Benj Hellie

Benj Hellie (University of Toronto) “That Which Makes the Sensation of Blue a Mental Fact,” with commentary by Adam Pautz (University of Texas—Austin).  The paper, commentary, and author's reply can be found here.

Comments

[David Chalmers]

I'm no expert on Moore, but most of this paper made good sense to me. I found section 1.5 hard to follow, though. First, what exactly is Moore's "grievous error", which you promise to highlight? It looks like the only candidate is the "overplaying" and or the "confusion" that you mention in the paragraph on pp. 19-20. But it's not clear to me exactly how this discussion suggests an error in the passage in question. In that passage, Moore's point seems to be (i) that B, which the idealists take to be a content (i.e. property) of the sensation is in fact an object of the sensation, via the awareness relation (it may or may not also be a content), and (ii) whether or not B is also a content of the sensation, the fact that it is an object already suggests that it is not "inseparable" in the relevant sense. Your point in response seems to be that Moore concedes that he cannot exclude the possibility that it is a content of the sensation. But I don't follow why this is supposed to undermine Moore's point, which appears to be made in full consciousness of this concession. Perhaps you disagree with Moore about his point (ii): i.e., you think that if B is both content and object, then it will still be "inseparable" in the relevant sense. If so, it would be interesting to hear more.

Also, I couldn't follow the reasons for saying that Moore held that B isn't surface blue. The rough idea seems to be that he held that B is a quality of a sense-datum. The bridging premise must be that Moore held that surface blue couldn't be a quality of a sense-datum. But what's the reason for thinking that? Perhaps you're using the term "surface blue" so that it's true by definition that surface blue is instantiated by surfaces in the environment (rather than merely appearing to be instantiated by them). But even then, is it clear that Moore thinks that properties of this sort can't be properties of sense-data?

[Benj]

Thanks for the questions, Dave.

Concerning the first, Maja Spener pressed me on a related point after I wrote this draft. I suspect that what is going on here is a conflation of epistemic and metaphysical possibility: what Moore has argued is that he has no evidence that B is a property of his sensation, so it might turn out that B is a property of something other than Moore's sensation. What he needs is that B is "separable" from his sensation: that B could be a property of something other than his sensation. He thinks he has the latter thanks to having the former, but it isn't obvious how.

Concerning the second, I wasn't trying to argue that in Moore's view, B isn't surface blue -- rather, the dual of that, that it's not the case that in Moore's view, B is surface blue. In particular, I think Moore is uncertain whether B is surface blue; indeed, he dithers about this point from about 1905 to at least 1942, finally making up his mind in 1957 that it is not. I should probably say that in the paper!

[susanna]

these are a couple of questions for adam plus a comment.

in your reply to benj you say on p. 12 that the relational view offers an intuitive account of hallucination, as compared with intentionalism. you write:

“why does it seem to you that you are aware of a blue square? The Relational View answer: because you are aware of a blue square! There is no explanatory gap here. But if one accepts this kind of argument for the Relational View, one is not relying on introspection, but on our apriori beliefs about what is intelligible and what is not. One is not here relying on one’s faculty of introspection, but one’s faculty of intuition or conceivability.”

(1) just to clarify the first step of the argument: Why exactly is there no e-gap between the explanandum (your seeming to be aware of a blue square) and the explanans (your being aware of a blue square)? it does not seem true in general that if you’re aware of X, then it seems to you that you’re aware of X. Were you making a more limited assumption of some sort for the case of experience – maybe something along the lines of the assumption that if you’re aware of having an experience, then it will seem to you that you are so aware? or was the thought rather that there really isn’t any difference in this case between seeming to have the experience and having it?

(2) on the argument that it isn’t introspection that’s being used:
is the thought here that if you took it as a point in favor of the Relational View that there is no explanatory gap between the explanandum (seeming to be aware of a blue square) and the explanans (being aware of a blue square), then

(a) you would be relying on both some comparison between these two things, and on an assessment of that comparison; and
(b) introspection cannot make such comparisons or assessments of them?

if that’s how the argument goes, I think it overshoots. it would count any claim that introspection supports a thesis about the metaphysical structure of experience as a mischaracterization, since one would always need to use a faculty of intuition to compare what introspection tells you the experience is like with the explanandum. well maybe that was just your point! but it is hard to see how anyone, even the people who think that introspection supports the Relational view, would think otherwise. their central claims would seem to be that introspection takes a stand on the structure of experience, and that the stand it takes is correct. those claims give introspection a pretty big role in an argument for the Relational view, even though that role is supplemented by whatever carries out the comparison.

(3) a comment: it seems like one pair of crucial issues in the debate about whether introspection favors the Relational view are (i) whether introspection takes a stand at all on the metaphysical structure of experience, and (ii) if it does, whether the stand it takes is correct. i guess there are really two kinds of modesty and immodesty in the vicinity. the Very-modest think introspection takes no stand at all; both the Less-modest and Less-immodest ones allow that it takes a stand but also allows that it is wrong; the Very-immodest think introspection takes a stand and is correct.

[Adam Pautz]

Hey Susanna, thanks so much for your helpful questions and comment on the exchange between Benj and me.

I had taken it that Benj had argued that Moore argued that we can know that the Relational View is correct on the basis of introspection. (It seems Benj only attributed the relational view to Moore. See his reply.) If this is correct, then introspection might provide us with *overall* justification for accepting the Relational View. This is a very strong claim. I considered the weaker claim that introspection might provide us with prima facie justification for accepting the Relational View. I raised some problems for this claim. Then I suggested that perhaps intuition and not introspection favors the Relational View over rival views, for instance Intentionalism. I said that Intentionalism faces a kind of gap problem, while the Relational View does not.

Susanna raises two questions and makes a comment about this. I’ll consider these in turn.

(1) Susanna’s first question was: why is there no gap problem for the Relational View?

Before I try to answer this question, let me correct something I said. In the passage of mine which Susanna reproduces, I wrote as if the analysandum were *seeming to be aware of a blue square*. Here I was being loose (my fault). As I set things up earlier, the true analysandum is *having a blue experience*. (There are worries about what this means, which I pass over.) So, the intuitions are:

[a] Intuitively, it’s possible to sensorily entertain a proposition yet not have a blue experience.

[b] Intuitively, it’s not possible to experience a blue thing and yet not have a blue experience.

But I don’t think this affects Susanna’s worry. Susanna’s worry may be applied, mutatis mutandis, to [b]. In general, “S is aware of a” and “a is F” do not entail “S has an F experience”. In illusion, you might be aware of a thing, which is in fact blue, and yet not have a blue experience, but a red experience instead. Perhaps this point applies to “item awareness” in general, even to “item awareness” of sense data. For instance, why couldn’t some one be aware of a sense datum, which is in fact blue, without having a blue experience, because they don’t notice its blueness? (Consider an individual in a psychophysical experiment, who is aware of a blue sense datum for an extremely short period of time.) A related point is made by Alex Byrne in “Intentionalism Defended”, p. 224.

Suppose we accept this point. Now consider the following view: having a blue experience = being aware of the blueness of a thing. (One can also imagine a “Positive Disjunctivist” version of this view according to which having a blue experience = being aware of the blueness of a physical object or the blueness of a sense datum.) This would not count as a version of the Relational View, as I understand that view. Nor would it count as a form of Intentionalism, as I understand that view. Now consider:

[b*] Intuitively, it’s not possible to be aware of the blueness of a thing and yet not have a blue experience.

I think that this claim is not vulnerable to the type of worry that Susanna raises. In this sense, this view, at least, does not face a gap problem. By contrast, I think that Intentionalism, the Property-Complex Theory, the Multiple Relation Theory, etc., do face such a problem. (Perhaps this is one reason, even if an unstated reason, for why some prefer Disjunctivism over some kind over Intentionalism.) If now we ask “why is this so?”, I guess I don’t have much to say. I just find the one thing intuitively possible, but not the other. In addition, I just find it counterintuitive that having a blue experience should be identical with standing in a relation to any kind of non-extended item.

Let me make another point here. I don’t think [a] and [b], or rather [a] and [b*], provide a strong argument against Intentionalism and for some other view, such as the Relational View. For once noticed, we see that this type of problem is ubiquitous. For instance, one of Armstrong’s arguments, in *A Theory of Universals*, against the Transcendent Universals Theory was that, intuitively, an object might bear the relation of instantiation to a non-spatial universal in Plato’s Heaven, and yet not be red. (This point of Armstrong’s is approvingly referred to by David Lewis in a footnote to *On the Plurality of World*.) But the same argument applies against almost any deep metaphysical theory of anything. So if you accept gap arguments, then you are precluded from giving any kind of serious metaphysical theory of anything, including experience. Gap arguments generalize – maybe they over-generalize. Perhaps this should suggest to us that we should not accept gap arguments. Another point is that, in any case, most people do not accept this type of gap argument as regards the mind-body problem, so they would not be swayed by it in other cases, including the present case. For my part, I find [a] and [b] persuasive, but I that there are countervailing arguments in favor of Intentionalism or something like it: arguments concerning the indeterminacy of perception (which by the way was one of Armstrong’s original motivations for Intentionalism), the Gestalt, imagery, and so on.

(2) Susanna’s second question was: why do I take it that the modal claims embedded in [a] and [b] (or perhaps better, [b*]) are supported by intuition, rather than introspection?

I confess I didn’t have any argument in mind here. The issue is difficult because ‘intuition’ and ‘introspection’ are technical terms for things we don’t have a good understanding of. I guess I just thought the following. When philosophers explain what they mean by “introspection”, they give a list of things you can know or be justified in accepting in this way: beliefs, tickles, and so on. When they explain what they mean by “intuition”, they give another list: that 2+2=4, that Zombies are possible, and so on. I just thought: the claims embedded in [a] and [b*] seem to fit better on the second list than on the first.

Susanna ends with a comment. She distinguishes between two issues: (i) whether introspection takes a stand at all on the metaphysical structure of experience, and (ii) if it does, whether the stand it takes is correct. I think that this is helpful. I guess I was arguing (against the claim which I wrongly took Benj to be attributing to Moore) that introspection does not take a stand on the truth of the Relational View, at least. That is to say: by introspection, you cannot get prima facie justification for accepting this view. Maybe intuition provides some support for this view over other views, but introspection does not. But, again, the issue is difficult because these categories are vague.

[susanna]

hi adam, thanks for the useful reply. it seems right that there is no gap between having a blue experience and being aware of the blueness of a thing.

i had a couple of small points in response to what you wrote. first, i was slightly puzzled by your suggestion that maybe we shouldn't be afraid of gaps - ie, that
maybe the fact that a theory T's explanans leaves a gap between it and the explanandum should not count against T. you cite most people's position on the mind-body problem, where they accept e-gaps as a fact of life and (if they're physicalists) set out to explain the gap away. but aren't you a dualist yourself? that we shouldn't fear gaps seems like a slightly odd thing for a dualist to say, since for a dualist a gap is just what you'd expect, given their position that when we say how physical stuff is related to phenomenal stuff, we are not in the business of giving metaphysical theories of the nature of the phenomenal. it's only the materialist who has to apologize for the gap, explaining why there is a gap even though the explanans really is telling us something about the nature of the explanandum. but maybe you weren't speaking qua dualist.

in any case, whether you were taking off your dualist hat or not, there seems to be slight tension between saying two things that you say. (this is the second verysmall point). on the one hand, you say, in effect, 'the fact that there's no e-gap counts in favor of the view V, even if not decisively in its favor'; but then you also seem to say 'there's a gap in other cases but that dsn't matter at all'. seems like
either gaps shld be dialectically relevant or dialectically irrelevant across the board!

[Adam Pautz]

Hey Susanna, thanks for the helpful follow-up questions.

The first question was: isn’t it odd for me, as a Dualist-sympathizer (and so, presumably, a friend of the gap in the mind-body case), to say that we shouldn’t be afraid of gaps in the philosophy of perception? The answer is that I am not sympathetic to Dualism because of the mind-body gap. I am sympathetic to Dualism because I accept an intentional theory of consciousness, and I think sensory intentionality is irreducible reasons independent of the gap (reasons which I explain in my exchange with William Lycan which is to appear next week in this conference). So, I think you can be a Dualist while rejecting the gap argument. It is not that I reject gap arguments. I am not sure what to think about gap arguments. I did mention one argument against gap arguments: they may overgeneralize in a bad way. The remark about how most people are Physicalists and so will not be swayed by gap arguments was not meant as an additional argument against gap arguments, but only a kind of sociological remark. I was not arguing from Physicalism to the failure of gap arguments, because of course I do not endorse Physicalism. I should have been more explicit about this.

The second question was: how can I say “the fact that there's no gap counts in favor of something like the Relational View, even if not decisively in its favor” and “there's a gap in other cases but that doesn't matter at all”. I reject the second claim: I think that gap arguments in other cases (like the Transcendent Universals Theory) count, but only weakly, because I don’t know what to say about them. Likewise, I think that the gap argument for something like the Relational View over Intentionalism counts, but only weakly, because I don’t know what to say about such arguments. (In my last post I said that I find the modal claims embedded in [a] and [b] persuasive, but I only meant this is in the psychological sense that I feel their pull. Again, I should have been clearer here.) Still, because it *may* be a good argument, I still think it is deserves to be factored into the balance sheet. It counts for something. But I think that, given other things like indeterminacy and the gestalt, the balance of consideration still tilts in favor of Intentionalism. Hope that makes better sense!