This is nicely done. As someone who has been involved in what, by now, must be cumulative years of discussion about interior and exterior decoration and, consequently, colour I am intrigued by this and the concept of colour constancy as well.
Anyway, really looking forward to something I have known about towards the end of the last century (great being able to say that), memory and learning. Thanks again Suzi.
I imagine you must have some fascinating stories about colour constancy from all those discussions. Those illusions like the blue and black dress (or is it the white and gold dress) always fascinate me.
Looking forward to exploring memory and learning next week, too! 😊
A thorough analysis of the Mary's Room thought experiment, with some relevant neuroscience thrown in. I particularly like the monkey experiment. I'd heard of others along those lines, but not that particular one.
I've never been concerned that Mary's Room challenges physicalism for exactly your final point. The premise just assumes Mary can have all the physical information without one of the pathways to it, personal experience. If she can't, then her learning anything new has no metaphysical implications.
The reality is none of us have Mary's knowledge, or what it might entail. So, even neuroscience complications aside, any statements about whether she'd be surprised or just thrilled at confirming her hypotheses, is pure guess work. It's a blank slate people can project their intuitions into, like most of these types of thought experiments.
That said, it's worth noting what even a Mary with contemporary knowledge about color and vision would know. She'd know the relations between the colors, that violet blends into blue, cyan, green, lime, yellow, orange, and red. She'd know that reds and yellows are more striking, stand out more, than blues and greens. She'd expect a clear sky and deep water to have similar hues, yet be distinct from the similar hues of the sun, daffodils, and bananas, or that of blood, roses, and ripe strawberries. She'd be able to anticipate the unease of walking into a completely red room.
There's an enormous amount about color experience she'd be able to anticipate. (Assuming of course that she'd be capable of it immediately, or ever.)
On information, I actually think of information as causation, or maybe a snapshot of causal processing. Which raises the question, if Mary does learn new information, what effects can it have, on her behavior, ability to write about it, etc? If none, then it's not "information" in the sense we normally mean with that word. I don't think a Mary with the profound knowledge in the premise would learn any new facts in any reasonable meaning of the word "fact" or "information". There'd be no update to her world model. But since I don't have her knowledge, I'll admit that's my intuition.
Thanks Mike! As usual, you've given us lots to think about here.
I'm curious about your framing of information as causation. This makes me wonder about cases where we have information that doesn't seem to directly cause behavioural changes. Like when we know the sun is actually white but we continue to see it as yellow. The information doesn't seem to have the same causation as other types of information. Would you consider that information of a different kind? Just curious what you think.
Good question. What I'd say is that learning the sun is white changed the state of your brain, so that you could later use it as an example here. :-) Information doesn't have to result in immediate behavior, just have an effect of some type.
I often think of it in terms of logic gates. One signal to an AND gate results in no further propagation of effects. But if that signal arrives in tandem with another, we get a propagated effect. And it's actually not true that the lone signal had no effect, just one that resulted in waste heat.
Where this view does get under some pressure is in a isolated system in a high entropy state. There's arguably a lot of hidden information in such a system, but without outside energy, little to nothing is happening. (This is why I added the weaselly qualifier "or a snapshot of causal process" above.) But again, if energy is ever added to that isolated system, it's hidden information would get its chance to have effects.
I should admit that I got the notion of information as causation from an offhand remark Eric Schwitzgebel once made on his blog in a discussion with Eric Borg. It summed up a relation I'd been wrestling with for several years. And fits with Gregory Bateson's idea of information as a difference that makes a difference.
That said, I'm always on the lookout for anything that actually dissociates the concepts of causation and information. I'll admit it seems a little too neat. But so far I haven't seen any.
It's also worth noting that this fits for physical information, information as physicists see it. Semantic information, information that provides actual knowledge to an agent, seems like a subset and so is a narrower concept.
Thank you, Suzi. (As always.) I think you've given me (and all of us) much to think on there.
One thing I did think about while reading this was "what is 'normal' colour vision, anyway?" in both objective and subjective terms.
A couple of anecdotal data points:
When I was being tested to join the military, I aced all the "colour blindness" tests. Scoring well above-average on the tests designed to test fine gradations of colour discrimination. Some of my friends (including some who passed civilian colour-blindness tests for civil aviation) 'failed' those tests - although, functionally, they felt themselves far from a naive understanding of what "colour blind" might mean.
Another: I drive what I've always regarded as a 'blue' car. (It's a 40yo car in dark metallic blue paint.) Even the registration papers for my car describe it as a "1983 Blue Toyota Celica Supra Sedan". Yet: many of my friends say my car is "grey" not "blue". Go figure. Further: when they see my car in photographs (even photos they've taken themselves) they see it as "blue" in the photos - but not "in real life".
I'm not sure what's going on in that 2nd example. Is it a definitional problem? I don't think so - when they see it via an 'objective' instrument (like the camera in their phone) they can see the 'blueness' - yet they seem unable to see it with their own eyes - while I can and always have. As have others. Which makes me think there's some kind of real perceptual difference happening. But what kind?
I don't have any theories or answers about that. Only questions.
Your military example reminds me of something I read recently. People who are colour blind actually detect camouflage better than those with normal colour vision. They are able to discriminate between shades of grey better which helps them notice subtle variations in brightness. So the military may actually benefit from having colour blind people.
As for your car example, I wonder if this is related to colour constancy.
Do you remember the black-and-blue dress (or was it a white-and-gold dress)? Depending on whether we assume the dress is in a natural lighting or under artificial lighting, affects how we perceive the colour of the dress.
In normal circumstances, our brain maintains colour constancy - keeping objects looking the same colour despite changes in lighting. But as Malcolm alludes to, photos flatten this process -- the camera has already done some of that work for us. It's a bit like how we might not notice a blue tint from fluorescent lights until we take a photo and suddenly everything looks obviously blue.
But we know people's brains handle colour constancy different (that's why half the population thinks the dress is blue and black and the other half think it's white and gold). There are genuine differences in how we perceive the same object.
Colour constancy might be part of it (I’ve spent way too much of my life on - strictly amateur - colour-managed workflow in photography, so I know exactly what you’re talking about there).
But I suspect Malcolm might have been closer when he talked about colour saturation. It might well be that the people I was talking about have a higher (than me) saturation threshold that has to be met in the blue part of the spectrum before they see a blue tint. And that camera-phones tend to boost saturation to make “more attractive” photos.
Modern cameras and camera-phones are enormously better at AWB (“automatic white balance”) when colour temperature (eg. direct sunlight vs shadowed light) or tint of light source varies (especially so under artificial light, these days). So I suspect that saturation levels might be more important here than colour casts.
It might also be that I’ve spent so much time trying to get colour-casts ‘just right’ between what’s on-screen vs a photographic print that I’m just more practiced at colour discrimination. (I'm also more prone than most, I think, to having the 'reference white' that guides my colour perception be context-dependent. I have to be aware of that and compensate for it when editing photos.)
It’s hard to know, because all of the above (and likely a good deal more) feed into differences in colour perception. But I am sure those differences exist.
Your experience with colour likely has had a big effect.
I've spent several years in a lab investigating human vision, primarily with colour. Part of this work was coding colour games for participants to play. This means I've spent many hours testing these games, which involved many, many hours looking at and trying to distinguish between different colours. I've often had the impression that I've got better at it over time.
I wonder whether something similar has happened with your ability to have a context-dependent reference white. It makes sense that it would be the case.
Interesting as they are, I don't think the developmental aspects add much to the intuition pump. It's about the difference between the perceived world and the lived experience. You could equally use any optical illusion, or the cultural ability to recognize vertical lines and right angles.
Hi Malcolm! Thanks for your interesting comment. I'm a little confused though - you mention that the developmental aspects don't add much, but then say Mary's Room is about the difference between the perceived world and lived experience. I'm having trouble seeing how the development aspects could be separate from lived experience and the perceived world. What am I missing here?
For me, Mary's Room is about the ineffability of qualia. It uses "red" as an example, but red is just an example. To what extent qualia are learnt, arise automatically as a result of physical development or have "always" been there as part of consciousness, is a supplementary question which needs to be addressed individually for each quale.
I see it now: you read Mary's Room as an exploration of the development of qualia. I'm not sure this works cos it's so artificial (and why was there so much emphasis on Mary being an all-knowing scientist when this requires only naivety?), tho it does serve to stimulate discussion.
Copilot says: It is designed to challenge physicalism, the view that everything about the human mind can be explained by physical processes alone.
Or even that "everything about the human mind can be explained", ie not ineffable.
I'd prefer to say that this universe is (obviously) awareness-friendly, but nobody can say whether that's a specific unknown property of space/time (dualism), or an emergent property of everything else (physicalism).
Another great one. The question that pops into my mind is whether Mary, knowing everything, knows how to teach herself to see colors and what teaching colors would involve. And then, of course, how to teach those poor monkeys how to see color - I'm not so comfortable drawing the ethical distinction between depriving humans and monkeys in the way most experimental scientists seem to be. I guess another approach would be to say, if you were trying to set up a computer simulation of the test so as to avoid the ethical concerns, what exactly would you try to encode the computer with?
The idea of Mary teaching herself to see colours is intriguing. Even with all her knowledge, would she know how to guide her own brain through the development process that normally happens with experiencing colour? I doubt it. It's a bit like knowing everything there is to know about how to play tennis, practicing the swinging motions, the standing positions, how to hold a racket, but never actually playing a game of tennis.
Your question about computer simulation is the key question, isn't it!? Could we do it -- could we create a computer simulation that perceives colour? What is missing if we can't?
I'm looking at all the questions you face on this one, all the comments... the price of being so fascinating and good at what you're doing is very high!
Thanks so much, Jack! You're right, the questions and comments are wonderfully thought-provoking, and I wish I had time to explore every fascinating thread. It's actually a lovely problem to have -- being part of such an engaged community of curious minds. I'm truly grateful for how thoughtful and generous everyone is with their insights and questions.
I have to forgive you for ignoring what I thought were some pretty good follow-up questions on the last one, where I asked if you knew anything about psychoacoustics and what that might do to the blind subject attempting to echolocate! I think if I had to pick the very best example of what Substack could be at it's very best, it would be your posts.
I did read your questions about psychoacoustics and echolocation -- they are fascinating questions and deserved a proper response. I've been caught up in work commitments this week, but I'm planning on getting to it (and some others) on the weekend when I can give it the attention it deserves.
And thank you for your incredibly kind words about the newsletter - they mean more than I can say!
It seems like you missed the point of the Jackson thought experiment: “Mary’s Room was intended to expose what science can’t explain. But maybe the more interesting thing it shows is how complex learning and knowledge is.” More interesting to you I suppose but an irrelevant cul-de-sac. This seems common among those who cling to a simple physicalism as foundational. I can make a similar criticism of EINSTEIN’S thought experiment of riding on a beam of light—holy cow! Can’t be physically done! He’d be ripped to shreds! Instead of color try doing this criticism with Mary getting a dose of poison ivy or learning she is allergic to peanuts. Does not take years an experience to understand that new experience.
I think we might have some common ground here. You mentioned in another comment to @wyrdsmythe you mentioned that "the best criticism of the Jackson mind experiment is that final leap, that therefore physicalism is false." I completely agree -- that leap isn't justified. The scientific evidence I presented here wasn't meant to prove or disprove physicalism. Rather, I think it simply suggests that Jackson's argument might not work as early Jackson intended.
Your Einstein comparison is interesting. Einstein's thought experiment about riding a light beam was primarily exploring logical and mathematical relationships -- to me, whether he'd be "ripped to shreds" doesn't affect the underlying mathematical insights he was pursuing. But with Mary's Room, the physical structures of the brain are central to the question being asked: would Mary learn anything new? If learning itself involves physical changes in the brain (as neuroscience suggests), then the thought experiment's premises might be internally inconsistent.
You raise an interesting point about poison ivy and peanut allergies. These immediate physical reactions are quite different from complex perceptual learning. But I wonder -- even in these cases, wouldn't the brain need certain physical structures to process and recognise these new sensations?
My point about Einstein was that thought experiments are, in many cases, wildly hypothetical. There is an assumption in such hypotheticals that all other circumstances about reality are held in place. Obviously Jackson’s and Einstein’s thought experiments could not be conducted in reality. Thus, the critique by some on this basis is missing the point. And that was my point. You further say that even in the examples I suggested that “wouldn't the brain need certain physical structures to process and recognize these new sensations?” Well yes, of course. I’m sorry, I fail to see your point here. I was merely attempting to provide a work around to the silly argument that the experiment could not be conducted in reality—which because it is wildly hypothetical anyway should not be necessary.
By the way, I also think argument that because Mary knows everything—implying some sort of omniscience that no one has—that she will recognize the color red. That is laughable. It’s similar to medieval arguments where one whistles up God to fill in what can’t be explained.
Let me see if I understand your argument correctly:
-- Thought experiments are meant to be hypothetical and don't need to be physically possible
-- My critique about brain development misses the point because, like Einstein's light beam example, we're meant to assume other aspects of reality are held constant
-- The fact that we need physical brain structures to process sensations is beside the point - that's just one of those aspects we're meant to hold constant in the thought experiment
1.) Yes, thought experiments are hypothetical. And many are not possible to duplicate physically. Some obvious examples are Hilary Putnam’s Brain in a vat, John Searles’ Chinese room, and Phillipa Foot’s Trolley problem among others.
2.) You’re asking if needing brain structures to process sensations is one of those aspects we are meant to hold constant in Jackson’s hypothetical. The implication of your question is puzzling. I suspect I have not made myself clear or I’m slow to understand where you are going with that fact. One can accept the fact that we need a physical brain in order to have a functioning mind (which I certainly do) and still see an interesting philosophical challenge in Jackson’s thought experiment. What I mean is having a brain-dependent mind is not a refutation of the thought experiment if that is where you are intending to go with that. Otherwise, yes, one needs “brain structures” and other complex physical activities in a brain “to process sensations.”
I previously alluded to a form of knowledge that keeps popping in debates over this thought experiment. I think that needs repetition. Mary has so-called “complete knowledge” of the functioning of the brain according to the rules of the experiment. This is often implied as “omniscient” knowledge. As I said previously, that interpretation is similar to medieval arguments where one whistles up God to fill in what can’t be explained. It is an expression of faith and nothing more—faith that complete knowledge (when and if it is attained in an imagined future) will justify a materialist interpretation of reality. It’s an interpretation of the ground rules of the thought experiment that—through circular logic—guarantees only one conclusion. It’s a unpersuasive argument.
I assume such mental gymnastics are inspired by concerns that Jackson’s thought experiment threatens a firmly held belief about reality. To maintain a belief in a certain form of physicalism one cannot fully acknowledge that Mary has a truly irreducible subjective experience. To do so, it is feared, opens the door to a nefarious dualism—or worse. I don’t think it does.
Quite simply it seems obvious that a causal explanation is not an experience. At present, objective explanations of the physical causes of subjective experience, no matter how detailed, simply cannot transform into a direct subjective experience or a form of knowledge permitting one to know that experience. I doubt that we could ever get to that point.
While academia seems fixated on determining whether or not consciousness is physical, perhaps there’s a better way to go. Consider the option of beginning from the premise that systemic causality never fails, simply to preserve the potential integrity of science. Though William of Occam (1287 – 1347) was mainly known for his preference for simple rather than complex explanations, that was probably the result of a more fundamental belief that worldly reason should not be mixed with otherworldly faith. As a friar he once barely escaped with his life from papal soldiers for accusing a sitting pope of blasphemy for trying to mix the two.
In any case I find it amusing how much effort people today expend to refute Jackson’s argument (including himself), when I also realize that these same people often back unfalsifiable and ultimately magical consciousness proposals Once a falsifiable consciousness proposal becomes empirically validated, I expect others to share my mirth about this.
Interesting as always! I like your reference to William of Occam. That's an interesting idea that his preference for simplicity might have stemmed from a deeper conviction about separating natural and supernatural explanations.
I've been wondering about your suggestion that we start with the premise that "systemic causality never fails." I like this, given it makes studying consciousness easier. But I do wonder what someone leaning toward a quantum explanation might say. Quantum mechanics surprised many physicists who held similar assumptions about causality.
Your point about the irony of people rejecting Jackson's argument while embracing unfalsifiable theories of consciousness made me smile. Of course, even if we had one, falsifiability only tells us when we're wrong, never when we're right. We can show that a theory is false, but never truly show it to be true -- we can only gather evidence that supports it.
I figured you might like that little anecdote on Occam Suzi! Actually the person who informed me about this is also my favorite consciousness theorist. Apparently as a Surrey University professor he’d see road signs for the Ockham village that gave this medieval friar his name. Curiosity led him to explore the man’s life and he used this material to write a historical account about how the simplicity theme was crucial to the progression of our hard sciences. Few seem to grasp that William’s mandate actually stemmed from his belief that mixing godliness with worldliness is erroneous. I used to let the audio kindle version of McFadden’s book take me off to sleep until I’d heard it all too many times. I consider the history of hard science to be relevant to the present state of soft science.
On quantum mechanics surprising physicists, definitely! And even today I don’t think physicists feel all that comfortable. Some notable physicists even attempt to escape its contradictions by positing that bajillions of actual worlds emerge from ours each moment, and presumably with bajillions emerging from each of them each moment as well, and so on for infinite regress. Thus even modern hard scientists seem to take spooky steps when sufficiently challenged. I do think I can provide one solid observation on the matter however. Either systemic causality never fails regarding QM, and so remains perfectly determinate even given that the “how” of this isn’t understood, or science is rendered obsolete in this sense because reality functions magically here. I stand with Occam and Einstein against magical interpretations.
Of course consciousness itself does tend to seem quite magical to us and so it isn’t surprising that some would associate it with the also magical seeming quantum mechanics. To me however this conjunction seems inordinately convenient. In the 90s McFadden wondered if QM might help explain various strange elements of biology, and since that time has helped found a now reasonably established field of quantum biology. With his first published book on the matter in 1999 he even planed to include the Penrose-Hameroff quantum consciousness proposal as the final chapter. After reading their book however he decided that it had too many problems. This is what got him thinking that consciousness might exist electromagnetically, and he did end up concluding that quantum biology book with his own non quantum EMF consciousness proposal.
On falsifiability, I think many consciousness theorists today inadvertently rely upon the incompleteness and messiness of what they propose as a way to stay in a game that’s almost entirely composed of incomplete and messy ideas. Here politics trumps evidence since evidence doesn’t apply. We can either be charitable about this today and call it “early science”, or not and say it’s “pseudo-science”. In hindsight I doubt that history will be kind.
The difference between standard consciousness proposals today and the electromagnetic proposal that I back, is not just falsifiability, but verifiability. Experts ought to be able to determine any truth to this particular theory just as they do under hard forms of science. In our last discussion regarding my proposed means of EMF consciousness testing, this is where we left off:
There's a hidden assumption here: red is a primary colour cos we have red-sensitive rods in the retina. But it goes thru a lot of processing before we perceive it and we don't actually perceive red as more fundamental than (eg) yellow, brown or even orange.
Yellow and brown are the same colour with different luminance. Why do we regard those as different but not make the same distinction for pink (=pale red) and blood (=dark red) or sky blue and cobalt? Nobody would say brown=dark yellow. So there's a learning and/or language layer.
I wonder if we could create a new colour simply by giving it a name? (I think we did this historically with orange, which is why we still say red hair not orange hair)
Oh! This is fascinating. You've hit on some really important points about colour perception! Brown and yellow are the same wavelength at different luminance levels. But we treat them as totally different colours. And yet, as you point out, we don't treat all brightness variations this way. As you say, our culture and language likely have a lot to do with how we categorize colours.
Your question about creating new colours simply by naming them is interesting. I've often wondered about this. I'm a tetrachromat -- someone with four cone types instead of the usual three -- so I can distinguish between more colours in the blue/green range than most people. But I don't feel like I see a completely new colour. Orange, yep! I think that is a different category from red and yellow (although I understand it is related to red and yellow). But aqua -- I don't think about aqua the same way as I think about orange. Aqua is just a blue-green. I find this strange. I even have a word for mid blue-green but I don't see it as a separate category.
It's something I've been interested in for decades.
I thought female tetrachromats had an extra red? (I see there's an extra green as well now)
How do you know this? Was it a DNA test or a visual test?
Of course having different variants of rhodopsin gets you nowhere unless your brain interprets them differently, and this must be novel wiring so it must be learnt: seeing it and SEEING IT are different so you need to train your discrimination. So if you can learn it, that suggests that at least some other colours are probably also learnt.
Do you find that printing or colour screens lack some colours?
re brown: of course the other possibility is that it's hard-wired.
Evolutionarily we needed red to spot our own and prey respiratory pigment, green for foliage, blue for water. There's plenty of brown in wild habitats too. Yellow and orange for fruit, but pink, sky blue, cobalt, cyan, purple etc. are largely ornamental (AIUI, purple fruits are primarily Australasian so not on our evolutionary pathway).
To be honest, I'm not 100% sure. But I seem to see things differently to my friends.
The first time I suspected, I was working in a vision lab. I was writing some code, trying to create a game to use in an experiment about just-noticeable differences. The stimulus needed to change just enough so that the difference was noticeable on only half the trials. I was trying to do this using colour on a special light display that can produce very specific wavelengths of light (more than a typical monitor can produce). I was using wavelength values given to me by another research from a past experiment. But I couldn't get it to work -- I kept noticing the difference between colours -- every time.
So, I asked a friend for some help. He just said "how are you doing that?" We guessed that I might be a tetrachromat. I haven't confirmed with DNA testing, so I'm not 100% sure.
But I notice that I see things differently to my friends. Some digital photos never look like the real thing. For me this is especially true for pictures with the ocean, rainbows, or sunsets. Ocean waves have a sunlight glow that doesn't show up in pictures and pictures of sunsets always look dull -- they never have that teal band between the yellow and the blue -- it's usually washed out like it's almost pale pink.
You've got me thinking though, I've always thought it was that I saw more colours in the blue/green range, but now I'm not so sure that's true. Maybe it's more blue/yellow?
> so I can distinguish between more colours in the blue/green range than most people. But I don't feel like I see a completely new colour.
If you have two variants of green but don't distinguish between them that should reduce your number of greens, not increase it, so you must "know" which green cone is which.
Your two blues will each combine with red and green to give different magentas and cyans. Yellow should be unaffected.
The other possibility is just training: if you've worked in demanding colour environments, maybe you've just learnt to use your 3 types of cones more effectively. Both sRGB of standard monitors and CMYK for printing are quite restricted (esp. in green-blue) compared to the whole visible spectrum, or even a wide gamut screen like you were using.
Many children enter kindergarten (typically age 5-6) already knowing basic color words, but not all do.
Prior preschool experience, home environment and exposure to color-focused activities, individual development pace, primary language(s) spoken at home, differences in color word teaching.
Educational research suggests that most children can reliably identify and name basic colors (red, blue, yellow, green) by age 3-4, but mastery of more complex colors (like burgundy, turquoise, etc.) develops later. Some kindergartners may still be working on consistently identifying even basic colors.
In an undergrad course in General Semantics, the prof passed around several color chips and asked us to write down the color word we use for each. The variance was stunning. I’ll never forget it—almost an out of body experience:)
What does it feel like to be a bat? It may depend a little on which bat. To know everything about colors is to learn the words your language(s) use to parse out the spectrum. As the sensation works its way through the midbrain the wavelengths get routed to the amygdala as well (is this right, Suzi?) and therefore emotionally laden—something learned from culture.
I keep seeing a philosophical problem arise when the argument fails to account for the relationship between language and ontology. Not to go all Sapir-Wharf on this, but can we ever separate consciousness from language? And since language is cultural, can we ever separate culture from sensation and perception?
On the first day Mary walked out of her house, she may have experienced red for the first time, but red would soon be wrapped in geography, language, religion, aesthetics—what does it mean to know your colors?
Suzi, I’m trying to learn the terms associated with all parts of the brain (hence, my provisional use of the word amygdala). So keep that in mind:) if you have any suggestions to help me master this task, I’m game for anything.
Thanks for these examples about colour learning and language! Developmental psychology and learning are fascinating. I noticed even my own children differed in the speed at which they learned their colours. One was fascinated with art -- wanted to own every different coloured pencil that the art store supplied, and the other was happy simply knowing the seven colours in the rainbow song.
I love the colour chip exercise (I might steal that one day). It's a great way to show how personal and variable our experiences (at least with colour naming) can be.
Interesting question about the amygdala! Yep -- you're right! The amygdala is involved in 'attaching' emotional significance to sensory stimuli, including colours. For example, culturally or individually significant colours (e.g., red for danger or romance, green for calmness) may elicit emotional responses mediated by the amygdala. But these emotional associations have a lot to do with learning and memory circuits interacting with the amygdala and visual pathways.
I'm really intrigued by your points about language, culture and perception. It reminds me of those findings that different languages carve up the colour spectrum differently.
As for learning brain anatomy -- one of my favourite resources is the Brain Atlas by Neurotorium. https://neurotorium.org/tool/brain-atlas/. It's a 3D interactive brain that allows you to zoom in and spin to see all the parts of the brain.
Good overview! I think the Ability Argument is right but see it only as Mary gaining experiential knowledge (aka facts). Exactly as your neuroscience analysis says, Mary cannot gain the necessary brain states without actually seeing color. And, of course, none of which refutes physicalism.
It seems to me that New Knowledge vs a New Ability is a distinction without a significant difference. I’d like to be open-minded about this but knowing THAT I can recognize a chess board as representing the game of chess and knowing HOW to play the game are—in my feeble brain—two different types of knowledge not differences in kind that makes the latter not a form of knowledge at all.
Oh, I thought I was disagreeing with you—wasn’t I? I think the best criticism of the Jackson mind experiment is that final leap, that therefore physicalism is false. That is an unjustified leap IMO. We can exist in one physical world and still Mary gains new knowledge.
upon a careful re-reading of your initial comment—nothing my friend. I guess I was itching for a fight. And, yes, nice to see you here too. There are, I’ve noticed in my recent forays into philosophy of mind, many versions of physicalism. I see no problem with being happy with one physical world (which I am) yet also seeing that complex physical phenomena create new ontologies that cannot be reduced. I’ve said that quite clumsily but I think you may get my point.
Get it and agree. I'm one of those who differentiate between materialism and physicalism, the difference being — exactly as you say — that emergence involves new high-level laws. As I usually state it, under physicalism, reduction doesn't always work.
My one qualm with the Mary's Room experiment is how it reinforces the idea that perception is what is foundational to consciousness. One small tweak makes this obvious...
Mary has spent her entire life trapped in a room that perfectly meets every homeostatic need. But despite this limitation, she has learned all the physical information there is to know about FEAR, from the hormonal interactions to the precise neural activity patterns it triggers.
What will happen when Mary is released from her perfectly safe room and is threatened for her life? Will she learn anything about FEAR or not?
Everything about color can be captured and represented via a smartphone. But a smartphone can capture nothing of fear. Feelings have no corresponding informational content beyond that felt experience.
Which should tell us something about the primacy of feelings in our conscious experience, no?
Yes agreed R.B. Perceptions like color should not be foundational to consciousness if our non-conscious computers can in a sense “perceive” input information that we might just as well call color. So then what might be foundational to consciousness? I think changing the scenario over to fear was an effective move. This brings an inherent dread or personal badness regarding existence. And of course you might have presented a scenario where Marry knew all there was to know about the taste of good food, except had somehow been prevented from ever having such experiences until her handlers permitted it. Thus she’d now be rewarded in that way. I consider the felt goodness to badness of personally existing to not just be foundational for consciousness, but to fully constitute the value of existing for anything, anywhere. Back when algorithms weren’t able to deal with more open circumstances very well, I suspect that this goodness to badness of existing evolved essentially as fuel from which to motivate an agency form of function.
I chose fear, but you could pick any of the primal feelings that we share with all animals with a similar brainstem: rage, care, lust, panic, etc.
And I agree with you in that every feeling is valenced: experienced as good or bad, to the degree that it deviates from or returns to any of our various homeostatic needs. But valence is essentially a quantity, which seems necessary but not sufficient to me as a foundation. The qualitative difference between feelings is also necessary to prioritize action in the face of uncertainty. This is what justifies the enormous energy costs of consciously processing feelings.
This engaging read touches on the question of whether we are merely organic machines, or something more. As Robert Sapolsky notes in "Determined" (in which he argues that free will is an illusion) no one has been able to put their finger on where consciousness -- the thing that makes choices -- lives in the brain. Others, like Rupert Sheldrake, are open to the idea that consciousness resides elsewhere, perhaps everywhere.
Mary's Room is just a (rather complicated and highly artificial) metaphor for what we all call "new experiences". Could you anticipate the taste of durian by combining knowledge of the chemistry and other people's descriptions? As they say: you have to be there. So it is open to replication but I don't see how you could design a test to answer the question.
I have to be honest - I had to Google what a durian was! For those who haven't encountered it either, it's a Southeast Asian fruit famous for its powerful (and controversial) smell and unique taste - people tend to either love it or hate it (apparently).
You make a great point about connecting this to everyday experiences. You're right - there does seem to be something special about 'being there'. But I've been wondering-- don't we need to 'be there' for everything we learn? At some point, we had to 'be there' to learn that the Earth revolves around the Sun or that water is a compound of hydrogen and oxygen. We've probably forgotten where, when, and who taught us these facts, but we had to 'be there' to learn it, right?
I was wondering the other day to what extent memories are raw or cooked.
Raw = remember the sensory input.
Cooked = remember the processed result.
When I was a child I disliked canned beans (we call them Heinz Beans, dunno what you call them - haricot beans in tomato sauce)
Anyway, as I got older "I learnt to like them" (as my mother used to say) and when I remembered back the taste no longer seemed unpleasant, so I was remembering the raw quale, not the cooked one which would have included the distaste.
If you think of other memories they're cooked to varying degrees: words of a song: if you remember the original artist singing it, it's raw. If you only remember the words and tune, it's cooked.
Most schoolwork is cooked, but there are probably a few raw nuggets.
Oh! This is fascinating. I'm very curious about this idea -- what's the difference between raw and cooked experience/memory? It reminds me of a thought experiment by Daniel Dennett about coffee tasters (which I'm planning on writing about in a couple of weeks).
Thanks for doing this series, Suzi, and getting me to finally read that paper. As you know, I don't think epiphenomenalism has many virtues, but I don't agree with physicalism either (and I find it funny that he Jackson thinks he's crusading against physicalism when he's practically a physicalist in my view). I don't think any of the standard arguments I've heard against Mary's room—the difference between 'knowledge' and 'know how'; that Mary's knowledge of some hypothetical completed science would have to include that of the experience of color—do anything to dispel the problem I'm having with the experiment. Maybe it's just me, but it seems the entire project of color science, of determining the 'objective' nature of color, depends on us having the 'subjective' experience of color in the first place. Of course you can't begin with the science and get to the experience; it's the other way around. I didn't think this was controversial. Or maybe I'm missing the point of the entire controversy, I don't know.
We talked about this once a little while back, although I think we were considering a slightly different sort of set up. Here I'm talking about phenomenally-experienced color, which I take to include black and white. So in other words, in a completely colorless world we wouldn't visually perceive boundaries to objects; we would be blind. Objects could only be known by the other senses. This topic is touched on in Jackson's paper, I just realized, and in it he brings up H.G. Wells' story "The Country of the Blind" which is about "a sighted person in a totally blind community. This person never manages to convince them that he can see, that he has an extra sense. They ridicule this sense as quite inconceivable, and treat his capacity to avoid falling into ditches, to win fights and so on as precisely that capacity and nothing more." Keep in mind the entire community must be incapable of seeing in order for this to work. So if we're all incapable of seeing color, effectively blind, how would we come to detect wavelengths? In other words, what would prevent us from saying of color-detecting creatures that they were reacting to certain textures, perhaps even concocting a whole complicated theory around texture, for instance? Or we might be inclined to say they were merely disposed to behave the way they do. Maybe you can figure out a way to come up with wavelengths from here, but I'm at a loss.
I think I can answer your question but let me first ask about what constraints you have in mind. Do we have our other senses, and do we have our same level of intelligence and curiosity?
FWIW: My reply got long, and I think your attention has turned to your recent post about idealism, so rather than post a long comment here, I made it into a Note.
Sorry, my bad! I wasn't clear in what I was saying above. I agree it seems like we might come up with the idea that heat is some sort of energy, but I meant "color" wavelengths (I realize I just said "wavelengths"...which is confusing.) BTW, sorry it took so long to get back to you. I was inundated for a minute there, and I'm finding I don't always see comments or replies in my email. Not sure what's going on.
Hey Tina, I'm glad you brought up the Wells story; it's a good one.
I might quibble over whether we should think about colour as the reason we see boundaries -- our visual system distinguishes between processing edges and processing colour -- but I feel like that might be missing the point.
The interesting question is, even if we had no visual experience of light (we were blind), would we still discover wavelengths? We already interact with wavelengths we can't see - we feel infrared as heat, some animals detect ultraviolet, and we've developed instruments to measure radio waves and X-rays.
To understand wavelengths, we need to understand spatial dimensions and movement over time. And it seems like we could develop these concepts even if we were blind.
We can feel movement; we can learn to understand space through touch and sound. We often represent waves with visual schematics, but waves like infrared radiation can be represented as sound. Even without being able to see light, we might still discover that there's something that travels in straight lines, reflects off surfaces, and carries energy. We might still figure out that this something has wave-like properties that can be measured and manipulated.
Yes, you're guessing right. I should have made it clear I wasn't thinking of color in any scientific sense, but more in a pre-theoretical sense. Also I see now that I should have made it clear I was thinking of color wavelengths—I omitted 'color' in talking about wavelengths above, which is confusing! I agree we could probably come up with a theory about heat and energy, given that those don't rely on vision. I can't imagine how we'd arrive at the concept of color wavelengths specifically. Sorry about the confusion!
All I meant to point out was the funny relationship our scientific understanding has to our perception, especially in this instance. It seems our science is rooted in our experiences, here, our perception of colors, and these experiences shape our ability to conceptualize and investigate the world around us, and yet at the same time these color experiences are deemed subjective.
Color is a funny one, isn't it!? What we perceive as colour doesn't really exist -- it's just wavelengths of light. But in order to perceive those wavelengths as colour, we need certain physical structures (photoreceptors, V4 neurons, etc.) -- without them we don't see colour at all.
This reminds me of a current debate currently happening in psychology/neuroscience. Much of our understanding of how the brains works started with subjective experience (or psychology). We created psychological models and then looked for their neural correlates. In essence, we've taken an outside-in approach to understanding the brain.
But could we flip this approach? Instead of starting with the psychology (or the subjective experience) and looking for its neural basis, could we start with the brain. For example, could we start with how photoreceptors, neural pathways, and brain circuits process wavelengths of light, and let that understanding inform our concepts? I do wonder whether this inside-out approach is possible.
I am delighted that you get my point and its implications, and so readily too.
Yes, color is a funny one. I can see why Jackson chose to talk about color instead of, say, sound, as science explicitly marks color as being subjective and therefore unreal in some sense. What I find strange about both the argument and the counter-arguments is the assumption that science can somehow be characterized as a collection of "information" and "facts" divorced from anyone ever experiencing anything. It's like everyone wants to render sensory perception completely irrelevant to science. For instance, Jackson says, "I think that there are certain features of the bodily sensations especially, but also of certain perceptual experiences, which no amount of purely physical information includes." If he's saying scientific descriptions exclude subjective experiences, well, duh. But he seems to be saying more than that. Why is no one pointing out that sensory experience is required for science? And that it must be taken as the data for which science is supposed to be accountable in order to be empirical—knowledge derived from sensory perception? Are we moving towards a purely semantic ontology with no empirical constraints? What is this bizarre trend of throwing doubt on the reliability of experience, going so far as to call it an illusion, while at the same time propping up science as "all there is to know". I'm thoroughly confused. Or am I just living in the dark ages while the rest of the world has gone post-empirical? Anyway, thanks, Suzi for understanding. Feel free to take these questions as rhetorical. :)
Due to a congenital toxoplasmosis infection, my son Jakob, who is now 41 years old, has severely damaged retinas, a seizure disorder, and various physical and intellectual impairments. He is a graduate of the legendary Perkins School for the Blind. Despite his various limitations he has until recently been a voracious, compulsive reader — especially of books about science, neuroscience in particular. He reads normal-print using what he calls 'facebook' technique — with the book about 3 inches from his face.
Earlier this this summer the cataract in Jakob's one eye that has useful vision had become so opaque that he could no longer see well enough to read. Two days ago, on November 18, in a long operation under general anesthesia, that cataract was removed. Because of the extreme fragility of the structural parts of his eye, no new lens was inserted. After the eye has had time to heal, Jakob & his medical team will determine whether a second operation should be done to insert a lens, or if he should wear glasses, or if he will just use the vision he has due to the removal of the cataract — which was, according to the surgeon, extremely dense. Having read this essay I'm now wondering how long that cataract has been developing, and how big a jolt his brain is getting from the (presumably) vastly increased supply of visual data to his retina. Over decades of taking Jake to see eye specialists, the main concern his always been his retinas — because toxoplasmosis attacks them differentially. I don't recall much discussion about cataracts before about a year ago, but who knows.
Jake has face-blindness. He recognizes people — including me, and his mother & sisters — by their voices. I had always assumed that his face-blindness came from damage to the portion of the brain that specializes in recognizing faces, but this essay makes me wonder if that deficit was the result of not getting the data needed to develop that ability.
It's too soon to know if Jakob will be able to read this essay on his laptop. If he can't, I will either read it to him or figure out how to have a synthetic voice read it for him. In any event I very much look forward to hearing his comments on it. Through Perkins, Jake knows deafblind people, and people who have had operations of various kinds — such as cochlear implants, or cornea transplants — to restore, or provide for the first time, new kinds sensory input. This is a subject he knows a lot about.
My son may not be a 'Mary,' raised in a black & white world who is now experiencing color for the first time, but I'm sure that he will consider her familiar, the kind of person who were his friends and classmates during his seven years at Perkins.
I recently wrote about some recent adventures Jakob & I have had in the land of toxo. This essay includes some photos of just a few of the hundreds of books that fill his apartment.
This is nicely done. As someone who has been involved in what, by now, must be cumulative years of discussion about interior and exterior decoration and, consequently, colour I am intrigued by this and the concept of colour constancy as well.
Anyway, really looking forward to something I have known about towards the end of the last century (great being able to say that), memory and learning. Thanks again Suzi.
Thanks, John!
I imagine you must have some fascinating stories about colour constancy from all those discussions. Those illusions like the blue and black dress (or is it the white and gold dress) always fascinate me.
Looking forward to exploring memory and learning next week, too! 😊
A thorough analysis of the Mary's Room thought experiment, with some relevant neuroscience thrown in. I particularly like the monkey experiment. I'd heard of others along those lines, but not that particular one.
I've never been concerned that Mary's Room challenges physicalism for exactly your final point. The premise just assumes Mary can have all the physical information without one of the pathways to it, personal experience. If she can't, then her learning anything new has no metaphysical implications.
The reality is none of us have Mary's knowledge, or what it might entail. So, even neuroscience complications aside, any statements about whether she'd be surprised or just thrilled at confirming her hypotheses, is pure guess work. It's a blank slate people can project their intuitions into, like most of these types of thought experiments.
That said, it's worth noting what even a Mary with contemporary knowledge about color and vision would know. She'd know the relations between the colors, that violet blends into blue, cyan, green, lime, yellow, orange, and red. She'd know that reds and yellows are more striking, stand out more, than blues and greens. She'd expect a clear sky and deep water to have similar hues, yet be distinct from the similar hues of the sun, daffodils, and bananas, or that of blood, roses, and ripe strawberries. She'd be able to anticipate the unease of walking into a completely red room.
There's an enormous amount about color experience she'd be able to anticipate. (Assuming of course that she'd be capable of it immediately, or ever.)
On information, I actually think of information as causation, or maybe a snapshot of causal processing. Which raises the question, if Mary does learn new information, what effects can it have, on her behavior, ability to write about it, etc? If none, then it's not "information" in the sense we normally mean with that word. I don't think a Mary with the profound knowledge in the premise would learn any new facts in any reasonable meaning of the word "fact" or "information". There'd be no update to her world model. But since I don't have her knowledge, I'll admit that's my intuition.
Excellent as always Suzi!
Thanks Mike! As usual, you've given us lots to think about here.
I'm curious about your framing of information as causation. This makes me wonder about cases where we have information that doesn't seem to directly cause behavioural changes. Like when we know the sun is actually white but we continue to see it as yellow. The information doesn't seem to have the same causation as other types of information. Would you consider that information of a different kind? Just curious what you think.
Hi Suzi,
Good question. What I'd say is that learning the sun is white changed the state of your brain, so that you could later use it as an example here. :-) Information doesn't have to result in immediate behavior, just have an effect of some type.
I often think of it in terms of logic gates. One signal to an AND gate results in no further propagation of effects. But if that signal arrives in tandem with another, we get a propagated effect. And it's actually not true that the lone signal had no effect, just one that resulted in waste heat.
Where this view does get under some pressure is in a isolated system in a high entropy state. There's arguably a lot of hidden information in such a system, but without outside energy, little to nothing is happening. (This is why I added the weaselly qualifier "or a snapshot of causal process" above.) But again, if energy is ever added to that isolated system, it's hidden information would get its chance to have effects.
I should admit that I got the notion of information as causation from an offhand remark Eric Schwitzgebel once made on his blog in a discussion with Eric Borg. It summed up a relation I'd been wrestling with for several years. And fits with Gregory Bateson's idea of information as a difference that makes a difference.
That said, I'm always on the lookout for anything that actually dissociates the concepts of causation and information. I'll admit it seems a little too neat. But so far I haven't seen any.
It's also worth noting that this fits for physical information, information as physicists see it. Semantic information, information that provides actual knowledge to an agent, seems like a subset and so is a narrower concept.
Thank you, Suzi. (As always.) I think you've given me (and all of us) much to think on there.
One thing I did think about while reading this was "what is 'normal' colour vision, anyway?" in both objective and subjective terms.
A couple of anecdotal data points:
When I was being tested to join the military, I aced all the "colour blindness" tests. Scoring well above-average on the tests designed to test fine gradations of colour discrimination. Some of my friends (including some who passed civilian colour-blindness tests for civil aviation) 'failed' those tests - although, functionally, they felt themselves far from a naive understanding of what "colour blind" might mean.
Another: I drive what I've always regarded as a 'blue' car. (It's a 40yo car in dark metallic blue paint.) Even the registration papers for my car describe it as a "1983 Blue Toyota Celica Supra Sedan". Yet: many of my friends say my car is "grey" not "blue". Go figure. Further: when they see my car in photographs (even photos they've taken themselves) they see it as "blue" in the photos - but not "in real life".
I'm not sure what's going on in that 2nd example. Is it a definitional problem? I don't think so - when they see it via an 'objective' instrument (like the camera in their phone) they can see the 'blueness' - yet they seem unable to see it with their own eyes - while I can and always have. As have others. Which makes me think there's some kind of real perceptual difference happening. But what kind?
I don't have any theories or answers about that. Only questions.
cameras are often set to increase saturation to make photos look more exciting so they amplify colours
Good point! Thanks, Malcolm
Thank you, Mike!
Your military example reminds me of something I read recently. People who are colour blind actually detect camouflage better than those with normal colour vision. They are able to discriminate between shades of grey better which helps them notice subtle variations in brightness. So the military may actually benefit from having colour blind people.
As for your car example, I wonder if this is related to colour constancy.
Do you remember the black-and-blue dress (or was it a white-and-gold dress)? Depending on whether we assume the dress is in a natural lighting or under artificial lighting, affects how we perceive the colour of the dress.
In normal circumstances, our brain maintains colour constancy - keeping objects looking the same colour despite changes in lighting. But as Malcolm alludes to, photos flatten this process -- the camera has already done some of that work for us. It's a bit like how we might not notice a blue tint from fluorescent lights until we take a photo and suddenly everything looks obviously blue.
But we know people's brains handle colour constancy different (that's why half the population thinks the dress is blue and black and the other half think it's white and gold). There are genuine differences in how we perceive the same object.
Colour constancy might be part of it (I’ve spent way too much of my life on - strictly amateur - colour-managed workflow in photography, so I know exactly what you’re talking about there).
But I suspect Malcolm might have been closer when he talked about colour saturation. It might well be that the people I was talking about have a higher (than me) saturation threshold that has to be met in the blue part of the spectrum before they see a blue tint. And that camera-phones tend to boost saturation to make “more attractive” photos.
Modern cameras and camera-phones are enormously better at AWB (“automatic white balance”) when colour temperature (eg. direct sunlight vs shadowed light) or tint of light source varies (especially so under artificial light, these days). So I suspect that saturation levels might be more important here than colour casts.
It might also be that I’ve spent so much time trying to get colour-casts ‘just right’ between what’s on-screen vs a photographic print that I’m just more practiced at colour discrimination. (I'm also more prone than most, I think, to having the 'reference white' that guides my colour perception be context-dependent. I have to be aware of that and compensate for it when editing photos.)
It’s hard to know, because all of the above (and likely a good deal more) feed into differences in colour perception. But I am sure those differences exist.
…Mike
Oh, this adds some interesting context, Mike!
Your experience with colour likely has had a big effect.
I've spent several years in a lab investigating human vision, primarily with colour. Part of this work was coding colour games for participants to play. This means I've spent many hours testing these games, which involved many, many hours looking at and trying to distinguish between different colours. I've often had the impression that I've got better at it over time.
I wonder whether something similar has happened with your ability to have a context-dependent reference white. It makes sense that it would be the case.
Interesting as they are, I don't think the developmental aspects add much to the intuition pump. It's about the difference between the perceived world and the lived experience. You could equally use any optical illusion, or the cultural ability to recognize vertical lines and right angles.
While we're on the subject:
https://gondwana-collection.com/blog/how-do-namibian-himbas-see-colour - sorry that's based on a BBC fake, try this:
https://www.sciencealert.com/humans-didn-t-even-see-the-colour-blue-until-modern-times-evidence-suggests
Finally: a link that links to published science:
https://www.newscientist.com/article/dn11759-russian-speakers-get-the-blues/
The history of the colour orange is also provocative.
Hi Malcolm! Thanks for your interesting comment. I'm a little confused though - you mention that the developmental aspects don't add much, but then say Mary's Room is about the difference between the perceived world and lived experience. I'm having trouble seeing how the development aspects could be separate from lived experience and the perceived world. What am I missing here?
For me, Mary's Room is about the ineffability of qualia. It uses "red" as an example, but red is just an example. To what extent qualia are learnt, arise automatically as a result of physical development or have "always" been there as part of consciousness, is a supplementary question which needs to be addressed individually for each quale.
I see it now: you read Mary's Room as an exploration of the development of qualia. I'm not sure this works cos it's so artificial (and why was there so much emphasis on Mary being an all-knowing scientist when this requires only naivety?), tho it does serve to stimulate discussion.
Copilot says: It is designed to challenge physicalism, the view that everything about the human mind can be explained by physical processes alone.
Or even that "everything about the human mind can be explained", ie not ineffable.
I'd prefer to say that this universe is (obviously) awareness-friendly, but nobody can say whether that's a specific unknown property of space/time (dualism), or an emergent property of everything else (physicalism).
Another great one. The question that pops into my mind is whether Mary, knowing everything, knows how to teach herself to see colors and what teaching colors would involve. And then, of course, how to teach those poor monkeys how to see color - I'm not so comfortable drawing the ethical distinction between depriving humans and monkeys in the way most experimental scientists seem to be. I guess another approach would be to say, if you were trying to set up a computer simulation of the test so as to avoid the ethical concerns, what exactly would you try to encode the computer with?
The idea of Mary teaching herself to see colours is intriguing. Even with all her knowledge, would she know how to guide her own brain through the development process that normally happens with experiencing colour? I doubt it. It's a bit like knowing everything there is to know about how to play tennis, practicing the swinging motions, the standing positions, how to hold a racket, but never actually playing a game of tennis.
Your question about computer simulation is the key question, isn't it!? Could we do it -- could we create a computer simulation that perceives colour? What is missing if we can't?
I'm looking at all the questions you face on this one, all the comments... the price of being so fascinating and good at what you're doing is very high!
Thanks so much, Jack! You're right, the questions and comments are wonderfully thought-provoking, and I wish I had time to explore every fascinating thread. It's actually a lovely problem to have -- being part of such an engaged community of curious minds. I'm truly grateful for how thoughtful and generous everyone is with their insights and questions.
I have to forgive you for ignoring what I thought were some pretty good follow-up questions on the last one, where I asked if you knew anything about psychoacoustics and what that might do to the blind subject attempting to echolocate! I think if I had to pick the very best example of what Substack could be at it's very best, it would be your posts.
I'm so sorry, Jack.
I did read your questions about psychoacoustics and echolocation -- they are fascinating questions and deserved a proper response. I've been caught up in work commitments this week, but I'm planning on getting to it (and some others) on the weekend when I can give it the attention it deserves.
And thank you for your incredibly kind words about the newsletter - they mean more than I can say!
This is quite a useful discussion. My answer to the Mary's room argument has always been that it just shows that an explanation is different from what it explains. This intuition is a combination of the three objections that you lay out. I've recently written about a recent grand unified theory of consciousness that might interest you and your readers: https://open.substack.com/pub/eclecticinquiries/p/a-grand-unified-theory-of-consciousness?r=4952v2&utm_campaign=post&utm_medium=web
Thanks, Ian!
It seems like you missed the point of the Jackson thought experiment: “Mary’s Room was intended to expose what science can’t explain. But maybe the more interesting thing it shows is how complex learning and knowledge is.” More interesting to you I suppose but an irrelevant cul-de-sac. This seems common among those who cling to a simple physicalism as foundational. I can make a similar criticism of EINSTEIN’S thought experiment of riding on a beam of light—holy cow! Can’t be physically done! He’d be ripped to shreds! Instead of color try doing this criticism with Mary getting a dose of poison ivy or learning she is allergic to peanuts. Does not take years an experience to understand that new experience.
Hi Matti!
I think we might have some common ground here. You mentioned in another comment to @wyrdsmythe you mentioned that "the best criticism of the Jackson mind experiment is that final leap, that therefore physicalism is false." I completely agree -- that leap isn't justified. The scientific evidence I presented here wasn't meant to prove or disprove physicalism. Rather, I think it simply suggests that Jackson's argument might not work as early Jackson intended.
Your Einstein comparison is interesting. Einstein's thought experiment about riding a light beam was primarily exploring logical and mathematical relationships -- to me, whether he'd be "ripped to shreds" doesn't affect the underlying mathematical insights he was pursuing. But with Mary's Room, the physical structures of the brain are central to the question being asked: would Mary learn anything new? If learning itself involves physical changes in the brain (as neuroscience suggests), then the thought experiment's premises might be internally inconsistent.
You raise an interesting point about poison ivy and peanut allergies. These immediate physical reactions are quite different from complex perceptual learning. But I wonder -- even in these cases, wouldn't the brain need certain physical structures to process and recognise these new sensations?
My point about Einstein was that thought experiments are, in many cases, wildly hypothetical. There is an assumption in such hypotheticals that all other circumstances about reality are held in place. Obviously Jackson’s and Einstein’s thought experiments could not be conducted in reality. Thus, the critique by some on this basis is missing the point. And that was my point. You further say that even in the examples I suggested that “wouldn't the brain need certain physical structures to process and recognize these new sensations?” Well yes, of course. I’m sorry, I fail to see your point here. I was merely attempting to provide a work around to the silly argument that the experiment could not be conducted in reality—which because it is wildly hypothetical anyway should not be necessary.
By the way, I also think argument that because Mary knows everything—implying some sort of omniscience that no one has—that she will recognize the color red. That is laughable. It’s similar to medieval arguments where one whistles up God to fill in what can’t be explained.
Thanks, Matti. I appreciate your push back.
Let me see if I understand your argument correctly:
-- Thought experiments are meant to be hypothetical and don't need to be physically possible
-- My critique about brain development misses the point because, like Einstein's light beam example, we're meant to assume other aspects of reality are held constant
-- The fact that we need physical brain structures to process sensations is beside the point - that's just one of those aspects we're meant to hold constant in the thought experiment
1.) Yes, thought experiments are hypothetical. And many are not possible to duplicate physically. Some obvious examples are Hilary Putnam’s Brain in a vat, John Searles’ Chinese room, and Phillipa Foot’s Trolley problem among others.
2.) You’re asking if needing brain structures to process sensations is one of those aspects we are meant to hold constant in Jackson’s hypothetical. The implication of your question is puzzling. I suspect I have not made myself clear or I’m slow to understand where you are going with that fact. One can accept the fact that we need a physical brain in order to have a functioning mind (which I certainly do) and still see an interesting philosophical challenge in Jackson’s thought experiment. What I mean is having a brain-dependent mind is not a refutation of the thought experiment if that is where you are intending to go with that. Otherwise, yes, one needs “brain structures” and other complex physical activities in a brain “to process sensations.”
I previously alluded to a form of knowledge that keeps popping in debates over this thought experiment. I think that needs repetition. Mary has so-called “complete knowledge” of the functioning of the brain according to the rules of the experiment. This is often implied as “omniscient” knowledge. As I said previously, that interpretation is similar to medieval arguments where one whistles up God to fill in what can’t be explained. It is an expression of faith and nothing more—faith that complete knowledge (when and if it is attained in an imagined future) will justify a materialist interpretation of reality. It’s an interpretation of the ground rules of the thought experiment that—through circular logic—guarantees only one conclusion. It’s a unpersuasive argument.
I assume such mental gymnastics are inspired by concerns that Jackson’s thought experiment threatens a firmly held belief about reality. To maintain a belief in a certain form of physicalism one cannot fully acknowledge that Mary has a truly irreducible subjective experience. To do so, it is feared, opens the door to a nefarious dualism—or worse. I don’t think it does.
Quite simply it seems obvious that a causal explanation is not an experience. At present, objective explanations of the physical causes of subjective experience, no matter how detailed, simply cannot transform into a direct subjective experience or a form of knowledge permitting one to know that experience. I doubt that we could ever get to that point.
While academia seems fixated on determining whether or not consciousness is physical, perhaps there’s a better way to go. Consider the option of beginning from the premise that systemic causality never fails, simply to preserve the potential integrity of science. Though William of Occam (1287 – 1347) was mainly known for his preference for simple rather than complex explanations, that was probably the result of a more fundamental belief that worldly reason should not be mixed with otherworldly faith. As a friar he once barely escaped with his life from papal soldiers for accusing a sitting pope of blasphemy for trying to mix the two.
https://www.goodreads.com/book/show/56643989-life-is-simple
In any case I find it amusing how much effort people today expend to refute Jackson’s argument (including himself), when I also realize that these same people often back unfalsifiable and ultimately magical consciousness proposals Once a falsifiable consciousness proposal becomes empirically validated, I expect others to share my mirth about this.
Hey Eric!
Interesting as always! I like your reference to William of Occam. That's an interesting idea that his preference for simplicity might have stemmed from a deeper conviction about separating natural and supernatural explanations.
I've been wondering about your suggestion that we start with the premise that "systemic causality never fails." I like this, given it makes studying consciousness easier. But I do wonder what someone leaning toward a quantum explanation might say. Quantum mechanics surprised many physicists who held similar assumptions about causality.
Your point about the irony of people rejecting Jackson's argument while embracing unfalsifiable theories of consciousness made me smile. Of course, even if we had one, falsifiability only tells us when we're wrong, never when we're right. We can show that a theory is false, but never truly show it to be true -- we can only gather evidence that supports it.
I figured you might like that little anecdote on Occam Suzi! Actually the person who informed me about this is also my favorite consciousness theorist. Apparently as a Surrey University professor he’d see road signs for the Ockham village that gave this medieval friar his name. Curiosity led him to explore the man’s life and he used this material to write a historical account about how the simplicity theme was crucial to the progression of our hard sciences. Few seem to grasp that William’s mandate actually stemmed from his belief that mixing godliness with worldliness is erroneous. I used to let the audio kindle version of McFadden’s book take me off to sleep until I’d heard it all too many times. I consider the history of hard science to be relevant to the present state of soft science.
On quantum mechanics surprising physicists, definitely! And even today I don’t think physicists feel all that comfortable. Some notable physicists even attempt to escape its contradictions by positing that bajillions of actual worlds emerge from ours each moment, and presumably with bajillions emerging from each of them each moment as well, and so on for infinite regress. Thus even modern hard scientists seem to take spooky steps when sufficiently challenged. I do think I can provide one solid observation on the matter however. Either systemic causality never fails regarding QM, and so remains perfectly determinate even given that the “how” of this isn’t understood, or science is rendered obsolete in this sense because reality functions magically here. I stand with Occam and Einstein against magical interpretations.
Of course consciousness itself does tend to seem quite magical to us and so it isn’t surprising that some would associate it with the also magical seeming quantum mechanics. To me however this conjunction seems inordinately convenient. In the 90s McFadden wondered if QM might help explain various strange elements of biology, and since that time has helped found a now reasonably established field of quantum biology. With his first published book on the matter in 1999 he even planed to include the Penrose-Hameroff quantum consciousness proposal as the final chapter. After reading their book however he decided that it had too many problems. This is what got him thinking that consciousness might exist electromagnetically, and he did end up concluding that quantum biology book with his own non quantum EMF consciousness proposal.
On falsifiability, I think many consciousness theorists today inadvertently rely upon the incompleteness and messiness of what they propose as a way to stay in a game that’s almost entirely composed of incomplete and messy ideas. Here politics trumps evidence since evidence doesn’t apply. We can either be charitable about this today and call it “early science”, or not and say it’s “pseudo-science”. In hindsight I doubt that history will be kind.
The difference between standard consciousness proposals today and the electromagnetic proposal that I back, is not just falsifiability, but verifiability. Experts ought to be able to determine any truth to this particular theory just as they do under hard forms of science. In our last discussion regarding my proposed means of EMF consciousness testing, this is where we left off:
https://suzitravis.substack.com/p/nagel-misunderstood/comment/77558139
There's a hidden assumption here: red is a primary colour cos we have red-sensitive rods in the retina. But it goes thru a lot of processing before we perceive it and we don't actually perceive red as more fundamental than (eg) yellow, brown or even orange.
Yellow and brown are the same colour with different luminance. Why do we regard those as different but not make the same distinction for pink (=pale red) and blood (=dark red) or sky blue and cobalt? Nobody would say brown=dark yellow. So there's a learning and/or language layer.
I wonder if we could create a new colour simply by giving it a name? (I think we did this historically with orange, which is why we still say red hair not orange hair)
Hey Malcolm!
Oh! This is fascinating. You've hit on some really important points about colour perception! Brown and yellow are the same wavelength at different luminance levels. But we treat them as totally different colours. And yet, as you point out, we don't treat all brightness variations this way. As you say, our culture and language likely have a lot to do with how we categorize colours.
Your question about creating new colours simply by naming them is interesting. I've often wondered about this. I'm a tetrachromat -- someone with four cone types instead of the usual three -- so I can distinguish between more colours in the blue/green range than most people. But I don't feel like I see a completely new colour. Orange, yep! I think that is a different category from red and yellow (although I understand it is related to red and yellow). But aqua -- I don't think about aqua the same way as I think about orange. Aqua is just a blue-green. I find this strange. I even have a word for mid blue-green but I don't see it as a separate category.
It's something I've been interested in for decades.
I thought female tetrachromats had an extra red? (I see there's an extra green as well now)
How do you know this? Was it a DNA test or a visual test?
Of course having different variants of rhodopsin gets you nowhere unless your brain interprets them differently, and this must be novel wiring so it must be learnt: seeing it and SEEING IT are different so you need to train your discrimination. So if you can learn it, that suggests that at least some other colours are probably also learnt.
Do you find that printing or colour screens lack some colours?
re brown: of course the other possibility is that it's hard-wired.
Evolutionarily we needed red to spot our own and prey respiratory pigment, green for foliage, blue for water. There's plenty of brown in wild habitats too. Yellow and orange for fruit, but pink, sky blue, cobalt, cyan, purple etc. are largely ornamental (AIUI, purple fruits are primarily Australasian so not on our evolutionary pathway).
To be honest, I'm not 100% sure. But I seem to see things differently to my friends.
The first time I suspected, I was working in a vision lab. I was writing some code, trying to create a game to use in an experiment about just-noticeable differences. The stimulus needed to change just enough so that the difference was noticeable on only half the trials. I was trying to do this using colour on a special light display that can produce very specific wavelengths of light (more than a typical monitor can produce). I was using wavelength values given to me by another research from a past experiment. But I couldn't get it to work -- I kept noticing the difference between colours -- every time.
So, I asked a friend for some help. He just said "how are you doing that?" We guessed that I might be a tetrachromat. I haven't confirmed with DNA testing, so I'm not 100% sure.
But I notice that I see things differently to my friends. Some digital photos never look like the real thing. For me this is especially true for pictures with the ocean, rainbows, or sunsets. Ocean waves have a sunlight glow that doesn't show up in pictures and pictures of sunsets always look dull -- they never have that teal band between the yellow and the blue -- it's usually washed out like it's almost pale pink.
You've got me thinking though, I've always thought it was that I saw more colours in the blue/green range, but now I'm not so sure that's true. Maybe it's more blue/yellow?
> so I can distinguish between more colours in the blue/green range than most people. But I don't feel like I see a completely new colour.
If you have two variants of green but don't distinguish between them that should reduce your number of greens, not increase it, so you must "know" which green cone is which.
Hmmm... yes, you're right. It seems I've made a few assumptions about what's going on. That's embarrassing! 😅
Your two blues will each combine with red and green to give different magentas and cyans. Yellow should be unaffected.
The other possibility is just training: if you've worked in demanding colour environments, maybe you've just learnt to use your 3 types of cones more effectively. Both sRGB of standard monitors and CMYK for printing are quite restricted (esp. in green-blue) compared to the whole visible spectrum, or even a wide gamut screen like you were using.
Many children enter kindergarten (typically age 5-6) already knowing basic color words, but not all do.
Prior preschool experience, home environment and exposure to color-focused activities, individual development pace, primary language(s) spoken at home, differences in color word teaching.
Educational research suggests that most children can reliably identify and name basic colors (red, blue, yellow, green) by age 3-4, but mastery of more complex colors (like burgundy, turquoise, etc.) develops later. Some kindergartners may still be working on consistently identifying even basic colors.
In an undergrad course in General Semantics, the prof passed around several color chips and asked us to write down the color word we use for each. The variance was stunning. I’ll never forget it—almost an out of body experience:)
What does it feel like to be a bat? It may depend a little on which bat. To know everything about colors is to learn the words your language(s) use to parse out the spectrum. As the sensation works its way through the midbrain the wavelengths get routed to the amygdala as well (is this right, Suzi?) and therefore emotionally laden—something learned from culture.
I keep seeing a philosophical problem arise when the argument fails to account for the relationship between language and ontology. Not to go all Sapir-Wharf on this, but can we ever separate consciousness from language? And since language is cultural, can we ever separate culture from sensation and perception?
On the first day Mary walked out of her house, she may have experienced red for the first time, but red would soon be wrapped in geography, language, religion, aesthetics—what does it mean to know your colors?
Suzi, I’m trying to learn the terms associated with all parts of the brain (hence, my provisional use of the word amygdala). So keep that in mind:) if you have any suggestions to help me master this task, I’m game for anything.
Hi, Terry (sorry for the slow reply).
Thanks for these examples about colour learning and language! Developmental psychology and learning are fascinating. I noticed even my own children differed in the speed at which they learned their colours. One was fascinated with art -- wanted to own every different coloured pencil that the art store supplied, and the other was happy simply knowing the seven colours in the rainbow song.
I love the colour chip exercise (I might steal that one day). It's a great way to show how personal and variable our experiences (at least with colour naming) can be.
Interesting question about the amygdala! Yep -- you're right! The amygdala is involved in 'attaching' emotional significance to sensory stimuli, including colours. For example, culturally or individually significant colours (e.g., red for danger or romance, green for calmness) may elicit emotional responses mediated by the amygdala. But these emotional associations have a lot to do with learning and memory circuits interacting with the amygdala and visual pathways.
I'm really intrigued by your points about language, culture and perception. It reminds me of those findings that different languages carve up the colour spectrum differently.
As for learning brain anatomy -- one of my favourite resources is the Brain Atlas by Neurotorium. https://neurotorium.org/tool/brain-atlas/. It's a 3D interactive brain that allows you to zoom in and spin to see all the parts of the brain.
Good overview! I think the Ability Argument is right but see it only as Mary gaining experiential knowledge (aka facts). Exactly as your neuroscience analysis says, Mary cannot gain the necessary brain states without actually seeing color. And, of course, none of which refutes physicalism.
It seems to me that New Knowledge vs a New Ability is a distinction without a significant difference. I’d like to be open-minded about this but knowing THAT I can recognize a chess board as representing the game of chess and knowing HOW to play the game are—in my feeble brain—two different types of knowledge not differences in kind that makes the latter not a form of knowledge at all.
Exactly so. (Hi, Matti. Nice to see you on Substack.)
Oh, I thought I was disagreeing with you—wasn’t I? I think the best criticism of the Jackson mind experiment is that final leap, that therefore physicalism is false. That is an unjustified leap IMO. We can exist in one physical world and still Mary gains new knowledge.
What is the perceived disagreement?
upon a careful re-reading of your initial comment—nothing my friend. I guess I was itching for a fight. And, yes, nice to see you here too. There are, I’ve noticed in my recent forays into philosophy of mind, many versions of physicalism. I see no problem with being happy with one physical world (which I am) yet also seeing that complex physical phenomena create new ontologies that cannot be reduced. I’ve said that quite clumsily but I think you may get my point.
Get it and agree. I'm one of those who differentiate between materialism and physicalism, the difference being — exactly as you say — that emergence involves new high-level laws. As I usually state it, under physicalism, reduction doesn't always work.
My one qualm with the Mary's Room experiment is how it reinforces the idea that perception is what is foundational to consciousness. One small tweak makes this obvious...
Mary has spent her entire life trapped in a room that perfectly meets every homeostatic need. But despite this limitation, she has learned all the physical information there is to know about FEAR, from the hormonal interactions to the precise neural activity patterns it triggers.
What will happen when Mary is released from her perfectly safe room and is threatened for her life? Will she learn anything about FEAR or not?
Everything about color can be captured and represented via a smartphone. But a smartphone can capture nothing of fear. Feelings have no corresponding informational content beyond that felt experience.
Which should tell us something about the primacy of feelings in our conscious experience, no?
Yes agreed R.B. Perceptions like color should not be foundational to consciousness if our non-conscious computers can in a sense “perceive” input information that we might just as well call color. So then what might be foundational to consciousness? I think changing the scenario over to fear was an effective move. This brings an inherent dread or personal badness regarding existence. And of course you might have presented a scenario where Marry knew all there was to know about the taste of good food, except had somehow been prevented from ever having such experiences until her handlers permitted it. Thus she’d now be rewarded in that way. I consider the felt goodness to badness of personally existing to not just be foundational for consciousness, but to fully constitute the value of existing for anything, anywhere. Back when algorithms weren’t able to deal with more open circumstances very well, I suspect that this goodness to badness of existing evolved essentially as fuel from which to motivate an agency form of function.
I chose fear, but you could pick any of the primal feelings that we share with all animals with a similar brainstem: rage, care, lust, panic, etc.
And I agree with you in that every feeling is valenced: experienced as good or bad, to the degree that it deviates from or returns to any of our various homeostatic needs. But valence is essentially a quantity, which seems necessary but not sufficient to me as a foundation. The qualitative difference between feelings is also necessary to prioritize action in the face of uncertainty. This is what justifies the enormous energy costs of consciously processing feelings.
This engaging read touches on the question of whether we are merely organic machines, or something more. As Robert Sapolsky notes in "Determined" (in which he argues that free will is an illusion) no one has been able to put their finger on where consciousness -- the thing that makes choices -- lives in the brain. Others, like Rupert Sheldrake, are open to the idea that consciousness resides elsewhere, perhaps everywhere.
Read my novel, THE SOUL GENE https://tinyurl.com/thesoulgene
Mary's Room is just a (rather complicated and highly artificial) metaphor for what we all call "new experiences". Could you anticipate the taste of durian by combining knowledge of the chemistry and other people's descriptions? As they say: you have to be there. So it is open to replication but I don't see how you could design a test to answer the question.
I have to be honest - I had to Google what a durian was! For those who haven't encountered it either, it's a Southeast Asian fruit famous for its powerful (and controversial) smell and unique taste - people tend to either love it or hate it (apparently).
You make a great point about connecting this to everyday experiences. You're right - there does seem to be something special about 'being there'. But I've been wondering-- don't we need to 'be there' for everything we learn? At some point, we had to 'be there' to learn that the Earth revolves around the Sun or that water is a compound of hydrogen and oxygen. We've probably forgotten where, when, and who taught us these facts, but we had to 'be there' to learn it, right?
And no airline will carry durian because they tend to overripen at low air pressure and stink the plane out! So you have to go to them!
we had to 'be there' to learn it, right? Yes.
I was wondering the other day to what extent memories are raw or cooked.
Raw = remember the sensory input.
Cooked = remember the processed result.
When I was a child I disliked canned beans (we call them Heinz Beans, dunno what you call them - haricot beans in tomato sauce)
Anyway, as I got older "I learnt to like them" (as my mother used to say) and when I remembered back the taste no longer seemed unpleasant, so I was remembering the raw quale, not the cooked one which would have included the distaste.
If you think of other memories they're cooked to varying degrees: words of a song: if you remember the original artist singing it, it's raw. If you only remember the words and tune, it's cooked.
Most schoolwork is cooked, but there are probably a few raw nuggets.
Oh! This is fascinating. I'm very curious about this idea -- what's the difference between raw and cooked experience/memory? It reminds me of a thought experiment by Daniel Dennett about coffee tasters (which I'm planning on writing about in a couple of weeks).
I thought I explained: raw is when you remember the raw sensory input. Cooked is when you remember your response.
Thanks for doing this series, Suzi, and getting me to finally read that paper. As you know, I don't think epiphenomenalism has many virtues, but I don't agree with physicalism either (and I find it funny that he Jackson thinks he's crusading against physicalism when he's practically a physicalist in my view). I don't think any of the standard arguments I've heard against Mary's room—the difference between 'knowledge' and 'know how'; that Mary's knowledge of some hypothetical completed science would have to include that of the experience of color—do anything to dispel the problem I'm having with the experiment. Maybe it's just me, but it seems the entire project of color science, of determining the 'objective' nature of color, depends on us having the 'subjective' experience of color in the first place. Of course you can't begin with the science and get to the experience; it's the other way around. I didn't think this was controversial. Or maybe I'm missing the point of the entire controversy, I don't know.
What about something like gamma rays or neutrinos? We cannot experience those, but our science discovered them.
We can't experience them directly, but posit their existence through what we assume to be their effects.
Right. Why couldn't that happen with color if it was something we didn't experience directly? (Gamma rays, for example, are a form of light.)
We talked about this once a little while back, although I think we were considering a slightly different sort of set up. Here I'm talking about phenomenally-experienced color, which I take to include black and white. So in other words, in a completely colorless world we wouldn't visually perceive boundaries to objects; we would be blind. Objects could only be known by the other senses. This topic is touched on in Jackson's paper, I just realized, and in it he brings up H.G. Wells' story "The Country of the Blind" which is about "a sighted person in a totally blind community. This person never manages to convince them that he can see, that he has an extra sense. They ridicule this sense as quite inconceivable, and treat his capacity to avoid falling into ditches, to win fights and so on as precisely that capacity and nothing more." Keep in mind the entire community must be incapable of seeing in order for this to work. So if we're all incapable of seeing color, effectively blind, how would we come to detect wavelengths? In other words, what would prevent us from saying of color-detecting creatures that they were reacting to certain textures, perhaps even concocting a whole complicated theory around texture, for instance? Or we might be inclined to say they were merely disposed to behave the way they do. Maybe you can figure out a way to come up with wavelengths from here, but I'm at a loss.
I think I can answer your question but let me first ask about what constraints you have in mind. Do we have our other senses, and do we have our same level of intelligence and curiosity?
FWIW: My reply got long, and I think your attention has turned to your recent post about idealism, so rather than post a long comment here, I made it into a Note.
https://substack.com/profile/195807185-wyrd-smythe/note/c-78394790
Sorry, my bad! I wasn't clear in what I was saying above. I agree it seems like we might come up with the idea that heat is some sort of energy, but I meant "color" wavelengths (I realize I just said "wavelengths"...which is confusing.) BTW, sorry it took so long to get back to you. I was inundated for a minute there, and I'm finding I don't always see comments or replies in my email. Not sure what's going on.
Hey Tina, I'm glad you brought up the Wells story; it's a good one.
I might quibble over whether we should think about colour as the reason we see boundaries -- our visual system distinguishes between processing edges and processing colour -- but I feel like that might be missing the point.
The interesting question is, even if we had no visual experience of light (we were blind), would we still discover wavelengths? We already interact with wavelengths we can't see - we feel infrared as heat, some animals detect ultraviolet, and we've developed instruments to measure radio waves and X-rays.
To understand wavelengths, we need to understand spatial dimensions and movement over time. And it seems like we could develop these concepts even if we were blind.
We can feel movement; we can learn to understand space through touch and sound. We often represent waves with visual schematics, but waves like infrared radiation can be represented as sound. Even without being able to see light, we might still discover that there's something that travels in straight lines, reflects off surfaces, and carries energy. We might still figure out that this something has wave-like properties that can be measured and manipulated.
Thanks Suzi,
Yes, you're guessing right. I should have made it clear I wasn't thinking of color in any scientific sense, but more in a pre-theoretical sense. Also I see now that I should have made it clear I was thinking of color wavelengths—I omitted 'color' in talking about wavelengths above, which is confusing! I agree we could probably come up with a theory about heat and energy, given that those don't rely on vision. I can't imagine how we'd arrive at the concept of color wavelengths specifically. Sorry about the confusion!
All I meant to point out was the funny relationship our scientific understanding has to our perception, especially in this instance. It seems our science is rooted in our experiences, here, our perception of colors, and these experiences shape our ability to conceptualize and investigate the world around us, and yet at the same time these color experiences are deemed subjective.
Thanks, Tina!
Color is a funny one, isn't it!? What we perceive as colour doesn't really exist -- it's just wavelengths of light. But in order to perceive those wavelengths as colour, we need certain physical structures (photoreceptors, V4 neurons, etc.) -- without them we don't see colour at all.
This reminds me of a current debate currently happening in psychology/neuroscience. Much of our understanding of how the brains works started with subjective experience (or psychology). We created psychological models and then looked for their neural correlates. In essence, we've taken an outside-in approach to understanding the brain.
But could we flip this approach? Instead of starting with the psychology (or the subjective experience) and looking for its neural basis, could we start with the brain. For example, could we start with how photoreceptors, neural pathways, and brain circuits process wavelengths of light, and let that understanding inform our concepts? I do wonder whether this inside-out approach is possible.
I am delighted that you get my point and its implications, and so readily too.
Yes, color is a funny one. I can see why Jackson chose to talk about color instead of, say, sound, as science explicitly marks color as being subjective and therefore unreal in some sense. What I find strange about both the argument and the counter-arguments is the assumption that science can somehow be characterized as a collection of "information" and "facts" divorced from anyone ever experiencing anything. It's like everyone wants to render sensory perception completely irrelevant to science. For instance, Jackson says, "I think that there are certain features of the bodily sensations especially, but also of certain perceptual experiences, which no amount of purely physical information includes." If he's saying scientific descriptions exclude subjective experiences, well, duh. But he seems to be saying more than that. Why is no one pointing out that sensory experience is required for science? And that it must be taken as the data for which science is supposed to be accountable in order to be empirical—knowledge derived from sensory perception? Are we moving towards a purely semantic ontology with no empirical constraints? What is this bizarre trend of throwing doubt on the reliability of experience, going so far as to call it an illusion, while at the same time propping up science as "all there is to know". I'm thoroughly confused. Or am I just living in the dark ages while the rest of the world has gone post-empirical? Anyway, thanks, Suzi for understanding. Feel free to take these questions as rhetorical. :)
Due to a congenital toxoplasmosis infection, my son Jakob, who is now 41 years old, has severely damaged retinas, a seizure disorder, and various physical and intellectual impairments. He is a graduate of the legendary Perkins School for the Blind. Despite his various limitations he has until recently been a voracious, compulsive reader — especially of books about science, neuroscience in particular. He reads normal-print using what he calls 'facebook' technique — with the book about 3 inches from his face.
Earlier this this summer the cataract in Jakob's one eye that has useful vision had become so opaque that he could no longer see well enough to read. Two days ago, on November 18, in a long operation under general anesthesia, that cataract was removed. Because of the extreme fragility of the structural parts of his eye, no new lens was inserted. After the eye has had time to heal, Jakob & his medical team will determine whether a second operation should be done to insert a lens, or if he should wear glasses, or if he will just use the vision he has due to the removal of the cataract — which was, according to the surgeon, extremely dense. Having read this essay I'm now wondering how long that cataract has been developing, and how big a jolt his brain is getting from the (presumably) vastly increased supply of visual data to his retina. Over decades of taking Jake to see eye specialists, the main concern his always been his retinas — because toxoplasmosis attacks them differentially. I don't recall much discussion about cataracts before about a year ago, but who knows.
Jake has face-blindness. He recognizes people — including me, and his mother & sisters — by their voices. I had always assumed that his face-blindness came from damage to the portion of the brain that specializes in recognizing faces, but this essay makes me wonder if that deficit was the result of not getting the data needed to develop that ability.
It's too soon to know if Jakob will be able to read this essay on his laptop. If he can't, I will either read it to him or figure out how to have a synthetic voice read it for him. In any event I very much look forward to hearing his comments on it. Through Perkins, Jake knows deafblind people, and people who have had operations of various kinds — such as cochlear implants, or cornea transplants — to restore, or provide for the first time, new kinds sensory input. This is a subject he knows a lot about.
My son may not be a 'Mary,' raised in a black & white world who is now experiencing color for the first time, but I'm sure that he will consider her familiar, the kind of person who were his friends and classmates during his seven years at Perkins.
I recently wrote about some recent adventures Jakob & I have had in the land of toxo. This essay includes some photos of just a few of the hundreds of books that fill his apartment.
https://open.substack.com/pub/johnsundman/p/toxoplasmosis-my-pazuzu-part-one?r=38b5x&utm_campaign=post&utm_medium=web
After Jakob has had time to recover from the surgery and process this essay I'll be sure to write up his impressions.