On panpsychism

PanpsychismPanpsychism is in the news. Check out, for instance, this Oxford University Press blog entry by Godehard Brüntrup and Ludwig Jaskolla. Brüntrup is the Erich J. Lejeune Chair at the Munich School of Philosophy, has published a monograph on mental causation, and is the author of a bestselling introduction to the philosophy of mind. Jaskolla, in turn, is a lecturer in philosophy of mind at the same school, his research focusing on the metaphysics and phenomenology of persons, the philosophy of psychology, and the philosophy of action.

In other words, these are serious people. And so is the paladino-par-excellence of panpsychism, NYU’s David Chalmers. Why, then, are they lending their weight to such a bizarre notion? Let’s talk about it.

The essay begins with a quote by neuroscientist Christof Koch, to the effect that “as a natural scientist, I find a version of panpsychism modified for the 21st century to be the single most elegant and parsimonious explanation for the universe I find myself in.” After which I am compelled to comment that, as a natural scientist myself, panpsychism seems to me both entirely unhelpful and a weird throwback to the (not so good) old times of vitalism in biology.

But according to Brüntrup and Jaskolla, panpsychism is enjoying a Renaissance these days, as attested, among other things, by physicist Henry Stapp’s A Mindful Universe, which embraces a version of the notion strongly reminescent of the thinking of Alfred North Whitehead (the same guy who came up with the quote that gives the name to this blog: “The safest general characterization of the European philosophical tradition is that it consists of a series of footnotes to Plato”).

As the OUP authors correctly note, panpsychism is far from being a new notion in philosophy. Before Whitehead, it was endorsed in one form or another by Giordano Bruno, Gottfried Leibniz, and Teilhard de Chardin, and in fact goes back to Thales of Miletus’ claim that “soul is interfused throughout the universe.”

(It was also a crucial notion in Stoic physics, and being myself a practicing Stoic, you would think I would be sympathetic to it. I’m not.)

In 1979, always controversial philosopher Thomas Nagel (also currently at NYU) wrote Mortal Questions, in which he argued that neither “reductive materialism” (as he calls a scientific approach to the study of mind) nor mind-body dualism (obviously) are going to solve the problem of consciousness, which, to him, almost seems like a miracle. Nagel, therefore, saw panpsychism as possibly “the last man standing” on the issue, winning by default, though it isn’t clear why what is essentially an argument from ignorance (science at the moment hasn’t the foggiest about how consciousness emerges when matters organizes in certain ways, therefore science will never know) should carry any weight whatsoever.

(Recall that Nagel was at it again, recently, with a new book criticizing the scientific approach to the understanding of the cosmos as permanently incomplete because it cannot explain either consciousness or morality. The book, Mind and Cosmos, was not favorably reviewed, either by scientists or by philosophers.)

But, as Brüntrup and Jaskolla correctly point out, it is with Chalmers’ 1996 The Conscious Mind that modern panpsychism gets serious, so to speak. And it is therefore time to actually get down to work and see what panpsychism is and why people might be tempted to take the notion seriously.

According to Brüntrup and Jaskolla, “panpsychism is the thesis that mental being is an ubiquitous and fundamental feature pervading the entire universe,” which is an idea that is actually harder to pin down than it sounds. Does that mean that the iPad on which I’m typing this is (partially) conscious? What about the coffee that I just drank as part of my morning intake of caffeine? What about every single atom of air in my office? Every electron? Every string (if they exist)?

The OUP piece goes on to provide a helpful summary of the two cardinal ideas underlying panpsychism. I will quote it verbatim, then we’ll get down to do some unpacking:

“The genetic argument is based on the philosophical principle ‘ex nihilo, nihil fit’ – nothing can bring about something which it does not already possess. If human consciousness came to be through a physical process of evolution, then physical matter must already contain some basic form of mental being. Versions of this argument can be found in both Thomas Nagel’s Mortal Questions (1979) as well as William James’s The Principles of Psychology (1890).”

“The argument from intrinsic natures dates back to Leibniz. More recently it was Sir Bertrand Russell who noted in his Human Knowledge: Its Scope and its Limits (1948): ‘The physical world is only known as regards certain abstract features of its space-time structure – features which, because of their abstractness, do not suffice to show whether the world is, or is not, different in intrinsic character from the world of mind.’ (Russell 1948, 240). Sir Arthur Eddington formulated a very intuitive version of the argument from intrinsic natures in his Space, Time and Gravitation (1920): ‘Physics is the knowledge of structural form, and not knowledge of content. All through the physical world runs that unknown content, which must surely be the stuff of our consciousness.’ (Eddington, 1920, 200).”

Okay, then, let us consider the “genetic argument” first. The “ex nihilo, nihil fit” bit is so bad that it is usually not taken seriously these days outside of theological circles (yes, it is a standard creationist argument!). If we did, then we would not only have no hope of any scientific explanation for consciousness, but also for life (which did come from non-life), for the universe (which did come from non-universe or pre-universe), and indeed for the very laws of nature (where did they come from anyway?).

Then there is a nonsensical bit: “If human consciousness came to be through a physical process of evolution, then physical matter must already contain some basic form of mental being.” To see why this is nonsense, just substitute any qualitatively novel physical or biological phenomenon for “consciousness” and any physical process whatsoever for “evolution” and you’ll see what I mean. The genetic argument, in other words, reduces to an argument from ignorance, or from personal incredulity. It is a non starter from which to build any serious philosophical or scientific position.

On to the second pillar, the “argument from intrinsic nature.”

First off, notice the name dropping and do not be too impressed by it. Just because Leibniz, or Russell (one of my favorite philosophers ever) said something, and even when that something is alost believed by, wait for it, a scientist!, it doesn’t mean it’s a good idea.

Second, notice that Russell doesn’t come anywhere near making an argument for panpsychism. At best, that quotation can be interpreted as a statement of agnosticism, and more probably as an example of a philosopher entertaining a notion without necessarily endorsing it (a lot of philosophy is like that).

What about Eddington’s “intuitive” form of the argument? “Physics is the knowledge of structural form, and not knowledge of content. All through the physical world runs that unknown content, which must surely be the stuff of our consciousness.”

Surely. I’m not positive if my physicist friends would agree that physics is the study of structural form but not content — whatever those two terms actually mean in this context. But if so, then this is simply an argument for the incompleteness, as a science, of fundamental physics. Which, of course, is why we have a number of other sciences that study “content,” chiefly — in the case of consciousness — biology.

Consciousness, so far as we know, is an evolved property of certain kinds of animal life forms equipped with a sufficiently complex neural machinery. There is neither evidence nor any reason whatsoever to believe that plants or bacteria are conscious, let alone rocks, individual molecules of water, or atoms.

Moreover, since at the very bottom matter dops not seem to be made of discrete units (there are no “particles,” only wave functions, possibly just one wave function characterizing the entire universe), it simply isn’t clear what it means to say that consciousness is everywhere. Is it a property of the quantum wave function? How? Can we carry out an experiment to test this idea?

One more try. Brüntrup and Jaskolla conclude their piece by quoting Freya Mathews, the author  of For the Love of Matter (2003), as presenting an argument that assumes that — somehow — we have privileged access to the ultimate nature of matter by virtue of being conscious. Here is how Mathews puts it: “the materialist view of the world that is a corollary of dualism maroons the epistemic subject in the small if charmed circle of its own subjectivity, and … it is only the reanimation of matter itself that enables the subject to reconnect with reality. This ‘argument from realism’ constitutes my defense of panpsychism.” (Mathews, 2003, 44)

You call this an “argument”? It seems to me to be a massive exercise in begging the question, accompanied by poetic but highly imprecise phrasing (materialism maroons the epistemic subject…). And I simply don’t get why dualism is supposed to be a corollary of the materialist view of the world — I would have thought the two to be incompatible. This isn’t the stuff of either good philosophy or good science.

I am left with just one question: why would anyone take any of this seriously at all? And one possible answer comes right at the end of the OUP post:

“Panpsychism paints a picture of reality that emphasizes a humane and caring relationship with nature due to its fundamental rejection of the Cartesian conception of nature as a mechanism to be exploited by mankind. For the panpsychist, we encounter in nature other entities of intrinsic value, rather than objects to be manipulated for our gain.”

I get it, panpsychism allows us to feel at one with nature because consciousness is everywhere, and that will make us better shepherds of nature itself. I got news: Nature is mind bogglingly bigger than humanity, and it will be here for eons after humanity will be gone. It doesn’t need us to feel connected with her in order to exist. Yes, we do need to take care of our own puny piece of Nature that we call Earth, for our own sake, if nothing else. But we can do that quite independently of either Cartesian dualism or New Age panpsychism. We can do it as material creatures endowed by evolution of the ability to reflect on what they are doing and decide whether it’s a good idea to do it.

(Much more on panpsychism can be found in the corresponding entry in the Stanford Encyclopedia of Philosophy, including a third category of arguments, not considered here, and a brief overview of standard counter-arguments.)

Advertisements

285 thoughts on “On panpsychism

  1. Robin Herbert

    Excellent article. I didn’t know much about the claims of panpsychism before. Seems like I might go back to not thinking about it much.

    It would seem that Naturalism is the, er, natural home for anyone who wants to be one with Nature. Not that we could be anything else but one with Nature.

    Liked by 1 person

  2. Disagreeable Me (@Disagreeable_I)

    Hi Massimo,

    what is essentially an argument from ignorance

    I don’t think anyone is making an argument from ignorance to claim that science will never understand consciousness. To me, the claim that science will never understand consciousness is much more like the claim that science will never solve the trolley problem. The (hard) problem of consciousness seems to me to be not a scientific but a philosophical issue. Science can study how the brain performs various functions, but I don’t think it can answer why it is that there is something it is like to be a person with a brain. This is because phenomenal experience is irreducibly subjective. All science has access to, in principle, are the objective facts, and it’s seems to many of us that no amount of objective facts add up to subjective experience.

    Unlike Nagel, however, I don’t think this means that science is in some sense “incomplete”, or that there is something “more” going on in the sense of something mystical or magical or whatever. I just think it means that the questions we ask ourselves about consciousness are philosophically confused, and we need philosophy to properly unpack them and ask the right questions. As you know, I think Dennett and his ilk have the right approach in this enterprise.

    To see why this is nonsense, just substitute any qualitatively novel physical or biological phenomenon for “consciousness”

    While I’m not impressed by the genetic argument either, I don’t think it is as bad as you, and I wouldn’t go so far as to use the word “nonsense”. The genetic argument shows that there is a smooth transition from things we would call conscious to things we would call unconscious. This implies that unconscious/conscious is more a continuum like small/big than a sharp category like even/odd. If consciousness is like small/big, then it’s not so crazy to suggest that everything might have the teensiest glimmer of consciousness, the way every lepton has at least some mass (but, even so, we wouldn’t say that an electron is massive).

    Where it breaks down is that the word “conscious” is used exclusively for things at one end of this spectrum, so to describe an electron as a little bit conscious is as absurd as to say it is very slightly gigantic. We need a different word, corresponding to “having mass”. It’s also clear from comparing it to non-living/living rather than small/big that just because there is a continuum doesn’t necessarily mean we can make sense of the suggestion that particles are conscious any more than that they are alive.

    If one of its two central pillars falls, then the entire construction comes down.

    Why do you say this? This doesn’t follow at all. This is just one argument for panpsychism. If this argument has weaknesses (and I agree that it does), there may yet be other arguments.

    I’m with you that the argument from intrinsic nature doesn’t work. It seems to me that you haven’t really argued against it, but that’s fair enough because the quotations you provided didn’t really argue for it! It all seems very wishy-washy and I don’t see that there is any serious argument there at all.

    I’m not positive if my physicist friends would agree that physics is the study of structural form but not content

    Which, of course, is why we have a number of other sciences that study “content,” chiefly — in the case of consciousness — biology.

    I don’t think that’s quite right. Structural form is the mathematics of physics, the underlying laws of physical reality. The content is what “breathes fire into the equations” and makes them real. This is the “baggage” Max Tegmark is talking about when he says that there is no baggage and reality is all mathematics. If there is content, then Tegmark is wrong, but I don’t think this content is what is studied in biology etc. The content is just what makes it all real as opposed to mathematics. It is not studied in any science, because there is nothing to study other than to note that the world the sciences study is real and not made up.

    There is neither evidence nor any reason whatsoever to believe that plants or bacteria are conscious, let alone rocks, individual molecules of water, or atoms.

    I think that’s a straw man. Even though panpsychists themselves might say these things are conscious, I think that this isn’t really capturing what they mean. I don’t think anybody believes these things are conscious in anything like the way a human being is conscious. The panpsychist view is just that these things can bear elements of consciousness out of which an actual consciousness can be built. I’ll agree with you that there is no evidence for this view, but I think it’s going too far to say there is no reason. There is the (rather weak) genetic argument, which suggests that maybe consciousness is a continuum like mass, and there’s the better (though still flawed) p-zombie argument of David Chalmers, which you didn’t mention in this article (a curious omission, but I understand you may not want to go through it again).

    Is it a property of the quantum wave function? How? Can we carry out an experiment to test this idea?

    Nobody knows, and it may be impossible to carry out such an experiment. The only thing a panpsychist claims is that there is something funny going on at a fundamental level in the way this universe is constructed that enables consciousness to exist. Whatever is going on, it is natural (and so this is not an appeal to the supernatural), but it has no known empirical consequences other than that we have phenomenal experience. This is no less reasonable than your view as a biological naturalist. On biological naturalism, there is something going on in human biology that enables consciousness to exist. Whatever is going on, it is natural, but it has no known empirical consequences other than that we have phenomenal experience. You can’t fairly fault the panpsychists for failing to grasp how it all works until you can yourself explain how exactly human biology produces consciousness.

    On Freya Mathews:

    You call this an “argument”? It seems to me to be a massive exercise in begging the question

    Agreed.

    I am left with just one question: why would anyone take any of this seriously at all?

    The p-zombie argument. That’s the beginning and the end of it for me at least. That and the realisation that it’s no less worthy of serious consideration than biological naturalism, which faces (to my mind) similar issues, in that both views assume without evidence and for flawed reasons that consciousness is substrate-dependent. The difference is only that for panpsychists, the substrate is the matter of this universe (as opposed to that of a p-zombie universe), whereas for biological naturalists, the substrate is neurons. In both cases, the substrate is supposed to have mysterious unknown properties that enable consciousness.

    More broadly, panpsychism is a natural consequence of (incorrectly, in my view) rejecting functionalism coupled with the (correct, in my view) insight that biology can never explain consciousness any more than the study of railway engineering can solve the trolley problem. Although I guess you could always turn to Cartesian dualism instead, if that’s any better.

    It doesn’t need us to feel connected with her in order to exist.

    Agreed. This last argument is also weak.

    Liked by 5 people

  3. Coel

    Presuming that it is on-topic to continue the previous discussion with DM,

    If you have simulated every physical aspect of a system, then every function physically realised in that system has a virtual analogue.

    I don’t understand what you mean by “simulate every physical aspect”. As I see it, if absolutely every physical aspect is reproduced then we’re talking about a replication. And, I would indeed regard a replication as conscious. To me the term “simulation” implies reproducing some salient aspects of a system but not others. I can simulate a bridge on a laptop but there are many physical aspects of the bridge that are not reproduced and thus the simulation cannot carry a truck across a river.

    In the same way that “supporting a truck” is a property of a real, physically extant thing — not of a simulation or computation — as I see it being conscious is a property of a real, physically extant brain.

    Like

  4. Philosopher Eric

    Thanks for indeed “giving ’em hell” Massimo. It’s wonderful to see you display agitation about this modern failure, even though many others would passively accept “new age excuses.”

    I personally doubt that humanity will ever solve the hard problem of consciousness, though I also consider this unnecessary. What I think we need are more useful sets of definitions for associated terms. For example, observe that in this modern age we still carry around Freud’s old “unconscious mind” term, or a strange “quasi consciousness” notion. That doesn’t help! Let’s instead say that there is conscious mind, as well as mind that isn’t conscious. And we’ll surely need functional models of mind if we are to get anywhere. Otherwise, like physics not so long ago, these mental sciences logically ought to remain “soft as jelly.”

    Like

  5. SocraticGadfly

    Per Coel, and to continue, “Substrate matters.” This is why I don’t argue this — or determinism, per my crack about deterministic panpsychism yesterday — with DM. Otherwise, per your last comment, I don’t see any formal difference between your “turn” and your suggesting a turn to Cartesian dualism.

    Like

  6. SocraticGadfly

    Per Massimo’s pull quote at the end, that’s the appeal for many panpsychists: It creates a strawman that only they are truly in touch with nature/the universe, etc. Like some Greens’ strawman against GMOs for similar reasons, also hearkening back to early 19th century Romanticism, that’s what it is.

    Like

  7. SocraticGadfly

    Also, DM, even if consciousness IS a “hard problem,” it in no way logically follows that panpsychism is therefore the answer. A clear refutation of that is the number of philosophers who agree with you that it is a hard problem but reject panpsychism.

    Like

  8. Disagreeable Me (@Disagreeable_I)

    Hi Coel,

    I don’t understand what you mean by “simulate every physical aspect”

    I mean represent every physical fact in the system, but without physically reproducing it. So that anything you can say about the physical system has a virtual counterpart in the simulation, but it doesn’t have a physical counterpart.

    Of course this is not realistic. I’m not aware of any computer simulation that simulates everything down to the tiniest detail. This might be practicable for very simple systems, such as that of a photon travelling in a vacuum.

    If we relax this criterion and only simulate some aspects of a physical system, then you ought to realise what I mean. If we simulate the collapse of a building 10m high, the 10m height of the building is represented in the simulation, but there is nothing physically in the simulating system that is 10m high. Similarly, in a simulation of a brain, there would be a representation of the way different neurons are connected, but we don’t physically have different actual components connected in this way.

    So, let’s suppose that the necessary and sufficient conditions for consciousness in a brain whatever they may be are among those represented and simulated in our simulation, but let’s stipulate that they are not physically present (so there are no actual neurotransmitters in the computer, but there may be an abstraction in the logic of the program that performs the same function).

    and thus the simulation cannot carry a truck across a river.

    Of course not. But it can carry a simulated truck across a simulated river. So anything you can say about a physical bridge has an analogue for the simulated bridge, unless you’re going into details that are not represented in the simulation (such as that the bridge is made of iron atoms).

    as I see it being conscious is a property of a real, physically extant brain.

    Right. But does a simulated brain (simulated in perfect detail with respect to all the functional aspects that are required for consciousness in a real brain) also have that property? You don’t know, right? And yet such a simulated brain could in principle drive a robot the way ours does our body. And so if it is not conscious, consciousness must be functionally inert.

    Like

  9. Disagreeable Me (@Disagreeable_I)

    Hi Socratic,

    Also, DM, even if consciousness IS a “hard problem,” it in no way logically follows that panpsychism is therefore the answer.

    I agree. I’m not saying panpsychism is the answer. It isn’t. Functionalism is. I’m only asking that you be fair to the panpsychists.

    Like

  10. SocraticGadfly

    I simply extend my previous comment, then. If functionalism is the answer, it by no means entails panpsychism as the likely answer within functionalism. Nor does it entail, under a naturalistic approach, special treatment (sic) for panpsychism.

    Like

  11. Disagreeable Me (@Disagreeable_I)

    Hi Socratic,

    Evidently, you’re not following me.

    If functionalism is the answer, it by no means entails panpsychism as the likely answer within functionalism.

    Of course it doesn’t. Panpsychism is not compatible with functionalism. Panpsychism is wrong. Nothing I have said is intended to imply that panpsychism is correct. I’m only attempting to defend against what I perceive to be unfair criticisms. Ultimately it still fails because the arguments that lead Chalmers to prefer panpsychism to functionalism are flawed.

    Like

  12. SocraticGadfly

    Well, then, all I can say, DM, is that your thought processes work a lot differently than mine. Because I would never invest such mental energy in defending “the right to a fair trial” of something I thought wrong.

    Like

  13. SelfAwarePatterns

    Well said Massimo! Panpsychism seems to be everywhere this week (Peter Hankins and Aeon both have articles on it.)

    It seems to me that panpsychism surrenders on the key question of what separates conscious systems from non-conscious ones. Incidentally, so does concluding that consciousness is an illusion. Deciding that everything is conscious, or that nothing is, both seem like avoidance of the question rather than solutions to it.

    I think the most productive view is that consciousness is a type of information processing involving a system that models its environment and itself, and the relation between the two, as a guide to its actions. Of course, this unavoidably means there’s an element of interpretation involved in whether or not a system is conscious, but that’s true of many concepts, such as life vs non-life (see viruses).

    Liked by 2 people

  14. SocraticGadfly

    Also, DM, if you think panpsychism is clearly wrong, presumably you came to that stance after giving it a fair hearing in your own thought processes, which would arguably make your own …. internal states 🙂 …. logically inconsistent.

    Like

  15. SocraticGadfly

    Two other quick notes.

    First, per Massimo, word substitution. “I think torture is wrong, but I so much think it hasn’t gotten a fair hearing that I’m going to strongly defend ‘a fair hearing.'”

    Maybe, per Dennettian subselves, some part of DM is defending more than just “a fair hearing.”

    There’s also the claim that panpsychism “hasn’t had a fair hearing.”

    Per all that Massimo cited, “tosh.”

    Like

  16. Disagreeable Me (@Disagreeable_I)

    Hi Socratic,

    Because I would never invest such mental energy in defending “the right to a fair trial” of something I thought wrong.

    I care very much about philosophy of mind. It’s something of a passion for me.

    What I value above all else in these matters is clarity of thought. I think that Chalmers writings on panpsychism are clearer than Massimo’s, even though I think Massimo is right that panpsychism is false. It annoys me to see Chalmers careful and thoughtful analysis of these issues dismissed so easily.

    Similarly, I have invested a lot of energy in trying to convince Coel that The Chinese Room argument — an argument I vehemently disagree with — is not as bad as he thought it was. In fairness to Coel, I think he somewhat came around, and now he gives that argument more credit than I do myself! I think in the end, we concluded that what Searle meant by “understand” was not what Coel thought he meant, and that Searle’s argument does seem to indicate that perhaps the CR doesn’t have a subjective experience of understanding — which is more or less all Searle was trying to show.

    Also, DM, if you think panpsychism is clearly wrong, presumably you came to that stance after giving it a fair hearing in your own thought processes, which would arguably make your own …. internal states🙂 …. logically inconsistent.

    There’s a joke in there somewhere, it seems, but I don’t get it. Sorry, I don’t know what you’re driving at.

    Liked by 1 person

  17. Disagreeable Me (@Disagreeable_I)

    Hi Socratic,

    First, per Massimo, word substitution. “I think torture is wrong, but I so much think it hasn’t gotten a fair hearing that I’m going to strongly defend ‘a fair hearing.’”

    A better analogy might be: “I also think torture is wrong, but I really think that you’re focusing too much on the evidence that it doesn’t work. Torture is wrong even if it does work, because … “.

    The reasons why we come to the conclusions we do are important. It matters how you justify your views.

    Like

  18. Daniel Kaufman

    The sort of thing that gives Philosophy a bad name. Definitely not what we need in these times, when Philosophy is already under attack from a number of quarters.

    Speaking of consciousness in something with no brain — or anything like it — is just a lot of nonsense, and trying to dodge that judgment by speaking of “mental being” — which I would argue is utterly meaningless — doesn’t help.

    This sort of intellectual train-wreck is best left alone. That so many people are buying into it — or even taking it seriously — says a lot about how a certain kind of New Agey spiritualism has managed to seep in to the public consciousness, albeit often unconsciously.

    Like

  19. Coel

    Hi Massimo,

    Consciousness, so far as we know, is an evolved property of certain kinds of animal life forms equipped with a sufficiently complex neural machinery.

    Agreed. And we can go further. We have some control over consciousness. We can turn it off and on by the simple expedient of flooding the neural machinery with sodium thiopental or propofol. Further we can (to some extent) alter and control qualia by introducing molecules such as ethanol, diamorphine, serotonin, LSD, nicotine, cocaine and all sorts of other things. What seems to matter is the particular molecular patterns of these substances, and how they interact with particular molecules in the neural machinery, thus affecting patterns of flows of ions. One can also change qualia by applying a magnetic field or electrical stimulation, which again affects the flow of ions.

    This is all pretty good evidence that “consciousness” is a property of particular dynamic patterns at the molecular and neural-network levels of description.

    Looking for additional and unknown properties at the level of particles such as electrons just seems unwarranted.

    Hi DM,

    And yet such a simulated brain could in principle drive a robot the way ours does our body. And so if it is not conscious, consciousness must be functionally inert.

    It would only make sense to describe something as functionally inert if it was an optional extra, as oppose to being an intrinsic part of the way a system is behaving.

    Let’s return to my analogy of turbulence in a liquid flow. One could manufacture some sort of control device, wherein turbulent liquid led to some output state, and this output state could be hooked up to some other device. The output state would then be different if the flow were laminar as opposed to turbulent.

    Now, one could then entirely simulate the physics of that device on a computer, and thus one could compute the output state, and so could instead hook up the simulation to the other device, with the exact same effect. The simulation would not have the property “turbulent”, though it would simulate turbulence. Would it then be fair to say that, in the original device, turbulence was “functionally inert” since the exact same effect could be arrived at without turbulence? No, I don’t think so, since way that effect is being arrived at in the original device does involve turbulence.

    In the same way, I would assert that consciousness is not functionally inert, in that it is an intrinsic part of the behaviour of a brain, even though a non-conscious simulation of that brain could arrive at the same output states.

    I think in the end, we concluded that what Searle meant by “understand” was not what Coel thought he meant, and that Searle’s argument does seem to indicate that perhaps the CR doesn’t have a subjective experience of understanding …

    Correct! I always thought the Chinese Room was about “understanding”, whereas I now gather that to many people understanding doesn’t count as “understanding” unless one is doing a conscious and sentient self-reflection on that understanding (which seems to me a whole ‘nother issue, and a way harder one).

    I may be at fault for not having realised this a lot earlier, but I can point to discussions of the CR by Searle himself (including the original 1980 one), and by Dennett and other notable philosophers that simply don’t make this clear (or even state it at all). (Maybe it’s regarded as too obvious to need stating, but it’s not at all obvious to some of us! Sometimes I wonder whether Dennett’s disagreement with Searle is because he, like me, is making a rather different interpretation of what “understanding” amounts to.)

    Like

  20. Disagreeable Me (@Disagreeable_I)

    Hi Coel,

    It would only make sense to describe something as functionally inert if it was an optional extra, as oppose to being an intrinsic part of the way a system is behaving.

    Agreed.

    If functionalism is true, then consciousness is just an intrinsic property of any system that behaves like a human brain. It doesn’t matter how that system is realised. If it behaves in the same way (internally as well as externally — so it’s not using lookup tables) then it’s conscious.

    Would it then be fair to say that, in the original device, turbulence was “functionally inert” since the exact same effect could be arrived at without turbulence?

    OK, that’s a pretty good analogy, and it’s making me think.

    If turbulence is our analogy, and there is such a thing as simulated turbulence, then by analogy there is such a thing as simulated consciousness. So, the important question is whether there is an importance difference between “simulated consciousness” and “real consciousness” — is there something it is like to be a simulated person? If there is, then simulated consciousness is just consciousness.

    In your example is the actual physical realisation of the turbulence always important in building a system with the behaviour you describe? Well, obviously not. Simulated turbulence will do. So while physical turbulence is an important part of the original system, actual turbulence is not required in an alternative design which behaves in the same way. It is the “actuality” of the original physical system that is functionality inert, not the system itself, if you see what I mean, in that you can get the same behaviour with a virtual analogue. In contrast, rigidity and tensile strength are important properties in construction materials. You can’t replace these materials with a computer simulating rigidity or tensile strength. The actuality of the rigidity and tensile strength is not functionally inert.

    I just think it’s very fishy to suppose that we have evolved consciousness and a rich interior life if it is possible that we could have evolved with just the same sophistication in behaviour and having no interior life at all. If it is possible in principle to build such a system (e.g. by simulating a brain), then it seems to me that such a system could evolve naturally also. This is what I mean by functionally inert — the idea that consciousness is not required to get behaviour like that of a human (especially if we’re not using shortcuts like lookup tables). If we don’t need to be conscious to behave as we do, then why are we conscious?

    Sometimes I wonder whether Dennett’s disagreement with Searle is because he, like me, is making a rather different interpretation of what “understanding” amounts to.)

    No, I don’t think so. I think Dennett understands Searle just fine. Dennett’s positions and mine are basically the same.

    Like

  21. michaelfugate

    Water – is the easy example. Does a water molecule contain wetness or a snowflake? Does a Hydrogen or Oxygen atom? Does a proton or electron?

    Liked by 3 people

  22. Massimo Post author

    DM,

    “I don’t think anyone is making an argument from ignorance to claim that science will never understand consciousness. To me, the claim that science will never understand consciousness is much more like the claim that science will never solve the trolley problem”

    I think that’s an (interesting) disanalogy. The trolleys problem makes no claims about the structure and function of the world, it is a thought experiment meant to explore people’s intuitions about ethical priorities. So, right, there can be no scientific answer to the dilemma, though of course there is plenty of empirical research documenting how people actually respond to it.

    But panpsychism is making a very definite claim about the nature of reality, and I find it interesting that you say that that sort of claim is outside the purview of science. When Democritus claimed that everything is made of irreducible basic units called atoms, was that also an untestable claim?

    “The genetic argument shows that there is a smooth transition from things we would call conscious to things we would call unconscious.”

    Actually, the argument assumes, but does not “show* such a smooth continuum. Whether consciousness (or life, or whatever) is a qualitative leap or a quantitative one is an empirical question, but the argument is “nonsense” because if applied to other properties of matter it would sound silly. Water, for instance, is not “wet” below a certain threshold in terms of number of molecules, because wetness requires a minimum assemblage of molecules to manifest itself as a physical property. No magic involved, but it would be silly (i.e., “nonsensical”) to claim that wetness has to go all the way through, to the level of electrons. Electrons ain’t wet.

    “panpsychism. If this argument has weaknesses (and I agree that it does), there may yet be other arguments.”

    You are probably right there, I told the authors’ metaphor of “pillars” too literally. I will edit accordingly, thanks for pointing it out.

    “Structural form is the mathematics of physics, the underlying laws of physical reality. The content is what “breathes fire into the equations” and makes them real.”

    Than physics does deal with content, since physics is not (just) mathematics, pace Tegmark.

    “Even though panpsychists themselves might say these things are conscious, I think that this isn’t really capturing what they mean. I don’t think anybody believes these things are conscious in anything like the way a human being is conscious”

    You are probably right, but that’s because it isn’t at all clear what they are saying to begin with.

    “The only thing a panpsychist claims is that there is something funny going on at a fundamental level in the way this universe is constructed that enables consciousness to exist.”

    There isn’t anything funny, there is something we don’t understand (yet?). Hence my characterization of the whole thing as an argument from ignorance, or from personal incredulity.

    “The p-zombie argument. That’s the beginning and the end of it for me at least”

    Well, you know what I think of that argument, so no need to revisit.

    “panpsychism is a natural consequence of (incorrectly, in my view) rejecting functionalism”

    No. I’m not a functionalist about life either — and before you say it, no I’m no vitalist! — and for similar reasons. Materials matter, not just how they are arranged. It’s a physical universe we live in, without certain material properties it would be just math. Oh right…

    Like

  23. Disagreeable Me (@Disagreeable_I)

    Hi Massimo,

    When Democritus claimed that everything is made of irreducible basic units called atoms, was that also an untestable claim?

    I grant that the trolley problem is not a perfect analogy, but neither is this.

    The problem is that, unlike atomism, and unlike moral claims, we have no reason to believe that phenomenal consciousness could ever have any empirical consequences. I fail to see how any finding in biology could falsify the hypothesis that a computer could be conscious in virtue of the program it is running. There are two sides to the analysis of brains (and pretty much everything) — how they work and what they are made of. Biology can study both, but I can’t see how it can study the question of which of these is required for consciousness, because consciousness is not something we can observe. You can’t make an apparently functional brain out of silicon and then test it for phenomenal consciousness.

    Perhaps I’m wrong here, but I’m just failing to conceive of what a scientific solution to the problem could possibly look like, and my intuition leads me to believe that it is hopeless. I mean, invent any kind of fantasy you like about what kind of experiment we might one day be able to carry out, and I don’t think you’ll be able to come up with a hypothetical outcome that would demonstrate anything about whether consciousness is substrate-independent. I mean this sincerely. Try it as an exercise. What kind of empirical finding could ever settle the question? Democritus might at least have imagined that we could build machines to cut things into finer and finer pieces until we suddenly hit a wall and could divide no further.

    No magic involved, but it would be silly (i.e., “nonsensical”) to claim that wetness has to go all the way through, to the level of electrons.

    OK. I agree that it is silly to say that such a thing certainly follows. I clearly agree with you that it doesn’t. But I don’t think it is silly to think that these kinds of argument suggest that it might be the case that simple systems can have some sort of elementary proto-consciousness, the way electrons have mass.

    Insofar as panpsychists find the genetic argument to be conclusive, I agree with you that they are silly.

    Than physics does deal with content, since physics is not (just) mathematics, pace Tegmark.

    Insofar as physics is not just doing mathematics, I think it’s acting much like a special science like biology. But I think physics is mostly mathematics.

    Anyway, I don’t think that gets to my point about content. I think what was meant in the original quote was that content is what makes the equations real, what distinguishes abstract equations from the real world. Science studies the structures, but I don’t think there can be a scientific study of the content if content is just what distinguishes abstract fantasy from reality.

    I’m not a functionalist about life either

    I think life is a disanalogy, because life isn’t a real objective thing in the universe which is either objectively there or objectively not. Life is a family resemblance word humans use to group together certain kinds of structures. There is no fact of the matter on whether a virus is alive or not. It is just a matter of convention whether we say it is or not. There is no fact of the matter on whether a self-replicating robot would be alive. You might say it isn’t, I might say it is. We’re just disagreeing on what “alive” means.

    But whether something is conscious does matter. If an AI is conscious, it has an inner life. If it isn’t, it doesn’t. This is not just a matter of definitions.

    Like

Comments are closed.