smellingAn intriguing paper from Benjamin D. Young claims that we can have phenomenal experiences of which we are unaware – although experiences of which we are aware always have phenomenal content. The paper is about smell, though I don’t really see why similar considerations shouldn’t apply to other senses.

At first sight the idea of phenomenal experience of which we are unaware seems like a contradiction in terms. Phenomenal experience is the subjective aspect of consciousness, isn’t it? How could an aspect of consciousness exist without consciousness itself? Young rightly says that it is well established that things we only register subconsciously can affect our behaviour – but that can’t include the sort of experience which for some people is the real essence of consciousness, can it?

The only way I can imagine subjectivity going on in my head without me experiencing it is if someone else were experiencing it – not a matter of me experiencing things subconsciously, but of my subconscious being a real separate entity, or perhaps of it all going on in the mind of alternate personality of the kind that seems to occur is Dissociative Identity Disorder (Multiple Personality, as it used to be called).

On further reflection, I don’t think that’s the kind of thing Young meant at all: I think instead he is drawing a distinction between explicit and inexplicit awareness. So his point is that I can experience qualia without having any accompanying conscious thought about those qualia or the experience.

That’s true and an important point. One reason qualia seem so slippery, I think, is that discussion is always in second order terms: we exchange reports of qualia. But because the things themselves are irredeemably first order they have a way of disappearing from the discussion, leaving us talking about their effable accompaniments.

Ironically, something like that may have happened in Young’s paper, as he goes on to discuss experiments which allegedly shed light on subjective experience. Smell is a complex phenomenon of course; compared with the neat structure of colours the rambling and apparently inexhaustible structure of smell space is daunting;y hard to grasp. However, smell conveniently has valence in a way that colours don’t: some smells are nice and some are nasty. Humans apparently vary their sniff rate partly in response to a smell’s valence and Young thinks that this provides an objective, measurable way into the subjectivity of the experience.

Beyond that he goes on to consider mating choice: it seems human beings, like other mammals, choose their mates partly on the basis of smell. I imagine this might be controversial to some, and some of the research Young quotes sounds amusingly naive. In answer to a questionnaire, female subjects rated body odour as an important factor in selecting a sexual partner; well yes, if a guy smells you’re maybe not going to date him, huh?

I haven’t read the study which was doubtless on a much more sophisticated level, and Young cites a whole wealth of other interesting papers. The problem is that while this is all fascinating psychologically, none of it can properly bear on the philosophical issue because qualia, the ultimate bearers of subjectivity, are acausal and cannot affect our behaviour. This is shown clearly by the zombie twin argument: my zombie twin has no qualia but his behaviour is ex hypothesi the same as mine.

Still, the use of valence as a way in is interesting. The normal philosophical argument is that we have no way of telling whether my subjective red is your subjective green: but it’s hard to argue that m subjective nasty is your subjective nice (unless we also hypothesise that you seek out nasty experiences and avoid nice ones?).

21 Comments

  1. 1. Mike Cox says:

    Peter, I don’t think the Zombie twin argument is at all conclusive. Do you not think for example, that an appreciation of literature,ie the enjoyment of a novel, requires consciousness and assumes relatively equivalent qualia in all involved parties.Readers and writer.

  2. 2. Sergio Graziosi says:

    Peter,
    once more, you seem to accept the p-zombie argument, and apparently espouse the view that therefore qualia can’t be causal. It took me some time, but I’ve found a comment you made, clarifying that you do see problems with the whole idea.

    You know that my appreciation for your efforts here is virtually unbounded (the highest possible), but (or therefore), I’m getting more and more puzzled by the number of times you state things like “qualia, the ultimate bearers of subjectivity, are acausal and cannot affect our behaviour”.
    Now, on the specific content: I agree that Young’s paper will not convince Chalmers and those similarly inclined, his efforts fall squarely in what you describe as the scientific/engineering approach. Yes, it will be easily brushed aside as inconsequential by those philosophers who happen to already believe there must be something more than matter as we know it. I am hoping that this is what you mean in the current post…

    Still more or less on the subject, Massimo Pigliucci has tackled p-zombies once more, arguing that “p-zombies are inconceivable“, useful also as a reminder of how many times the whole argument has been shredded into pieces (from many different angles). This time, Pigliucci focuses on the idea of “metaphysically possible”, and I’m guessing you’d enjoy reading his contribution.

    Should we take (acausal) qualia, p-zobies and the “hard problem” seriously? Of course! Should we accept this trinity (and its dualistic consequences) a-critically? Well, you know that I see the latter as a rhetorical question! ;-)

    Also: there is clear, readily available (and subjectively verifiable) evidence that qualia (in the sense of phenomenal experience) are causal. Say that I fling something at you in such a way that it will enter your visual field before hitting you, but it’s small and you still might not notice it. We try this a few times, you’ll (attempt to) dodge my projectile if and only if you’ll see it consciously, right? Why is this not enough to confirm the causality of conscious experience is beyond me.
    [If you don't like the idea of being used as a target(!), there are a gazillion studies that used masked or close-to-the-threshold stimuli, and I believe they all confirm the same causal relation]
    Sometimes you’ll even dodge before knowing what’s happening (or be ‘primed’ by sub-threshold stimuli), that’s interesting, isn’t it? And ties-in well with the current topic: “explicit and inexplicit awareness”, there IS a difference, isn’t there?
    (with apologies for my obvious partisanship!)

  3. 3. Tom Clark says:

    Sergio:

    “Also: there is clear, readily available (and subjectively verifiable) evidence that qualia (in the sense of phenomenal experience) are causal. Say that I fling something at you in such a way that it will enter your visual field before hitting you, but it’s small and you still might not notice it. We try this a few times, you’ll (attempt to) dodge my projectile if and only if you’ll see it consciously, right? Why is this not enough to confirm the causality of conscious experience is beyond me.”

    The causal story that’s available from the 3rd person perspective is that it’s the neural processes associated with being conscious of the projectile that are responsible for dodging, not experiential qualia. Of course, if the qualia and the processes are identical, then qualia inherit all the causal powers of their neural correlates. But that identity hasn’t been established.

    From our perspective on the world (including our own behavior) as conscious subjects, we can’t help but take our experience of the approaching object as the cause of our dodging: if I hadn’t seen it, I wouldn’t have dodged. This lines up very reliably with the 3rd person causal story involving neural processes. But I think these are parallel, non-interacting accounts of behavior, such that qualia don’t add causally to what neurons are doing. Rather they constitute a qualitative representational account of goings-on for the subject that can be communicated in ordinary language to others.

    Since qualia don’t appear in 3rd person public space (experience isn’t observable, unlike its physical correlates), they aren’t in a position to be epiphenomenal or a-causal. Instead I’d suggest they are more accurately conceived of as causally orthogonal.

  4. 4. Mike Cox says:

    I would argue that Peter’s blog could not have been produced by his p zombie twin and indeed was largely a product of his p conciousness
    and it’s fascination for him.

  5. 5. Sergio Graziosi says:

    Tom:
    thanks, this forced me to clarify my thoughts. We don’t fundamentally disagree, but the details are still sketchy, and may actually divide us.

    I’ll summarise my hunch here (sketchy alert!): on the first level, inexplicit/pre-conscious perceptions are causal; leaving the projectile case aside, it happens a lot with driving and/or in my case riding on two wheels, we sometimes react to something before being aware of what happened (specifically, I’ve somehow managed to retain control even when the front wheel lost adherence, and more than once). Personally, on many occasions I reacted before knowing what I was doing and why, therefore there is a dissociation between full, self aware perception (the second order that Peter refers to) and some sorts of causal perception (some stimuli can “cause” a complex reaction before reaching full consciousness, and still become conscious nevertheless).

    Second: I think it’s reasonable to expect that full, self aware perception is needed to learn from experience (for example, learn how to play chess), making standard qualia (second order?) also “causal” of something much more complex than ‘simple’ motor-reactions.

    Yes, you may say, but causal how?
    I’m not sure I understand your “orthogonal” claim, my own position is that we are talking about the same thing, but using two separate (parallel or orthogonal, I wouldn’t stretch the metaphor!) epistemological reference points. Qualia from within, and neural mechanisms from a third person point of view need in the end to be the same thing. For sure, we can’t demonstrate that this identity is the case, and for good reason, as the physiological explanations we have are not even close to sketchy; but I still found nothing in the qualia/p-zombie/hard problem camp that is able to dismiss this expectation in a convincing way.
    I see their argument as “yes, but we have a set of alternative (and fantastically problematic) expectations, how about them?” at best, and I’m OK with it: it’s perfectly reasonable to discuss about these other possibilities as well.

    Mike:
    :-) I agree with enthusiasm! I guess you are familiar with Dennett’s Zimbo argument: you managed to replicate it in one sentence, I am impressed.

    PS All: I rarely participate here because I’m afraid I might be unable to follow up adequately. If I’ll fail to respond, please accept my apologies, I might just be too busy.

  6. 6. john davey says:

    Sergio

    Referring to good old projectiles won’t help you establish that qualia are causal. The only way to establish that qualia are causal is to use them as a behavioural reference – e.g I’m hungry (feeling), so I eat. I feel the need to drink/sleep/defaecate etc, all feelings with unique shapes and which condition subsequent activity.

    John

  7. 7. john davey says:

    “The normal philosophical argument is that we have no way of telling whether my subjective red is your subjective green”

    To which the obvious biological argument is : why would they be substantially different when the machinery that produces them is much the same ? In fact we know when the relatively minor differences occur : colour blindness is not exactly a new disorder.

    Whether or not a biological object (such as a brain) produces a biological output (such as a colour sensation) that is uniform between humans is a not a question of logic and is certainly not a philosophical question. It is a scientific question. The belief that questions like this are philosophical quandaries is just plain wrong.

  8. 8. Tom Clark says:

    Sergio:

    “Qualia from within, and neural mechanisms from a third person point of view need in the end to be the same thing.”

    Well, it’s the great hope of physicalism to show that this is the case, but it needn’t *be* the case. That is, there is no a priori metaphysical necessity that this identity holds, only an understandable desire for unification. There are naturalistic alternatives to physicalism, e.g., psycho-physical parallelism, that imo do a better job of respecting the subjective reality of experience and the current neuroscientific evidence connecting consciousness to the representational functions of its associated brain processes.

  9. 9. Sergio Graziosi says:

    Tom and John,
    I’ve obviously allowed myself to get carried away. Yes, yet another quick and raw thought experiment won’t help establishing one view or the other. I was doing two things (always a bad idea!), hastily as well:
    1) clarifying to myself the issue with pre-conscious perceptions, from an angle I hadn’t explored so far.
    2) grinding my own axe on the plausibility of identity and physicalism. The great hope needs to (will inevitably) remain a hope until we’ll have a more complete picture of the supposedly “easy” problems. After that, I expect the game to change, one way or the other.

    I haven’t found “current neuroscientific evidence connecting consciousness to the representational functions of its associated brain processes”, not evidence that fully convinced me (and I do hope to find it, so I set my evidence bar quite high, to compensate for the biasing drive of my own wishes); IOW, I can’t wait for someone to propose a detailed model on how representational functions really work, but I don’t think we’re there (IIT got close, but its core is theoretical, not empirical, and has both gaps and flaws to my eyes), if you do have some references that you find compelling, please do share!

    Thanks to all,
    Sergio

  10. 10. Mike.cox says:

    I would also strongly suggest that p-consciousness is a necessary condition of the (in my view erroneous) concept of free will. A concept that whatever it’s validity,has had,and continues to have a huge causal impact on human behaviour and experience.

  11. 11. Peter says:

    Mike,
    No, I agree – that is, I don’t believe the zombie argument is actually correct. But I take it to be sort of authoritative in establishing what the philosophical conception of qualia is.
    It seems we might be developing a parallel psychological conception, as has happened width ‘epiphenomenon’, where the psychological conception is less rigorous but as a result is actually usable rather than leading directly into insoluble difficulty…

  12. 12. Peter says:

    Sergio,

    “you seem to accept the p-zombie argument”

    I accept it as one of the key arguments that set out the problem of qualia as generally understood, I don’t accept that it’s completely valid. For what it’s worth I don’t believe anything real (by which I mean non-abstract) can lack causal powers. The thing is, if we simply reconceive qualia as causally unproblematic entities within the sensory system we evade the Hard Problem rather than solving it

  13. 13. Sergio Graziosi says:

    Thanks Peter,
    we agree on this: one has to address the qualia/p-zombie/hard problem trinity. I don’t agree that the p-zombie thought experiment demonstrates that the hard problem is indeed as hard as it’s claimed, but that’s another story.
    Enjoy,
    Sergio

  14. 14. Tom Clark says:

    Sergio:

    “IOW, I can’t wait for someone to propose a detailed model on how representational functions really work, but I don’t think we’re there (IIT got close, but its core is theoretical, not empirical, and has both gaps and flaws to my eyes), if you do have some references that you find compelling, please do share!”

    Agreed we’ve got a long way to go, but I think it’s pretty clear that consciousness is associated with higher-level, recursive and nested representations that enable flexible behavior and learning in a changing environment, as opposed to reflex and programmed behavioral routines. See for instance Dehaene, S., Changeux, J.P. Experimental and theoretical approaches to conscious processing, Neuron, 2011,70, 200-27, which is available online. In the previous post here (Quanta and Qualia) I suggest why the logic of being a representational system at our level might entail the existence of qualia for the system, see replies 8, 10 and 13. Click on my name for some papers on consciousness.

  15. 15. Sergio Graziosi says:

    Agreed we’ve got a long way to go, but I think it’s pretty clear that consciousness is associated with higher-level, recursive and nested representations that enable flexible behavior and learning in a changing environment, as opposed to reflex and programmed behavioral routines.

    Tom, you have just produced a mini-abstract of the theoretical paper that I’ve submitted for PR a week ago. I’ll keep my fingers crossed, although you gave me one more reason to regret not citing Dehaene and Changeux (I love their work, but due to words limits I couldn’t even try to discuss it in full, and just throwing it in didn’t feel right).

  16. 16. Tom Clark says:

    Cool, good luck Sergio! Would of course love to see your paper sometime.

  17. 17. Sergio Graziosi says:

    Thanks Tom, I hope you’ll be able to read it via the official channels; I can’t avoid thinking it has decent publication chances, a very unwise & unsubstantiated attitude :-/.

  18. 18. Kevin Z. says:

    Peter: I’m not sure I follow how it could be possible for qualia to be acausal. If they were acausal, how could we even be talking about them? It seems to me that, if you and I experience qualia and then talk about it, those qualia have affected our (objectively visible) behavior. Can you tell me what I’m missing?

  19. 19. Peter says:

    It is pretty strange, Kevin, and some would say it’s just nonsense. But the claim is that qualia are over and above the physics (otherwise Mary the Colour Scientist would know about them, and my zombie twin would experience them).

    Although it sounds odd, we can talk about things that have no causal powers: for example I can talk about purple unicorns, which have never caused anything (because they don’t exist). More helpfully perhaps, I can talk about aspects of an experience. So the red colour does all the actual causing; its participation in a subjective experience doesn’t add any new causal powers but it’s a different aspect which I can talk about and which isn’t part of the account given by physics.

    Or something like that!

    This is what I thought ten years ago; not very well expressed and I might be a little less sceptical now, but I think I was groping in the right sort of place.

  20. 20. john davey says:

    “But the claim is that qualia are over and above the physics”

    I think it is very important to remember what physics is. Physics is an academic mathematical modelling discipline. It is not “more real” than reality, merely an attempt to model in syntactic terms the very semantic universe about us. To that extent, everything that is in the universe is “beyond physics”, as it is more than just a symbol in a mathematical system, it is the thing the actual symbol represents. Space and time are just scalars in physics, but in the universe they are far more than that. They are stuff, real stuff, that happens to have interesting mathematical properties and a comfortable linearity.

    But time and space are more than numbers, and physics cannot hope to fully represent them, except in a formal mathematical sense. Time and space remain mostly mysteries and we are incapable of more than the mathematical sledgehammer to try to split them up a bit for a better view. The semantic question – what is time , or what is space – physics cannot answer any more than religion or philosophy. The difference with physics is it can predict how time and space react with each other with astonishing accuracy – undermining those same religious explanations.

    So qualia are no different – they are no more “beyond physics” than time or space, but unlike time or space have not surrendered to mathematical analysis. If they do, they will then join everything else in the universe currently modelled by physics and – not to put too fine a point on it – not in the least bit fully understood as a consequence.

  21. 21. google api tutorial says:

    This blog was… how do you say it? Relevant!! Finally I’ve found something which helped me.
    Thank you!

Leave a Reply