Surveying Loose Talk

(Note: This post began its life as a comment to Josh's recent post on Justin and Edouard's paper last night. Since a technical glitch prevented publishing it, I wrote a full blog post on Ethics Etc, where I occasionally blog. However, since I suspect there isn't much overlap in the readership, I asked Thomas if I could cross-post here as well for my target audience, and he graciously invited me to do so. Justin and Edouard - sorry for accentuating the negative. -A)

This is the first in a series of posts about recent work in experimental philosophy. I will be examining some persistent general issues with the different experimental approaches by way of looking at particular papers in some detail. I’ll begin with ‘Two Conceptions of Subjective Experience’ by Justin Sytsma and Edouard Machery. The problem that the study highlights is that everyday language is often vague, ambiguous, or just spoken loosely, so that we can’t draw conclusions about people’s concepts just by looking at what they say in response to prompts. We first need to tease out just what people mean, and this can’t be done in a survey that doesn’t allow for a back-and-forth between the researcher and the subject. This would be a problem even if experimentalists solved all the other problems raised by myself and others.

Here’s how the abstract of the paper begins:

Do philosophers and ordinary people conceive of subjective experience in the same way? In this article, we argue that they do not and that the philosophical concept of phenomenal consciousness does not coincide with the folk conception. We first offer experimental support for the hypothesis that philosophers and ordinary people conceive of subjective experience in markedly different ways.

There’s more in the paper, but since my focus is going to be on the alleged experimental support, I will not discuss the rest of the paper.

Sytsma and Machery’s (S&M) thesis is that while philosophers conceive of perceptual experience and bodily sensations as belonging in the same category of subjective experience, unified by the fact that there something that it’s like to be in the relevant states, the folk don’t do so. Their evidence for this comes from the following online study. Both philosophers and non-philosophers were presented with the following prompts about a robot (and corresponding ones about a normal undergraduate):

Jimmy is a relatively simple robot built at a state university. He has a video camera for eyes, wheels for moving about, and two grasping arms with touch sensors that he can move objects with. As part of a psychological experiment, he was put in a room that was empty except for one blue box, one red box, and one green box (the boxes were identical in all respects except color). An instruction was then transmitted to Jimmy. It read: “Put the red box in front of the door.” Jimmy did this with no noticeable difficulty. Did Jimmy see red?

Jimmy is a relatively simple robot built at a state university. He has a video camera for eyes, wheels for moving about, and two grasping arms with touch sensors that he can move objects with. As part of a psychological experiment, he was put in a room that was empty except for one blue box, one red box, and one green box (the boxes were identical in all respects except color). An instruction was then transmitted to Jimmy. It read: “Put the red box in front of the door.” When Jimmy grasped the red box, however, it gave him a strong electric shock. He let go of the box and moved away from it. He did not try to move the box again. Did Jimmy feel pain when he was shocked?

When it came to the normal human case, there was no significant difference between philosophers and the folk: most agreed that the person saw red and felt pain. But in the case of the robot, most philosophers answered that the robot neither saw red nor felt pain, while most non-philosophers answered negatively to the pain question but positively to the seeing red question. S&M conclude:

On average, the folk (but not philosophers) are willing to ascribe the perceptual state of seeing red to a simple robot. Given the illustrative centrality of the example of the redness of red to the philosophical concept of phenomenal consciousness, our results indicate that philosophers’ concept of phenomenal consciousness is not how the folk understand subjective experience. (p. 13)

Now, there is a pretty obvious response to this: maybe ’seeing’ is ambiguous between differential responsiveness to the visible features of the environment and such sensitivity + phenomenal experience, or, better yet, the first of these concepts gets triggered in loose talk. It’s not that the folk and philosophers have a different concept: rather, the ordinary term can express either, and the folk latch on to the former in the context of the study. Consequently, there is no one-to-one correspondence between language and thought, and we can’t draw any conclusions about the latter by looking at the former before we’ve established one by resolving ambiguities, taming vagueness, or tightening loose talk.

I raised the issue in a Q&A a couple of years ago (as did a number of commenters on the Experimental Philosophy blog when the study was posted there recently), and it turned out, unsurprisingly, that this was not a new concern to the authors. Given that the response is so frequent, S&M address it in the paper. As they nicely put it, according to it ’seeing red’ has both informational and phenomenal readings, and the folk select the former, unlike the philosophers. So what, according to them, is wrong with this natural thought?

Their first response is that the distinction between the two readings is ‘ad hoc’. This is a strange thing to say. Roughly speaking, a distinction is ad hoc if the only use it has is to respond to a criticism. But the fact that ’seeing’ and other sensing verbs have an informational reading should be obvious anyway, and it would be shocking if nobody had pointed it out before in the literature. After all, we talk like this all the time: the baby alarm ‘hears’ the baby cry, and alerts the parents, the missile ’senses’ the heat pattern and redirects itself, and so on. Ambiguity is probably too strong a term to describe this - as I’ll suggest, there’s some reason to think that this is just loose talk, a convenient shorthand that we (the folk) recognize as strictly speaking false. Compare this to the loose use of ‘knows’ - we use ‘know’ for accidentally true belief all the time, but are willing to withdraw the claim if some literalist busybody pushes us. Surely (please tell me) nobody would parade a survey showing this as a discovery about the folk concept of knowledge and its divergence from the philosophers’ one.

What triggers one reading rather than another is a separate issue. I don’t think the burden of proof on why the folk pick out one rather than another reading is on the ambiguity/loose talk theorist, who might not be particularly interested in theorizing about people’s actual use of words. You wouldn’t demand that of other ambiguity claims, would you? But if I had to guess, I’d say most people are being charitable and choosing the least demanding reading - “Ummm…. if the robot didn’t see the box was red, it couldn’t choose the right one”. If S&M were serious about testing for this, they would run a study using prompts in which the ambiguity or looseness is naturally resolved one way or another, and compare the results. I wonder what simply adding “literally/strictly speaking see red” would do? (If that made a difference, it would support the loose talk interpretation.) Or “did Jimmy a) exhibit sensitivity to different wavelengths of light/visual information, b) exhibit sensitivity to redness, c) see red (tick as many boxes as you think apply)”. Sure, this is not trivial - but it’s not that hard either (waiving for the moment general problems with surveys as a tool for getting at people’s concepts).

Second, S&M claim that the pattern of findings about the folk doesn’t support the ambiguity interpretation, because in that case, they claim, “we would expect that a reasonable proportion of the folk would answer negatively to the question “Did Jimmy see red?”—as philosophers do—while the remainder would answer positively” (p. 16). Now, why would we expect this? Suppose the folk note that the robot is sensitive to colour and charitably pick the informational reading when answering the question. What pattern of responses would you expect, when everybody is in the same context? You’d expect the vast majority to pick the same sense. And that’s just what we find. Only in a different context where the phenomenal reading was made salient would you expect ‘a reasonable proportion’ of the folk answering negatively. Maybe tell a story about a blind person, who is able to detect red surfaces with the help of a super-duper scanner. Would the folk think this person was seeing red? I doubt it. Then a completely parallel story with Jimmy. My hunch is that you’d get a reasonable proportion of not seeing red - especially if alternative responses were available.

Finally, S&M claim that the comments the minority who say the robot doesn’t see red don’t suggest they make use of a phenomenal reading. But again, the most natural reading of the comments is that the subjects are indeed groping toward the phenomenal reading. S&M relate that they say things like “I’m not sure how color is understood by robots and computers”; “seeing is a human attribute”; “seeing is something which animals do.” Now, if they weren’t restricted by the survey method, they could have asked them “What do you mean by ‘a human attribute’? What’s the relevant difference between humans and machines?”, and so on. Alas, the dogma of experimental philosophy, I’m afraid, says that this is contaminating the data, or something. So we don’t know what they were getting at. But it’s pretty reasonable, I think, to assume on the basis of such responses that the people who didn’t think the robot literally saw red figured that you need to be a creature that has conscious experiences.

So, in short, the three responses S&M offer to the ambiguity or loose talk criticism are extremely weak - if anything, the evidence they appeal to supports that interpretation. Is it a surprise that people with a philosophical training gravitate toward a narrow, literal interpretation of a term, while people without such training prefer a loose usage that works well for practical purposes? Not in the least. Why not, then, accept this simple and general hypothesis about the entirely predictable set of data, rather than reach for a story that attributes massive error to generations of really, really smart members of the folk whose curiosity about these matters led them to reflect on them professionally? And why not properly test for the obvious competing hypothesis, when at least the first steps would be very easy to take, even within the constraints of a survey methodology?

Comments

[Justin Sytsma]

Antti,

I appreciate your comments. Let me first note that everyone doing work in this area, and that includes Edouard and I, recognize that these are difficult questions that we are trying to answer and that a great deal of work will be involved in answering them. The three studies we report on here (following-up on our (2009), following-up on Knobe and Prinz’s (2008)) are part of a large ongoing project. As I noted in the comments on the previous post that your post was going to be a response to, I have since done further research specifically related to the objection you focus on:

“Finally, in some subsequent work I’ve explored this type of objection further, looking at how people understand states like seeing red. Long story short, it seems (perhaps not too surprisingly) that non-philosophers generally hold a naïve view about colors, treating the colors that they are acquainted with in ordinary visual perception as mind-independent qualities of the objects seen. But, starting from this view of color perception, the split between an information reading and a phenomenal reading of ‘see red’ does not get much traction.

I’m happy to say more about that work and that response, here, if you would like; especially as you find our other responses unconvincing.

I appreciate your suggestions for other follow-up studies; in fact, I intend to run them (they’ve been added to my experiments-to-do list). But, why do you think that we aren’t serious about testing the objection you focus on? This seems rather unfair of you to imply. I have run and am running a number of studies following up on the research reported in the “Two Conceptions” article (studies that would have, unfortunately, bloated that article beyond reasonable bounds if they were included there). In fact, one follow-up study that I’ve done pilot testing on and am planning to run on a larger pool of subjects in the near future is rather similar to what you suggest: After giving participants a variation on the robot red case on the first page, on a second page I asked them whether Jimmy saw red as they do (almost everyone who said that Jimmy saw red, said that he saw red as they do). In another pilot study that developed from some studies using the case of imagining that Josh Knobe suggested on a previous thread, I ran a within-subjects design for the robot and human red cases (I found no difference in scores for the robot seeing red when they saw the robot first versus when the saw the human first). Now, of course, one could raise issues with each of these studies as well and argue about what they show us. Heck, I do that myself for my studies (I am tempted to say it is part of my philosophical temperament)! But, I think they help fill in the picture. And, as I said, I will also plan on running the study you suggest – I very much like the addition of “strictly speaking” (or some such construction)! I don’t think that you really meant it in the tone of helpful suggestion, but I’m going to take it that way and gladly.

I take it that we have a rather deeper disagreement than that you don’t find the responses to the informational/phenomenal objection given in “Two Conceptions” to be compelling. As you write at the beginning of your post, “we can’t draw conclusions about people’s concepts just by looking at what they say in response to prompts. We first need to tease out just what people mean, and this can’t be done in a survey that doesn’t allow for a back-and-forth between the researcher and the subject.” As you don’t argue the point here, I won’t either. Let me just say, however, that I find this to be unduly pessimistic. I think that a variety of methods, including starting with survey methods, can cast light on the folk theory of consciousness. I think that the work that has been done in the past few years by experimental philosophers on this topic demonstrates that. Is the work finished? No. Do we have a final answer? No. Are we slowly making progress? I think yes. In fact, I think you have contributed to that by suggesting further studies to run. Now, this might be more slow and plodding than you like, but science is like that. And, you might well be correct that to really answer these questions we’re going to need to employ other methods, perhaps methods that allow for a back-and-forth between the researcher and the subject. Of course, unlike the suggestion you make in your 2007 Philosophical Explorations article, I think that you can do back-and-forth without “[giving] up the detached stance of the experimentalist observer” – one can use structured interviews, for example. In fact, this is an approach that I am very keen to pursue; unfortunately, it is quite time consumer to do well (and so will have to wait until I am done with the job market before I can commit the time to it).

Finally, I am rather confused by one of the comments you make in the last paragraph. You state that our story “attributes massive error to generations of really, really smart members of the folk whose curiosity about these matters led them to reflect on them professionally.” Could you clarify? What do you think is the massive error that we attributed to folk-turned-philosophers?

Cheers,
Justin

[Antti Kauppinen]

Thanks, Justin - that's useful. I'm curious to see the follow-up studies, and the preliminary result you mention is intriguing.

The most critical thing I said was that you didn't seriously try to test the loose talk/ambiguity hypothesis. There's three reasons I think so: a) you were aware of the hypothesis (since it seems just about everybody made the response), b) you could easily have come up with the sort of modifications I proposed, basically just by asking yourself "Under what conditions would I be likely to deny that a robot sees red?", and c) instead, you pretty much invite loose talk by formulating the prompt in anthropomorphic terms ("Jimmy", "psychological experiment" (??), "he") - as if you had asked yourself, "Under what conditions would I be most likely to say that a robot sees red?". This suggests to me that you were hunting for a 'counterintuitive', newsworthy result, and weren't too concerned with what it actually means. I mean, to me it says that it's easier to get non-philosophers to talk loosely than philosophers, and that's just not an interesting difference between the two.

Now, I should emphasize that I don't mean to single out your study as somehow below par. I think the same concerns apply to much other research in the area - but I need to read the stuff more carefully before naming any names. Clearly, your effort to rule out the alternative hypothesis meets the standards of the field. My problem is in the first instance with those standards, not your work as such.

As to the primary quality view of colour, could you explain why you think it helps your case? I would, again, have thought the opposite. You couldn't get an informational reading if there was no information to be detected. The folk, according to your results, assumes the information is out there. Contrast this with sensations like pain. This, I presume, is not seen as a feature of the environment. Hence, no informational reading is available, and most everybody focuses on the phenomenal.

The massive error you seem to attribute is that there is a pre-theoretical understanding we all share about there being something it's like to have perceptual experience. You explicitly argue against Chalmers (p. 33), whose remark is very much in the tradition. And I'm saying: either people like Nagel, Chalmers (and hell, probably Descartes and Hume) got it all wrong, at least in terms of having a burden of proof against ordinary experience, or you've picked up on a superficial feature of how we sometimes talk about perception.

Finally, I think many of the things Eddy and Thomas say in response to my old paper, like doing structured interviews, are on the right track, and I'd really like to see people pay more attention to their suggestions if they do experimental studies. And as long as you're not interviewing for Trinity College Dublin next month, good luck on the goddamn job market!

[Justin Sytsma]

Hi Antti,

Our feeling with this paper is that while there were many more experiments to be run directly testing the objections we note to our first study, we wanted to explore the study in another way, devoting much of our space to the second and third studies and developing the valence hypothesis that we put forward. We also felt that the responses we gave to the informational/phenomenal objection were largely sufficient, perhaps reflecting that it does not strike me as nearly as plausible as it strikes others. (I’ll try to say a bit more about your objections to those responses later today.)

With regard to (C), Jonathan Livengood and I have run the robot red scenario changing the anthropomorphic language (except for the name “Jimmy” which is based on the robot Shakey). This was part of the third study in our paper on experiments on reference -- http://philsci-archive.pitt.edu/archive/00004634 (interestingly, our argument in that paper fits nicely with the concern with language that you express here). Here is the updated vignette that we used:

“Jimmy (shown below) is a relatively simple robot built at a state university. Jimmy is equipped with a scent detector, video camera, wheels for moving about, and two grasping arms with touch sensors for moving objects. As part of an experiment, Jimmy was put in a room that was empty except for one blue box, one red box, and one green box (the boxes were identical in all respects except color). Jimmy was instructed to put the blue box in front of the door. Jimmy did this with no noticeable difficulty. The test was repeated on three consecutive days with the order of the boxes shuffled. Each time Jimmy performed the task with no noticeable difficulty.

Did Jimmy see blue?”

The results were virtually identical to that reported in “Two Conceptions.” Additionally, Adam Arico and I have been working on a series of studies following up on a distinct objection he has raised concerning the presumed complexity of the robots sensory apparatus. In those studies we use a variation on the vignette above, again avoiding personal pronouns and so on, and preliminary results for seeing blue are similar to what is reported in “Two Conceptions.” (We plan on replacing the name as well in upcoming versions of this study; I currently favor simply calling Jimmy “the robot.”)

No worries about feeling singled out: I very much value the type of objections you raise and appreciate the suggestions. I think these are quite difficult questions, in large part because the language just doesn’t seem to make the distinctions that we want when it comes to phenomenal consciousness.

The reason that I think that the folk holding a naïve view of colors helps our case is that it seems to collapse the distinction between an informational and phenomenal reading for “seeing red.” I find that the distinction rests on drawing a distinction with regard to the term “red,” here – there is the information sense and the phenomenal sense, or the distinction between something like physical red (red-as-light-reflectance, for example) and phenomenal red (red-as-phenomenally-experienced-by-us). This distinction, however, does not seem to be made on the naïve view of color. The red that people are acquainted with is not thought to be phenomenal, but to be a quality of the object, and that quality of the object is the quality being described in a scientific account in terms of light reflectance, for example.

Interestingly, and perhaps rather surprisingly, it turns out that a majority of people (although not quite as large of a majority) treat pains the same way – they aren’t seen as features of the environment, but as features of the body outside of the brain. That is, they seem to hold a naïve view about pains as well! I am currently finishing up a draft of the paper that discusses these results and will send it to you if you are interested; I would love to get your suggestions on following-up on these follow-up studies!

Thanks for the clarification on the error bit. Yes, I think that Chalmers is mistaken insofar as he takes this to be the pre-theoretical understanding. In the paper that I am working on, I track this down more explicitly for Dennett – there are many indications that he holds that qualia are part of our current “folk theory of consciousness.” He seems to think that this reflects the Lockean secondary quality view being popularized and entering the common wisdom. At the time, though, thinkers like Descartes did not think that this view was part of our pre-theoretical understanding. For example, as Pier Bayle mockingly wrote of the Cartesian view of colors: “Ever since the beginning of the world, all mankind, except perhaps one out of two hundred millions, has firmly believed that bodies are colored, and this is an error.”

[Edouard Machery]

Interesting post, Antti! It looks like you have the beginning of a reply to our forthcoming paper!

A few clarifications are in order, though.

1. You mischaracterize our stance with respect to our studies. We do not think we have shown beyond any reasonable doubt that the folk conception of subjective experience differs from philosophers'.

Here is what we say: "What we have offered in this article in an initial empirical account of the common-sense conception of subjective experience. We have also presented preliminary evidence that it differs significantly from the philosophical concept of phenomenal consciousness. In contrast to philosophers’ emphasis on the phenomenality of subjective mental states, for the folk, subjective states seem to be primarily states with a valence. We are aware that much work remains to be done to establish our hypothesis about the folk conception of subjective experience."

So, yes, we are aware that more work remains to be done, and as Justin explained above, much work has already been done since this paper was working. But at the very least our results suggest that the folk conception might differ from philosophers'. And given the philosophical implications of this claim (according to our argument), this is noteworthy.

2. You mischaracterize our first reply when you say that "Their first response is that the distinction between the two readings is ‘ad hoc’." What we say is that "if the objection is to avoid being blatantly ad hoc, the critic owes us an explanation for the folk’s divergence from philosophers in choosing the informational reading over the phenomenal reading of “seeing red.”" What could be ad hoc is not the distinction itself, but asserting *without rationale* that the folk would prefer one reading while philosophers would prefer another one.

3. You fail to mention what is perhaps the strongest reason to resist the ambiguity objection.

As we put it in the paper, "Finally, while post-hoc explanations of our data that are compatible with the assumption that the folk’s understanding of subjective experience coincides with philosophers’ can be given, our positive account of the folk’s understanding of subjective experience suggests that this assumption is mistaken."

That is, we have a positive account of the folk conception of subjective experience, one which is supported by several striking findings (variation across and within modalities). This conception differs from philosophers'.

4. You write: "I don’t think the burden of proof on why the folk pick out one rather than another reading is on the ambiguity/loose talk theorist, who might not be particularly interested in theorizing about people’s actual use of words. You wouldn’t demand that of other ambiguity claims, would you?"

I am not following here. We say ""seeing red" means a single think for ordinary people", you say "nope, it means two different things, but it just happens that one sense is overwhelmingly preferred by ordinary people, a fact that conveniently happens to explain your results." Surely, to substantiate you answer, you want at least the beginning of a story about why this sense is preferred.

5. "I’d say most people are being charitable and choosing the least demanding reading." This reply won't work since it makes the wrong prediction for study 3. The same point applies to the reply to our second rejoinder. If you were right, we would expect a different body of answer in study 3. Fitting data with post hoc stories is harder than you think.

6. You write: "again, the most natural reading of the comments is that the subjects are indeed groping toward the phenomenal reading." This strikes me as false: the most natural reading as we say in the paper is that for some people, only animals and living creatures can see. Nothing to do with phenomenality.

7. You conclude: "So, in short, the three responses S&M offer to the ambiguity or loose talk criticism are extremely weak - if anything, the evidence they appeal to supports that interpretation."

Now, measuring evidence is a difficult matter, but I suspect you are working with the wrong picture. Your views about evidence strength seem to be that if pieces of evidence are not inconsistent with a view, i.e., if they fail to falsify it, they are weak evidence. Clearly our results fail to falsify the idea that philosophers' concept of consciousness is the folk, and we never pretended they do. However, this is clearly an inadequate way of measuring evidence.

A better measure compares the likelihood of hypotheses in light of data. It turns out that contrary to what your post suggests, we have hard evidence (a fact you conveniently fail to mention) that our results are unexpected by philosophers: As we predicted, philosophers make the wrong prediction about ordinary people's answers in the seeing red case (while their predictions are well calibrated for the other cases). What does it suggest? That if you think that "seeing red" has two different readings, as many contemporary philosophers plainly do (but as noted by Justin, not ancient philosophers), you would expect ordinary people to give about the same answers as you. So, for philosophers, our results in study 1 are surprising and unexpected, contrary to what you suggest in your post. Which is not to say that they can't be accommodated. Which challenges your claim about the evidence being "very weak."

8. You write: "there’s some reason to think that this is just loose talk, a convenient shorthand that we (the folk) recognize as strictly speaking false."

I didn't see any evidence for this in your post.

Cheers

e

[Antti Kauppinen]

Justin, I'd be very happy to take a look at your follow-up studies - though in fairness, my main motivation is to have more grist for my mill in writing a response paper for Phil Studies, which I'm now considering doing!

Thanks for explaining why you think the naive view of colour is relevant. Here's why I'm unmoved: I never thought that 'red' was the issue, but *seeing*. That is, it is the perceptual verbs rather than the object descriptions that have informative and phenomenal readings. And of course I should have thought a bit more and realized that sensation verbs can have the informational reading as well, precisely to the extent that pains etc. are thought of as objective features of the body that can be detected. (This is probably what the minority who say Jimmy feels pain think of.) One reason why the informational reading in this case could be expected to be weaker is that pain-talk doesn't seem to be factive. We don't flinch at a claim to feel pain in a phantom limb in the same way we do when someone claims to see a non-existent limb.

Well, actually, as soon as I say that, I can hear some loose talk: we don't actually hesitate to say, while watching a movie in which a character clearly has a hallucination, something like "He was in shock, because he saw a dagger floating in the air". So all right, we can have a merely phenomenal reading of sensing verbs as well, but that doesn't compromise their status as factive, does it?

(Hope this comes through now - didn't work yesterday.)

[Justin Sytsma]

“I never thought that 'red' was the issue, but *seeing*. That is, it is the perceptual verbs rather than the object descriptions that have informative and phenomenal readings.”

I don’t think that the perceptual verb can be pulled apart from the object description and still have a phenomenal reading, at least not without changing to a different sense of “phenomenal.” Our concern in the article is with phenomenally conscious mental states—or roughly mental states thought to have distinctive phenomenal qualities in virtue of which those states are phenomenally conscious. I take it that on the phenomenal reading of “seeing red” the relevant phenomenal quality is redness; as such, if the quality wasn’t characterized as being a quality of the perceiver’s mental state, we wouldn’t have a phenomenal reading.

Now, you might give a variation on the informational/phenomenal objection, giving an no-acquaintance/acquaintance objection, for example: You might think that there is a reading of “seeing red” on which it only matters that you get the information about the object and that you might get information about the qualities of objects without being acquainted with those qualities. Thus, Jimmy might somehow learn which box is red without ever being acquainted with redness and you could, I suppose, claim that this is naturally called “seeing red.” Alternatively, you might read “seeing red” as requiring that you be acquainted with that quality (however you characterize that quality – i.e., whether you characterize it as phenomenal or non-phenomenal).

The work I was calling on directly bears on the first objection, but not the second. Unlike the informational/phenomenal objection, however, the no-acquaintance/acquaintance objection doesn’t seem to have much surface plausibility to me. The main issue for me is that unless you revert back to the informational/phenomenal objection, it is unclear that the information-gaining pulls apart from the acquaintance, as the acquaintance seems to be the means of information-gaining in this instance. Thus, it is unclear to me why we should think that people read “seeing” in a way that involves making visual judgments using visual equipment but as getting that information in some other way than through visual acquaintance. I suppose that people might think that the robot was “cheating” in some way – maybe it can’t see red, but can see blue and green and inferred which box was red; or maybe it knows that the different paints dampen sound differently and it heard which box was red; or maybe it was just fed the answers – but in these cases I wouldn’t expect them to say that Jimmy saw red.

[Antti Kauppinen]

Thanks for the response, Edouard! Here's some replies:

1) The main claim of yours that I take exception to is simply this: "our results indicate that philosophers’ concept of phenomenal consciousness is not how the folk understand subjective experience." I don't think the results indicate that, since you haven't (according to me) ruled out a plausible alternative hypothesis. The criticism isn't that more work is needed; it is that this study should and could have been designed differently to reduce the likely effect of loose talk.

2) (and 4) As I noted in the original post, I'm not sure if a defender of two readings needs not only to point out the evident possibility, but also provide an account for what influences the folk and the philosophers in a different direction. In any case, on the loose talk hypothesis, the answer is ready to hand: philosophers, in general, are apt to read thought experiments strictly and literally. Though I might in an everyday context say "I knew Cork was going to lose the All-Ireland final", I wouldn't say such a thing in an epistemology classroom (well, except as an example of loose talk).

3) I don't think that works dialectically. You do provide a positive account of how people conceive of subjective experience - but it's built on the assumption that the folk don't conceive of perception in the same way as they do feeling and sensation. You can hardly appeal to that account in arguing against the claim that folk do, after all, have a unified concept of subjective experience, can you?

5) The loose talk hypothesis, as such, fits experiment 3 just fine. It is to be expected that a lot of variables influence just how loose the talk is going to be. As you know, the folk are easily influenced by using scientific jargon - I believe Eddy and co have a study on free will that shows this nicely. If I had to guess (and again I'm not sure if I do have to), I'd say people think of banana or vomit smell as something less objective than the scientific-sounding isoamyl acetate, and hence are less likely to opt for an informational reading. As I've pointed out in response to Justin, thinking of colour as something out there seems to be presupposed by an informational reading.

6) Well, you have no evidence for that, since you didn't engage in dialogue with them. The question is: *why* do they think only humans and animals can see red? Is this meant to be a brute fact on your account? You're the ones who are trying to convince us that these people don't answer in the negative because of a phenomenal reading, and you haven't begun to do so.

7) When you ask philosophers what they think others would say, they may again be too charitable, or, what's perhaps more likely, think of what people would say after a bit of clarificatory conversation. Or maybe they're just directly focused on what people think about the cases - in which case, for all your study shows, they may be just right.

8) A simple way to get evidence for this would be to run the sort of study I proposed in the original post, but I'm too lazy to do so. If someone wants to co-author a paper, let me know - I've got more ideas than I have time to write now.

[Edouard Machery]

Antti

Surely this is a fallacy: "I don't think the results indicate that, since you haven't (according to me) ruled out a plausible alternative hypothesis." Experimental results very rarely rule out alternative hypotheses (i.e., crucial experiments are very rare). What matters is the comparative likelihood of the hypotheses in light of the data together perhaps with the priors of the hypotheses.

I feel that your misunderstanding of how the evaluation of empirical hypotheses works affects most of your replies. Consider just your reply to 5. Sure, together with some auxiliary assumptions, your hypothesis might be consistent with the data, but again mere consistency is irrelevant. What matter is first whether your hypothesis make them as plausible as our hypothesis and, second, whether the auxiliary assumptions you need to appeal to (and you need to be explicit about these in order to be able to compare seriously the hypotheses - you just cannot handwave) have a high or a low prior probability. An example where the required auxiliary assumption has a very low prior probability is give by your explanation of philosophers' answers: "perhaps more likely, [they] think of what people would say after a bit of clarificatory conversation." This strikes me as a very curious interpretation of the task: they were just asked what people would say, nothing more, nothing less.

But perhaps you disagree with my way of assessing the competing views? Could this be the issue?

To answer another of your questions, yes this is meant to be a brute fact; this is in fact what people literally say (only animals can see), and I suppose it is charitable to interpret them as meaning just this.

Anyway, I hope this will turn out to be a paper, since I am eager to read the full-blown discussion.

E

[Antti Kauppinen]

Edouard,

I'm sure you're right that strictly speaking ruling out a hypothesis
is too much to ask. So let me rephrase the claim in terms of
probabilities. Your hypothesis that philosophers and the folk have a
different concept (or conception, let's not worry about that), which
has a low prior probability (what was it that explained that?). The
alternative, which a lot of people appear to have put to you, is that
the folk do after all have the same concept, but speak more loosely
than philosophers when it conveys potentially practically useful
information. Given that this patently goes on with other terms (as we
all should know from our own usage in non-philosophical contexts),
this hypothesis has a high prior probability. Both hypotheses fit the
data. I can't see how you can say in this situation that the data
indicate the truth of your hypothesis. If you can, we do indeed have a
very different conception of how to evaluate theories.

The question about why philosophers mispredict what the folk will say
is a very different one from the original. It is not about conception
of experience, but about predicting other people's answers. I admit to handwaving about this, but then again, I don't think it poses any challenge. Philosophers talk about charity, psychologists talk about egocentric bias, but either way, both agree that - not just in this case but perfectly generally - people's predictions and attributions of mental states to others are tilted towards their own. (Goldman and others take this to support simulation theory - but that's a whole different debate.) This is exactly what you find. So where's the surprise?

Of course, were someone to publish a widely read paper on the prevalence of loose talk among the folk, it's possible that philosophers would begin to think that the folk are more willing to credit others with knowledge/perception/whatever, at least verbally, and would consequently predict a divergence between their own and the folk's responses. This would be consistent with thinking that they share the same concept. In any case, your results suggest some awareness of this, as philosophers do think the folk are somewhat more likely to answer positively than they are themselves.

Anyway, I think it is likely that we'll get a chance to argue about these issues on paper. I much appreciate the feedback, which has already made the stakes and issues much clearer.