I agree with David Chalmers’ basic argument: “Physical descriptions of the world characterize the world in terms of structure and dynamics. From truths about structure and dynamics, one can deduce only further truths about structure and dynamics”. Mary’s knowledge argument is a little weak because it carries a lot of implied assumptions: Mary is a human (which implies consciousness), she already can see a scale of grays, etc. I prefer the example of an alien creature from another dimension who is given all the physical laws governing our universe, so that it can simulate life on Earth. At most the alien would be able to describe what Chalmers calls “philosophical zombies”. There is no way for that alien to figure out that those earthly creatures have conscious feelings (other than by their vocal claims that they do, which the alien would not be able to understand anyway). So the complete knowledge of physical laws would still not be sufficient for the alien to have a complete understanding of what is going on, and new laws would need to be added to complete the description
if you're a physicalist and adhere to the church of magical mass-energy which can arise as colors sounds etc then yes the physical alien can eventually manipulate the physical to experience all those things directly
You make an assumption yourself: what the outcome of said alien simulation would yield. You can’t. As otherwise, there would be no point in running the simulation in the first place. E.g., such aliens might be able to replay parts of such a simulation in their heads and experience human consciousness themselves.
"complete understanding" what you want ? You want scientist to show you qualia by opening up brain ? It's so obvious that qualia which is building block of consciousness is first person experience. But it doesn't imply that it is correct and fundamental reality , there are lot of experiments that shows that qualia can be misleading and they are just modifications of actual objective reality
The Mary argument may be disadvantaged as a potential defeater of physicalism by virtue of the fact that Mary is already a conscious being, ie Mary already knows what is it like to have experiential or phenomenological consciousness. A knowledge argument which might side-stop this could be to ask: what if you feed an artificially intelligent entity all the conceivable data of a purely physical nature about a human being, and then asked it to produce the most complete description of the human being, would the AI know that the human being was conscious (or experienced itself as being conscious) at all? If yes, then how precisely does the AI extrapolate consciousness from the purely physical data (without having any prior/pre-existing direct experience of consciousness itself)? If no, then isn't the purely physicalist description of the human being necessarily incomplete?
Professor Jackson says that there is a new way to experience the information, but the argument is not that there isn't a new way, it is that the experience does not come through any way that a physicalist can come up with.
One major problem for panpsychism is that once you ascribe conscious properties to all matter you lose the capacity to define boundaries and such behavioural differences from non-organic matter to organic beings. Further you have to then attribute all cognitive capacities to inanimate objects from memory to disorders of consciousness which brings you to ridiculous conclusions such as rocks experience states like psychosis or deficits in theory of mind which are by definition observable behavioural or at least expressive definitions of intentional active beings
Why? We attribute gravitational properties to all matter, yet no one explicitly expects to observe pebbles orbiting mountains. Why wouldn't it be possible for all matter to have a "consciousness property", yet observable consciousness as we understand it only emerges when a concentration gradient is high enough relative to some other not yet understood factor.
@@christophercavill2664 You do in space - you expect pebbles to orbit asteroids. On Earth you expect pebbles to be influenced by all the matter nearby, ie the Earth as well as the mountain.
My internal examiner for my PhD in 1996 was Howard Robinson, and the external was Prof. E.J.Lowe. They both seemed quite happy with my insistence that qualia really exist and need an adequate account. What we are witnessing here is a throw-back to JJC Smart being able to distinguish between lettuce and cabbage leaves, but unable to explain in what respect. If conscious experience is just an illusion, we need an explanation of how that illusion works.
A rebuttal of the knowledge argument against physicalism was made nearly 2 decades ago. Firstly, there is a false premise that Mary will suddenly acquire colour knowledge once released as evidence would support the alternative scenario that she would continue to see the world monochromatically because acquisition of perceptual knowledge outside previous possible experience requires direct interaction with coloured objects. Crucially and this is the linchpin for physicalism her ability to acquire this new colour experience will depend on adult neural plasticity which will permit encoding of new action perception algorithms. I find it extraordinary that this obstacle against the knowledge argument has been marginalised
Great conversation as always. Jackson's reformulation of what Mary gains when leaving her colorless room, in terms of either immediate or derived information - in order to say the information is in some sense the same, with only the access to it changing - had me paused for a moment. But after the pause it occurred to me that the access is precisely what we're worried about. Which is to say the knowledge argument regards knowledge of data, rather than the broader category of information. It makes sense to say that a written text on the neurophysiology of color perception and the phenomenal experience of seeing color are two channels of data aiming at, in some sense, the same information. But what we were supposed to care about was whether her gaining access to the latter channel of data gave her knew knowledge. And framed in terms of 'knowledge of data', its clear she does. Regarding Jackson's newfound representationalism and analytic functionalism. One can get very far with a representationalist account of exteroceptive perception, as the objects perceived are in some sense almost always representations. With some elegant intellectual contortions one could argue that color is primarily just another dimension of distinguishing those representations. Interoceptive perception/sensation is more difficult to give a representationalist treatment, since the objects of those states, sensations, lack the obvious boundaries that exteroceptive perceptions have. Jackson and Frankish highlight their difficulty in being able to characterize those states, of pain for example, and that they are nearly always associated with some sort of behavioral inclination, as being fine evidence for a broadly functionalist treatment. Sensations are nothing more than functional inclinations toward particular behavior. There is much to say regarding this sort of perspective, mostly negative in my opinion, but in the interest of brevity I'll offer just one seemingly counterexample. Consider the distinction between sharp pain and dull pain. This is a distinction that seems only to be locatable along a phenomenal 'what it feels like' axis. Both are pains, both might dispose one to avoidance behavior, and yet there is an obvious phenomenal distinction. I'm unsure what a representationalist or analytic functionalist perspective could offer as a treatment of this. If anyone has any ideas, I'd love to hear them. Cheers to all.
"It makes sense to say that a written text on the neurophysiology of color perception and the phenomenal experience of seeing color are two channels of data aiming at, in some sense, the same information." Its not the same. Because written text is description of perception. If you describe someone, weight / height / heir color / eye color / skin color/ job / birth date and so one. That description will not be that person!!! If you take selfie and send it to different city, it is not situation where you travel to another city. This is absurd of materialism, when someone dont understand what is map of city vs real city. It looks like: reality -> percpetion vs reality -> perception -> text Assuming there is reality. We can put it in simpler form : perception vs description of perception ( aka usefull ficiton ) Description of reality is not reality. Science is description of behaviour of perception.
@@mehowop Hey man, thanks for the reply. Ultimately I think we agree. I don't think they are the same thing. That's ultimately what I meant by "access" mattering. The very thing that's supposed to matter when it comes to phenomenal experience just is the access, which has an experiential dimension that the theoretical dimension can't have. So all good on that. If I'm being particularly picky, I might argue with your analogy though. A theoretical explanation of perception isn't merely a description of it. This is where the "in some sense" in my comment comes in. A theory attempts to explain how and why that phenomenon occurs or can occur, it doesn't just provide a list of it's properties as a description might, and your example person's properties does. I see what you're driving at though, and am ultimately in agreement. Take care, fellow traveler.
Marys system of learning is totally shattered. She would run from that abusive black and white house. Her previous knowledge would be seen as an illusion. She would be insecure about all the knowledge she learned in that house. She may become so overwhelmed with the vastness of this new information that she eventually runs back to the house because at least she can predict the information from the house.
I think that to talk of phenomenal experience as a representation of our physical senses' interactions with the physical world makes sense, but doesn't get to the heart of the mind-body problem. The real nuts and bolts of what that characterisation describes is say photons hitting eyeballs causing a reaction in the brain's visual system, neurons interacting and then 'somehow' the experience of seeing a tree manifesting. The 'somehow' remains unaddressed. For Mary her neural interactions have patterns caused by photons of limited wave-lengths hitting her eyeballs within the black and white room. Then new neural patterns are formed when she's exposed to the world of other colours, manifesting different 'what it is like colour experiences. This is new knowledge manifested by new interactions with the world creating new neural patterns. Which is how the what it is likeness of all new sensory knowledge is learned. She's learned what it is like for her senses to interact with a yellow lemon, rather than what it like to interact with black words on white pages describing the neurophysics she now directly experiences. You can't get away from the experiential what it is likeness of knowledge, because knowing involves experiencing. And I think Philip is right that the what it is likeness of pain is what pain is. Other species and human ancestors who didn't have language feel it, babies feel distress. The behaviour associated with feeling pain arises from the experience of feeling pain and the desire to make it stop because it feels horrible. Otherwise when I take a headache pill for a headache, the behaviour of the taking of the headache pill is identical to the pain of the headache.
Maybe there is no 'somehow' that remains unaddressed. A complex neuron interaction IS the experience of seeing a tree. I know that this "feels" inadequate, because we intuitively imagine something like a "lifeless" machine when exploring the concept of a complex interaction of huge number of small parts, which cannot "feel" or "think", but this is just it - an intuition. We do not have an experience of all the possibilities that a "complex interaction" can produce. And if you accept this simple principle than there is really no "hard problem of consciousness".
@@ivanvnucko3056 Yes that's a legitimate What If... explanation which could turn out to be right. Novel physical properties do emerge from complex physical systems, and this could be the case for conscious experience too. What gives me pause is that physicalism has this explanatory standard model of particles and forces which theoretically reduces everything there is (and in principle can predict everything that could be) which doesn't mention or predict consciousness. And while it can theoretically fully account for everything that physical brains do, there's this explanatory gap as to how consciousness also somehow emerges. A gap which physicalist science doesn't know how to address except by noting neural correlation in ever finer detail, because physicalist science deals with what can be observed and measured. It might be we do find a way around that and the hard problem dissolves, or it might be our current notion of what is fundamental is incomplete as Philip would argue. Or if we take the representational nature of conscious experience seriously, our mental models of the real world we see as trees, and brains and understand as particles and forces, while functionally useful, might be a million miles from the ontological reality they represent to us. For now there's no way to falsify all these types of What Ifs... Physicalism I think has the advantage of parsimony and this amazing explanatory standard model on its side, but on the other side maybe its apparent inability in principle to integrate consciousness signals the need for an expanded or different model. Who knows :)
@@gert8439 nicely put, just a small "but": standard model doesn't "predict" any of the complex macro scale objects and events for us. Even given our "everyday" physics is fairly complete, no human can predict the existence of a chair or a car from it. I think it is not a reasonable demand to have.
@@ivanvnucko3056 Fair point, I was thinking in terms of Laplace's Demon, or a thought experiment Mary who had complete physical knowledge of everything in the universe at a specific moment (including quantum probability). Even manmade objects like my chair causally rooted in the actions of physical neurons ought to in principle be reducible and predictable. Or is my dodgy understanding of the science behind physicalism leading me astray?
@@gert8439 In principle they "ought to be reducible/predictable" but given 'Laplace's Demon' and 'Laplace's Mary' are theoretical thought experiment constructs (probably not possible in reality), I have no confidence we know what they would feel or predict. The huge gap between "demon like strict determinism" of our though experiments based on the most advanced physical theories and our tiny real world abilities (we spend billions of dollars and months of supercomputer time to calculate some couple of particles on nanosecond time scales) comfortably accommodates all the fuzzy human concepts of "free will" or "consciousness".
This could be subtitled "Why Frank became Frankish"! I'm very interested in why Frank Jackson changed his mind. I hope to add a more serious comment later. Peter
Thank you for considering my question ("There are parallels to Mary's story, e.g. one substituting completed panpsychism for physics. Is it reasonable to see the original as significant without giving the others similar weight?") Keith wondered if one might take this approach to show that only some sort of representationalism can meet the knowledge argument's challenge. For what it is worth, my feeling is that it leads to a resolution of that challenge in a way that is consistent with a very broad range of views (so you might even say that the challenge is, in a sense, illusory!) Once we generalize what Mary learns beforehand to subsets of all possible verbally-expressible knowledge, we have a trilemma: A) Mary is always surprised; B) Mary is never surprised; C) whether or not she is surprised depends on which subset of written knowledge she has mastered. If you go with B, I don't see any other way of putting it but that you are saying that there is nothing to see here: the intuition around which the argument is built is wrong. (I'm not saying it is mistaken to adopt this position, but it is not mine.) If you go with C - for example, insisting that Neuroscientist Mary will be surprised while Panpsychist Mary will not be - then you cannot use the argument against (in this case) physicalism without begging the question, unless you have an independent reason for determining who is surprised and who is not - but having that would render the knowledge argument itself moot. If you go with A (and, as it happens, I do), then we should not conclude anything more than that qualia are not communicable in words; as Russell almost said, "It is obvious that a man who can see knows things which a blind man cannot know; but a blind man can know the whole of written knowledge. Thus the knowledge which other men have and he has not is not to be found in any library." It is not hard to see why this would be so: we don't have complete conscious control of our minds (e.g. we cannot either produce or prevent the effects of heroin purely through willpower), and even if we did, we certainly do not have full access to other people's minds. As a consequence of the latter, we cannot develop a vocabulary for communicating qualia [1] (which is not the same as having a vocabulary for talking about qualia - a vocabulary with which Mary is already familiar.) Somewhat ironically, the very feature of qualia which is supposed to make them a problem for physicalism is not only reconcilable with it, but also allows physicalism to defuse the supposed threat raised by the knowledge argument. As for the question of what information Mary gains, I think it would be memories of what colors are like. Personally, I don't hold that she gains abilities so much as she gains the information to use abilities that she had from birth, learned over the hundreds of millions of years of evolution beginning with the first sense organs. So, is this dualism, after all? Well, it divides the world into things we can communicate verbally and things we cannot, and if that is a dualism, so be it. --- [1] In "What RoboMary Knows", Dennett posits his robots all have the same design, and know it, and also have full control over their own detailed physical state. We can suppose that such beings could develop a language for communicating qualia.
Contrary Mary is an expert at experiencing her visual senses. She has perfect eyesight and acuity and ability to differentiate on sight all shades of colour. Her phenomenological experiences of colour are perfectly clear and accurate and complete. She then goes into a room and studies the neuroscience of vision for the very first time. Does she learn anything new? Has she gained new knowledge?
The story of Mary shows one thing: that there is a plot and that there is a play. There is the Mathematical description and there is the reality. The mathematical description of a ball thrown in the air is not the same as the ball thrown in the air. In the same way, electro-chemical transmitters and the neurons and ganglia are not the experiences that the thrower of the ball encounters. All evidence is that something has these experiences. What, exactly, do the physicalists say that something is? Well, it’s a configuration of the brain, viz., brain states. Is it true that there are separate brain states for each experience of a human being? Is it true that for each emotional state we can point to specific brain and hormonal states that are reproducible? If all this were really true, the physicalists should be able to read minds after their program is realized. In that case, the physicalist could list the brain states and the thoughts and emotions implied by each brain state. I'm assuming a certain threshold that forces discretization of the brain states. But the fact is, there are only finitely many possible brain states in our universe and, therefore, only finitely many possible humanly recognizable thoughts at a given segment of time. So, what happens when we develop AGI, for instance, and the computer comes up with an idea outside of the humanly possible number of thoughts? Will we be able to understand the new thought at least as a thought? If you don't like that idea, what about if we list all the possible human thoughts and use some mechanical means of forming a new thought that is different from all the thoughts on our list? Will we not be able to understand that thought as a thought? For to understand the basic import of the thought the statement of the thought would have to be5among the finite number of thoughts possible to humans. Would the thought when read appear to us as a random or nonsense sentence? That is, would we not be able to detect that intelligence could have created the thought thus stated? To be a little more specific, suppose that we generated elementary thoughts in increasing order about integer numbers for instance. Let’s say we look at numbers like 1.000000000000000 x 10^15, 1.000000000000001 x 10^15, …, 9.999999999999999 x 10^15. If each distinct number has its own configuration of neural pattern corresponding to the thought, there will have to be a rational number that we could not recognize as a rational number because all the possible neural patterns were used up. However, if it is not true that a mind presented with the numbers as given eventually reaches an impasse, then every mental phenomenon would not have its own unique brain state, and there would have to be something in addition to the finite brain involved in the phenomenon of thinking about the numbers. An intuition is that the mind would not reach such an impasse, and therefore, there must be something in addition. If so, what seems the most likely at the present time? A secondary notion is that there are possible humanly thinkable thoughts that could never be thought "in reality" within our universe that in a sense reside within our universe. One might ask, when the universe is summed up, are these thoughts represented? Or is there a realm in which they exist? I find problems with the notion of Platonic Forms, but is there any other known way to account for these unrealized ideas?
What if Mary’s knowledge of _all_ of the physical facts about brains and light and so forth leads her away from Physicalism before her experience of seeing red?
The more Mary learns about colours, the more she realises it’s knowledge she does not have first person access to and therefore she is presumably aware she is lacking in first person knowledge about phenomenal colours…and if she doesn’t know that, she definitely ain’t no expert on colours.
I wonder if this is a recapitulation of the Measurement Problem in physics. Or actuality maybe it’s the Uncertainty Principle. Mary can only know so much about the world that does not contain her. Once she is in the world it’s a fundamentally different world. The mistake is to expect she could possibly know beforehand. Maybe it’s only just classic probability theory. She knows statistics but she won’t and can’t and shouldn’t know until she looks at the coin if it landed heads or tails.
I suspect that Goff was being overly accommodating to the others and their materialist position. We know that someone seeing a colour for the first time will find out what it is like. How could that not be true? Now it seems unfashionable to admit it. The physicalists' dismissal of intrinsic experiential qualities needs to be defended more explicitly and more rigorously. The visual illusions they fall back on show that we can misjudge what we are looking at, but not that there is no conscious experience of vision. That would take a very strong demonstration, and one that I cannot even imagine at this stage. Pain is an illusion? Go on ...?
Is Frank Jackson's account of phenomenology sympathetic to Gibson's theory of direct perception? There is something in his explanation that rings similar.
At ~30:50, Jackson says, "Something everyone knows is this: people who have perceptual experiences navigate the world much more successfully than people who don't [have perceptual experiences]." What? No, that's not true. At least, it is unjustified. Why would I believe that? Zombies navigate the world just as successfully as real people!
Excellent. 👍 The knowledge argument seems to rely on a confusion between Type (the general concept of 'red') and Token (a particular instance of 'red'.) Only tokens carry information. Mary doesn't have the latter wrt colour experiences until she leaves the B&W room.
mind: Although the meaning of “mind” has already been provided in Chapter 05 of “A Final Instruction Sheet for Humanity”, it shall prove beneficial to further clarify that definition here in the Glossary. It is NOT implied that mind is the sum of the actual thoughts, the sensations, the memories, and the abstract images that inhabit the mental element (or the “space”) that those phenomena occupy, but the faculty itself. This mental space has two phases: the potential state (traditionally referred to as the “unconscious mind”), where there are no mental objects present (such as in deep sleep or during profound meditation), and the actualized state (usually referred to as the “conscious mind”), where the aforementioned abstract objects occupy one’s cognition (such as feelings of pain). Likewise, the intellect and the pseudo-ego are the containers (or the “receptacles”) that hold conceptual thoughts and the sense of self, respectively. It is important to understand that the aforementioned three subsets of consciousness (mind, intellect, and false-ego) are not gross, tangible objects. Rather, they are subtle, intangible objects, that is, objects that can be perceived solely by an observant subject. The three subsets of consciousness transpire from certain areas of the brain (a phenomenon known as “strong emergence”), yet, as stated above, are not themselves composed of gross matter. Only a handful of mammal species possess intelligence (that is, abstract, conceptual thought processes), whilst human beings alone have acquired the pseudo-ego (the I-thought, which develops in infancy, following the id stage). Cf. “matter, gross”, “matter, subtle”, “subject”, and “object”. In the ancient Indian systems of metaphysics known as “Vedānta” and “Sāṃkhya”, mind is considered the sixth sense, although the five so-called “external” senses are, nonetheless, nominally distinguished from the mind, which is called an “internal” sense. This seems to be quite logical, because, just as the five “outer” senses involve a triad of experience (the perceived, the perception, and the perceiver), so too does the mind comprise a triad of cognition (the known, the knowing, and the knower). See also Chapter 06 of "F.I.S.H". P.S. There is much confusion (to put it EXTREMELY mildly) in both Western philosophy and in the so-called “Eastern” philosophical traditions, between the faculty of mind (“manaḥ”, in Sanskrit) and the intellect (“buddhiḥ”, in Sanskrit). Therefore, the following example of the distinction ought to help one to understand the difference between the two subtle material elements: When one observes a movie or television show on the screen of an electronic device that one is holding in one’s hands, one is cognizing auditory, textural, and visual percepts, originating from external objects, which “penetrate” the senses of the body, just as is the case with any other mammal. This is the component of consciousness known as “mind” (at least according to the philosophical terminology of this treatise, which is founded on Vedānta, according to widely-accepted English translations of the Sanskrit terms). However, due to our intelligence, it is possible for we humans (and possibly a couple of other species of mammals, although to a far less-sophisticated degree) to construct conceptual thoughts on top of the purely sensory percepts. E.g. “Hey - look at that silly guy playing in the swimming pool!”, “I wonder what will happen next?”, or “I hate that the murderer has escaped from his prison cell!”. To provide an even more organic illustration of how the faculty of mind “blends” into the faculty of the intellect, consider the following example: When the feeling of hunger (or to be more precise, appetite) appears in one’s consciousness, that feeling is in the mind. When we have the thought, “I’m hungry”, that is a conceptual idea that is a manifestation of the intellect. So, as a general rule, as animals evolve, they develop an intellectual faculty, in which there is an increasingly greater perception of, or KNOWLEDGE of, the external world (and in the case of at least one species, knowledge of the inner world). In addition to these two faculties of mind and intellect, we humans possess the false-ego (“ahaṃkāraḥ”, in Sanskrit). See Chapter 10 of "F.I.S.H" regarding egoity.
Type/Token is an interesting approach Richard. I’ve often wondered and argued that the first person quality of consciousness is actually mundane, and actually about geometry. It is fundamental, but not in the way that panpsychics mean. It is more simple and spatial. A hole in a fence has a point of view. It’s this hole not that hole. Consciousness has nothing to do with that bit. So yeah, like you say, type is not token.
She has neither and then she gains both no? One through direct experience and one through extrapolation of that experience. So it changes nothing about the knowledge argument.
This has to be a troll. Although Goff is a genius, and I'm a fan, but his rebuttals to physicalism, here, were terrible. "What about Mary's CURIOSITY!??" That's IT, Goff? WTF!!! Oh, and the big finale from Goff, "It just SEEMS like there's something MORE to FEELING pain... you know what I mean?" Wow! Again, Goff you're a genius, but you really let those of us down who wanted to see a good take down of materialism. almost this ENTIRE VIDEO was excellent arguments FOR MATERIALISM, of which our only hope, Goff, did NOTHING ABOUT! Sorry, for my Internet rage, Goff. Big fan, I promise.
I’m on the side of the physicalists, but in Goff’s defense I’ll say that the Knowledge Argument is THE argument. It’s asking a lot that he come up with something better than that.
I mean, the knowledge argument is good, but physicalism dies because it is internally inconsistent. It's over for physicalism. You cannot start with consciousness, then define something else as non-conscious, then say that your consciousness IS that non-conscious stuff. Inconsistent. What's more, it fails on modal grounds. You can know all physical facts yet from them not get a single quality of the world or a single experience. Not only is that the case in our world but in possible worlds, as well. If physicalism were true then it would be necessary and because it's the opposite of that, it fails. This is just the start, as there are other facts that show physicalism as untenable. Materialism, though related, is also doomed. Goff is a complete disappointment, at least in this vid. @@aaronshure3723
@@robertblankenship5000 I'm not convinced by any demand for consistency or total knowledge. I start from the assumption that I am not the world but a part of it. And I assume that it is unlikely that the world will satisfy my personal demands for satisfaction. This is supported by how stubbornly not only the world, but my mind seems to resist my attempts to control them. I also find a priori arguments modal or otherwise to be a poor starting place. I mean if a self contradiction implies infinite conclusions, I think you should take all logical arguments with a grain of salt. Goff has gotten me to question whether or not my attitude leads necessarily to physicalism. Kastrup too. They both want the world described by science to be taken as a version of real. So that's the part I'm working on.
@@aaronshure3723The knowledge argument suffers from an assumption: that knowledge (in physicalism) can only come from communication with the world outside the brain. But this isn’t true. There possibly is also communication from within the brain.
@@falklumointeresting. In a similar vein, Frankish has noted that there is also “internal behavior”. I came up with the term “skull dichotomy” fallacy. It’s the assumption that the skull is some sort of boundary between “inner” and “outer”world. The skull dichotomy is at its strongest when talking about fMRI studies. You need to ask yourself does this fMRI show me something different than what I learn from talking to the subject? Does this fMRI show me something more than I can learn from the rest of the body? So it could be that Mary’s neocortex knows something that her eyes have yet to experience. Or her blabbity blab part of her brain has knowledge that the thingymabob part of her brain doesn’t yet know.
how would to seem to Mary if the visual input suddenly included pain or sound? if a ball suddenly turned from red to gratitude? why doesnt that happen? we have no clue at all other than "but thats wrong, that couldnt happen". and yet syntetesia happen. people hear colors. seeing red or blue is not about knowledge. its not a new fact. its just one of many inner experiences we can have. those that have done drugs probably have seen colors we never normally experiences. its not nee knowledge in itself.
That is a good question. Would she actually experience red or would she experience a shade of grey? Would the rods ever have been activated in her dreams? If you stare at green long enough then look away, you will see yellow (or something like that). How would that work with Mary if you just allowed her to see green and no other color?
Not taking sides between physicalism and panpsychism, but the knowledge argument is severely flawed. Here is why: Imagine a human reading and learning everything possible about four-dimensional space. Then we know we would still be missing something important: the ability to “feel” 4D. We humans can only experience 3D. But the difference would be material: biological brain structures to allow the brain to capture 4D. (In AI research, we would say to build a latent representation of 4D data.) So, what is missing is latent space information. A brain function. This does NOT allow to deduce that consciousness must have a non physical side. Only to deduce that consciousness needs internal state which may well be physical.
Mary (before) knows everything worth saying about red. She knows how to talk to you about it, she knows how to write poetry about it, she knows why you can’t see it if you’re color blind, and she knows how to cure your color blindness if it can be cured. The only thing she doesn’t know is a fact, not about the world, but about herself. It’s not actually that much or that important. If you want to build your metaphysical castle on that one grain of sand, have at it, but it seems misguided to me.
I don’t know if you’re seeing the point of the knowledge argument. If all the pertinent physical facts associated with seeing red are insufficient at getting at what it’s like to see red, then there’s a fundamental epistemic gap between physical stuff and experiences. This is a problem assuming you don’t want to grant consciousness this special status of being this one irreducibly emergent thing in the world.
@@Sam-hh3ryIs it some spooky emergent thing to say a hole in a fence has a different point of view from a different hole in a fence? They don’t see but they have points of view. It’s so trivial that it seems special. It’s just particular v universal. Not even subjective v universal. Just the inevitable result of picking one point out. That’s what Mary lacks.
@@aaronshure3723 your analogy doesn't work and misses the point. there's nothing about looking at something from a particular angle that can't be empirically predicted and verified (based on, for example, the size of the hole, your distance and angle, the structure of your eye, etc.). on the other hand, there is absolutely nothing about the phenomenal character of ANY given experience that can be empirically predicted or verified.
@@Sam-hh3ry anesthesiologists and drug dealers beg to differ about the predictability of subjective states, but keep enjoying your ineffable mystery mate
@@aaronshure3723 loool you have absolutely no idea what you're talking about. I guess you should start by learning the meaning of the word 'phenomenal' and familiarize yourself with concepts like 'hard problem' and 'explanatory/epistemic gap'. I assure that drugs and anesthetics have literally nothing to do with what I'm talking about.
People sometimes bring up Frank Jackson's conversion to physicalism as a point to weaken the knowledge argument, as if the argument was so bad the creator had to abandon it. So, it was interesting hearing that he switched to illusionism, and he still thinks the knowledge argument needs to be taken seriously by physicalists. I wonder if the deeper disagreement isn't between physicalists and anti-physicalists, but between those who take the epistemic gap presented by qualia seriously, and those who don't.
I remember studying this in school and my thoughts on it don't seem to have changed. Experience and knowledge are different things. There is a physiological difference between what happens when we learn about a particular thing versus when we experience that particular thing. The idea that because this difference exists then physicalism is false is just bizarre to me. It would be like reading a review of a novel that describes every thing that happens, nothing being left out but also nothing being quoted, and then saying you've read the novel. Well, you haven't, and you certainly wouldn't think so after reading it. You wouldn't have experienced the author's voice, the rhythm and flow of their writing, or the imagery and dialogue in the way that only your mind could imagine from reading the actual text. Honestly, I'm amazed that anyone was ever swayed by Professor Jackson's argument. It is inventive, but it isn't persuasive.
You’ve completely missed the point of the knowledge argument. If knowledge of a what it’s like to have a particular experience requires actually being in the associated physiological state, then there is an epistemic gap between knowing the physical facts about that physiological state and actually being in that state. All we’ve done is reframe things from talking about what it’s like to see red to what it’s like to be in the physiological state associated with seeing red. It’s the fact that there’s anything it’s like at all to be in a particular state that is the issue.
@@micdavey It seems to falsify physicalism if you consider physicalism the view that everything should be reducible and describable in terms of physical concepts. If we can't know anything about consciousness working purely from physical states, then I don't know what it would mean to call it physical.
@@Sam-hh3ry Well, for one we might have to expand our understanding of physical concepts. I mean, we don't fully understand how consciousness works. It can be a physical phenomena but not currently explanatory. And because we can't explain something today, that doesn't mean it's unexplainable. I'm also not trying to say that physicalism can't be false, just that the knowledge argument doesn't falsify it. And I guess just to be clear, we do know a lot about what consciousness is like working from physical states, we just can't know exactly what it's like based our current understanding of the physical states that determine states of subjective experience, and we might never know. The tl;dr here is that it's too early to claim what Jackson claims --that's all.
@@micdavey We know nothing about consciousness working purely from physical states. Literally everything we know about consciousness requires comparing our personal experiences against correlated physical states. If the premise of the knowledge argument is correct, then this gap can't be bridged even in principle. I don't know what kind of scientific theory we can expect to have when we can't measure, quantify, or model the phenomenon in question. And if we can't do those things, I don't know what it means to nonetheless call it physical.
If Mary has the neural wiring capable of distinguishing/perceiving colour before exiting the b&w room, once she observes the novel sense data (i.e. that which is interpreted/perceived as polychromatic; e.g. particular photon distributions or artificial stimulation of the brain), she will become aware of this novel experience, and the encoding/memory of this awareness in the neural circuitry is itself new knowledge. From a physicalist perceptive, the phenomenological experience is irrelevant; only the agent's model/belief in the phenomenological experience.
1:49:00 - the knowledge argument has no implications for the philosophy of mental properties; there is no new knowledge endowed by mental properties (phenomenological consciousness). The "knowledge" i.e. belief/model of mental properties is encoded in the brain, and their existence is functionally irrelevant (empirically unobservable).
Phenomenal consciousness is empirically unobservable, yes. That is the whole point and the reason that there is an epistemic gap in the first place. If the move is to deny the existence of phenomenal consciousness altogether, that’s a valid direction, but you’ve done nothing to justify it.
The phenomenological experience itself is not new knowledge, at least not from an empirical point of view with respect to information (or without additional assumptions such as substance dualism or epiphenomenalism). I therefore think the concept of an epistemic gap here is invalid (irrelevant/misleading), philosophy of mind is not primarily concerned with the experience of qualia A (monochromatic) vs qualia B (polychromatic), it is concerned with experience vs non-experience. Under physicalism broadly construed (e.g. Chalmers) the phenomenological experience has associated (mapped) knowledge encoded in the physical system.
Why does South Korea produce the greatest Women Golfers? That's not a small group like the Swedish Tennis Players and the Australian Philosophers. Each year, more and more great Women Golfers from South Korea.
I am an expert on consciousness. I don’t know what it’s like to see red, have pain, fall in love, play football, enjoy Pink Floyd, taste BBQ ribs or read French poetry. My physicalist philosopher supervisor says my expertise on consciousness will be complete just as soon I acknowledge I don’t exist…or at least come up with a logical proof of my own non-existence. At which point my PhD will instead be awarded to a “ghost in the machine,” in other words, one of those rare students who believes in dualism or some kind of Kantian idealism.
Orange isn’t a real thing. It becomes real when the combination of light frequencies hit your eye’s cones fire in a peculiar way, the neural processing behind the retina to the visual cortex. In short, orange is a lie. It doesn’t exist outside your brain. You’re Leonardo Da Vinci. You know colour better than most. You see colour but until recently, all the maps you’ve read were all printed in black and white. Today you visited the library of the Sforzas for the first time and saw a glorious map of Milan in black and other colours. All you got War was an easier way to represent the same information. The map maker is leveraging the peculiarities of the human brain to pack information.
when you aren't seeing red or imagining red, do you have knowledge of what red is like? It isn't anywhere in your current subjective experience, it is only stored in the memory ready to be accessed. So for you to know what red it like, wouldn't it need to be stored in the same format it is experienced it, e.g. wouldn't there need to be some non-physical qualia storage that your consciousness can access? That is qualia that isn't being experienced, which immediately begs the question of why ordinary qualia is experienced. If knowledge of what red is like is stored in a purely physical way in the brain when you aren't seeing or imagining red, what would prevent Mary from getting that knowledge by studying a brain of someone who isn't actively experiencing red but has that knowledge? Part of that stored knowledge must be non-physical if Mary can't learn it from all physical facts, which leads to my first paragraph.
Memory is not like a computer disk. There is no "format" and storing. The process of remembering is a reconstruction of a particular state the brain was already before in. When you do not see the color red, you can actively reconstruct a simulation of that experience - if your brain was once in that state we call "seeing red", there are possibly still the neural connections needed to actualize similar state (not exactly the same, and due to pruning, maybe some of these connections are already gone) without actual photons of particular wavelength hitting your retina. The problem with this kinds of arguments is - they are mostly contingent on folk intuition. I would even go further and point out the "begging the question" in it, pointing to phrases such as "is stored in a purely physical way", which presupposes some kind of physical/non-physical duality of "things". It is maybe intuitive, but what would it even mean for something to exist (or be stored) in "non-physical" way?
@@ivanvnucko3056 it presupposes dualism because I am pointing out that there is an inconsistency with it. If it is possible to reconstruct the experience of red after seeing it, why wouldn't Mary be able to do that? Why wouldn't Mary be able to bring about the same neutral connections if she has perfect knowledge of what those connections are? Those connections are not experience - yet Mary is still unable to have the knowledge corresponding to having those connections. That tells me that there is something wrong with the thought experiment - there is a physical kind of knowledge that Mary can't access. The only way to avoid that is to assume knowledge of what it is like to see red is also stored in some non physical storage. I do think there is something wrong with the thought experiment - all Mary knows concerning subjective experience is indexical, she by definition can only have her own experience. If she had someone else's experience, she would be someone else. You can't know a subjective experience, you can only have it.
@@carnap355 "If it is possible to reconstruct the experience of red after seeing it, why wouldn't Mary be able to do that?" - because she never experienced it before, that was the whole point of my comment. The experience itself is just a state (in time of course, not static) of the brain, caused partly by light of particular wavelength interval hitting her retina, which creates some new connections (the experience itself and also the reference to language construct - the word red, if someone tells Mary "see, this thing is red"), which then can be used to try to reconstruct it "from memory". But before that there aren't those connections in Mary's brain. The fact that she "knows" a lot of shit about red is another potentially huge set of connections, but the experience itself cannot be among them. With our state of technology we are not able to "create connections" on demand, not with that level of precision - I think some manipulations are already possible. Nonetheless experience is a different set of connections to knowledge. I honestly do not understand why this is so baffling to so many people. It is perfectly intuitive and understandable to most people that one can perfectly know music theory, and guitar construction theory and all about playing styles and finger movements and so on, but it is a completely different skill to practically take a guitar and fluently play with own fingers. Experience of red is exactly the same thing. Both types of connections (knowledge and motor) are physical, just a different set.
@@ivanvnucko3056 and she can't enter that state without actually seeing red, not because she is limited by being physical but because that's how knowledge of perceptual experiences is formed in a human being. I agree with you there. But a dualist would use one of few arguments such as why is she surprised, what magically prevents the appropriate connections to form when Mary has perfect understanding of someone else's brain experiencing red, and so on. I was trying to say those fail because you can formulate those about knowledge that must be physical even under dualism, since otherwise there would be qualia that isn't being experienced
Reflectory theory of human behavior. Nothing special. Just standard analysis, an attempt to break the complicated phenomenon down to elementary functions and events. Human mind is composed of various images and associations. I consider it completed now. We can proceed to applications such as fixing widespread mental disorders or creating conscious machines.
I agree with David Chalmers’ basic argument: “Physical descriptions of the world characterize the world in terms of structure and dynamics. From truths about structure and dynamics, one can deduce only further truths about structure and dynamics”.
Mary’s knowledge argument is a little weak because it carries a lot of implied assumptions: Mary is a human (which implies consciousness), she already can see a scale of grays, etc. I prefer the example of an alien creature from another dimension who is given all the physical laws governing our universe, so that it can simulate life on Earth. At most the alien would be able to describe what Chalmers calls “philosophical zombies”. There is no way for that alien to figure out that those earthly creatures have conscious feelings (other than by their vocal claims that they do, which the alien would not be able to understand anyway). So the complete knowledge of physical laws would still not be sufficient for the alien to have a complete understanding of what is going on, and new laws would need to be added to complete the description
if you're a physicalist and adhere to the church of magical mass-energy which can arise as colors sounds etc then yes the physical alien can eventually manipulate the physical to experience all those things directly
You make an assumption yourself: what the outcome of said alien simulation would yield. You can’t. As otherwise, there would be no point in running the simulation in the first place. E.g., such aliens might be able to replay parts of such a simulation in their heads and experience human consciousness themselves.
"complete understanding" what you want ? You want scientist to show you qualia by opening up brain ? It's so obvious that qualia which is building block of consciousness is first person experience. But it doesn't imply that it is correct and fundamental reality , there are lot of experiments that shows that qualia can be misleading and they are just modifications of actual objective reality
The Mary argument may be disadvantaged as a potential defeater of physicalism by virtue of the fact that Mary is already a conscious being, ie Mary already knows what is it like to have experiential or phenomenological consciousness. A knowledge argument which might side-stop this could be to ask: what if you feed an artificially intelligent entity all the conceivable data of a purely physical nature about a human being, and then asked it to produce the most complete description of the human being, would the AI know that the human being was conscious (or experienced itself as being conscious) at all? If yes, then how precisely does the AI extrapolate consciousness from the purely physical data (without having any prior/pre-existing direct experience of consciousness itself)? If no, then isn't the purely physicalist description of the human being necessarily incomplete?
Professor Jackson says that there is a new way to experience the information, but the argument is not that there isn't a new way, it is that the experience does not come through any way that a physicalist can come up with.
One major problem for panpsychism is that once you ascribe conscious properties to all matter you lose the capacity to define boundaries and such behavioural differences from non-organic matter to organic beings. Further you have to then attribute all cognitive capacities to inanimate objects from memory to disorders of consciousness which brings you to ridiculous conclusions such as rocks experience states like psychosis or deficits in theory of mind which are by definition observable behavioural or at least expressive definitions of intentional active beings
Why? We attribute gravitational properties to all matter, yet no one explicitly expects to observe pebbles orbiting mountains.
Why wouldn't it be possible for all matter to have a "consciousness property", yet observable consciousness as we understand it only emerges when a concentration gradient is high enough relative to some other not yet understood factor.
@@christophercavill2664 You do in space - you expect pebbles to orbit asteroids. On Earth you expect pebbles to be influenced by all the matter nearby, ie the Earth as well as the mountain.
My internal examiner for my PhD in 1996 was Howard Robinson, and the external was Prof. E.J.Lowe. They both seemed quite happy with my insistence that qualia really exist and need an adequate account. What we are witnessing here is a throw-back to JJC Smart being able to distinguish between lettuce and cabbage leaves, but unable to explain in what respect. If conscious experience is just an illusion, we need an explanation of how that illusion works.
A rebuttal of the knowledge argument against physicalism was made nearly 2 decades ago. Firstly, there is a false premise that Mary will suddenly acquire colour knowledge once released as evidence would support the alternative scenario that she would continue to see the world monochromatically because acquisition of perceptual knowledge outside previous possible experience requires direct interaction with coloured objects. Crucially and this is the linchpin for physicalism her ability to acquire this new colour experience will depend on adult neural plasticity which will permit encoding of new action perception algorithms.
I find it extraordinary that this obstacle against the knowledge argument has been marginalised
Great conversation as always.
Jackson's reformulation of what Mary gains when leaving her colorless room, in terms of either immediate or derived information - in order to say the information is in some sense the same, with only the access to it changing - had me paused for a moment. But after the pause it occurred to me that the access is precisely what we're worried about. Which is to say the knowledge argument regards knowledge of data, rather than the broader category of information. It makes sense to say that a written text on the neurophysiology of color perception and the phenomenal experience of seeing color are two channels of data aiming at, in some sense, the same information. But what we were supposed to care about was whether her gaining access to the latter channel of data gave her knew knowledge. And framed in terms of 'knowledge of data', its clear she does.
Regarding Jackson's newfound representationalism and analytic functionalism. One can get very far with a representationalist account of exteroceptive perception, as the objects perceived are in some sense almost always representations. With some elegant intellectual contortions one could argue that color is primarily just another dimension of distinguishing those representations. Interoceptive perception/sensation is more difficult to give a representationalist treatment, since the objects of those states, sensations, lack the obvious boundaries that exteroceptive perceptions have. Jackson and Frankish highlight their difficulty in being able to characterize those states, of pain for example, and that they are nearly always associated with some sort of behavioral inclination, as being fine evidence for a broadly functionalist treatment. Sensations are nothing more than functional inclinations toward particular behavior. There is much to say regarding this sort of perspective, mostly negative in my opinion, but in the interest of brevity I'll offer just one seemingly counterexample. Consider the distinction between sharp pain and dull pain. This is a distinction that seems only to be locatable along a phenomenal 'what it feels like' axis. Both are pains, both might dispose one to avoidance behavior, and yet there is an obvious phenomenal distinction. I'm unsure what a representationalist or analytic functionalist perspective could offer as a treatment of this. If anyone has any ideas, I'd love to hear them.
Cheers to all.
"It makes sense to say that a written text on the neurophysiology of color perception and the phenomenal experience of seeing color are two channels of data aiming at, in some sense, the same information."
Its not the same. Because written text is description of perception.
If you describe someone, weight / height / heir color / eye color / skin color/ job / birth date and so one. That description will not be that person!!!
If you take selfie and send it to different city, it is not situation where you travel to another city. This is absurd of materialism, when someone dont understand what is map of city vs real city.
It looks like:
reality -> percpetion vs reality -> perception -> text Assuming there is reality.
We can put it in simpler form :
perception vs description of perception ( aka usefull ficiton )
Description of reality is not reality. Science is description of behaviour of perception.
@@mehowop Hey man, thanks for the reply. Ultimately I think we agree. I don't think they are the same thing. That's ultimately what I meant by "access" mattering. The very thing that's supposed to matter when it comes to phenomenal experience just is the access, which has an experiential dimension that the theoretical dimension can't have. So all good on that. If I'm being particularly picky, I might argue with your analogy though. A theoretical explanation of perception isn't merely a description of it. This is where the "in some sense" in my comment comes in. A theory attempts to explain how and why that phenomenon occurs or can occur, it doesn't just provide a list of it's properties as a description might, and your example person's properties does. I see what you're driving at though, and am ultimately in agreement.
Take care, fellow traveler.
Marys system of learning is totally shattered. She would run from that abusive black and white house. Her previous knowledge would be seen as an illusion. She would be insecure about all the knowledge she learned in that house. She may become so overwhelmed with the vastness of this new information that she eventually runs back to the house because at least she can predict the information from the house.
I think that to talk of phenomenal experience as a representation of our physical senses' interactions with the physical world makes sense, but doesn't get to the heart of the mind-body problem. The real nuts and bolts of what that characterisation describes is say photons hitting eyeballs causing a reaction in the brain's visual system, neurons interacting and then 'somehow' the experience of seeing a tree manifesting. The 'somehow' remains unaddressed.
For Mary her neural interactions have patterns caused by photons of limited wave-lengths hitting her eyeballs within the black and white room. Then new neural patterns are formed when she's exposed to the world of other colours, manifesting different 'what it is like colour experiences. This is new knowledge manifested by new interactions with the world creating new neural patterns. Which is how the what it is likeness of all new sensory knowledge is learned. She's learned what it is like for her senses to interact with a yellow lemon, rather than what it like to interact with black words on white pages describing the neurophysics she now directly experiences.
You can't get away from the experiential what it is likeness of knowledge, because knowing involves experiencing.
And I think Philip is right that the what it is likeness of pain is what pain is. Other species and human ancestors who didn't have language feel it, babies feel distress. The behaviour associated with feeling pain arises from the experience of feeling pain and the desire to make it stop because it feels horrible. Otherwise when I take a headache pill for a headache, the behaviour of the taking of the headache pill is identical to the pain of the headache.
Maybe there is no 'somehow' that remains unaddressed. A complex neuron interaction IS the experience of seeing a tree. I know that this "feels" inadequate, because we intuitively imagine something like a "lifeless" machine when exploring the concept of a complex interaction of huge number of small parts, which cannot "feel" or "think", but this is just it - an intuition. We do not have an experience of all the possibilities that a "complex interaction" can produce. And if you accept this simple principle than there is really no "hard problem of consciousness".
@@ivanvnucko3056 Yes that's a legitimate What If... explanation which could turn out to be right. Novel physical properties do emerge from complex physical systems, and this could be the case for conscious experience too.
What gives me pause is that physicalism has this explanatory standard model of particles and forces which theoretically reduces everything there is (and in principle can predict everything that could be) which doesn't mention or predict consciousness. And while it can theoretically fully account for everything that physical brains do, there's this explanatory gap as to how consciousness also somehow emerges. A gap which physicalist science doesn't know how to address except by noting neural correlation in ever finer detail, because physicalist science deals with what can be observed and measured.
It might be we do find a way around that and the hard problem dissolves, or it might be our current notion of what is fundamental is incomplete as Philip would argue. Or if we take the representational nature of conscious experience seriously, our mental models of the real world we see as trees, and brains and understand as particles and forces, while functionally useful, might be a million miles from the ontological reality they represent to us.
For now there's no way to falsify all these types of What Ifs... Physicalism I think has the advantage of parsimony and this amazing explanatory standard model on its side, but on the other side maybe its apparent inability in principle to integrate consciousness signals the need for an expanded or different model. Who knows :)
@@gert8439 nicely put, just a small "but": standard model doesn't "predict" any of the complex macro scale objects and events for us. Even given our "everyday" physics is fairly complete, no human can predict the existence of a chair or a car from it. I think it is not a reasonable demand to have.
@@ivanvnucko3056 Fair point, I was thinking in terms of Laplace's Demon, or a thought experiment Mary who had complete physical knowledge of everything in the universe at a specific moment (including quantum probability).
Even manmade objects like my chair causally rooted in the actions of physical neurons ought to in principle be reducible and predictable. Or is my dodgy understanding of the science behind physicalism leading me astray?
@@gert8439 In principle they "ought to be reducible/predictable" but given 'Laplace's Demon' and 'Laplace's Mary' are theoretical thought experiment constructs (probably not possible in reality), I have no confidence we know what they would feel or predict. The huge gap between "demon like strict determinism" of our though experiments based on the most advanced physical theories and our tiny real world abilities (we spend billions of dollars and months of supercomputer time to calculate some couple of particles on nanosecond time scales) comfortably accommodates all the fuzzy human concepts of "free will" or "consciousness".
The laugh at the end is priceless.
Love the pods, guys!
This could be subtitled "Why Frank became Frankish"!
I'm very interested in why Frank Jackson changed his mind. I hope to add a more serious comment later. Peter
Thank you for considering my question ("There are parallels to Mary's story, e.g. one substituting completed panpsychism for physics. Is it reasonable to see the original as significant without giving the others similar weight?") Keith wondered if one might take this approach to show that only some sort of representationalism can meet the knowledge argument's challenge. For what it is worth, my feeling is that it leads to a resolution of that challenge in a way that is consistent with a very broad range of views (so you might even say that the challenge is, in a sense, illusory!)
Once we generalize what Mary learns beforehand to subsets of all possible verbally-expressible knowledge, we have a trilemma: A) Mary is always surprised; B) Mary is never surprised; C) whether or not she is surprised depends on which subset of written knowledge she has mastered.
If you go with B, I don't see any other way of putting it but that you are saying that there is nothing to see here: the intuition around which the argument is built is wrong. (I'm not saying it is mistaken to adopt this position, but it is not mine.)
If you go with C - for example, insisting that Neuroscientist Mary will be surprised while Panpsychist Mary will not be - then you cannot use the argument against (in this case) physicalism without begging the question, unless you have an independent reason for determining who is surprised and who is not - but having that would render the knowledge argument itself moot.
If you go with A (and, as it happens, I do), then we should not conclude anything more than that qualia are not communicable in words; as Russell almost said, "It is obvious that a man who can see knows things which a blind man cannot know; but a blind man can know the whole of written knowledge. Thus the knowledge which other men have and he has not is not to be found in any library."
It is not hard to see why this would be so: we don't have complete conscious control of our minds (e.g. we cannot either produce or prevent the effects of heroin purely through willpower), and even if we did, we certainly do not have full access to other people's minds. As a consequence of the latter, we cannot develop a vocabulary for communicating qualia [1] (which is not the same as having a vocabulary for talking about qualia - a vocabulary with which Mary is already familiar.) Somewhat ironically, the very feature of qualia which is supposed to make them a problem for physicalism is not only reconcilable with it, but also allows physicalism to defuse the supposed threat raised by the knowledge argument.
As for the question of what information Mary gains, I think it would be memories of what colors are like. Personally, I don't hold that she gains abilities so much as she gains the information to use abilities that she had from birth, learned over the hundreds of millions of years of evolution beginning with the first sense organs.
So, is this dualism, after all? Well, it divides the world into things we can communicate verbally and things we cannot, and if that is a dualism, so be it.
---
[1] In "What RoboMary Knows", Dennett posits his robots all have the same design, and know it, and also have full control over their own detailed physical state. We can suppose that such beings could develop a language for communicating qualia.
Contrary Mary is an expert at experiencing her visual senses. She has perfect eyesight and acuity and ability to differentiate on sight all shades of colour. Her phenomenological experiences of colour are perfectly clear and accurate and complete.
She then goes into a room and studies the neuroscience of vision for the very first time. Does she learn anything new? Has she gained new knowledge?
I wish Frank's "fred's apples" thought experiment had taken off instead of "mary's red room", it does much better in avoiding question begging.
The story of Mary shows one thing: that there is a plot and that there is a play. There is the Mathematical description and there is the reality. The mathematical description of a ball thrown in the air is not the same as the ball thrown in the air. In the same way, electro-chemical transmitters and the neurons and ganglia are not the experiences that the thrower of the ball encounters.
All evidence is that something has these experiences. What, exactly, do the physicalists say that something is? Well, it’s a configuration of the brain, viz., brain states.
Is it true that there are separate brain states for each experience of a human being? Is it true that for each emotional state we can point to specific brain and hormonal states that are reproducible? If all this were really true, the physicalists should be able to read minds after their program is realized.
In that case, the physicalist could list the brain states and the thoughts and emotions implied by each brain state. I'm assuming a certain threshold that forces discretization of the brain states. But the fact is, there are only finitely many possible brain states in our universe and, therefore, only finitely many possible humanly recognizable thoughts at a given segment of time.
So, what happens when we develop AGI, for instance, and the computer comes up with an idea outside of the humanly possible number of thoughts? Will we be able to understand the new thought at least as a thought?
If you don't like that idea, what about if we list all the possible human thoughts and use some mechanical means of forming a new thought that is different from all the thoughts on our list? Will we not be able to understand that thought as a thought? For to understand the basic import of the thought the statement of the thought would have to be5among the finite number of thoughts possible to humans. Would the thought when read appear to us as a random or nonsense sentence? That is, would we not be able to detect that intelligence could have created the thought thus stated?
To be a little more specific, suppose that we generated elementary thoughts in increasing order about integer numbers for instance. Let’s say we look at numbers like 1.000000000000000 x 10^15, 1.000000000000001 x 10^15, …, 9.999999999999999 x 10^15. If each distinct number has its own configuration of neural pattern corresponding to the thought, there will have to be a rational number that we could not recognize as a rational number because all the possible neural patterns were used up.
However, if it is not true that a mind presented with the numbers as given eventually reaches an impasse, then every mental phenomenon would not have its own unique brain state, and there would have to be something in addition to the finite brain involved in the phenomenon of thinking about the numbers.
An intuition is that the mind would not reach such an impasse, and therefore, there must be something in addition. If so, what seems the most likely at the present time?
A secondary notion is that there are possible humanly thinkable thoughts that could never be thought "in reality" within our universe that in a sense reside within our universe. One might ask, when the universe is summed up, are these thoughts represented? Or is there a realm in which they exist?
I find problems with the notion of Platonic Forms, but is there any other known way to account for these unrealized ideas?
What if Mary’s knowledge of _all_ of the physical facts about brains and light and so forth leads her away from Physicalism before her experience of seeing red?
In other words, does the “complete knowledge” objection beg the question?
The more Mary learns about colours, the more she realises it’s knowledge she does not have first person access to and therefore she is presumably aware she is lacking in first person knowledge about phenomenal colours…and if she doesn’t know that, she definitely ain’t no expert on colours.
I wonder if this is a recapitulation of the Measurement Problem in physics. Or actuality maybe it’s the Uncertainty Principle. Mary can only know so much about the world that does not contain her. Once she is in the world it’s a fundamentally different world. The mistake is to expect she could possibly know beforehand. Maybe it’s only just classic probability theory. She knows statistics but she won’t and can’t and shouldn’t know until she looks at the coin if it landed heads or tails.
If the answer to Mary's room is pain is nothing more than its functions ...
I suspect that Goff was being overly accommodating to the others and their materialist position. We know that someone seeing a colour for the first time will find out what it is like. How could that not be true? Now it seems unfashionable to admit it. The physicalists' dismissal of intrinsic experiential qualities needs to be defended more explicitly and more rigorously. The visual illusions they fall back on show that we can misjudge what we are looking at, but not that there is no conscious experience of vision. That would take a very strong demonstration, and one that I cannot even imagine at this stage. Pain is an illusion? Go on ...?
Is Frank Jackson's account of phenomenology sympathetic to Gibson's theory of direct perception? There is something in his explanation that rings similar.
At ~30:50, Jackson says, "Something everyone knows is this: people who have perceptual experiences navigate the world much more successfully than people who don't [have perceptual experiences]." What? No, that's not true. At least, it is unjustified. Why would I believe that? Zombies navigate the world just as successfully as real people!
Would love a follow-up with Howard on the argument!
Excellent. 👍 The knowledge argument seems to rely on a confusion between Type (the general concept of 'red') and Token (a particular instance of 'red'.) Only tokens carry information. Mary doesn't have the latter wrt colour experiences until she leaves the B&W room.
mind:
Although the meaning of “mind” has already been provided in Chapter 05 of “A Final Instruction Sheet for Humanity”, it shall prove beneficial to further clarify that definition here in the Glossary. It is NOT implied that mind is the sum of the actual thoughts, the sensations, the memories, and the abstract images that inhabit the mental element (or the “space”) that those phenomena occupy, but the faculty itself. This mental space has two phases: the potential state (traditionally referred to as the “unconscious mind”), where there are no mental objects present (such as in deep sleep or during profound meditation), and the actualized state (usually referred to as the “conscious mind”), where the aforementioned abstract objects occupy one’s cognition (such as feelings of pain).
Likewise, the intellect and the pseudo-ego are the containers (or the “receptacles”) that hold conceptual thoughts and the sense of self, respectively. It is important to understand that the aforementioned three subsets of consciousness (mind, intellect, and false-ego) are not gross, tangible objects. Rather, they are subtle, intangible objects, that is, objects that can be perceived solely by an observant subject. The three subsets of consciousness transpire from certain areas of the brain (a phenomenon known as “strong emergence”), yet, as stated above, are not themselves composed of gross matter. Only a handful of mammal species possess intelligence (that is, abstract, conceptual thought processes), whilst human beings alone have acquired the pseudo-ego (the I-thought, which develops in infancy, following the id stage). Cf. “matter, gross”, “matter, subtle”, “subject”, and “object”.
In the ancient Indian systems of metaphysics known as “Vedānta” and “Sāṃkhya”, mind is considered the sixth sense, although the five so-called “external” senses are, nonetheless, nominally distinguished from the mind, which is called an “internal” sense. This seems to be quite logical, because, just as the five “outer” senses involve a triad of experience (the perceived, the perception, and the perceiver), so too does the mind comprise a triad of cognition (the known, the knowing, and the knower). See also Chapter 06 of "F.I.S.H".
P.S. There is much confusion (to put it EXTREMELY mildly) in both Western philosophy and in the so-called “Eastern” philosophical traditions, between the faculty of mind (“manaḥ”, in Sanskrit) and the intellect (“buddhiḥ”, in Sanskrit). Therefore, the following example of the distinction ought to help one to understand the difference between the two subtle material elements:
When one observes a movie or television show on the screen of an electronic device that one is holding in one’s hands, one is cognizing auditory, textural, and visual percepts, originating from external objects, which “penetrate” the senses of the body, just as is the case with any other mammal. This is the component of consciousness known as “mind” (at least according to the philosophical terminology of this treatise, which is founded on Vedānta, according to widely-accepted English translations of the Sanskrit terms).
However, due to our intelligence, it is possible for we humans (and possibly a couple of other species of mammals, although to a far less-sophisticated degree) to construct conceptual thoughts on top of the purely sensory percepts. E.g. “Hey - look at that silly guy playing in the swimming pool!”, “I wonder what will happen next?”, or “I hate that the murderer has escaped from his prison cell!”.
To provide an even more organic illustration of how the faculty of mind “blends” into the faculty of the intellect, consider the following example: When the feeling of hunger (or to be more precise, appetite) appears in one’s consciousness, that feeling is in the mind. When we have the thought, “I’m hungry”, that is a conceptual idea that is a manifestation of the intellect.
So, as a general rule, as animals evolve, they develop an intellectual faculty, in which there is an increasingly greater perception of, or KNOWLEDGE of, the external world (and in the case of at least one species, knowledge of the inner world). In addition to these two faculties of mind and intellect, we humans possess the false-ego (“ahaṃkāraḥ”, in Sanskrit). See Chapter 10 of "F.I.S.H" regarding egoity.
Type/Token is an interesting approach Richard. I’ve often wondered and argued that the first person quality of consciousness is actually mundane, and actually about geometry. It is fundamental, but not in the way that panpsychics mean. It is more simple and spatial. A hole in a fence has a point of view. It’s this hole not that hole. Consciousness has nothing to do with that bit. So yeah, like you say, type is not token.
She has neither and then she gains both no? One through direct experience and one through extrapolation of that experience. So it changes nothing about the knowledge argument.
This has to be a troll. Although Goff is a genius, and I'm a fan, but his rebuttals to physicalism, here, were terrible. "What about Mary's CURIOSITY!??" That's IT, Goff? WTF!!! Oh, and the big finale from Goff, "It just SEEMS like there's something MORE to FEELING pain... you know what I mean?" Wow! Again, Goff you're a genius, but you really let those of us down who wanted to see a good take down of materialism. almost this ENTIRE VIDEO was excellent arguments FOR MATERIALISM, of which our only hope, Goff, did NOTHING ABOUT! Sorry, for my Internet rage, Goff. Big fan, I promise.
I’m on the side of the physicalists, but in Goff’s defense I’ll say that the Knowledge Argument is THE argument. It’s asking a lot that he come up with something better than that.
I mean, the knowledge argument is good, but physicalism dies because it is internally inconsistent. It's over for physicalism. You cannot start with consciousness, then define something else as non-conscious, then say that your consciousness IS that non-conscious stuff. Inconsistent. What's more, it fails on modal grounds. You can know all physical facts yet from them not get a single quality of the world or a single experience. Not only is that the case in our world but in possible worlds, as well. If physicalism were true then it would be necessary and because it's the opposite of that, it fails. This is just the start, as there are other facts that show physicalism as untenable. Materialism, though related, is also doomed. Goff is a complete disappointment, at least in this vid. @@aaronshure3723
@@robertblankenship5000 I'm not convinced by any demand for consistency or total knowledge. I start from the assumption that I am not the world but a part of it. And I assume that it is unlikely that the world will satisfy my personal demands for satisfaction. This is supported by how stubbornly not only the world, but my mind seems to resist my attempts to control them. I also find a priori arguments modal or otherwise to be a poor starting place. I mean if a self contradiction implies infinite conclusions, I think you should take all logical arguments with a grain of salt. Goff has gotten me to question whether or not my attitude leads necessarily to physicalism. Kastrup too. They both want the world described by science to be taken as a version of real. So that's the part I'm working on.
@@aaronshure3723The knowledge argument suffers from an assumption: that knowledge (in physicalism) can only come from communication with the world outside the brain. But this isn’t true. There possibly is also communication from within the brain.
@@falklumointeresting. In a similar vein, Frankish has noted that there is also “internal behavior”.
I came up with the term “skull dichotomy” fallacy. It’s the assumption that the skull is some sort of boundary between “inner” and “outer”world. The skull dichotomy is at its strongest when talking about fMRI studies. You need to ask yourself does this fMRI show me something different than what I learn from talking to the subject? Does this fMRI show me something more than I can learn from the rest of the body?
So it could be that Mary’s neocortex knows something that her eyes have yet to experience. Or her blabbity blab part of her brain has knowledge that the thingymabob part of her brain doesn’t yet know.
very good
how would to seem to Mary if the visual input suddenly included pain or sound? if a ball suddenly turned from red to gratitude? why doesnt that happen? we have no clue at all other than "but thats wrong, that couldnt happen". and yet syntetesia happen. people hear colors. seeing red or blue is not about knowledge. its not a new fact. its just one of many inner experiences we can have. those that have done drugs probably have seen colors we never normally experiences. its not nee knowledge in itself.
If Mary had a brain implant that stimulates the same part of the brain as seeing red, will she learn something new when she sees red through her eyes?
That is a good question. Would she actually experience red or would she experience a shade of grey? Would the rods ever have been activated in her dreams? If you stare at green long enough then look away, you will see yellow (or something like that). How would that work with Mary if you just allowed her to see green and no other color?
Not taking sides between physicalism and panpsychism, but the knowledge argument is severely flawed. Here is why: Imagine a human reading and learning everything possible about four-dimensional space. Then we know we would still be missing something important: the ability to “feel” 4D. We humans can only experience 3D. But the difference would be material: biological brain structures to allow the brain to capture 4D. (In AI research, we would say to build a latent representation of 4D data.) So, what is missing is latent space information. A brain function. This does NOT allow to deduce that consciousness must have a non physical side. Only to deduce that consciousness needs internal state which may well be physical.
Mary (before) knows everything worth saying about red. She knows how to talk to you about it, she knows how to write poetry about it, she knows why you can’t see it if you’re color blind, and she knows how to cure your color blindness if it can be cured. The only thing she doesn’t know is a fact, not about the world, but about herself. It’s not actually that much or that important. If you want to build your metaphysical castle on that one grain of sand, have at it, but it seems misguided to me.
I don’t know if you’re seeing the point of the knowledge argument. If all the pertinent physical facts associated with seeing red are insufficient at getting at what it’s like to see red, then there’s a fundamental epistemic gap between physical stuff and experiences. This is a problem assuming you don’t want to grant consciousness this special status of being this one irreducibly emergent thing in the world.
@@Sam-hh3ryIs it some spooky emergent thing to say a hole in a fence has a different point of view from a different hole in a fence? They don’t see but they have points of view. It’s so trivial that it seems special. It’s just particular v universal. Not even subjective v universal. Just the inevitable result of picking one point out. That’s what Mary lacks.
@@aaronshure3723 your analogy doesn't work and misses the point. there's nothing about looking at something from a particular angle that can't be empirically predicted and verified (based on, for example, the size of the hole, your distance and angle, the structure of your eye, etc.). on the other hand, there is absolutely nothing about the phenomenal character of ANY given experience that can be empirically predicted or verified.
@@Sam-hh3ry anesthesiologists and drug dealers beg to differ about the predictability of subjective states, but keep enjoying your ineffable mystery mate
@@aaronshure3723 loool you have absolutely no idea what you're talking about. I guess you should start by learning the meaning of the word 'phenomenal' and familiarize yourself with concepts like 'hard problem' and 'explanatory/epistemic gap'. I assure that drugs and anesthetics have literally nothing to do with what I'm talking about.
People sometimes bring up Frank Jackson's conversion to physicalism as a point to weaken the knowledge argument, as if the argument was so bad the creator had to abandon it. So, it was interesting hearing that he switched to illusionism, and he still thinks the knowledge argument needs to be taken seriously by physicalists.
I wonder if the deeper disagreement isn't between physicalists and anti-physicalists, but between those who take the epistemic gap presented by qualia seriously, and those who don't.
I remember studying this in school and my thoughts on it don't seem to have changed. Experience and knowledge are different things. There is a physiological difference between what happens when we learn about a particular thing versus when we experience that particular thing. The idea that because this difference exists then physicalism is false is just bizarre to me. It would be like reading a review of a novel that describes every thing that happens, nothing being left out but also nothing being quoted, and then saying you've read the novel. Well, you haven't, and you certainly wouldn't think so after reading it. You wouldn't have experienced the author's voice, the rhythm and flow of their writing, or the imagery and dialogue in the way that only your mind could imagine from reading the actual text. Honestly, I'm amazed that anyone was ever swayed by Professor Jackson's argument. It is inventive, but it isn't persuasive.
You’ve completely missed the point of the knowledge argument. If knowledge of a what it’s like to have a particular experience requires actually being in the associated physiological state, then there is an epistemic gap between knowing the physical facts about that physiological state and actually being in that state. All we’ve done is reframe things from talking about what it’s like to see red to what it’s like to be in the physiological state associated with seeing red. It’s the fact that there’s anything it’s like at all to be in a particular state that is the issue.
@@Sam-hh3ry No, I get that. What I'm refuting is that the epistemic gap falsifies physicalism, not that the gap doesn't exist.
@@micdavey It seems to falsify physicalism if you consider physicalism the view that everything should be reducible and describable in terms of physical concepts. If we can't know anything about consciousness working purely from physical states, then I don't know what it would mean to call it physical.
@@Sam-hh3ry Well, for one we might have to expand our understanding of physical concepts. I mean, we don't fully understand how consciousness works. It can be a physical phenomena but not currently explanatory. And because we can't explain something today, that doesn't mean it's unexplainable. I'm also not trying to say that physicalism can't be false, just that the knowledge argument doesn't falsify it. And I guess just to be clear, we do know a lot about what consciousness is like working from physical states, we just can't know exactly what it's like based our current understanding of the physical states that determine states of subjective experience, and we might never know. The tl;dr here is that it's too early to claim what Jackson claims --that's all.
@@micdavey We know nothing about consciousness working purely from physical states. Literally everything we know about consciousness requires comparing our personal experiences against correlated physical states. If the premise of the knowledge argument is correct, then this gap can't be bridged even in principle. I don't know what kind of scientific theory we can expect to have when we can't measure, quantify, or model the phenomenon in question. And if we can't do those things, I don't know what it means to nonetheless call it physical.
If Mary has the neural wiring capable of distinguishing/perceiving colour before exiting the b&w room, once she observes the novel sense data (i.e. that which is interpreted/perceived as polychromatic; e.g. particular photon distributions or artificial stimulation of the brain), she will become aware of this novel experience, and the encoding/memory of this awareness in the neural circuitry is itself new knowledge. From a physicalist perceptive, the phenomenological experience is irrelevant; only the agent's model/belief in the phenomenological experience.
1:49:00 - the knowledge argument has no implications for the philosophy of mental properties; there is no new knowledge endowed by mental properties (phenomenological consciousness). The "knowledge" i.e. belief/model of mental properties is encoded in the brain, and their existence is functionally irrelevant (empirically unobservable).
Phenomenal consciousness is empirically unobservable, yes. That is the whole point and the reason that there is an epistemic gap in the first place. If the move is to deny the existence of phenomenal consciousness altogether, that’s a valid direction, but you’ve done nothing to justify it.
The phenomenological experience itself is not new knowledge, at least not from an empirical point of view with respect to information (or without additional assumptions such as substance dualism or epiphenomenalism). I therefore think the concept of an epistemic gap here is invalid (irrelevant/misleading), philosophy of mind is not primarily concerned with the experience of qualia A (monochromatic) vs qualia B (polychromatic), it is concerned with experience vs non-experience. Under physicalism broadly construed (e.g. Chalmers) the phenomenological experience has associated (mapped) knowledge encoded in the physical system.
@@richardbrucebaxter I can't help but feel my reply is identical as I wrote above.
"Belief" and "transparent" here seems even more woo woo than "spirit."
Why does South Korea produce the greatest Women Golfers? That's not a small group like the Swedish Tennis Players and the Australian Philosophers. Each year, more and more great Women Golfers from South Korea.
I am an expert on consciousness. I don’t know what it’s like to see red, have pain, fall in love, play football, enjoy Pink Floyd, taste BBQ ribs or read French poetry. My physicalist philosopher supervisor says my expertise on consciousness will be complete just as soon I acknowledge I don’t exist…or at least come up with a logical proof of my own non-existence. At which point my PhD will instead be awarded to a “ghost in the machine,” in other words, one of those rare students who believes in dualism or some kind of Kantian idealism.
Orange isn’t a real thing. It becomes real when the combination of light frequencies hit your eye’s cones fire in a peculiar way, the neural processing behind the retina to the visual cortex. In short, orange is a lie. It doesn’t exist outside your brain.
You’re Leonardo Da Vinci. You know colour better than most. You see colour but until recently, all the maps you’ve read were all printed in black and white.
Today you visited the library of the Sforzas for the first time and saw a glorious map of Milan in black and other colours. All you got War was an easier way to represent the same information. The map maker is leveraging the peculiarities of the human brain to pack information.
when you aren't seeing red or imagining red, do you have knowledge of what red is like? It isn't anywhere in your current subjective experience, it is only stored in the memory ready to be accessed. So for you to know what red it like, wouldn't it need to be stored in the same format it is experienced it, e.g. wouldn't there need to be some non-physical qualia storage that your consciousness can access? That is qualia that isn't being experienced, which immediately begs the question of why ordinary qualia is experienced.
If knowledge of what red is like is stored in a purely physical way in the brain when you aren't seeing or imagining red, what would prevent Mary from getting that knowledge by studying a brain of someone who isn't actively experiencing red but has that knowledge? Part of that stored knowledge must be non-physical if Mary can't learn it from all physical facts, which leads to my first paragraph.
Memory is not like a computer disk. There is no "format" and storing. The process of remembering is a reconstruction of a particular state the brain was already before in. When you do not see the color red, you can actively reconstruct a simulation of that experience - if your brain was once in that state we call "seeing red", there are possibly still the neural connections needed to actualize similar state (not exactly the same, and due to pruning, maybe some of these connections are already gone) without actual photons of particular wavelength hitting your retina.
The problem with this kinds of arguments is - they are mostly contingent on folk intuition. I would even go further and point out the "begging the question" in it, pointing to phrases such as "is stored in a purely physical way", which presupposes some kind of physical/non-physical duality of "things". It is maybe intuitive, but what would it even mean for something to exist (or be stored) in "non-physical" way?
@@ivanvnucko3056 it presupposes dualism because I am pointing out that there is an inconsistency with it. If it is possible to reconstruct the experience of red after seeing it, why wouldn't Mary be able to do that? Why wouldn't Mary be able to bring about the same neutral connections if she has perfect knowledge of what those connections are? Those connections are not experience - yet Mary is still unable to have the knowledge corresponding to having those connections. That tells me that there is something wrong with the thought experiment - there is a physical kind of knowledge that Mary can't access. The only way to avoid that is to assume knowledge of what it is like to see red is also stored in some non physical storage. I do think there is something wrong with the thought experiment - all Mary knows concerning subjective experience is indexical, she by definition can only have her own experience. If she had someone else's experience, she would be someone else. You can't know a subjective experience, you can only have it.
@@carnap355 "If it is possible to reconstruct the experience of red after seeing it, why wouldn't Mary be able to do that?" - because she never experienced it before, that was the whole point of my comment. The experience itself is just a state (in time of course, not static) of the brain, caused partly by light of particular wavelength interval hitting her retina, which creates some new connections (the experience itself and also the reference to language construct - the word red, if someone tells Mary "see, this thing is red"), which then can be used to try to reconstruct it "from memory". But before that there aren't those connections in Mary's brain. The fact that she "knows" a lot of shit about red is another potentially huge set of connections, but the experience itself cannot be among them. With our state of technology we are not able to "create connections" on demand, not with that level of precision - I think some manipulations are already possible. Nonetheless experience is a different set of connections to knowledge. I honestly do not understand why this is so baffling to so many people. It is perfectly intuitive and understandable to most people that one can perfectly know music theory, and guitar construction theory and all about playing styles and finger movements and so on, but it is a completely different skill to practically take a guitar and fluently play with own fingers. Experience of red is exactly the same thing. Both types of connections (knowledge and motor) are physical, just a different set.
@@ivanvnucko3056 and she can't enter that state without actually seeing red, not because she is limited by being physical but because that's how knowledge of perceptual experiences is formed in a human being. I agree with you there. But a dualist would use one of few arguments such as why is she surprised, what magically prevents the appropriate connections to form when Mary has perfect understanding of someone else's brain experiencing red, and so on. I was trying to say those fail because you can formulate those about knowledge that must be physical even under dualism, since otherwise there would be qualia that isn't being experienced
Knowledge Chauvinism
Reflectory theory of human behavior.
Nothing special. Just standard analysis, an attempt to break the complicated phenomenon down to elementary functions and events. Human mind is composed of various images and associations.
I consider it completed now. We can proceed to applications such as fixing widespread mental disorders or creating conscious machines.