ESR’s New Take on Qualia
http://esr.ibiblio.org/?p=1192#more-1192
ADDED: Even if you disagree with ESR’s take, and many will, this is the clearest definition I have seen on what qualia is. So it should present a useful starting point, even for those who strongly disagree, to argue from.
Mary’s Room is invalid because it sneaks in a false assumption: that if something is knowledge, then it can be learned by study alone. As a counterexample to the story of Mary’s Room, consider the story of Marty’s Bicycle, which I have just made up.
Marty is a brilliant physicist who has suddenly become very interested in bicycles. Unfortunately, his leg is in a cast, so he can’t ride one. While he waits for it to heal, he studies their physics, watches people ride bikes, reads guides on bike riding, and interviews top racers—in fact, he studies everything there is to study about bicycles. When the cast is finally taken off, he buys a bicycle, gets on, and immediately falls over.
The ability to recognize red objects is like the skill of riding a bicycle—it can only be acquired by doing it, not by study, because study can only train the linguistic centers of the brain, not the visual processing centers (Mary’s room) or the balance centers (Marty’s bicycle). This is merely an accident of how our brains work; one could easily imagine a robot that could be told how to recognize red objects or balance a bicycle without having to try it first.
This is a standard reply, known as the Lewis-Nemirow “Ability Hypothesis”. See here for a critique.
I once read an anecdote about a person who was determined to prove that “purely theoretical” knowledge was just as good as actual experience, and to prove it, he was going to teach himself how to swim by reading about it. After acquiring what he felt was a sufficient “theoretical” knowledge of swimming, he jumped into a pool and was immediately able to swim, much to the amazement of onlookers.
It’s probably false, though.
I don’t believe it. If I were one of the onlookers, I wouldn’t be amazed; I just wouldn’t believe his claim never to have swum before.
I was quite surprised to learn one day, after riding bicycles for years, that you steer by leaning to the side. This is information that is quite important for riding a bicycle that had never passed through the verbal part of my mind. And when I wanted to test this, I couldn’t ask my body about it. Instead I had to pay attention while riding to see what I did; also, I was able to lean or turn the handlebars and see what happens. The communication channel between the verbal part of the mind and the mechanical memory is quite narrow. That’s not to say that it’s impossible to send information in either direction, but that is a skill that is quite rare, not a matter of deciding to acquire “theoretical knowledge.” Also, there’s the problem that since people don’t normally learn this way, they don’t record instructions.
Swimming is a lot easier to figure out prior to doing than bike riding is. Falling, for example, is rather more difficult. It is highly unlikely that someone could do this for any sport or activity that involves significant interactive feedback (e.g. balance or aim).
It’s impossible to jump into a pool without learning something about swimming. At the very least, you start learning about how the water responds to your body. And this will remain true even if you don’t consciously try. (There are experiments, for example, where people “learn” to twitch a muscle they can’t consciously control because twitching it becomes correlated with turning off an annoying sound.)
Furthermore, it’s been my experience that people who think they “can’t explain something” really aren’t trying. If you really try (and I’m sure people have), you can come up with pointers that greatly demystify swimming, even if they require you to be in a pool to start making sense.
I don’t know what I would call the hard limits of a first-timer. If they swam at a competitive level the first time, and I had good reason to believe they’d never practiced, I’d have to admit I underestimated what you can learn without swimming. But it’s not unreasonable that someone who was both a quick learner and was told of a known motion that closely approximates good swimming, could swim on the first try.
The first time I went ice skating a couple years ago, I flailed around dangerously for a bit, and was unable to exert much control over my path. I comically ran into a female friend and wound up in an unflattering position, and it was forceful/painful enough for all present to understand that I hadn’t done it on purpose...
Then my roommate took me aside and explained what motion to do, with a few hand gestures. After about 45 seconds of explanation, it clicked in my mind, and I went out and was able to skate around quite gracefully in comparison to a few minutes before. I was surprised at how well it worked.
I suspect that this falls into the “quick learner who is told of a motion that approximates good skating,” though.
Yep, sounds exactly right.
What you describe is what Tim Gallwey calls the ‘inner game’. It is, to simplify a bit, training your intuitive subconscious without letting your conscious awareness interfere. Here is a video of him coaching a woman who has never touched a tennis racket to serve using the technique.
Another similar technique is drawing on the right side of the brain.
Good analogy!
I don’t think the question is about ability. Mary already has the ability to recognize red objects by using a light detector, for example. The question is that, when seeing red for the first time, it seems you’ve learned something new—i.e. what it feels like to see red.
The question boils down to whether subjective experiences are a form of knowledge.
This applies to humans, who have fixed systems for different kinds of learning. If marty or mary were AI’s they should be able to gain abilities through knowledge directly.
I basically agree with what you said, but I would go one step further.
That is, we don’t learn about redness as such. When we get the sensory stimuli of redness, we aren’t learning the ability to recognize red objects as such. What we are learning is things and causes in the world and redness is just one attribute of the sensory data. It is useful, because it makes the prediction of things in the world more accurate.
Let’s take one type of cause in the world that we learn: objects. When we are exposed to visual stimuli of a certain type of object, we learn to expect it to have some type of common pattern. One aspect of this is the color of the object. If the color is nearly always red, we learn to predict that this type of object is typically red. So our prediction of this type of cause in the world is enhanced by recognizing this attribute of the sensory data.
If the sensory data lacks color completely, or if the color is not what we expect it to be, the prediction of our brain is not as accurate. But we might be able to still make the correct inference of the world, it’s just a bit harder. If the color is completely different of what we’ve learned it to be, we are typically still able to recognize the object, but we are surprised about the color, because it doesn’t match our representation. Our attention is drawn to the peculiar color.
The high level representation of the object of certain type, then, is composed of these different aspects of what we have seen before. These different aspects are things like the pattern of the visual stimuli, the intensity of the visual signal, the color of the signal, etc. A person who has been exposed to color stimuli all her life, has these extra bits of information about objects of the world she has learned about. So “sensation of red” isn’t a thing or place in the brain. It’s not even a process in the brain. Redness (and color stimuli in general) is part of our knowledge of the world. It’s part of the objects and categories we have learned. It’s part of the representation of these objects in our brain.
So when Mary, in her black and white room, tries to learn about colors, she cannot directly influence her representation of the world in her brain. She would have to twiddle every piece of knowledge in her head with regards of its color.
As an example, let’s say she has seen a lot of roses from her black and white TV. (Let’s assume for simplicity that roses were pretty much always red.) Now, from her sources, she learns that roses are typically red. She would then have to locate the representation of roses in her head and add a probabilistic adjustment if the color of the visual stimuli is red. It’s something like: “if the visual pattern is like this and if the color is red, then it’s likely that’s it’s a rose.” Or more precisely, it’s some sort of conditional probabilistic table.
So for Mary to learn redness before being exposed to color stimuli, she would have to modify all the bits of knowledge of the world that she has acquired through visual data. This would mean all types of objects and other visual information she has. She would have to add this conditional clause to every type of object she has learned where the color red has any significance to the inference of the object. She would have to add this attribute also to every specific knowledge of the world she has too.
So in summary, this thought experiment seems to me like a huge misunderstanding of how we learn about the world.
Liked this a lot, though it was basically my view already, so I’m decidedly biased.
Interestingly, I don’t think he bluntly said the (perhaps) most fundamental part of his position: what the color red looks like is a fact about Mary’s brain, not about the color red. This also explains how qualia are essentially incommunicable: I don’t have Mary’s brain, so I can’t really observe what it is like for information to be processed in that brain.
It seems that, in theory, an intelligence capable of reconstructing brains, wiring them into its own consciousness, and running them through stimuli could in fact understand their qualia, though such a procedure may not be meaningfully possible.
I agree with this.
It was about the effect that actually seeing the color red had on Mary’s brain. As opposed to learning about the color red from other sources. His claim was that there was something that could only be learned about the color red by actually seeing it, and that was the effect seeing it has on your mind.
That’s the issue. I don’t think you learn something about the color red; I think you learn something about your mind. There is no objective thing that red looks like; it is the result of certain neurons firing. If Mary’s brain were wired differently, the experience of seeing red light would be different if it even existed. The only relevant thing about red light is that it stimulates certain neurons; if you knew absolutely nothing about it except that, you could fully describe explain Mary’s experience of seeing red (in theory, of course; neuroscience ain’t there yet). On the other hand, you couldn’t explain the experience with unlimited information about red light plus only the fact that it stimulated Mary’s retina.
That doesn’t explain why qualia are incommunciable, since others stuff that happens in brains is communicale. We expect Mary to understand everything except qualia.
They are incommunicable because you cannot occupy any arbitrary brainstate you want to. See the ability hypothesis.
To repeat my point: that doens’t explain why other metnal content is communicable.
Other mental content is expressible propositionally, whereas to duplicate someone’s qualia you have to literally duplicate their brainstate (or at least the functional equivalent of their brainstate).
When I say “Greece is in Europe” I am making a propositional claim that is relatively independent of the precise details of how my brain represents that belief, versus how your brain does. Maybe I am picturing a map, while you are remembering Greece’s EU membership. For the purposes of communicating that proposition, nobody cares about what your neural activity looks like when you think about Greece being in Europe. The description is higher level: many brainstates map to a single belief state. This is what makes it communicable.
To duplicate something, you of course have to duplicate it. That isn’t really the point. What you are tacitly assuming is that qualia cannot be communicated, know or understood without duplication. ie, there is something special about them in that regard. That is how that approach fails as a dissolution. Rather than showing there is nothing special about qualia, it assumes there is something special.
I don’t know about yours, but my qualia don’t look like neural activity.
(blink) Can you clarify how you can know something like this about peterdjones’ qualia and not know it about simplicio’s? That is, what evidence do you have for the former that isn’t equally evidence for the latter?
Other mental content is communicable because it’s a lot simpler. All of the data of an entire human brain state, even considering only a snapshot taken at one instant in time, is many orders of magnitude of quantity/complexity beyond what the conscious mind is capable of considering and understanding.
Is there independent evidence for that?
Is there a rule, in the Mary’s room argument, against Mary doing some brain/electricity experiments on herself? Of course, she would be careful to use non-red tools and change her blood somehow so it isn’t red either.
I would expect that such a capable scientist could learn what it feels like to see red objects without any red light going through her eyeballs.
Unless I am mistaken, ESR either thinks that these experiments are not allowed, or hasn’t thought of that possibility. This makes me believe that ESR’s idea that he disagrees with Dennett might be based on reasoning from different ground rules than Dennett.
The way I see it, it’s possible to learn ways to produce the sensation of red, but still learn something upon seeing red.
If each brain is wired up differently from birth, but in ways that achieve similar functionality, then Mary could only learn aggregate and past statistics about the ways brains responded to seeing red. She could make a good guess about the range of responses her own brain is expected to have upon seeing red. But until she sees red, she doesn’t know which response within that range is the one she gets.
As I understood it, Mary is, in the thought experiment, supposed to be very nearly omniscient—she knows EVERY PHYSICAL FACT about color and human perception of color.
My question is whether she’s allowed (before exiting the room) to wire her brain up to a machine that stimulates her neurons exactly as if there was a red thing in front of her.
Would you agree that if she were allowed to use such a machine, then she would know which response her brain in fact would have?
Yes, I do, but I think that turns it into an unhelpful question. If she already knows enough to stimulate her neurons to generate what it feels like “exactly as if there was a red thing in front of her”, then of course she won’t learn anything from seeing red. But knowing enough to do that raises the question of how she could know she’s generating the right feeling, which just regresses to the original dilemma. It’s like saying, “If Mary knows everything, does she know a subset of everything?”
I thought the point of the scenario was to ask if Mary learns anything from seeing red if, before that, she received sensory data in all modes except color sight—that is, she could feel or smell or even see black and white images.
The above reformulation is still fair because, for example, she can run simulations of other people that have information about their reactions to red, and information about her own brain, and she can extrapolate to her internal experience. But I would still say that when she sees red, she resolves residual uncertainty, for example, about which of several branching Marys she is.
As I understand it, the point of the scenario is intended to be an argument against physicalism. Qualia are claimed to be nonphysical, because Mary knew “every physical fact” about color and human color perception, but still (it is claimed) learned something when she saw red for the first time.
Dennett, if I undersand him correctly, argues that the “knowing every physical fact” is a very strong hypothesis, which the exposition of the scenario doesn’t pay enough attention to. With so much information and comprehension, Mary is not very much like a human; our intuition that she would say “wow” is not trustworthy.
In particular, I don’t think that we actually disagree—your posts seem to come from a physicalist, rather than dualist, conception of the world.
Funny how those highly unlikely borderline cases whisk away a lot of the confusion, huh? No committed physicalist can postulate a serious difference between seeing something red and having a virtual red-thing pumped into your optic nerve, I would hope. I think that’s a far more useful scenario than thinking about someone suddenly be able to see colour. In fact, you could probably keep moving a step towards your magical ‘inner perceiver’ and asking whether ‘it’s a real experience of redness’. That’s not to say that qualia are fundamentally dualistic, simply that it’s idle to discuss that question while we clearly don’t have the capacity to get anywhere near it at this point.
I’m always surprised to see people arguing for or against the existence of ‘qualia’ or similar as though they’re making some sort of contribution to the field. Particularly when it’s such nonsense as ‘if brain science doesn’t come up with an explanation for these thingies, it’s not complete!’
It sounds like many of us agree: the experience of qualia is a real physical phenomenon. One that occurs in our brain somehow and is incommunicable because we don’t mutually observe what happens in each other’s brains. (Or even our own except in a limited way.)
So what’s the other point of view? Why do dualists want to insist that there is a non-physical aspect to qualia?
One such point of view is that currently all existing explanations for qualia—even the ones that try to be materialistic—are actually dualist. For instance, suppose we have a materialist-computational explanation of the brain, which explains how each neuron performs a certain type of computation, and how certain activity patterns in the neurons correspond to specific mental states. We find the model to be good and to accurately predict what people will do. So far, so good.
Yet here we run into a problem—we can now fully explain the brain and one’s doings in terms of the computational model, without needing to include actual qualia or subjective feelings in the explanation at all. The model works just as well regardless of whether or not we experience phenomenal consciousness, or are p-zombies. In trying to explain mental states, we have actually explained them away and created a theory that has no need for them.
Therefore, the argument goes, assuming we don’t deny the existence of qualia entirely, we must dualists. After all, we now have two models, one physical model which doesn’t include qualia, and one which assumes their existence but provides no physical explanation.
No. We can now assume that we can fully explain the brain without qualia. If qualia are in fact reductionist, any full understanding of the brain will show how they exist and what causes them. “Assume we have a perfect model of how the brain works that doesn’t imply qualia” is no different from asking, “Assume that qualia are essentially dualist.” If you can actually generate this model, you have an overwhelming counterargument. But we can’t just suppose we have a perfect model of the brain without qualia, because whether such is actually possible is the exact question we are trying to answer.
To foresee a counterargument, this statement quoted (if I’m reading it correctly) also assumes that accurately predicting what people do is the same thing as fully explaining the brain. This creates the possibility that qualia exist, but if they are an effect of behaviour rather than a direct, independent cause, you can pretend they don’t by assumption. I think a full explanation of the brain needs to explain everything the brain actually does, rather than merely predicting behaviour. I’m not arguing that qualia have no causal effect; I think the word “cause” gets too muddied to break down at this point, given current science. Only that if they did not have a clear effect that is not explained by another proxy, you could exclude by definition and they could still very much exist.
If this is the dualist argument, then I wouldn’t find it compelling but just confused about what reductionism does.
In the reductionist worldview, higher level things can be explained in terms of lower levels. For example, table salt can be explained in terms of the molecules sodium and chlorine, which in term can be explained in terms of electrons and protons, which in turn can be explained in terms of quarks, etc. Having explained table salt in terms of quarks doesn’t mean that its been explained away though; table salt exists just as well as it did before. (There is the argument that categories of things and our frameworks for things don’t really exist, but I don’t think that is what this debate is about.)
Instead, I’m pretty sure the dualists are asserting that there is some aspect of qualia that cannot be captured physically. So their idea is that the experience of qualia may occur via a physical process (neuron activity) but that it possesses some additional non-physical quality.
I suspect that our paradigm of what the world is like has changed so much that we can’t really relate to what the dualist/materialism debate used to be about. I hypothesize that in the past, we thought of “physical things” as the set of things we could touch and manipulate tactilely, but thoughts appeared to be a completely different category. We now all understand that thoughts occur in the brain and are neural activity. (For me, thoughts occur as sounds usually, so when I’m thinking I am activating different memories of sounds that I have, that are in turn associated with and handles for the ideas I’m thinking about. )
I think this view that thoughts = neural activity is pretty mainstream now … I wonder if dualism is really obsolete then?
WTF? AFAICT that just means the whole notion of “qualia” is just begging the question, in precisely the same way that p-zombies and Chinese room arguments do.
You dont see colours or feel pains?
The comments in this thread have been concerned with “the experience of redness” and whether this exists or not or is a real physical phenomenon or not. The answers appear to be straight-forward (yes, and yes) and I am left wondering what the dualism debate is all about. Below, I even argued that dualism may be obsolete.
Dualism is obsolete if dualism = {the experience of redness is a metaphysical phenomenon}. But then I remembered some ideas (15 years ago for me) from, “Zen and the art of motorcycle maintenance”. Dualism is not really about whether neuronal activity can account for all subjective experience. (It can, and a dualist need not object to this.) It’s about whether redness exists.
Redness does exist, so redness is not the best example of qualia for understanding dualism. Consider the quale emphasized in Pirsig’s book, “quality”. We all agree quality exists in some sense (we’d all like to have quality posts on LW) but dualists want to think that “quality” exists in a real sense out in the territory. A quality book and a quality cheese have something in common; an essence of “quality” attached to them.
My two cents on the issue: I think that some qualia might exist in the territory, in some complicated way. But if they do, they’re physical and we just haven’t figured out how yet. I’ve babbled before about how I think qualia like “beauty” are concerned with the subjective experience of patterns, so these qualia are going to be more difficult to describe physically than, say, the experience of redness. We have no well-developed scientific theory for pattern recognition. We can write a code that recognizes a pattern (e.g., the letter “A”), but we don’t have any idea what a pattern is.
I think I’m beginning to remember...(from “Zen and the Art of Motorcycle Maintenance,” which was 15 years ago, for me)...
Everyone, dualists and physical materialists alike, may agree that the experience of redness is a real physical phenomenon (as something that occurs in the brain). But isn’t the qualia/dualism debate about whether a red object possesses a property (quality) of “redness”?
Am I being down-voted because you don’t think objects can possess qualia (like redness or beauty, etc) or because you don’t agree this is what the debate between physical materialists and dualists is about? I think it’s important to remember what the debate is originally about, because the way it’s being framed here, it doesn’t seem like there would be one.
Not downvoting you, but the debate isn’t about that. It would be more about “even if you knew all the physical facts about the brain, how those things implement the various algorithms the brain implements, etc, would there still be some aspect of redness that ‘transcends’ that, an epiphenomenon that one needs ADDITIONAL facts to learn about from the outside?” (well, okay, I guess that isn’t exactly true of all flavors of dualism… Since the flavors of dualism that have the dual properties actually have physical effects and so on would be a bit different, but then arguably one could even rephrase that sort of belief so that it’s not even a form of dualism anymore, or becomes an epiphenomena style of dualism)
What you’ve described is consistent with what I find on the internet and what s being discussed on this thread. I just thought there might be more to it..
Ahh, I was fully on board until the ‘incommunicable’. Not even sure what it means to say something’s uncommunicable by definition, except that it seems to introduce a brand new element of mystery. Is the ‘quale’, then, the element of the redness that doesn’t go across when I tell the other person about the letterbox? Sounds like putting a name on ‘the stuff we don’t fully understand yet’ - not helpful.
To be honest though, it’s not a particularly interesting question. We are physics. All we’re doing is shedding light on the particular oddities that arise from experiencing the brain from the inside—there are no deep insights here for me.
On the incommunicability of qualia:
First, I disagree that qualia are inherently incommunicable, they just always have been.
As children, we learn what a tree is when someone points to a tree and says, “tree”. In this way, “what a tree is” is communicable.
Red is communicable too. However, what you experience when you experience redness has not been communicable, because it happens inside the brain. When someone says they observe something red, you can only assume that they have some experience analogous to your own experience.
But there’s no reason why we can’t observe the experience indirectly, eventually, as neuroscience develops. Hooked up to the right machine, it could indicate if I’m “experiencing redness”, either because I’m observing something red or imagining something red.
No. This means that the label “tree” is communicable, and that with any luck, native ability to separate natural kinds will fill in what the label is meant to contain. The definition of “tree” is also communicable, but it doesn’t involve pointing and isn’t closely wedded to the internal mental understanding of treeness. (My mental understanding of treeness is something like “plant that holds its leaves or needles a ways off the ground with a large stick of wood”. The definition involves exciting things like “apical dominance”.)
Helpful. We need to put labels on Stuff We Don’t Understand yet so that we know what we are talking about when we try t understand it.
Peter,
As a general strategy for considering a black box, great. As a vehicle for defining a mysterious ‘something’ you want to understand, potentially useful but dangerous. Labelling can make a job harder in cases where the ‘thing’ isn’t a thing at all but a result of your confusion. ‘Free will’ is a good example. It’s like naming an animal you plan to eat: makes it harder to kill.
Ben
But we don’t know that qualia aren;t anything and we don’t know that about free will either.
Downvoted for lack of description of the (topic of the) link.
Seconded, I also downvoted for lack of summary. And ESR is known for talking out of his ass, so I’m not going to bother reading that unless I’m sure this is much more insightful than his typical writing.
seems to me that ESR is basically right, except, I’m not sure Dennet would even disagree. Maybe he’ll reply in a comment?
Yes, by now ESR was corrected in the comments and has mentioned that he probably misread Dennett.
I’m not sure either, but at the least his explanation of what he thinks qualia is is clearer than any other that I have seen.
It has been my perspective that the experience (and thus existence) of qualia is completely consistent with physical materialism.
It was in a thread about theism, which I don’t want to pick up, but in a tangent about qualia here I argued that “beauty” existed (physically) when an observer has a special sort of experience in response to the subjective observation of a pattern. The special sort of experience—just now read from “ESR’s New Take on Qualia”—is incommunicable but an experience we learn over time to associate with the label “beauty”.
Probably my comment that I linked to is too confusing. The confusing part is that I was trying to explain where the “meaningfulness” comes from in our experience of qualia. I was describing a hypothesis that the feeling of “Wow!” comes from the (subjective) experience of “significant” pattern (subjectively considered significant). So “Wow!” is a qualia response you may have in response to the experience of the qualia “beauty”. So “beauty” is a feeling about the pattern of whatever is being thought beautiful, and the “wow!” is a feeling about the pattern of beauty (the experience of beauty) itself.