 'goad'. I would think that he would have been responding to something like the idea of the consequentiality of something like 'perceptual knowledge' as a type of knowledge with specifically significant inhering consequentiality and as reducing the impact of psychological association grounded in fact. What I don't see, but seems to be emphasized in this thread discussion, is why Dretske's notion of non-epistemic seeing is to be placed in the context of phenomenological differentiation; why it wouldn't just put a stop to discussion of the background phenomenology (other than one might have a different sort of phenomenology of specificity associated with some specific judgements associated with contents).

2010-06-26
Dretske on seeing
Matthew: you say "What I don't see . . . is why Dretske's notion of non-epistemic seeing is to be placed in the context of phenomenological differentiation"

How completely does Dretske mean to sever the logical and conceptual ties between non-epistemic seeing and believing?  I understand that he doesn't think seeing can be defined in terms of a disposition to believe (as Daryl emphasizes) but it may nevertheless constitute a potential warrant for belief.  I haven't yet had time to read your papers, Daryl, but if one wanted to deny that seeing can be defined in terms of belief but allow the possibility of its acting as warrant for belief, that would require phenomenality, at least if warrant must be conscious.

If that is right, then consider de re belief about x.  In order to serve as warrant for a such a belief, mustn't a visual state differentiate or single out x?

2010-06-26
Dretske on seeing
Hi Philippe:
3) Inclined to think that this case shows that the ordinary notion of seeing O in the sense relevant to perception doesn't require O causing distinguishable visual experience from its surround -- not so impressed that Dretske added a disjunct. Inclined to think there is a way we use 'see' on which this requirement may be imposed: as when we mean 'spot'. But this is not a notion that concerns perception in a passive sense but rather observation in an active sense.

4) Issues are confounding your characterization of the case: mode of presentation, or sensitivity to epistemic method, or genericity of 'can distinguish'. Root analysis is that for Bill to discriminate x and y is for him to recognize that m <> n, where 'm' and 'n' have senses which are modes of presentation of x and y respectively. 

5) A relativized judgement is something like the following -- 1940: the allies are on the ropes; Britain is at siege; Hitler is pushing deep into the Soviet Union. Or -- let's look at it from their point of view: Hellie is the one who must be stopped. Obviously I don't judge from my own point of view that Britain is at siege or that Hellie is the one who must be stopped. The thought is that saying '1940' or 'let's take up their point of view' shift around various indices, so what counts as true gets shifted, and I can judge things with contents that are sort of projections onto the shifted index by ostensibly judging things with normal contents.

No, and no. If Bill judges that O is F as a response to an act of looking, then O looks F to Bill (then, and given the information in the background of his judgement). 


You are right that that is trivial. Also not what I am up to. The analysis is of /copular/ verb 'look' in terms of /agentive/ verb look. The claim, again, is that O looks F to S means: assuming a certain salient body of information: taking up S's perceptual situation: going by looking: O is F. I say nothing about what S believes, or about any conditions relevant to causing S to believe that o is F. Nor do I say anything about S's 'experience of O'. 

The objection is that the differentiation constraint is unmotivated once we have seen the relation between perceptual conditions and what it is to look F. After all, the following seems to be entirely coherent: one has a certain body of information of an unspecified sort; part of one's perceptual condition is that one sees X, Y, and Z; going by looking, one judges that X is blue, that Y is blue, and that Z is blue. Where is the contradiction?

2010-06-26
Dretske on seeing
Ah yeah, also: why do it pragmatically? Somewhat inclined to think that semantics is about the stable power of linguistic entities to influence the common ground; pragmatics is about other influences on common ground. My feeling is that the conditional approach seems to assume that too many words are uttered.

2010-06-26
Dretske on seeing
Hi Ray, not seeing how Dretske can coherently maintain that claim in light of the requirement that you only see O if O 'looks different' from its surround.
Inclined to think I see the gorilla whether or not I get any de re knowledge of him. Seems like a theoretical claim to deny this; not sure what benefits are to be gained by mooshing together the perceptual and the epistemic in this way. 

2010-06-26
Dretske on seeing
Reply to Mohan Matthen
Hi Mohan,
somewhat inclined to think this may be an issue to legislate away. Not entirely clear on the ups and downs of one or other choice.

2010-06-28
Dretske on seeing
Reply to Benj Hellie
Hi Benj,

I do agree with you that there is a penumbra of legislatable issues here.  But there is also an important core notion.  Seeing x should give a subject the ability to act relative to x, where acting relative to x can include such physical actions as touching or pointing towards it, sensorimotor actions such as tracking or attending to it, and epistemic actions such as investigating it or recording its characteristics.  It seems to me that something like visual differentiation is needed for any of these.



2010-06-28
Dretske on seeing
Reply to Mohan Matthen
Hi Mohan, agreed that there is a connection between seeing and the abilities. Not sure if it is equivalence. May be that the abilities in regard to O are a sufficient condition for seeing O. May also be sufficient if one's situation in regard to O is very similar to a situation in which one has the abilities in regard to O. Really uncertain what could ground things coming out one way or another here. 

2010-06-28
Dretske on seeing
Reply to Benj Hellie
I don't agree, Benj, about visual differentiation being sufficient but not necessary. 

For me to have a visually guided ability to make contact with (or flee) a tiger, to track it, to form perceptual beliefs about it, and to find it, it is necessary for me to visually differentiate it from its background.  I can't do any of these things if I have failed to penetrate the tiger's camouflage.  My point is simply that while the locution 'see X' should to some extent be legislated, there is at least one sense in which it demands visual differentiation of X from its background. 

Grice's sense does not seem to require this, because he allows the Pricean form: I see M if M causes (in the right sort of way) a sense-datum with which I am acquainted.  In that sense, I can certainly see a tiger that I fail to differentiate.  However, Grice sometimes also uses a de re 'looks' statement to describe the effect on the perceiver of the thing that s/he sees -- and this is maybe closer to what Dretske had in mind. 

Now I don't know whether what I have just says implies that the sense of 'sees' that was Grice's analysandum is different from the sense of 'sees' that was Dretske's.  It might just imply that Grice was wrong -- certainly that is how e.g. Snowdon 1980 understands the matter (if I understand him aright).  Perhaps some legislation would help sort this out.

2010-06-28
Dretske on seeing
Reply to Mohan Matthen
Hi Mohan, 
Agreed that we could legitimately legislate usage which tracks either the differentiation or Pricean phenomena. Inclined to think that they come apart rather rarely, and that there is a sort of modal instability to situations in which they do. So I'm inclined to think that the broader Pricean phenomenon is probably a bit more 'rounded out', though maybe the narrower differentiation phenomenon is more 'what we care about'.

Doubt that there's enough in ordinary use to sort this issue out. In my view, the point of having a theory involving these perceptual relational phenomena is to enable us to explain various modal and causal facts about perceptual actions like looking, watching, etc. Where exactly we want to set the boundaries of the explananda is an issue that may be of little importance to the core functioning of the theory.

2010-06-28
Dretske on seeing
Reply to Benj Hellie
Benj, since we seem now to be in complete agreement, let's just congratulate each other on having discovered another thoroughly brilliant philosopher and all around bon vivant.

Small point: the phenomena are Gricean not Pricean -- as you know, Grice defended Price on sense-data and 'L-phenomena' but added the causal requirement.  (Very confusing, those angrez.)

2010-06-29
Dretske on seeing
Reply to Benj Hellie
Benj,

RE:  "Doubt that there's enough in ordinary use to sort this issue out. In my view, the point of having a theory involving these perceptual relational phenomena is to enable us to explain various modal and causal facts about perceptual actions like looking, watching, etc. Where exactly we want to set the boundaries of the explananda is an issue that may be of little importance to the core functioning of the theory."

I'm a little hesitant about your last sentence when we're looking at cases where the subject's reports don't match up with what is independently known.  (This hints at the primary reason that I don't think the non-epistemic thesis holds much promise in maneuvering with skepticism, incidentally.) 

Case 1 (after Dretske):  "Didn't you see the cufflinks?  They were right there in plain view."  "No, I didn't see them.  I don't know how, but I just didn't see them."

Case 2 (after Wright):  "Don't you see the moth on the bark [points to an area of the tree bark inches away]."  "No, I don't see it.  I just don't see it."  "There.  Right there!"  [Entomologist touches area of bark and moth flies away]."  "Oh!  Now I see it."

Case 3: (after Kneale):  [Tailor]: "I can darn this hole in your coat so that no one will be able to see it [the darn], but this other hole is pretty large.  I'll do my best."  [Customer, later]: "I don't know how you did it, but this is great work!  I can't see either darn!"  [Tailor]:  "I glad you're pleased, sir!  [mutters under breath:  I can see both of them.]

Although Cases 2 and 3 involve visual camouflage whereas Case 1 involves what I call "psychological" camouflage, all three cases clearly satisfy Dretske's visual differentiability condition--cases of non-epistemic seeing in which "seeing a bug in this fundamental way is like stepping on a bug" (Dretske).  Yet, it seems counterintuitive to say that the subject saw the object in question, unless we mean somthing trivial by "seeing," e.g., having light reflected off of the object into one's working eyes, etc.  So, I think that "setting the boundaries of the explananda" in these cases is critical to the core of any plausible theory of perception.  Specifically, visual differentiability varies from person to person under suitably uniform public conditions.  The boundaries will vary; some imposing no constraints on the theory while others pose serious challenges.




  



2010-06-29
Dretske on seeing
Reply to Daryl Close
Hi Daryl,
Zeno Vendler's monumental 1957 paper 'Verbs and times' distinguishes between two senses of 'see': one of them has a clear 'sudden change' or 'eventive' reading and the other has a clear 'stative' reading. The former can be sensibly swapped out for 'notice' or 'spot', the latter can't. 

As I understand the progression of the discussion, our quarry is what is required for seeing in the stative sense. I am inclined to think that your examples pertain to seeing in the eventive sense. 

(As it happens, I agree entirely that discrimination in the most robust sense is needed for eventive seeing. However I worry that Dretske may have allowed opinions concerning eventive seeing to slop over into his judgements about stative seeing.)

2010-06-30
Dretske on seeing
Reply to Mohan Matthen
It seems to me that seeing, as Hume recognized, involves both impressions ( non-eventive seeing), and ideas, which focus on some ingredient of the visual array which plain seeing, as distinct from noticing, presents. What ideas replicate, from our impressions, is what struck as, what we noticed in wht we saw. Humw thought ideas were instantly obtaned from imperessions, since we r always aware of what matters to us. So the Dreske distintion between seeing and noticing has Humean roots.   

2010-06-30
Dretske on seeing
Reply to Mohan Matthen
 Isn't non epistemic seeing supposed to follow from Sperling's experiment for Dretske,
which proves that we are conscious of more than we can 'phenomenally' judge or know?
I thought it was supposed to be connected with a kind of iconic (analogue) memory very short term?

2010-07-01
Dretske on seeing
Annette, I am not sure I understand.  The target of Dretske's object-seeing is an external object, not an element of the visual array.  And I think he would allow, or even insist, that one can see such an object without noticing it.  Thus, visually differentiating an object from its background need not be a conscious action of the perceiver -- it may be accomplished sub-personally. 

Aspasia, Sperling's experiment is indeed supposed to demonstrate some kind of iconic memory -- but does it indicate a non-segmented display?  I am not clear that it does.  Nine numbers are seen; three are remembered.  Does anybody say that Sperling shows that you simply see a pixel-array?  Also, since the Sperling experiment probes a subject-report of what is in the display, I don't see why it is non-epistemic. The subject believes that s/he has seen such-and-such numbers.   

2010-07-01
Dretske on seeing
Reply to Mohan Matthen
Dear Mohan,  Sorry my post was a bit garbled. I thought there could be a link between non-epistemic seeing, contrasted with seeing as distinguishing, and Hume's contrast between ideas and impressions. He says "all the perceptions of the mind are double, and appear both as impressions and ideas." When we see, there will always be more in one's visual field than what we notice and recall, the ideas we take away from that complex impression will not exhaust its content. Some call this remainder non-conceptual content.Hume thought ideas will always be there when we have sense impressions, although their main role is to enable us to think of things in their absence. But why drag Hume into your Dreske musings?   Annette  

2010-07-02
Dretske on seeing
Reply to Mohan Matthen

I wanted to raise a point in a general way: I think there is an element as an almost covert point (he would let people do whatever they wanted with mirrors, coins etc.) raised against the figuring of beliefs into determination of objects in Dretske's simplifying use of objects as determinants in seeing. Echos of that type of point would be (say) that some specific grounds of a justified belief may figure in instances of invalid reasoning which might also cite those grounds. (William Alston has a nice quotation to that affect from Marshall Swain in his 'An Internalist Externalism '(230, Alston's 'Epistemic Justification).) By the way, in the connected footnote, Alston cautions that (what he calls) experiential grounds are unlike beliefs in that there is '... no propositional or factive ground to serve as a ground rather than the experience itself. One who does take experiences to be essentially propositional attitudes will find the same problem with doxastic grounds'.
I think there is as well a separate issue over allowing oneself belief in the appearance of phenomenality.

2010-07-02
Dretske on seeing
Dear Annette,  The contrast between non-epistemic seeing and visual-field seeing is very interesting.  Here are a couple of thoughts:

a.  In what some philosophers think of as pure sensation -- personally, I don't believe that there is such a thing -- there are no objects.  Consequently, there is nothing substantial to believe.  You can believe "Red there at 2 o'clock", but if this is taken as a statement about your private visual field, the believing is implicit in the visual consciousness, and consequently there is no interesting distinction between epistemic and non-epistemic seeing.  If on the other hand, it is taken (non-Humeanly) as a statement about an external occurrence of red in public space, I don't think it is even well-formed, or meaningful.

b.  For Hume: red is a general term that applies to a plurality of particulars.  When he says "all perceptions of the mind are double, and appear both as impressions and ideas," he is actually implying that the general term red applies to both an impression and an idea.  Which in turn implies that the impression was, so to speak, conceptual -- red is implicit in it.

c.  What happens when you fail to take red away from an impression?  Suppose there is a shade of red you see only once, and do not recall -- so that it is not in fact part of any idea?  I think Hume must mean that "all perceptions of the mind are potentially double."  And this vindicates conceptual content in the above sense. 

Best, Mohan

2010-07-02
Dretske on seeing
Matthew, I find this very difficult to follow:  too many embedded causes for my tiny brain.  Are you saying that whether you see x or or see y is partially determined by what you believe?  Can you explain why you think this is so?  Mohan

2010-07-02
Dretske on seeing
Reply to Mohan Matthen
Mohan, I'm not saying that. I'm saying (and I thought I was following through on your way of putting things) something about that species of determinacy (centered on object seeing) that interested Dretske that remained in the absence of beliefs or merely alongside beliefs and alongside propositions. I was trying to undermine (which is what the grounds argument gives me) any connection with the phenomenology. I may regret saying this—it's uncouth—or it's even a kind of an imposing of the Millian phenomenon, stretching the non-descriptive (and that's me suppressing the obvious links to causal theories). My point about grounds being inconclusive (in the sense of the passage I was referring to) as a support to reasoning should be familiar enough from elsewhere, but I'll give the whole passage I mentioned and perhaps you'll say whether you think it's relevant:

Smith and Jones have the same evidence (grounds) for belief that p, where the evidence consists of the proposition p v (p&q). Both subjects come to believe that p on this basis of the evidence (and no other evidence). In the case of Smith, the mechanism for generating the belief is an inference which instantiates a tendency to invalidly infer p from any sentence of the form 'p v q'. In the case of Jones, the mechanism is an inference which is based on an internalized valid inference schema (of which several are possible). It seems clear to me that only Jones has a justified belief that p, even though they have the same grounds.

(Alston quoting Swain on Alston)

2010-07-02
Dretske on seeing
Ah, that's clearer now -- thanks Matthew.  Alston says that one person can go validly while another goes invalidly from the same grounds to the same conclusion. 

I am still having some trouble, though, in figuring out how this point would transfer to perception.  You quote Alston as saying:
... no propositional or factive ground to serve as a ground rather than the experience itself. One who does take experiences to be essentially propositional attitudes will find the same problem with doxastic grounds. 
I half-agree with the first part of this: when I conclude on the basis of my visual experience of the chocolate that it is brown, I may have only my experience of the chocolate to go on.  Now suppose (as I think) that the experience is indeed a propositional attitude: the experience that leads me to the belief that the chocolate is brown has content "That is brown".  

This is where I get stuck.  Are there supposed to be multiple routes from the perceptual experience to the conclusion?  (Perhaps there are, but there is only one instinctive route, I think.)  You say that this undermines the connection with the phenomenology.  What connection?  And how is it undermined?

2010-07-05
Dretske on seeing
Reply to Mohan Matthen
Hi Mohan, I'll come back to that Alston quote. I think Dretske's simple seeing is to act as an admonishment, like some 'whatever' clause in conversation. Of course although that's refined it may not be adequate ('seeing' to be legislated in terms of use as proposed earlier whereas I'm just emphasizing what I thought of as the goad part). By phenomenology I mean things that collect like experience acting as qualitative inputs, character, underpinnings to representational content, comic residues etc. (as Snowdon uses  'phenomenology' very helpfully contrasting a 'looks as if' statement with just a 'tentative judgement'), and answering your last two questions, I think that groupings of those things I mentioned support judgements at least relating to test cases which should be more than just their equivalents in some statement/statements; which, depending on how one weighs statements against propositions, reintroduces the Alston point. But first, isn't it true that Dretske's simple seeing is part of an informational context which is never realized up to its full potential according to Dretske? And a point is, one can have examples involving experiential grounds subject to a kind of Dretskean condition. Alston runs on:

... consider two persons, A and B, who come to believe that a collie is in the room on the basis of qualitatively identical visual experiences. But A recognizes the dog as a collie on the basis of distinctively collie features, whereas B would take any largish dog to be a collie.

The goad part in all of this was—I thought—that something (not content) should mediate assignments or ascriptions for which some full description probably shouldn't be available in attempting recovering that thing. In Alston, and this is where you might feel there is some disconnection, that might be some fully discriminatory mechanism working in the appropriation of refined grounds as a way of taking into account a belief input. Or as Alston also puts it—now in the context of subject-relativity—'making justification hang on what the subject "had to go on" by way of support, rather than what the subject actually went on...'.

2010-07-07
Dretske on seeing
Reply to Mohan Matthen
The Alston example is really interesting- I thought that if we get rid of
an additional 'conceptual' judgement acting on the content of perceptual experience and treat
the perceptua discriminatoryl experience as non further analysable we won't
treat the case of looking a tomato as a case of an inference from grounds to conclusion. Do U think such a move can be allowed? But then what happens with the Alston example which as U suggest
seems to work best for logical connectives for which U have rules of inference?
Can't we think of a perceptual judgment as one that 'binds' the different sense modalities which is not
'conceptual' and nothing more than that?

2010-07-10
Dretske on seeing
Reply to Mohan Matthen
The person who sees a collie and recognises it as such, as disticint from the one who knows only collies as large dogs, may have the same thing in their visual field (which,Mohan, is not like a clockface but is three dimensional) yet identify it as  collie on different grounds, both having noticed the size, only one the special features in which collies differ in appearance from other large dogs. That is a difference in their prior experiential and conceptual knowldege. But if what they see is both a collie and, with it, a German shepherd, the differences will strike them, so that they will realize that not all large dogs are collies. Perceptual content is always different from conceptual or classificatory content,  though sometimes it can lead us to enrich the latter, as when we see that collies are dfferent from German shepherds. As Hume said, impressions are not the same as ideas, and somtimes the impression (of a dog) can come with rhe idea of "collie, not any other large dog," sometimes not, only with the idea of " large dog, likely a collie." Thanks, Annette Baier.   

2010-07-10
Dretske on seeing
Hello Annette,  Interesting example. 

You say: "Perceptual content is always different from conceptual or classificatory content".  My position is that descriptive perceptual content is classificatory content (though obviously not all classificatory content is perceptual).  That's the Sensory Classification Thesis in chapter 1 of my 2005, and it's what I meant when I was talking about red in my earlier post. 

But this is not to deny your general point, which, I take it, is that perceptual content can be the object of classification by the perceiver.  Dretske has a wonderfully rich and fecund discussion of this sort of point in chapter 6 of his 1981 -- I kind of think that this is his most brilliant work.  (Fred: Are you following this discussion?  Are you blushing?)

Coming back to Hume: does he think that the idea of a collie is a complex of visual ideas?

cheers,

Mohan



2010-07-10
Dretske on seeing
Reply to Mohan Matthen
Dear Mohan, I suppose that for Hume the idea of a collie is a complex idea, but whether it is a complex of ideas of simpler spatial parts or some other sort of complex, say of an animal of a certain sort, depends on whether it is as you say a visual idea, the idea of the look of a collie, or not. Hume certainly needs ideas which are not copies of any sense impression or complex of them. The idea of an idea, for example, or the idea of inference, or of thinking. I have just given a talk at the AAS conference in Sydney defending Hume against conceptualists like McDowell (who was there, and unconvinced.) Thanks, Annette