A dubious argument for panpsychism

At Aeon, philosopher Philip Goff argues for panpsychism:

Panpsychism is crazy, but it’s also most probably true

It’s a short essay that only takes a couple of minutes to read.

Goff’s argument is pretty weak, in my opinion, and it boils down to an appeal to Occam’s Razor:

I maintain that there is a powerful simplicity argument in favour of panpsychism…

In fact, the only thing we know about the intrinsic nature of matter is that some of it – the stuff in brains – involves experience… The theoretical imperative to form as simple and unified a view as is consistent with the data leads us quite straightforwardly in the direction of panpsychism.

…the brains of organisms are coloured in with experience. How to colour in the rest? The most elegant, simple, sensible option is to colour in the rest of the world with the same pen.

Panpsychism is crazy. But it is also highly likely to be true.

I think Goff is misapplying Occam’s Razor here, but I’ll save my detailed criticisms for the comment thread.

656 thoughts on “A dubious argument for panpsychism

  1. keiths: Your house is an external object. If it is not represented in your brain…

    I think petrushka meant it a bit differently: The house is sure represented, but the representation cannot be read off from the brain.

    This is my own view anyways.

  2. Erik: It’s behind paywall. I want to debunk it, but not by giving them money…

    I mostly share your instinct of wanting to debunk it — except that I don’t think it would be worth the effort.

  3. keiths: Your house is an external object. If it is not represented in your brain, how are you able to draw the floorplan from the inside of a CIA lab, where your house is not in view?

    That’s just silly. Does a spider have a representation of a web in it’s brain, or does it have a brain evolved to make certain kinds of turns under certain circumstances?

    You have the ability to do certain things, one of which is to describe your house. A set of sequential behavioral triggers is not a representation.

    You may wish to disagree, and at this point in our understanding, I cannot offer a formal proof that my assertion is true. But it makes a difference if you are engaged in AI research or in development of learning machines.

    I could be completely wrong in my understanding of what you mean by representation, but what I think you mean is a set of data that can be mapped to points in the external object. I don’t think brains have that kind of memory.

  4. petrushka:

    There is no representation of external objects or events in brains. That is not a philosophical position, it is a fact.

    keiths:

    Suppose I were to abduct you and take you by black helicopter to an undisclosed location run by the CIA, where you were asked to draw the rough floorplan of your house. I suspect you’d be able to do it; most people would.

    Your house is an external object. If it is not represented in your brain, how are you able to draw the floorplan from the inside of a CIA lab, where your house is not in view?

    petrushka:

    That’s just silly. Does a spider have a representation of a web in it’s brain, or does it have a brain evolved to make certain kinds of turns under certain circumstances?

    So your argument is “spiders don’t represent their webs, therefore humans don’t represent their houses”? Now that’s a silly argument.

  5. petrushka:

    You have the ability to do certain things, one of which is to describe your house. A set of sequential behavioral triggers is not a representation.

    Imagine a wooden ramp the width of a bowling ball, about ten feet tall, with a ski-jump style “hook” at the bottom. The ramp is placed in the middle of a gymnasium floor, with no surrounding obstacles. You release a bowling ball from the top; there are lips on the ramp to keep the ball from rolling off the sides. The ball rolls down the ramp and takes the jump. What happens next?

    The ball arcs through the air, landing on a spot on the floor. You mark that spot with an X. You then cover that X, and the surrounding area, with egg cartons full of fresh eggs. You take the ball back to the top of the ramp and release it again. What happens? What can you say about the state of the eggs afterwards?

    If you’re a normal person, you are able to visualize that entire scenario and predict what will happen to the eggs.

    You mentally represent the ramp, the ball, the gymnasium floor, the eggs, and the physics of all of the above, in order to predict the outcome.

    It ain’t just “a set of sequential behavioral triggers” sans representations.

  6. keiths:

    As if your comment about beer bottles were relevant to the question of whether petrushka is a functionalist.

    walto:

    Of course it’s relevant, as any reader of Searle would have understood. In any case, behaviorism is generally considered an antecedent to functionalism.

    Neither of those things is relevant. If someone matches the definition of a functionalist, then they’re a functionalist.

    Whether behaviorism is an antecedent to functionalism doesn’t change that, and neither does their opinion regarding the feasibility of using beer bottles, carburetors, or back issues of TV Guide to do what brains do.

  7. keiths:

    Humans didn’t bring stars into existence, Neil. They existed long before us.

    Neil:

    That is not actually relevant.

    keiths:

    Sure it is. It shows that your statement is wrong:

    There is no “way things are”. There is only the way that we say things are.

    Neil:

    Actually, no. Rather, it shows a problem in how you conceive our relation to the world.

    What problem? Do you actually deny that stars existed before humans, and that that was the way things were?

  8. petrushka,

    I read the description of the hard problem and still don’t see what the problem is.

    I’m not sure I can help you, then. Even people who deny the hard problem (Dennett, for instance) generally understand why it is seen as a problem by others.

  9. petrushka:
    You have the ability to do certain things, one of which is to describe your house. A set of sequential behavioral triggers is not a representation.

    Maybe. But this does not mean representations don’t exist under other definition. For example, keiths said, “Imagine a wooden ramp…” When you do that, you have a specific representation (of a wooden ramp) that you perceive by introspection.

    Interestingly, as I read KN’s article, looks like “neural representation” is defined there in the way you do: The definition includes (and seems to depend on) behaviors triggered. Basically the article debunks itself by its own definitions, no help needed.

    ETA: Yup, and (commendably) they see it themselves: “Before getting to the empirical evidence, a few caveats are in order. […] our account is limited to sensory, uncoupled, and motor neural representations. We are not giving an account of what is distinctive about mental representations. We are not attempting to explain full-blown mental or linguistic intentionality, including the ability to represent nonexistent objects like unicorns or abstractions like numbers, the possession of non-natural meaning in Grice’s (1957) sense, or the ability to attach different senses to the same referent (Frege 1892). ”

    Already the term “neural representation” raised doubts whether they could be talking about anything interesting. They correctly use the term to exclude everything truly interesting about representations. They use the term to describe an intermediate point in the perception-reaction chain. That’s uninteresting.

  10. KN stated,

    Kantian Naturalist: When neuroimaging (first CAT, then MRI and fMRI) improved to the point where neuroscience could be integrated with cognitive science, there was a slow shift from thinking about mental representations as symbols to neural representations as icons.

    The article states,

    There has been a lot of research on this topic, but it is clear that neural representations are not sufficient for conscious awareness. Hence, we will not be entering debates about consciousness, and reject any view in which consciousness is necessary for the existence of representational content.

    So, contra KN, there has been no shift from mental representations to neural representations. The authors of the article recognize with complete clarity that they are talking about something unrelated to mentality, awareness, and consciousness. KN of course insists on wild conflation, but that’s just KN as usual.

  11. Kantian Naturalist: I recently read two fascinating papers that bear on this: The cognitive neuroscience revolution and From symbols to icons: the return of resemblance in the cognitive neuroscience revolution. (Sadly both papers are behind paywalls but I can send PDFs to those interested.)

    The latter is freely available and an interesting read. But false. From the article,

    The history of philosophy has given rise to two major rival views about the nature of mental representation (cf. Waskan 2006). On one view—the intellectual provenance of TCS—mental representations are language-like entities and cognition is a matter of what Hobbes famously called ratiocination. In contrast to this, an older tradition—associated with figures like Aristotle, the Scholastics, and the British empiricists—holds mental representation to be founded on similarity to the mind’s objects

    The falsity here is the statement that these two views are rivals to each other. They are not.

    Language-like entities (symbols), being arbitrary, can have any shape. For example, the word “house” represents a house, without being similar to it in any way nor having any necessary connection to it. Take French “maison” – meaning the same, but what’s the similarity? Just the “s” is the same, but does this in any way represent a house necessarily? Nope.

    Further, being arbitrary and able to have any shape, symbols can even appear in the form of direct resemblance. For example, whether you write “house” on a piece of paper or draw the shape resembling a house, they both refer to a house. There is no contradiction, competition, rivalry, nor mutual exclusivity.

    Some people are more concretist-literalist, others more capable of abstract thought. The latter know better how language, mind, and consciousness works.

  12. keiths: So your argument is “spiders don’t represent their webs, therefore humans don’t represent their houses”? Now that’s a silly argument.

    Brains, however big, are about doing stuff, not about storing blueprints.

  13. keiths: You mentally represent the ramp, the ball, the gymnasium floor, the eggs, and the physics of all of the above, in order to predict the outcome.

    The word represent is the sticking point. I do not think of seeing or visualizing as equivalent to having a representation. My thought is that brain states are more like DNA. There is no blueprint in DNA, and there are no blueprints in brains.

    Brains do not store images.

  14. The reason I do not think much of the hard problem is not because I don’t think it is a problem, or that I don’t think it is hard, but because I don’t think it is a special class or category of problem.

    For the same reason I don’t think abiogenesis is a special kind of problem.

  15. keiths:
    keiths:

    walto:

    Neither of those things is relevant.If someone matches the definition of a functionalist, then they’re a functionalist.

    Whether behaviorism is an antecedent to functionalism doesn’t change that, and neither does their opinion regarding the feasibility of using beer bottles, carburetors, or back issues of TV Guide to do what brains do.

    As usual, everybody is wrong but keiths. And WHY does he think the SEP article is mistaken when it says that Skinnerian behaviorism was a precursor to functionalism? Well, because he said something different from that, and that’s a goddam nuff. He’s batting 1000 I tell you! Even with the whole world arrayed against him! A titan among boys.

  16. petrushka: The reason I do not think much of the hard problem is not because I don’t think it is a problem, or that I don’t think it is hard, but because I don’t think it is a special class or category of problem.

    It’s about like asking what “2” really consciously is.

    Not a meaningless question, but why is it especially important?

    Glen Davidson

  17. walto:

    As usual, everybody is wrong but keiths. And WHY does he think the SEP article is mistaken when it says that Skinnerian behaviorism was a precursor to functionalism?

    Who said the SEP article was wrong? Dang, walto.

    Read this again:

    Neither of those things is relevant.If someone matches the definition of a functionalist, then they’re a functionalist.

    Whether behaviorism is an antecedent to functionalism doesn’t change that…

  18. petrushka, today:

    The reason I do not think much of the hard problem is not because I don’t think it is a problem…

    petrushka, yesterday:

    I read the description of the hard problem and still don’t see what the problem is.

  19. petrushka,

    Brains, however big, are about doing stuff, not about storing blueprints.

    “Representation” is not synonymous with “blueprint”.

    In my bowling ball example, the ball is mentally represented as a solid, spherical object of a certain size, weight and toughness. The representation isn’t a blueprint for making bowling balls, and it doesn’t need to be.

  20. Glen, re the hard problem:

    Not a meaningless question, but why is it especially important?

    Because consciousness is important, and being able to explain consciousness matters to us.

    There are even ethical implications. Petrushka alluded to that when he wrote:

    I think we value computers precisely because they can be guilt free slaves.

    “Guilt-free” because “not capable of suffering”. But what if we create machines that are capable of suffering?

  21. petrushka: The reason I do not think much of the hard problem is not because I don’t think it is a problem, or that I don’t think it is hard, but because I don’t think it is a special class or category of problem.

    Exactly. It’s reification.

  22. The intereesting thing (to me) is that we have living human beings who report not having experiences that most of us consider essential to being human.

    Empathy is one example, but I can think of a less controversial example.

    There is the case of the artist who due to a stroke-like event, lost his color vision.

    Not just his ability to see color, but his ability to remember the experience of color.

    To me, this is an irrefutable demonstration of a definable bit of brain tissue being responsible for a specific experience of qualia. The problem posed by this is hard and unsolved, but no more so than other problems in understanding brains.

  23. keiths: “Guilt-free” because “not capable of suffering”. But what if we create machines that are capable of suffering?

    See Asimov, Phillip K. Dick and many others.

  24. petrushka,

    See Asimov, Phillip K. Dick and many others.

    Right. The ethical issues are real. Even if the hard problem weren’t intellectually interesting to us, we’d have moral reasons for investigating it.

  25. petrushka,

    To me, this is an irrefutable demonstration of a definable bit of brain tissue being responsible for a specific experience of qualia. The problem posed by this is hard and unsolved, but no more so than other problems in understanding brains.

    It’s a much harder problem. Compare it to the ability to do addition, for instance.

    We don’t yet know the details of how human brains do arithmetic, including addition. But we’ve built machines with the right capabilities, and we know why they’re able to do addition. The ability of matter to do arithmetic is not a mystery to us. It’s just information processing in both cases.

    The mystery concerns how information processing gives rise to felt experiences such as pain. A toothache appears to be more than just information processing. A toothache it isn’t just the knowledge that there’s something wrong with a tooth. It hurts!

  26. petrushka:

    The reason I do not think much of the hard problem is not because I don’t think it is a problem, or that I don’t think it is hard, but because I don’t think it is a special class or category of problem.

    Alan:

    Exactly. It’s reification.

    You need to bone up on the concept of reification, Alan.

  27. keiths: The mystery concerns how information processing gives rise to felt experiences such as pain.

    I understand the mystery, but it’s not a magical mystery. It is, like all unsolved problems, difficult, but nothing is added by insisting on some woo-like level of difficulty.

  28. petrushka:

    I understand the mystery, but it’s not a magical mystery.

    Who said it was magical? It’s “the hard problem”, not “the magical, unsolvable problem”.

    It is, like all unsolved problems, difficult…

    It’s not true that all unsolved problems are difficult. Problems vary enormously in terms of ease or difficulty. The problem of explaining why a certain pattern of neural activity actually hurts is much more difficult than the problem of explaining why a different pattern of neural activity accomplishes the addition of two numbers.

    In the latter case, we are explaining one third-person phenomenon in terms of another. That is routine in science. Explaining first-person phenomenology in terms of third-person neural activity is another beast entirely, and it’s what makes the hard problem hard.

    …but nothing is added by insisting on some woo-like level of difficulty.

    Again, that’s not what I’m doing. I’m quite optimistic that the hard problem will be solved without the need for any “woo”.

  29. I’m not optimistic in the short run. We have maps of brain structure an function, and we are building emulators based on the maps. At the moment that looks a bit like cargo cult engineering. Maps are not the territory.
    Finding an alternate substrate is not going to be easy, and that is the hard part of the problem. Electronics are not faster than chemistry. That is a misconception left over from the days when neurons were thought of as transistors. In fact, there have been attempts to use DNA for computation.

  30. keiths:

    I’m quite optimistic that the hard problem will be solved…

    walto:

    What is the basis of your optimism?

    It’s based on a few things:

    1. The hard problem might already have been solved (or shown to be illusory, which amounts to the same thing). Perhaps one of these guys (Dennett, Tononi, Graziano, etc.) has truly solved it and is just waiting for the laggards (including me) to recognize that.

    2. There may be a crucial insight or change of perspective that suddenly renders the hard problem much easier. Breakthroughs like this are not uncommon in science. Perhaps we’re just waiting for an “Einstein of consciousness” to come along and supply us with that crucial insight.

    3. There are intelligences much greater than ours on the horizon (augmented humans and artificial intelligences). Even if we can’t solve the problem, I’m optimistic that they will be able to.

  31. keiths,

    For what it’s worth, my pessimism is based on the fact that no heavyweight problem of philosophy has ever been “solved.”. The ” hard problem” doesn’t seem like a scientific problem to me…and so I think it’s basically doomed.

  32. walto,

    For what it’s worth, my pessimism is based on the fact that no heavyweight problem of philosophy has ever been “solved.”.

    That’s at least partly because philosophical problems often get reclassified as scientific problems once someone figures out how to approach them empirically.

    Think of philosopher Auguste Comte’s infamous pronouncement regarding the composition of stars:

    On the subject of stars, all investigations which are not ultimately reducible to simple visual observations are … necessarily denied to us. While we can conceive of the possibility of determining their shapes, their sizes, and their motions, we shall never be able by any means to study their chemical composition or their mineralogical structure … Our knowledge concerning their gaseous envelopes is necessarily limited to their existence, size … and refractive power, we shall not at all be able to determine their chemical composition or even their density… I regard any notion concerning the true mean temperature of the various stars as forever denied to us.

    Within just a few decades, Comte was proven wrong and those questions became answerable by science.

  33. I don’t think the composition of stars is an age-old philosophical problem. The nature of consciousness is.

  34. People have been wondering (and speculating) for millennia about the nature of stars, walto.

  35. Good for them. They often wonder about curing the common cold too. I’ve got hundreds of philosophy books in my house, and thousands of papers.

    Not a single mention of what stars are made of. Sorry.

  36. What does speed of computation have to do with the hard problem?

    This:

    The problem of qualia can only be attacked via emulation or duplication in a non biological substrate. Thinking about it gets nowhere.

    There are two problems in emulation. One is figuring out what neurons and brains do. The other is replicating this in another medium. We are a ways away from solving the performance problem. And neurons are not the only things that needs to be emulated.

  37. keiths:

    But what does relative speed have to do with the hard problem?

    petrushka:

    This:

    The problem of qualia can only be attacked via emulation or duplication in a non biological substrate. Thinking about it gets nowhere.

    How will we know that we’ve “duplicated” qualia on a non-biological substrate? What test can you perform to determine the presence or absence of qualia?

  38. walto:

    I’ve got hundreds of philosophy books in my house, and thousands of papers.

    Not a single mention of what stars are made of. Sorry.

    You’re making my point for me. It’s been reclassified as a scientific problem, just as I described above.

  39. walto,

    In any case, Comte’s statement is about what we can and cannot know. That’s epistemology, and epistemology is obviously philosophy.

  40. walto: For what it’s worth, my pessimism is based on the fact that no heavyweight problem of philosophy has ever been “solved.”. The ” hard problem” doesn’t seem like a scientific problem to me…and so I think it’s basically doomed.

    With some haggling over what counts as a “heavyweight problem,” I’m inclined to agree. Scientific problems are solved by devising a technique of measurement (if the solution doesn’t involve a scientific revolution) or by devising a new way of thinking about measurement (which is one way about what scientific revolutions are). Philosophical problems aren’t like that.

    But philosophical problems certainly do change over time, in response to new developments in science, art, religion, and politics. The modern problem of political philosophy, how to reconcile the demands of liberty and equality, would have been unintelligible to Aristotle. The modern problem of philosophy of mind, how to understand intentionality and consciousness in light of mechanistic science, would have been unintelligible to pretty much everyone before Descartes invented it.

    A different but related point that bears on the hard problem of consciousness: as usually understood the hard problem hinges on what is conceivable. Can we conceive of functional structures without any accompanying awareness or feeling? But what one finds to be conceivable varies with background knowledge. I’m not really sure how intuitively plausible the worries about the hard problem will look if one evaluates in light of our currently best available cognitive science and mindfulness practices.

  41. keiths:
    walto:

    You’re making my point for me.It’s been reclassified as a scientific problem, just as I described above.

    No, that’s just wrong. What stars are made of is an empirical question, and it was one that we were technologically incapable of answering until recently. What minds are made of is not an empirical question. At one time, all scientific questions were called “natural philosophy”–but at least since the 18th Century that has not been the case. To do philosophy nowadays, one really needs to be able to tell the difference between empirical and legitimately philosophical matters. What stars are made of was always of the former type, what consciousness is, has never been, and I believe never will be. Congrats on your optimism, but no heavyweight philosophical question has ever really been ‘solved’ by science or anything else.

    In my view, these are legitimate questions that have meaningful answers, but they aren’t empirical questions and never will be.

  42. Kantian Naturalist: With some haggling over what counts as a “heavyweight problem,” I’m inclined to agree. Scientific problems are solved by devising a technique of measurement (if the solution doesn’t involve a scientific revolution) or by devising a new way of thinking about measurement (which is one way about what scientific revolutions are). Philosophical problems aren’t like that.

    But philosophical problems certainly do change over time, in response to new developments in science, art, religion, and politics. The modern problem of political philosophy, how to reconcile the demands of liberty and equality, would have been unintelligible to Aristotle. The modern problem of philosophy of mind, how to understand intentionality and consciousness in light of mechanistic science, would have been unintelligible to pretty much everyone before Descartes invented it.

    A different but related point that bears on the hard problem of consciousness: as usually understood the hard problem hinges on what is conceivable. Can we conceive of functional structures without any accompanying awareness or feeling? But what one finds to be conceivable varies with background knowledge. I’m not really sure how intuitively plausible the worries about the hard problem will look if one evaluates in light of our currently best available cognitive science and mindfulness practices.

    This is better, but my pessimism continues. Consider the questions you mentioned about liberty and equality. They’re no closer to receiving anything like a consensus answer than they ever were.

Leave a Reply