Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Does non-mental supervenience exist?

  1. Nov 26, 2009 #1
    This is an example of supervenience:

    "a wall supervenes on its bricks"

    It seems to me that the words "supervenes on" can be replaced with "consists of". A wall consists of its bricks. This makes clear that "supervenience" is about giving a label(wall) to a group of objects(bricks). So supervenience is actually nothing more than the act of label-giving, and this is a mental activity that takes place inside minds. One sees a collection of bricks and gives it a label, "wall". Thats supervenience.

    Am i wrong?
    Anyone know an example of non-mental supervenience?


    Btw the reason im asking is because im wondering what use the common phrase "mind supervenes on brain" has. If it just means "mind is a label given to brain", and it doesnt describe any kind of physical relation or causation, then its no different from a phrases like "the mind dreams about clouds" or "i see colors". All of those phrases describe mental activities and none of them has anything to do with, let alone supports, physicalism.
    Last edited: Nov 26, 2009
  2. jcsd
  3. Nov 26, 2009 #2


    User Avatar
    Gold Member

    The use of supervenience as a term is more complicated than this. But you would be right that it has come to stand for a certain intellectual position on the issue of emergence and systems properties.

    It is a shorthand way of saying that sometimes something must of course depend on the stuff out of which it is made, but the nature of that dependence can be opaque. Given this state of physical affairs, we will have this state of mental affairs. And this is pretty much all we can say.

    This is not the technical definition, but rather the way the term became used in brain~mind debates.

    It also tacitly supports a reductionist substance ontology - the belief that all causality acts from the bottom-up, reality is just a collection of atoms that may get arranged in complex ways. And then because simple bottom-up causality is clearly inadequate for capturing the full causality of complex systems, supervenience becomes the justification for jumping to dualism.

    So a wall can be just a stack of bricks. But can you really believe a mind is just a bunch of neural firing? No, well you will have to join us dualists then.

    So you would be right. Supervenience is about a lack of real causal explanation. Either way you go from "bottom-up emergence", you lose.

    Systems thinking recognises that reality involves both bottom-up and top-down causality in interaction. So what "emerges" is the thing in the middle, or what some might call the whole.

    This goes right back to Aristotle's four causes. And the standard example give is a house made of bricks. The bricks are only the substantive cause of the house. There is also an effective cause in the builder, a formal cause in the designer, and the final cause in the very purpose of there ever being the house.

    You can see perhaps how the four causes are split into bottom-up and bottom-down actions.

    The bricks and the builder are bottom-up - the constructive causes. While the plan and the purpose are top-down - the constraints which shaped the construction.

    Even physics breaks down this way - into the materials and the laws, the initial conditions and the boundary constraints.

    As to consciousness, the source of the problem is that so many people use the word freely without having a clue about the neurological, psychophysical and anthropological complexities bound up in being "an aware and self-aware human".

    It is exactly like the days when people believe in a life force that must animate matter. Vitalism. There must be some mysterious substance or essence which gets added to the physical mixture.

    But now who does not understand life as a complex systems property? Who cannot see the mix of substantial, formal, effective and final causes involved?

    Everyone talks as if we don't yet have this level of information about mental processes. But frankly, they just have not put the time into studying what can be known.
  4. Nov 26, 2009 #3
    Damn, Apeiron...is there a subject you don't know a lot about? What is consciousness? What is mind? Have you ever read anything about Carl Woese? He was written some interesting things about the flaws of reductionism. Sorry for all the questions but you seem to know a lot about this.
  5. Nov 26, 2009 #4


    User Avatar
    Gold Member

    Well, I don't know about Carl Woese. :rolleyes:

    A quick google reveals he split the Archaea off from bacteria and was an RNA world proponent. More recently he rails against the molecular paradigm in biology.

    And I guess that does show my comment "who does not believe in a systems approach to biology" would be a little too hasty in some quarters. The recent hunt for a gene for this, a gene for that, would be an example of science being swamped by simple-minded reductionism again.

    Though even then, no molecular biologist that I ever came across really believed biology was just bottom-up chemistry, or even genetics. Yet in mind science, that was sadly a too common view.

    If you want some name-dropping, a few years back I quizzed Francis Crick exactly about this - why he and Koch were putting forward such crudely bottom-up theories of consciousness. The 40 hertz thalamic synchrony, the pyramidal cells, the mysteriously connected claustrum, and other candidate "consciousness-causing" structures of the time.

    He was very reasonable. He said well we've got to put forward hypotheses then let them get shot down experimentally. Yet he was unshakeable about the idea that some specific physical structure within the brain must be responsible for the "trick" of consciousness lighting up within a material processing structure. He was strikingly naive about systems principles.

    But others like Chalmers, Penrose and even Dennett displayed an even greater basic lack of psychology and neuroscience understanding.

    The truth is that everyone wants to talk about consciousness but hardly anyone will do everything it takes to be able to talk about it scientifically - on the basis of known data and using sound theoretical frameworks.

    Relating this to supervenience, scientifically I found that mind science would need to supervene on theoretical biology, which in turn supervenes on thermodynamics and systems science.

    But the mainstream approach expects mind science to supervene on computationalism or perhaps quantum mechanics.

    So I am arguing that it needs to be a pyramid of holism or systems thinking with mind science as the cherry on that cake of increasing complexity. Yet the Cricks, Chalmers and Dennetts see a different hierarchy of theory. High level bottom-up theory would supervene on foundational bottom-up theory.
    Last edited: Nov 26, 2009
  6. Nov 26, 2009 #5
    Very interesting. And Crick is a giant.

    Here is an article by Woese on reductionism and the philosophy of biology. He is saying a lot of the things you are saying. I dont know if you willl agree with him though.


    And this between Dyson and Weinberg:

    weinberg said something like the explanatory arrow always points down.
    Last edited: Nov 26, 2009
  7. Nov 26, 2009 #6


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi pftest,
    This isn't exactly correct, but it's very close. We generally use the term "supervenience" to denote how a property or phenomena is dependant on a physical structure. SEP is always a good reference for this. They state:

    In cognitive science, we generally suppose that a mental state (meaning the phenomenal mental state or the phenomenon of experience) is supervenient on the physical state (meaning the objectively measurable physical state including the positions or locations of all neurons or other material on which the mental state is dependant). Thus, in order for you to experience something different than you experienced just a moment before (these are changes in your mental state) then there must be a corresponding change in your physical state (a change in which neuron is firing for example).

    This concept is not without contention, though I won't pretend to understand the details. If you look farther down the page referenced above, they quote Chalmers, who amoung others, would contest the concept of supervencience.
    I understand Chalmers backs up his view in "The Conscious Mind" (book). I'd be interested in understanding how this argument is defended. As far as I'm aware, most of the philosophical community accepts supervenience, and I'm not sure how Chalmers can avoid it.
    Last edited by a moderator: Apr 24, 2017
  8. Nov 26, 2009 #7


    User Avatar
    Gold Member

    I certainly agree with his argument that "reductionism" has two senses.

    The kind of reduction we all must do in science, or modelling generally, is of the generalising or information reduction kind. The epistemological kind I would call it.

    So when we want to represent "the world", we actually want to boil away all the particulars and extract the universals. Describe in terms of principles, laws, equations. The world can then be recreated by measurements and predictions.

    Then there is the second kind of reductionism which is ontological. This is when the world is seen as being reducible to the fundamentally small and atomistic, rather than the fundamentally coherent and systematic.

    On Weinberg vs Dyson, you are obviously a fan of Dyson. I've never actually thought of him as an actual systems guy. As in having a theory of systems as opposed to just knowing reductionism is not enough.
  9. Nov 26, 2009 #8


    User Avatar
    Gold Member

    The zombies argument is basically that I can imagine every fact about a brain being true, and then that brain not actually being conscious (even if the zombie is doing processing so that it gives every appearance of being conscious).

    But as I say, does that ability to imagine such a case stem from a detailed understanding of brain neurology, or instead a basic ignorance of it?

    Where I personally get to in this debate is that supervenience - in its strict technical sense - does bog down when we get to the necessary redness of red.

    So I can sort of imagine a brain that does everything much the same as mine (so your brain for example) and yet it is difficult (well, impossible really) to say why my particular red kind or redness might not be some other kind of hue experience for you. Perhaps if I could see inside your head, your red would look perhaps bluish, or even like a colour I never imagined.

    And indeed we know there are the colour blind who do see colour differently. And also tetrachromats who would see a range of reds where I only see a single shade.

    Now I would still say that if you had exactly my brain, with exactly its same history of development and experience, then the mental state should exactly supervene. Your red and my red would have to be identical. It just seems logical.

    Yet even then, there is a "why red" question. Why do we both now see red, and not some other tinge? What is it about the circuitry that forces such a particular response?

    So while zombies are implausible, you can get down to this residual issue of how red as a particular phenomenal response - rather than the generality of there being a phenomenal response - can be accounted for in causal terms. As a state A that supervenes on a state B.

    That is, we can imagine with all plausibility that it does supervene, but not why it visibly must supervene.

    How do you get out of that bind? I would say it shows we do eventually reach limits of modelling. Eventually there are just things which for us - trapped within our own subjectivity - just are. We could rerun our developmental history and might come up with a slightly different brain, a slightly red. But that is not practical as an experiment. And what cannot be varied, what cannot be changed and seen from different angles, just exists for us.

    And yet, we have ended up with something so tiny, and so much else now accounted for. Do we get all dualistic like Chalmers and use the fact of a limit to throw out everything used to reach that limit?

    As to how Chalmers avoids arriving at commonsense positions rather than at gee whiz stories about panpsychic dualism, let me guess? Could it have something to do with how you build an academic career?
  10. Nov 26, 2009 #9
    I take it your a systems guy.

    I like Weinberg too. I have learned a lot from both. It was a Dyson article which first introduced me to Woese. They are both anti-reductionist. Is that rare in physics? I was just seeing if you had an opinion on this kind of physics based philosophy of reductionism versus the philosophy of biology and its inherent complexity. At least some physicists seem to be crossing over to biology and bringing this philosophy with them. And Woese said there is or was a lot of physics envy in biology.

    Woese from that article:

    Last edited: Nov 26, 2009
  11. Nov 26, 2009 #10


    User Avatar
    Gold Member

    I think things are more complex than being anti-reductionist.

    How I see it is that reductionism (in the "all is just a bunch of particles" sense) is a brilliant and proven method of modelling. It works. It inspires technology like computers.

    But then there is also this other possible brand of modelling - one that does not have a great productive track record as yet, but may be better at accounting for "the whole" of things.

    So diagnosing what is actually missing from the "too simple" models of reductionism is a first step. Then working out how to represent the missing factors in a different way of modelling - call a systems approach to give it a name - and that would be a way forward to another level of modelling reality.

    Schrodinger of course wrote the classic on how "biology is bigger than physics". So how there is more to life than reductionism.

    Polyani is another common cite because he got an article into Science. Same with Anderson.

    In modern times, speaking from within the camp of hardcore physics, I would say Gell Mann is excellent, so is Davis, so is Laughlin.

    I'm glad to be introduced to Woese now, but he seems peripheral to the reductionism~systems debate as much as it has been publically debated.

    But anyway, a very key message is that both sides can be right.

    Of course, philosophically, systems thinking is righter! But it is also to date far less technically impressive. It helps you know things, but not so much lets you do things.

    This is also why perhaps it is difficult. You have to learn two different methods of thought. And many people find it hard enough learning one. :-0
  12. Nov 26, 2009 #11


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi apeiron,
    The zombie argument only suggests, as near as I can tell, that it is feasible for a behavior to exist without experience. I think people take it too far and try to suggest that there could be a physical state with a known mental state, and that physical state could also not support that mental state. But these two are very different. Suggesting that it is feasible for a behavior to exist without the experience is very different than suggesting a mental state may or may not exist given a specific physical state. And that’s why I mentioned above that if put into those terms, into the literal terms provided by SEP (that physical state B ALWAYS creates a mental state A) then I think everyone will be forced to agree.

    If on the other hand, it’s being suggested in the literature, that physical state B may or may not support mental state A, then I’d like to understand what the argument is. When speaking of physical states, I’m refering to what might also be called the “micro states”, in that even every partical is in the same physical state. Also, mental state here means the specific state of phenomenal experience. So if we have a physical state B, and we know it corresponds to a mental state A, then every occurrence of physical state B will create the same mental state A. I don’t see that being argued in the literature.

    Take for example, Maudlin who suggests the “supervenience thesis” (I’ve not heard another philosopher suggest it as a “thesis”, but it seems reasonable to say so).
    I think the first issue is how to define supervenience. I’m not certain that Chalmers for example, defines it the same way. Do you know how Chalmers defines it? Please provide references if possible. If you have his book, “The Conscious Mind”, check pages 127 through 129. Especially where he says, “Experience is not a fundamental property that physicists need to posit in their theory of the external world; physics forms a closed, consistent theory even without experience [such as is suggested by the p-zombie thought experiment]. Given the possibility of a zombie world, there is a clear sense in which experience is superfluous to physics as it is usually understood."

    I'll have to read over what Chalmers is talking about here. I've not read it in a while. But if you can suggest why Chalmers rejects supervenience, it would be appreciated.
  13. Nov 27, 2009 #12


    User Avatar
    Gold Member

    In the bit you cite, Chalmers is trying to pave the way for his (bonkers) idea that consciousness might be another fundamental aspect of material reality. So just as electromagnetic charge and magnetism got combined in a single better theory, and then matter and energy, so conscious experience might just be another physical aspect of nature.

    Chalmers actually wants to have it both ways.

    The zombie argument is used to created the initial impression of dualism. He shows (he claims) that nothing in intermediate levels of materialistic explanation, such as neural synchrony, would require conscious experience in a brain. So therefore causal explanations collapse all the way to the ground floor of material theory - fundamental physics, QM perhaps.

    So first knock out all the intervening possible levels of bottom-up explanation. And then either the very bottom level ToE of material modelling will have to eventually include consciousness as a property along with charge, mass, etc (the monadic outcome) or we will have a naturalistic dualism in which there is the physics ToE and some separate mind ToE with its own very simple and physics-like psychophysical laws.

    So "why" Chalmers rejects supervenience is to allow him to make outlandish speculations about consciousness as a fundamental material essence rather than having to accept the duller idea that is is just a very complex system. One which he would have to study some neurology and psychology to be able to make any interesting comments about.

    Sorry to be cynical, but I was there when he made his first splash at Tuscon 1. And I had dealings with him for a few years following. He knew how to make a career in his particular field.
  14. Nov 27, 2009 #13


    User Avatar
    Gold Member

    Given your interest in zombie approaches, a good book giving the ideas a work out is Robert Kirk's Zombies and Consciousness (Clarendon2005).

    Here is the review I did for the Journal of Consciousness Studies (and I still agree with myself!)


    Having popularised philosophical zombies in the 1970s, Nottingham University’s Robert Kirk now thinks it is past time to kill them off. But despatching the undead was never going to be easy. Zombies and Consciousness has two aims. First, Kirk hopes to show that the notion of a zombie – a person of flesh and blood but without the inner light of experience – lacks logical conceivability; it is incoherent and thus cannot be used as grounds for proposing a ‘Hard Problem’ of consciousness. Second, he wants to go the next step and show that ordinary physicalism can completely account for consciousness. He identifies a set of psychological functions, each of which is plausibly physical in its causation and, when bundled together, should result in a conscious being with no further (mental or otherwise non-physical) aspects left dangling. Consciousness would be just the sum of these material activities and nothing more.

    The zombie story is that it is possible to imagine a creature which is exactly the same as you and me in every physical detail. It would have a brain that processes information and act as if it can “see”, “think”, “imagine”, and “feel”. But the final essential ingredient would be missing. It would not benefit from a parade of subjective experiences – qualia. All would be dark inside. Kirk tells how he originally became a zombie enthusiast following a naive question from a first-year student at a tutorial. As he thought about it, the materialist position just melted away and for some years he was an ardent convert.

    Zombies do not actually have to exist. Just the fact that we agree the idea to be a logical possibility opens the door to Cartesian dualism. If all the physical circuitry does not necessarily entail the mental states, then physicalism is not up to the job of explaining consciousness. The mind is still the ghost in the machine. It’s old idea, as Kirk acknowledges. In the 1930s, G.F. Stout used zombie-type examples to argue against epiphenomenalism. Stout said it was “incredible to Common Sense” that there could be human bodies lacking mental experiences that would still go through the motions of making and using telephones and telegraphs, writing and reading books, speaking in Parliament, even arguing about materialism.

    For a long time, zombies played only a minor role within consciousness studies. Searle’s Chinese Room – which appealed better to the artificial intelligence community – hogged the limelight. But Chalmers (1996) put zombies centre stage in the late 1990s when he used them to argue that a physicalist approach to mind could never work. The logical conceivability of zombies proved there was an explanatory gap between the objective realm of the brain and the subjective one of mind. In philosophical circles, however, dualism is a monster that many would like to see buried once and for all. So there would be plenty of people to cheer on Kirk if he has now put paid the zombies that he used to love. Does he succeed?

    As a warm up exercise, he starts with the jacket fallacy. Some properties can be imagined as capable of being shrugged off - like a person might remove a jacket. Nothing essential changes. The wearer is merely now sans jacket. However other properties are central to the definition of what it is to be that thing. The performance of a car cannot be altered except by fiddling with some physical part of its machinery. It makes no sense to think of two cars identical in every mechanical detail, yet one has “performance”, and the other lacks it. Of course this Rylean category-error style rejection of dualism may be considered facile. To zombie-mongers, the performance of a car is merely an emergent property. The mind appears actually different in kind. Stronger arguments are required.

    Kirk then mounts his own e-qualia argument (“e” for epiphenomenal the reader is left to presume). His angle is that e-qualia are a necessary corollary of zombiedom. For zombies to exist, e-qualia would also have to exist as the precise type of mental experience that the zombies so sadly lack. And then comes the clever bit. If we feel that his description of these e-qualia is incredible and quite contrary to common sense, the original idea of a zombie must be incredible too. The downfall of one is automatically the demise of the other, for they would be two sides of the same coin.

    So what would e-qualia have to be? In the spirit of zombiedom, we would begin with the suggestion that the world has a part that is physical and closed in its causation. That is, every physical event within it is a result of other physical events. This physical part would then be used to construct a body with a brain that does complex cognitive processing. Just like a zombie. Then along comes the extra bit that makes this creature instead conscious. It would enjoy a parade of non-physical e-qualia. For some reason the physical activity would generate an epiphenomenal, causally inert, glow of experience. The qualia are there but they have no effects and do no cognitive work. This logically is what we must suppose if e-qualia are then to be the kind of thing that can be stripped away without altering the physical activities of a brain in any way.

    So Kirk presents us with a conscious being that can be turned into a zombie through the loss of its e-qualia. It all seems conceivable - so far. At this point Kirk hopes to snatch the rug from under the argument. Consciousness involves one further necessary aspect he says. We apparently have epistemic access to our mental states. They are the subject of much noticing, attending, remarking and comparing. How, for example, could we ever choose between the taste of two wines unless we had access to qualia that are the subject of the mental contrast? We have no good explanation of this kind of access, he says, but even zombie enthusiasts feel that we possess it. Therefore e-qualia, in the sense of qualia which are so completely epiphenomenal they do not even do indirect work by way of being noticed and acted upon, cannot exist. If this kind of pure epiphenomenalism is inconceivable, as even zombie “ultras” must admit, then zombies become inconceivable as well.

    Kirk dismisses the obvious counter-argument. Zombie supporters would reply that zombies are able physically to feign all our complex mental responses. This would be so by definition. They would process a lot of information and give every outward impression of admiring two fine wines. They would fake a sense of attentional effort and deliberation if necessary. So again there would be a dualistic position in which a physical world of itself cannot entail the presence of mental states. The existence of qualia remain extra to any materialistic story. This misses the point, says Kirk. We know we have access to our conscious states, so the idea of zombies as just us minus e-qualia is the thing that is inconceivable. It is not about what level of clever behavioural simulation might be possible but about whether we really have something inessential that can be stripped away.

    Does his argument work? Well not really. Perhaps I am missing something here myself, but it could be true that we really need our mental states to operate and yet it is also conceivable that a zombie might not. This is the essence of the argument. Of course we are actually conscious (we think). But it is unclear how that consciousness is entailed by any physical mechanism. It is a case of in for a penny, in for a pound. Once we grant that a zombie can feign the presence of everything else, like perception, thought and imagery, then why not also epistemic access? Thus it remains open to us to suppose that we have e-qualia and are merely being fooled by our remarkable cognitive machinery into believing our mental states are both necessary and causal to our intellectual functioning.

    Kirk’s case is not helped by his principal follow-up argument, the sole-pictures story (a pun on soul-pictures). He asks us to imagine a zombie whose physical processes produce the particular epiphenomenal effect that it has qualia-type pictures, like little television images, playing on the soles of its feet. Whatever the zombie feigned seeing would in fact show up somewhere as an actual activity. The point is that we would not expect the zombie to have an epistemic relation to this activity merely because it happened to be occurring. But this is weak. Flickering images projected on the soles of the feet would be a physical process as Kirk himself agrees. And the dualistic position is that, being mental, the epiphenomenal states under discussion are of a different kind. Res cogitans not res extensa. Thus it is not where they show up that matters -– either in the head, or on the feet – but the fact they exist.

    So Kirk has not killed off his zombies. Must we then believe in the resulting explanatory gap? Not at all! I would argue that the hard problem is created by a basic assumption of the brand of logic that philosophers generally choose to deploy. Being axiomatic to the logic, this same logic can hardly be used to defeat it. The arguments of Kirk, Chalmers and others depend on a “mechanical” logic based on the law of the excluded middle. Everything is either a this or a that, one kind of thing or another. Crisp binary divisions are taken for granted. But there is an alternative or indeed complementary view, which I call organic (Kahn, 1960; Peirce. 1980), where middles only become excluded in the course of a process. On this view the question becomes, not how the physical could ever produce the mental, but how they ever became separated.

    In the organic view, everything begins mixed together as a vagueness – the unbounded apeiron, or naked Aristotelian potential. Then this vagueness divides dichotomously. It tends towards opposed limits. The Peircean firstness of monadic vagueness becomes the secondness of a dyadic separation through interaction. Then, out of this separating, arises the thirdness, the triadic richness, of hierarchical complexity. A bootstrapping story of 1, 2, 3.

    This is not the place to defend organicism as an alternative logic. But we can sketch its key consequences for theories of mind. It implies that every dichotomous outcome begins in the commonality of a vagueness. And the division does not bring absolute separations, only relative ones. Limits arise, but they are limits that can only be approached, never actually reached. To fully attain them would be to break the world apart and leave no middle ground of interaction. If we follow this logic, which could be said to exude limits rather than exclude middles, we can see that the apparent opposition of mind and matter is in fact an outcome of the dichotomous separation of a single potential. Although the two may now seem far apart as kinds, they must remain connected in terms of causality. They are the mutual product of a process of dependent co-arising or paticca samuppada (Macy, 1991). As a necessary fact of logic, therefore, the physical and the mental are to be regarded as joint products of a process of development. There can be no hard problem because, like figure and ground or yin and yang, one could not exist without the other.

    This easy victory does have its troublesome consequences. The same logic requires that all of the physical world must be connected to the mental world in some real manner. This does not necessarily entail panpsychism; the idea that the material world – objects such as stars, rocks and water molecules – has qualia. But it does lead us to pansemiosis, a view of reality organised in a holistic or hierarchical fashion by a top-down, mind-like in the broadest sense, knowing. This is not so outlandish as it may sound. Physics already has universal laws that look down to constrain every local event. Relativity and quantum theory are both observer-dependent models of reality.

    For the moment it does not really matter how the concept of mind would be deconstructed under a pansemiotic and organic approach to the modelling of the wider world. It is enough to show that zombies and their detached e-qualia are highly dependent on a system of logic that assumes what it then proves. There is hidden tautology in the arguments of this book as well as in those of zombie enthusiasts. Mechanical logic is in its way dichotomistic. But because of its reliance on the law of the excluded middle, mechanical logic leaves no option but to say that reality is either dualistic or monadic. Either the world is made of fundamental twonesses – such as chance and necessity, stasis and change, atom and void, discrete and continuous, substance and form, simple and complex, particular and general, matter and mind – or one of these two is taken as the fundamental and the other the derived or constructed. Every one of the above mentioned dichotomies has been the subject of Hard Problem type wrangling. Is the world fundamentally continuous or discrete, random or determined, a flux or a frozen spacetime block, a formless chora or the shadow cast by Platonic ideas? It is simply the nature of the beast. A discourse founded on the law of the excluded middle has no choice but to vacillate between monism and dualism, finding neither satisfactory when it comes to the deep ontological questions.

    So it would be astonishing if Kirk, armed with standard logic, could fulfil his first aim and finally dispose of zombies with the dualism they imply. To start with the physical and then to try to build up to the mental is a doomed project because the connecting middle ground that must bridge the gap has already been excluded in the formation of the dichotomy. Zombies and e-qualia may be incredible to common sense, but dualism remains the inevitable destination for this way of thinking.

    The second half of the book is taken up by Kirk’s other aim; an attempt to define consciousness in terms of a bundle of functions. He reviews the rise of awareness in the animal kingdom and says the essence of subjective awareness is being a decider. This ability to decide involves a “basic package” which includes processes such the initiation and control of behaviour, the acquisition, interpretation and retention of information, the assessment of situations, and the choice of alternatives guided by goals. Then, to ensure this basic package of cognitive skills is conscious, there has to be one final thing – directly active information. What comes into the mind must have immediate impact and gain processing priority. What he is hoping to achieve here is to outline a set of functions which, when combined together, would leave out nothing that a mind is capable of doing. You could hand over this wish list to a clever hardware engineer and get back a conscious system. If his list sounds believably complete and implementable, we should find it easier to accept that mind is material.

    Why does this approach seem so inadequate? Again because it is mechanical – based on the atomistic and reductionist approach by which humans build machines. Kirk is saying the mind can be created by putting together a system of particulars. Each of the functions is some particular skill, a component or a module. By careful choice of particulars, a mind can be constructed. An organic metaphysics suggests quite a different approach, for it treats mind as a fundamental category. Mind is an extreme to match that other extreme, brute inanimate matter, and so is a general rather than a particular. It has to be approached in terms of its universal laws rather than as a set of locally contingent specifics.

    This is the message we should be taking from the ‘Hard Problem’. The material realm is indeed not enough. Mind is something other. But this does not mean we have to accept dualism. What we have are two limits approached from a shared middle ground. As scientists we should aim to model each kind of limit in terms of universal laws. A heap of particulars would always be the wrong approach for dealing with something that is actually fundamental. Can mind in fact be treated as a universal? Yes, of course. An organic metaphysics – such as Peircean semiotics, for example – treats mind as the upper boundary, the realm of downward acting constraints. The whole that shapes up the parts. Organicism works as philosophy and it also works as science. Once we know what we are looking for, we can appreciate the progress already made towards modelling the universal laws of mind with anticipatory systems (Grossberg, 1995; Rosen, 1985), autopoietic systems (Maturana and Varela, 1992), complex adaptive systems (Waldrop, 1992), and hierarchy theory (Salthe, 1993; Pattee, 2000). All these approaches share similar principles and lead towards generalised mathematical ideas. And while they have been prompted by the need to explain (mainly) biological complexity, there is no reason why they cannot be extended to cover physical simplicity – the “simple” world of particles, stars and universes. This is the future of consciousness studies, in my opinion anyway. We are working towards a theory of how wholes can organise their parts, regardless of whether these wholes are organisms or entire worlds.

    Kirk’s zombies are lumps of physics that have lost their minds and no amount of mechanical complexity is ever going to restore them. But the organicist’s idea of mindfulness as the organising, constraining, downwardly-acting, aspect of a dichotomised reality could bring mind back to the entirety of existence. Now that would be quite an achievement for consciousness studies, wouldn’t it?

    - Chalmers, David (1996), The Conscious Mind (Oxford: Oxford University Press).
    Grossberg, Stephen (1995), ‘The attentive brain’, American Scientist, 83(5), pp. 438-449.
    - Kahn, Charles (1960), Anaximander and the Origins of Greek Cosmology (New York: Columbia University Press).
    - Macy, Joanna (1991), Mutual Causality in Buddhism and General Systems Theory (Albany, NY: State University of New York Press).
    - Maturana, Humberto and Varela, Francisco (1992), The Tree of Knowledge (Boston, MA: Shambhala).
    - Pattee, Howard (2000), ‘Causation, control, and the evolution of complexity’. In P.B. Anderson, C. Emmeche, N.O. Finnemann and P.V Christiansen (eds.), Downward Causation (Aarhus University Press).
    - Peirce, Charles (1980), Selected Writings (Mineola, NY: Dover).
    - Rosen, Robert (1985), Anticipatory Systems (New York: Pergamon).
    - Salthe, Stanley (1993), Development and Evolution (Cambridge, MA: MIT Press).
    - Stout, G.F. (1931), Mind and Matter (Cambridge: Cambridge University Press).
    - Waldrop, Mitchell (1992), Complexity (New York: Simon and Schuster).
  15. Nov 27, 2009 #14
    I read the quote but im having trouble visualising what it means. Could you give an example of purely physical supervenience?

    Before we use the term supervenience on the mind-body issue, i want to be sure that it has something to do with the physical, and that its not a purely mental activity.

    It still sounds like a "consists of" relationship. The mind consists of (or = )the physical state, therefor any change to the physical state equals a change to the mental state.
    Last edited by a moderator: Apr 24, 2017
  16. Nov 27, 2009 #15


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi pftest,
    Absolutely. Consider that pressure and temperature of a gas are properties that are supervenient on the atoms or molecules that constitute the gas.
  17. Nov 27, 2009 #16


    User Avatar
    Gold Member

    This is a nice example, but does it support the idea of "there cannot be an A-difference without a B-difference”?

    If we have an ideal gas, we can change the pressure/temperature by changing either the number of particles or the size of the container. And also the average particle energy of course.

    So here you are arguing that the macrostate (pressure/temperature) is supervenient on the microstates (all the possible arrangements of a particular number of particles with a particular energy in a particular container).

    And it seems we have got to the heart of the description when we can simplify the modelling of the system until it just is a macrostate~microstate locked down supervenience relationship. The microstates depend on at least three variables (particle number, container size, average energy density). But all that gets reduced/generalised to the B concept of a microstate. Then this can be exactly related to an emergent macrostate.

    So good physical modelling. Yet we can also see that supervenience seems to have been imposed as a property of the model rather than emerging neccessarily as a fact about the reality.

    We seem to be starting with the A-level macrostate and then creating a description of "what lies beneath" in the kinds of terms that would allow a completely supervenient relation. Which is kinda top-down. A constraint of what can be at the lower level.

    So this is how science has extracted an important example of a supervenient relationship in nature. The A emerges completely from the B - if the B is specified in terms suitable to allow that to happen.

    Now if we want to apply that modelling technique to minds and brains, can we pull off such a stunt? I call it a stunt, but it would of course be a model that might be worthy of a Nobel.
  18. Nov 28, 2009 #17
    If the gas pressure consists of atoms, then isnt "gas pressure" not just a label given to a collection of atoms? Just like "wall" is a label given to a collection of bricks. We could do without the "pressure" and "wall" labels and just describe the whole system in terms of its components and their interactions.

    The statement "there cannot be an A-difference without a B-difference", is true because A differs only from B in terms of the label it has (its labelled A, while the other is labelled B).
    Last edited: Nov 28, 2009
  19. Nov 28, 2009 #18


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi apeiron,
    Yes, it’s a very nice example. Though I won’t claim to have come up with it myself. This issue certainly comes up regarding the kinematic theory of gases. Frigg wrote in “What is Statistical Mechanics”
    Ref: http://www.romanfrigg.org/writings/What_is_SM.pdf

    Take for example, the various measurable physical parameters of a gas. Here’s a list that’s not exhaustive:
    - temperature
    - pressure
    - density
    - internal energy
    - enthalpy
    - entropy

    One can specify the physical state of a gas by defining any two of the above. So if I know the internal energy and the entropy of a gas, then I can determine the physical state and all the other physical parameters above. In order for there to be a change to any measurable property of the gas such as…
    - Any of the above physical parameters
    - viscosity
    - ratio of specific heats
    - compressibility factor
    - thermal conductivity
    - etc…
    …there has to be a change in the micro state of the gas.

    Sorry, but I don’t really know what you’re getting at here. I work as a mechanical engineer, and what I can tell you for sure is that the measurable properties of a gas for example, are determined by the state as mentioned above. Note that given a gas has a specific temperature and pressure, we’ve defined the macro state, but we HAVEN’T defined the microstate. There can be numerous (infinite?) micro states all of which have the same macro state. However, if we see a change to the macro state, there MUST be a change that has occurred to the micro state.
  20. Nov 28, 2009 #19


    User Avatar
    Science Advisor
    Homework Helper
    Gold Member

    Hi pftest,
    The A property (pressure for example) is not just a label, it is a measurable property. A wall is not the best example to describe the use of the term supervenience, but if I had to use it, I'd say something like the density or hardness of the wall is supervenient on the micro states of the bricks. I don't like that example, it's not a good one, but there you go. Hopefully the responce to apeiron above also helps.
  21. Nov 28, 2009 #20


    User Avatar
    Gold Member

    The point I was hoping to make was that strong supervenience appears in practice to be created top-down in our modelling. We define the local or micro-scale in terms suitable to yield us our global or macro-scale. And statistical mechanics is an example of just this.

    So it seems to me that we start with two kinds of description that don't dovetail - a set of macro-properties and a second set of micro-properties. Then we generalise until we are instead working with macro and micro states. We turn what does not have a necessary supervenient logic into something that has been abstracted to an extent where it does.

    So the macro properties are pressure, temperature and volume. And the micro properties are then the kind of variables you list here, and more....

    Real gas particles of course come with all sorts of complications, such as valence, dipoles, atomic weight. We could have a mix of atoms (which may lead to explosive complications!). We may have phase transitions so the particles go liquid or solid at certain temperature/pressures, leaving us with a less than ideal gas.:smile:

    So there is huge variety possible in the micro-properties. And most of this complicating possibility we want to abstract away to be left with micro-states upon which macro-states can be seen to be logically supervenient.

    So this is the WHOLE point. Supervenience is more a fact about our models than a fact about reality.

    When it comes to the behaviour of gases, it is indeed quite easy to create this kind of supervenience. Gas particles barely interact with each other. If the gas mixture is pure, all the atoms/molecules are near as dammit identical. All the constraints are thus "external" - summed up as the volume created by the walls of a container and the degree of negentropy present as part of the initial conditions.

    But if we want to pull the same trick with the brain~mind connection, well we are dealing with something much more complicated obviously. We may want supervenience, a scientific hierarchy of explanation, in some sense, but it is going to have to be a clever model.

    Then my further point is that supervenience is taken to be about just "bottom-up" causality. The ideal gas is so stripped down that all top-down constraints are abstracted away. This is clearly unreal. After all, some experimenter has to build the container, confine the particles. It just does not fluctuate into existence (even if the MODEL suggests such things could happen given infinite time).

    When it comes to consciousness, or complex systems in general, it is so clear that we are also dealing with top down constraints, or downwards causation, that the success of the ideal gas approach is even more misleading philosophically.

    With complexity, the macro creates the micro just as much as the micro creates the macro. It is not the case that one is emergent or supervenient on the other. The causality is mutual and so "supervenient" - one atop the other - just becomes a wrong word.

    So to answer the OP, perhaps the best that can be said is only a very non-mental example, like an ideal gas, in fact comes at all close to a pure supervenient relationship in our modelling. The mental would be a particularly bad candidate for jamming into this kind of simple-minded reduction to bottom-up causality.

    I could go off on another tangent about how the whole actually does shape its parts when it comes to the brain~mind. There is this issue of molecular turnover in brains.

    Well, whoops, I already have....
    (published in both Lancet Neurology and Science and Consciousness Review)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook