Symposium on Louise Richardson’s “Flavour, Taste and Smell” (Mind & Language 28 (3), 322-341)

I’m very glad to be able to kick off the first of a series of symposia that Brains will be hosting on papers from Mind & Language. This month our target article is “Flavour, Taste and Smell”, by Louise Richardson of York University in the UK.

Louise’s article, which thanks to the kind cooperation of Wiley-Blackwell is available freely at the preceding link, discusses the relationship between scientific investigation of the senses and our everyday way of individuating the sense modalities. Specifically, she argues that the commonsense idea that flavors are tasted rather than smelled is not refuted by scientific findings showing that olfactory receptors play a significant role in our perception of flavor.

Similar to the format made familiar by the Online Consciousness Conference, our symposium will begin with a short introductory video from Louise, followed by written commentaries from Fiona Macpherson, Mohan Matthen, Matthew Nudds, and Barry C. Smith. Louise will have up a response to the commentaries within the next day or so, but the discussion may begin in earnest in the meantime.

***

Louise’s introductory video:

Her paper: “Flavour, Taste and Smell”

Commentaries:

Added June 7: Louise’s response to Fiona, Mohan, and Matt

This post will be open for comments through the end of June. Thanks in advance to all those who participate!

06. June 2013 by John Schwenkler
Categories: Mind & Language Symposia, naturalism, perception | 39 comments

Comments (39)

  1. Hello to all! I’m looking forward to hearing more from commentators and other participants. I’ll be posting my initial responses to these first three commentaries very soon.

  2. Hi all,

    Thanks very much for the excellent discussion so far.

    I wonder if we can’t make some progress in understanding Louise’s position by distinguishing between several different ways in which science might be thought to correct (or confirm) our sensory judgments.

    First, the *generality* of the proposed scientific correction seems very significant: it’s one thing to say that a judgment categorizing some *particular* sensory episode — say, that I have just tasted something lemony rather than smelling it — could be falsified on scientific grounds; but quite another to say that we could do this about a whole *class* of sensory judgments, which is the kind of thing that scientists seem to be doing in e.g. the passages that Louise near the start of her paper. Certainly Louise’s position doesn’t require ruling out the first sort of correction: to pick up on one of Mohan’s examples, denying that this is possible would be like denying that it’s possible for me to discover “scientifically” that the sky isn’t really blue right now, but that it looks blue to me because my eyeglasses have blue lenses. Instead, her target is the more ambitious idea that science could reveal us to be *massively* mistaken in our ways of categorizing sensory episodes, much as it might be thought to show us that the sky isn’t *ever* blue, and things only look that way because of how light is scattered as it moves through the atmosphere. (A difficulty, though, is that there’s a whole range of possibilities in between being massively mistaken and just being mistaken in some particular case. I imagine Louise would say: the more massive the supposed mistake, the more difficult it will be to show that it is part of our everyday conception of the senses. That is, it seems more plausible that we could discover scientifically that smell always plays some very important role in our perception of the flavor of *lemon*, than that we could discover that it always plays such a role in flavor perception quite generally.)

    Second, and just as importantly, we need also to attend to how the proposed scientific correction is supposed to proceed. On one model, what happens is something like the following: scientists classify sensory episodes (and/or capacities) in a deeply explanatory way; this classification doesn’t match our everyday conception of the senses; and as a consequence, we ought to conclude that this conception mistaken. And clearly Louise is unhappy with this kind of picture: she is arguing that science can’t refute the everyday conception of the senses in the same way as, say, we learned that we were wrong to think that the earth is stationary, or that the sky is a dome separating the earth from the waters. But as she says in her reply, that doesn’t commit her to saying that our everyday conception of the senses is *altogether* static, and I think her position can allow that experimental findings could be *part* of the basis on which that conception changes over time. However, in such a process of change that everyday conception would remain largely intact: we wouldn’t *stop* being attentive to whatever “surface features” of sensory episodes underlie our sensory judgments at present, but would only be a bit more careful in how we made use of them; and any changes in our everyday conception would likely be informed by a different set of practical interests than those that drive scientific classifications. A possible analogy here would be with the discovery that there are mountains under the sea: perhaps people used to use “mountain” only to refer to elevations on the land and used a different term to categorize undersea land masses, but upon discovering that these masses were really no different from mountains except in where they were located, they altered their conception of mountains to classify them all together. Yet this wouldn’t mean that scientific findings had *refuted* that old conception (at least in my telling, it’s not as if people thought that the ocean’s surface was perfectly flat), but only that it had contributed some data that led that conception to shift.

    Okay, I’ve gone on long enough. Is any of this on the right track?

    • Thanks, John. I think you’re right that it’s useful to think about different ways in which science might be thought to correct or confirm our sensory judgements. The commentaries so far have helped me in doing so!

      I’ll pick up, for now, on just one of the distinctions that you suggest we should make, namely in how any supposed scientific correction is supposed to proceed. I think one useful distinction is between

      (A) The way in which if naturalism is true, the truth or falsity of sensory judgements can be just read off a certain sort of data, namely, data about similarities and differences in how perceivings are produced. That’s what I had in mind by talking of science ‘telling’ us the modality to which a perceiving belongs, or ‘overturning’ sensory judgements.

      And

      (B) The way in which if non naturalism is true, ECS can be shifted about, in the light of scientific findings, in ways that might lead to changes in sensory judgements. As you put it, John, non-naturalism about ECS does not commit one to saying that ECS is ‘altogether static’. But the shifting about that occurs will depend upon (i) the particular variety of non-naturalism in question and (ii) answers to other contested questions. (And, of course, neither (i) nor (ii) can be ‘read off’ the sort of data mentioned in (A) either). As John puts it, the experimental findings are only part of the basis of how the conception changes, if non-naturalism is correct.

      I hope Mohan doesn’t mind if I borrow something he said in a useful (to me!) email exchange we had in advance of the symposium. He raised the possibility that “science does not tell us [how to determine the modality to which a perceiving belongs] but may give us material that is relevant to us making decisions about this…’. I think Fiona was saying something similar on page 8 of her comments. Perhaps I can say: the sort of shifting about that can occur if non-naturalism is true is too dependent on other factors to count as science simply telling us that our sensory judgements are false, or overturning them.

      There are lots of further distinctions to be made, within (B).

      • Thanks, Louise, this is very helpful indeed. The distinction you draw is just the one I was after in my second long paragraph. But now what if a naturalist were to respond that you’ve caricatured her position: that science can overturn our sensory judgments only in the way that it can overturn judgments of any other sort, i.e. against the background of a host of other considerations that can be upheld and revised in a range of different ways in order to fit the phenomena? (I imagine this is pretty vague and needs a lot of explication, which I’m happy to provide when I’ve got more time.)

  3. A couple of clarifications in response to John Schwenkler.

    My point about the look of the sky was simply that discovering that there is no blue hemisphere above us does not clearly refute the LOOK of the sky, because it is not clear that things look that way. The connection to Louise’s paper is that, like her, I think we have to be cautious in reporting how things look (or seem). It’s possible to overstate what perceptual experience tells us, and thus possible to exaggerate the corrective role of science.

    As for the possibility of massive error, however, my position is that we can indeed be led to modify our everyday conception of the senses, and that scientific discoveries can be instrumental in getting us to do so. So for instance, reading about the McGurk effect can lead us visually to attend to a speaker’s mouth movements. This change in our behaviour can lead us to conclude that vision has a hitherto unnoticed role in how we perceive speech. (I think this is compatible with everything Fiona says, though it introduces some new considerations.)

    Finally, I want to agree with Louise that there is a good bit of commonality between her approach and mine.

  4. Hi, John. Thanks very much for your follow-up.

    Two quick thoughts. I hope not too quick!

    First, perhaps (A) is a caricature in that if it’s true, no-one is a naturalist, in this sense. I’m not sure what that would do to my overall position. It wouldn’t obviously undermine the main point I want to make.

    But second, maybe I can say a bit more, and thus accommodate the naturalist response you mention. I’d be pleased to hear what you think! I said in my previous comment that on a ‘naturalist’ view, one can ‘read off ‘answers about how to categorise perceivings from a certain sort of data. ‘Read off from the data’ is the problem, right? That it can’t be as simple as that is made clear by, for example, the fact that there’s some dispute (within science) about whether to interpret this sort of data (about similarities and differences in how perceivings are produced, etc.) using the notion of a sensory system dedicated to perceiving flavour, or the sensory systems dedicated to taste and to smell. And, of course, there’s the more general point that you might have had in mind (and see Fiona’s commentary) that one can never simply ‘read off’ any conclusion from ‘the phenomena’: it might be true that you can always, in principle, choose to revise some other element in your web of beliefs, perhaps guided by Quine’s advice, as Fiona suggests. In this way, decisions will be made within science about which similarities and differences in how perceivings are produced are the ones that make the classificatory difference. So, the data always comes to us, from science, interpreted.

    According to naturalism, once these scientific decisions have been made about how to interpret the data, we know what sensory judgements we should make. We can ‘read off’ the answers to questions about how to categorise perceivings from the *interpreted* data.

    The point about the way in which science can play a role in changing sensory judgements on a naturalist view then, is not so much that only non-naturalism has a role for background beliefs in changing one’s views in the face of new information. Rather, the route from the (interpreted!) data and a change in sensory judgements will itself be much more indirect, and subject to the influence of further considerations, if non-naturalism is true.

    An interesting aspect of Mohan’s suggestion is that it shows, also, that there are ways in which ‘science’ can play a role in changing sensory judgements, if non-naturalism is true, that is not a matter of accommodating the (interpreted) data but changing one’s practice in some way. I’m also intrigued by the possibility, which I see in reading Matt’s comments, that scientific findings might impact upon ECS in quite significant ways, yet without changing sensory judgements.

    What do you think? Is that at all helpful?

  5. Hi Louise,

    I was wondering how your account of ECS would handle new technologies like mind-brain interfacing, particularly “sensory substitution” devices that use the tongue as an electrical interface to the brain. This blog article gives an overview:

    http://www.psychologytoday.com/blog/the-fallible-mind/201107/see-your-tongue

    In a nutshell, it seems to me like the ECS would never have “predicted” that it would be possible to “see” through the tongue. And it’s not just playing with modality switching. This technology allows us to see into spectrums not previously available to humans such as infrared and ultraviolet. Again, my take is that ECS would not have predicted that this would be possible. Thus, it seems like either ECS is wrong, or must be modified significantly in light of these technological discoveries. My view is that such technologies will force us to “co-evolve” our everyday notions of perception as being more plastic than otherwise conceived (assuming that the ECS is more “rigid”).

    I apologize if one of the commentators has already asked this question; I am traveling at the moment and didn’t have time to read all the commentary.

    Thanks,

    Gary

    • Hi Gary. Thanks very much for your question! And no, nobody else has asked it, yet. Here are some thoughts:

      I’m not convinced that it’s right to say that the device you mention allows you to see with the tongue. In fact, Fiona ran a really interesting conference on sensory substitution and augmentation earlier this year and it seemed to me that there was very little agreement about whether the experiences one has when using such a device are experiences in the substituted modality. And, of course, whether or not one agrees that the experiences one has with, say the BrainPort device are visual perceptual experiences is going to be determined, in part, by the view one has of what makes an experience visual. If a non-naturalism about ECS is true that says something like ‘an experience visual if it involves surface feature x,’, and the experiences one has when using the device have feature x, then those experiences, according to ECS, will be visual. I’m not sure if any such non-naturalistic view is true, but I think I’ve described a possibility.

      However, whatever the best thing is to say about the sorts of experiences users of SSDs have, either now or in the future, I agree with you that our everyday conception of the senses would never have allowed us to *predict* that those sorts of perceivings were possible. But what does that tell us?

      One thing it doesn’t seem to tell us that any of the sensory judgements that are underpinned by ECS are false, or that ECS has, up to then, included false empirical commitments. I take it that successful sensory substitution, in an obvious respect, changes the world. It brings about new perceivings, in a new way. So it strikes me that the occurrence of successful sensory substitution is not going to correct any such sensory judgements in a straightforward way, but will rather give us a new bunch of perceivings to categorise. I agree with you that if the world changes in this way, ECS is likely to evolve, too, to accommodate the change. That doesn’t seem to me to impact greatly on whether we should think that naturalism or non-naturalism is true of ECS.

      Does that go any way towards answering your question?

  6. >Does that go any way towards answering your question?

    Yes it does, thank you!

    A few words, though.

    >I’m not convinced that it’s right to say that the device you mention allows you to see with the tongue.

    I should have been more careful in my writing because I am quite aware of these phenomenological/introspective controversies, and this is why I used quotations around the first usage of “see” to illustrate that I was using the term more metaphorically, as in “I see what you mean” (meanings are clearly not ‘visible’ in any obvious sense). It strikes me that in light of Lakoff & Johnson as well as Hofstadter’s newest book, it would be quite interesting to see (!) how metaphor infects ECS. Visual metaphors (seeing = knowing) are ubiquitous and deeply embedded in our minds ,so I wonder how the asymmetries in what modalities are analogically dominant influences concept-change in ECS e.g. visual metaphors are more influential for describing basic knowledge than, say, smelling.

    Furthermore, your response sparked an idea about the empirical commitments of everyday persons. It seems that one motivation for denying that everyday persons are “wrong” about how their minds work is that they don’t know any better and couldn’t possibly know any better (particularly if we are examining pre-scientific peoples). But if we assume naturalism is true, then perhaps we could capture the above motivation by distinguishing between accountability and blame. That is, we could coherently say that laypersons are “wrong” without thereby *blaming* them for being wrong. In other words, we could acknowledge laypersons are “wrong” but just strip away the negative connotations of being wrong such as that they should know better, or should be ‘punished’ by a forced conceptual restructuring, etc. Just a thought.

    • Thanks for these interesting thoughts, Gary! Just one point on your last paragraph.

      I think you’re right that it’s desirable that any view of ECS and the senses shouldn’t blame people, in everyday thought and talk about the senses, for getting things wrong. And I think your thought about how a naturalist might accommodate that thought is interesting.

      However, I don’t think that, according to non-naturalism, in our everyday thought and talk about the senses (or in the thought and talk of pre-scientific people) we/they ‘don’t know any better and couldn’t possibly know any better’. It’s rather that on a non-naturalistic view of ECS, as I understand it, that which underlies our sensory judgements is likely to be non-committal, or neutral about lots of things: it’s just not in the business of, say, predicting the possibility of certain sorts of sensory substitution devices, any more than (as might be argued) folk psychology, more generally, is in the business of explaining why we dream, or the occurrence of mental illness. It doesn’t, as Mohan puts it in his commentary, ‘take a stand’ on such things. As such, it’s consistent with our being told all about these sorts of things by scientific psychology, or neuroscience (for instance).

      It’s no commitment of non-naturalism (as I understand it) that ‘folk’ or ‘prescientific people’ or any of us can’t and won’t (and don’t) learn a lot about perception from psychology and neuroscience. But it’s not clear to me that this is best understood as learning that impacts upon whatever underlies our sensory judgements, still less that it will force us to change those judgements (though I have allowed, in the paper, but more so in response to commentaries and above that in certain circumstances science might contribute to some such change!).

  7. Hello Louise,

    Thanks for doing this discussion. My question is really just a plea for clarification. I didn’t understand exactly what was being argued for, and maybe this stemmed from an initial confusion on my part about the meaning of the ‘puzzle about the sweets’, and what sort of answers we could expect to it.

    The puzzle is:

    “what is it that’s missing when you hold your nose when trying the sweets?” or “what does holding your nose when eating sweets do?”

    And 2 potential answers are:

    1. What’s missing when you hold your nose is one aspect or part of tasting. Holding your nose impairs your ability to taste the flavours of the sweets

    2. What’s missing is something olfactory—an olfactory experience, or an olfactory component of the multimodal experience of flavour. Holding your nose, accordingly, prevents the sense of smell from playing its usual role in flavour perception.

    So my simplistic response to this set up is that the puzzle seems ambiguous. Do you want an explanation of the physical mechanism that explains why holding your nose impairs your experience of the flavour of the sweets. Or do you want an exploration of our concept of, or what we mean by ‘tasting’, ‘smelling’ or the ‘sense of taste’ or the ‘sense of smell’?

    Lets say that the findings in the psychology of flavour perception relate to the first sort of explanation (although some would say that the we can find out about our concepts through empirical methods too). Then when you say that you want to argue that: “findings in the psychology of flavour perception do not settle [the puzzle]. Which is to say: they do not show that it is wrong to think that flavours are just tasted”, it feels like a debate is being set up on the basis of an ambiguity in the initial puzzle, and the sorts of answers we might be interested in.

    I have a feeling that my simplistic response misses something in particular in relation to the notion of a ‘sensory judgement’. I wonder if initially you could say a bit more about what you intended ‘the puzzle’ to be about?

    Thanks!

    • Hello Chris! I think the ‘simplistic’ response you mention is an appropriate one, so thanks for giving me the chance to clarify.

      Findings in the psychology of flavour perception (considered broadly, to include a certain amount if interpretation of the basic data) *do* settle the question of which physical mechanism(s) are prevented from doing their usual job in flavour perception, when you hold your nose. (Although, as I understand it, there is disagreement about how to interpret the data.)

      What interests me, though, is that it’s sometimes said that such findings show that people in general are wrong to think that flavours are just tasted, or that the experience of flavour is not an olfactory experience (or whatever).

      I think that something like the ambiguity you mention is present in these claims, too: perhaps people mean, when they say these things, that only the physical mechanism specific to taste is involved in flavour perception. But perhaps they don’t, and perhaps the truth of what they say is unaffected by which mechanisms are involved. If anything like the latter is true then any move from the findings to claims that people are wrong to think that flavours are tasted (and so on) is going to be much less easy to make.

      In a sense (!), then, you’re right that ‘a debate is being set up on the basis of an ambiguity in the initial puzzle’. Or at least, I hoped to have made it seem that this was, indeed, a puzzle not least because it’s not clear what sort of thing people are talking and thinking about, when they talk about the sense or modality to which perceivings of flavour belong.

      I wonder, though, if behind your question might be the thought that the best response to the puzzle about the sweets is just to say that some or all of the terms ‘flavour’ or ‘sense’ or ‘modality’ are ambiguous. Is that what you had in mind?

      • Hi Louise, thanks for your reply.

        So I read the quotes you use at the beginning of the paper from Rozin and from Auvrey and Spence as saying, roughly: People have some notion of what’s involved in the taste sensations they experience, or if you like, what their sense of taste consists in – they basically think it’s to do with stuff that goes on in the mouth, and, moreover, don’t intuitively think of what they associate with the sense of smell – i.e. the nose(!) having much if any role, or that they underestimate its role or whatever. I take it that you aren’t arguing against this.

        But rather you have something more specific in mind by ‘sensory judgements’? You seem to be saying: yes people might have all these mistaken beliefs about what’s involved in tasting, but their sensory judgements are, in some other respect not mistaken. So part of what I would be interested to hear is a more precise characterisation from you of what these judgements are, and how they are to be distinguished from the the beliefs we have (which also seem to use our sensory concepts) that on the face of it the psychology findings show to be mistaken in some respects.

        If the above is right then it seems to me that setting up the argument you want to make as: ‘Scientists claim our judgments about taste are mistaken, but they are (or might be) wrong’, is unhelpfully strong. And maybe the argument is more like: Scientists aren’t being subtle enough about what they say we the ‘folk’ are mistaken about.

        Hope this isn’t way off target, and it would be great to hear more from you!

        • Thanks for the follow-up, Chris. I think I see what you have in mind, but do say if I’m overlooking something.

          Firstly, reading your comments made me think that I have something *less* specific in mind than you might imagine! Sensory judgments are just judgments about the modality to which some perceiving belongs, or to which some perceivings belong. Judgments like ‘flavours are tasted’ or ‘I am seeing my cup of coffee’. I don’t give any precise characterization of such judgments because, in a way, what I want to say is that it’s not clear how one should do so. The point I hoped to make was that when people make these judgments, it’s not clear whether they are committing themselves to anything, with regards to the mechanisms and processes that underlie their perceivings. And it’s only if they do have such commitments, that there’s a straightforward route from the sorts of findings that Rozin and Auvray and Spence discuss, to the claim that sensory judgments are false.

          >You seem to be saying: yes people might have all these mistaken beliefs about what’s involved in tasting, but their sensory judgments are, in some other respect not mistaken.

          Not quite. What I want to say is: people might have mistaken beliefs about what’s involved in tasting, but without a defence of a particular view of our everyday conception of the senses, it’s not clear whether this entails that their sensory judgments are false.

          > …it seems to me that setting up the argument you want to make as: ‘Scientists claim our judgments about taste are mistaken, but they are (or might be) wrong’, is unhelpfully strong. And maybe the argument is more like: Scientists aren’t being subtle enough about what they say we the ‘folk’ are mistaken about.

          One clarification: I’m not especially keen to be read as saying that scientists are right or wrong about anything, here. What I’m really concerned with is whether one should take claims like ‘people are in constant error about the modality to which perceivings of flavour belong’ at face value, that is, as entailing that some of our sensory judgments are false, which I think some philosophers think that they do. And what I hoped to argue was: it’s not clear whether we should, because it’s not clear what we commit ourselves to in making sensory judgments.

          I do think your comment about how to read the quotes at the beginning of the paper is helpful. If you’re right that the authors I quote have in mind merely that we folk get things wrong about the mechanisms of flavour perception, then it’s quite apparent that this is only going to entail that sensory judgments are false, given some assumptions about the role of thinking about such mechanisms in everyday thought and talk about the senses.

  8. Wow–what a fruitful discussion!

    Thank you Louise for such a stimulating paper.

    I’m not sure I fully get the non-naturalist picture here (I guess this is clarificatory). What is a “surface feature”?

    My thinking was that it’s like the painfulness or “hurtiness” of pain. And one might argue that if something doesn’t hurt, it’s just not pain. And that’s something no science could undermine. “Hurtiness” is a surface feature–one with no “hidden essence”–and it is what makes something a pain. And all this is folk-psychological–we know this in virtue of grasping the folk “theory” of pain.

    Does this capture the idea of a surface feature, and does this provide a decent parallel for your case about taste and smell? Or is there something else at play here?

    Thanks!

    • Hi Josh. Nice to sort-of meet you! Thanks for your question.

      I wonder if one reason why the non-naturalist picture is difficult to see is that it encompasses quite a lot of different, possible, positive views? Non-naturalism, as such, is just the *negative* view that the everyday conception of the senses isn’t committed to the distinction between the senses being a matter of differences in the way that perceivings are brought about, of a sort that await discovery by science.

      My idea, in talking about ‘surface features’ was that there is a family of positive views about ECS, that count as non-naturalistic in the negative sense, that say something like this:

      “What makes the difference between the senses, according to ECS, *is* just that feature in terms of which we (by which I mean, for want of a better word the ‘folk’) pick out the various senses. That is, some relatively superficial (or ‘surface’) feature.”

      The contrast here is with a view that claims that the feature in terms of which the folk distinguish the various senses is one thing, and the feature in terms of which they are *really* distinguished is quite another, and is a ‘deep’ feature in that it’s ‘to be discovered’.

      Your example of the hurtiness of pain is apposite because one member of this family of views says that the relevant surface/superficial feature is an aspect of phenomenal character. But this is not the only member of the family. Other potential surface or superficial features will be ones that are accessible to us, but need not be ones about which we are authoritative, or which are entirely transparent to us.

      I’m not sure whether this sort of non-naturalist view has to go along with the suggestion that we (the folk) know how the senses are distinguished in virtue of grasping a theory. What we grasp may be too loose and partial an affair to count as a theory, as Matt suggests in his commentary.

      I hope that helps a bit!

      • Thanks, Louise! And nice to sort-of-meet you as well!

        That does make it clearer to me.

        Here’s another question, again, perhaps helping me locate this debate, but also raising a more substantive worry.

        David Lewis’s view of folk psychology, sometimes called “analytic” functionalism, holds that mental terms are defined by the the platitudes used by the folk (I take it you know all this, but I’ll lay it out to make the point). We find out what we and others know about certain terms, what we all know that others know, and so on. This gives us the folk-psychological meaning of our mental state terms. Now, Lewis thought that the analytic platitudes one would get would involve commonsense functional roles. But even if we don’t think one gets a commonsense functionalism, is this platitudinous method non-naturalist, in your sense? That is, in analytic functionalism non-naturalist?

        It seems to me that it is non-naturalist, because if science discovers sensory processes that do not fit the platitudinous definitions, it is not talking about the senses. It may, however, find those neural processes that realize the platitudinous definitions, thus showing how the senses, as folk-psychologically characterized, are just certain neural processes. But science can’t show that we’re wrong about the senses.

        However, science could, on this view, show that nothing in the brain answers to our folk-psychological definitions. Then we might learn that there are no senses! But science may find processes that aren’t too far from the senses: they take in and transduce certain kinds of information, they involve certain specialized organs, etc. Call these processes “schmenses”.

        So perhaps non-naturalism, construed in the Lewis-y way, is right. Then, science can’t tell us that we’re wrong about the senses. But science can tell us about a better idea: the schmenses. So it can change our beliefs about how we detect the world, but not about our use of the term “senses,” “smelling,” “tasting,” etc. But who cares about these terms if we’ve got a better way to talk, one using the scientifically grounded schmenses language. So even if one wins the language battle, one loses the ontological war. Hence, non-naturalism can’t protect the senses in the way you suggest.

        Does this make sense? Sorry if I’m reproducing something already covered. I must confess a rather quick read of the full debate!

        • Thanks, Josh. Yes: I think what you say does locate the debate very well and is a response that will look very appealing to some! That is, people might say, ‘sure, maybe non-naturalism is correct and the folk categorise perceivings in a way that’s insensitive (or perhaps, sensitive only in indirect and messy ways) to scientific data. But so much the worse for the folk’s current way of categorizing perceivings, because the science gives us a better way of categorizing.’ I take it that this is a kind of eliminativism about senses.

          In terms of the current paper, if this sort of eliminativism is the right response to non-naturalism then my overall claim is unaffected: the science won’t be correcting or confirming sensory judgments so much as changing the subject. But more broadly speaking, as you imply, if I want to defend some sort of non-naturalism then I’d probably better have some response to the eliminativist option! Matt (Nudds) has argued that central to the project of distinguishing the senses should be some view of what the distinction is *for*, i.e., why we make it, perhaps what interests our making it serves. I suppose that some such view is going to be the best way of responding to the eliminativist’s suggestion that we’d be better off with schmenses anyway.

          • Great! Thanks so much, Louise–You’ve done a great job presenting and then holding up your end of a number of blog conversations. That’s a lot of work!

            Cheers,

            Josh

  9. Hi Louise,

    I’ve just read your paper. It’s got me thinking! I’m sorry if I repeat anything that anyone else has already said; I haven’t had time to read all the commentaries yet.

    I think that the arguments you provide in sections 4 and 5 carry a lot of force. What I’m not quite clear about is what precisely is involved in endorsing what you call naturalism about the senses, or perhaps rather why one should accept what you seem to say is required for naturalism. Given that you refer to a couple of things I say in section 1.1 you would seem to be suggesting that I endorse naturalism about the senses. However, although I still accept those remarks you quoted, I am not so sure that I am committed to other claims to which you seem to think a naturalist should be committed. The main one – and it seems to figure centrally in your paper – is that the naturalist can be contrasted with the non-naturalist in so far as the former takes the view that we are ordinarily committed to certain claims about the senses and that science might show that we are wrong to be so committed.

    To begin with I do think that we are committed to some beliefs about the senses that science cannot overturn. The examples I am thinking of involve modal claims. Consider the examples in your first paragraph: I see the keys but I could smell them (if I get close enough); I hear the bus but I could see it if I opened the curtains. However, I don’t think that we can or could hear colours whatever people may think that science tells us or might tell us. I think that it is not a metaphysical possibility. Indeed, I think such suggestions are conceptually confused. If some scientist or otherwise were to cite synaesthesia, it is the job of the philosopher to provide some conceptual clarification (although for various reasons I don’t think that what has been said about ‘gold’, ‘water’ and tiger’ should be pressed too far here). What this indicates is that we do have some commitments, conceptual truths if you like, regarding the way that we think of the senses.

    However, these cases are rather different from the one that you are interested in. It is true that I have some thoughts about how we should think of taste and smell, but I would not say that I am committed to them in the way that I am committed to the impossibility of hearing colours. Indeed, I think that because of the kind of not unusual examples you cite, I should not commit myself to too much without thinking further (as I said at the end of the paper of mine you quoted: I think that it is clearer what we should say about snakes tasting volatile molecules than in our own case.)

    Minimally, I think that we should consult the natural sciences in order to inform our thinking about the senses, and especially certain cases. If I am a naturalist about the senses then that is what I should be committed to minimally. If you think that scientific data is able to inform debate in some way about the senses, and even to persuade one of a certain view about a particular sense, and, in as much, about the senses more generally, then one might fairly be called a naturalist. (So, for instance, I think that what scientists have told us about thermal imaging – that it detects infra-red radiation – has helped us to come to a better view of what kind of sense it is (regardless of what we might have previously thought).) But I do not think that science will be able to give us the final answer in all cases, and for reasons you give, maybe not in the flavour case. I would have thought a non-naturalist is better regarded as someone who endorses the opposite claim that there is no point in looking to the natural sciences at all because they cannot help us in any way. It may be that I am not drawing the distinction as you do. But then I ask why we should burden the naturalist with anything stronger than this commitment.

    This does not mean that science might not or cannot help us to change the beliefs we do hold. As it happens, I think that it might and, indeed, does.

    Regarding the former, suppose we are inclined to the view that flavour perception is cross-modal or bi-modal because of the everyday examples you give, and that science counterfactually had discovered that the receptors in the nose that contributed to flavour perception were distinct from those that were responsible for the detection of external smells. That would be interesting. It seems to me that it might give people more reason to think that these receptors were all part of a unitary gustatatory system, and hence that we taste flavours, than our having actually discovered, as you say, that the same receptors as are used to detect external odours are responsible. Of course, this does not rebut your argument that the scientific evidence we have does not tell us whether we taste flavours or partly smell them. My point here is that the evidence that science might have provided could have had some implications for what we should think about perceiving flavours. This seems to me to be naturalistic.

    Regarding the latter, and I think that we might just disagree on this, I do think that there is a good case of where science has provided information that helps to change our minds about how we think or should think about the senses. (At least it does if we tend to think that heat sensations are part of touch.) Well it has changed my mind! I think that data from psychophysics and neuroscience about how our thermal receptors work and what they detect give us good reason to think that we have a heat detecting sense that is quite different from a tactile sense, indeed as different from touch as any of the other senses is different from touch. I even dare to suggest that if people knew more about how it works they would agree. So I suppose that would make me a naturalist about the senses in your sense, although, as I say, I am not convinced that anyone would have to follow me in this to endorse a view that is reasonably thought of as naturalism about the senses.

    Part of the disagreement I suspect is that you sometimes suggest (e.g. p.326) that naturalism should hold that science can tell us what the conditions are that we should use to individuate the senses. Someone might argue that. But I think one might be considered a naturalist if one just accepts that scientific evidence can help us to get clearer about what kind of sense something is given that we do individuate the senses by means of the conditions that we do use, and even help to change our mind about the kind of sense we thought something was.

    • Hello Richard. I’m really pleased to see you here! Thank you very much for your comments.

      First, I think I share your intuition that hearing colour is an impossibility of some sort, although I suspect this might be something of a grammatical point: as Geoffrey Warnock puts it, ‘’sound’ is the tautologous accusative of the verb ‘to hear’’. If that’s right, one could never rightly say that one heard colour, for instance, or odour. And science couldn’t overturn that. (Or at least, not in any direct way. Perhaps one can imagine science having a role in some gradual change in linguistic usage, here.)

      Second, and more importantly, naturalism and non-naturalism, and varieties thereof. One small clarification: what I called naturalism is a view not about the senses, per se, but about our everyday conception of the senses (or ECS, to use Mohan’s helpful abbreviation). ECS is just whatever it is that underlies our judgments about the modality to which a perceiving belongs. And, as you say, thus understood, the naturalist thinks that ECS involves commitments that can be straightforwardly overturned by scientific findings about, for instance, which sensory systems are involved in a perceiving. (Of course, there are questions about how to distinguish such systems, too: perhaps the relevant differences are just a matter of receptor types, or perhaps they are functionally individuated, as you have previously, and helpfully, suggested.)

      As I’m using the terms, the non-naturalist doesn’t have to deny that, as you put it, ‘we should consult the natural sciences in order to inform our thinking about the senses’. For one, it is open to a non-naturalist to accept that science will affect beliefs about the senses that are peripheral to ECS, in that they don’t play any or much of a role in our sensory judgments. But also, as I say in the paper, there are possible non-naturalist views on which we ‘the folk’ distinguish the senses in terms of some relatively superficial feature, and consulting relevant sciences might convince us that, for instance, some unexpected perceivings have that feature, leading to their ‘relabeling’ (see Fiona’s commentary for a worry about that, and also my response and exchange with John earlier in the comments). And finally, it seems likely that on most possible non-naturalist views there are ways in which, conceivably, science might play a role in changing ECS somewhat indirectly.

      So far, discussion on this symposium has convinced me that there is a great deal more to say about ways in which, if non-naturalism is true, science might impact our thinking about the senses (and sometimes also ECS, and thus our sensory judgments). I think you’re right then, to say that you and I are drawing the distinction between naturalism and non-naturalism in different ways, because as I’ve drawn it, non-naturalism is not the view that there is ‘no point in looking to the natural sciences because they cannot help us in any way’.

      It may be, of course, that there is some other distinction than the one I’ve tried to make, which for some purposes would be more fruitful. And I can see that the term ‘non-naturalism’ might better evoke the idea that sciences are wholly irrelevant to our thought and talk about the senses, and in that way, perhaps it would have been good to have chosen a different name for the view. The distinction I wanted to make, though, was between a view of ECS on which science very straightforwardly confirms or corrects sensory judgments, and a view on which that’s not the case.

      So, I wonder what your views is on *how* data from psychophysics and neuroscience changed your mind about whether the perception of heat and cold belongs to the tactile modality? On what I call a naturalist view, the data (interpreted in some way by science) can play this roll because we folk have based our judgments that heat is perceived by touch on a mistaken commitment to the thought that perceivings of heat are brought about in just the same way as (other) tactile perceivings. But I can see that that wouldn’t be the only way in which such data might contribute to a change of heart, here. I’d be interested to hear more about what you think!

  10. Hi everyone, really great discussion going on in here!

    I wanted to push a bit on some of the themes that have come up already in a slightly different way. Suppose we move back into the brain and we discover what kind of neural activation is elicited in the sweets case. Suppose that we find out that when we hold our nose we find activation only in areas associated with taste, and that when we open the nose we find activation in both the areas associated with taste as well as those associated with olfaction, and so forth in the other cases (let’s ignore for a moment the recent finding that you get discernible activation in visual areas for stimuli presented in other modalities, or if you don’t want to ignore that we can change the example to talk about sensory specific activation patterns: the pattern associated with visual processing, the pattern associated with olfactory processing, etc). Alternatively suppose that we find that in both cases we find activation of only the areas(/patterns) associated with taste. I would take either of these findings to settle the issue on way or the other, and it seems very different from finding out taste and olfaction share neurons, or that . So, in the first case, can the non-naturalist continue to say that olfaction plays no role in taste perception? We might even imagine that one is able to discern finer details in one’s phenomenology, perhaps coming to be able to introspectively distinguish the olfactory phenomenology from the gustatory phenomenology in what they had previously taken to be solely gustatory.

    • I don’t see how such a finding could have the straightforward consequence you suppose, Richard. We know that certain patterns of neural activation are associated with what we ordinarily characterize as smell and taste, respectively. But couldn’t our everyday conception also accommodate the idea that the brain activity standardly associated with smelling is sometimes involved in tasting instead?

      • Hi John, thanks for the response.

        I guess I was thinking something along the following lines. Let’s assume that our everyday conception of the senses involves their being individuated by a distinctive kinds of phenomenal character. Suppose also that we come to think that this phenomenal character is either identical to or nomologically supervenient on certain kinds of brain activity. We might come to have good evidence that neural pattern A is associated with auditory phenomenal character while neural pattern B is associated with gustatory phenomenal character. So then, were we to discover that when we taste sweets with our nose unplugged we get A+B but when we do it with it plugged we get only A then we ought to conclude that in one case we have olfactory and gustatory phenomenal character, while in the other we do not. I mean it is logically possible, I suppose, that one could maintain that in the first case neural activity B did not instantiate olfactory processing, but that seems like a very drastic thing to say (in the sense of being at odds with current neural theory). (I am neglecting a third option –that there are ‘multi-modal’ contents– for simplicity but we could expand it to include it)

        For comparison we can look at the McGurk effect. I take it that this effect might be thought of as suggesting that speech is not just heard and that visual phenomenology plays an important role in speech perception. Would it make sense for someone who wanted to defend our everyday conception of the senses (which I assume is committed to the claim that speech is only heard) that the activity in the visual areas (which we ordinarily think of as subserving visual processes and instantiating(/giving rise to) visual phenomenology) is actually doing auditory processing or instantiating auditory phenomenology? Maybe if the brain works very differently than we now think it does, but otherwise I don’t think this is a very promising move.

        You might say (and I think this is what motivated your comment) that our everyday conception of the senses is neutral about which brain areas are at work, after all it is supposed to be *ordinary*. But if we are to take Louise’s challenge seriously (i.e. to show what our everyday conception of the senses commits us to that is falsified (or could be falsified) by science) then we need to translate some of our everyday talk into language that might possibly be disconfirmed or confirmed. In this case we need to translate talk about phenomenal character into talk about representations and ultimately neural activity (one might wonder whether introducing ‘phenomenal character’ is already doing some translating). So our everyday conception of the senses is committed (indirectly) to the claim that only olfactory representations are instantiated in smelling, visual representations in seeing, gustatory representations in tasting, etc; and science can confirm or disconfirm that.

        • Hi Richard! Thanks very much for your contribution, and sorry for the slow reply.

          My first response, to your first comment, is very much the same as John’s (Hi John!): I wasn’t sure sure why ‘moving back into the brain’ would help. So I’ll focus on your helpful second comment.

          I think what you’ve offered is another way in which, on a view of ECS that is consistent with non-naturalism (about ECS) science may, given certain other assumptions and answers to other questions, lead us to change some of our sensory judgments. I’ve acquired quite a list of such ways from the discussion so far! Thanks! Yours is particularly interesting. It says: suppose we folk distinguish the senses in terms of feature F. And suppose it turns out that F = G. Then if it is discovered that (say) some of what we thought were tastings have the G characteristic of smellings then, clearly, those perceivings also have the F characteristic of the smellings. And thus, by the folk’s own lights, those perceivings are smellings. Is that right?

          So, all this is premised on the assumption that what, according to ECS, distinguishes the senses is phenomenal character. I’m not committed to that being the case, but here’s some thoughts, given that assumption.

          (1) I wonder if, even supposing that brain activity/phenomenal character identities are discovered, it’s all that plausible to think that ways we have of categorising or ‘typing’ phenomenal character, from the first personal perspective, will map onto ways of categorising phenomenal character that are most important in the scientific study of such things.

          (2) It’s not clear to me that experiencing the McGurk effect reveals to subjects that visual phenomenology plays an important role in speech perception. After all, everyone who experiences the effect realises at the time that they are both seeing and hearing: the surprise is the effect that one appears (in some way) to have on the other.

          (3) My knowledge is very slim, in this regard, but would it be so at odds with what is known about how the brain works to say that (sometimes?) ‘activity in the visual areas…is actually doing auditory processing or instantiating auditory phenomenology’? Kiverstein Farina and Clark seem to suggest not in their forthcoming ‘Substituting the Senses’.

          (4) I worry that your example might be worse for me than you suggest, in that it might put some pressure on the distinction I’m trying to make between naturalism and non-naturalism about ECS. That’s because it looks as if it’s an example on which the sensory modalities are natural kinds with, if you like, a discoverable ‘essence’, after all: the difference that ‘makes the difference’ between the senses, on the view you offer, is a difference in types of brain activity, which is to be discovered or uncovered by science.

  11. Thanks Louise – just to follow up

    I seem to recall that some editions of the OED give “hearing sounds with one’s ears” as an example of a pleonasm. I’m not sure that this would be a merely grammatical point; it has something to do with the meaning of ‘ears’ and ‘hearing’; ears are, by definition, what we hear with and hearing, by definition, takes place by means of ears. ‘Tasting with one’s mouth’ might seem to have a similar ring. But I take it that if you’re right then this is wrong because we taste in part with our noses.

    I realize that you are focusing on an everyday conception of what a sense is and how we should individuate the senses and perceptual experiences and so on. So if you are right that this conception might be a non-naturalistic one and thus cannot be revised by science, then we would all have this conception (whatever we might think). So do you then think that those of us who believe there are more than the five senses have another conception of the senses as well? Or do we have the ECS and are able to understand all of the other senses we think exist by our ECS non-naturalistically construed?

    I’m also wondering what the content of the naturalistically construed ECS would be. What would the folk ordinarily have to believe about the senses that has to be revisable by science for the ECS to be naturalistic? There seem to be a number of possibilities, but here are two:

    1. Our ECS is such that the folk think science might give us new information about the kinds of conditions that we should use to individuate the senses. I think that you note the one’s that are usually referred to in the literature – content, character, sensory organ, physical stimuli. It looks like the folk do not use any more than these.

    2. Our ECS is such that the folk think science might give us new information about what is actually going one when some organism – us when we taste/smell flavours or some other creature when it seems to use a sixth or nth sense – such that we modify the way we think of that sense, e.g. classify it in a different way.

    These, especially the first, would seem to be quite sophisticated contents/conceptions for the ordinary folk to have.

    Regarding your use of ‘naturalism’, I think it is fairly standard in the literature to think of naturalism as having (at least) two aspects – methodological and metaphysical. Very roughly speaking, the first is that we should look to the sciences to help resolve philosophical problems and the second is that we might try to show how putatively problematic features are natural features. You focus on the first, but what about the second? I wonder whether there is some instability in the other sort of non-naturalism you give. Take your example of lakes and mountains. Are these natural properties? If they are why shouldn’t this be a form of naturalism. If they are not why doesn’t this become a form of conventionalism? After all, think of the convention of calling certain protuberances in Scotland ‘Munros’; ‘lake’ and ‘mountain’ might be used as conventions in a similar way. So it sounds surprising to call the view a form of non-naturalist non-conventionalism.

    • Thanks for following up, Richard: it’s much appreciated. Four further follow-ups from me!

      First, on whether we ‘taste in part with our noses’. The data shows that receptors in the nasal cavity are crucially involved in normal flavour perception. But I think that which body part one perceives ‘with’ might mean, in certain circumstances, something different, for instance, the body part one actively uses to explore that which one is perceiving, or the body part ‘from which’ one seems to perceive.

      Second, on how a non-naturalist about ECS should construe the beliefs of those who believe there are more than five human senses. The short answer to this one is that I’m not sure! But (see also my response to Fiona) I think at least part of what a non-naturalist will have to say is that ‘sense’ and related terms are polysemous.

      Third, on the content of a naturalistically construed ECS. As I say in the paper, I take it that the nature of ECS may be opaque to those who have it (I guess the fact that one can *ask* about the nature of ECS goes someway to demonstrating its opacity. Although perhaps you think, like Fiona, that non-naturalism about ECS is wholly implausible anyway.) I tried to avoid characterizing the naturalist view as one on which ordinary people have implausibly sophisticated contents or conceptions. Specifically, I suggested that the naturalist might say that according to the folk, the senses are distinguished by something to do with how different perceivings are produced by the effect of the environment on our internal perceptual equipment (whatever that might turn out to be).

      Finally, on methodological and metaphysical naturalism. I’m not sure I have a very good sense of what makes a feature a natural one. But I take your point (and thank you for it) that if the features that we folk distinguish the senses in terms of are natural features then ‘non-naturalism’ may be a misleading term for the sort of view I sketch, in this way as well as the way you suggested in your previous comment. And in fact one thing I’d like to think about more is whether all non-naturalistic views (in the misleading sense) are committed to being antirealists about the senses as you suggest (in your 2011) that Matt is.

  12. Hi Louise,

    Thank you for your comments on my response to your paper. I’d like to ask you something about each of the three comments that you make.

    First, in response for my urging that you distinguish the “experience question” (the question of which type of experience you are having) from the “modality question” (the question of which type of modality is being used), you suggest that there are two uses of “modality”. There is the one that you say that I use, according to which a modality is “a mechanism or process or sensory system that produces perceivings”. If one has this sense in mind, you agree with me that the type of modality being used and the type of experience being had on a particular occasion can come apart. The second sense of “modality” is the one that you say that you have in mind, according to which a modality is a “capacity to have a certain sort of perception/perceptual experience.” And you claim that on this sense, it is harder to suppose that the type of modality being used and the type of experience being had on a particular occasion can come apart. Therefore, I take it that you would say that you don’t have to distinguish the experience and the modality questions.

    I don’t think that you can avoid this so easily. I agree that having a certain sort of mechanism or process or sensory system is different from having a capacity. (Although note that having a capacity to smell will at the very least typically involve (i) having a mechanism that allows one to smell, (ii) having a capacity to undergo a process that produces perceivings of a certain sort, and (iii) involve having a smell sensory system.) I also agree that it is harder to suppose that if one had a capacity to smell then one wouldn’t have a capacity to have olfactory experiences. However, in the puzzle about the sweets we are not asking whether the person with the sweetie in their mouth has a capacity to smell or taste. I take it that we are supposing that the person that we are considering in the puzzle of the sweets has the capacity to do both. We are asking which of his or her capacities is being exercised: his or her smell or taste capacity. Regarding the exercise of capacities, it seems to me that we can differentiate two questions of just the sort that I proposed: (a) is one exercising one’s capacity to smell or one’s capacity to taste, or (b) is one having an olfactory or taste experience. The reason to think this is exactly the same as I gave in my initial response to you: one could hold that which capacity one is exercising depends on which sensory organ one is using or which proximal stimulus is stimulating you, while one could think that which type of experience one is having depends on the representational content or phenomenal character of the experience. If one thought this then it would follow that sometimes the exercise of one’s capacity to smell could produce taste experiences. And more generally, sometimes the exercise of one’s capacity to sense in a particular modality could produce experiences that were experiences of a different modality. Therefore, I can’t see how your response allows you to avoid being clearer about the set up of the issue by distinguishing these two questions and being clearer about which one you are asking.

    Second, I agree that you do make the point in your paper that “according to certain members of this family of views [non-conventionalist non-naturalism] science might lead us to ‘re-label’ some perceivings” (p328). Given how you spell out non-conventionalist non-naturalism, you have to. I think that the fact that non-naturalism, thus conceived, entails this point totally undermines your thought that science cannot tell us that in the puzzle of the sweets that we are smelling. It precisely could tell us that one or even many instances that we took to be tasting were instances of smelling. But if that is right, then the main thing that you set out to show in your paper is false. I don’t understand why you are not more troubled by this. Do you think that there is another claim in the vicinity to be salvaged?

    With regard to this point, here is one way that you could reply. I think that you could explain the non-conventionalist non-naturalist position slightly differently. Of course you will want to stick by the claim that according to non-conventionalist non-naturalism science can’t tell which feature it is that makes a sense a sense. But you should, I think, ditch the claim that science could reveal to us previously hidden features relevant to determining which modality is being used to perceive. One might stick more closely by O’Shaughnessy’s claim that some things have no hidden nature, and claim that the senses have no hidden nature relevant to determining which modality they are. After all, O’Shaughnessy’s example of the mountain that you cite, and which you want to treat as analogous, is such that being a certain elevation from the surround is the feature that makes a mountain a mountain. That is just not something that we ever have to wait for science to tell us that we got right or wrong. In a similar manner, you could claim that what makes the senses the type of sense that they are is not something that science could reveal. If you held this then I think that you would have to deny that use of the nose is a surface feature of a modality, and therefore cannot be the feature that makes a modality smell. In fact, I think that the only features that one could hold to be features of the senses that science could not correct us about is the nature of the phenomenal character (and perhaps the representational content) of the experiences produced by the senses. So you would have to hold that the folk’s conception of the senses individuates them by the conscious nature of the experiences that they produce. I think that such a conception of the senses is possible, however, I cannot see a good reason to think that this is the folk’s conception of them. I’d be interested to know if you’d want to respond to me in the manner that I have just outlined.

    Third, what you say in response to my third point re-iterates points that you made in your paper. I understand the view that you want to promote. I hope that you don’t mind me pushing again two questions that I asked you in my comments. (1) What do you say about someone like me who has completely stopped saying that there are five senses and has done so for about twenty years? I do not revert to saying that there are five in any situation. In fact, I take great delight in correcting people! Do I still believe that there are five and only five senses and that there could only be five? Was my belief that there are just five senses correctable? (2) Why should one think that there are beliefs about the number of the senses, or beliefs about the individuation of the senses, that cannot be revised? As I noted previously, if there are any beliefs that are not open to revision then one would expect these to be beliefs in the domain of logic or mathematics. There are reasons—albeit highly controversial reasons—to think that these beliefs might be immune to revision as I pointed out in my commentary on your paper. However, no such reasons seem forthcoming in the case of our beliefs about the senses. That people frequently say that there are five senses does not provide a reason that I think should motivate us to think this.

    • Thanks, Fiona, for your questions and for being so generous in your attention to my paper. I apologize for the length of what follows!

      (I) On your first question: I agree, of course, that we’re asking, in the flavour case, which capacities are being exercised when one perceives flavour. And we suppose that relevant perceivers have both capacities. However, I still think that on the sense of ‘modality’ I’m using, the experience and modality questions don’t come apart in the way you suggest.

      As I see it, understanding a sensory modality as a perceptual capacity is to understand it as something one has if and only if one has the potential to have a certain sort of perceiving or perceptual experience. Capacities or potentials, thus understood, are not individuated independently of that which they are capacities or potentials *for*.

      This isn’t to assume that modalities are individuated by phenomenal character. It might be that kinds of perceiving, or perceptual experience, are individuated by proximal stimulus or sense organs, for instance. But then, so will the capacity to have that perceptual experience be individuated in the same way: so the modality question and the experience question will be answered together.

      It also isn’t to deny that having a perceptual experience involves having some functioning sensory equipment: it’s just that this use of ‘modality’ doesn’t presuppose that modalities (and perceptual experiences) are distinguished in terms of differences in this sort of equipment.

      Thus understood, to ask which modality perceivings of flavour belong to is to ask whether the perceiving is an exercise of one capacity (for having gustatory perceptual experiences) or two capacities (for having gustatory experiences and for having olfactory perceptual experiences).

      On your view, we ought to understand modalities or capacities in such a way that they are individuated independently of the kind of perceiving to which they give rise. I think what we might have here is a substantive disagreement rather than a failure to distinguish two questions, on my part. I’m not sure that your view is the neutral starting point I ought to have adopted in the paper.

      (II) On why I’m not more troubled by your second objection: Non-naturalism is characterized by the following negative claim, (N): it is not the case that our everyday conception of the senses involves commitments about differences in the way in which perception is produced by the effect of the environment on our internal perceptual equipment.

      Given the truth of (N), scientific findings about such differences would not speak directly to the question of which modality some perceiving belongs to. *This* is, I take it, ‘the main thing I set out to show in the paper’ –that only if naturalism is true, do these findings straightforwardly determine, for instance, the solution to the puzzle about the sweets.

      Now, as you, Mohan, Matt and others have pointed out to me, there are nevertheless various really interesting ways in which scientific findings *might*, more or less indirectly, affect our sensory judgments, if N is true. However Whether they do, and how they do this will depend, amongst other things, on which positive view of how the senses are distinguished one adopts, and there are very many such views consistent with N.

      If this is right then one who wants to argue that there is a route from the scientific data to the claim that flavours are perceived with the sense of smell will have to either defend naturalism, or (i) defend a particular positive view of how the senses are distinguished that is consistent with N and also (ii) show how, on this positive view, the data corrects the relevant sensory judgement(s). I’m not especially troubled by accepting this because it’s still a view on which the scientific data *itself*, doesn’t settle the puzzle.

      I agree with you, as of course I must, that some of the positive views consistent with N are going to be ones on which the route from the data to a change in sensory judgments will be shorter than on other such views. For instance, some surface features will, as you suggest, be ones that it is less plausible to think that science will ‘correct us about’. But I’m not committed to any particular positive view, and it’s not clear which, if any, one should commit to.

      A note on a side issue: it’s not obvious to me that ‘use of the nose’ has to be understood as the sort of feature about which science will straightforwardly tell us whether we are right or wrong. That’s because ‘use of the nose’ might mean something like ‘sniffing’ or ‘getting the nose (i.e., the external, visible part of it) into the right place to perceive’. In which case, the involvement of receptors located in the roof of the nasal cavity, for instance, will not be at all obviously relevant to whether a perceiving involves the ‘use of the nose’.

      (III) In answer to your two questions:

      (1) No. I don’t think (and I don’t think I’ve committed myself to thinking) that you still believe that there are five senses! If non-naturalism is correct, then I can say either (i) you’re wrong to think that there are more than five human senses, or (ii) you’re using ‘sense’ or ‘modality’ in a way that diverges from the everyday conception, or (iii) some positive view of how to distinguish the senses, consistent with N, is true, and the scientific data impacts upon the everyday conception in one of the interesting, indirect ways mentioned above. (The reason I reiterated points I made in the paper, in my previous response, is because I wanted to make clear that nothing in the paper forces me to say that you believe, even just at the weekend, that there are five senses.)

      (2) I don’t think that non-naturalism about ECS entails that there are beliefs about the senses that cannot be revised, in the way in which someone *might* think that some beliefs in the domain of logic or maths are.

      Consider, for instance, Matt’s conventionalist view. As I understand this view, it doesn’t entail that senses other than the familiar five are inconceivable, and thus the fact that we can easily imagine such things just doesn’t speak to the truth of this view. Rather, if Matt’s view is right, sensory judgments aren’t (or at least, aren’t straightforwardly) revisable *by findings about which sensory systems (or mechanisms, or processes) produce which perceivings.* That’s because, on his view, what we’re distinguishing between, when we distinguish between the senses, are ‘ways of perceiving’, which are sets of conditions that have to be satisfied for us to perceive something. We group these conditions into the five sets that we do as a matter of convention, but in a way that’s constrained by the practical purposes to which we put the distinction.

      If this is right, the sort of findings mentioned in the last paragraph will have at best an indirect impact on our sensory judgments. But it’s consistent with this that if we were quite different (for instance, if we had spiderman’s powers) or if the world were quite different (for instance, if light behaved quite differently to how it in fact behaves) we would find it useful to identify senses quite differently to the way in which we currently do. And we can imagine this being the case.

      Or, consider the sort of view that Matt offers me in his response to my paper. He suggests that I might want to say that our everyday conception of the senses is a loose set of platitudes that cannot be refined into a rigorous theory. Again, if this is right, there’s going to be no straightforward revision of sensory judgments on the basis of the above-mentioned findings. But this is not to say that our beliefs about the senses are unrevisable in the way that (maybe) 2+2=4 is unrevisable.

      I’m sure there’s still lots more to be said—thank you for pushing me to say a bit more of it, anyway!

  13. Louise, I thought it might be fun at this late stage of the discussion to introduce a related but unrelated question. Forgive me if this takes us too far afield.

    Consider pain. You might think that there is an everyday conception of pain (ECP), if there is of anything. Yet there is a controversy about what it is. This controversy apparently raged in medieval times, and continues today.

    Some people hold that pain is a sensation of its own, a separate feeling that accompanies, but is distinct from, other sensations. If I cut myself on a knife, for instance, there is the tactile feel of the knife, and accompanying it, there is a pain sensation. The latter is distinct from the former.

    Others think that pain is an uncomfortably intense (“painful”) sensation in another modality. For example, a very loud sound or a very cold liquid can produce pain.

    Now I am sure that you and everybody else will assent to the examples above. But the question is this. Is there an ECP that unites these kinds of pain? Is one of the above theories right and the other wrong? Or are there two kinds of pain?

    The question overlaps with some of those you consider (I think). If pain is a single modality, the pain that arising from touching a hot stove is of the same kind as that of eating a very hot chilli, and thus to a different modality from either the neutral sensation of heat that one experiences in a tepid bath or that of tasting a warm poached egg. If, on the other hand, pain is simply a characteristic of other experiences (for example, that they are uncomfortably intense), then these two kinds of pain belong apart; one is tactile, the other gustatory.

    Do you think that these sorts of questions can be answered by an everyday conception? Or do you think that either science, or alternatively guided experience, is needed?

    • Hi Mohan. Thanks very much for returning, and for introducing such an interesting issue!

      So first, your question: my own view is that there is no reason to suppose that there is an everyday conception of pain that settles the question of whether there are (a) distinct experiences of pain or (b) just painful (perhaps because intense) experiences in other modalities. But then, I don’t think that the everyday conception of the senses settles the question of which modality flavour perceivings belong to either. I do think that what view is true of the everyday conception of the senses will partly determine if and how we can be corrected, by science, about judgments of flavour. Furthermore, whilst, as you say, we surely have some sort of everyday conception of pain (in that we have lots of beliefs about it, say) it seems as if this ‘ECP’ might be entirely neutral about the sort of classificatory question you ask. It seems plausible to me that neither sort of answer to the pain question you raise would be a correction of an everyday or folksy judgement.

      The interesting question, I suppose, is how one *would* settle this question. I wish I knew! I suspect that lots of different considerations will be relevant. One is the independently interesting and difficult question of how to count experiences on particular occasions. If pain is perceiving (in any modality) painfully, then feeling the slice of the knife, and feeling the pain, are probably just one experience. If pain is a distinct sensation, then perhaps one would want to say that there are two experiences here: one of pain, and one (a tactile experience) of slicing. Perhaps there might also be a view on which counting experiences is a context relative affair. In which case, it might be that in some contexts you’d want to say that pain is an aspect of an experience in another modality and in other contexts that pain is itself a distinct, additional experience.

      It also seems likely that this is the kind of issue one might be driven to take a stand on in virtue of other philosophical views one holds. Here, I think I can relate the point back to the paper! Whilst ECP might not take a stand on the question you mention, I can imagine someone thinking that there is a much more general ‘everyday conception of the mental’, that commits we folk to categorising mental items in terms of the mechanisms and processes that produce them. In which case, the right place to look to answer your question, in quite a direct way, would be scientific data about the mechanisms and processes that produce pain. (I suppose that, thinking about your original response to my paper, one might also look to use this sort of scientific data about pain less directly, to *contribute* to answering the pain question, even if one thought that the everyday conception of the mental was not thus committed. But then one would need to put some work into explaining the relevance of the data.)

      I can also imagine someone saying: it really doesn’t matter what the everyday conception of the mental is: it’s just *true* that we should categorise mental items in terms of the mechanisms and processes that produce them. But that looks to open the door to a kind of eliminativism about mental items, or at least about the categories of mental item that we usually identify. (This is an issue that came up in discussion with Josh, earlier.)

      I think I may have drifted too far away both from the paper *and* your pain question, now!

  14. Hi Louise (and everybody else) I’m a little late jumping in on this interesting discussion; I only found out about it a few days before setting off on a trip. I just got caught up on the back and forth, as well as the original article, and there is much to think about. Let me add a few ingredients to the stew here.

    Probably not surprisingly, my reactions are probably closest to those of Fiona Macpherson and Richard Gray.

    With Macpherson, on reading the target article, I’m a bit at a loss exactly what the specific claims of the nonnaturalist reading of the ECS is. As I see it, in order to not be threatened with correction or elimination by discoveries in the sciences, the ECS would need not to be committed to any of the things science might uncover details about.  In the case of flavor, you claim that the details of the mechanism of the sensory process do not contravene anything that the common sense view of the senses is committed to.  OK, but this is a double-edged sword. The more commitments the ECS refrains from making, the less it seems to be saying anything about anything. 

    Part of my confusion here might be the negative framing of the argument. There are many claims about what science cannot do and what a common sense view of the senses is not committed to. You tell us that science can’t tell us that we’re wrong about the nature of taste perception because the common sense view isn’t committed to anything detailed about the mechanism of sense perception. In other words, common sense can’t be shown to be wrong here because it wasn’t attempting to be _right_. OK, fine enough, but is it trying to be right about anything about the senses, and if so what, exactly?

    I’m largely in agreement with what Gray describes as his own minimal commitment to naturalism.  I found your clarification helpful: “As I’m using the terms, the non-naturalist doesn’t have to deny that, as you [Gray] put it, ‘we should consult the natural sciences in order to inform our thinking about the senses’. For one, it is open to a non-naturalist to accept that science will affect beliefs about the senses that are peripheral to ECS, in that they don’t play any or much of a role in our sensory judgments.” With Macpherson, though, I’m not clear at all on what the metric of centrality/peripheralness is. 

    This question is important because of the relationship of senses to explanation. You make it clear that whatever the nonnaturalist view of the senses is, the view is not about scientific explanation. You write of Barry Smith’s take on all of this, that “it’s the centrality of the claim that flavours are perceived partly olfactorily to the psychology of flavour perception that justifies its wide acceptance” (336). You note that, 

    >One way in which to respond to this argument would be to question the assumption that usefulness in science is sufficient to determine truth. We saw, in Section 1.2, that according to the Nuddsian conventional view of our everyday conception of the senses, our distinguishing them as we usually do—so that flavours count as tasted, and not, partly, smelled—is useful too. The usefulness in question is not of course usefulness to science. The interests served are the everyday ones of interpreting and predicting the behaviour of other perceivers. On Nudds’ view, we distinguish the senses as we do precisely because doing so serves those interests. But even if one disagrees that the serving of these interests explains our making the distinction in the way Nudds suggests, one might still agree that these interests are, nevertheless, served by our making it in this way. (ibid).

    As you indicates in a footnote, in my first paper on this topic, I had the goal of trying to make the concept of a sense scientifically useful. But I agree with you that this isn’t the _only_ way that a notion of the senses might be useful. (Indeed, in the years that have passed since that first paper on the topic, I’ve found myself coming to a much more pluralistic view of the senses.  We don’t have any more a univocal concept of “sense” than we do of “species,” or “gene,” or many other central concepts in science, especially those concepts that also play a role in our everyday interactions with others.) 

    What seems to be at play here is a distinction between “scientific (naturalistic)” explanation and “nonscientific (nonnaturalistic)” explanation. Further, than distinction largely turns on the invocation (in the case of naturalism) or not (in the case of nonnaturalism) of some “hidden nature” in explanations of the two sorts. Scientific explanation involves hidden natures, whereas nonnaturalist explanation only invokes manifest marks. Is this right?  If so, my Sellars/Hanson/Churchland “theory-ladenness of perception” worry would be that this distinction begs the question. What is “hidden” or not in the senses is  not something that is independent of our theories of the senses.  To those of us in the Aristotelean West, the facts of the balance sense are largely “hidden” and we do not, as a culture, count the balance sense among our basic senses. For the Anlo-Ewe of West Africa (if sensory anthropologist Guerts is correct), the situation is different. Their ECS “reveals” balance to them in a way that we don’t experience it. And, when it comes to taste, umami is now part of my experience of tasting that was “hidden” before.  So, is this naturalism/nonnaturalism distinction culturally relative or am I understanding what is meant by that which is hidden and not, in the case of senses?

    • Hello Brian. Thank you very much for contributing to the symposium, especially during a busy time for you! To keep things simple(ish) I’ll reply to your two comments separately.

      I think that the negative framing of the argument has made things difficult for me, and for someone reading the paper, in various ways. However, perhaps unfortunately, my purposes on the paper were largely negative! In discussions of flavour perception, it seemed to me that a move was sometimes made, very quickly, and without explanation, from scientific data about the shared underpinnings of flavour perception and smell, to claims about the modality to which one ought to think that perceivings of flavour belong. I take it that this has to involve some assumptions about the nature of our everyday conception of the senses. If naturalism is true then the very direct route from the data about flavour perception to the claims about categorising perceivings looks justified. But naturalism, thus understood, is far from obviously true. I don’t take this point to hang on a defence of the truth of some particular positive view of ECS consistent with non-naturalism. In fact, the variety of views consistent with non-naturalism is part of the point: there are a lot of possible views of ECS, consistent with non-naturalism, on which, at best, the move from the data to the claims about categorisation will need explaining. (I’ve been convinced by the discussion here that it’s important that I don’t say that if non-naturalism is true, science cannot play *any* role in changing our sensory judgments.)

      Up to a point, then, I’d like to resist your question about what ECS *is* trying to do, since it wasn’t my intention in the paper to defend any such positive view (nor even non-naturalism). But partly in the interests of not being overly infuriating, I’ll try to say something more!

      Minimally, ECS is just whatever it is that underlies our sensory judgments. So what it tries to do is categorise perceivings. The non-naturalist might say that in so doing, it’s not really trying to be right about anything: Richard Gray interprets Matt Nudds’ view as a kind of anti-realism. If that’s right, then Matt, whose view is consistent with non-naturalism, might say that whilst (everyday) sensory judgments are truth-apt, and useful, they aren’t trying to pick out any independently existing kinds of thing.

      I suspect though, that it’s not compulsory for a non-naturalist to be an antirealist. For one, they might say that whilst the sensory modalities of ECS are not natural kinds, they are still *real.* Alternatively, they might resist even the claim that the sensory modalities are not natural kinds. They might say: the sensory modalities are kinds that are perfectly natural (by whatever measure of naturalness is relevant) and ECS picks out those kinds.

      But as you rightly say, the non-naturalist view is not about scientific explanation, and furthermore, if non-naturalism is correct, neither is ECS any sort of *competitor* to scientific explanation of perception. I am, like you, also tempted by the kind of pluralism you mention: I think that’s got to be part of the story. But of course, that doesn’t make moves from scientific data to claims about how one ought to categorise perceivings any simpler, or less in need of defence.

      >I’m not clear at all on what the metric of centrality/peripheralness is.
      I meant, in making this somewhat rough and ready distinction, that a belief about perception or the senses is peripheral to ECS to the extent that it doesn’t play a role in bringing about sensory judgments. I didn’t have in mind any particular way of measuring peripheralness: my intention was just to point out that we folk might have beliefs about the senses (such as the idea that there are four basic ‘taste’ qualities) that don’t impact upon our judgments about the modality with which we perceive.
      >…is this naturalism/nonnaturalism distinction culturally relative or am I understanding what is meant by that which is hidden and not, in the case of senses?
      That information about the Anlo-Ewe is very interesting! Thanks! I intended the appeal to ‘hidden natures’ to invoke the idea that what makes a thing (such as a sense) what it is, is something that awaits scientific discovery. I think that goes along quite nicely with naturalism, considered as a view about ECS and its commitments. What I had in mind, then, is that a naturalist might say something about the senses analogous to what Lycan says about belief, i.e. that “…the word ‘belief’ points dimly towards a natural kind that we have not yet fully grasped and that only mature psychology will reveal. “ (Judgement and Justification, p32) I think it’s readily apparent that one might not agree with Lycan here, in that one might not accept that the features that make something a belief await discovery by mature psychology. Similarly for the senses, on a non-naturalist understanding of ECS: they don’t have a hidden nature just in the respect that we don’t await mature psychology (say) telling us what makes them what they are.

      I find it hard to place the Anlo-Ewe’s identification of a ‘balance sense’ in this picture. If one is a naturalist, and thinks that this ‘balance sense’ is a sense in just the same respect as are the other senses that ECS (somewhat sloppily) identifies, then you will have to say that the Anlo-Ewe identify a sense that it has taken scientific investigation to reveal to us. Whatever allowed them to recognize this ‘sense’ need not be that ‘hidden nature’ which science would reveal: perhaps there’s just some aspect of experience, say, that Westerners overlook, but which is salient, for cultural reasons, to the Anlo-Ewe.

      If one is a non-naturalist, then how one responds to this case will depend a great deal on one’s positive views. If one adopts Matt’s view for instance, then I suppose one will just have to say that the Anlo-Ewe have different conventions.

  15. Since this is a different line of thought, let me add this as a separate comment:

    I also want to add my voice to Macpherson’s affirmation that she thinks in terms of more than 5 senses as an unexceptional matter of course. One nice thing that has occurred in the almost 30 years since I first became aware of the cognitive sciences has been an enrichment of my sensory phenomenology. A number of injuries and reactions to pharmaceuticals has also opened up experiences of which I was unaware of as a teenager.  As Heidegger notes, much is revealed when a tool is broken; by the same token, you can learn a lot about your mind/brain when your body is broken. I would put the experience of eating with a blocked nose into this general category. But I routinely and consistently think of balance, pain, and proprioception as senses on par with vision and hearing. I also spend a lot of time thinking about pleasurable touch, thirst, hunger and even pheromones similarly.

    Why do we (in the Aristotelean West) answer the question “How many senses do we have?” with the answer “Five”? Because that’s been the refrain for quite a long time. Although Richardson and others have their doubts that it’s this simple (and it’s not), my own view is that this situation is similar to what happens when you ask Americans about whether there’s any pattern to when “i” goes before “e” and when the other way around.  A very high percentage of Americans will quickly respond with “‘I’ before ‘e,’ except when it follows ‘c'” (with a number of variations). The problem is that this rule has so many exceptions that it is practically useless! (See “http://en.wikipedia.org/wiki/I_before_E_except_after_C”).  Why do we have the rule? Because sometimes having a poor rule is more useful than not having one.  Why do we say we have five senses? Because it’s actually a pretty confusing topic and having a poor answer is sometimes more useful than not having one.

    Finally, what’s a “cod-theory”? I’ve never come across this term before. Even Google has failed me!

    • >I also want to add my voice to Macpherson’s affirmation that she thinks in terms of more than 5 senses as an unexceptional matter of course.

      Sure! I do hope I’ve made clear that I don’t think that it’s false that there are those who genuinely believe that there are more than five senses! Your suggestion here reminds me of Mohan’s comment, right at the start of this symposium, about how familiarity with scientific data might change our everyday conception in some really interesting, indirect ways. Mohan’s example was of the way in which participation in some studies might lead one to change one’s perceptual practice, and then, in turn, to change how one categorises some perceivings.

      What I was most keen to emphasise in the paper was the lack of a direct route from facts about the processes, mechanisms and systems involved in perceiving, to the correction of any of our sensory judgments, unless one makes some decidedly non-obvious assumptions about what we’re up to in making those judgments. You seem to be hinting, here, that you also think that this route is somewhat indirect: you’ve been convinced, in part, to change your judgments due to the way in which the phenomenal character of your experiences have changed with greater familiarity with cognitive science. I wonder why that is. What do you think the role of phenomenal character is in making sensory judgments?

      I also find the view you articulate of why we answer ‘five’ to the counting question interesting. I suspect it’s a common view that the explanation for thought and talk about five senses is something that is, in this way, quite uninteresting! And of course I’ve done nothing at all to rule out that this is all there is to be said. But I guess I’m still not clear about why one *has* to say this, or something like it. Or at least, why one wouldn’t think it worth considering whether there might be more to it. In talking about his favoured philosophical method, Austin writes, in his ‘Plea for excuses’ of the way in which our words embody ‘all the distinctions that men have found worth drawing, and the connections they have found worth marking, in the lifetimes of many generations’. These distinctions have, as he remarks, ‘stood up to the long test of the survival of the fittest’. I don’t take it I’m doing ordinary language philosophy, but the quotation is nevertheless appropriate because I start from a position of being interested in why we make the distinction between the senses in the way we tend to, and the assumption that there’s an interesting and difficult question to be asked about why and how we do so.

      >Finally, what’s a “cod-theory”? I’ve never come across this term before. Even Google has failed me!

      When I first read this, I googled too and initially thought, with some embarrassment, that I must have invented a silly expression. I’m relieved to say that I have now found the expression used elsewhere! As I understand it, it just means a theory that’s somewhat hastily put together, or not very well thought through. (I guess it might be a somewhat local expression!) Here, chosen somewhat at random, is one page the expression is used as I use it:
      http://www.ukjusticepolicyreview.org.uk/what-we-ve-been-up-to/makingsenseofcrimetrends

  16. Hi Louise,

    Thank you for your careful consideration of my comments. Here are some further thoughts on each of our discussion points (I) – (III).

    (I) I think that we are agreed that on your view to possess a sensory modality is to possess a capacity, and capacities are not individuated independently of that which they are a capacity for. You want to say that to possess the smell modality is to possess the capacity for having smell experiences. To ask whether the fruitiness of the sweetie is tasted or smelled is to ask which capacity is being exercised: one’s taste or smell capacity. And that’s to ask whether one’s capacity for having taste or smell experiences is being exercised. In this sense, you are defining the sensory modalities in virtue of the experiences that they produce. I think that this can all be true and yet in the puzzle of the sweets the modality question (which modality is being used) and the experience question (which type of experience is one having) can still come apart. Let me explain why.

    Suppose that one exercises one’s capacity to X. Does it follow that one has X-ed? Suppose that one exercises one’s capacity to have a smell experience. Does it follow that one has had a smell experience? On one plausible view of capacities, the answer is no. On this view, capacities can be fallible. The philosopher that I know of who has discussed the fallibility of capacities at most length recently—and perceptual capacities at that—is John McDowell (Perception as a Capacity for Knowledge, Marquette University Press, 2011). According to McDowell, capacities to X can be fallible. To have the capacity to X means that you can sometimes X, but there can be defective exercises of the capacity. For example, to use McDowell’s example, suppose that you have the capacity to throw a basketball through a hoop. This means that you can sometimes throw a basketball through a hoop. Perhaps it even means that you can frequently throw a basketball through a hoop. It does not mean that you can always do it if you try. If that’s what the capacity entailed, then no one on Earth would have ever have had the capacity because no one, I take it, has ever not missed a shot. We can acknowledge that you have a capacity to throw a basketball through a hoop even though sometimes you miss. When you exercise a capacity to X and you fail to X, McDowell says the exercise of your capacity is defective. When you exercise your capacity to throw a basketball through a hoop and you miss, you have exercised your capacity, but you have done so in a defective manner. (McDowell goes on to argue that perception is a fallible capacity to get into positions in which one has indefeasible warrant for certain beliefs. However, the merits of this conception of perception is tangential to the debate here.)

    Let us return to the possession of the modality of smell. Suppose that your view, Louise, is right, and that this is a capacity to have smell experiences. Now consider the puzzle of the sweets. One can ask whether people who experience the fruitiness are exercising their capacity to have smell experiences. Suppose that we think that they are exercising their capacity to have smell experiences. That they are so doing does not entail that they are having smell experiences, for capacities are fallible and one can exercise one’s capacity to have a smell experience and yet not have a smell experience. If that is right, then the question of which modality is being used is distinct from the question of which type of experience is being had. If one wanted to resist this conclusion then one would have to deny that one’s capacity to have smell experiences was fallible. I wonder if this is the route that you wish to take?

    (II) You said in your response to me that, “Non-naturalism is characterized by the following negative claim, (N): it is not the case that our everyday conception of the senses involves commitments about differences in the way in which perception is produced by the effect of the environment on our internal perceptual equipment.” If this is what you take non-naturalism to be, then I agree that science could not tell us in the puzzle of the sweets that the use of the nose means that we are using our sense of smell to detect the fruitiness of the sweeties. However, I’m exceptionally surprised that this is what you say that you take non-naturalism to be. And I don’t think that this is what you take it to be in your paper.

    In your paper, you say that naturalism is the view according to which “what individuates the senses [is] … correctable (or confirmable, in theory) by science” (p. 326). And you say that, “there is more than one sort of property that might be thought to constitute the ‘hidden nature’ of the senses, discoverable, on this view, by science” (p. 326). This seems right as a characterisation of naturalism. One might then think that according to non-naturalism there is more than one sort of property that might be thought to characterise the “non-hidden nature” of the senses—a property that nonetheless is one that will not be revealed to us by science. Indeed the first line in your paper that states what non-naturalism is suggests this very conception: “non-naturalism, our everyday conception of the senses does not carry correctable, empirical commitments about how the senses are individuated” (p. 326). You do not allude to the idea that the view is that there is one particular property—the use of certain sorts of internal equipment—which is not taken to be constitute the “non-hidden nature” of the senses. Indeed, when you are describing non-conventionalist non-naturalism you suggest that one form of the view is that the senses are to be identified by surface features, where one of these could be taken to be “the use of certain parts of the body (the ones we usually think of as sense organs) in perceiving” (p. 328). And you go on directly to say “Since my purpose here is not to argue for non-naturalism about our everyday conception, still less for any particular variety of non-naturalism, pursuing these suggestions will have to wait for another occasion” (p. 328). This suggests that you take non-naturalism to be any view according to which there are non-hidden, surface features that individuate the senses. Indeed, you say “*Any* such view, on which our concepts of the senses were ‘surface’ concepts, would be a non-naturalistic view of our everyday conception: it would be a view according to which our everyday conception does not have empirical commitments about what individuates the senses that await correction or confirmation by science.” (p. 328).

    I think that you are right to elucidate non-naturalism as you do in your paper. Therefore, I do not accept what you say in your last response to me that “Non-naturalism is characterized by the following negative claim, (N): it is not the case that our everyday conception of the senses involves commitments about differences in the way in which perception is produced by the effect of the environment on our internal perceptual equipment.” If I am right about what non-naturalism is, then the point that I made in my last response to you stands. If you agree that according to non-naturalism “science might lead us to ‘re-label’ some perceivings” (p328) then one could be a non-naturalist and think that science could tell us in the puzzle about the sweets that we were using the modality of smell. It is only by holding a very particular form of non-naturalism that has substantial commitments about what does not individuate the senses that one can resist this conclusion. But you say explicitly in your paper that you are not looking to defend any particular form on non-naturalism and that it is holding non-naturalism in general that shows that the puzzle of the sweets could not be addressed by science.

    (III) Thanks for your comments about (III). They helped me to understand your view better. I like your suggestions for thinking that the judgements about the natures of the senses might be unrevisable but not on the grounds that one might think that logic or maths beliefs are: they are just convention or they can’t be refined into a rigorous theory. But I cannot see a good reason to see that the senses are like this and so I see no inclination to think that they are. And I’d want positive reasons to motivate me to think this. Perhaps you think that unless we have positive reasons to deny that the senses are like this then you want to leave open this possibility. So perhaps we just have different starting points here and do not feel the force of the dialectical move that we are each trying to impose on the other. What to do then? I’m not sure! I suspect that much will end up turning, as you yourself have identified, on whether it is right to think that there is continuity between the language of science and ordinary language. I think that there is, though I have not offered in depth reasons to think that this is true.

    • Thanks, Fiona. I am also unsure, re. III, how we would go about persuading each other of the force of the dialectical moves me disagree on! However, I find in immensely helpful that we’ve identified this point of disagreement.

      One quick point on (I). It strikes me that there are limits to what one can plausibly say counts as the exercise of a capacity, short of what it’s for, that are relevant to our discussion about individuating capacities. Roughly: it’s one thing to say that the exercise of a capacity can give rise to defective instances of that which it is a capacity for, and another thing to say that the exercise of a capacity can give rise to something totally different.

      In a bit more detail: Were I to have a capacity to throw a basketball through a hoop, perhaps that very capacity could yield something short of my doing so: a failed attempt of some kind. Similarly, perhaps my capacity for seeing can give rise to visual illusions, or hallucinations. In contrast, I’m not sure I can make sense of the idea of a capacity for throwing basketballs through hoops yielding scoring a goal in a game of football: doing so would not be a defective instance of what the capacity was for, but another activity altogether. I don’t see how a capacity that yielded goal-scorings could be a capacity for throwing basketballs through hoops. Likewise, a perceiving of flavour doesn’t seem to be a *defective* instance of that which an olfactory capacity is for. So, unless flavour perceivings are olfactory perceivings, I don’t see how a capacity that yielded perceivings of flavour could be an olfactory capacity.

  17. Hi all,

    In keeping with my original plan (and to free Louise from the very hard work of addressing all your feedback!), I’m going to close the public comments on this symposium. Many thanks to Louise for her willingness to be the target of the first of these symposia, and also to everyone else who participated, especially Fiona, Mohan, and Matt.

    Stay tuned for future events …