Architectonics and the Hard Problem

twinsCan we solve the Hard Problem with scanners? This article by Brit Brogaard and Dimitria E. Gatzia argues that recent advances in neuroimaging techniques, combined with the architectonic approach advocated by Fingelkurts and Fingelkurts, open the way to real advances.

But surely it’s impossible for physical techniques to shed any light on the Hard Problem? The whole point is that it is over and above any account which could be given by physics. In the Zombie Twin though experiment I have a physically identical twin who has no subjective experience. His brain handles information just the way mine does, but when he registers the colour red, it’s just data; he doesn’t experience real redness. If you think that is conceivable, then you believe in qualia, the subjective extra part of experience. But how could qualia be explained by neuroimaging; my zombie twin’s scans are exactly the same as mine, yet he has no qualia at all?

This, I think, is where the architectonics come in. The foundational axiom of the approach, as I understand it, is that the functional structure of phenomenal experience corresponds to dynamic structure within brain activity; the operational architectonics provide the bridge . (I call it an axiom, but I think the Fingelkurts twins would say that empirical research already provides support for a nested hierarchical structure which bridges the explanatory gap. They seem to take the view that operational architectonics uses a structured electrical field, which on the one hand links their view with the theories of Johnjoe McFadden and Sue Pockett, while on the other making me wonder whether advances in neuroimaging are relevant if the exciting stuff is happening outside the neurons.) It follows that investigating dynamic activity structures in the brain can tell us about the structure of phenomenal, subjective experience. That seems reasonable. After all, we might argue, qualia may be mysterious, but we know they are related to physical events; the experience of redness goes with the existence of red things in the physical world (with due allowance for complications). Why can’t we assume that subjective experience also goes with certain structured kinds of brain activity?

Two points must be made immediately. The first is that the hunt for Neural Correlates of Consciousness (NCCs) is hardly new. The advocates of architectonics, however, say that approaches along these lines fail because correlation is simply too weak a connection. Noticing that experience x and activation in region y correlate doesn’t really take us anywhere. They aim for something much harder-edged and more specific, with structured features of brain activity matched directly back to structures in an analysis of phenomenal experience (some of the papers use the framework of Revonsuo, though architectonics in general is not committed to any specific approach).

The second point is that this is not a sceptical or reductive project. I think many sceptics about qualia would be more than happy with the idea of exploring subjective experience in relation to brain structure; but someone like Dan Dennett would look to the brain structures to fully explain all the features of experience; to explain them away, in fact, so that it was clear that brain activity was in the end all we were dealing with and we could stop talking about ‘nonsensical’ qualia altogether.

By contrast the architectonic approach allows philosophers to retain the ultimate mystery; it just seeks to push the boundaries of science a bit further out into the territory of subjective experience. Perhaps Paul Churchland’s interesting paper about chimerical colours which we discussed a while ago provides a comparable case if not strictly an example.

Churchland points out that we can find the colours we experience mapped out in the neuronal structures of the brain; but interestingly the colour space defined in the brain is slightly more comprehensive than the one we actually encounter in real life. Our brains have reserved spaces for colours that do not exist, as it were. However, using a technique he describes we can experience these ‘chimerical’ colours, such as ‘dark yellow’ in the form of an afterglow. So here you experience for the first time a dark yellow quale, as predicted and delivered by neurology. Churchland would argue this shows rather convincingly that position in your brain’s colour space is essentially all there is to the subjective experience of colour. I think a follower of architectonics would commend the research for elucidating structural features of experience but hold that there was still a residual mystery about what dark yellow qualia really are in themselves, one that can only be addressed by philosophy.

It all seems like a clever and promising take on the subject to me; I do have two reservations. The first is a pessimistic doubt about whether it will ever really be possible to deliver much. The sort of finding reported by Churchland is the exception more than the rule. Vision and hearing offer some unusual scope because they both depend on wave media which impose certain interesting structural qualities; the orderly spectrum and musical scale. Imaginatively I find it hard to think of other aspects of phenomenal experience that seem to be good candidates for structural analysis. I could be radically wrong about this and I hope I am.

The other thing is, I still find it a bit hard to get past my zombie twin; if phenomenal experience matches up with the structure of brain activity perfectly, how come he is without qualia? The sceptics and the qualophiles both have pretty clear answers; either there just are no qualia anyway or they are outside the scope of physics. Now if we take the architectonic view, we could argue that just as the presence of red objects is not sufficient for there to be red qualia, so perhaps the existence of the right brain patterns isn’t sufficient either; though the red objects and the relevant brain activity do a lot to explain the experience. But if the right brain activity isn’t sufficient, what’s the missing ingredient? It feels (I put it no higher) as if there ought to be an explanation; but perhaps that’s just where we leave the job for the philosophers?

47 thoughts on “Architectonics and the Hard Problem

  1. At the very least it’s good someone is trying to something novel on the empirical side.

    Seems like a good idea to solve the easy part of the Hard Problem before deciding anything about the fundamental nature (or not) of consciousness.

  2. My issue with the idea of philosophical zombies, at least the variety whose physical brain is identical to that of a conscious twin, is that the premise begs the question, inherently assuming some form of substance dualism (as well as epiphenomenalism). The idea of zombies wouldn’t make me doubt this endeavor.

    I have no idea if this approach will work, but even if someone is a skeptic about qualia (I have to admit I am, at least toward any ghostly version), that doesn’t mean there shouldn’t be some physical correlate for what causes us to have the feeling or intuition of qualia.


  3. But how could qualia be explained by neuroimaging; my zombie twin’s scans are exactly the same as mine, yet he has no qualia at all?

    I would suggest that probably because contemporary analytical techniques like “neuroimaging” are probably as relevant to minds as the use of leeches was to 18th century medicine.

    JBD

  4. Do these ‘zombie’ examples ever have the zombie verbally claiming to have experience, qualia and real redness anyway? Or does the zombie somehow know it’s a zombie – so that raises an interesting question about the example. How does it know that?

    Seems more likely it wouldn’t know.

  5. The zombie twin, ex hypothesi, behaves exactly like me, so assuming for the sake of argument that I believe in qualia he claims to have them and he talks and writes about them exactly as I do. There cannot be any physical evidence of him knowing he is a zombie, though maybe he discusses the possibility in a light-hearted way…

    That seems odd, but then nothing I ever say or write was caused by qualia anyway because they have no causal powers.

    This is the sort of stuff Dennett points to in order to suggest the whole thing is hopeless nonsense.

  6. Peter: “nothing I ever say or write was caused by qualia anyway because they have no causal powers.”

    Well, if qualia are identical with their neural correlates, then they have causal powers. I’m curious if you have a theory of qualia that would rule out the identity claim, of which I’m skeptical myself.

  7. We’re all agreed that we hate brain/computer analogies but this remark from the Brogaard/Gatzia article:

    “On this view, the cortical system can perform many operations, involving large-scale local and global cortical networks, characterized by different spatial and temporal parameters, using one or more operational modules. These operational modules are not directly connected with structural or static functional modules in the brain. Rather, they have a dynamic nature – they are metastable – and exist in their own operational space-time, i.e., an abstract space and time the brain constructs each time an operational module emerges”

    does sound a bit like a computer running a virtual machine. Of course if we claim that “operational modules are not directly connected with structural or static functional modules in the brain” we have to ask if they are (A) indirectly connected or (B) unconnected. Because we are not doing theology we assume (A) there is an indirect connection. What mediates that indirect connection between the hardware and the virtual machine? Our hated brain/computer analogy suggests human brains run some sort of operating system on the hardware and the virtual machines are programs running within the operating system. A quale, such as the experience of perceiving the redness of an object, is a program running on the virtual machine, which in turn is running on the base operating system which is in turn running on the hardware.

    To put it another way, qualia are sensory and other percepts reformatted for use by the conscious mind. What then is the conscious mind? What is consciousness? A simple answer is consciousness is the suite of capabilities we lose when we go under general anesthesia. We lose the ability to perceive our own bodies. We lose the ability to form or access memories. We lose the ability to perceive the external environment using our senses. We lose the ability to integrate those sources of information, so as (for example) to see a person and remember her name. I could list more, but I hope those suffice to make my point, which is that consciousness is just s suite of capabilities. Seen this way, there does not seem to me to be anything about consciousness that requires a qualitatively different kind of explanation than does respiration or digestion.

  8. For me, the way out of the zombie dilemma is this:

    Zombies as usually defined are not possible/conceivable in the actual world we live in… because of the logical problems that result in the conclusion that qualia can have no causal powers.

    However, the world described by physics is not the actual world–it’s a very good model for understanding many aspects of the world. I can believe in qualia by postulating that there is a missing dimension not included in the usual physical model of the world.

    So it’s not that zombies are conceivable in this world. It’s that describing reality in purely physical terms leaves out something: in effect, physics is a zombie model of the world.

    Analogy: Suppose someone claims it’s conceivable there could be a creature just like us in all respects, except it contains 2 dimensions instead of 3. After all, the people we see on the TV screen certainly appear to be real! But this is of course absurd: they are actually 2D representations of real people.

    Likewise, the “zombies” described in the philosophic literature cannot be real; they can only be representations of humans that exist in a model in which there is no qualia dimension. The real world has such a dimension… we just haven’t figured out what it is or how it works 🙂

  9. Yeah – it’s probably more the absence of experience that makes experience compelling. People talk about experiencing the red of a rose and how do you explain that experience? But if you imagine it in terms of A: Experiencing a rose and B: Experiencing A (experiencing that experiencing) – but the key component is that there is no C: Experiencing B.

    Without that there is no internal mapping a recursion pattern – ‘A’ seems inexplicable because the observation in progress that is B is invisible for lack of C. Things ‘just are’. It’s kind of like how we don’t question our view into a TV episode. Things just are. But on the rare occasion a TV show pans back and shows (the first) cameraman, suddenly ‘just is’ is less compelling.

    Absent C, it seems like A is the ‘hard question’ to explain. Because of how invisible B is.

  10. Callan


    Do these ‘zombie’ examples ever have the zombie verbally claiming to have experience, qualia and real redness anyway?

    You hit the nail on the head .. a true zombie, for whom consciousness is not cognitively possible, would never talk about consciousness. Any more than a dog would talk about mathematics.

    But physically identical means causally identical. So two beings constructed of the same stuff will always have the same properties, including consciousness, assuming (as seems likely) brains cause consciousness. So zombies on the model of this thought experiment don’t even pass the test of being a legitimate thought experiment because the premises are obviously false – namely that identical objects have can have different properties.

    J

  11. Peter

    “That seems odd, but then nothing I ever say or write was caused by qualia anyway because they have no causal powers.”

    This seems to be based upon the assumption that the current partial account of the universe proposed by physics (that part that excludes mental phenomena) trumps all other potential accounts, including those accounts that may arise in the future. Would it not be more sensible to just say “there is no physical account of mental phenomena currently” ?

    J

  12. I wrote one long response to this, but thought it would be TL:DR. This long response probably is as well. And too at odds with what everyone else is saying to find a welcome. But sometimes I can’t help myself. Sorry.

    If zombies don’t exist, then the thought experiment doesn’t tell us anything. What would a real world example that made the same point look like? Occam counselled against inventing entities to make our explanations work. I still think this is sage advice.

    “Qualia”, as Searle says, is just a fancy word for conscious states. Using it has the unfortunate consequence that we appear to have two things to explain, i.e. conscious states and qualia, when in fact we only have one thing to explain, i.e. conscious states with qualitative features.

    Causality is a metaphysical concept and it puzzles me that philosophers keep trying to build ontologies on the basis of metaphysics. Its what I expect from a theologian.

    The Hard Problem is an extension of the mind-body problem which emerges (Searle again) from the implicit dualism of scientific materialism. A truly monistic worldview eliminates the mind-body problem; and the issue of causality (if we have not yet weeded it out of our narrative) becomes irrelevant. In an monistic ontology there is no special kind of causality needed, because there is only *one kind of stuff*.

    If we are still dividing the world into mental and physical, this is Cartesian. Even if we claim that only one or other of the two is real, the division precedes the truth-claim and the division *is dualism*. Almost all the comments I see here seem to be assuming a Cartesian Dualism with respect to mind and body. I think the Fingelkurts subtly do this also.

    If you are still wondering how the mind and body interact, then you haven’t abandoned Descartes – meaning you are about 200 years out of date. But if you still define reality *through* causality, if you are looking for a causal relation between brain and mind, then you have another Descartes sized problem, because there is no such thing as causality. Look through the laws of physics and point out the equation for causality; or point out the expression within any equation which stands for causation. They don’t exist. Kant did not make causality *real*, he made it manageable. Why do so many philosophers (including Searle) treat causality as real? I think I know, but this comment is already way too long!

    Fairly sophisticated thinkers talking about causality with respect to mind and body, or mental and physical phenomena as though there is an ontological difference, seems frankly weird in 2016.

    We’re left with an epistemological problem which is the same epistemological problem as this: the nutrients from the food I digest are only available to nourish me; they cannot nourish anyone else. Nutrition is a localised, first-person phenomenon; only I can feel what it is like to be nourished by the food I eat. We could make this into a Hard Problem by asking how a “nutrient” can possibly “nourish” me. Nourishment can only make sense relative to the individual and can only be experienced by individuals. There is something that it is like to be nourished by a nutrient – so nourishment a qualitative aspect – nutro-qualia. What is it about nutrients that make them nourishing? Blah blah. It’s a language game.

    Any internal biological process is susceptible to this treatment if we are only willing to treat it as an abstraction, and insist that the abstraction requires a special explanation. As I read Chalmers article on the Hard Problem, he was saying that particular conscious states (hunger, fear etc) are *not* the Hard Problem. In fact these are easily explained by neurobiological processes. The Hard Problem revolves around the abstraction conscious-ness. The Hard Problem of digestion is nourishment; the Hard Problem of the circulatory system is circulation; and so on. No abstractions can be explained in terms of concrete examples of the type. I cannot explain fruit by showing you an apple and a orange. I cannot understand the abstract notion of fruit by studying certain fruit in isolation. I have to understand the abstraction qua abstraction; and as such the abstraction will always be over and above any concrete example. There is something that it is like to be a fruit. Probably I end up taking a Wittgensteinian “family resemblance” approach to defining a fruit, probably via George Lakoff’s work on categorisation. Has anyone done this with consciousness?

    So no, we’ll never explain the abstraction “consciousness” purely in terms of concrete examples of the type “conscious state”. That’s not how abstractions work. Given this, what use is the abstraction “consciousness”? It’s vaguely defined, constantly erroneously reified, misunderstood, and the source of unending argument. Seems to me that we’re fixated on this particular abstraction purely as a legacy concept with deep roots in Christian theology. And just as we seem reluctant to finally abandon Descartes, we seem to have some reluctant to abandon theology too.

    If there is a distinction between mental phenomena and physical phenomena (and I do not admit that there is) then it is a matter how we come to *know* about them. It’s analogous to the distinction between vision and hearing. As far as I know, no one is proposing that we require a distinct ontology to account for these two sensual modes: both are manifestations of the one ontology, though we know about them and describe them in different ways. Sound and light are emergent manifestations of one kind of stuff, structured in different ways.

    The article cited has crossed out of philosophy and into science. We have a perfectly good philosophy of mind. Mind is an emergent property of processes in the brain; conscious states are subjective, qualitative etc. There is no other plausible philosophical position. However, we cannot say mind is caused by brain, because causation is a metaphysical overlay on perception, not a feature of reality. There is a simply a patterned evolution of things over time.

    A fully developed science of how consciousness emerges is something else again. We’re really just at the beginning of the quest to understand this and a long way from having any definite answers, though some promising leads have turned up. But we do know that the answers will be details of how consciousness emerges from the functioning of neurons and that causation won’t be a part of the explanation. So in the meantime it makes sense to study neurons, groups of neurons, and brains, and first-person reports of consciousness. Posing hypotheses and knocking them down until we work it out. For now, we simply do not know.

  13. the idea that mental beings can emerge out of complex processes in the brain is fundamentally flawed and can never be tested. If material processes operating through known physical laws generate as a side effect sentient minds how can we ever measure the existence of those mental beings? Mental beings are private and can never be directly observed and are only inferred from their behavior, but according to emergence they have no effect on behavior since all their effects are really the byproduct of brain processes. Therefore conscious entities which merely pop into existence as epiphenomenal side effects of the brain software disappear into nothing since they cannot be measured, observed or inferred in any way. This is eliminative materialism whereby we achieve consistency by throwing away the introspective evidence.
    There is to be fair another form of emergence which aims to show that consciousness is actually efficacious and that it at least has some unique influence on motor action. Since pain and pleasure evolved it must have an evolutionary purpose. John Searle calls this emergence 2 as he says below:

    “This conception of causal emergence, call it “emergent 1,” has to be distinguished from a much more adventurous conception, call it “emergent2.” A feature F is emergent2 if F is emergent 1 and F has causal powers that cannot be explained by the causal interactions of a, b, c…If consciousness were emergent 2, then consciousness could cause things that could not be explained by the causal behaviour of the neurons. The naive idea here is that consciousness gets squirted out by the behaviour of the neurons in the brain, but once it has been squirted out, it then has a life of its own. […O]n my view consciousness is emergent1, but not emergent2. In fact, I cannot think of anything that is emergent2, and it seems unlikely that we will be able to find any features that are emergent2, because the existence of any such features would seem to violate even the weakest principle of the transitivity of causation”

    I have to agree with Searle here. Emergent 2 is a worthy goal because it gives real causal power to qualia such as pleasure and pain.but it makes no sense. If the second to second reverberating of neural impulses is responsible for the squirting out of the conscious self and its ongoing maintenance then surely it is also responsible for every decision made by it.

    Is there any choice but eliminative materialism, one which can have the efficacy of consciousness and qualia and still be logically conistent? The only alternative is dualistic. The subjective self is a fundamental enduring non-emergent entity which experiences the qualitative virtual reality content created by the brain and can interact with the brain. I discuss in this paper. http://philpapers.org/rec/SLETLO-2
    Only subjective agents that exist with the same fundamental elemental status as say an electron can truly be efficacious and therefore have effects that can be measured and known. This implies that the known physical laws are not complete. However, with what is known about the probabilistic nature of quantum physics it is certainly possible that real mental entities can exert an independent force thereby influencing the probability distributions of matter.

  14. Lorenzo


    the idea that mental beings can emerge out of complex processes in the brain is fundamentally flawed and can never be tested. If material processes operating through known physical laws generate as a side effect sentient minds how can we ever measure the existence of those mental beings?

    same way was we measure everything else. Through theory and process.


    Mental beings are private and can never be directly observed

    Nothing is “directly” observed. Ever.


    Therefore conscious entities which merely pop into existence as epiphenomenal side effects of the brain software disappear into nothing since they cannot be measured, observed or inferred in any way.

    We’ll disregard the “epiphenominal” bit for now, but your comment is evidently wildly wrong. If I see a man snoring I can ‘infer’ he’s asleep. If I see a man with his brains blown out I can ‘infer’ he conscious states are inactive.

    As for “observation”, that is also evidently not true – when I converse with somebody I can “observe” their conscious mental states. I just can’t experience them. Which makes it like everything else I ever observe.

    As for “measure” that is also not true. Measurement = (theory + process ) -> number. Anaesthetists are doing it now.


    “This conception of causal emergence”

    I think emergence theories of consciousness are going nowhere. They are an attempt to square the reductionist circle. They clearly have a broad, qualitative appeal – bigger brains are more conscious than smaller ones, and its unlikely a single neuron will be conscious. But ultimately they are an attempt to construct a megastructure from a microstructure which is never going to work. (Searle even gives the example of surface tension which is very poor. Surface tension is quickly predictable, modellable and reducible in physics)


    This implies that the known physical laws are not complete.

    They are incomplete and always will be (as proof of the alternative is not available)

    JBD

  15. John

    you can infer a man is in pain when he pulls his hand away from a hot stove. can we infer the same for a fish? it seems to me that the pain actually causes my withdrawal reaction and I infer that in someone else the pain causes the withdrawal. But in any theory of emergence the pain is really a side effect and has no influence on the neural firings.

  16. Michael in #10:

    “…consciousness is just a suite of capabilities. Seen this way, there does not seem to me to be anything about consciousness that requires a qualitatively different kind of explanation than does respiration or digestion.”

    What needs explanation is why just this suite of capabilities, and not the capabilities carried out by unconscious processes, is accompanied by, or is identical to, conscious experience. If you can say why qualitative states arise when the neural processes associated with them are active, then you’ll have solved the hard problem.

  17. @Tom #9: I think the definition of qualia precludes their being identical with their neural correlates.

    I’m sort of a sceptic though, or rather I think that what people mistake for qualia is just the reality of actual experience. That in itself is somewhat mysterious, but as I may have said before, it isn’t the mystery people think.

  18. Peter in 20:

    “…I think that what people mistake for qualia is just the reality of actual experience.”

    So you’re saying we have experience but not qualia? I thought qualia were the basic, non-decomposable qualitative elements of experience, e.g., the redness of red. So if experience is real, so are qualia, the way I think of them.

    Then the question arises of whether qualia (and therefore experiences) are identical or not with various physical and/or functional goings-on. If they are, then experience has all the causal powers of those goings-on, thus avoiding epiphenomenalism. If they aren’t, then either epiphenomenalism threatens, or (avoiding epiphenomenalism) one can accept that there are parallel, causally non-interacting stories about behavior: one third-person, involving physically specifiable processes, and one subjective, involving conscious experience.

    http://www.naturalism.org/philosophy/consciousness/respecting-privacy

  19. Lorenzo


    can we infer the same for a fish?

    Yes. If fish had hands ..and similar neurological equipment to us. I kept fish so I don’t have a problem with the idea they experience pain.


    But in any theory of emergence the pain is really a side effect and has no influence on the neural firings.

    It sthis about emergence or is this about autonomous responses and reflexes ..ie the hand on the stove that short-circuits and doesn’t go via the brain ? Reflex response is rare but yes, pain does not affect it. It is a damage limitation exercise. Yes, the pain doesn’t move the hand but it might teach us a lesson or two about putting our hand near the stove again.

    J

  20. Peter


    I’m sort of a sceptic though, or rather I think that what people mistake for qualia is just the reality of actual experience.

    You can’t get away with that Peter .. can you unpack that, as Bryan Magee might say ? ..

    J

  21. Tom in #19

    What sort of explanation do you think is called for? If you assume that consciousness is a natural phenomenon you could be seeking an evolutionary sort of explanation about what species develop consciousness, what advantages cause consciousness to be favored by natural selection, what genetics encode the information needed to guide the construction of a brain capable of consciousness and so on. Or you could be seeking a neurological sort of explanation about what structures and dynamics govern the operation of consciousness. Or you could be seeking a logical sort of explanation about how the various information streams such as perception, memory, imagination etc. are integrated into an understanding about the past, present and future states of the world as it affects survival and reproductive fitness. I suspect there are other kinds of explanation I could suggest, but I think that both philosophically and scientifically we have to be in agreement about both what we mean by consciousness and what constitutes an explanation of it in order for progress to be made. I think a lot of what makes the “hard problem” seem hard is lack of this agreement.

    I like “the suite of capabilities we lose when we go under general anesthesia” as a definition of consciousness because I believe it should command at least grudging assent from most and because I believe it to be free of religious and philosophical baggage. I’m not sure what you mean by “qualitative states” but if you mean what commenters on this blog usually mean by qualia I’m not sure if there is a definition of qualia that commands assent sufficient to make the word a useful tool with which to think.

  22. Tom,

    How about qualitative states ‘arise’ from B: A neural process reacting to A: nerve information from external senses. But there is no C: A neural process identifying that B is occurring. So A seems to ‘happen by itself’. And thus is inexplicable – a hard problem.

  23. Peter

    I agree with you on many things in that reflection .. qualia, by being a noun, perhaps masks the probable reality that mental aspectual shape within a conscious field is a function and not a “thing”. but the boundaries between “subjective” and “objective” aren’t as clear cut as some might think.

    Space has a definite mental aspect, a definitiVe subjective shape. Visualisation of space is a ‘quale’. Yet space visualisation, whilst suffering from incommunicability like other quales, is viewed as ‘objective’ and indeed seems amenable to much quantitative analysis. like time. I dread reading Mr Dennett et al, but I wonder if the computationalists – whilst knocking “redness”- have given much thought to “spaceness” – which happens to be the basis of physics. Or, for that matter, “timeness”.

    J

  24. John Davey #17

    I agree with everything you say to Lorenzo up to this point,

    “I think emergence theories of consciousness are going nowhere.”

    Reductionism is clearly the best way to think about *substance* as demonstrated time and again by the physical sciences. Substance reductionism is wildly successful. Beyond any reasonable doubt we live in a world *made of* one kind of stuff. But reductionism is lousy at explaining the properties of complex objects, i.e. what that one kind of stuff is *made into*. Complex objects span ~100 orders of magnitude of mass, length, and energy. And reductionism fails to account for this.

    If we take a structure reductionist approach then we very quickly meet with failure. Structure reductionism cannot even explain the properties of molecules in terms of properties of atoms! It certainly does not explain conscious states. As you yourself point out, John, *neurons are not conscious*. Nor is any structural feature of the brain conscious in its own right. As soon as you start reducing the brain, consciousness *disappears*. Consciousness is irreducible. Most structures are, in the sense that as soon as you start to disassemble a structure, the complex object and its properties disappear.

    Some combination of substance reductionism and structure anti-reductionism is the *only* philosophical position that still provides us with options for making progress in a science of mind that includes consciousness. Of course some philosophers and scientists are happy to eliminate consciousness from their science of the mind, but that is another story. Even if our current science of mind is “unpromising”, this does not change the philosophical situation. If we acknowledge consciousness, then we have to also acknowledge that it is only a feature of a certain level of structural complexity. Consciousness is *irreducible*, because no less-complex structure has the property of having conscious states.

    However, if this analysis is correct, then we have a problem. Many neuroscientists are committed to metaphysical reductionism–the view that reductionism is the only viable methodology in science. This is obviously a hindrance to making scientific progress when studying an irreducible feature of reality. When the philosophy of mind *has* to be structure anti-reductive and scientists pursue a structure reductionist methodology, then *of course* progress will be uneven or non-existent.

  25. Jayarava

    I think I disagree. I thing many physical features are entirely reducible to snaller physical structures. One of the classic cases is the “wetness” of a body of liquid. This is cited by Searle as an example of reducibility where the aggregate of the superstructure (wetness) is not contained in a unit of the microsructure (e.g water molecule)

    Wetness is a property of a relatively small handful of molecules though. A few molecules don’t possess a wetness we could feel, but the intramolecular interactions that create liquidity are evident in a small collection of molecules : short range order, long range disorder, which makes liquids neither solid nor gas. Thus an analysis of collections of molecules predicts the property of liquidity.

    No amount of analysis of neurons – if such analysis is bound to current physics and chemistry – is ever going to yield consciousness though. That means there is a fundamental difference between consciousness and liquidity, namely that the current analytical toolsets of chemistry and biology cannot establish if the relationship between consciousness and neurons is reductive. They may be : but at this time it’s not possible to say. Certainly, as you say, consciousness is not semantically reducible to electrochemistry in any shape or form, unlike liquidity and water molecules. There may nonetheless be a weaker reductive link.

    J

  26. Michael in 24:

    “I’m not sure what you mean by “qualitative states” but if you mean what commenters on this blog usually mean by qualia I’m not sure if there is a definition of qualia that commands assent sufficient to make the word a useful tool with which to think.”

    Yes qualia are qualitative states like the sensation of pain, the look and feel of colors, tastes, smells, sounds, etc. – all the sensory experiences via which the world appears to us. Experience is generally taken to be what constitutes consciousness, not the physical or functional correlates of experience or the capacities associated with having it. If the latter were what’s meant by consciousness, then there would be no special difficulty about explaining it (and since you take it to be the latter, you see no special difficulty).

    The explanation called for by the hard problem is one that shows why it’s the case that the physical and functional goings-on associated with experience necessitate its existence for the system instantiating those goings-on. But if one doesn’t accept the prima facie distinction between experience and its physical-functional correlates and associated capacities, then of course there is no need for such an explanation.

    About the characteristics of qualitative states, see http://www.naturalism.org/philosophy/consciousness/killing-the-observer#toc–informational-characteristics-of-qualia-Rb1ifiQD

  27. John #29

    “I think I disagree. I thing many physical features are entirely reducible to snaller physical structures.” [sic]

    In order to properly disagree with me, you would need to show that all physical features are entirely reducible. If any features remain that are not, then you agree with me to some extent. We just need to figure out to what extent. I think you *completely* agree with me, but let’s work through it.

    We agree that *analysis*, the reductive methodology par excellence, is not sufficient for exploring consciousness. Thus you do seem to agree with me that reductionism cannot explain consciousness and that we must rely more on methods deriving from antireductionism.

    However, you do hedge your statement about analysis. It’s almost as if you think that reduction might work in principle, but that there are epistemic barriers in place. Reduction “in principle” is a very popular narrative in modern philosophy, though a very weak philosophical view. If I read your hedge right, this would make you some kind of epistemic antireductionist on a bedrock of ontic reductionism. Probably you would relate to the weaker forms of this view which argue for practical limitations due to the state of our knowledge or our ability to measure, rather than the stronger forms which argue for an intrinsic epistemic limit like “the universe doesn’t allow us to know”. In this case we are only agreed about the limitations of analysis as a method – you are more of a methodological antireductionist, without necessarily accepting that this limitation implies something about the epistemology or ontology of consciousness. I’m not convinced that this is a coherent position.

    We’ve left the question open of whether consciousness has features that can be explained by features of neurons. You think some features of some systems are reducible, but you cannot decide whether this applies to all features of all systems, and in particular you think we are not in a position to say regarding the reducibility of consciousness. However, as you pointed out in your last comment you go along with the majority in believing that no single neuron, nor any simple arrangement of neurons, nor any sub-system of the brain taken in isolation, possesses the feature or property of having conscious states. No single neuron is conscious. Conscious states only occur in the brains of certain animals and birds that are above a certain threshold: C. elegans with just 304 neurons is almost certainly not conscious. Homo sapiens with 85 billion neurons is conscious. Where the threshold is and how narrow it is remain to be seen.

    If you accept the way I’ve stated it, then you have effectively repudiated reductionism methodologically, epistemically, and ontologically. If no single neuron is conscious, and I don’t think you are disputing this, then the brain as a whole definitely has a feature that no single neuron has, i.e. it is conscious. No simple aggregate of non-conscious neurons can be conscious. And this repudiates reductionism as an explanation of consciousness, unless, like Dennett you are willing to skip the neuron and go for a sub-cellular explanation. Which as this stage is no better than hand waving. Do any reductive explanations allow us to ignore intermediate levels of description completely? I cannot think of any.

    If no single neuron possesses the features or properties of having conscious states, then reductionism has already failed to explain consciousness. Some explanation that involves structure antireductionism is already required, and all that is left is to negotiate which version of it is best. If as you state, “analysis of neurons… won’t yield consciousness”, then you have already arrived at the this same conclusion. Thus, you seem to completely agree with me, but to be hesitant to say so.

    The difference between a vat of 85 billion unconnected non-conscious neurons and a conscious brain is how the neurons are arranged and connected, then it is the arrangement itself (the structure) that makes the difference. The structure is a feature of reality in a structure antireductionist view. In other words we can safely move towards ontological structure antireductionism, because it is the only view that makes sense of what we currently know.

    I’m sure we both agree that the world is made up from quantum fields bundled into particles. So we our ontology has to combine substance reductionism with structure antireductionism. And this is the best of all possible views on ontology 🙂

  28. Jayarava: In reference to 15 above, I think for mind and dualism there is a mind conceit that philosophers have. I could show a layman four different parts of body organs : Liver, Kidney, Heart and Brain, and unless they were Mary The Organ Scientist who knew all of the facts etc., they would be clueless how the four distinct functions of the organs emerge: Bile, Renal, Pumping and Mind or each is simply a function that can be assigned a dualist nature by the layman. Better put the philosophers who lack knowledge of brain function are hopelessly stuck in this last remaining dualism.

    I would say that the bundle of neurons we call mind/consciousness is no different than the bundle of circuits a consumer calls an iPhone or “iPhonishness”. It’s complex and unless you are an engineer you would have no clue how to unbundle the functions and explain them down to the hardware and software levels.

  29. VicP #32

    Assigning a function to an organ is not a simple as you make it sound. A heart may pump blood, but when you say that its *function* is to pump blood then you are imposing something observer-relative on it. This is one of the examples Searle uses in The Construction of Social Reality, e.g.

    “Functions are never intrinsic to the physics of any phenomenon but are assigned from outside by conscious observers and users… when, in addition to saying ‘the heart pumps blood’ we say, ‘the *function* of the heart is to pump blood,’ we are doing something more than recording intrinsic facts” (1995: 14).

    Admittedly the heart is a difficult example to grasp, but in the context of the book it does become clear that this is the case. The importance of this distinction becomes more obvious as he builds up his philosophy of social reality. Social facts are all functions of this type, imposed on objects by declarative speech acts by authorised persons (i.e. people who have had a function imposed on them by the collective…).

    You seem to be saying that laypeople would naively assume a dualism between the organ and the function of the organ. This doesn’t ring true to me. The naive observer would be more likely to *identify* the object with the imposed function, at least in part because the function is tied into our values. There are plenty of metaphorical expressions, especially metonyms, in English that suggest naive commentators do just this. Or maybe you think my head is not screwed on right?

    Nor is this scenario the essence of Western mind/body dualism, which has its roots in Christian theology. Descartes formalises the long-held duality the way he does to try to account for two conflicting realities of his day: the apparent successes of the new mechanistic science (body as mechanism); and the need to leave open a door for God (mind as a function of the *soul* rather than the brain). An organ/function duality is observer relative; whereas the Cartesian duality is intrinsic. But Descartes was clearly wrong about this intrinsic dualism and so he is now deeply unfashionable, but Searle says he lurks everywhere in the background.

    It’s not laypeople we are concerned with in this critique, but the whole enterprise of Scientific Materialism – i.e. *professional* scientists and philosophers. Materialists divide the world into two ontological categories–mental and physical–and claim that only physical phenomena, and only those physical phenomena at the bottom of the hierarchy of description, are *real*. Everything else, any complex phenomenon or “mental” phenomenon, is *not real*. Eliminativists in particular insist that consciousness *does not exist*; that mind *is* matter. The distinction is claimed to be monistic, but in fact it fundamentally relies on the applying the Cartesian duality prior to applying the real/non-real judgement. Scientific Materialism is fundamentally dualistic!

    True monism does not allow the prior distinction. There *is* no ontological distinction between mental and physical phenomena (here I part ways with Searle’s way of putting it). Monism does allow an epistemic distinction. There are epistemic distinctions between phenomena that present themselves to our awareness via different sensory modalities. Some of these fall into the epistemic category of “mental” and some into the category of “physical”. But the categories are defined in Lakoffian terms, i.e. by relatedness to a prototype. And they are not mutually exclusive. Emotional states, for example, simultaneously involve both types of thoughts and types of bodily sensations. The mistake seems to be confusing an epistemic distinction for an ontic distinction.

  30. Jayarava


    In order to properly disagree with me, you would need to show that all physical features are entirely reducible.

    Not really, I did qualify by saying that many physical features are entirely reducible to snaller physical structures”


    It’s almost as if you think that reduction might work in principle, but that there are epistemic barriers in place.

    Yep. But ‘reduction’ in this context would not flow through low-level-physics based upon orthodox contemporary “Space,Time,Matter” equations to the high-level goal, it would have to have a different form.

    It really is speculation on my part – but as a matter of science I don’t think it can be ruled out. I’ll be no more definite than that.


    This would make you some kind of epistemic antireductionist on a bedrock of ontic reductionism

    Wow. I’ll take that as a complement. Possibly ?


    However, as you pointed out in your last comment you go along with the majority in believing that no single neuron, nor any simple arrangement of neurons, nor any sub-system of the brain taken in isolation, possesses the feature or property of having conscious states. No single neuron is conscious.

    Well .. the answer is I don’t know. Like everybody else. Every neuron might not be totally conscious, but they might be a bit conscious. Similarly I clearly disagree with you that a ‘simple’ arrangement of neurons is not conscious – it might be, once we’ve decided what a ‘simple arrangement’ is.

    I also don’t know what the actual role of neurons is in the brain. Like everyboy else. Everybody talks about neurons – neurons do this, they do that .. but the “atomic” causal agents of consciousness – if there are any- don’t have to be neurons. It could be anything. Neurons are just structures – in our terms they are small but in molecular terms they are massive. Who knows what the hell they do ?


    If no single neuron is conscious, and I don’t think you are disputing this, then the brain as a whole definitely has a feature that no single neuron has,

    That is not a failure of reductionism. A body of liquid has ‘liquidity’ which is something no single water molecule has. But by analyzing collections of molecules of water, liquidity is a predictable property of the superstructure. Liquidity is reducible.


    analysis of neurons… won’t yield consciousness

    It won’t, using a conventional physical/biochemical approach. But that is not to say a modified approach (which I admit is difficult to visualise) might work.


    The difference between a vat of 85 billion unconnected non-conscious neurons and a conscious brain is how the neurons are arranged and connected, then it is the arrangement itself (the structure) that makes the difference.

    I would say the start point is the “arrangement” and I would agree. It’s the brain , the “arrangement” that produces the consciousness – what roles the “bits” play is far from clear. It may be reducible, in the sense that a series of smaller components can be legitimately thought to aggregate to the higher level of consciousness. But i may well be that the starting point is the arrangement, without which no consciousness can commence – in which case the neurons cannot be viewed as reducible components of consciousness, but rather as drops of water in the sea which is necessary for the boat to float.


    I’m sure we both agree that the world is made up from quantum fields bundled into particles.

    Not really. Quantum fields bundled into particles is a particular perspective of physics. I’m not a reductionist in that sense and I don’t think the world is just physics, which is a particular discipline of the species.

    JBD

  31. Jayarava: I said “I could show a layman four different PARTS…” Or if I showed someone a square inch of a liver, kidney, heart or brain, it would be impossible to know how its natural functions emerge of chemical production, filtration, muscle or mind respectively unless you analyzed it down to the cellular level and were a scientist. For the first three, the emergence can be done relatively easily scientifically as opposed to the fourth to which they are still relatively blind.

  32. Tom in #30

    As i read

    “The notion of a first-person perspective, when construed in a certain sense, arguably helps to perpetuate the intuition that experience includes categorically private facts, facts that are inaccessible and unsubsumable by any sort of shared, objective, third-person understanding.”

    in the article you referenced, I realized that I have never had such an intuition. I realize that because my sensory apparatus is different from (although similar to) those of other people each person’s sensory experience will be unique. I also realize that the ability of human beings to agree on for example the redness of stop signs indicates that the the uniqueness is relatively trivial.

    Given that I have not had such an intuition I wondered how common that intuition actually is, and whether people who have it vary in some systematic way from people who don’t. Is it possible that the intuition is just the sense that some people have that they are special? Could it be mere vanity that makes some of us think that our experiences are ineffably special?

  33. ‘“globalists” posited that primary visual consciousness emerges from the inter-related activation of vast areas of the brain. In contrast, the so-called “localizationists” proposed that visual consciousness does emerge from specific and relative small brain circuits…’

    http://europepmc.org/articles/PMC4927634

    My own thoughts regarding the ineffable bits of qualia would tend to disinterest. If a facial expression is a “single” quale (as in experiment 2 above), it represents a high level summary of individual features. Other animals assess human facial expressions, presumably using similar lower level processing,
    http://europepmc.org/articles/PMC4830442
    and alter their actions depending on the results. Our type of conscious access (if it is that different) to this information is still is atop these non-linguistic faculties. This being the case, access will be not be to a verbal or non-verbal concept, but to a “feel”. The existence of synaesthesia, and the ease of our adjustment to novel modalities eg picture writing on the skin, suggests to me that the quality of the “feel” is pretty arbitrary. Or consider the sensation we experience following activation of a single type of olfactory receptor that has not previously been stimulated. It may or may not be unique (the human olfactory space is probably only ~20 dimensional). If unique, swapping around the smell qualia with another odourant (a la inverted spectrum stuff) doesn’t impress us like the Mary story.

  34. Hi Michael (36),

    I was raising the issue in 30 of whether there are first person, subjective facts about experience (what’s called “phenomenal knowledge”) in addition to 3rd person physical facts. Here’s Chalmers on that:

    “The classic statement of the knowledge argument against materialism has been given by Frank Jackson (1982). Mary knows everything that can be stated in physical terms about the physical processes that are in any way relevant to color vision. But Mary has never experienced colors other than black, white, and shades of grey. It seems that Mary has complete physical knowledge, but she does not have complete phenomenal knowledge: in particular, she does not know what it is like to see red. Jackson argues that Mary knows all the physical facts, but does not know all the facts. When she sees red for the first time, she learns a new fact concerning what it is like to see red. So there are facts over and above the physical facts, and materialism is false. In particular, phenomenal facts — facts about the character of conscious experience — are nonphysical facts, and phenomenal properties are nonphysical properties.”
    http://consc.net/papers/knowledge.html

    In the paper I linked to in 30 I was trying to make the point that it’s very difficult to specify, even to oneself, what factual knowledge one has of, say, red, so I was trying to cast doubt on the very idea of phenomenal knowledge. Qualia are the first-person subjective terms in which we know the *world*, so it isn’t as if we have propositional knowledge about individual qualia, although of course we can recognize colors, sounds, tastes, etc. and know how they relate to one another along various qualitative dimensions.

    Still, that there may not be private phenomenal facts leaves open, it seems to me, whether experience is straightforwardly physical. There are no accepted accounts showing that it *is* straightforwardly physical, although physicalists often assume it must be. For instance Sean Carroll, good physicist that he is, says

    “.. I don’t think downward causation is of any help to attempts to free the phenomenon of consciousness from arising in a completely conventional way from the collective behavior of microscopic physical constituents of matter.”
    http://www.preposterousuniverse.com/blog/2016/09/08/consciousness-and-downward-causation/

    But of course he has no suggestions in The Big Picture or elsewhere about what such conventional arising consists of, only the blithe conviction that it must.

  35. When [Mary] sees red for the first time, she learns a new fact concerning what it is like to see red.

    I always find the phrase “what it’s like to see red” as an example problematic because of the role played by “red” in English – the color of blood, stop signs, capes, heating elements, each associated with some activity possibly having high emotional content.

    My starting point for “what it’s like to see red” is the occurrence of a certain neural activity pattern (NAP) consequent to visual sensory stimulation by light with a SPD from a defined set. We respond to the NAP by saying “I’m seeing red”, but that’s esentially just a learned response to occurrence of the NAP. The attendant “mental image” that we associate with occurrence of the NAP is, as far as I know, an as yet unexplained “phenomenal experience” (PE) with as yet unexplained function.

    I assume that any emotional response attendent to occurrence of the NAP (eg, “Wow!) is due to associations with either the word “red” or the NAP itself, not with the presumed PE. Absent such associations, why would one PE evoke more emotion than any other?

    In short, for the first time, emergent Mary “experiences” the “red” NAP, but only in Sellars’ nonepistemic sense of an “undergoing” – here, a neural event in her brain. As far as I know, the exact nature of any associated “phenomenal experience” is unknown. So, I don’t see how one gets from such thin gruel to a meaty “non-physical fact”. And my understanding is that Jackson came to question that leap as well.

    I agree with Tom that the possible absence of phenomenal facts doesn’t cement physicalism, but it seems to me that it’s up to skeptics to provide negative arguments much better than Mary-type thought experiments in support of their skepticism.

  36. Charles in 40:

    “…the possible absence of phenomenal facts doesn’t cement physicalism, but it seems to me that it’s up to skeptics to provide negative arguments much better than Mary-type thought experiments in support of their skepticism.”

    Physicalists suppose there is just one kind of stuff – that which plays a fundamental role in explaining *all* phenomena, even if some loose ends like consciousness haven’t quite been pinned down. Consciousness will eventually fall to physicalism, they are quite sure. From whence comes this assumption and expectation? From the success of reductive physicalist explanations and the lack of evidence for any other fundamental sort of stuff.

    But the lack of any agreed-upon physicalist account (thus far) of how experiential properties (qualitiativeness and subjectivity) arise from or are identical to physical properties (non-qualitative and objective) suggests that such confidence might be misplaced. What I’d like to see physicalists acknowledge (and some do) is that the problem is, on the face of it, quite unlike standard reductions in science, in which the all the properties in question are in the public domain. The very idea of the physical, after all, is that which is mind-independent, self-existent, spatio-temporal. Whether this *idea*, this *conception* of the world as one kind of stuff can incorporate seamlessly, as physical phenomena themselves, the mind-dependent elements that give us the very idea of the physical, is an open question.

  37. Tom (39) and Charles (40)

    The first time I heard of Mary the Color Scientist I though that truly complete knowledge of the visual system would include the ability to simulate the neural activity pattern associated with the appropriate visual stimulation and thereby cause the associated phenomenal experience. Even within the terms of the thought experiment I don’t think Mary would learn anything new. Regarding physicalism, I don’t see how phenomenal consciousness can be completely non-physical, because of the way it can be affected by anesthesia, strangulation, blunt head trauma and the like. I agree with Charles that the seeming absence of non-physical components for other biological processes (digestion, respiration etc.) place the burden of proof on those who claim a nonphysical component for consciousness. I don’t think Mary The Color Scientist Sitting In Her Chinese Room Imagining What It’s Like To Be A Bat stories help them to meet that burden. The trouble is that because the phenomena whose existence they are trying to prove are non-physical, stories are all they have.

    I really think, as I said above, that qualia might have been invented by mere vanity, men imagining that their experiences are special and unique just because those experiences are their experiences and they as people are special and unique.

  38. Tom re.41
    To crudely lump all of the complexities of physical knowns and unknowns as just ‘one kind of stuff’ seems deliberately belittling and simplistic.
    Why do you feel the need to do that? I can only think that it is due to your desire for there to be something ethereal.

  39. Richard in 43:

    Yes I agree that “stuff” (a folk physicalist term) doesn’t do justice to the physical as we now know it scientifically, but reductive accounts generally assume that there’s a bottom-level set of mind-independent properties that are involved in all physical phenomena. So the question about physicalism is whether consciousness can be understood as Sean Carroll would have it, as arising “in a completely conventional way from the collective behavior of microscopic physical constituents of matter.”

    I don’t desire the ethereal, just a transparent naturalistic account of how the qualitative and subjective fits together with the quantitative and objective. Whether our conception of the world as “all physical” can do that is an open question, seems to me. Seeing that our basic epistemic situation is that of creatures who model reality using mind-dependent representations (qualia, concepts, numbers) might explain why we won’t, contrary to the physicalist assumption and expectation, find those representations in the physical world as modeled by them. We’ll only find their physical and functional correlates.

  40. Tom re 44
    Thanks for your reply and reference to Sean Carroll. It is unclear from his recent blog as to whether he supports the belittling view of physical reductionism relating to consciousness, ie ‘a bunch of atoms and particles, mindlessly obeying the laws of physics.’ As again, this completely ignores the whole complex and vast spectrum of physical systems, which are able to react and interact from the micro to the macro, body to brain and vice versa, making us as we are. Examining the evolution of how all these non-conscious processes function and interact is the way to understand awareness, memory and consciousness. Not by reducing the physical to one kind of stuff or a bunch of atoms and particles, as that is just silly. Most of our physical function is autonomic and non-conscious, which some people ignore and do not even consider as part of themselves. The comparatively small amount of conscious somatic activity interfering when awake, being the interaction and reaction between our internal physical autonomic needs and variations of the physical external world. As Libet and others have found, conscious somatic interaction can prevent via our senses and memory what had unconsciously, autonomically been required and initiated. Possibly the duality that many confuse with their need for a believers dualism.

  41. Richard (43)

    My understanding of “physicalists suppose there is just one kind of stuff” is that there is no “soul stuff” or “spirit” and that the phenomena that theologians, spiritualists, ghost busters and the like explain using ectoplasm or the Holy Spirit or other paranormal kinds of stuff can be explained using atoms and molecules, the normal as opposed to the paranormal kind of stuff. I suppose when you start talking about strings and fields and things the normal/paranormal distinction might start to become problematic…

  42. Michael re 46
    Some physicalists do not think that a bunch of atoms can provide consciousness, so they leave the door open to the non-physical or pansychism, not me I hasen to add. Hence my reply to Tom of my understanding of the interaction, between the internal nonconscious zombie like autonomic action, and the conscious somatic action that allows us to function in the world as we do including sleep.

Leave a Reply

Your email address will not be published. Required fields are marked *