Folk Psychology and Phenomenal Consciousness Chair: Rick Gawne, University of Western Michigan Presenter: Justin M. Sytsma, University of Pittsburgh Justin’s video presentation is hosted at his website Commentator: Adam J. Arico, University of Arizona Adam’s paper or a larger version of his video Share this:FacebookTwitterEmailRedditPrintPinterestTumblrLinkedInLike this:Like Loading... Related 24 Comments Justin and Adam, Great Exchange! I have a clarification question for Adam about the first part of his comment. I did not really follow what the worry was. I thought that Justin’s point was the following. Dennett takes it for granted that ordinary people report being acquainted with properties of their mental states. For instance, they view redness as a property of their experience of things. In substance, Dennett proposes that consciousness theorists have to account for the fact that people make such verbal reports. However, they should not assume that these verbal reports are true, and thus they do not have to account for the properties of experiences that would make these reports true. This is the heterophenomenology strategy. (BTW, this strikes me as a familiar strategy to explain away judgments that are worrisome for some philosophical theory: instead of taking them to be true, explain causally how people could make such judgments.) Justin points out that it is far from clear whether Dennett’s assumption is true. That is, it is far from clear whether ordinary people report being acquainted with properties of their experiences (e.g., the redness of the experience of red things, etc.). If Justin is correct, there is no need to appeal to the heterophenomenology strategy. If I understood your criticism correctly, you object that Justin conflates Dennett’s claim about the ordinary people’s understanding of the sensuous properties (redness etc.) expressed in statements like “This shirt is red” and Dennett’s claim about people’s understanding of the science (of colors, I presume) – see about 5:30 in your comment. I do not see where the latter comes from. I would have thought that only the former was relevant, and that this was Justin’s target. Could you clarity this? Edouard PS: I will let Justin reply to the (good) points you make about the paper I wrote with him. Adam, Thank you for your comments. I hope that some of your concerns have been answered in creating the presentation. In particular, your first major concern is well taken and I hope that I have better clarified the relationship between Dennett’s suggestion that the folk hold a secondary quality view and the claim that the folk have the concepts of phenomenal consciousness and qualia. If not, I suspect that this will come up in later comments! Your second major point is that Edouard and I go beyond agnosticism in “Two Conceptions of Subjective Experience,” making the substantive claim that, by and large, the folk do not share the philosophical concept of phenomenal consciousness. That is absolutely true, although caution is certainly warranted here. I take it that our study provides tentative support for the claim and that currently the weight of the evidence is on the side of the folk not sharing the philosophical concept. Of course, there is more research to be done and this might change with future results. Your third point follows up on the second, offering an alternative explanation of the finding in our first study. As I understand it, the objection is that it might be that the folk have the concept of phenomenal consciousness, but have a lower threshold than philosophers for attributing phenomenal consciousness to an agent; thus, the folk (but not philosophers) think that the simple robot is phenomenally conscious and this is reflected in their saying that the robot can see red. An explanation of why the folk do not say that the robot can feel pain is then needed and the suggestion is that it might be because they think that the technological hurdles for this are greater than for seeing colors. This is, I think, possible and well worth investigating. In fact, I am very pleased to be able to say that Adam and I have been working on a study to test this objection (or begin to test it) and hope to have some results in the near future. I think that this reflects the promise of experimental philosophy in that rather than just arguing back and forth we can work together to gather the empirical data that (we hope) will resolve the dispute! With regard to your fourth point: I agree that having shown that the folk, by and large, find it possible for pains to exist unfelt does not necessarily mean that they locate the pains in the afflicted body part. But, this response is predicted by the hypothesis that the folk are naïve realists about pains and, as such, I think it offers some support for the hypothesis. Further support comes from Studies 3 and 4. Hi Edouard, Thanks for the question. Hopefully I can clear up any confusion. I think we all agree that there are two claims of interest: 1. “The folk express beliefs about having phenomenal experiences (e.g., ‘That shirt looks red to me’)” 2. “The folk possess a Lockean understanding of sensual qualities (i.e., they think these properties are mind-dependent, not ‘out-there-in-the-world’)” I think we all also agree that Dennett makes both claims. My concern is simply that since holding 1 does not require or depend on holding 2, that Justin’s data undermine claim 2 doesn’t discredit those who endorse claim 1. I agree, Justin’s intended target is Dennett’s endorsement of 1. My objection was simply that there needed to be more work shown for how that endorsement depends, in some important way, on also endorsing 2. It’s certainly not the case that Dennett only thinks the folk express beliefs about phenomenal experiences because he also thinks they take colors/pains to be mind-dependent. You say that Justin is showing that “it is far from clear whether ordinary people report being acquainted with properties of their experiences.” But my point is that he’s going about this in a roundabout way, which may, or may not, actually get to the intended destination. His data aren’t about people reporting being acquainted with the properties of their experience so much as their about where the folk are locating the properties of experience with which they are acquainted. While I grant that this gets at Dennett’s endorsement of the above claim 2, I was having some trouble seeing how this also gets at Dennett’s endorsement of claim 1. Does that clarify things a bit? Interesting experiments, and congratulations on the user-friendly online presentation! My concern is to do with the wording used in the experiments, where I think there’s some ambiguity or descriptions could encourage particular responses. Any difference between philosophers and non-philosophers could then be due to their degree of acquaintance with conventions in the literature. I’ll comment on two examples: 1) When asked if the robot “sees” red, participants may interpret this as a question about whether it could detect which object is red (or appears red to humans), which is distinct from an experience as of seeing the colour red. People might talk about a CCTV camera being able to “see” some parts of a shop and not others, for example. So could the folk have a concept of phenomenal consciousness that does include what it’s like to experience redness (or feel pain), operating in parallel with a concept of the ability to detect colours, etc, but not consider the former concept relevant to questions about a task involving simple detection (for which the robot is equipped)? 2) If told that people are being “distracted” from their pain, the implication is that the pain is still there, as otherwise they couldn’t be distracted from it, it would just stop. Could this influence responses? Adam, Thanks for this useful reply. I agree with you that Justin’s evidence bears on claim 2. (Incidentally, I would not put it in exactly these terms, since being mind-independent and being in the head aren’t identical.) Now, my worry is that *1* is ambiguous between two claims: 1a. People report experiencing red things. 1b. People report being acquainted with the properties *of their experiences* (that is, of their mental states), such as the redness of their experiences of firemen’s trucks. I suppose Justin does not want to deny 1a. (If he wanted to do so, though, your point would clearly hold.) What he needs to deny it seems to me (Justin?) is 1b. But if 2 is false, then 1b should also be false, it seems to me. Edouard Bryony, thanks for your comments! Choosing the best wording for studies like these is difficult and your concern is well placed. Your first worry is something that has come up a number of times. Edouard and I note the objection in our original paper and give a couple of responses to it (the paper is available on my website, “Two Conceptions of Subjective Experience”); I give a further response in “Phenomenological Obviousness and the New Science of Consciousness” and this response is further supported by the studies discussed in this talk. The objection, if I understand you correctly, is that “seeing red” is ambiguous. On one reading (which we called “informational”), a creature sees red if its behavior is responsive to the distinction between red things and non-red things. As the robot performs the discrimination task, it seems that it sees red in this sense. On a second reading (the “phenomenal” reading), however, to see red is to have the appropriate phenomenal experience. The claim is then that it might be that the non-philosophers we surveyed focused on the informational sense of “seeing red,” while the philosophers focused on the phenomenal sense. While this is possible, it is not clear to me why one would expect the folk to differ from philosophers in this way; absent an explanation, here, the objection is rather ad hoc. Regardless, the data doesn’t seem to support this account. If it is correct that the folk recognize both of these readings of the phrase “seeing red,” then we would expect that a reasonable proportion of them would read the question in the phenomenal sense and deny that the robot sees red. This is not the case, however. Almost all non-philosophers (84.6%) gave an answer equal or superior to 4 on this question. A second reason for doubting this objection is that we also asked subjects to explain their responses and these explanations did not suggest the informational/phenomenal distinction; in particular, the explanations given by those who denied that the robot saw red did not suggest that they understood this in a phenomenal way. Further, it is not clear that the folk hold the view of colors that this objection calls on. I find that the ambiguity found in the phrase “see red” reflects a theoretical view of color that the folk plausibly lack: The ambiguity primarily attaches to the term “red” (not “see”) and supposes a distinction between physical red (red-as-light-reflectance, for example) and phenomenal red (red-as-phenomenally-experienced-by-us). I think that there is reason to doubt that most people make this distinction, however. In fact, I think most hold the contrasting naïve view of color. On this view it is held that colors are real, mind-independent properties of things in the world; as such, the red we experience is taken to be the red of the physical object. This is supported by the first study reported on in this talk. That is an interesting point about “distracted” and one I hadn’t thought about. I suspect that the line could be changed to remove that wording without altering the results, as the implication seems quite subtle to me; but, suspicions are cheap and I’ll have to run a study to be sure. Thanks for drawing this to my attention! Adam and Edouard, I think Edouard has it right. Certainly, I have no problem with 1a: People report experiencing red things. In the cases we are interested in, they specifically report seeing them. Put another way, I have no desire to dispute that people are acquainted with red objects and their qualities; rather, I think the issue centers on how they understand the relevant qualities, such as redness. As you note, Adam, there are many possibilities for this. Nonetheless, I think that if people think that the relevant qualities are mind-independent qualities of objects out-there-in-the-world, then they don’t think that those qualities are qualia (on the standard understanding of the term). I think that my first study suggests that this is how the folk, by and large, understand colors. Now, Dennett seems to hold that the folk believe in qualia and that they take a specific stance toward them, holding a secondary quality view that banishes the relevant qualities from the world and puts them in the head. As I think my study suggests against the weaker claim (any qualia view), I also think it suggests against Dennett’s stronger claim. Justin, Adam, and company–wonderful presentations and discussion! Great stuff, very well done. And props to Richard–great work in setting all this up! Justin– I wonder if you might get different results on the primary/secondary thing if you asked folk about errors/illusions/hallucinations. For example, a red car looks blue under certain lighting. Is the car blue? Is the blue in your mind? And so forth. My guess is you’d get results pushing in the other direction–it would be neat to find out. This raises the question of ambiguity or even inconsistency in the folk view. They may give one answer to one sort of case, and another in a different sort of case. (I suspect that this is what happens, for what it’s worth.) What do we do, as consciousness researchers, if the folk view is ambiguous or inconsistent? Do we clean it up as best we can and take that as our explanandum? Perhaps this is what various “arm chair” philosophers are up to? About the pain studies: Again, I wonder what the folk have to say about phantom pains. But there may be another confound here. Perhaps pains can only be *directly* accessed. In the conjoined and transplant cases, there seems to be a means of direct access. Perhaps it’s part of the definition of phenomenal quality that it can only be directly accessed (one thinks of Rorty and privacy here). So “where” the pain is might be less important than how it’s accessed. You might ask subjects about limbs hooked up by some kind of broad-band wire. Wonder if they’ll say the pain is shared? Anyway, great stuff all around! Josh– I should start by noting the your comments on the Experimental Philosophy blog were a big help in pushing me to get clear on how Dennett fit in as the central example for the talk (there is a footnote to that effect in the written draft of the paper)! The cases of non-veridical perception are very interesting here. I’ve started gathering some data on imagination using some probes I developed with Josh Knobe, but that is perhaps not the most central example for this type of case and much work remains to be done! My suspicion, however, is that disjunctivists basically have it right in their guesses about the common-sense view: In general, non-philosophers are naïve realists about ordinary perception, but treat cases of hallucination (or dreams, etc.) differently. My guess is that non-philosophers actually have basically the same model for cases like hallucination, treating them like perception but with the object being mental. That is, I suspect that many hold that when you hallucinate a ripe tomato, for example, you “see” a (non-real) hallucinatory tomato in your “mind’s eye”; just as the redness that you are aware of in the case of ordinary perception is thought to be a quality of the real tomato, the redness that you are aware of in the case of hallucination is thought to be a quality of the non-real tomato. If this is correct, then I would expect to see a reversal of the scores seen in the first study I reported if I asked them about a hallucination of a tomato instead of a real tomato, for example. Sounds like an interesting case to try! I’m not sure that this would necessarily indicate any inconsistency in the folk view, however. I suspect that the type of view outlined above is likely to seem a bit strange to philosophers (as it seems disjunctivism often does), but I find it to be a rather natural view when I take my philosopher hat off. I would expect something similar, as well, for pains. In fact, I’ve been planning on running something concerning phantom limbs and claims by Putnam, for example, that you cannot have a pain hallucination. My guess is that, by and large, the folk would happily distinguish between ordinary perception of a pain and a phantom pain, treating the former as a quality of a real object and the latter as a quality of a non-real object. I suppose that if an account like this turns out to be a good description of the folk view, then the motivating question that many of us have been talking about (Do the folk have the concept of phenomenal consciousness?) is probably not the most fruitful question to ask if understood as a yes/no. But, then, if we get to that point the question will have served us well. Thanks for directing me to the two other papers, Justin. A possible explanation a critic could offer would be that (for the folk) discrimination is generally sufficient for “seeing red”. The ambiguity (which for me is in “seeing”, rather than “red”) wouldn’t lead to bimodal responses, as all but the more cautious or pedantic (the majority you report) would be willing to say that the robot could “see red”. Philosophers are more likely to interpret the phrase in a technical sense that requires conscious experience, following conventions in the literature, so would apply different criteria. The folk could have a concept of phenomenal consciousness, yet have difficulty articulating the role it plays in guiding their responses. The failure (in your small sample of 3 responses) to indicate a phenomenal reading clearly, compared to philosophers, is not evidence that there is no phenomenal interpretation going on in those who deny that the robot can see (or potentially in others if asked about conscious experience rather than seeing). The mention of humans/animals in two out of three comments and the implication that computers and robots are different to humans in the third are at least consistent with a phenomenal reading. A minor point is that you say the details of the story don’t support the claim that the robot’s visual system is more complex than its touch system, but it is said to have “a video camera for eyes”, i.e. a video camera functioning as eyes would, presumably seeing things accurately and in detail, whereas “touch sensors” might be expected to register contact only to guide grasping – there is no reason to suppose that the sensors can register (rather than conduct) electrical current or sense pain unless we are told they can. We are told that Jimmy is a “relatively simple robot built at a state university”. I doubt the folk would expect a relatively simple robot to be capable of feeling pain or anger? If told a robot has a scent sensor, the folk may be willing to believe that it could detect the scent of a banana, depending on what the sensor is designed to detect, but this is arguably less plausible than using a video camera to “see”, if only because it’s a less familiar function, which could account for the difference in responses to seeing and smelling. A final point on hedonic value: it’s not obvious to me that colours don’t have hedonic value. People have colours they love and hate, just as they might love or hate the smell of bananas, or have no strong feelings either way. I think there is also an extraneous variable in the smell experiment using isoamyl acetate, in that it sounds as though it could be a simple chemical that the scent sensor is designed to detect, whereas the smell of vomit or banana might be expected to be made up of a more complex mixture of chemicals. Those smells would also be more variable, depending on what had been vomited, or the ripeness of the bananas, and a sensor might be expected to detect individual key chemicals that could discriminate between vomit and bananas, say, rather than detecting “vomit” or “bananas” themselves. If given the name of a chemical, people might also assume it smells unpleasant, so a lack of positive or negative valence can’t be taken for granted just because it’s unfamiliar. I’d be interested to see the results if the robot had to detect something neutral or unfamiliar that participants would expect to be as complex and variable a smell as vomit. I’m with you in thinking valence is well worth investigating, though and look forward to seeing what you come up with next. Bryony, lots to think about, here! For many considerations like these, I expect that the best way forward is to run further studies and at a couple points I have studies in the works; for some, though, I think the weight of evidence clearly leans in one direction. For the ambiguity objection that you raise to my first study with Edouard, in particular, I think that there is good reason (both historically and from considerations of the transparency of experience) to suspect that the folk are naïve realists about colors and this is supported by the first study I reported on in this talk. If that is correct, then it difficult for me to see how the informational/phenomenal distinction would get made by the folk. Thus, while your responses to our other two responses are each individually reasonable, I think that taken together the weight of evidence favors our interpretation. The point about complexity is well-taken. I think this relates most clearly to the objection that Adam raised: It might be that the folk have the concept of phenomenal consciousness, have a lower threshold than philosophers for thinking that an agent is phenomenally conscious, and think that seeing red is easier to implement than feeling pain. As you note, it is plausible that video cameras more readily suggest the relevant ability than touch sensors. Adam and I are currently putting a study together that would test this, leaving Jimmy basically the same but increasing the complexity of his grasping arms / touch sensors. Hopefully this will cast some light on the issue. (Let me note, briefly, that Jonathan Livengood and I replicated this study replacing language like “video camera for eyes” with just “video camera” and got a near identical result for Jimmy seeing.) That is a very interesting point about the complexity of the smells! I don’t think that this objection has been called to our attention before, but it seems like a plausible confound and would predict our pattern of results (on two quite reasonable assumptions: 1. that subjects will think that Timmy has a sophisticated enough smeller that they wouldn’t expect him to be less likely to be able to smell the complex olfactory stimuli compared to the simple olfactory stimuli; and, 2. that subjects are likely to think that Jimmy has a less sophisticated smeller and hence might be expected to be less likely to be able to smell the complex olfactory stimuli compared to the simple olfactory stimuli). That strikes me as a rather compelling point and one that I think could only be answered by running further studies, either to clarify this example or to show a similar point in another modality (Edouard and I have been putting together a similar experiment for hearing sounds that, as it turns out, controls for the confound that you note). Hi everyone! This has been an interesting discussion to follow. Justin, I am wondering if you have any controls built in to the study in order determine the level of competance with physics that your participants have? It seems to me that part of Dennett’s argument relies on his claim that it is folk aquainted with physical theory that hold the representation theory. Folk who are largely ignorant of physical theory might be naive realists. If so your study might be tracking folk theories of physics rather than folk theory of mind (at least in the color case). Richard, While what Dennett says suggests that he thinks the view is quite widespread (in talking about “laypeople” and the “folk theory of consciousness” for example), it might well be that what he means to say (or what he might want to say after reconsidering) is that he was thinking of a somewhat more sophisticated group. I imagine that the people he tends to interact with (including his students) tend to have well above average levels of training in philosophy, cognitive science, maybe physics and so on. I have no doubt that such training tends to shift people’s views on this issue… and I have some evidence to suggest that. I haven’t asked subjects specifically about their training in physics. I have asked about training in philosophy on each of the studies, however, and about training in psychology, neuroscience, and computer science on some of them. There is a very clear trend with regard to training in philosophy: Basically, the more training in philosophy a subject reports, the more likely they are to differ from the average response found for non-philosophers for the critical cases. For example, for the case of Jimmy seeing red, the following relationship held for the average responses for different levels of training in philosophy: no training in philosophy > 1-3 classes > 4-6 classes > 7+ classes > undergraduate degree in philosophy > graduate work > professor and/or Ph.D. If I remember correctly, I also found a similar trend for training in psychology, although it was fairly small, but didn’t see much difference conditioning on training in neuroscience or computer science (although I don’t think the sample sizes for either of these were very large on any of the studies). Asking about training in physics would be interesting, however; I’ll have to include that in some of the future studies. In the section that you quote from Dennett (slide 3) he says “Locke’s conception of the secondary qualities has become a part of the layperson’s interpretaion of the sciences” (this is a paraphrase). So it seems to me that if you really want to test Dennett’s claim you need to include something about physics. For what its worth, I suspect that you’ll find that even people who have had some training in physics will still hold the naive realist view and so that Dennett is wrong about this (though he may have been right about it for the previous generation). Interesting; looking at that passage again, it is somewhat odd. The first line reads: “Locke’s way of defining secondary qualities has become part of the standard layperson’s interpretation of science, and it has its virtues, but it also gives hostages” (1991, 371). I guess I find it odd that one would expect that the standard layperson (which I suppose here means basically non-scientist) had any interpretation of science. Maybe that is good reason to read Dennett as meaning to say something more like “the standard educated non-scientist with some training is the sciences.” That certainly would make the claim more plausible, although I suspect that it will depend on how much training they have and in what sciences. Just from anecdotal evidence, I expect that you are right that training in physics won’t make much of a difference. (This topic has come up a few times with graduate students doing philosophy of physics and, despite having undergraduate degrees in physics, they seemed to find it strange that one would hold the Lockean conception.) In my experience the view is most prevalent in certain branches of psychology and neuroscience (and is readily found in texts dealing with perception). If that is what Dennett means, however, then it would seem like there should be an added layer of complexity to his descriptions of the folk theory of consciousness and the data stages for heterophenomenology. One might reasonably hold that there are a couple of different folk theories—those with training in the sciences tending toward one view, those without such training tending toward another—and we would need to know which folk theory a given person held if we wanted to accurately interpret some of their beliefs: If the subject holds the Lockean folk theory, then we should (perhaps) interpret her statements about “seeing red,” for example, in terms of beliefs about the experiences themselves; but if she instead holds the naïve realist folk theory, then we probably shouldn’t. Yeah, Justin, I think something like that does end up being what he would be committed to amd that’s why I think you should include quesyions about physics in your study (in particular questions that get at whether they have a mostly Newtonian conception of physics). Interesting paper, odd experimental design. One would expect the simple question “Do you think that a ripe tomato would still be red even if no-one was around to see it” to be paired with the simple question “Do you think a pain would still be painful even if no-one was feeling it”, or something like that. Instead it is contrasted with a question preceded by a lengthy vignette clearly designed to favor the intuition that there is an unfelt pain in this case. Pretty clearly the vignette loads the scales here (one could equally have used a vignette tending to the opposite effect), so it’s hard to know what whether the data allegedly supporting the claim that subjects hold a naive view of pain is due to subject’s views or to the use of the vignette. Philosophically: it seems to be that experiments in this vicinity may shed light on whether subjects hold a certain sort of realist view about color and pain. But it’s a big step from there to shedding light on whether they hold certain common philosophical views of color experience and pain experience, such as qualia-based views. Of course many philosophers are both qualia theorists concerning color/pain experience and realists concerning (external) colors and pains. So I think it would be a good idea to keep these issues more cleanly separated. Hello Justin, First off, great presentation, I enjoyed it a lot and appreciate the time you spent creating it. Now to my concern. I am skeptical of the claim that one could tease out someone’s deep intuitions concerning the qualia/sense-data theory of phenomenal experience by simply asking them a series of questions that aren’t followed up by many other counter examples and arguments. In my experience, it is sometimes only through a Socratic dialogue that phenomenalist intuitions can be unearthed. In my opinion, how someone answers a questionnaire would not necessarily be indicative of their true beliefs concerning these philosophical intuitions considering that it would likely be possible to change the layperson’s minds given they were in a room with very argumentative philosophy professors. I mean, how do you know that a series of carefully crafted follow-up questions and counter-examples wouldn’t turn these naive realists into phenomenalists? In other words, I think that taking a laypersons gut instinct towards a few questions will not tell you how they really feel, given that what really counts is what side of the debate the layperson would agree with *after having actually participated in a heated philosophical debate* and not a impersonal verbal questionnaire. Looking forward to your response, Gary Williams Thanks for the recent comments. I apologize for being slow to respond; I’ve been out of town for a funeral (my grandfather passed away) and have had limited internet access. I promise full responses will be posted tomorrow, however! Richard, While I agree that this is a better position for him to be committed to, and I think it is where he should end up on reflection, that admission would seem sufficient to do the work I wanted out of the experimental studies (for purposes of this talk). This would seem to undercut Dennett’s discussion of the folk theory of consciousness; further, it would cast serious doubt on the quick move from people’s utterances to beliefs about conscious experiences. If we accept that a chunk of non-philosophers tend toward one type of view, another chunk another, then if we want to give the best interpretation of their utterances in this area we will have to do some additional work—we will have to first figure out what their “folk theory of consciousness” is. But, that is essentially the conclusion that I am arguing for. David, I appreciate your concern here. The one you suggest is an approach that I considered, but I found it rather difficult to get the pain questions to line up in parallel with the color study without sounding rather awkward. Note, for example, that the question you suggest does not seem to be quite parallel with the color case: “pain” corresponds with “ripe tomato” and “red” with “painfulness,” but it would seem that what we want is more like, “Do you think that the stubbed toe still has a pain even when no-one is feeling it?” That sounds rather awkward to me, but maybe that isn’t a problem (or, maybe the awkwardness should be seen as indicative of something about the common-sense view). My basic concern was that pain language seems to be something of a mess, with people sometimes using “pain” for the feeling of pain and sometimes for the object being felt and suggesting a perceptual model. Of course, it is often assumed that the common-sense view of pain does not follow a perceptual model (see for example Aydede’s introduction to the recent edited book on pain). My starting point, then, was to try to draw out the inclination to a perceptual view of pain and so devised a couple of cases where one would answer differently depending on your view. I hope I didn’t stack the deck too badly in devising these scenarios (although, as noted in a previous response, I will make some changes to the language on the vignette for the second study and repeat). If so, hopefully this will come to light with further work on the topic. In particular, I will take your comments as indicating that I might well have been too quick to go a different direction on the studies about pain and will take another look into doing pain cases that run in parallel with the color study! With regard to the philosophical point, I’m not sure I agree that the step is quite so big as you make it out to be. It depends, I guess, on exactly what you mean by “color experience”: if this is understood as “experience of color” and non-philosophers, by and large, hold that the color that the experience is of is a mind-independent property of the object that they are looking at, then it strikes me as misleading to refer to that quality as a qualia. Certainly, if we construe “qualia” broadly enough, then it would be silly to deny that the folk have a belief in qualia. But, in this paper I was specifically targeting the rather strong view that Dennett seems to suggest that the folk have, under which qualities like colors are banished from the external world and relocated as qualia in the mind/brain. Gary, Thank you—I’m glad you enjoyed the presentation! I have no doubt that you are correct here: Anyone who has taught an intro philosophy of mind course that looks at these topics will likely attest that discussion and argumentation can turn people into phenomenalists (or lead them to a qualia view). I am not sure that this is best described as those intuitions being “unearthed,” however, so much as instilled. But, how readily a qualia-view can be instilled in a naïve subject strikes me as a rather different question from what those subjects’ naïve view was. My interest here is with the latter. This is not because I want to say that the naïve subjects are right, or because I want to put difficult philosophical questions like this to a popular vote, but because claims are sometimes made about the views of naïve subjects (that is the focus of this talk) and because it is sometimes said that phenomenal consciousness or qualia are just obvious or phenomenologically obvious (which I didn’t talk about here). With regard to the latter, I am perfectly happy to accept your point that many naïve subjects can be brought around to a qualia view with some philosophical argumentation. I take that to mean, however, that qualia weren’t just obvious to them, but that they found the arguments for qualia compelling. This should then direct our attention to those arguments (and away from the claims of obviousness). On the first point, I appreciate the difficulty in raising parallel question, but it does seem to me that the vignettes are functioning as a sort of mini-argument for the existence of unfelt pains and shared pains. And I don’t think the fact that subjects presented with an argument for P agree with P is strong evidence that they pretheoretically hold P. So playing it straight with relatively parallel questions might well be a good idea. On the second point, again, many philosophers hold that there are both qualia (had by experiences) and colors (had by external objects). On this view, the color that the experience is of is a property of the object, but the quality that the experience itself possesses is a quale. So the discovery that ordinary subjects think that colors are mind-independent properties of objects is no evidence that they don’t believe in qualia. As far as I can tell the evidence says very little about their view of qualia: it just tells us that they aren’t anti-realists about color. Of course this ipso facto tells us that they aren’t anti-realists about color who relocate color as qualia (as you say at the end) — but it’s the view about color that’s doing all the work. It’s an interesting result all the same, but I think it’s a bit misleading to label it as a result about the folk’s view of phenomenal consciousness. That is interesting: In constructing the vignettes for the unfelt pain and shared pain vignettes, I was attempting to set up scenarios where the relevant questions could be asked (as most, if not all, cases of pains are felt and unshared). But, from another perspective, in that it seems at least plausible to say that there was an unfelt pain or a shared pain, these scenarios could also be seen as counterexamples or mini-arguments. I’m convinced — not that this undermines the results, but that it is good reason to follow-up with a more straightforward approach. With regard to the second point, I take it that one could read the *of* in a relational way in which there is only one quality at issue (the quality that figures in the experience being the quality had by the external object), or such that there is both a quale and the quality had by the external object. The objection, if I understand correctly, is then that subjects might hold (something like) the latter view and be answering the questions with regard to the quality had by the external object. (Note: This is a similar objection to the one raised above against my first study with Edouard discussed in the talk.) This certainly is possible; perhaps one might even argue that it has some plausibility for the first three questions. For the spectrum inversion question, however, it would seem rather strange to both hold a view like the one you suggest and to answer that question with regard to the quality had by the external object. Still possible, but I don’t find it very plausible. Nonetheless, your point is well taken. it is certainly very difficult to pin-down people’s views on these issues, especially if you want to be careful not to alter them in the process. I would be very interested in your thoughts on how best to go about it!! Comments are closed.